Compare commits

..

173 Commits

Author SHA1 Message Date
Anas Nashif
d4e799d77a Zephyr 1.6.0
Change-Id: Iccecc12218132ff1eae209d2dd17edaf71b94d5a
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
2016-12-03 13:31:44 +00:00
Anas Nashif
c3d9ce8e27 release-notes: minor updates
Change-Id: Ic93535da51462cffbe6b19f79034f82a0344fe27
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
2016-12-03 13:31:34 +00:00
Anas Nashif
af4e5d44f2 sanitycheck: update footprint data
Change-Id: I2a0b480b42bcb8daf8a941848eb59f35ffbf844a
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
2016-12-02 22:58:35 -05:00
Andrew Boie
18928e7f60 nios2: fix irq_lock/unlock ordering bug
Memory accesses could be reordered before an irq_lock() or
after an irq_unlock() without the memory barriers.

See commit 15bc537712 for the
ARM fix for a complete description of the issue and fix.

Change-Id: I1d96fe0088d90150f0888c2893d017155fc0a0a7
Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
(cherry picked from commit bba445b31f)
2016-12-03 00:05:35 +00:00
Benjamin Walsh
d39e0a92c4 arc: fix irq_lock/unlock ordering bug
Memory accesses could be reordered before an irq_lock() or after an
irq_unlock() without the memory barriers.

See commit 15bc537712 for the ARM fix for
a complete description of the issue and fix.

Change-Id: I056afb0406cabe0e1ce2612904e727ccce5f6308
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-12-02 18:07:57 +00:00
Benjamin Walsh
c9235e2833 x86: fix irq_lock/unlock ordering bug
Memory accesses could be reordered before an irq_lock() or after an
irq_unlock() without the memory barriers.

See commit 15bc537712 for the ARM fix for
a complete description of the issue and fix.

Change-Id: Ic92a6b33f62a938d2252d68eccc55a5fb07c9114
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
(cherry picked from commit 1f8125a416)
2016-12-02 18:07:29 +00:00
Marcus Shawcroft
d97299c9fa arm: fix irq_unlock() ordering bug
Add the missing memory clobber to irq_unlock() in order to prevent the
compiler reordering memory operations over the unlock.

Change-Id: If1d664079796618ed247ff5b33b8b3f85fb7e680
Signed-off-by: Marcus Shawcroft <marcus.shawcroft@arm.com>
2016-12-02 16:38:41 +00:00
Marcus Shawcroft
6e2fbead3a kernel: Fix ARM irq_lock() ordering bug.
The inline asm definition of irq_lock() on the ARM architecture marks
the ASM as volatile which prevents the compiler from removing the
isntruction but does provide any information to the compiler to
prevent the inline ASM instruction being re-ordered relative to other
instructions.  The instruction used in irq_lock() do not touch memory,
however in order to acheive their intended purpose they must be
ordered relative to other memory access instruction.  This is acheived
by adding the "memory" clobber.

Instances of the compiler inappropriately re-ordering irq_lock() calls
relative to other instructions without this patch can be observed in
the code generated for k_sleep() on NRF51 target boards.

Signed-off-by: Marcus Shawcroft <marcus.shawcroft@arm.com>
Change-Id: I9d42d54cd9a50e8150c10ce6715af7ca2f5cfe51
(cherry picked from commit 15bc537712)
2016-12-02 15:03:29 +00:00
Mahavir Jain
9adaf59d36 work_q: delayed work cancel returns incorrect status
If delayed work is already submitted or completed, then subsequent
cancel should return -EINVAL as return status.

Fixes ZEP-1373.

Change-Id: I16bbacca7e31a5a5d8e5a89e729d70302ada6223
Signed-off-by: Mahavir Jain <mjain@marvell.com>
(cherry picked from commit 45f2ef653d)
2016-12-02 12:51:04 +00:00
Anas Nashif
fcbd5e7bc7 release notes: update release notes for 1.6
Change-Id: Ibb560b0382aadc76d393d76ab71dbef2b268ecf6
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
2016-12-01 23:15:54 -05:00
Anas Nashif
4311814a05 Zephyr 1.6.0-rc4
Change-Id: Ib357207570e7f03434e66da748439983c1cdf27e
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
2016-12-01 21:30:20 -05:00
Ramesh Thomas
70c57b2b96 doc: power_mgmt: Update PM doc with latest changes
Revise the document to reflect the latest changes including
the updated concept of SOC interface instead of PMA. Simplified
and enhanced areas that were known to cause confusion. Added
descriptions of new APIs and usages.

Jira: ZEP-1386
Change-Id: I5fa74d85245924f512c22d9d977dd0c9ea62b6ce
Signed-off-by: Ramesh Thomas <ramesh.thomas@intel.com>
2016-12-01 21:12:40 -05:00
Jithu Joseph
bdd6b19a52 usb :mass_storage: correct the license details
Certain structures and defines in this file are from
from mbed's implementation. The file header is updated
as per this.

Change-Id: I688917cdd17cfc8b27d5b78181ced90df73c9efd
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
2016-12-01 21:12:32 -05:00
Benjamin Walsh
a599dbd128 kernel: add emphasis to nano_sem_take/k_sem_take return code difference
The reversal of the meaning of a value of 0 from k_sem_take vs
nano_sem_take has caused some issue when porting code from the legacy
API to the new API, so put some emphasis on this difference.

- Add a note in the API description.
- Put the call to k_sem_take and the reversal of the return value inside
  of nano_sem_take on one line so that grepping on it shows the
  reversal.

Change-Id: I2f4ba58dc087176d68b55371fa6e367b72559e70
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-12-01 21:12:32 -05:00
Anas Nashif
4344299c00 [REVERTME]boards: omit frame pointer for ARC boards
The current ARC GCC compiler used in Zephyr SDK v0.8.2 generates
incorrect code when using the "-fno-omit-frame-pointer" option. This bug
should have been fixed in the 2016.03 release of the compiler.

Jira: ZEP-1243, ZEP-1403

Change-Id: I0901f55973c1ea37491b07bf625d0d1918803f3e
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
2016-12-01 21:12:32 -05:00
Inaky Perez-Gonzalez
0307d6ea5f scrips/kconfig: reduce impact of getenv() buffer overflow
getenv() returns an string of unknown size, so Coverity warns that it
might be used to overflow the stack in the call chain off
conf_read_simple().

To avoid that, wisdom says copy to an string of known size and pass
that.

Change-Id: I9e468de0ae66429062027f58fe0a0a4e1197218f
Coverity-ID: 150819
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
2016-12-01 23:37:47 +00:00
David B. Kinder
f38cbb5744 release-notes: fixed reST errors
Change-Id: I02c02dfcfd5b12ba5df745b42eed3b55401d8fb0
Signed-off-by: David B. Kinder <david.b.kinder@intel.com>
2016-12-01 00:41:29 +00:00
Anas Nashif
4c0d57ed3e release-notes: Update and cleanup
Change-Id: Iffd4c117f5ac7696652f6458009821fb351205f8
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
2016-11-30 15:53:00 -05:00
Maureen Helm
4540aa0877 release-notes: Update executive summary and drivers
Change-Id: I90c6cc53a78696a49a1e64d513b6fe2a5b7c7ccf
Signed-off-by: Maureen Helm <maureen.helm@nxp.com>
2016-11-30 12:05:44 -06:00
Kien Dinh
1c7c4dd43e release-notes: update executive summary
Change-Id: I445beb8ea475d6f381baf9b4825ebd7a675721fb
Signed-off-by: Kien Dinh <kien.t.dinh@intel.com>
2016-11-30 04:39:38 +00:00
Anas Nashif
a16bc64bf8 Zephyr 1.6.0-rc3
Change-Id: I6c1592a77a7ad0a5bee28e03967345999353c4e8
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
2016-11-28 22:42:21 -05:00
Benjamin Walsh
3d37868d09 arm: fix bug when Zero Latency Interrupts are enabled
An IRQ would always register as a ZIL interrupt.

Change-Id: If82a85f472a60512745652aacc7e8b7dfacaa268
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-11-28 16:00:04 -05:00
Benjamin Walsh
98a001e1f8 arc: fix missing _firq_stack symbol when INIT_STACKS=y and NUM_BANKS=1
There is no FIRQ stack in the system in this case, so do not initialize
it.

Change-Id: I8bc068ce43ac8a39909994d8cc01ba0c6a17f4ae
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-11-28 16:00:04 -05:00
Benjamin Walsh
bf2eb5542a kernel: remove K_TIMING thread flag
The fact that a thread is timing out was tracked via two flags: the
K_TIMING thread flag bit, and the thread's timeout's
delta_ticks_from_prev being -1 or not. This duplication could
potentially cause discrepancies if the two flags got out-of-sync, and
there was no benfits to having both.

Since timeouts that are not parts of a thread rely on the value of
delta_ticks_from_prev, standardize on it.

Since the K_TIMING bit is removed from the thread's flags, K_READY would
not reflect the reality anymore. It is removed and replaced by
_is_thread_prevented_froM_running(), which looks at the state flags that
are relevant. A thread that is ready now is not prevented from running
and does not have an active timeout.

Change-Id: I902ef9fb7801b00626df491f5108971817750daa
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-11-28 16:00:04 -05:00
Benjamin Walsh
51859b8ea0 kernel/arch: move common thread.flags definitions to common file
Also remove NO_METRIC, which is not referenced anywhere anymore.

Change-Id: Ieaedf075af070a13aa3d975fee9b6b332203bfec
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-11-28 16:00:04 -05:00
Ramesh Thomas
7aa536789e quark_se_ss: power_mgmt: Fix a bug in call to sleep instruction
When sleep instruction is called with interrupts enabled, the
interrupt priority threshold bits need to be set. Only interrupts
with equal or higher priority will wake the sleep. Currently it
is set to 0 unintentionally and only priority 0 interrupt can
wake the sleep.

Jira: ZEP-1349
Change-Id: I927e259345cc37c5ecc4dfdcde996dd16443e61b
Signed-off-by: Ramesh Thomas <ramesh.thomas@intel.com>
2016-11-28 16:00:04 -05:00
Ramesh Thomas
05a0c6fef0 quark_se: power_mgmt: Fixes a cpu context save bug
The cpu context save function was manipulating stack and
returning to C caller. This can corrupt stack if the calling
function has data saved and it pops before entering deep
sleep. Moved sleep functions into assembly to avoid this.

Jira: ZEP-1345
Change-Id: I8a6d279ec14e42424f764d9ce8cbbef32149fe84
Signed-off-by: Ramesh Thomas <ramesh.thomas@intel.com>
2016-11-28 16:00:04 -05:00
Jithu Joseph
9471d1f6c8 samples: event_collector: Remove redundant check
Removes a redundant check flagged by coverity.

Coverity-CID: 152005

Change-Id: I8cc3a64c42e04a2d52deed11d9022ed4a49baaa7
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
2016-11-28 16:00:04 -05:00
Inaky Perez-Gonzalez
09f4f54e72 tests/ztest/mock: remove usage of legacy k_fifo_get()
Legacy FIFO operations were failing and thus the TC was failing to run.

Stop using k_fifo_get() for allocation and use a bitmap allocator. A
couple of the bitmap operations should be moved to a common header
once ZEP-1347 is completed.

Passes on all arches and boards, whitelist removed; ARM excluded
though due to missing bitfield implementation as per ZEP-82.

Note there is a false checkpatch positive in the decl of
sys_bitfield_find_first_clear().

Change-Id: I5d43f804d6bec3a464124accbe3be238f9cade82
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
2016-11-28 16:00:00 -05:00
Inaky Perez-Gonzalez
c2fe55bbe5 libc/minimal: snprintf(): KILL negative len parameter
snprintf() implements the ability to foce a negative value through the
(unsigned) size_t len parameter to allow the formatter to use a
maximum size string.

This is point less, we don't have as much memory and this is a recipe
for all kinds of vulnerabilities.

Kill the whole thing, the testcase it represents and thank Coverity
for finding this thing. Whatever use it had before, it has no more.

Change-Id: If422246548664699d8aa328a1b9304ef13cab7ea
Coverity-ID: 131625
Coverity-ID: 131626
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
2016-11-28 15:53:37 -05:00
Benjamin Walsh
70028dd97f arm/frdm_k64f: fix boot issue with MBED bootloader and INIT_STACKS=y
Initializing the interrupt stack before initializing (turning off) the
watchdog on the FRDM board pushed the initialization of the watchdog too
late, causing it to fire and reset the board. The board would be kept in
a reboot loop.

Move the initialization of the watchdog earlier: this runs on the main
stack now, instead of the interrupt stack, the same stack the interrupt
stack initalization code runs on.

Change-Id: Ic0006f4f4f4090393571d8355a80dc9390c9fbc6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
(cherry picked from commit eba017632a)
2016-11-28 18:34:02 +00:00
Szymon Janc
d15b758632 Bluetooth: GATT: Fix primary service discovery response
Applications expect service end handle as attribute value in userdata
on discovery response callback.

Jira: ZEP-1354

Change-Id: I664da4a7e054a531ad1c2c8cbc74367cb679ff03
Signed-off-by: Szymon Janc <ext.szymon.janc@tieto.com>
2016-11-27 14:36:54 -05:00
Anas Nashif
999c15d1b5 release notes: update highlights
Change-Id: Ieec2b3de4770e1717a826054e8264d1978ed23a8
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
2016-11-27 14:36:03 -05:00
Carles Cufi
d108946a1b Bluetooth: hci_uart: Fix init order in hci_uart bootup
bt_enable_raw() needs to be called before spawning the tx thread,
otherwise there might be an HCI command processed from the UART
before the HCI driver has been opened and therefore initialized.

Change-Id: I050158bd48bebaf8fa2cf6b11efb54b531f70079
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
2016-11-24 10:52:13 +00:00
Anas Nashif
97ab403573 release-notes: cleanup entries
Change-Id: I12bc617d6886050114d9f96cdf5935e36e79552a
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
2016-11-24 10:51:51 +00:00
Johan Hedberg
63538db423 release-notes: Add Bluetooth changes
Add changes to the Bluetooth subsystem since 1.5.

Change-Id: I3a0554c7bc3c5d22cfb244c83152f3805809fd99
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
2016-11-24 10:53:37 +02:00
Javier B Perez
272ec5e219 release notes: add release notes doc
Added release-notes.rst file.
Added MAINTAINER entry for the release notes.

Change-Id: Ia2e7645c905e709f014e7d971337bcfc9cb59597
Signed-off-by: Javier B Perez <javier.b.perez.hernandez@intel.com>
2016-11-24 07:06:27 +00:00
Benjamin Walsh
82804fe115 arm: fix early boot on Cortex-M0 with init stack
The assembler was passed immediate values that are too large for the
limited Cortex-M0 thumb assembly. Load values in registers instead of
using immediate values.

Change-Id: Ib5541c92dea03e0efb1b88ab91eeb408d151a71b
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-11-24 02:33:59 +00:00
Vincenzo Frascino
672dc9cc89 arm: Enable REBOOT when RUNTIME_NMI is selected
This patch enables REBOOT when RUNTIME_NMI is selected via defconfig
file. This action is required to prevent compilation errors.

Change-Id: I06869cb86b1abc151974df66797a0b25ee62e166
Signed-off-by: Vincenzo Frascino <vincenzo.frascino@linaro.org>
(cherry picked from commit cc288335fbf9353b9c8af64fccefe96daa13dd8e)
2016-11-23 16:11:05 +00:00
Luiz Augusto von Dentz
74d75f2bd5 Bluetooth: L2CAP: Fix possibly reading past the end of buffer
If the original buffer cannot be reused, either by no having enough
space for user data or if is fragmented, it can in fact be smaller than
both the segment buffer and MPS.

Change-Id: I59a537aff59c5d56b2883e9bd51f3a1a3932d348
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
2016-11-22 20:03:24 -05:00
Luiz Augusto von Dentz
c5a40d60bb Bluetooth: L2CAP: Fix segmentation
The segments need to be limited by the minimun of the segment buffer
tailroom and tx MPS not the original buf length.

Change-Id: I580a3bb61aa190ac0cdd3717bc06fd6e6e668304
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
2016-11-22 20:02:48 -05:00
Luiz Augusto von Dentz
189e5d0006 Bluetooth: L2CAP: Fix regression with move to k_sem API
k_sem_take return differ from nano_sem_take since it return 0 for
successful case instead of 1.

Change-Id: Ia39cd624d56dbc1c8e7f3558244bebf765da191d
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
2016-11-22 20:01:55 -05:00
Szymon Janc
039a130861 Bluetooth: Kconfig: Remove deprecated dependency on NANO_TIMEOUT
This is no longer needed after switch to unified kernel.

Change-Id: Ie1f8dadb3f2e43ae6ccfbfaf1f754196f3237471
Signed-off-by: Szymon Janc <ext.szymon.janc@tieto.com>
2016-11-22 20:01:03 -05:00
Szymon Janc
61ec28adb6 Bluetooth: tests: Fix Makefiles comments
Those tests are now build with unified kernel.

Change-Id: Idbc42bb77060cea0130d62cccdf2e40aeee89128
Signed-off-by: Szymon Janc <ext.szymon.janc@tieto.com>
2016-11-22 20:00:14 -05:00
Szymon Janc
4e38776774 Bluetooth: Kconfig: Remove deprecated dependency for ECC support
After switch to unified kernel this is no longer needed.

Change-Id: If9877d3fa038dd873011fb780c7e767e150647ae
Signed-off-by: Szymon Janc <ext.szymon.janc@tieto.com>
2016-11-22 20:00:13 -05:00
Inaky Perez-Gonzalez
14dc173c1f libc: remove useless code in _prf()
Coverity reported a block of deadcode in _prf() that seems to be a
leftover carcass from a previous time. Replaced with a comment in case
someone decides it was needed back.

Change-Id: Id97e84f3279f807e6188371f27f6af157e6d5038
Coverity-ID: 131631
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
2016-11-22 19:58:47 -05:00
Inaky Perez-Gonzalez
0a48547bc6 tests/drivers/pci_enum: move to ztest and run in HW when possible
This test case just exercises the PCI enumeration and there is no real
way to test success/failure other than running it and the kernel not
crashing.

Moved to ztest.

Retag so it is actually ran on QEMU/x86 and galileo once we deploy in
the HW pool. Note this means that we need to force CONFIG_PCI on
Qemu/x86, which can run this testcase.

Change-Id: I85b64800f7d989357927b4a25777041047293b34
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
2016-11-22 19:35:30 -05:00
Ramesh Thomas
16f5611f3e tests: power_mgmt: Fix wrong value being passed to post_ops func
The app passes the index into an array storing power states
instead of the power state to _sys_soc_power_state_post_ops

Jira: ZEP-1341
Change-Id: I6ddf0a2dbadfd06aafbcafa88be7441e99694a51
Signed-off-by: Ramesh Thomas <ramesh.thomas@intel.com>
2016-11-22 19:34:58 -05:00
Vincenzo Frascino
7336b2a978 arm: Fix CONFIG_RUNTIME_NMI behavior
Zephyr kernel is unable to compile when CONFIG_RUNTIME_NMI is enabled in
defconfig on ARM's architectures.

This patch addresses the following issues:
* In nmi.c _DefaultHandler() is referencing a function
(_ScbSystemReset()) not defined in Zephyr. This has now been replaced
with sys_arch_reboot.
* nmi.h is included in ASM files and due to the usage of "extern" the
compilation ends with an error. Added the directive _ASMLANGUAGE to
prevent the problem.

Jira: ZEP-1319
Change-Id: I7623ca97523cde04e4c6db40dc332d93ca801928
Signed-off-by: Vincenzo Frascino <vincenzo.frascino@linaro.org>
2016-11-22 19:34:52 -05:00
Benjamin Walsh
42cf1ab802 kernel/mbox: add missing dummy thread timeout init
It was possible for a dummy thread to be not timing, but not having
timeout.delta_ticks_from_prev not be -1 at the same time, which is a big
no-no.

Use _init_thread_base() to do a full initialization of the dummy thread.

Fixes ZEP-1312.

Change-Id: I16a2373be3329c142cf26f5dca6bfdbe6014ac5e
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-11-22 19:34:52 -05:00
Benjamin Walsh
05291174df kernel: streamline initialization of _thread_base and timeouts
Move _thread_base initialization to _init_thread_base(), remove mention
of "nano" in timeouts init and move timeout init to _init_thread_base().
Initialize all base fields via the _init_thread_base in semaphore groups
code.

Change-Id: I05b70b06261f4776bda6d67f358190428d4a954a
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-11-22 19:34:52 -05:00
Benjamin Walsh
6c8409c083 arc: support interrupt/FIRQ stacks with CONFIG_INIT_STACKS
Use the main stack during very early boot so that we can call memset on
the interrupt and FIRQ stacks. Iniitalize the them before one of them is
used for the rest of the pre-kernel initialization.

Change-Id: Ib57856a66273dda9382e08fa91da5a54847b77c2
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-11-22 19:34:52 -05:00
Benjamin Walsh
bf591e9edf arm: support interrupt stack with CONFIG_INIT_STACKS
Use the main stack during very early boot so that we can call memset on
the interrupt stack. Initialize the interrupt stack before it is used
for the rest of the pre-kernel initialization.

Change-Id: I6fcc9a08678afdb82e83465cda1c7a2a8c849c9b
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-11-22 19:34:52 -05:00
Vinayak Chettimada
cf05794924 kernel: declare main and idle stack as globals
Renamed main_stack and idle_stack, to _main_stack and
_idle_stack, respectively, and made them globals. This does
not affect performance. They are still kept kernel private
symbols and not part of kernel API.

This will allow these symbols to be referenced in calls to
stack_analyse misc functions to profile stack usage in
applications.

Change-id: Id6b746c5cfda617c26901c6e62c3e17114471f57
Signed-off-by: Vinayak Chettimada <vinayak.kariappa.chettimada@nordicsemi.no>
2016-11-22 19:34:52 -05:00
Benjamin Walsh
60a31d6ed1 arc: use one of the interrupt stacks during early init
Same issue as with ARM. ARC can use _Swap() though, because the call to
it is serial, not generating a low-priority exception and interrupts are
locked until the main() thread is context-switched into and the
interrupt stack is released.

Fixes ZEP-1310.

Change-Id: Ie1f27f7ad0502191ca2867b5400d6e0bfb7f0fc6
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-11-22 19:34:52 -05:00
Benjamin Walsh
6b6572629d arm: use interrupt stack during early init
The ARM Cortex-M early boot was using a custom stack at the end of the
SRAM instead of the interrupt stack. This works as long as no static
data that needs a known initial value occupies that stack space. This
has probably not been an issue because the .noinit section is at the
very end of the image, but it was still wrong to use that region of
memory for that initial stack.

To be able to use the interrupt stack during early boot, the stack has
to be released before an interrupt can happen. Since ARM Cortex-M uses
PendSV as a very low priority exception for context switching, if a
device driver installs and enables an interrupt during the PRE_KERNEL
initialization points, an interrupt could take precedence over PendSV
while the initial dummy thread has not yet been context switched of and
thus released the interrupt stack. To address this, rather than using
_Swap() and thus triggering PendSV, the initialization logic switches to
the main stack and branches to _main() directly instead.

Fixes ZEP-1309

Change-Id: If0b62cc66470b45b601e63826b5b3306e6a25ae9
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-11-22 19:34:52 -05:00
Benjamin Walsh
96fc793c25 kernel: add support for switching to main thread without _Swap()
It's possible that an architecture needs a custom way of switching to
the main() task, rather than using _Swap() with a dummy thread.

Change-Id: I14e9bc67be35174ff16209bcea27b18a069ff754
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-11-22 19:34:52 -05:00
Benjamin Walsh
a30942dbb8 kernel/arch: remove unused uk_task_ptr parameter from _new_thread()
Artifact from microkernel, for handling multiple pending tasks on
nanokernel objects.

Change-Id: I3c2959ea2b87f568736384e6534ce8e275f1098f
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-11-22 19:34:52 -05:00
James Fagan
963d04d67e aio: Configuration issue in aio driver
The reference and polarity fields of the config struct which is passed
to qm_set_config are never cleared, meaning the wrong configuration
may be written if aio_set_config is called for different sets of pins
in the same program.

This patch clears these fields in aio_cmp_disable to prevent such an
issue.

Change-Id: I8feabae1f3d9fa4c7260d94c1ec919ef2fb84bfb
Signed-off-by: James Fagan <james.p.fagan@intel.com>
2016-11-22 19:34:27 -05:00
Marcus Shawcroft
b792e4277f gpio: Add doxygen markup for internal only definitions.
Change-Id: Ibc01d35199c643e186b58282af971975c0999e8b
Signed-off-by: Marcus Shawcroft <marcus.shawcroft@arm.com>
2016-11-22 19:34:08 -05:00
Marcus Shawcroft
b006b1bb9a gpio: Document public API return codes.
Change-Id: Ief1cd5c231b9b87ac7a0a80c2c005560fd29652b
Signed-off-by: Marcus Shawcroft <marcus.shawcroft@arm.com>
2016-11-22 19:34:08 -05:00
Tomasz Bursztyka
0fc5801607 clock_control: NRF5 Kconfig option should be available only on NRF5
Change-Id: If6ba2f5a03967ca2ddca8d4bc211bc0c55ae0312
Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com>
2016-11-22 19:33:51 -05:00
Mahavir Jain
271ab7d583 doc: fix ring_buffer code snippet
Change-Id: I5c1b08aea675ae9a9f636a1e8f0de5684dc403de
Signed-off-by: Mahavir Jain <mjain@marvell.com>
2016-11-22 19:33:39 -05:00
Flavio Santes
9f0e4d2a90 tinycrypt/sha256: Array compared to NULL has no effect
This commit fixes the issue reported by Coverity: an array compared
against NULL is always false.

Coverity-CID: 143715
Coverity-CID: 143730

Change-Id: Ie3c87f892c2b2a337981125e2a92c37c579d4b38
Signed-off-by: Flavio Santes <flavio.santes@intel.com>
2016-11-22 19:33:31 -05:00
Flavio Santes
979aedc2d3 tinycrypt/hmac: Array compared to NULL has no effect
This commit fixes the issue reported by Coverity: an array compared
against NULL is always false.

Coverity-CID: 143687
Coverity-CID: 143737
Coverity-CID: 143740

Change-Id: Id94a144c47b3377876695e86da8c0c33a989ec99
Signed-off-by: Flavio Santes <flavio.santes@intel.com>
2016-11-22 19:33:31 -05:00
Iván Briano
850877b95d ext hal qmsi: Avoid QMSI specific flags when QMSI is not used
The way the build system works, the Makefiles under ext/hal/* are being
included unconditionally, so anything they add to the build flags needs
to be protected by the correct configuration value.

Change-Id: I238e04cd836dd9e4c5d83040822039c68abb6b17
Signed-off-by: Iván Briano <ivan.briano@intel.com>
2016-11-21 17:49:34 -05:00
Anas Nashif
8f0b4d7f4d Zephyr 1.6-rc2
Change-Id: I930d11d2e41af3b77513531b8944ba77b5c5b278
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
2016-11-19 13:20:39 -05:00
Benjamin Walsh
796a6bb4d8 samples: configure philosophers with more than 32 priorities
To have one project use more than 32 priorities. The preempt priorities
are also aligned so that they straddle two priority bitmaps.

Change-Id: I0f0862110d876e40fde45a0d105b769e8603d644
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-11-18 18:46:15 -05:00
Benjamin Walsh
385e02ba52 kernel: support for more than 32 total priorities
In addition to more priorities taking more memory to host them, finding
the next thread to run when it is not cached is slower since each extra
set of 32 priorities maps to a loop iteration. That loop is remove
entirely when the number of priorities is less than 32 (31 + the idle
thread).

Fixes ZEP-1303.

Change-Id: I3205df90d379a0f4456ff1d7f1aaa67ad2cddf15
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-11-18 18:46:15 -05:00
Juan Manuel Cruz
5a5d878252 win-build: Fixes a kconfig incompatibility for Windows
In windows systems the rename() function fails if the new name
of the original file corresponds to a file that already exists.

The fix removes the new file before renaming the original one.

Jira: ZEP-980

Change-Id: Ib3a43db86c0dd3fabb592f53ea7619eb5738bb65
Signed-off-by: Juan Manuel Cruz <juan.m.cruz.alcaraz@intel.com>
2016-11-18 18:43:41 -05:00
Juan Manuel Cruz
cf6e5cf730 enc28j60: Fixes an issue reading/writing long frames from SPI
Jira: ZEP-1302
Change-Id: Ia58d51aee14281aaeb2f8e85fbbf8c250eae8e06
Signed-off-by: Juan Manuel Cruz <juan.m.cruz.alcaraz@intel.com>
2016-11-18 18:43:41 -05:00
Allan Stephens
7f1d5a47e4 kernel: Minor optimization to kernel event logger timestamping
Rewrites the timestamping logic to always generate timestamps
via a function pointer that is initialized to sys_cycle_get_32(),
but can be changed to point to a user-supplied function. This
eliminates the need for an if/then/else construct in every place
that a timestamp is generated.

Change-Id: Id11f8c41b193a93cece16565978a525056010f0e
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
2016-11-18 18:43:41 -05:00
Allan Stephens
aca2baa43a doc: Revise kernel event logger documentation
Makes the purpose and capabilities of the kernel event logger
clearer, and leaves much of the low-level detail relating to
use of the configuration options and APIs to the configuration
option guide and API guide, respectively. Also corrects some
bugs in the example code for retrieving event information.

Also updates the API guide to make a clear distinction between
the general purpose event logger framework and the kernel event
logger (which is a specific instance of this framework).

Change-Id: I924f65092b2b0e5050af13376b5da85a6cdc1a65
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
2016-11-18 18:43:40 -05:00
Allan Stephens
bf77e2616d doc: Fix up API descriptions for kernel event logger
Prepares the kernel event logger APIs for inclusion in the
API guide. Also corrects a couple of other issues:

* Gets rid of obsolete thread monitor code.
* Renames "timer_func" global variable to "_sys_k_timer_func"
  to align it with kernel naming conventions.

Change-Id: I93d403f83ae44ff45dda489c2ead7bfec6ce1fa3
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
2016-11-18 18:43:40 -05:00
Allan Stephens
cfa7ad5c4a kernel: Ensure event logger APIs convert timeouts to millseconds
Event logger APIs still express timeout delays in ticks;
need to convert to milliseconds when using unified kernel APIs.

Change-Id: I5fab66be660621cd2029417eaff3758e3ef4ba2c
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
2016-11-18 18:43:40 -05:00
Juan Manuel Cruz
30414fd866 sensor: fixes program hangs in the apds9960 sample
If an i2c transaction fails the sample will hang the program into
an infinite circle.

This commit will remove the infinite circle and report back the
error code from the i2c transaction.

Change-Id: I38d350a805af6bec43f2fa8d4af6ce4e3cc27662
Coverity-CID: 151991
Coverity-CID: 151992
Signed-off-by: Juan Manuel Cruz <juan.m.cruz.alcaraz@intel.com>
2016-11-18 18:43:40 -05:00
Juan Manuel Cruz
7d21b5402c sensor: fixes dead code in the apds9960 sample
If the gpio or spi devices are not found there is no
need to keep the device busy in a loop for this particular
sample.

Since it is not possible to continue execution it is better
to simply end the application.

Change-Id: Ie25ea970a479db2a2f339ca2b37f88541a45ef97
Coverity-CID: 151973
Signed-off-by: Juan Manuel Cruz <juan.m.cruz.alcaraz@intel.com>
2016-11-18 18:43:40 -05:00
Vinicius Costa Gomes
3ed3f29223 iot/zoap: Fix decoding of 16-bit delta
When an option code or length representation is encoded in a 16-bit
value, the access was wrong.

Coverity-CID: 151963

Change-Id: Ie7741998cbde348ccf490a6686e68a1ace99920e
Signed-off-by: Vinicius Costa Gomes <vinicius.gomes@intel.com>
2016-11-18 18:43:40 -05:00
Anas Nashif
e9a4431362 tests: test CONFIG_KERNEL_DEBUG and CONFIG_ASSERT
Enable this option to test any usage of structs and variables inside
macros.

Change-Id: I6ec64fb865e87fc0771ae10f0c4eb63f6144c88a
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
2016-11-18 18:43:40 -05:00
Benjamin Walsh
56816bdbbf kernel: fix obsolete access to fields in K_DEBUG() calls
When moving arch-specific thread structure to arch-agnostic, some field
accesses were missed when used in K_DEBUG statements, which are turned
off by default.

Change-Id: Ife0f49b8185a0db468deab73555f7034f20ca3e8
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-11-18 18:43:40 -05:00
Benjamin Walsh
420594122a kernel: fix thread prio and stack size types in some APIs
Prio should be an int, since values are small integers, not a fixed-size
int32_t. It aligns with the prio parameters of the other APIs.

Stack size should be size_t.

Change-Id: Id29751b86c4ad7a7c2a7ffe446c2a96ae83c77bf
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-11-18 18:43:40 -05:00
Inaky Perez-Gonzalez
0efe69ec31 test_static_idt: fix unininitialized variable
The divide-by-zero test was using an uninitialized variable that
Coverity was unhappy about. Simple fix to just initialize to any non
zero value.

Change-Id: I9e5865a99e7a8eb3ee52421cc3dcb6717dca1ad1
Coverity-ID: 152053
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
2016-11-18 18:43:40 -05:00
Inaky Perez-Gonzalez
8173b881ba tests/legacy/kernel/test_libs: use memcpy() vs strncpy()
Coverity complained about the use of strncpy() to fill up a buffer of
size N with a string of the same size didn't leave room for the final
\0.

This is a valid concern; however, the usage is valid too, as the
writer intended to create a pattern that later can be tested--addind a
\0 would break the pattern.

So instead, use memcpy() for the same function.

Change-Id: If52d02ce41731348f4a2d750c79f9e1c51f3afcf
Coverity-ID: 151947
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
2016-11-18 18:43:40 -05:00
Inaky Perez-Gonzalez
e7c091bba2 scrips/kconfig: use snprintf() vs sprintf()
Coverity reported 150819 issue, which steams off Flex generated code
from zconf.l in which sprintf() was use. Because of that, the
conf_read_simple() @name parameter could be used to overrun
zconf_open() @fullname by crafting SRCTREE and KCONFIG_ALLCONFIG
environment variables.

Change-Id: I2cff817dccafe0e06b35636bbb7be95e062410af
Coverity-ID: 150819
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
2016-11-18 18:43:40 -05:00
Inaky Perez-Gonzalez
4c11ae8e60 test_fp_saring/nanokernel: fix uninitialized variable
Coverity complains about using an uninitialized variable; there is no
reason to do so; thus fixed to avoid maintaining a whitelist.

Change-Id: I657f9e7d46b1b9b091e36638c1951b93903fbec3
Coverity-ID: 152048
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
2016-11-18 18:43:40 -05:00
Inaky Perez-Gonzalez
af941d5832 test_map: initialize memory block
This is not a strictly necessary initialization, as the data is not
used, but will keep Coverity happy. It being a testcase, there are no
size or speed penalties to the overall kernel and avoids having to
manage a whitelist for an issue in scanning tools.

Change-Id: I0ddcf43ca1114356d58f93de57232864246ffe07
Coverity-ID: 152052
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
2016-11-18 18:43:40 -05:00
Inaky Perez-Gonzalez
32de7848c4 test_map: fix uninitialized area
Coverity complained about the code using an uninitialized chunk of
memory; harmless, but fixed to avoid having to whitelist.

Change-Id: I5c890ff78fab2799b882b8e4a25c15476702d132
Coverity-ID: 152049
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
2016-11-18 18:43:40 -05:00
Inaky Perez-Gonzalez
afe1621713 test_static_idt: fix uninitialized variable
Coverity complains about an easy-to-fix uninitialized variable.

Change-Id: I04bf670c7137df25165d4e37f2f7df2d4004c478
Coverity-ID: 152050
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
2016-11-18 18:43:40 -05:00
Inaky Perez-Gonzalez
42b956050a test_fp_sharing: fix uninitialized variable
Coverity complains about this (harmless) issue, so simple fix.

Change-Id: Ibac952157cb0541dbd150d681515280091409864
Coverity-ID: 152051
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
2016-11-18 18:43:40 -05:00
Inaky Perez-Gonzalez
04af679197 samples/kernel_event_logger: initialize variable
Fix usage of an uninitialized variable detected by Coverity.

In theory GCC should pick up this situation, but it does not. I've
experimented with adding -Wextra and -Wuninitialized but I cannot get
GCC to complain. I might be missing something else, but in the
meantime, this is a simple fix to remove this issue.

Change-Id: I6fec37719719dfaf7077ce1f464605c93efa8ea2
Coverity-ID: 152054
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
2016-11-18 18:43:40 -05:00
Anas Nashif
b082e12265 kernel: remove v2 usage and rename KERNEL_V2_DEBUG
Change-Id: I6b3f07714322ad79aeec2342621a4cddfe84cb2c
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
2016-11-18 18:43:40 -05:00
Paul Sokolovsky
3cbabecfcc pinmux: Make default init priority be between GPIO's prio and device prio.
Pinmux driver almost certainly should be initialized before the
rest of hardware devices (which may need specific pins already
configured for them), and usually after generic GPIO drivers.
Thus, its priority should be between KERNEL_INIT_PRIORITY_DEFAULT
(default 40) and KERNEL_INIT_PRIORITY_DEVICE (default 50). Thus,
we set PINMUX_INIT_PRIORITY to 45.

There are exceptions to the rule above for particular boards. For
example, BOARD=galileo has GPIO and pinmuxer on I2C bus and thus
overrides PINMUX_INIT_PRIORITY to be much higher. Note that while
PINMUX_INIT_PRIORITY was defined previously (at 60), it was used
only for galileo, which overrides it anyway.

This fix was prompted by investigation why eth_ksdk driver was
non-functional after kernel priorities re-hashing: both eth_ksdk
and pinmux used the same priority, and eth_ksdk happened to run
before pinmux. While bumping eth_ksdk priority would help in the
particular case, the same would likely reoccur with other drivers
like I2C, SPI, etc.

Change-Id: Ie5ca3135c1ee2fe8d9cf48d5c12e62eac63487f7
Signed-off-by: Paul Sokolovsky <paul.sokolovsky@linaro.org>
2016-11-18 18:40:17 -05:00
Inaky Perez-Gonzalez
4d2ad79207 legacy/kernel/test_{static_id,stackprot}: 'fatal fault' is not a failure
By default,  when a  'fatal fault'  message is seen in the output of any
testcase,  it is consider an inmediate fatal condition and the test case
is aborted.

However,  in all such cases,  the testcase is provoking the situation to
verify the condition is caught.  In this case it shall NOT be considered
a fatal fault and the default overriden to allow it to proceed.

Change-Id: Id4e9138e5f0fcb8cd77efbb1831897fb0946ba20
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
(cherry picked from commit da5f3a5c89)
2016-11-18 19:42:58 +00:00
Allan Stephens
0cddc4b665 doc: Minor cosmetic tweaks for kernel API descriptions
Change-Id: Ie989b45b19e5e70958301dd8d903cf2876709f5a
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
2016-11-17 21:32:10 -05:00
Allan Stephens
4cdbfbf9e2 doc: Add descriptions for clock-related helper macros
Also fixes up Kernel Primer examples to use these macros.

Change-Id: Ib1bc9e3f85ab75f81986bc3930fb287266a886b5
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
2016-11-17 21:32:10 -05:00
Allan Stephens
c38888bccb doc: Revise timer example to use workqueue instead of alert
Rewrites the example of a timer's expiry routine offloading
processing that can't be done at interrupt level. The example
now submits work to the system workqueue directly, rather than
using an alert. This saves footprint by eliminating the need
for alert-related API support that isn't needed. (This is a
true savings, since the alert code just called the same
workqueue APIs the example now calls directly.)

Change-Id: I378e40aef33014f2c75c4f57531f75247d50e479
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
2016-11-17 21:32:10 -05:00
Allan Stephens
be0db01093 doc: Fix up API descriptions for ring buffers
Change-Id: I82453c1fb5365d7dfe35cb1bc9eba50c71a47b17
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
2016-11-17 21:32:10 -05:00
Allan Stephens
a9cd7c0498 doc: Fix up API description for IRQ_CONNECT()
Change-Id: I5ea1bd28f355d78c724948568c160ef1b32b5eb5
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
2016-11-17 21:32:10 -05:00
Allan Stephens
6af438c440 doc: Fix up return value descriptions for kernel APIs
Return value descriptions using the "@retval" tag now reflect
the fact that they appear on a separate line from the value
they are describing.

Change-Id: I3e3e347d133ad998e7db50a99369d41cbfb9efcc
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
2016-11-17 21:32:10 -05:00
Allan Stephens
0ae966be58 doc: Improve descriptions of workqueue APIs
The API guide now does a better job of explaining how to use
a workqueue. Also hides information about workqueue internals
and fixes several errors and omissions.

Change-Id: I6492c1c6105c258ce98365ca33059d8f32c1be41
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
2016-11-17 21:32:10 -05:00
Allan Stephens
33d0716c21 doc: Improve descriptions for some user-supplied functions
The API guide now does a better job of explaining how to correctly
write these functions.

Change-Id: Ib1df55eb28fa408f3f786f122353e37505002f07
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
2016-11-17 21:32:10 -05:00
Allan Stephens
a176dd3274 doc: Enable Kernel Primer links to macro-type APIs
Also adds a link to function-type API that was missing.

Change-Id: Ie671ad2f239cdca3ac1a2eb33248dfecfa251c79
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
2016-11-17 21:32:09 -05:00
Baohong Liu
ed8c6e2d1f samples: grove_lcd: stop the app if device binding fails
Proceed to LCD programming only if device binding succeeds.
Otherwise, dereferencing a NULL pointer will happen. This
was caught by Coverity.

Coverity-CID: 151986

Change-Id: Ibdb658f530203428aa3e53f358e0788fc1502b06
Signed-off-by: Baohong Liu <baohong.liu@intel.com>
2016-11-17 21:17:13 -05:00
Javier B Perez
c280ce6bf9 test: power states: fix dead code issue
Coverity detected some constant value in the vars, due to the
exclusive config select in the code.

Change-Id: Id27b658f3cd70dce626fef054457a9c726b3b957
CID: 151974, 151972 and 151971
Signed-off-by: Javier B Perez <javier.b.perez.hernandez@intel.com>
2016-11-17 21:17:13 -05:00
Sergio Rodriguez
67d49c2344 drivers: gpio_dw: Remove contradictory if statement evaluation
This fixes an always false evaluation of the gpio I/O direction

This issue was reported by Coverity

Coverity-CID: 151978

Change-Id: I93ec3319a3f18d564c961a5cbd9dcc9c60efbeb7
Signed-off-by: Sergio Rodriguez <sergio.sf.rodriguez@intel.com>
2016-11-17 21:17:13 -05:00
Sergio Rodriguez
f3c2664e53 drivers: gpio_atmel: Fix erronous if statement
The GPIO_INT_ACTIVE_LOW value is  zero so the mask assignement is
never executed. Using the bit complement GPIO_INT_ACTIVE_HIGH the
proper mask is assigned

This issue was reported by Coverity

Coverity-CID: 151966

Change-Id: Ibc7d2e4c3ebee249b5ab9719f8177cc14c0d1d33
Signed-off-by: Sergio Rodriguez <sergio.sf.rodriguez@intel.com>
2016-11-17 21:17:13 -05:00
Sergio Rodriguez
22bbdc2f85 soc: stm32f1: gpio: Fix unnecessary else statement
The bitfield determining the I/O direction already defines the pin
as either input or output, cannot be none or both at the same time

This issue was reported by Coverity

Coverity-CID: 151970

Change-Id: I18d5387139d6834004ba3269c5b54176bdc97ea7
Signed-off-by: Sergio Rodriguez <sergio.sf.rodriguez@intel.com>
2016-11-17 21:17:13 -05:00
Baohong Liu
3b626c5e15 samples: button: stop the app if device binding fails
Stop the app from running if device binding fails. Otherwise,
dereferencing NULL pointer will happen. This was caught by
Coverity.

Coverity-CID: 151988

Change-Id: I8245d938498a51123249fbd069935900ad660314
Signed-off-by: Baohong Liu <baohong.liu@intel.com>
2016-11-17 21:17:13 -05:00
Flavio Santes
794a47dedf tests/tinycrypt: Fix wrong sizeof argument in test_ccm_mode (2nd)
This commit fixes an issue in the test_ccm_mode.c file:

sizeof(data) is used to compute the length of the array pointed to
by the 'uint8_t *data' pointer.

At the same function scope, there is a variable (dlen) that already
specifies the required length, so we use that variable instead of
the 'sizeof' function call.

This issue was not reported by Coverity, although is worth to fix it.

Change-Id: I27cbf8c7000a4189a42d193f6445996d4b852aa6
Signed-off-by: Flavio Santes <flavio.santes@intel.com>
2016-11-17 21:17:13 -05:00
Flavio Santes
e203a64000 tests/tinycrypt: Fix wrong sizeof argument in test_ccm_mode
This commit fixes the wrong sizeof argument error reported by
Coverity.

Coverity-CID: 152032

Change-Id: I2ee3089b4b840f4a1b8ba0303e92a3311c07ffeb
Signed-off-by: Flavio Santes <flavio.santes@intel.com>
2016-11-17 21:17:13 -05:00
Flavio Santes
25a40b19ea tests/tinycrypt: Fix dead code issue (2nd)
This commit fixes the dead code issue reported by Coverity.

Coverity-CID: 151977

Change-Id: Iaa31c032456f48e1af1d1c9d722f051ac5519ccf
Signed-off-by: Flavio Santes <flavio.santes@intel.com>
2016-11-17 21:17:13 -05:00
Flavio Santes
9366bc161b tests/tinycrypt: Fix dead code issue (1st)
This commit fixes the dead code issue reported by Coverity.

Coverity-CID: 151975

Change-Id: I449341d1f540abe149e8ad9197a64d52cd5722cd
Signed-off-by: Flavio Santes <flavio.santes@intel.com>
2016-11-17 21:17:13 -05:00
Benjamin Walsh
602295a88f dlist: fix SYS_DLIST_FOR_EACH_SAFE when operating on empty list
There was no check to see if the head of a list was empty before trying
to fetch the next node in the list. The fix is added to
sys_dlist_peek_next() so that it also return NULL if the node parameter
is NULL, in addition to being the tail of the list.

Since the value is not used until the second iteration of the loop, and
there will be no second iteration if the list is empty, as long as the
CPU does allow reading at address 0, this was not causing any issues.

Our ARC targets did not seem to like that.

Fixes ZEP-1263 and ZEP-1297.

Change-Id: I07ca16592d206d13662226d1249f487ee78c06aa
Signed-off-by: Benjamin Walsh <walsh.benj@gmail.com>
2016-11-17 21:17:13 -05:00
Flavio Santes
829c3ceb12 tests/tinycrypt: Fix wrong sizeof argument
Fix the issue reported by Coverity: wrong sizeof argument.

Coverity-CID: 152042

Change-Id: I5d593ba54bf8f69f3c9d41a8b2878827d1cc186a
Signed-off-by: Flavio Santes <flavio.santes@intel.com>
2016-11-17 21:17:12 -05:00
Anas Nashif
9f36bbc07a kernel: event_logger: use POST_KERNEL instead of NANOKERNEL
NANOKERNEL is obsolete and this kernel service is still using it causing
deperecaton warnings. Move it to POST_KERNEL

Change-Id: I17fabd080645f93a8599f4ea25da844e1ec5f4bb
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
2016-11-17 21:17:12 -05:00
Anas Nashif
145a4c93fa Revert "build: Handle ALL_LIBS dependencies correctly"
This reverts commit 608abd987c.

This change is breaking build dependencies.

Change-Id: Id8e9dbfc14b72933c402d25847615cddbfaca40d
Jira: ZEP-1291
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
2016-11-17 21:17:12 -05:00
Jithu Joseph
0ce96e850d tests: libs: Fix string overflow
This fixes a string overflow past the end of a buffer
which was reported by coverity.

Coverity-CID: 152044

Change-Id: I5b331135e338fa43b5589a9488b06367e8cad5a7
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
2016-11-17 21:17:12 -05:00
Baohong Liu
27d7a9e29f drivers: cc2520: fix variable type mismatching issue
The variable type mismatching was caught by LLVM.

Jira: ZEP-1179

Change-Id: If26c881d207a6cedc52b7589c5d7ebb2040c7ab7
Signed-off-by: Baohong Liu <baohong.liu@intel.com>
2016-11-17 21:17:12 -05:00
Baohong Liu
391ec95f38 net: ip: fix variable type mismatching issue
The variable type mismatching was caught by LLVM.

Jira: ZEP-1179

Change-Id: I92ca14b7a2c0507a86a6b6abaa567a5091622ad1
Signed-off-by: Baohong Liu <baohong.liu@intel.com>
2016-11-17 21:17:12 -05:00
Baohong Liu
cb409a67fd samples: net: fix a memcmp len error
The memcmp is a comparison between two strings or buffers.
So, the length should be the buffer length, not the length
of the pointer to the buffer. This was caught by LLVM.

Jira: ZEP-1179

Change-Id: I7fd6b199686b19e7f4a2e1288897483e69ad091e
Signed-off-by: Baohong Liu <baohong.liu@intel.com>
2016-11-17 21:17:12 -05:00
Baohong Liu
cf4ce62590 net: 802.15.4: Fix a variable type mismatching issue
This variable type mismatching was caught by LLVM.

Jira: ZEP-1179

Change-Id: I891dc9d55055292e6a749f300e995798040d0b24
Signed-off-by: Baohong Liu <baohong.liu@intel.com>
2016-11-17 21:17:12 -05:00
Jithu Joseph
3188ed4e64 samples :usb : Check return value fix
This commit fixes a missing function return check reported by
Coverity.

Coverity-CID: 151949

Change-Id: Iedf090b7f2ded9f20ff6d796f1cd5c02990b0a4e
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
2016-11-17 21:17:12 -05:00
Sergio Rodriguez
f0523be409 samples: drivers: gpio: Exit from testcase if device not found
Exiting from the test case when the gpio device is not found, this
to avoid a null pointer dereference

This issue was reported by Coverity

Coverity-CID: 151980

Change-Id: I44f13131d44c7c093781e1f11f8481e7ef8175c9
Signed-off-by: Sergio Rodriguez <sergio.sf.rodriguez@intel.com>
2016-11-17 21:17:12 -05:00
Baohong Liu
db089efe48 tests: benchmark: fix a string format issue
A popular issue "format is not a string literally" was
caught by LLVM. Let's make it a string literally.

Jira: ZEP-1179

Change-Id: I2b4a5aef750b772504bf0e6f005dab2ff9ac3e7c
Signed-off-by: Baohong Liu <baohong.liu@intel.com>
2016-11-17 21:17:12 -05:00
Sergio Rodriguez
628ecfa4d7 samples: drivers: i2c fram : Exit from testcase if device not found
Exiting from the test case when the gpio device is not found, this
to avoid a null pointer dereference

This issue was reported by Coverity

Coverity-CID: 151982

Change-Id: Ifaed47b2b48359dacfdb3111ca2895d5912779e6
Signed-off-by: Sergio Rodriguez <sergio.sf.rodriguez@intel.com>
2016-11-17 21:17:12 -05:00
Baohong Liu
f8d3ac0130 drivers: rtc: fix enum type mismatching issue
The enum type mismatching was caught by LLVM.

Jira: ZEP-1179

Change-Id: I50b68e201ef6fb18a02eeda2a2e7548dad3f358c
Signed-off-by: Baohong Liu <baohong.liu@intel.com>
2016-11-17 21:17:12 -05:00
Baohong Liu
853c11885c tests: spi: add return value check
Add function return value check. This was caught by
Coverity.

Coverity-CID: 151950

Change-Id: Iee550e15d124f05f0b0514fdad22d06c617beac2
Signed-off-by: Baohong Liu <baohong.liu@intel.com>
2016-11-17 21:17:12 -05:00
Benjamin Walsh
f1c880aa8a kernel/arm: fix race condition when setting _Swap() return value
There was a possible race condition when setting the return value of a
thread that is pending, from an ISR.

A kernel function causes a thread to pend, with the following series of
steps:

- disable interrupts
- move current thread to wait_q
- call _Swap

Depending if running on M3/4 or M0+, _Swap will either issue a svc #0,
or pend PendSV directly. The same problem exists in both cases.

M3/4:
__svc will:
- enable interrupts
- trigger __pendsv

M0+:
_Swap() will enable interrupts.

__pendsv will:
- save register context including PSP into the thread struct

If an interrupt occurs between interrupts being enabled them and
__pendsv saving PSP, and the ISR sets the pending thread's return value,
this will happen:

- sees the thread in a wait_q
- removes it
- makes it ready
- calls _set_thread_return_value
- _set_thread_return_value looks at the thread's saved PSP to poke
  the value

In this scenario, PSP hasn't yet been updated by __pendsv so it's a
stale value from the previous context switch, resulting in unpredictable
word on the stack getting set to the return value.

There is no way to fix this issue and still have the return value being
delivered directly in the pending thread's exception stack frame, in the
M0+ case. There will always be a window between the unlocking of
interrupts and PendSV being handled. On M3/4, it could be possible with
the mix of SVC and PendSV, since the exception stack frame is created in
the __svc handler. However, because we want to keep the two
implementations as close as possible, and there were talks of moving
M3/4 to using PendSV only, to save an exception, the approach taken
solves both cases.

The approach taken is similar to the ARC and Nios2 ports, where
there is a field in the thread structure that holds the return value.
_Swap() then loads r0/a1 with that value just before returning.

Fixes ZEP-1289.

Change-Id: Iee7e06fe3f8ded84aff918fd43408c7f589344d9
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-11-17 11:03:40 -05:00
Sergio Rodriguez
653328c6e8 samples: aio comparators: Use expected pointer type in printf
The data structure member being used  is character array,
dereferencig this array gives **char instead of the expected
*char type.

This issue was reported by Coverity

Coverity-CID: 152030
Coverity-CID: 152033

Change-Id: Ied67e4b2d47017e6ad5e40b9b6fca1b496c483ed
Signed-off-by: Sergio Rodriguez <sergio.sf.rodriguez@intel.com>
2016-11-17 11:03:40 -05:00
Sergio Rodriguez
d88617db9b drivers: gpio: Remove contradictory if statement evaluation
This fixes an always false evaluation of the gpio I/O direction

This issue was reported by Coverity (CID 150821).

Change-Id: I6c0e9fe405cbd3e35454a81754fa0b1c721691f0
Signed-off-by: Sergio Rodriguez <sergio.sf.rodriguez@intel.com>
2016-11-17 11:03:40 -05:00
Sergio Rodriguez
385770cf21 drivers: gpio_ss: Remove contradictory if statement evaluation
This fixes an always false evaluation of the gpio I/O direction

This issue was reported by Coverity

Coverity-CID: 151833

Change-Id: Ie952d6f50c0383d5631325b69e8e8b234c67c4b8
Signed-off-by: Sergio Rodriguez <sergio.sf.rodriguez@intel.com>
2016-11-17 11:03:40 -05:00
Sergio Rodriguez
cf00c1c184 drivers: gpio_k64: Remove contradictory if statement evaluation
This fixes an always false evaluation of the gpio I/O direction

This issue was reported by Coverity

Coverity-CID: 151834

Change-Id: I033e368b2e91d888f2e8a797490df757513c3906
Signed-off-by: Sergio Rodriguez <sergio.sf.rodriguez@intel.com>
2016-11-17 11:03:40 -05:00
Flavio Santes
6a7b6679b1 sensors/mcp9808: Evaluate sensor_channel_get return code
The sensor_channel_get return code is now evaluated.

Change-Id: Ib931d6caba65af7195bad53c62e6e5a3033b49e8
Signed-off-by: Flavio Santes <flavio.santes@intel.com>
2016-11-17 11:03:40 -05:00
Flavio Santes
3893503f49 sensors/mcp9808: Evaluate sensor_sample_fetch return code
sensor_sample_fetch return code is now evaluated.

Coverity-CID: 151957

Change-Id: I79b9f44c79ac13e8d7da55c9e3866ad504a4a450
Signed-off-by: Flavio Santes <flavio.santes@intel.com>
2016-11-17 11:03:40 -05:00
Luiz Augusto von Dentz
b42719243d Bluetooth: GATT: Fix using out of scope variable
This fixes defect found by coverity: 152027 Pointer to local outside
scope.

Change-Id: I50f196a04363ffa6e6654b71a9a1d89034580413
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>
2016-11-17 16:02:30 +00:00
Allan Stephens
fd45cb4567 kernel: Enhance naming of memory pool configuration options
Replaces confusing (and excessively long) configuration option
names with more intuitive names. Also enhances the description
of each option to clarify its use.

Change-Id: If4d4541407627482b1e90302cfc9df3bc8130d44
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
2016-11-16 18:06:57 -05:00
Allan Stephens
9f0045a30a doc: Incorporate kernel APIs into API documentation guide
Change-Id: Ib5e5aa14534af4789d8247e6096913e09731f5bb
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
2016-11-16 18:06:57 -05:00
Allan Stephens
662c8bee81 doc: Various corrections to Kernel Primer
* Ensures all references to kernel functions are correctly
  tagged so they will auto-link to the API guide.

* Adds references to a few functions and macros that were
  omitted.

Change-Id: I26ccd9c29ea123db2807f2df4d05d574932c6849
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
2016-11-16 18:06:57 -05:00
Allan Stephens
6d0fa01492 doc: Various corrections to doxygen info for Kernel APIs
Most kernel APIs are now ready for inclusion in the API guide.
The APIs largely follow a standard template to provide users
of the API guide with a consistent look-and-feel.

Change-Id: Ib682c31f912e19f5f6d8545d74c5f675b1741058
Signed-off-by: Allan Stephens <allan.stephens@windriver.com>
2016-11-16 18:06:57 -05:00
Iván Briano
afe118fa1d drivers spi_ss: Fix setting of wrong config for SPI 1
Jira: ZEP-1287

Change-Id: I3678631aa5843e769b8e1611734767fa6264b9af
Signed-off-by: Iván Briano <ivan.briano@intel.com>
2016-11-16 14:35:35 +00:00
Vincenzo Frascino
247a2a0671 console: Fix unreachable code condition
This patch fixes an unreachable code condition in the uart_console
driver.

If UART_CONSOLE_DEBUG_SERVER_HOOKS was not defined
handled_by_debug_server in console_out was always 0.

This issue was reported by Coverity (CID 131627).

Change-Id: I4376c3e5b3e68220218df6aabd91b6a8900ca31f
Signed-off-by: Vincenzo Frascino <vincenzo.frascino@linaro.org>
2016-11-16 08:32:25 -05:00
Vincenzo Frascino
cb9033d10f sensor: Fix Unchecked return value in bma280 driver
This patch fixes two "Unchecked return value" conditions into the bma280
driver.

The issue was reported by Coverity (CID 151953).

Change-Id: I2e595b67619411594cec527f358f6c3d3d034550
Signed-off-by: Vincenzo Frascino <vincenzo.frascino@linaro.org>
2016-11-16 08:31:24 -05:00
Sergio Rodriguez
bc7e0455c8 tests: crypto: Fix unchecked return value on CTR PRNG test case
This issue was reported by Coverity (CID 151952)

Change-Id: I59a20a3ccbe606ef634db98ac6cc6889a3973ec3
Signed-off-by: Sergio Rodriguez <sergio.sf.rodriguez@intel.com>
2016-11-16 07:39:31 -05:00
Sergio Rodriguez
71e85e390b drivers: pwm: Fix uninitialized pointer
This fixes an uninitialized pointer being pass and evaluated by
a subsequent function

This issue was reported by Coverity (CID 150824)

Change-Id: If1f636a44cc675b56e426b1de85895b74ba7105e
Signed-off-by: Sergio Rodriguez <sergio.sf.rodriguez@intel.com>
2016-11-16 07:39:31 -05:00
Johan Hedberg
eee3a430dc Bluetooth: Use convenience macros for timeout durations
Using the K_* macros makes it easier to read what exactly the various
timeouts are.

Change-Id: Ia405d3760b8e600af7e33a7221ef6ec717708973
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
2016-11-16 07:11:21 -05:00
Vinayak Chettimada
32ef1480e9 Bluetooth: Controller: Remove unused util functions
Change-id: I7b691d082d080239c35b63221e3c6c7aa93ed58e
Signed-off-by: Vinayak Chettimada <vinayak.kariappa.chettimada@nordicsemi.no>
2016-11-16 07:10:51 -05:00
Vinayak Chettimada
39410d79fe Bluetooth: Controller: Fix incorrect irq priority check
External interrupts are indexed from value 16, wherein
0 to 15 are ARM cortex M exceptions. Fixed code in
_irq_is_priority_equal to fetch correct external
interrupt line ISR priority.

Change-id: I9cfd411480e78dfc9635e72d14df9d667a9d8400
Signed-off-by: Vinayak Chettimada <vinayak.kariappa.chettimada@nordicsemi.no>
2016-11-16 07:05:58 -05:00
Ramesh Thomas
40c944f0b6 tests: power_states: Update testcase.ini to include arc
testcase.ini was not building for ARC. This app would
run on x86 and arc.

Change-Id: I961d56079aa1db7d84e0fcc87780ba11d7f4d831
Signed-off-by: Ramesh Thomas <ramesh.thomas@intel.com>
2016-11-15 21:23:59 -05:00
Ramesh Thomas
efcdfce517 samples: power_mgmt: Remove platform filtering of testcases
Remove redundant platform filtering and only use SOC filtering

Change-Id: Ib823e076a874ce61a235eca63eebb7f19d2fdd30
Signed-off-by: Ramesh Thomas <ramesh.thomas@intel.com>
2016-11-15 21:23:59 -05:00
Baohong Liu
e50f05df3e samples: button: fix variable type mismatching issue
The variable type mismatching was caught by LLVM.

Jira: ZEP-1179

Change-Id: I084406601badc64c257cbdd82b9c8b7509549303
Signed-off-by: Baohong Liu <baohong.liu@intel.com>
2016-11-15 21:09:46 -05:00
Baohong Liu
3f0fcedf00 drivers: bmi160: fix variable type mismatching issue
The variable type mismatching was caught by LLVM.

Jira: ZEP-1179

Change-Id: I1193a946ea5814510e6c07668c5d05a5d91445a8
Signed-off-by: Baohong Liu <baohong.liu@intel.com>
2016-11-15 21:08:44 -05:00
Baohong Liu
baab38500c samples: usb: fix variable type mismatching issue
The variable type mismatching was caught by LLVM.

Jira: ZEP-1179

Change-Id: I402c348af142342e37e93619c4da6e3a5bfd82da
Signed-off-by: Baohong Liu <baohong.liu@intel.com>
2016-11-15 20:38:34 -05:00
Sergey Kiselev
06869d4499 sensors: bme280: fix typo in reading trimming parameters
Change-Id: I32e72c2845bd06b10585ac8048f67ac754c2a6d6
Signed-off-by: Sergey Kiselev <sergey.kiselev@intel.com>
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
2016-11-15 20:38:17 -05:00
Vincenzo Frascino
de55b9f73a sensor: Fix Unchecked return value issues in bme280 driver
This patch fixes unchecked return value conditions in the bme280
driver.

This issue was reported by Coverity (CID 151961, 151959, 151955).

Change-Id: I3a2dfbabd41ae52b00fa512a40e00c2e36c3b5ca
Signed-off-by: Vincenzo Frascino <vincenzo.frascino@linaro.org>
2016-11-15 19:43:09 -05:00
Vincenzo Frascino
eb7910b9d1 sensor: Fix less-than-zero comparison in bmi160 driver
This patch fixes a less-than-zero comparison of an unsigned value
condition present in bmi160 driver.

This issue was reported by Coverity (CID 152002, 152003).

Change-Id: I703066519652ac1ecdd9ddf7e97ec7dcbe2a9e27
Signed-off-by: Vincenzo Frascino <vincenzo.frascino@linaro.org>
2016-11-15 19:43:09 -05:00
Ramesh Thomas
95f8f6f3e0 samples: power_mgmt: Remove redundant sample power_hooks
This sample was created intially when there were no other
samples in place to enable the CONFIG flags to build code
inside those flags. However, those CONFIG flags are now
guarded with corresponding "SUPPORTED" flags which are
enabled based in Kconfigs of socs based on their support
for that power feature. This app is for x86 and those
features will not get enabled for this configuration. If
it is still required, then we would need to fake such
support in Kconfig.board of qemu_x86. Removing it, because
those flags will get enabled by sample and test apps of
socs that support the power features, causing code inside
them to get built.

Change-Id: I647be9289a49d69880811abee499a4efd61bbc6a
Signed-off-by: Ramesh Thomas <ramesh.thomas@intel.com>
2016-11-15 19:43:09 -05:00
Ramesh Thomas
b65e208171 samples: power_mgmt: Cleanup and update with new pm interface
Cleaned up and removed some unnecessary code to avoid
distraction from main sample implementation. Updated some
logic based on new PM interface in soc area. Updated README
to indicate it supports x86 and ARC and updated sample
output of both architectures.

Change-Id: I1c9c8348dae403b7ca6fe17ab867e3fbef06ae60
Signed-off-by: Ramesh Thomas <ramesh.thomas@intel.com>
2016-11-15 19:43:09 -05:00
Baohong Liu
2ccf0ad045 tests: spi_test: fix variable type mismatching issue
The variable type mismatching was caught by LLVM.

Jira: ZEP-1179

Change-Id: I37934ef2ee47c521a78086564876843794688d55
Signed-off-by: Baohong Liu <baohong.liu@intel.com>
2016-11-15 19:43:09 -05:00
Benjamin Walsh
efc7ffde75 kernel/arm: fix missing interrupt lock around _is_next_thread_current()
This reverts commit

	"kernel/arm: add comment about _is_next_thread_current"

and fixes the interrupt locking issue.

The comment would have been right if only reads were done the ready
queue, but that is not the case. It turns out that the comment was written
ignoring the fact that _is_next_thread_current() updates the next thread
cache when fetching the next thread.

Change-Id: I21c9230f85f4f87a6bbf14fd4a9eb7e19b59f8c5
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
2016-11-15 19:17:06 -05:00
Szymon Janc
bee0fd0601 Bluetooth: Fix address type use for passive scanning
This fix using incorrect address type for passive scanning with
privacy enabled. Controller was not reporting directed advertising
to RPA address due to public type being used for passive scan.

This was affecting TC_CONN_GCEP_BV_01_C, TC_CONN_ACEP_BV_01_C and
TC_CONN_DCEP_BV_01_C qualification test cases.

Jira: ZEP-1200

Change-Id: Icc316441fcac1a72d75f9ade27a99030efc846b9
Signed-off-by: Szymon Janc <ext.szymon.janc@tieto.com>
2016-11-15 19:12:07 -05:00
Mariusz Skmara
c6c73c1b2e Bluetooth: Fix not sending L2CAP Connection Parameters Update Request
This fixes issue that L2CAP Connection Parameters Update Request was
not sent. There was check that used LE features of host controller
to determine if L2CAP procedure or LL shall be used. It was failing
with 4.2 controller. The check shall test if remote supports
LL Connection Parameters Request Procedure. If it's not supported,
then L2CAP Connection Parameters Update Procedure will be used.

Closes ZEP-1220

1/4   L2CAP   TC_LE_CPU_BV_01_C      PASS
2/4   GAP     TC_CONN_CPUP_BV_01_C   PASS
3/4   GAP     TC_CONN_CPUP_BV_02_C   PASS
4/4   GAP     TC_CONN_CPUP_BV_03_C   PASS

Change-Id: I61ad544d9568ca6306a845e05c1a2e28d1693ab4
Signed-off-by: Mariusz Skamra <mariusz.skamra@tieto.com>
2016-11-15 19:11:21 -05:00
Vinayak Chettimada
ae495b7618 Bluetooth: Controller: Fix incorrect auto variable init
Coverity analysis discovered NULL pointer being
dereferenced when passing a auto variable. The variable is
now correctly assigned with address of a valid default
value variable. As per design, the dereferencing will not
happen as the master role does not use the passed parameter
only slave role uses it to prepare the connection parameter
request PDU.

Change-id: I3f8519b23a83cb8c50c7fba81810eff7737ff74a
Signed-off-by: Vinayak Chettimada <vinayak.kariappa.chettimada@nordicsemi.no>
2016-11-15 19:11:21 -05:00
Vinayak Chettimada
3b22494192 Bluetooth: Controller: Fix observer filter_policy field size
Coverity analysis discovered that observer filter policy
field was 1 bit, whereas valid range for extended scanner
filter policy feature implemented in controller is 0 to 3.
Increase the bit field size from 1 to 2.

Change-Id: Id4b2e354961dfb3b45f72fa4e0ab18de7425bbb5
Signed-off-by: Vinayak Chettimada <vinayak.kariappa.chettimada@nordicsemi.no>
2016-11-15 19:11:21 -05:00
Vinayak Chettimada
c4a5b9c74e Bluetooth: Controller: Fix HCI Reset Command implementation
Added implementation for HCI Reset Command. Implementation
gracefully disables any running advertiser, observer, and/
or connection roles, and it resets controller context members.
The HCI Reset Command is implemented in such a way that
driver instances shared with other sub-systems and
application is not disturbed and instance/references used
by Bluetooth Controller are gracefully returned back.

Jira: ZEP-1282

Change-id: Ifb9ae6807736b5ec2d9f346cf2a590322056bcee
Signed-off-by: Vinayak Chettimada <vinayak.kariappa.chettimada@nordicsemi.no>
Signed-off-by: Carles Cufi <carles.cufi@nordicsemi.no>
2016-11-15 19:11:21 -05:00
Szymon Janc
ca23390e84 Bluetooth: Kconfig: Fix BR/EDR dependencies
BR/EDR code should have minimal impact on LE code so to keep it simple
just require peripheral and central to be enabled when selecting BR/EDR
support.

Fix following Kconfig warning:

warning: (NETWORKING_WITH_BT && BLUETOOTH_BREDR) selects
    BLUETOOTH_L2CAP_DYNAMIC_CHANNEL which has unmet direct dependencies
    (BLUETOOTH && BLUETOOTH_HCI && BLUETOOTH_HCI_HOST && BLUETOOTH_CONN
    && BLUETOOTH_SMP)

Change-Id: I7f7cb8794def0df6daaa4abfe4596df460f1a2b2
Signed-off-by: Szymon Janc <ext.szymon.janc@tieto.com>
2016-11-15 13:28:21 -05:00
Gil Pitney
2644d370c8 cc3200: Remove CPU_HAS_FPU from cc3200 Kconfig.soc
Though Cortex-M4 could optionally have a floating point unit,
the MCU in the cc3200 in fact does not have an FPU.

Enabling CPU_HAS_FPU caused applications built with CONFIG_FLOAT=y
to crash during an early call to enable_floating_point().

This patch was validated by running microPython, which is one
such application.

Change-Id: I8bfd42c456524e152cbbb983001d9540d93fbe98
Signed-off-by: Gil Pitney <gil.pitney@linaro.org>
2016-11-15 16:43:32 +00:00
Benjamin Walsh
bef829cfc0 kernel: remove last instances of tNANO in comments
Change-Id: I3d533b819422d4b754afb81d3ea67c03bc7f5630
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
(cherry picked from commit 59a382e339)
2016-11-15 01:30:21 +00:00
Benjamin Walsh
c625aa2636 kernel/arm: add comment about _is_next_thread_current
Normally, _is_next_thread_current() must be called with interrupts
locked, but the ARM interrupt exit code does not have to do that. Add
explanation why.

Change-Id: Id383b47a055fdd6fbd5afffa52772e92febde98f
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
(cherry picked from commit dfa7ca4ee5)
2016-11-15 01:30:13 +00:00
Benjamin Walsh
4329e5e24a kernel: fix typo in comment
Change-Id: I1919fd7b0ae3cb3ac434acc2dceddf3afb4a975b
Signed-off-by: Benjamin Walsh <benjamin.walsh@windriver.com>
(cherry picked from commit ba26678fc6)
2016-11-15 01:30:06 +00:00
Inaky Perez-Gonzalez
4e4ac94f90 tests/drivers/adc: move to ztest to actually test
This TC is only exercising the API, as we don't have a feedback loop
mechanism to verify whichever values are fed to the ADC.

Fixed the loop to complete after 10 runs; on each run, print the
values and actually report the difference between them. With no inputs
connected (aka: floating), they should be reporting noise relatively
close to the previous reading, so we might want to use this delta as a
testing pattern (assert if the delta is higher than some value, but
I've seen variations as high as 40M units). For now, the test is just
happy with being able to read them.

The buffer has been re-typed to uint32_t so we can iterate over it
without casting tricks -- it requires then only a single cast when
initializing sample.buffer (which shall be a void* anyway).

Duplicated the buffer, so we can flip/flop between two buffers to
compare against the entries read in the previous run.

v4: fixed missed warnings

Change-Id: If6b48b92231007202d74f5c042f6d0cf3fdcb60a
Signed-off-by: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
(cherry picked from commit 9e1df6f21f)
2016-11-15 00:47:57 +00:00
Szymon Janc
8d10dc63fc Bluetooth: Fix use of deprecated PRIMARY init level
Fix following warning:

  CC      subsys/bluetooth/host/monitor.o
In file included from zephyr/include/drivers/loapic.h:58:0,
                 from zephyr/include/drivers/ioapic.h:22,
                 from zephyr/include/drivers/sysapic.h:20,
                 from zephyr/include/arch/x86/irq_controller.h:33,
                 from zephyr/include/arch/x86/arch.h:28,
                 from zephyr/include/arch/cpu.h:23,
                 from zephyr/include/kernel.h:2458,
                 from zephyr/include/zephyr.h:20,
                 from zephyr/subsys/bluetooth/host/monitor.c:24:
zephyr/subsys/bluetooth/host/monitor.c: In function
    '_deprecation_check_sys_init_bt_monitor_init0':
zephyr/include/device.h:130:16: warning: '_INIT_LEVEL_PRIMARY' is
    deprecated [-Wdeprecated-declarations]
  static struct device_config _CONCAT(__config_, dev_name) __used \
                ^
zephyr/include/device.h:245:2: note: in expansion of macro
    'DEVICE_AND_API_INIT'
  DEVICE_AND_API_INIT(dev_name, drv_name, init_fn, data, cfg_info, \
  ^
zephyr/include/init.h:69:2: note: in expansion of macro 'DEVICE_INIT'
  DEVICE_INIT(_SYS_NAME(init_fn), "", init_fn, NULL, NULL, level, prio)
  ^
zephyr/subsys/bluetooth/host/monitor.c:193:1: note: in expansion of
    macro 'SYS_INIT'
 SYS_INIT(bt_monitor_init, PRIMARY, MONITOR_INIT_PRIORITY);
 ^
zephyr/include/device.h:48:31: note: declared here
 static __deprecated const int _INIT_LEVEL_PRIMARY = 1;

Change-Id: I0960bfddddfd1105daf3bb8cc1114e9a25840f2c
Signed-off-by: Szymon Janc <ext.szymon.janc@tieto.com>
2016-11-15 00:46:23 +00:00
Johan Hedberg
ce596d3c54 Bluetooth: doc: Fix reference to documentation location
The Bluetooth documentation is found in doc/subsystems/bluetooth and
not in doc/bluetooth.

Change-Id: I7e7010b5ae4a26ea552d75f1a095baec18d02630
Signed-off-by: Johan Hedberg <johan.hedberg@intel.com>
2016-11-14 10:49:53 +02:00
Anas Nashif
61b596b0e5 Zephyr 1.6.0-rc1
First release candidate of 1.6.

Change-Id: I01e8524c163f7e984405e2feecc0118db1605ab8
Signed-off-by: Anas Nashif <anas.nashif@intel.com>
2016-11-13 10:16:28 -05:00
16156 changed files with 2289955 additions and 1560263 deletions

View File

@@ -1,20 +1,22 @@
--mailback
--no-tree
--emacs
--summary-file
--show-types
--max-line-length=80
--min-conf-desc-length=1
--typedefsfile=scripts/checkpatch/typedefsfile
--ignore BRACES
--ignore PRINTK_WITHOUT_KERN_LEVEL
--ignore SPLIT_STRING
--ignore VOLATILE
--ignore CONFIG_EXPERIMENTAL
--ignore PREFER_KERNEL_TYPES
--ignore AVOID_EXTERNS
--ignore NETWORKING_BLOCK_COMMENT_STYLE
--ignore DATE_TIME
--ignore MINMAX
--ignore CONST_STRUCT
--ignore FILE_PATH_CHANGES
--ignore SPDX_LICENSE_TAG
--exclude ext
--exclude net/ip/contiki

View File

@@ -1,145 +0,0 @@
# SPDX-License-Identifier: GPL-2.0
#
# clang-format configuration file. Intended for clang-format >= 4.
#
# For more information, see:
#
# Documentation/process/clang-format.rst
# https://clang.llvm.org/docs/ClangFormat.html
# https://clang.llvm.org/docs/ClangFormatStyleOptions.html
#
---
AccessModifierOffset: -4
AlignAfterOpenBracket: Align
AlignConsecutiveAssignments: false
AlignConsecutiveDeclarations: false
#AlignEscapedNewlines: Left # Unknown to clang-format-4.0
AlignOperands: true
AlignTrailingComments: false
AllowAllParametersOfDeclarationOnNextLine: false
AllowShortBlocksOnASingleLine: false
AllowShortCaseLabelsOnASingleLine: false
AllowShortFunctionsOnASingleLine: None
AllowShortIfStatementsOnASingleLine: false
AllowShortLoopsOnASingleLine: false
AlwaysBreakAfterDefinitionReturnType: None
AlwaysBreakAfterReturnType: None
AlwaysBreakBeforeMultilineStrings: false
AlwaysBreakTemplateDeclarations: false
BinPackArguments: true
BinPackParameters: true
BraceWrapping:
AfterClass: false
AfterControlStatement: false
AfterEnum: false
AfterFunction: true
AfterNamespace: true
AfterObjCDeclaration: false
AfterStruct: false
AfterUnion: false
#AfterExternBlock: false # Unknown to clang-format-5.0
BeforeCatch: false
BeforeElse: false
IndentBraces: false
#SplitEmptyFunction: true # Unknown to clang-format-4.0
#SplitEmptyRecord: true # Unknown to clang-format-4.0
#SplitEmptyNamespace: true # Unknown to clang-format-4.0
BreakBeforeBinaryOperators: None
BreakBeforeBraces: Custom
#BreakBeforeInheritanceComma: false # Unknown to clang-format-4.0
BreakBeforeTernaryOperators: false
BreakConstructorInitializersBeforeComma: false
#BreakConstructorInitializers: BeforeComma # Unknown to clang-format-4.0
BreakAfterJavaFieldAnnotations: false
BreakStringLiterals: false
ColumnLimit: 80
CommentPragmas: '^ IWYU pragma:'
#CompactNamespaces: false # Unknown to clang-format-4.0
ConstructorInitializerAllOnOneLineOrOnePerLine: false
ConstructorInitializerIndentWidth: 8
ContinuationIndentWidth: 8
Cpp11BracedListStyle: false
DerivePointerAlignment: false
DisableFormat: false
ExperimentalAutoDetectBinPacking: false
#FixNamespaceComments: false # Unknown to clang-format-4.0
# Taken from:
# git grep -h '^#define [^[:space:]]*FOR_EACH[^[:space:]]*(' include/ \
# | sed "s,^#define \([^[:space:]]*FOR_EACH[^[:space:]]*\)(.*$, - '\1'," \
# | sort | uniq
ForEachMacros:
- 'FOR_EACH'
- 'FOR_EACH_FIXED_ARG'
- 'RB_FOR_EACH'
- 'RB_FOR_EACH_CONTAINER'
- 'SYS_DLIST_FOR_EACH_CONTAINER'
- 'SYS_DLIST_FOR_EACH_CONTAINER_SAFE'
- 'SYS_DLIST_FOR_EACH_NODE'
- 'SYS_DLIST_FOR_EACH_NODE_SAFE'
- 'SYS_SFLIST_FOR_EACH_CONTAINER'
- 'SYS_SFLIST_FOR_EACH_CONTAINER_SAFE'
- 'SYS_SFLIST_FOR_EACH_NODE'
- 'SYS_SFLIST_FOR_EACH_NODE_SAFE'
- 'SYS_SLIST_FOR_EACH_CONTAINER'
- 'SYS_SLIST_FOR_EACH_CONTAINER_SAFE'
- 'SYS_SLIST_FOR_EACH_NODE'
- 'SYS_SLIST_FOR_EACH_NODE_SAFE'
- 'Z_GENLIST_FOR_EACH_CONTAINER'
- 'Z_GENLIST_FOR_EACH_CONTAINER_SAFE'
- 'Z_GENLIST_FOR_EACH_NODE'
- 'Z_GENLIST_FOR_EACH_NODE_SAFE'
- '_WAIT_Q_FOR_EACH'
#IncludeBlocks: Preserve # Unknown to clang-format-5.0
IncludeCategories:
- Regex: '.*'
Priority: 1
IncludeIsMainRegex: '(Test)?$'
IndentCaseLabels: false
#IndentPPDirectives: None # Unknown to clang-format-5.0
IndentWidth: 8
IndentWrappedFunctionNames: false
JavaScriptQuotes: Leave
JavaScriptWrapImports: true
KeepEmptyLinesAtTheStartOfBlocks: false
MacroBlockBegin: ''
MacroBlockEnd: ''
MaxEmptyLinesToKeep: 1
NamespaceIndentation: Inner
#ObjCBinPackProtocolList: Auto # Unknown to clang-format-5.0
ObjCBlockIndentWidth: 8
ObjCSpaceAfterProperty: true
ObjCSpaceBeforeProtocolList: true
# Taken from git's rules
#PenaltyBreakAssignment: 10 # Unknown to clang-format-4.0
PenaltyBreakBeforeFirstCallParameter: 30
PenaltyBreakComment: 10
PenaltyBreakFirstLessLess: 0
PenaltyBreakString: 10
PenaltyExcessCharacter: 100
PenaltyReturnTypeOnItsOwnLine: 60
PointerAlignment: Right
ReflowComments: false
SortIncludes: false
#SortUsingDeclarations: false # Unknown to clang-format-4.0
SpaceAfterCStyleCast: false
SpaceAfterTemplateKeyword: true
SpaceBeforeAssignmentOperators: true
#SpaceBeforeCtorInitializerColon: true # Unknown to clang-format-5.0
#SpaceBeforeInheritanceColon: true # Unknown to clang-format-5.0
SpaceBeforeParens: ControlStatements
#SpaceBeforeRangeBasedForLoopColon: true # Unknown to clang-format-5.0
SpaceInEmptyParentheses: false
SpacesBeforeTrailingComments: 1
SpacesInAngles: false
SpacesInContainerLiterals: false
SpacesInCStyleCastParentheses: false
SpacesInParentheses: false
SpacesInSquareBrackets: false
Standard: Cpp03
TabWidth: 8
UseTab: Always
...

View File

@@ -1,31 +0,0 @@
codecov:
notify:
require_ci_to_pass: yes
coverage:
precision: 2
round: down
range: "70...100"
status:
project: yes
patch: yes
changes: no
#ignore:
# - "tests/**/*"
# - "samples/**/*"
# - "ext/hal/**/*"
parsers:
gcov:
branch_detection:
conditional: yes
loop: yes
method: no
macro: no
comment:
layout: "reach, diff, flags, files, footer"
behavior: default
require_changes: no

View File

@@ -1,67 +0,0 @@
# EditorConfig: https://editorconfig.org/
# top-most EditorConfig file
root = true
# All (Defaults)
[*]
charset = utf-8
end_of_line = lf
insert_final_newline = true
trim_trailing_whitespace = true
max_line_length = 80
# Assembly
[*.S]
indent_style = tab
indent_size = 8
# C
[*.{c,h}]
indent_style = tab
indent_size = 8
# Python
[*.py]
indent_style = space
indent_size = 4
# Perl
[*.pl]
indent_style = tab
indent_size = 8
# YAML
[*.{yml,yaml}]
indent_style = space
indent_size = 2
# Shell Script
[*.sh]
indent_style = space
indent_size = 4
# Windows Command Script
[*.cmd]
end_of_line = crlf
indent_style = tab
indent_size = 8
# Valgrind Suppression File
[*.supp]
indent_style = space
indent_size = 3
# CMake
[{CMakeLists.txt,*.cmake}]
indent_style = space
indent_size = 2
# Makefile
[Makefile]
indent_style = tab
# Device tree
[*.{dts,dtsi,overlay}]
indent_style = tab
indent_size = 8

4
.gitattributes vendored
View File

@@ -3,7 +3,3 @@
.gitattributes export-ignore
.gitignore export-ignore
.mailmap export-ignore
# Tell linguist that generated test pattern files should not be included in the
# language statistics.
*.pat linguist-generated=true

View File

@@ -1,39 +0,0 @@
---
name: Bug report
about: Create a report to help us improve Zephyr
title: ''
labels: bug
assignees: ''
---
**Describe the bug**
A clear and concise description of what the bug is.
What have you tried to diagnose or workaround this issue?
**To Reproduce**
Steps to reproduce the behavior:
1. mkdir build; cd build
2. cmake -DBOARD=board\_xyz
3. make
4. See error
**Expected behavior**
A clear and concise description of what you expected to happen.
**Impact**
What impact does this issue have on your progress (e.g., annoyance, showstopper)
**Screenshots or console output**
If applicable, add a screenshot (drag-and-drop an image), or console logs
(cut-and-paste text and put a code fence (\`\`\`) before and after, to help
explain the issue.
**Environment (please complete the following information):**
- OS: (e.g. Linux, MacOS, Windows)
- Toolchain (e.g Zephyr SDK, ...)
- Commit SHA or Version used
**Additional context**
Add any other context about the problem here.

View File

@@ -1,20 +0,0 @@
---
name: Enhancement
about: Suggest enhancements to existing features
title: ''
labels: Enhancement
assignees: ''
---
**Is your enhancement proposal related to a problem? Please describe.**
A clear and concise description of what the problem is.
**Describe the solution you'd like**
A clear and concise description of what you want to happen.
**Describe alternatives you've considered**
A clear and concise description of any alternative solutions or features you've considered.
**Additional context**
Add any other context or graphics (drag-and-drop an image) about the feature request here.

View File

@@ -1,20 +0,0 @@
---
name: Feature request
about: Suggest an idea for this project
title: ''
labels: feature request
assignees: ''
---
**Is your feature request related to a problem? Please describe.**
A clear and concise description of what the problem is.
**Describe the solution you'd like**
A clear and concise description of what you want to happen.
**Describe alternatives you've considered**
A clear and concise description of any alternative solutions or features you've considered.
**Additional context**
Add any other context or graphics (drag-and-drop an image) about the feature request here.

View File

@@ -1,51 +0,0 @@
---
name: RFC / Proposal
about: Submit an RFC / Proposal
title: ''
labels: RFC
assignees: ''
---
## Introduction
This section targets end users, TSC members, maintainers and anyone else that might
need a quick explanation of your proposed change.
### Problem description
Why do we want this change and what problem are we trying to address?
### Proposed change
A brief summary of the proposed change - the 10,000 ft view on what it will
change once this change is implemented.
## Detailed RFC
In this section of the document the target audience is the dev team. Upon
reading this section each engineer should have a rather clear picture of what
needs to be done in order to implement the described feature.
### Proposed change (Detailed)
This section is freeform - you should describe your change in as much detail
as possible. Please also ensure to include any context or background info here.
For example, do we have existing components which can be reused or altered.
By reading this section, each team member should be able to know what exactly
you're planning to change and how.
### Dependencies
Highlight how the change may affect the rest of the project (new components,
modifications in other areas), or other teams/projects.
### Concerns and Unresolved Questions
List any concerns, unknowns, and generally unresolved questions etc.
## Alternatives
List any alternatives considered, and the reasons for choosing this option
over them.

View File

@@ -1,16 +0,0 @@
license:
main: apache-2.0
report_missing: true
category: Permissive
copyright:
check: true
exclude:
extensions:
- yml
- yaml
- html
- rst
- conf
- cfg
langs:
- HTML

View File

@@ -1,65 +0,0 @@
# Copyright (c) 2020 Linaro Limited.
# SPDX-License-Identifier: Apache-2.0
name: Documentation GitHub Workflow
on: [pull_request]
jobs:
build:
runs-on: ubuntu-latest
steps:
- name: Update PATH for west
run: |
echo "::add-path::$HOME/.local/bin"
- name: checkout
uses: actions/checkout@v2
- name: install-pkgs
run: |
sudo apt-get install -y ninja-build doxygen
- name: cache-pip
uses: actions/cache@v1
with:
path: ~/.cache/pip
key: ${{ runner.os }}-doc-pip
- name: install-pip
run: |
pip3 install setuptools
pip3 install 'breathe>=4.9.1,<4.15.0' 'docutils>=0.14' \
'sphinx>=1.7.5,<3.0' sphinx_rtd_theme sphinx-tabs \
sphinxcontrib-svg2pdfconverter 'west>=0.6.2'
pip3 install pyelftools canopen progress
- name: west setup
run: |
west init -l . || true
- name: build-docs
run: |
source zephyr-env.sh
make htmldocs
tar cvf htmldocs.tar --directory=./doc/_build html
- name: upload-build
uses: actions/upload-artifact@master
continue-on-error: True
with:
name: htmldocs.tar
path: htmldocs.tar
- name: check-warns
run: |
if [ -s doc/_build/doc.warnings ]; then
docwarn=$(cat doc/_build/doc.warnings)
docwarn="${docwarn//'%'/'%25'}"
docwarn="${docwarn//$'\n'/'%0A'}"
docwarn="${docwarn//$'\r'/'%0D'}"
# We treat doc warnings as errors
echo "::error file=doc.warnings::$docwarn"
exit 1
fi

View File

@@ -1,112 +0,0 @@
# Copyright (c) 2020 Linaro Limited.
# SPDX-License-Identifier: Apache-2.0
name: Doc build for Release or Daily
# Either a daily based on schedule/cron or only on tag push
on:
schedule:
- cron: '50 22 * * *'
push:
tags:
- '*'
jobs:
build:
runs-on: ubuntu-latest
steps:
- name: Update PATH for west
run: |
echo "::add-path::$HOME/.local/bin"
- name: Determine tag
id: tag
run: |
# We expect to get here either due to a schedule event in which
# case we are doing a daily build of the docs, or because a new
# tag was pushed, in which case we are building docs for a release
if [ ${GITHUB_EVENT_NAME} == "schedule" ]; then
echo ::set-output name=TYPE::daily;
echo ::set-output name=RELEASE::latest;
elif [ ${GITHUB_EVENT_NAME} == "push" ]; then
# If push due to a tag GITHUB_REF will look like refs/tags/TAG-FOO
# chop of 'refs/tags' so RELEASE=TAG-FOO
echo ::set-output name=TYPE::release;
echo ::set-output name=RELEASE::${GITHUB_REF/refs\/tags\//};
else
exit 1
fi
- name: Configure AWS Credentials
uses: aws-actions/configure-aws-credentials@v1
with:
aws-access-key-id: ${{ secrets.AWS_ACCESS_KEY_ID }}
aws-secret-access-key: ${{ secrets.AWS_SECRET_ACCESS_KEY }}
aws-region: us-east-1
- name: checkout
uses: actions/checkout@v2
- name: install-pkgs
run: |
sudo apt-get install -y ninja-build doxygen
- name: cache-pip
uses: actions/cache@v1
with:
path: ~/.cache/pip
key: ${{ runner.os }}-doc-pip
- name: install-pip
run: |
pip3 install setuptools
pip3 install 'breathe>=4.9.1,<4.15.0' 'docutils>=0.14' \
'sphinx>=1.7.5,<3.0' sphinx_rtd_theme sphinx-tabs \
sphinxcontrib-svg2pdfconverter 'west>=0.6.2'
pip3 install pyelftools
- name: west setup
run: |
west init -l . || true
- name: build-docs
env:
DOC_TAG: ${{ steps.tag.outputs.TYPE }}
run: |
source zephyr-env.sh
make DOC_TAG=${DOC_TAG} htmldocs
- name: check-warns
run: |
if [ -s doc/_build/doc.warnings ]; then
docwarn=$(cat doc/_build/doc.warnings)
docwarn="${docwarn//'%'/'%25'}"
docwarn="${docwarn//$'\n'/'%0A'}"
docwarn="${docwarn//$'\r'/'%0D'}"
# We treat doc warnings as errors
echo "::error file=doc.warnings::$docwarn"
exit 1
fi
- name: Upload to AWS S3
env:
RELEASE: ${{ steps.tag.outputs.RELEASE }}
run: |
echo "DOC_RELEASE=[$RELEASE]"
if [ "$RELEASE" == "latest" ]; then
export
echo "publish latest docs"
aws s3 sync --quiet doc/_build/html s3://docs.zephyrproject.org/latest --delete
echo "success sync of latest docs"
else
DOC_RELEASE=${RELEASE}.0
echo "publish release docs: ${DOC_RELEASE}"
aws s3 sync --quiet doc/_build/html s3://docs.zephyrproject.org/${DOC_RELEASE}
echo "success sync of rel docs"
fi
if [ -d doc/_build/doxygen/html ]; then
echo "publish doxygen"
aws s3 sync --quiet doc/_build/doxygen/html s3://docs.zephyrproject.org/apidoc/${RELEASE} --delete
echo "success publish of doxygen"
fi

View File

@@ -1,32 +0,0 @@
name: Scancode
on: [pull_request]
jobs:
scancode_job:
runs-on: ubuntu-latest
name: Scan code for licenses
steps:
- name: Checkout the code
uses: actions/checkout@v1
- name: Scan the code
id: scancode
uses: zephyrproject-rtos/action_scancode@v3
with:
directory-to-scan: 'scan/'
- name: Artifact Upload
uses: actions/upload-artifact@v1
with:
name: scancode
path: ./artifacts
- name: Verify
run: |
if [ -s ./artifacts/report.txt ]; then
report=$(cat ./artifacts/report.txt)
report="${report//'%'/'%25'}"
report="${report//$'\n'/'%0A'}"
report="${report//$'\r'/'%0D'}"
echo "::error file=./artifacts/report.txt::$report"
exit 1
fi

View File

@@ -1,65 +0,0 @@
# Copyright (c) 2020 Linaro Limited.
# SPDX-License-Identifier: Apache-2.0
name: Zephyr West Command Tests
on:
push:
paths:
- 'scripts/west-commands.yml'
- 'scripts/west_commands/**'
pull_request:
paths:
- 'scripts/west-commands.yml'
- 'scripts/west_commands/**'
jobs:
build:
runs-on: ${{ matrix.os }}
strategy:
matrix:
python-version: [3.6, 3.7, 3.8]
os: [ubuntu-latest, macos-latest, windows-latest]
steps:
- name: checkout
uses: actions/checkout@v2
- name: Set up Python ${{ matrix.python-version }}
uses: actions/setup-python@v1
with:
python-version: ${{ matrix.python-version }}
- name: cache-pip-linux
if: startsWith(runner.os, 'Linux')
uses: actions/cache@v1
with:
path: ~/.cache/pip
key: ${{ runner.os }}-pip-${{ matrix.python-version }}
restore-keys: |
${{ runner.os }}-pip-${{ matrix.python-version }}
- name: cache-pip-mac
if: startsWith(runner.os, 'macOS')
uses: actions/cache@v1
with:
path: ~/Library/Caches/pip
# Trailing '-' was just to get a different cache name
key: ${{ runner.os }}-pip-${{ matrix.python-version }}-
restore-keys: |
${{ runner.os }}-pip-${{ matrix.python-version }}-
- name: cache-pip-win
if: startsWith(runner.os, 'Windows')
uses: actions/cache@v1
with:
path: ~\AppData\Local\pip\Cache
key: ${{ runner.os }}-pip-${{ matrix.python-version }}
restore-keys: |
${{ runner.os }}-pip-${{ matrix.python-version }}
- name: install pytest
run: |
pip3 install pytest west pyelftools canopen progress
- name: run pytest-win
if: runner.os == 'Windows'
run: |
cmd /C "set PYTHONPATH=.\scripts\west_commands && pytest ./scripts/west_commands/tests/"
- name: run pytest-mac-linux
if: runner.os != 'Windows'
run: |
PYTHONPATH=./scripts/west_commands pytest ./scripts/west_commands/tests/

40
.gitignore vendored
View File

@@ -7,46 +7,24 @@
*.swp
*.swo
*~
build*/
!doc/guides/build
!tests/drivers/build_all
cscope.*
.dir
/*.patch
# The .cache directory will be used to cache toolchain capabilities if
# no suitable out-of-tree directory is found.
.cache
outdir
outdir-*
scripts/basic/fixdep
scripts/gen_idt/gen_idt
scripts/gen_offset_header/gen_offset_header
scripts/kconfig/conf
scripts/kconfig/mconf
scripts/kconfig/zconf.hash.c
scripts/kconfig/zconf.lex.c
scripts/kconfig/zconf.tab.c
doc/_build
doc/doxygen
doc/xml
doc/html
doc/boards
doc/samples
doc/latex
doc/themes/zephyr-docs-theme
sanity-out*
bsim_bt_out
sanity-out/
scripts/grub
doc/reference/kconfig/*.rst
doc/doc.warnings
.*project
.settings
.envrc
.vscode
hide-defaults-note
# Tag files
GPATH
GRTAGS
GTAGS
TAGS
doc/reference/kconfig/CONFIG_*
doc/reference/kconfig/index.rst
tags
.idea

View File

@@ -1,57 +0,0 @@
# All these sections are optional, edit this file as you like.
[general]
ignore=title-trailing-punctuation, T3, title-max-length, T1, body-hard-tab, B3, B1
# verbosity should be a value between 1 and 3, the commandline -v flags take precedence over this
verbosity = 3
# By default gitlint will ignore merge commits. Set to 'false' to disable.
ignore-merge-commits=true
# Enable debug mode (prints more output). Disabled by default
debug = false
# Set the extra-path where gitlint will search for user defined rules
# See http://jorisroovers.github.io/gitlint/user_defined_rules for details
extra-path=scripts/gitlint
[title-max-length-no-revert]
line-length=75
[body-min-line-count]
min-line-count=1
[body-max-line-count]
max-line-count=200
[title-starts-with-subsystem]
regex = ^(?!subsys:)(([^:]+):)(\s([^:]+):)*\s(.+)$
[title-must-not-contain-word]
# Comma-separated list of words that should not occur in the title. Matching is case
# insensitive. It's fine if the keyword occurs as part of a larger word (so "WIPING"
# will not cause a violation, but "WIP: my title" will.
words=wip
[title-match-regex]
# python like regex (https://docs.python.org/2/library/re.html) that the
# commit-msg title must be matched to.
# Note that the regex can contradict with other rules if not used correctly
# (e.g. title-must-not-contain-word).
#regex=^US[0-9]*
[max-line-length-with-exceptions]
# B1 = body-max-line-length
line-length=72
[body-min-length]
min-length=3
[body-is-missing]
# Whether to ignore this rule on merge commits (which typically only have a title)
# default = True
ignore-merge-commits=false
[body-changed-file-mention]
# List of files that need to be explicitly mentioned in the body when they are changed
# This is useful for when developers often erroneously edit certain files or git submodules.
# By specifying this rule, developers can only change the file when they explicitly reference
# it in the commit message.
#files=gitlint/rules.py,README.md

4
.gitreview Normal file
View File

@@ -0,0 +1,4 @@
[gerrit]
host=gerrit.zephyrproject.org
port=29418
project=zephyr.git

View File

@@ -3,7 +3,7 @@
#
# FIXME: all these should match the relative filename
#
^(?P<filename>([\-:\\/\w\.])+[/\\]doc[/\\]reference[/\\]bluetooth.rst):(?P<lineno>[0-9]+): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]$
^(?P<filename>[-._/\w]+/doc/api/bluetooth.rst):(?P<lineno>[0-9]+): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]$
^[ \t]*$
^[ \t]*\^$
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]$
@@ -15,7 +15,7 @@
#
# bt_gatt_discover_params unnamed struct definition
#
^(?P<filename>([\-:\\/\w\.])+[/\\]doc[/\\]reference[/\\]bluetooth.rst):(?P<lineno>[0-9]+): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^(?P<filename>[-._/\w]+/doc/api/bluetooth.rst):(?P<lineno>[0-9]+): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
@@ -27,7 +27,7 @@
#
# Bluetooth GATT unnamed struct definition
#
^(?P<filename>([\-:\\/\w\.])+[/\\]doc[/\\]reference[/\\]bluetooth.rst):(?P<lineno>[0-9]+): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^(?P<filename>[-._/\w]+/doc/api/bluetooth.rst):(?P<lineno>[0-9]+): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
@@ -36,33 +36,3 @@
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected end of definition. \[error at [0-9]+]
^.*bt_gatt_read_params.__unnamed__.*
^[- \t]*\^
#
# Bluetooth mesh unnamed struct definition
#
^(?P<filename>([\-:\\/\w\.])+[/\\]doc[/\\]reference[/\\]bluetooth.rst):(?P<lineno>[0-9]+): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected end of definition. \[error at [0-9]+]
^.*bt_mesh_model.__unnamed__.*
^[- \t]*\^
#
# Bluetooth mesh pub struct definition
#
^(?P<filename>([\-:\\/\w\.])+[/\\]doc[/\\]reference[/\\]bluetooth[/\\]mesh[/\\]access.rst):(?P<lineno>[0-9]+): WARNING: Invalid definition: Expected end of definition. \[error at [0-9]+]
^.*bt_mesh_model_pub.*
^[- \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected end of definition. \[error at [0-9]+]
^.*bt_mesh_model_pub.*
^[- \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected end of definition. \[error at [0-9]+]
^.*bt_mesh_model_pub.*
^[- \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected end of definition. \[error at [0-9]+]
^.*bt_mesh_model_pub.*
^[- \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected end of definition. \[error at [0-9]+]
^.*bt_mesh_model_pub.*
^[- \t]*\^

View File

@@ -1,25 +0,0 @@
#
# crypto unnamed struct definition
#doc/reference/crypto/index.rst
^(?P<filename>([\-:\\/\w\.])+[/\\]doc[/\\]reference[/\\]crypto[/\\]index.rst):(?P<lineno>[0-9]+): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected end of definition. \[error at [0-9]+]
^.*cipher_ctx.mode_params.*
^[- \t]*\^
#
^(?P<filename>([\-:\\/\w\.])+[/\\]doc[/\\]reference[/\\]crypto[/\\]index.rst):(?P<lineno>[0-9]+): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected end of definition. \[error at [0-9]+]
^.*cipher_ctx.key.*
^[- \t]*\^

View File

@@ -1,15 +0,0 @@
#
# Display
#
#
# include
#
^(?P<filename>([\-:\\/\w\.])+[/\\]doc[/\\]reference[/\\]display_api.rst):(?P<lineno>[0-9]+): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected end of definition. \[error at [0-9]+]
^.*mb_image.__unnamed__
^[- \t]*\^

View File

@@ -1,6 +0,0 @@
#
^(?P<filename>([\-:\\/\w\.])+[/\\]doc[/\\]reference[/\\]file_system[/\\]index.rst):(?P<lineno>[0-9]+): WARNING: Duplicate declaration(.*)
#
^(?P<filename>([\-:\\/\w\.])+[/\\]doc[/\\]reference[/\\]peripherals[/\\]dma.rst):(?P<lineno>[0-9]+): WARNING: Duplicate declaration(.*)
#
^(?P<filename>([\-:\\/\w\.])+[/\\]doc[/\\]reference[/\\]peripherals[/\\]sensor.rst):(?P<lineno>[0-9]+): WARNING: Duplicate declaration(.*)

View File

@@ -0,0 +1,18 @@
#
# KERNELVERSION not being defined in local builds, kill that warning,
# can ignore it
#
^.*/Kconfig.zephyr:[0-9]+: warning: The symbol KERNELVERSION references the non-existent environment variable KERNELVERSION.*
#
# Documentation generation, early message
#
^cd .* && doxygen doc/doxygen.config
^srctree=.* SRCARCH=\w+ python scripts/genrest/genrest.py .*$
# This cuts the sphinx build line; has to be separate because in the
# middle, we have removed the KERNELVERSION one and a full regex won't match
^sphinx-build -t \w+ -b html -d [-._/\w]+ -q \. .*
#
# Documentation generation, footer message
#
^[ \t]*
^Build finished. The HTML pages are in [-._/\w]+

View File

@@ -1,15 +0,0 @@
#
# Display
#
#
# include
#
^(?P<filename>([\-:\\/\w\.])+[/\\]doc[/\\]reference[/\\]misc_api.rst):(?P<lineno>[0-9]+): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected end of definition. \[error at [0-9]+]
^.*json_obj_descr.__unnamed__
^[- \t]*\^

View File

@@ -1,70 +0,0 @@
#
# Networking
#
#
# include/net/net_ip.h warnings
#
^(?P<filename>([\-:\\/\w\.])+[/\\]doc[/\\]reference[/\\]networking.rst):(?P<lineno>[0-9]+): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected end of definition. \[error at [0-9]+]
^.*in[_6]+addr.in[46]_u
^[- \t]*\^
#
# include/net/net_mgmt.h
#
^(?P<filename>([\-:\\/\w\.])+[/\\]doc[/\\]reference[/\\]networking.rst):(?P<lineno>[0-9]+): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected end of definition. \[error at [0-9]+]
^.*net_mgmt_event_callback.__unnamed__
^[- \t]*\^
#
# include/net/buf.h
#
^(?P<filename>([\-:\\/\w\.])+[/\\]doc[/\\]reference[/\\]networking.rst):(?P<lineno>[0-9]+): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected end of definition. \[error at [0-9]+]
^.*net_buf.__unnamed__
^[- \t]*\^
#
# include/net/ieee802154.h
#
^(?P<filename>([\-:\\/\w\.])+[/\\]doc[/\\]reference[/\\]networking.rst):(?P<lineno>[0-9]+): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected end of definition. \[error at [0-9]+]
^.*ieee802154_req_params.__unnamed__
^[- \t]*\^
#
# include/net/net_context.h
#
^(?P<filename>([\-:\\/\w\.])+[/\\]doc[/\\]reference[/\\]networking[/\\]net_context.rst):(?P<lineno>[0-9]+): WARNING: Invalid definition: Expected end of definition. \[error at [0-9]+]
^.*net_context.options
^[- \t]*\^
#
# include/net/net_stats.h
#
^(?P<filename>([\-:\\/\w\.])+[/\\]doc[/\\]reference[/\\]networking[/\\]net_stats.rst):(?P<lineno>[0-9]+): WARNING: Invalid definition: Expected end of definition. \[error at [0-9]+]
^.*net_stats_tc.[a-z]+
^[- \t]*\^
#
# stray duplicate definition warnings
#
^(?P<filename>([\-:\\/\w\.])+[/\\]doc[/\\]reference[/\\]networking[/\\]net_if.rst):(?P<lineno>[0-9]+): WARNING: Duplicate declaration(.*)

View File

@@ -0,0 +1,12 @@
#
# Sensor value unnamed struct definition
#
^(?P<filename>[-._/\w]+/doc/api/io_interfaces.rst):(?P<lineno>[0-9]+): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected end of definition. \[error at [0-9]+]
^.*sensor_value.__unnamed__.*
^[- \t]*\^

View File

@@ -1,7 +1,7 @@
#
# UART unnamed struct definition
#doc/api/peripherals/uart.rst
^(?P<filename>([\-:\\/\w\.])+[/\\]doc[/\\]reference[/\\]peripherals[/\\]uart.rst):(?P<lineno>[0-9]+): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
#
^(?P<filename>[-._/\w]+/doc/api/io_interfaces.rst):(?P<lineno>[0-9]+): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
@@ -13,19 +13,3 @@
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected end of definition. \[error at [0-9]+]
^.*uart_device_config.__unnamed__.*
^[- \t]*\^
#
^(?P<filename>([\-:\\/\w\.])+[/\\]doc[/\\]reference[/\\]peripherals[/\\]uart.rst):(?P<lineno>[0-9]+): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected identifier in nested name. \[error at [0-9]+]
^[ \t]*
^[ \t]*\^
^(?P=filename):(?P=lineno): WARNING: Invalid definition: Expected end of definition. \[error at [0-9]+]
^.*uart_event.data
^[- \t]*\^

View File

@@ -18,15 +18,3 @@ Keren Siman-Tov <keren.siman-tov@intel.com> <keren.siman-tov@intel.com>
Naga Raja Rao Tulasi <tulasi.r@tcs.com> <tulasi.r@tcs.com>
Felipe Neves <ryukokki.felipe@gmail.com> <ryukokki.felipe@gmail.com>
Amir Kaplan <amir.kaplan@intel.com> <amir.kaplan@intel.com>
Anas Nashif <anas.nashif@intel.com> <anas.nashif@intel.com>
Ruud Derwig <Ruud.Derwig@synopsys.com> <Ruud.Derwig@synopsys.com>
Flavio Arieta Netto <flavio@exati.com.br>
Nishikant Nayak <nishikantax.nayak@intel.com>
Justin Watson <jwatson5@gmail.com>
Johann Fischer <j.fischer@phytec.de>
Jun Li <jun.r.li@intel.com>
Xiaorui Hu <xiaorui.hu@linaro.org>
Yannis Damigos <giannis.damigos@gmail.com> <ydamigos@iccs.gr>
Vinayak Kariappa Chettimada <vinayak.kariappa.chettimada@nordicsemi.no> <vinayak.kariappa.chettimada@nordicsemi.no> <vich@nordicsemi.no> <vinayak.kariappa@gmail.com>
Sean Nyekjaer <sean@geanix.com> <sean.nyekjaer@prevas.dk>
Sean Nyekjaer <sean@geanix.com> <sean@nyekjaer.dk>

View File

@@ -1,75 +0,0 @@
language: c
compiler: gcc
env:
global:
- ZEPHYR_SDK_INSTALL_DIR=/opt/sdk/zephyr-sdk-0.11.3
- ZEPHYR_TOOLCHAIN_VARIANT=zephyr
- MATRIX_BUILDS="5"
matrix:
- MATRIX_BUILD="1"
- MATRIX_BUILD="2"
- MATRIX_BUILD="3"
- MATRIX_BUILD="4"
- MATRIX_BUILD="5"
build:
cache: false
cache_dir_list:
- ${SHIPPABLE_BUILD_DIR}/ccache
pre_ci_boot:
image_name: zephyrprojectrtos/ci
image_tag: v0.11.8
pull: true
options: "-e HOME=/home/buildslave --privileged=true --tty --net=bridge --user buildslave"
ci:
- export CCACHE_DIR=${SHIPPABLE_BUILD_DIR}/ccache/.ccache
- >
if [ "$IS_PULL_REQUEST" = "true" ]; then
./scripts/ci/run_ci.sh -c -b ${PULL_REQUEST_BASE_BRANCH} -r origin -m ${MATRIX_BUILD} -M ${MATRIX_BUILDS} -p ${PULL_REQUEST};
else
./scripts/ci/run_ci.sh -c -b ${BRANCH} -r origin -m ${MATRIX_BUILD} -M ${MATRIX_BUILDS};
fi;
- ccache -s
on_failure:
- >
if [ "$IS_PULL_REQUEST" = "true" ]; then
./scripts/ci/run_ci.sh -f -b ${PULL_REQUEST_BASE_BRANCH} -r origin -m ${MATRIX_BUILD} -M ${MATRIX_BUILDS} -p ${PULL_REQUEST};
else
./scripts/ci/run_ci.sh -f -b ${BRANCH} -r origin -m ${MATRIX_BUILD} -M ${MATRIX_BUILDS};
fi;
on_success:
- >
if [ "$IS_PULL_REQUEST" = "true" ]; then
./scripts/ci/run_ci.sh -s -b ${PULL_REQUEST_BASE_BRANCH} -r origin -m ${MATRIX_BUILD} -M ${MATRIX_BUILDS} -p ${PULL_REQUEST};
else
./scripts/ci/run_ci.sh -s -b ${BRANCH} -r origin -m ${MATRIX_BUILD} -M ${MATRIX_BUILDS};
fi;
branches:
only:
- master
- v*-branch
- topic-*
integrations:
notifications:
- integrationName: slack_integration
type: slack
recipients:
- "#ci"
branches:
only:
- master
on_success: never
on_failure: never
- integrationName: email
type: email
recipients:
- builds@zephyrproject.org
branches:
only:
- master
on_success: never
on_failure: always
on_pull_request: never

View File

@@ -1,80 +0,0 @@
indent_with_tabs = 2 # 1=indent to level only, 2=indent with tabs
input_tab_size = 8 # original tab size
output_tab_size = 8 # new tab size
indent_columns = output_tab_size
indent_label = 1 # pos: absolute col, neg: relative column
indent_switch_case = 0 # number
#
# inter-symbol newlines
#
nl_enum_brace = remove # "enum {" vs "enum \n {"
nl_union_brace = remove # "union {" vs "union \n {"
nl_struct_brace = remove # "struct {" vs "struct \n {"
nl_do_brace = remove # "do {" vs "do \n {"
nl_if_brace = remove # "if () {" vs "if () \n {"
nl_for_brace = remove # "for () {" vs "for () \n {"
nl_else_brace = remove # "else {" vs "else \n {"
nl_while_brace = remove # "while () {" vs "while () \n {"
nl_switch_brace = remove # "switch () {" vs "switch () \n {"
nl_brace_while = remove # "} while" vs "} \n while" - cuddle while
nl_brace_else = remove # "} \n else" vs "} else"
nl_func_var_def_blk = 1
nl_fcall_brace = remove # "list_for_each() {" vs "list_for_each()\n{"
nl_fdef_brace = add # "int foo() {" vs "int foo()\n{"
#
# Source code modifications
#
mod_paren_on_return = ignore # "return 1;" vs "return (1);"
mod_full_brace_if = add # "if() { } else { }" vs "if() else"
#
# inter-character spacing options
#
sp_sizeof_paren = remove # "sizeof (int)" vs "sizeof(int)"
sp_before_sparen = force # "if (" vs "if("
sp_after_sparen = force # "if () {" vs "if (){"
sp_inside_braces = add # "{ 1 }" vs "{1}"
sp_inside_braces_struct = add # "{ 1 }" vs "{1}"
sp_inside_braces_enum = add # "{ 1 }" vs "{1}"
sp_assign = add
sp_arith = add
sp_bool = add
sp_compare = add
sp_assign = add
sp_after_comma = add
sp_func_def_paren = remove # "int foo (){" vs "int foo(){"
sp_func_call_paren = remove # "foo (" vs "foo("
sp_func_proto_paren = remove # "int foo ();" vs "int foo();"
sp_inside_fparen = remove # "func( arg )" vs "func(arg)"
sp_else_brace = add # ignore/add/remove/force
sp_before_ptr_star = add # ignore/add/remove/force
sp_after_ptr_star = remove # ignore/add/remove/force
sp_between_ptr_star = remove # ignore/add/remove/force
sp_inside_paren = remove # remove spaces inside parens
sp_paren_paren = remove # remove spaces between nested parens
sp_inside_sparen = remove # remove spaces inside parens for if, while and the like
sp_brace_else = add # ignore/add/remove/force
sp_before_nl_cont = ignore
sp_cmt_cpp_start = add
sp_brace_typedef = add # }typedefd_name -> } typedefd_name
cmt_sp_after_star_cont = 1
#
# Aligning stuff
#
align_with_tabs = FALSE # use tabs to align
align_on_tabstop = TRUE # align on tabstops
align_enum_equ_span = 4 # '=' in enum definition
align_struct_init_span = 0 # align stuff in a structure init '= { }'
align_right_cmt_span = 3
align_nl_cont = TRUE
sp_pp_concat = ignore # ignore/add/remove/force

File diff suppressed because it is too large Load Diff

View File

@@ -1,508 +0,0 @@
# CODEOWNERS for autoreview assigning in github
# https://help.github.com/en/articles/about-code-owners#codeowners-syntax
# Order is important; for each modified file, the last matching
# pattern takes the most precedence.
# That is, with the last pattern being
# *.rst @nashif
# if only .rst files are being modified, only nashif is
# automatically requested for review, but you can manually
# add others as needed.
# Do not use wildcard on all source yet
# * @galak @nashif
/.known-issues/ @nashif
/.github/ @nashif
/.github/workflows/ @galak @nashif
/arch/arc/ @vonhust @ruuddw
/arch/arm/ @MaureenHelm @galak @ioannisg
/arch/arm/core/aarch32/cortex_m/cmse/ @ioannisg
/arch/arm/core/aarch64/ @carlocaione
/arch/arm/include/aarch32/cortex_m/cmse.h @ioannisg
/arch/arm/include/aarch64/ @carlocaione
/arch/arm/core/aarch32/cortex_a_r/ @MaureenHelm @galak @ioannisg @bbolen @stephanosio
/arch/common/ @andrewboie @ioannisg @andyross
/soc/arc/snps_*/ @vonhust @ruuddw
/soc/nios2/ @nashif @wentongwu
/soc/arm/ @MaureenHelm @galak @ioannisg
/soc/arm/arm/mps2/ @fvincenzo
/soc/arm/atmel_sam/sam3x/ @ioannisg
/soc/arm/atmel_sam/sam4e/ @nandojve
/soc/arm/atmel_sam/sam4s/ @fallrisk
/soc/arm/atmel_sam/same70/ @nandojve
/soc/arm/atmel_sam/samv71/ @nandojve
/soc/arm/bcm*/ @sbranden
/soc/arm/infineon_xmc/ @parthitce
/soc/arm/nxp*/ @MaureenHelm
/soc/arm/nordic_nrf/ @ioannisg
/soc/arm/qemu_cortex_a53/ @carlocaione
/soc/arm/st_stm32/ @erwango
/soc/arm/st_stm32/stm32f4/ @idlethread
/soc/arm/st_stm32/stm32mp1/ @arnopo
/soc/arm/ti_simplelink/cc13x2_cc26x2/ @bwitherspoon
/soc/arm/ti_simplelink/cc32xx/ @vanti
/soc/arm/ti_simplelink/msp432p4xx/ @Mani-Sadhasivam
/soc/arm/xilinx_zynqmp/ @stephanosio
/soc/xtensa/intel_s1000/ @sathishkuttan @dcpleung
/arch/x86/ @andrewboie
/arch/nios2/ @andrewboie @wentongwu
/arch/posix/ @aescolar @daor-oti
/arch/riscv/ @kgugala @pgielda @nategraff-sifive
/soc/posix/ @aescolar @daor-oti
/soc/riscv/ @kgugala @pgielda @nategraff-sifive
/soc/riscv/openisa*/ @MaureenHelm
/soc/x86/ @andrewboie
/arch/xtensa/ @andrewboie @dcpleung @andyross
/soc/xtensa/ @andrewboie @dcpleung @andyross
/boards/arc/ @vonhust @ruuddw
/boards/arm/ @MaureenHelm @galak
/boards/arm/96b_argonkey/ @avisconti
/boards/arm/96b_avenger96/ @Mani-Sadhasivam
/boards/arm/96b_carbon/ @idlethread
/boards/arm/96b_meerkat96/ @Mani-Sadhasivam
/boards/arm/96b_nitrogen/ @idlethread
/boards/arm/96b_neonkey/ @Mani-Sadhasivam
/boards/arm/96b_stm32_sensor_mez/ @Mani-Sadhasivam
/boards/arm/96b_wistrio/ @Mani-Sadhasivam
/boards/arm/arduino_due/ @ioannisg
/boards/arm/cc1352r1_launchxl/ @bwitherspoon
/boards/arm/cc26x2r1_launchxl/ @bwitherspoon
/boards/arm/cc3220sf_launchxl/ @vanti
/boards/arm/disco_l475_iot1/ @erwango
/boards/arm/frdm*/ @MaureenHelm
/boards/arm/frdm*/doc/ @MaureenHelm @MeganHansen
/boards/arm/google_*/ @jackrosenthal
/boards/arm/hexiwear*/ @MaureenHelm
/boards/arm/hexiwear*/doc/ @MaureenHelm @MeganHansen
/boards/arm/ip_k66f/ @parthitce
/boards/arm/lpcxpresso*/ @MaureenHelm
/boards/arm/lpcxpresso*/doc/ @MaureenHelm @MeganHansen
/boards/arm/mimxrt*/ @MaureenHelm
/boards/arm/mimxrt*/doc/ @MaureenHelm @MeganHansen
/boards/arm/mps2_an385/ @fvincenzo
/boards/arm/msp_exp432p401r_launchxl/ @Mani-Sadhasivam
/boards/arm/nrf*/ @carlescufi @lemrey @ioannisg
/boards/arm/nucleo*/ @erwango
/boards/arm/nucleo_f401re/ @idlethread
/boards/arm/qemu_cortex_a53/ @carlocaione
/boards/arm/qemu_cortex_r*/ @stephanosio
/boards/arm/qemu_cortex_m*/ @ioannisg
/boards/arm/xmc45_relax_kit/ @parthitce
/boards/arm/sam4e_xpro/ @nandojve
/boards/arm/sam4s_xplained/ @fallrisk
/boards/arm/sam_e70_xplained/ @nandojve
/boards/arm/sam_v71_xult/ @nandojve
/boards/arm/v2m_beetle/ @fvincenzo
/boards/arm/olimexino_stm32/ @ydamigos
/boards/arm/sensortile_box/ @avisconti
/boards/arm/steval_fcu001v1/ @Navin-Sankar
/boards/arm/stm32l1_disco/ @karlp
/boards/arm/stm32*_disco/ @erwango
/boards/arm/stm32f3_disco/ @ydamigos
/boards/arm/stm32*_eval/ @erwango
/boards/common/ @mbolivar-nordic
/boards/deprecated.cmake @tejlmand
/boards/nios2/ @wentongwu
/boards/nios2/altera_max10/ @wentongwu
/boards/arm/stm32_min_dev/ @cbsiddharth
/boards/posix/ @aescolar @daor-oti
/boards/posix/nrf52_bsim/ @aescolar @wopu-ot
/boards/riscv/ @kgugala @pgielda @nategraff-sifive
/boards/riscv/rv32m1_vega/ @MaureenHelm
/boards/shields/ @erwango
/boards/x86/ @andrewboie @nashif
/boards/xtensa/ @nashif @dcpleung
/boards/xtensa/intel_s1000_crb/ @sathishkuttan @dcpleung
/boards/xtensa/odroid_go/ @ydamigos
# All cmake related files
/cmake/ @tejlmand @nashif
/CMakeLists.txt @tejlmand @nashif
/doc/ @dbkinder
/doc/guides/coccinelle.rst @himanshujha199640 @JuliaLawall
/doc/CMakeLists.txt @carlescufi
/doc/scripts/ @carlescufi
/doc/guides/bluetooth/ @joerchan @jhedberg @Vudentz
/doc/guides/dts/ @galak @mbolivar-nordic
/doc/reference/bluetooth/ @joerchan @jhedberg @Vudentz
/doc/reference/devicetree/ @galak @mbolivar-nordic
/doc/reference/resource_management/ @pabigot
/doc/reference/networking/can* @alexanderwachter
/drivers/debug/ @nashif
/drivers/*/*cc13xx_cc26xx* @bwitherspoon
/drivers/*/*mcux* @MaureenHelm
/drivers/*/*stm32* @erwango
/drivers/*/*native_posix* @aescolar @daor-oti
/drivers/adc/ @anangl
/drivers/adc/adc_stm32.c @cybertale
/drivers/bluetooth/ @joerchan @jhedberg @Vudentz
/drivers/can/ @alexanderwachter
/drivers/can/*mcp2515* @karstenkoenig
/drivers/clock_control/*nrf* @nordic-krch
/drivers/counter/ @nordic-krch
/drivers/console/semihost_console.c @luozhongyao
/drivers/counter/counter_cmos.c @andrewboie
/drivers/counter/maxim_ds3231.c @pabigot
/drivers/crypto/*nrf_ecb* @maciekfabia @anangl
/drivers/console/*mux* @jukkar
/drivers/display/ @vanwinkeljan
/drivers/display/display_framebuf.c @andrewboie
/drivers/dac/ @martinjaeger
/drivers/dma/*dw* @tbursztyka
/drivers/dma/*sam0* @Sizurka
/drivers/dma/dma_stm32* @cybertale
/drivers/eeprom/ @henrikbrixandersen
/drivers/eeprom/eeprom_stm32.c @KwonTae-young
/drivers/entropy/*rv32m1* @MaureenHelm
/drivers/entropy/*gecko* @chrta
/drivers/espi/ @albertofloyd @franciscomunoz @scottwcpg
/drivers/ps2/ @albertofloyd @franciscomunoz @scottwcpg
/drivers/kscan/ @albertofloyd @franciscomunoz @scottwcpg
/drivers/peci/ @albertofloyd @franciscomunoz @scottwcpg
/drivers/ethernet/ @jukkar @tbursztyka @pfalcon
/drivers/entropy/*litex* @mateusz-holenko @kgugala @pgielda
/drivers/flash/ @nashif @nvlsianpu
/drivers/flash/*nrf* @nvlsianpu
/drivers/flash/*spi_nor* @pabigot
/drivers/flash/*stm32* @superna9999
/drivers/gpio/ @mnkp @pabigot
/drivers/gpio/*ht16k33* @henrikbrixandersen
/drivers/gpio/*lmp90xxx* @henrikbrixandersen
/drivers/gpio/*stm32* @erwango
/drivers/gpio/*sx1509b* @pabigot
/drivers/gpio/*litex* @mateusz-holenko @kgugala @pgielda
/drivers/hwinfo/ @alexanderwachter
/drivers/i2c/*litex* @mateusz-holenko @kgugala @pgielda
/drivers/i2s/i2s_ll_stm32* @avisconti
/drivers/i2c/i2c_shell.c @nashif
/drivers/ieee802154/ @jukkar @tbursztyka
/drivers/ieee802154/ieee802154_rf2xx* @jukkar @tbursztyka @nandojve
/drivers/interrupt_controller/ @andrewboie
/drivers/interrupt_controller/intc_gic.c @stephanosio
/drivers/*/intc_vexriscv_litex.c @mateusz-holenko @kgugala @pgielda
/drivers/ipm/ipm_mhu* @karl-zh
/drivers/ipm/Kconfig.nrfx @masz-nordic @ioannisg
/drivers/ipm/Kconfig.nrfx_ipc_channel @masz-nordic @ioannisg
/drivers/ipm/ipm_cavs_idc* @dcpleung
/drivers/ipm/ipm_nrfx_ipc.c @masz-nordic @ioannisg
/drivers/ipm/ipm_nrfx_ipc.h @masz-nordic @ioannisg
/drivers/ipm/ipm_stm32_ipcc.c @arnopo
/drivers/led/ @Mani-Sadhasivam
/drivers/led_strip/ @mbolivar-nordic
/drivers/lora/ @Mani-Sadhasivam
/drivers/modem/ @mike-scott
/drivers/pcie/ @andrewboie
/drivers/pinmux/stm32/ @idlethread
/drivers/pinmux/*hsdk* @iriszzw
/drivers/pwm/*litex* @mateusz-holenko @kgugala @pgielda
/drivers/sensor/ @MaureenHelm
/drivers/sensor/ams_iAQcore/ @alexanderwachter
/drivers/sensor/ens210/ @alexanderwachter
/drivers/sensor/hts*/ @avisconti
/drivers/sensor/lis*/ @avisconti
/drivers/sensor/lps*/ @avisconti
/drivers/sensor/lsm*/ @avisconti
/drivers/sensor/st*/ @avisconti
/drivers/serial/uart_altera_jtag_hal.c @wentongwu
/drivers/serial/*ns16550* @andrewboie
/drivers/serial/*nrfx* @Mierunski @anangl
/drivers/serial/Kconfig.litex @mateusz-holenko @kgugala @pgielda
/drivers/serial/uart_liteuart.c @mateusz-holenko @kgugala @pgielda
/drivers/serial/Kconfig.rtt @carlescufi @pkral78
/drivers/serial/uart_rtt.c @carlescufi @pkral78
/drivers/serial/Kconfig.xlnx @wjliang
/drivers/serial/uart_xlnx_ps.c @wjliang
/drivers/serial/*xmc4xxx* @parthitce
/drivers/net/ @jukkar @tbursztyka
/drivers/ptp_clock/ @jukkar
/drivers/pwm/*rv32m1* @henrikbrixandersen
/drivers/pwm/pwm_shell.c @henrikbrixandersen
/drivers/spi/ @tbursztyka
/drivers/spi/spi_ll_stm32.* @superna9999
/drivers/spi/spi_rv32m1_lpspi* @karstenkoenig
/drivers/timer/apic_timer.c @andrewboie
/drivers/timer/arm_arch_timer.c @carlocaione
/drivers/timer/cortex_m_systick.c @ioannisg
/drivers/timer/altera_avalon_timer_hal.c @wentongwu
/drivers/timer/riscv_machine_timer.c @nategraff-sifive @kgugala @pgielda
/drivers/timer/litex_timer.c @mateusz-holenko @kgugala @pgielda
/drivers/timer/xlnx_psttc_timer* @wjliang @stephanosio
/drivers/timer/cc13x2_cc26x2_rtc_timer.c @vanti
/drivers/timer/cavs_timer.c @dcpleung
/drivers/usb/ @jfischer-phytec-iot @finikorg
/drivers/usb/device/usb_dc_stm32.c @ydamigos @loicpoulain
/drivers/video/ @loicpoulain
/drivers/i2c/i2c_ll_stm32* @ldts @ydamigos
/drivers/i2c/i2c_rv32m1_lpi2c* @henrikbrixandersen
/drivers/i2c/*sam0* @Sizurka
/drivers/i2c/i2c_dw* @dcpleung
/drivers/*/*xec* @franciscomunoz @albertofloyd @scottwcpg
/drivers/watchdog/*gecko* @oanerer
/drivers/watchdog/wdt_handlers.c @andrewboie
/drivers/wifi/ @jukkar @tbursztyka @pfalcon
/drivers/wifi/eswifi/ @loicpoulain
/dts/arc/ @vonhust @ruuddw @iriszzw
/dts/arm/atmel/sam4e* @nandojve
/dts/arm/atmel/samr21.dtsi @benpicco
/dts/arm/atmel/sam*5*.dtsi @benpicco
/dts/arm/atmel/same70* @nandojve
/dts/arm/atmel/samv71* @nandojve
/dts/arm/atmel/ @galak
/dts/arm/broadcom/ @sbranden
/dts/arm/infineon/ @parthitce
/dts/arm/qemu-virt/ @carlocaione
/dts/arm/st/ @erwango
/dts/arm/ti/cc13?2* @bwitherspoon
/dts/arm/ti/cc26?2* @bwitherspoon
/dts/arm/ti/cc3235* @vanti
/dts/arm/nordic/ @ioannisg @carlescufi
/dts/arm/nxp/ @MaureenHelm
/dts/arm/microchip/ @franciscomunoz @albertofloyd @scottwcpg
/dts/arm/silabs/efm32gg11b* @oanerer
/dts/arm/silabs/efm32_jg_pg* @chrta
/dts/arm/silabs/efr32bg13p* @mnkp
/dts/arm/silabs/efm32jg12b* @chrta
/dts/arm/silabs/efm32pg12b* @chrta
/dts/riscv/microsemi-miv.dtsi @galak
/dts/riscv/rv32m1* @MaureenHelm
/dts/riscv/riscv32-fe310.dtsi @nategraff-sifive
/dts/riscv/riscv32-litex-vexriscv.dtsi @mateusz-holenko @kgugala @pgielda
/dts/arm/armv7-r.dtsi @bbolen @stephanosio
/dts/arm/armv8-a.dtsi @carlocaione
/dts/arm/xilinx/ @bbolen @stephanosio
/dts/xtensa/xtensa.dtsi @ydamigos
/dts/xtensa/intel/ @dcpleung
/dts/bindings/ @galak
/dts/bindings/can/ @alexanderwachter
/dts/bindings/iio/adc/st*stm32-adc.yaml @cybertale
/dts/bindings/serial/ns16550.yaml @andrewboie
/dts/bindings/*/nordic* @anangl
/dts/bindings/*/nxp* @MaureenHelm
/dts/bindings/*/openisa* @MaureenHelm
/dts/bindings/*/st* @erwango
/dts/bindings/sensor/ams* @alexanderwachter
/dts/bindings/*/sifive* @mateusz-holenko @kgugala @pgielda @nategraff-sifive
/dts/bindings/*/litex* @mateusz-holenko @kgugala @pgielda
/dts/bindings/*/vexriscv* @mateusz-holenko @kgugala @pgielda
/dts/posix/ @aescolar @vanwinkeljan @daor-oti
/dts/bindings/sensor/*bme680* @BoschSensortec
/dts/bindings/sensor/st* @avisconti
/include/ @nashif @carlescufi @galak @MaureenHelm
/include/drivers/adc.h @anangl
/include/drivers/can.h @alexanderwachter
/include/drivers/counter.h @nordic-krch
/include/drivers/dac.h @martinjaeger
/include/drivers/display.h @vanwinkeljan
/include/drivers/espi.h @albertofloyd @franciscomunoz @scottwcpg
/include/drivers/bluetooth/ @joerchan @jhedberg @Vudentz
/include/drivers/flash.h @nashif @carlescufi @galak @MaureenHelm @nvlsianpu
/include/drivers/led/ht16k33.h @henrikbrixandersen
/include/drivers/interrupt_controller/ @andrewboie
/include/drivers/interrupt_controller/gic.h @stephanosio
/include/drivers/pcie/ @andrewboie
/include/drivers/hwinfo.h @alexanderwachter
/include/drivers/led.h @Mani-Sadhasivam
/include/drivers/led_strip.h @mbolivar-nordic
/include/drivers/sensor.h @MaureenHelm
/include/drivers/spi.h @tbursztyka
/include/drivers/lora.h @Mani-Sadhasivam
/include/drivers/peci.h @albertofloyd @franciscomunoz @scottwcpg
/include/app_memory/ @andrewboie
/include/arch/arc/ @vonhust @ruuddw
/include/arch/arc/arch.h @andrewboie
/include/arch/arc/v2/irq.h @andrewboie
/include/arch/arm/aarch32/ @MaureenHelm @galak @ioannisg
/include/arch/arm/aarch32/cortex_a_r/ @stephanosio
/include/arch/arm/aarch64/ @carlocaione
/include/arch/arm/aarch32/irq.h @andrewboie
/include/arch/nios2/ @andrewboie
/include/arch/nios2/arch.h @andrewboie
/include/arch/posix/ @aescolar @daor-oti
/include/arch/riscv/ @nategraff-sifive @kgugala @pgielda
/include/arch/x86/ @andrewboie @wentongwu
/include/arch/common/ @andrewboie @andyross @nashif
/include/arch/xtensa/ @andrewboie
/include/sys/atomic.h @andrewboie @andyross
/include/bluetooth/ @joerchan @jhedberg @Vudentz
/include/cache.h @andrewboie @andyross
/include/canbus/ @alexanderwachter
/include/tracing/ @wentongwu @nashif
/include/debug/ @nashif
/include/device.h @wentongwu @nashif
/include/devicetree.h @galak
/include/display/ @vanwinkeljan
/include/dt-bindings/clock/kinetis_mcg.h @henrikbrixandersen
/include/dt-bindings/clock/kinetis_scg.h @henrikbrixandersen
/include/dt-bindings/dma/stm32_dma.h @cybertale
/include/dt-bindings/pcie/ @andrewboie
/include/dt-bindings/usb/usb.h @galak @finikorg
/include/fs/ @nashif @wentongwu
/include/init.h @andrewboie @andyross
/include/irq.h @andrewboie @andyross
/include/irq_offload.h @andrewboie @andyross
/include/kernel.h @andrewboie @andyross
/include/kernel_version.h @andrewboie @andyross
/include/linker/app_smem*.ld @andrewboie
/include/linker/ @andrewboie @andyross
/include/logging/ @nordic-krch
/include/net/ @jukkar @tbursztyka @pfalcon
/include/net/buf.h @jukkar @jhedberg @tbursztyka @pfalcon
/include/posix/ @pfalcon
/include/power/power.h @wentongwu @nashif
/include/ptp_clock.h @jukkar
/include/shared_irq.h @andrewboie @andyross
/include/shell/ @jakub-uC @nordic-krch
/include/sw_isr_table.h @andrewboie @andyross
/include/sys_clock.h @andrewboie @andyross
/include/sys/sys_io.h @andrewboie @andyross
/include/toolchain.h @andrewboie @andyross @nashif
/include/toolchain/ @andrewboie @andyross
/include/zephyr.h @andrewboie @andyross
/kernel/ @andrewboie @andyross
/lib/fnmatch/ @carlescufi
/lib/gui/ @vanwinkeljan
/lib/open-amp/ @arnopo
/lib/os/ @andrewboie @andyross
/lib/posix/ @pfalcon
/lib/cmsis_rtos_v2/ @nashif
/lib/cmsis_rtos_v1/ @nashif
/lib/libc/ @nashif @andrewboie
/modules/ @nashif
/kernel/device.c @andrewboie @andyross @nashif
/kernel/idle.c @andrewboie @andyross @nashif
/samples/ @nashif
/samples/basic/minimal/ @carlescufi
/samples/basic/servo_motor/boards/*microbit* @jhe
/lib/updatehub/ @nandojve @otavio
/samples/bluetooth/ @jhedberg @Vudentz @joerchan
/samples/boards/intel_s1000_crb/ @sathishkuttan @dcpleung @nashif
/samples/display/ @vanwinkeljan
/samples/drivers/can/ @alexanderwachter
/samples/drivers/display/ @vanwinkeljan
/samples/drivers/ht16k33/ @henrikbrixandersen
/samples/drivers/lora/ @Mani-Sadhasivam
/samples/drivers/counter/maxim_ds3231/ @pabigot
/samples/net/ @jukkar @tbursztyka @pfalcon
/samples/net/dns_resolve/ @jukkar @tbursztyka @pfalcon
/samples/net/lwm2m_client/ @rlubos
/samples/net/mqtt_publisher/ @jukkar @tbursztyka
/samples/net/sockets/coap_*/ @rveerama1
/samples/net/sockets/ @jukkar @tbursztyka @pfalcon
/samples/net/updatehub/ @nandojve @otavio
/samples/sensor/ @MaureenHelm
/samples/shields/ @avisconti
/samples/subsys/logging/ @nordic-krch @jakub-uC
/samples/subsys/shell/ @jakub-uC @nordic-krch
/samples/subsys/mgmt/mcumgr/smp_svr/ @aunsbjerg @nvlsianpu
/samples/subsys/usb/ @jfischer-phytec-iot @finikorg
/samples/subsys/power/ @wentongwu @pabigot
/samples/userspace/ @andrewboie
/scripts/coccicheck @himanshujha199640 @JuliaLawall
/scripts/coccinelle/ @himanshujha199640 @JuliaLawall
/scripts/kconfig/ @ulfalizer
/scripts/sanity_chk/expr_parser.py @nashif
/scripts/gen_app_partitions.py @andrewboie
/scripts/dts/ @ulfalizer @galak
/scripts/release/ @nashif
/scripts/ci/ @nashif
/arch/x86/gen_gdt.py @andrewboie
/arch/x86/gen_idt.py @andrewboie
/scripts/gen_kobject_list.py @andrewboie
/scripts/gen_syscalls.py @andrewboie
/scripts/net/ @jukkar @pfl
/scripts/process_gperf.py @andrewboie
/scripts/gen_relocate_app.py @wentongwu
/scripts/requirements*.txt @mbolivar @galak @nashif
/scripts/tests/sanitycheck/ @aasthagr
/scripts/tracing/ @wentongwu
/scripts/sanity_chk/ @nashif
/scripts/sanitycheck @nashif
/scripts/series-push-hook.sh @erwango
/scripts/west_commands/ @mbolivar-nordic
/scripts/west-commands.yml @mbolivar-nordic
/scripts/zephyr_module.py @tejlmand
/scripts/user_wordsize.py @cfriedt
/scripts/valgrind.supp @aescolar @daor-oti
/share/zephyr-package/ @tejlmand
/share/zephyrunittest-package/ @tejlmand
/subsys/bluetooth/ @joerchan @jhedberg @Vudentz
/subsys/bluetooth/controller/ @carlescufi @cvinayak @thoh-ot
/subsys/bluetooth/mesh/ @jhedberg @trond-snekvik @joerchan @Vudentz
/subsys/canbus/ @alexanderwachter
/subsys/cpp/ @pabigot @vanwinkeljan
/subsys/debug/ @nashif
/subsys/dfu/ @nvlsianpu
/subsys/tracing/ @nashif @wentongwu
/subsys/debug/asan_hacks.c @vanwinkeljan @aescolar @daor-oti
/subsys/disk/disk_access_spi_sdhc.c @JunYangNXP
/subsys/disk/disk_access_sdhc.h @JunYangNXP
/subsys/disk/disk_access_usdhc.c @JunYangNXP
/subsys/disk/disk_access_stm32_sdmmc.c @anthonybrandon
/subsys/fb/ @jfischer-phytec-iot
/subsys/fs/ @nashif
/subsys/fs/fcb/ @nvlsianpu
/subsys/fs/fuse_fs_access.c @vanwinkeljan
/subsys/fs/littlefs_fs.c @pabigot
/subsys/fs/nvs/ @Laczen
/subsys/logging/ @nordic-krch
/subsys/logging/log_backend_net.c @nordic-krch @jukkar
/subsys/mgmt/ @carlescufi @nvlsianpu
/subsys/mgmt/smp_udp.c @aunsbjerg
/subsys/net/buf.c @jukkar @jhedberg @tbursztyka @pfalcon
/subsys/net/ip/ @jukkar @tbursztyka @pfalcon
/subsys/net/lib/ @jukkar @tbursztyka @pfalcon
/subsys/net/lib/dns/ @jukkar @tbursztyka @pfalcon
/subsys/net/lib/lwm2m/ @rlubos
/subsys/net/lib/config/ @jukkar @tbursztyka @pfalcon
/subsys/net/lib/mqtt/ @jukkar @tbursztyka @rlubos
/subsys/net/lib/coap/ @rveerama1
/subsys/net/lib/sockets/socketpair.c @cfriedt
/subsys/net/lib/sockets/ @jukkar @tbursztyka @pfalcon
/subsys/net/lib/tls_credentials/ @rlubos
/subsys/net/l2/ @jukkar @tbursztyka
/subsys/net/l2/canbus/ @alexanderwachter @jukkar
/subsys/net/*/openthread/ @rlubos
/subsys/power/ @wentongwu @pabigot
/subsys/random/ @dleach02
/subsys/settings/ @nvlsianpu
/subsys/shell/ @jakub-uC @nordic-krch
/subsys/storage/ @nvlsianpu
/subsys/testsuite/ @nashif
/subsys/usb/ @jfischer-phytec-iot @finikorg
/tests/ @nashif
/tests/application_development/libcxx/ @pabigot
/tests/arch/arm/ @ioannisg @stephanosio
/tests/benchmarks/cmsis_dsp/ @stephanosio
/tests/boards/native_posix/ @aescolar @daor-oti
/tests/boards/intel_s1000_crb/ @dcpleung @sathishkuttan
/tests/bluetooth/ @joerchan @jhedberg @Vudentz
/tests/bluetooth/bsim_bt/ @joerchan @jhedberg @Vudentz @aescolar @wopu-ot
/tests/posix/ @pfalcon
/tests/crypto/ @ceolin
/tests/crypto/mbedtls/ @nashif @ceolin
/tests/drivers/can/ @alexanderwachter
/tests/drivers/counter/ @nordic-krch
/tests/drivers/counter/maxim_ds3231_api/ @pabigot
/tests/drivers/flash_simulator/ @nvlsianpu
/tests/drivers/gpio/ @mnkp @pabigot
/tests/drivers/hwinfo/ @alexanderwachter
/tests/drivers/spi/ @tbursztyka
/tests/drivers/uart/uart_async_api/ @Mierunski
/tests/kernel/ @andrewboie @andyross @nashif
/tests/lib/ @nashif
/tests/lib/cmsis_dsp/ @stephanosio
/tests/net/ @jukkar @tbursztyka @pfalcon
/tests/net/buf/ @jukkar @jhedberg @tbursztyka @pfalcon
/tests/net/lib/ @jukkar @tbursztyka @pfalcon
/tests/net/lib/http_header_fields/ @jukkar @tbursztyka
/tests/net/lib/mqtt_packet/ @jukkar @tbursztyka
/tests/net/lib/coap/ @rveerama1
/tests/net/socket/socketpair/ @cfriedt
/tests/net/socket/ @jukkar @tbursztyka @pfalcon
/tests/subsys/fs/ @nashif @wentongwu
/tests/subsys/settings/ @nvlsianpu
/tests/subsys/shell/ @jakub-uC @nordic-krch
# Get all docs reviewed
*.rst @nashif
*posix*.rst @aescolar @daor-oti

View File

@@ -1,78 +0,0 @@
# Contributor Covenant Code of Conduct
## Our Pledge
In the interest of fostering an open and welcoming environment, we as
contributors and maintainers pledge to making participation in our project and
our community a harassment-free experience for everyone, regardless of age, body
size, disability, ethnicity, sex characteristics, gender identity and expression,
level of experience, education, socio-economic status, nationality, personal
appearance, race, religion, or sexual identity and orientation.
## Our Standards
Examples of behavior that contributes to creating a positive environment
include:
* Using welcoming and inclusive language
* Being respectful of differing viewpoints and experiences
* Gracefully accepting constructive criticism
* Focusing on what is best for the community
* Showing empathy towards other community members
Examples of unacceptable behavior by participants include:
* The use of sexualized language or imagery and unwelcome sexual attention or
advances
* Trolling, insulting/derogatory comments, and personal or political attacks
* Public or private harassment
* Publishing others' private information, such as a physical or electronic
address, without explicit permission
* Other conduct which could reasonably be considered inappropriate in a
professional setting
## Our Responsibilities
Project maintainers are responsible for clarifying the standards of acceptable
behavior and are expected to take appropriate and fair corrective action in
response to any instances of unacceptable behavior.
Project maintainers have the right and responsibility to remove, edit, or
reject comments, commits, code, wiki edits, issues, and other contributions
that are not aligned to this Code of Conduct, or to ban temporarily or
permanently any contributor for other behaviors that they deem inappropriate,
threatening, offensive, or harmful.
## Scope
This Code of Conduct applies both within project spaces and in public spaces
when an individual is representing the project or its community. Examples of
representing a project or community include using an official project e-mail
address, posting via an official social media account, or acting as an appointed
representative at an online or offline event. Representation of a project may be
further defined and clarified by project maintainers.
## Enforcement
Instances of abusive, harassing, or otherwise unacceptable behavior may be
reported by contacting the project team at conduct@zephyrproject.org.
Reports will be received by Kate Stewart (Linux Foundation) and Amy Occhialino
(Intel). All complaints will be reviewed and investigated, and will result in a
response that is deemed necessary and appropriate to the circumstances. The
project team is obligated to maintain confidentiality with regard to the
reporter of an incident. Further details of specific enforcement policies may
be posted separately.
Project maintainers who do not follow or enforce the Code of Conduct in good
faith may face temporary or permanent repercussions as determined by other
members of the project's leadership.
## Attribution
This Code of Conduct is adapted from the [Contributor Covenant][homepage], version 1.4,
available at https://www.contributor-covenant.org/version/1/4/code-of-conduct.html
[homepage]: https://www.contributor-covenant.org
For answers to common questions about this code of conduct, see
https://www.contributor-covenant.org/faq

View File

@@ -1,39 +0,0 @@
Contribution Guidelines
#######################
As an open-source project, we welcome and encourage the community to submit
patches directly to the project. In our collaborative open source environment,
standards and methods for submitting changes help reduce the chaos that can result
from an active development community.
This document briefly summarizes the full `Contribution
Guidelines <http://docs.zephyrproject.org/latest/contribute/index.html>`_
documentation.
* Zephyr uses the permissive open source `Apache 2.0 license`_
that allows you to freely use, modify, distribute and sell your own products
that include Apache 2.0 licensed software.
* There are some imported or reused components of the Zephyr project that
use other licensing and are clearly identified.
* The Developer Certificate of Origin (DCO) process is followed to
ensure developers are following licensing criteria for their
contributions, and documented with a ``Signed-off-by`` line in commits.
* Zephyr development workflow is supported on Linux, macOS, and Windows,
(with a few exceptions).
* Source code for the project is maintained in the GitHub repo:
https://github.com/zephyrproject-rtos/zephyr
* Issue and feature tracking is done using GitHub issues in this repo.
* A Continuous Integration (CI) system runs on every Pull Request (PR)
to verify several aspects of the PR including Git commit formatting,
Coding Style, sanity checks builds, and documentation builds.
* The `Zephyr devel mailing list`_ is a great place to engage with the
community, ask questions, discuss issues, and help each other.
.. _Zephyr devel mailing list: https://lists.zephyrproject.org/g/devel

123
Kbuild Normal file
View File

@@ -0,0 +1,123 @@
# vim: filetype=make
ifneq ("$(wildcard $(MDEF_FILE))","")
MDEF_FILE_PATH=$(strip $(MDEF_FILE))
else
ifneq ($(MDEF_FILE),)
MDEF_FILE_PATH=$(strip $(PROJECT_BASE)/$(MDEF_FILE))
endif
endif
ifeq (${CONFIG_NUM_COMMAND_PACKETS},)
CONFIG_NUM_COMMAND_PACKETS=0
endif
ifeq (${CONFIG_NUM_TIMER_PACKETS},)
CONFIG_NUM_TIMER_PACKETS=0
endif
ifeq (${CONFIG_NUM_TASK_PRIORITIES},)
CONFIG_NUM_TASK_PRIORITIES=$(CONFIG_NUM_PREEMPT_PRIORITIES)
endif
ifeq ($(ARCH),x86)
TASKGROUP_SSE=" TASKGROUP SSE"
endif
define filechk_prj.mdef
(echo "% WARNING. THIS FILE IS AUTO-GENERATED. DO NOT MODIFY!"; \
echo; \
echo "% CONFIG NUM_COMMAND_PACKETS NUM_TIMER_PACKETS NUM_TASK_PRIORITIES"; \
echo "% ============================================================="; \
echo " CONFIG ${CONFIG_NUM_COMMAND_PACKETS} ${CONFIG_NUM_TIMER_PACKETS} ${CONFIG_NUM_TASK_PRIORITIES}"; \
echo; \
echo "% TASKGROUP NAME";\
echo "% ==============";\
echo " TASKGROUP EXE";\
echo " TASKGROUP SYS";\
echo " TASKGROUP FPU_LEGACY";\
echo $(TASKGROUP_SSE);\
echo; \
if test -e "$(MDEF_FILE_PATH)"; then \
cat $(MDEF_FILE_PATH); \
fi;)
endef
misc/generated/sysgen/prj.mdef: $(MDEF_FILE_PATH) \
include/config/auto.conf FORCE
$(call filechk,prj.mdef)
sysgen_cmd=$(strip \
$(PYTHON) $(srctree)/scripts/sysgen \
-i $(CURDIR)/misc/generated/sysgen/prj.mdef \
-o $(CURDIR)/misc/generated/sysgen/ \
)
misc/generated/sysgen/kernel_main.c: misc/generated/sysgen/prj.mdef \
$(srctree)/scripts/sysgen
$(Q)$(sysgen_cmd)
define filechk_configs.c
(echo "/* file is auto-generated, do not modify ! */"; \
echo; \
echo "#include <toolchain.h>"; \
echo; \
echo "GEN_ABS_SYM_BEGIN (_ConfigAbsSyms)"; \
echo; \
cat $(CURDIR)/include/generated/autoconf.h | sed \
's/".*"/1/' | awk \
'/#define/{printf "GEN_ABSOLUTE_SYM(%s, %s);\n", $$2, $$3}'; \
echo; \
echo "GEN_ABS_SYM_END";)
endef
misc/generated/configs.c: include/config/auto.conf FORCE
$(call filechk,configs.c)
targets := misc/generated/configs.c
targets += include/generated/offsets.h
always := misc/generated/configs.c
always += include/generated/offsets.h
ifeq ($(CONFIG_MDEF),y)
targets += misc/generated/sysgen/kernel_main.c
always += misc/generated/sysgen/kernel_main.c
endif
define rule_cc_o_c_1
$(call echo-cmd,cc_o_c_1) $(cmd_cc_o_c_1);
endef
OFFSETS_INCLUDE = $(strip \
-include $(CURDIR)/include/generated/autoconf.h \
-I $(srctree)/include \
-I $(CURDIR)/include/generated \
-I $(srctree)/kernel/unified/include \
$(OFFSETS_INCLUDE_KERNEL_LOCATION) \
-I $(srctree)/lib/libc/minimal/include \
-I $(srctree)/arch/${ARCH}/include )
cmd_cc_o_c_1 = $(CC) $(KBUILD_CFLAGS) $(OFFSETS_INCLUDE) -c -o $@ $<
arch/$(ARCH)/core/offsets/offsets.o: arch/$(ARCH)/core/offsets/offsets.c $(KCONFIG_CONFIG)
$(Q)mkdir -p $(dir $@)
$(call if_changed,cc_o_c_1)
define offsetchk
$(Q)set -e; \
$(kecho) ' CHK $@'; \
mkdir -p $(dir $@); \
$(GENOFFSET_H) -i $(1) -o $@.tmp; \
if [ -r $@ ] && cmp -s $@ $@.tmp; then \
rm -f $@.tmp; \
else \
$(kecho) ' UPD $@'; \
mv -f $@.tmp $@; \
fi
endef
include/generated/offsets.h: arch/$(ARCH)/core/offsets/offsets.o \
include/config/auto.conf FORCE
$(call offsetchk,arch/$(ARCH)/core/offsets/offsets.o)

18
Kconfig
View File

@@ -1,8 +1,20 @@
# General configuration options
# Kconfig - general configuration options
#
# Copyright (c) 2014-2015 Wind River Systems, Inc.
# SPDX-License-Identifier: Apache-2.0
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#
mainmenu "Zephyr Kernel Configuration"
source "Kconfig.zephyr"

View File

@@ -1,458 +1,49 @@
# General configuration options
# Kconfig - general configuration options
#
# Copyright (c) 2014-2015 Wind River Systems, Inc.
# Copyright (c) 2016 Intel Corporation
# SPDX-License-Identifier: Apache-2.0
menu "Modules"
source "$(CMAKE_BINARY_DIR)/Kconfig.modules"
source "modules/Kconfig"
endmenu
# Include Kconfig.defconfig files first so that they can override defaults and
# other symbol/choice properties by adding extra symbol/choice definitions.
# After merging all definitions for a symbol/choice, Kconfig picks the first
# property (e.g. the first default) with a satisfied condition.
#
# Shield defaults should have precedence over board defaults, which should have
# precedence over SoC defaults, so include them in that order.
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# $ARCH and $BOARD_DIR will be glob patterns when building documentation.
source "boards/shields/*/Kconfig.defconfig"
source "$(BOARD_DIR)/Kconfig.defconfig"
source "$(SOC_DIR)/$(ARCH)/*/Kconfig.defconfig"
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#
config KERNELVERSION
string
option env="KERNELVERSION"
source "boards/Kconfig"
source "$(SOC_DIR)/Kconfig"
source "arch/Kconfig"
source "kernel/Kconfig"
source "dts/Kconfig"
source "drivers/Kconfig"
source "net/Kconfig"
source "misc/Kconfig"
source "lib/Kconfig"
source "subsys/Kconfig"
osource "$(TOOLCHAIN_KCONFIG_DIR)/Kconfig"
menu "Build and Link Features"
menu "Linker Options"
choice
prompt "Linker Orphan Section Handling"
default LINKER_ORPHAN_SECTION_WARN
config LINKER_ORPHAN_SECTION_PLACE
bool "Place"
help
Linker puts orphan sections in place without warnings
or errors.
config LINKER_ORPHAN_SECTION_WARN
bool "Warn"
help
Linker places the orphan sections in output and issues
warning about those sections.
config LINKER_ORPHAN_SECTION_ERROR
bool "Error"
help
Linker exits with error when an orphan section is found.
endchoice
config CODE_DATA_RELOCATION
bool "Relocate code/data sections"
depends on ARM
help
When selected this will relocate .text, data and .bss sections from
the specified files and places it in the required memory region. The
files should be specified in the CMakeList.txt file with
a cmake API zephyr_code_relocate().
config HAS_FLASH_LOAD_OFFSET
bool
help
This option is selected by targets having a FLASH_LOAD_OFFSET
and FLASH_LOAD_SIZE.
if HAS_FLASH_LOAD_OFFSET
config USE_DT_CODE_PARTITION
bool "Link application into /chosen/zephyr,code-partition from devicetree"
help
When enabled, the application will be linked into the flash partition
selected by the zephyr,code-partition property in /chosen in devicetree.
When this is disabled, the flash load offset and size can be set manually
below.
# Workaround for not being able to have commas in macro arguments
DT_CHOSEN_Z_CODE_PARTITION := zephyr,code-partition
config FLASH_LOAD_OFFSET
# Only user-configurable when USE_DT_CODE_PARTITION is disabled
hex "Kernel load offset" if !USE_DT_CODE_PARTITION
default $(dt_chosen_reg_addr_hex,$(DT_CHOSEN_Z_CODE_PARTITION)) if USE_DT_CODE_PARTITION
default 0
help
This option specifies the byte offset from the beginning of flash that
the kernel should be loaded into. Changing this value from zero will
affect the Zephyr image's link, and will decrease the total amount of
flash available for use by application code.
If unsure, leave at the default value 0.
config FLASH_LOAD_SIZE
# Only user-configurable when USE_DT_CODE_PARTITION is disabled
hex "Kernel load size" if !USE_DT_CODE_PARTITION
default $(dt_chosen_reg_size_hex,$(DT_CHOSEN_Z_CODE_PARTITION)) if USE_DT_CODE_PARTITION
default 0
help
If non-zero, this option specifies the size, in bytes, of the flash
area that the Zephyr image will be allowed to occupy. If zero, the
image will be able to occupy from the FLASH_LOAD_OFFSET to the end of
the device.
If unsure, leave at the default value 0.
endif # HAS_FLASH_LOAD_OFFSET
config TEXT_SECTION_OFFSET
hex
prompt "TEXT section offset" if !BOOTLOADER_MCUBOOT
default 0x200 if BOOTLOADER_MCUBOOT
default 0
help
If the application is built for chain-loading by a bootloader this
variable is required to be set to value that leaves sufficient
space between the beginning of the image and the start of the .text
section to store an image header or any other metadata.
In the particular case of the MCUboot bootloader this reserves enough
space to store the image header, which should also meet vector table
alignment requirements on most ARM targets, although some targets
may require smaller or larger values.
config HAVE_CUSTOM_LINKER_SCRIPT
bool "Custom linker scripts provided"
help
Set this option if you have a custom linker script which needed to
be define in CUSTOM_LINKER_SCRIPT.
config CUSTOM_LINKER_SCRIPT
string "Path to custom linker script"
depends on HAVE_CUSTOM_LINKER_SCRIPT
help
Path to the linker script to be used instead of the one define by the
board.
The linker script must be based on a version provided by Zephyr since
the kernel can expect a certain layout/certain regions.
This is useful when an application needs to add sections into the
linker script and avoid having to change the script provided by
Zephyr.
config CUSTOM_RODATA_LD
bool "(DEPRECATED) Include custom-rodata.ld"
help
Note: This is deprecated, use Cmake function zephyr_linker_sources() instead.
Include a customized linker script fragment for inserting additional
data and linker directives into the rodata section.
config CUSTOM_RWDATA_LD
bool "(DEPRECATED) Include custom-rwdata.ld"
help
Note: This is deprecated, use Cmake function zephyr_linker_sources() instead.
Include a customized linker script fragment for inserting additional
data and linker directives into the data section.
config CUSTOM_SECTIONS_LD
bool "(DEPRECATED) Include custom-sections.ld"
help
Note: This is deprecated, use Cmake function zephyr_linker_sources() instead.
Include a customized linker script fragment for inserting additional
arbitrary sections.
config KERNEL_ENTRY
string "Kernel entry symbol"
default "__start"
help
Code entry symbol, to be set at linking phase.
config LINKER_SORT_BY_ALIGNMENT
bool "Sort input sections by alignment"
default y
help
This turns on the linker flag to sort sections by alignment
in decreasing size of symbols. This helps to minimize
padding between symbols.
endmenu
menu "Compiler Options"
config NATIVE_APPLICATION
bool "Build as a native host application"
help
Build as a native application that can run on the host and using
resources and libraries provided by the host.
choice
prompt "Optimization level"
default NO_OPTIMIZATIONS if COVERAGE
default DEBUG_OPTIMIZATIONS if DEBUG
default SIZE_OPTIMIZATIONS
help
Note that these flags shall only control the compiler
optimization level, and that no extra debug code shall be
conditionally compiled based on them.
config SIZE_OPTIMIZATIONS
bool "Optimize for size"
help
Compiler optimizations will be set to -Os independently of other
options.
config SPEED_OPTIMIZATIONS
bool "Optimize for speed"
help
Compiler optimizations will be set to -O2 independently of other
options.
config DEBUG_OPTIMIZATIONS
bool "Optimize debugging experience"
help
Compiler optimizations will be set to -Og independently of other
options.
config NO_OPTIMIZATIONS
bool "Optimize nothing"
help
Compiler optimizations will be set to -O0 independently of other
options.
endchoice
config COMPILER_OPT
string "Custom compiler options"
help
This option is a free-form string that is passed to the compiler
when building all parts of a project (i.e. kernel).
The compiler options specified by this string supplement the
predefined set of compiler supplied by the build system,
and can be used to change compiler optimization, warning and error
messages, and so on.
endmenu
choice
prompt "Error checking behavior for CHECK macro"
default RUNTIME_ERROR_CHECKS
config ASSERT_ON_ERRORS
bool "Assert on all errors"
help
Assert on errors covered with the CHECK macro.
config NO_RUNTIME_CHECKS
bool "No runtime error checks"
help
Do not do any runtime checks or asserts when using the CHECK macro.
config RUNTIME_ERROR_CHECKS
bool "Enable runtime error checks"
help
Always perform runtime checks covered with the CHECK macro. This
option is the default and the only option used during testing.
endchoice
menu "Build Options"
config KERNEL_BIN_NAME
string "The kernel binary name"
default "zephyr"
help
This option sets the name of the generated kernel binary.
config OUTPUT_STAT
bool "Create a statistics file"
default y
help
Create a stat file using readelf -e <elf>
config OUTPUT_DISASSEMBLY
bool "Create a disassembly file"
default y
help
Create an .lst file with the assembly listing of the firmware.
config OUTPUT_DISASSEMBLE_ALL
bool "Disassemble all sections with source. Fill zeros."
default n
depends on OUTPUT_DISASSEMBLY
help
The .lst file will contain complete disassembly of the firmware
not just those expected to contain instructions including zeros
config OUTPUT_PRINT_MEMORY_USAGE
bool "Print memory usage to stdout"
default y
help
If the toolchain supports it, this option will pass
--print-memory-region to the linker when it is doing it's first
linker pass. Note that the memory regions are symbolic concepts
defined by the linker scripts and do not necessarily map
directly to the real physical address space. Take also note that
some platforms do two passes of the linker so the results do not
match exactly to the final elf file. See also rom_report,
ram_report and
https://sourceware.org/binutils/docs/ld/MEMORY.html
config BUILD_OUTPUT_HEX
bool "Build a binary in HEX format"
help
Build a binary in HEX format. This will build a zephyr.hex file need
by some platforms.
config BUILD_OUTPUT_BIN
bool "Build a binary in BIN format"
default y
help
Build a binary in BIN format. This will build a zephyr.bin file need
by some platforms.
config BUILD_OUTPUT_EXE
bool "Build a binary in ELF format with .exe extension"
help
Build a binary in ELF format that can run in the host system. This
will build a zephyr.exe file.
config BUILD_OUTPUT_S19
bool "Build a binary in S19 format"
help
Build a binary in S19 format. This will build a zephyr.s19 file need
by some platforms.
config BUILD_NO_GAP_FILL
bool "Don't fill gaps in generated hex/bin/s19 files."
config BUILD_OUTPUT_STRIPPED
bool "Build a stripped binary"
help
Build a stripped binary. This will build a zephyr.stripped file need
by some platforms.
config APPLICATION_DEFINED_SYSCALL
bool "Scan application folder for any syscall definition"
help
Scan additional folders inside application source folder
for application defined syscalls.
config MAKEFILE_EXPORTS
bool "Generate build metadata files named Makefile.exports"
help
Generates a file with build information that can be read by
third party Makefile-based build systems.
config LEGACY_DEVICETREE_MACROS
bool "Allow use of legacy devicetree macros"
help
Allows use of legacy devicetree macros which were used in
Zephyr 2.2 and previous versions, rather than the devicetree.h
API introduced during the Zephyr 2.3 development cycle.
endmenu
endmenu
menu "Boot Options"
config IS_BOOTLOADER
bool "Act as a bootloader"
depends on XIP
depends on ARM
help
This option indicates that Zephyr will act as a bootloader to execute
a separate Zephyr image payload.
config BOOTLOADER_SRAM_SIZE
int "SRAM reserved for bootloader"
default 16
depends on !XIP || IS_BOOTLOADER
depends on ARM || XTENSA
help
This option specifies the amount of SRAM (measure in kB) reserved for
a bootloader image, when either:
- the Zephyr image itself is to act as the bootloader, or
- Zephyr is a !XIP image, which implicitly assumes existence of a
bootloader that loads the Zephyr !XIP image onto SRAM.
config BOOTLOADER_MCUBOOT
bool "MCUboot bootloader support"
select USE_DT_CODE_PARTITION
help
This option signifies that the target uses MCUboot as a bootloader,
or in other words that the image is to be chain-loaded by MCUboot.
This sets several required build system and Device Tree options in
order for the image generated to be bootable using the MCUboot open
source bootloader. Currently this includes:
* Setting TEXT_SECTION_OFFSET to a default value that allows space
for the MCUboot image header
* Activating SW_VECTOR_RELAY on Cortex-M0 (or Armv8-M baseline)
targets with no built-in vector relocation mechanisms
config BOOTLOADER_ESP_IDF
bool "ESP-IDF bootloader support"
depends on SOC_ESP32
help
This option will trigger the compilation of the ESP-IDF bootloader
inside the build folder.
At flash time, the bootloader will be flashed with the zephyr image
config BOOTLOADER_KEXEC
bool "Boot using Linux kexec() system call"
depends on X86
help
This option signifies that Linux boots the kernel using kexec system call
and utility. This method is used to boot the kernel over the network.
config BOOTLOADER_CONTEXT_RESTORE
bool "Boot loader has context restore support"
default y
depends on SYS_POWER_DEEP_SLEEP_STATES && BOOTLOADER_CONTEXT_RESTORE_SUPPORTED
help
This option signifies that the target has a bootloader
that restores CPU context upon resuming from deep sleep
power state.
config REBOOT
bool "Reboot functionality"
select SYSTEM_CLOCK_DISABLE
help
Enable the sys_reboot() API. Enabling this can drag in other subsystems
needed to perform a "safe" reboot (e.g. SYSTEM_CLOCK_DISABLE, to stop the
system clock before issuing a reset).
config MISRA_SANE
bool "MISRA standards compliance features"
help
Causes the source code to build in "MISRA" mode, which
disallows some otherwise-permitted features of the C
standard for safety reasons. Specifically variable length
arrays are not permitted (and gcc will enforce this).
endmenu
menu "Compatibility"
config COMPAT_INCLUDES
bool "Suppress warnings when using header shims"
default y
help
Suppress any warnings from the pre-processor when including
deprecated header files.
endmenu
source "ext/Kconfig"
source "tests/Kconfig"
#
# The following are for Kconfig files for default values only.
# These should be parsed at the end.
#
# Board defaults should be parsed after SoC defaults
# because board usually overrides SoC values.
#
source "arch/*/soc/*/Kconfig.defconfig"
source "boards/*/*/Kconfig.defconfig"

66
LICENSING.rst Normal file
View File

@@ -0,0 +1,66 @@
.. _zephyr_licensing:
Licensing of Zephyr Project components
######################################
The Zephyr kernel tree imports or reuses packages, scripts and other
files that are not covered by the :download:`Apache License
<../LICENSE>`. In some places there is no LICENSE file or way to put
a LICENSE file there, so we describe the licensing in this document.
- *kconfig* and *kbuild*
*Origin:* Linux Kernel
*Licensing:* *GPLv2*
- *scripts/{checkpatch.pl,checkstack.pl,get_maintainers.pl,spelling.txt}*
*Origin:* Linux Kernel
*Licensing:* *GPLv2*
- *ext/fs/fat/*
*Origin:* FatFs is a file system based on the FAT file system specification. This is
provided by ELM Chan http://elm-chan.org/fsw/ff/00index_e.html
*Licensing*:
Copyright (C) 2016, ChaN, all right reserved.
FatFs module is an open source software. Redistribution and use of FatFs in
source and binary forms, with or without modification, are permitted provided
that the following condition is met:
1. Redistributions of source code must retain the above copyright notice,
this condition and the following disclaimer.
This software is provided by the copyright holder and contributors "AS IS"
and any warranties related to this software are DISCLAIMED.
The copyright owner or contributors be NOT LIABLE for any damages caused
by use of this software.
- *ext/hal/cmsis/*
*Origin:* https://github.com/ARM-software/CMSIS.git
*Licensing*: :download:`CMSIS_END_USER_LICENCE_AGREEMENT <../ext/hal/cmsis/CMSIS_END_USER_LICENCE_AGREEMENT.pdf>`
- *ext/hal/ksdk/*
*Origin:* http://kex.nxp.com
*Licensing*: 3-clause BSD (see :download:`source
<../ext/hal/ksdk/drivers/fsl_rtc.h>`)
- *ext/hal/nordic/*
*Origin:*
*Licensing*: 3-clause BSD (see :download:`source <../ext/hal/nordic/mdk/nrf51.h>`)
- *ext/hal/qmsi/*
*Origin:* https://github.com/quark-mcu/qmsi/releases
*Licensing*: 3-clause BSD (see :download:`source <../ext/hal/qmsi/include/qm_common.h>`)

458
MAINTAINERS Normal file
View File

@@ -0,0 +1,458 @@
Originally from the Linux Kernel.
# Licensed under the terms of the GNU GPL License version 2
Descriptions of section entries:
P: Person (obsolete)
M: Mail patches to: FullName <address@domain>
R: Designated reviewer: FullName <address@domain>
These reviewers should be CCed on patches.
L: Mailing list that is relevant to this area
W: Web-page with status/info
Q: Patchwork web based patch tracking system site
T: SCM tree type and location.
Type is one of: git, hg, quilt, stgit, topgit
S: Status, one of the following:
Supported: Someone is actually paid to look after this.
Maintained: Someone actually looks after it.
Odd Fixes: It has a maintainer but they don't have time to do
much other than throw the odd patch in. See below..
Orphan: No current maintainer [but maybe you could take the
role as you write your new code].
Obsolete: Old code. Something tagged obsolete generally means
it has been replaced by a better system and you
should be using that.
F: Files and directories with wildcard patterns.
A trailing slash includes all files and subdirectory files.
F: drivers/net/ all files in and below drivers/net
F: drivers/net/* all files in drivers/net, but not below
F: */net/* all files in "any top level directory"/net
One pattern per line. Multiple F: lines acceptable.
N: Files and directories with regex patterns.
N: [^a-z]tegra all files whose path contains the word tegra
One pattern per line. Multiple N: lines acceptable.
scripts/get_maintainer.pl has different behavior for files that
match F: pattern and matches of N: patterns. By default,
get_maintainer will not look at git log history when an F: pattern
match occurs. When an N: match occurs, git log history is used
to also notify the people that have git commit signatures.
X: Files and directories that are NOT maintained, same rules as F:
Files exclusions are tested before file matches.
Can be useful for excluding a specific subdirectory, for instance:
F: net/
X: net/ipv6/
matches all files in and below net excluding net/ipv6/
K: Keyword perl extended regex pattern to match content in a
patch or file. For instance:
K: of_get_profile
matches patches or files that contain "of_get_profile"
K: \b(printk|pr_(info|err))\b
matches patches or files that contain one or more of the words
printk, pr_info or pr_err
One regex pattern per line. Multiple K: lines acceptable.
Note: For the hard of thinking, this list is meant to remain in alphabetical
order. If you could add yourselves to it in alphabetical order that would be
so much easier [Ed]
Maintainers List (try to look for most precise areas first)
-----------------------------------
ARC ARCHITECTURE
M: Ruud Derwig <Ruud.Derwig@synopsys.com>
M: Chuck Jordan <Chuck.Jordan@synopsys.com>
M: Benjamin Walsh <benjamin.walsh@windriver.com>
S: Supported
F: arch/arc/
F: include/arch/arc/
F: boards/arc/
ARM ARCHITECTURE
M: Maureen Helm <maureen.helm@nxp.com>
M: Kumar Gala <kumar.gala@linaro.org>
S: Supported
F: arch/arm/
F: include/arch/arm/
F: boards/arm/
ARM CORTEX MICROCONTROLLER SOFTWARE INTERFACE STANDARD (CMSIS)
M: Maureen Helm <maureen.helm@nxp.com>
M: Kumar Gala <kumar.gala@linaro.org>
S: Supported
F: ext/hal/cmsis/
BOARDS/ARC - ARDUINO 101 SSS
M: Anas Nashif <anas.nashif@intel.com>
S: Supported
F: boards/arc/arduino_101_sss/
BOARDS/ARC - EM Starterkit
M: Chuck Jordan <Chuck.Jordan@synopsys.com>
S: Supported
F: boards/arc/em_starterkit/
BOARDS/ARC - QUARK SE C1000 SS Devboard
M: Anas Nashif <anas.nashif@intel.com>
S: Supported
F: boards/arc/quark_se_c1000_ss_devboard/
BOARDS/ARM - 96Boards CARBON
M: Amit Kucheria <amit.kucheria@linaro.org>
M: Ricardo Salveti <ricardo.salveti@linaro.org>
S: Supported
F: boards/arm/96b_carbon/
BOARDS/ARM - 96Boards NITROGEN
M: Amit Kucheria <amit.kucheria@linaro.org>
S: Supported
F: boards/arm/96b_nitrogen/
BOARDS/ARM - ARDUINO 101 BLE
M: Johan Hedberg <johan.hedberg@intel.com>
S: Supported
F: boards/arm/arduino_101_ble/
BOARDS/ARM - CC3200 LAUNCHXL
M: Gil Pitney <gil.pitney@linaro.org>
S: Supported
F: boards/arm/cc3200_launchxl/
BOARDS/ARM - NXP FRDM-K64F
M: Maureen Helm <maureen.helm@nxp.com>
S: Supported
F: boards/arm/frdm_k64f/
BOARDS/ARM - NXP Hexiwear
M: Maureen Helm <maureen.helm@nxp.com>
S: Supported
F: boards/arm/hexiwear_k64/
BOARDS/ARM - NORDIC NRF51 REDBEAR BLENANO
M: Ricardo Salveti <ricardo.salveti@linaro.org>
S: Supported
F: boards/arm/nrf51_blenano/
BOARDS/ARM - NORDIC NRF52 PCA10040
M: Carles Cufi <carles.cufi@nordicsemi.no>
S: Supported
F: boards/arm/nrf52_pca10040/
BOARDS/ARM - NUCLEO-64 F401RE Devboard
M: Amit Kucheria <amit.kucheria@linaro.org>
M: Ricardo Salveti <ricardo.salveti@linaro.org>
S: Supported
F: boards/arm/nucleo_f401re/
BOARDS/ARM - ARM LTD V2M Beetle
M: Vincenzo Frascino <vincenzo.frascino@linaro.org>
S: Supported
F: boards/arm/v2m_beetle/
BOARDS/NIOS2 - ALTERA MAX10
M: Andrew Boie <andrew.p.boie@intel.com>
S: Supported
F: boards/nios2/altera_max10/
BOARDS/X86 - ARDUINO 101
M: Anas Nashif <anas.nashif@intel.com>
S: Supported
F: boards/x86/arduino_101/
BOARDS/X86 - Galileo
M: Anas Nashif <anas.nashif@intel.com>
S: Supported
F: boards/x86/galileo/
BOARDS/X86 - QUARK D2000 Devboard
M: Anas Nashif <anas.nashif@intel.com>
S: Supported
F: boards/x86/quark_d2000/
BOARDS/X86 - QUARK SE C1000 Devboard
M: Anas Nashif <anas.nashif@intel.com>
S: Supported
F: boards/x86/quark_se_c1000/
BLUETOOTH
M: Johan Hedberg <johan.hedberg@intel.com>
M: Luiz Augusto von Dentz <luiz.dentz@gmail.com>
M: Szymon Janc <szymon.janc@gmail.com>
S: Supported
W: https://www.zephyrproject.org/doc/subsystems/bluetooth/bluetooth.html
F: subsys/bluetooth/
F: include/bluetooth/
F: drivers/bluetooth/
F: samples/bluetooth/
F: tests/bluetooth/
F: doc/subsystems/bluetooth/
BLUETOOTH CONTROLLER
M: Vinayak Chettimada <vinayak.kariappa.chettimada@nordicsemi.no>
M: Carles Cufi <carles.cufi@nordicsemi.no>
S: Supported
F: subsys/bluetooth/controller/
CC3200 SDK
M: Gil Pitney <gil.pitney@linaro.org>
S: Supported
F: ext/hal/cc3200sdk/
CC32XX SOC - TI SIMPLELINK
M: Gil Pitney <gil.pitney@linaro.org>
S: Supported
F: arch/arm/soc/ti_simplelink/
DOCUMENTATION
M: Kinder, David <david.b.kinder@intel.com>
M: Perez-Gonzalez, Inaky <inaky.perez-gonzalez@intel.com>
S: Supported
F: doc/
FILE SYSTEM
M: Ramesh Thomas <ramesh.thomas@intel.com>
M: Kuo-Lang Tseng <kuo-lang.tseng@intel.com>
S: Supported
F: ext/fs/
F: subsys/fs/
F: include/fs/
F: include/fs.h
F: samples/fs/
FLASH DRIVER
M: Baohong Liu <baohong.liu@intel.com>
M: Kuo-Lang Tseng <kuo-lang.tseng@intel.com>
S: Supported
F: drivers/flash/
INTERRUPTS
M: Andrew Boie <andrew.p.boie@intel.com>
S: Supported
F: drivers/interrupt_controller/
F: arch/arc/core/
F: arch/arm/core/
F: arch/nios2/core/
F: arch/x86/core/
F: include/irq.h
F: include/arch/x86/arch.h
F: include/arch/arm/cortex_m/irq.h
F: include/arch/nios2/arch.h
F: include/arch/arc/arch.h
F: include/arch/arc/v2/irq.h
F: include/drivers/loapic.h
F: include/drivers/ioapic.h
F: include/drivers/mvic.h
KERNEL CORE
M: Benjamin Walsh <benjamin.walsh@windriver.com>
M: Allan Stephens <allan.stephens@windriver.com>
S: Supported
F: kernel/
F: include/nanokernel.h
F: include/microkernel.h
F: include/microkernel/
F: include/misc/
F: include/toolchain/
F: include/atomic.h
F: include/cache.h
F: include/init.h
F: include/irq.h
F: include/irq_offload.h
F: include/kernel_version.h
F: include/linker-defs.h
F: include/linker-tool-gcc.h
F: include/linker-tool.h
F: include/section_tags.h
F: include/sections.h
F: include/shared_irq.h
F: include/sw_isr_table.h
F: include/sys_clock.h
F: include/sys_io.h
F: include/toolchain.h
F: include/zephyr.h
KINETIS SOFTWARE DEVELOPMENT KIT (KSDK)
M: Maureen Helm <maureen.helm@nxp.com>
S: Supported
F: ext/hal/ksdk/
KNOWN ISSUES
M: Anas Nashif <anas.nashif@intel.com>
M: Inaky Perez-Gonzalez <inaky.perez-gonzalez@intel.com>
M: Javier B Perez <javier.b.perez.hernandez@intel.com>
F: .known-issues/
MAINTAINERS
M: Javier B Perez <javier.b.perez.hernandez@intel.com>
M: Anas Nashif <anas.nashif@intel.com>
M: Perez-Gonzalez, Inaky <inaky.perez-gonzalez@intel.com>
S: Supported
F: MAINTAINERS
MBEDTLS
M: Sergio Rodriguez <sergio.sf.rodriguez@intel.com>
M: Jithu Joseph <jithu.joseph@intel.com>
M: Kuo-Lang Tseng <kuo-lang.tseng@intel.com>
S: Supported
F: ext/lib/crypto/mbedtls/
F: samples/net/mbedtls_sslclient/
F: tests/crypto/test_mbedtls/
NETWORKING
M: Jukka Rissanen <jukka.rissanen@linux.intel.com>
S: Supported
W: https://www.zephyrproject.org/doc/subsystems/networking/networking.html
F: net/ip/
F: include/net/
F: samples/net/
F: tests/net/
NETWORK APPLICATIONS
M: Flavio Santes <flavio.santes@intel.com>
S: Supported
F: samples/net/dns_client/
F: samples/net/nats_clients/
F: samples/net/paho_mqtt_clients/
NETWORK BUFFERS
M: Johan Hedberg <johan.hedberg@intel.com>
M: Jukka Rissanen <jukka.rissanen@linux.intel.com>
S: Supported
W: https://www.zephyrproject.org/doc/subsystems/networking/buffers.html
F: net/buf.c
F: include/net/buf.h
F: tests/net/buf/
NIOS II
M: Andrew Boie <andrew.p.boie@intel.com>
S: Supported
F: arch/nios2/
F: include/arch/nios2/
F: drivers/serial/uart_altera_jtag.c
F: drivers/timer/altera_avalon_timer.c
F: tests/kernel/test_intmath/
F: boards/nios2/
NORDIC MDK
M: Carles Cufi <carles.cufi@nordicsemi.no>
S: Supported
F: ext/hal/nordic/mdk/
POWER MANAGEMENT
M: Ramesh Thomas <ramesh.thomas@intel.com>
M: Kuo-Lang Tseng <kuo-lang.tseng@intel.com>
S: Supported
F: arch/x86/core/crt0.S
F: include/device.h
F: include/init.h
F: include/power.h
F: kernel/microkernel/k_idle.c
F: kernel/nanokernel/device.c
F: samples/power/
QMSI
M: Anas Nashif <anas.nashif@intel.com>
S: Supported
F: ext/hal/qmsi/
QMSI DRIVERS
M: Sergio Rodriguez <sergio.sf.rodriguez@intel.com>
M: Baohong Liu <baohong.liu@intel.com>
M: Kuo-Lang Tseng <kuo-lang.tseng@intel.com>
S: Supported
F: drivers/*/*qmsi*
F: drivers/*/*/*qmsi*
QUARK D2000 SOC
M: Anas Nashif <anas.nashif@intel.com>
S: Supported
F: arch/x86/soc/intel_quark/quark_d2000/
QUARK SE C1000 SOC
M: Anas Nashif <anas.nashif@intel.com>
S: Supported
F: arch/x86/soc/intel_quark/quark_se_c1000/
QUARK X1000 SOC
M: Anas Nashif <anas.nashif@intel.com>
S: Supported
F: arch/x86/soc/intel_quark/quark_x1000/
RELEASE NOTES
M: Anas Nashif <anas.nashif@intel.com>
M: Javier B Perez <javier.b.perez.hernandez@intel.com>
M: Kinder, David <david.b.kinder@intel.com>
S: Supported
F: release-notes.rst
SANITYCHECK
M: Andrew Boie <andrew.p.boie@intel.com>
S: Supported
F: scripts/sanitycheck
F: scripts/expr_parser.py
F: scripts/sanity_chk/
SENSOR DRIVERS
M: Bogdan Davidoaia <bogdan.m.davidoaia@intel.com>
M: Laurentiu Palcu <laurentiu.palcu@intel.com>
M: Murtaza Alexandru <alexandru.murtaza@intel.com>
M: Vlad Dogaru <vlad.dogaru@intel.com>
S: Supported
W: https://www.zephyrproject.org/doc/subsystems/sensor.html
F: include/sensor.h
F: drivers/sensor/
F: samples/sensor/
STM32CUBE SDK
M: Erwan Gouriou <erwan.gouriou@linaro.org>
S: Supported
F: ext/hal/st/stm32cube/
STM32F4X SoC FAMILY and DRIVERS
M: Amit Kucheria <amit.kucheria@linaro.org>
M: Ricardo Salveti <ricardo.salveti@linaro.org>
S: Supported
F: arch/arm/soc/st_stm32/stm32f4/
F: drivers/pinmux/stm32/
F: drivers/gpio/*stm32*
F: drivers/clock_control/*stm32f4*
TINYCRYPT
M: Constanza Heath <constanza.m.heath@intel.com>
M: Flavio Santes <flavio.santes@intel.com>
S: Supported
F: ext/lib/crypto/tinycrypt/
F: tests/crypto/
USB
M: Jithu Joseph <jithu.joseph@intel.com>
S: Supported
F: subsys/usb
F: drivers/usb
F: samples/usb
X86 ARCH
M: Benjamin Walsh <benjamin.walsh@windriver.com>
M: Allan Stephens <allan.stephens@windriver.com>
S: Supported
F: arch/x86/
F: include/arch/x86/
F: boards/x86/
ZOAP
M: Vinicius Costa Gomes <vinicius.gomes@intel.com>
S: Supported
F: lib/iot/zoap/
F: samples/net/zoap_client/
F: samples/net/zoap_server/
F: tests/net/zoap/
THE REST
M: Anas Nashif <anas.nashif@intel.com>
M: Kumar Gala <kumar.gala@linaro.org>
L: devel@lists.zephyrproject.com
T: git https://gerrit.zephyrproject.org/r/a/zephyr
S: Buried alive in reporters
F: *
F: */

1346
Makefile

File diff suppressed because it is too large Load Diff

149
Makefile.inc Normal file
View File

@@ -0,0 +1,149 @@
# vim: filetype=make
#
UNAME := $(shell uname)
ifeq (MINGW, $(findstring MINGW, $(UNAME)))
DQUOTE = '
# '
PROJECT_BASE ?= $(shell sh -c "pwd -W")
else
DQUOTE = "
# "
PROJECT_BASE ?= $(CURDIR)
endif
ifdef BOARD
KBUILD_DEFCONFIG_PATH=$(wildcard $(ZEPHYR_BASE)/boards/*/*/$(BOARD)_defconfig)
ifeq ($(KBUILD_DEFCONFIG_PATH),)
$(error Board $(BOARD) not found!)
endif
else
$(error BOARD is not defined!)
endif
# Choose a default output directory if one wasn't supplied. Note that
# PRISTINE_O depends on whether this is default or not. If building
# in-tree, we want to remove the whole outdir and not just the BOARD
# specified (thus "pristine"). Out of tree, we can obviously remove
# only what we were told to build.
ifndef O
PRISTINE_O = outdir
O = $(PROJECT_BASE)/outdir/$(BOARD)
else
PRISTINE_O = $(O)
endif
# Turn O into an absolute path; we call the main Kbuild with $(MAKE) -C
# which changes the working directory, relative paths don't work right.
# Need to create the directory first to make realpath happy
ifneq ($(MAKECMDGOALS),help)
$(shell mkdir -p $(O))
override O := $(realpath $(O))
endif
export ARCH MDEF_FILE QEMU_EXTRA_FLAGS PROJECT_BASE
override CONF_FILE := $(strip $(subst $(DQUOTE),,$(CONF_FILE)))
SOURCE_DIR ?= $(PROJECT_BASE)/src/
override SOURCE_DIR := $(realpath $(SOURCE_DIR))
override SOURCE_DIR := $(subst \,/,$(SOURCE_DIR))
override SOURCE_DIR_PARENT := $(patsubst %, %/.., $(SOURCE_DIR))
override SOURCE_DIR_PARENT := $(abspath $(SOURCE_DIR_PARENT))
override SOURCE_DIR_PARENT := $(subst \,/,$(SOURCE_DIR_PARENT))
export SOURCE_DIR SOURCE_DIR_PARENT
ifeq ("$(origin V)", "command line")
KBUILD_VERBOSE = $(V)
endif
ifndef KBUILD_VERBOSE
KBUILD_VERBOSE = 0
endif
ifeq ($(KBUILD_VERBOSE),1)
Q =
S =
else
Q = @
S = -s
endif
export CFLAGS
zephyrmake = +$(MAKE) -C $(ZEPHYR_BASE) O=$(1) \
PROJECT=$(PROJECT_BASE) SOURCE_DIR=$(DQUOTE)$(SOURCE_DIR)$(DQUOTE) $(2)
BOARDCONFIG = $(O)/.board_$(BOARD)
DOTCONFIG = $(O)/.config
all: $(DOTCONFIG)
$(Q)$(call zephyrmake,$(O),$@)
ifeq ($(findstring qemu_,$(BOARD)),)
qemu:
@echo "Emulation not available for this platform"
qemugdb: qemu
else
qemu: $(DOTCONFIG)
$(Q)$(call zephyrmake,$(O),$@)
qemugdb: $(DOTCONFIG)
$(Q)$(call zephyrmake,$(O),$@)
endif
debug: $(DOTCONFIG)
$(Q)$(call zephyrmake,$(O),$@)
flash: $(DOTCONFIG)
$(Q)$(call zephyrmake,$(O),$@)
ifeq ($(MAKECMDGOALS),debugserver)
ARCH = $(notdir $(subst /$(BOARD),,$(wildcard $(ZEPHYR_BASE)/boards/*/$(BOARD))))
-include $(ZEPHYR_BASE)/boards/$(ARCH)/$(BOARD)/Makefile.board
-include $(ZEPHYR_BASE)/scripts/Makefile.toolchain.$(ZEPHYR_GCC_VARIANT)
BOARD_NAME = $(BOARD)
export BOARD_NAME
endif
debugserver: FORCE
$(Q)$(CONFIG_SHELL) $(ZEPHYR_BASE)/scripts/support/$(FLASH_SCRIPT) debugserver
initconfig outputexports: $(DOTCONFIG)
$(BOARDCONFIG):
@rm -f $(O)/.board_*
@touch $@
ram_report: initconfig
$(Q)$(call zephyrmake,$(O),$@)
rom_report: initconfig
$(Q)$(call zephyrmake,$(O),$@)
menuconfig: initconfig
$(Q)$(call zephyrmake,$(O),$@)
help:
$(Q)$(MAKE) -s -C $(ZEPHYR_BASE) $@
# Catch all
%:
$(Q)$(call zephyrmake,$(O),$@)
KERNEL_CONFIG = $(ZEPHYR_BASE)/kernel/configs/unified.config
$(DOTCONFIG): $(BOARDCONFIG) $(KBUILD_DEFCONFIG_PATH) $(CONF_FILE)
$(Q)$(CONFIG_SHELL) $(ZEPHYR_BASE)/scripts/kconfig/merge_config.sh \
-q -m -O $(O) $(KBUILD_DEFCONFIG_PATH) $(KERNEL_CONFIG) $(CONF_FILE) \
$(wildcard $(O)/*.conf)
$(Q)$(MAKE) $(S) -C $(ZEPHYR_BASE) O=$(O) PROJECT=$(PROJECT_BASE) oldnoconfig
pristine:
$(Q)rm -rf $(PRISTINE_O)
PHONY += FORCE initconfig
FORCE:
.PHONY: $(PHONY)

View File

@@ -1,86 +0,0 @@
.. raw:: html
<a href="https://www.zephyrproject.org">
<p align="center">
<img src="doc/images/Zephyr-Project.png">
</p>
</a>
<a href="https://bestpractices.coreinfrastructure.org/projects/74"><img
src="https://bestpractices.coreinfrastructure.org/projects/74/badge"></a>
<img
src="https://api.shippable.com/projects/58ffb2b8baa5e307002e1d79/badge?branch=master">
The Zephyr Project is a scalable real-time operating system (RTOS) supporting
multiple hardware architectures, optimized for resource constrained devices,
and built with security in mind.
The Zephyr OS is based on a small-footprint kernel designed for use on
resource-constrained systems: from simple embedded environmental sensors and
LED wearables to sophisticated smart watches and IoT wireless gateways.
The Zephyr kernel supports multiple architectures, including ARM Cortex-M,
Intel x86, ARC, Nios II, Tensilica Xtensa, and RISC-V, and a large number of
`supported boards`_.
.. below included in doc/introduction/introduction.rst
.. start_include_here
Getting Started
***************
Welcome to Zephyr! See the `Introduction to Zephyr`_ for a high-level overview,
and the documentation's `Getting Started Guide`_ to start developing.
Community Support
*****************
Community support is provided via mailing lists and Slack; see the Resources
below for details.
.. _project-resources:
Resources
*********
Here's a quick summary of resources to help you find your way around:
* **Help**: `Asking for Help Tips`_
* **Documentation**: http://docs.zephyrproject.org (`Getting Started Guide`_)
* **Source Code**: https://github.com/zephyrproject-rtos/zephyr is the main
repository; https://elixir.bootlin.com/zephyr/latest/source contains a
searchable index
* **Releases**: https://github.com/zephyrproject-rtos/zephyr/releases
* **Samples and example code**: see `Sample and Demo Code Examples`_
* **Mailing Lists**: users@lists.zephyrproject.org and
devel@lists.zephyrproject.org are the main user and developer mailing lists,
respectively. You can join the developer's list and search its archives at
`Zephyr Development mailing list`_. The other `Zephyr mailing list
subgroups`_ have their own archives and sign-up pages.
* **Nightly CI Build Status**: https://lists.zephyrproject.org/g/builds
The builds@lists.zephyrproject.org mailing list archives the CI
(shippable) nightly build results.
* **Chat**: Zephyr's Slack workspace is https://zephyrproject.slack.com. Use
this `Slack Invite`_ to register.
* **Contributing**: see the `Contribution Guide`_
* **Wiki**: `Zephyr GitHub wiki`_
* **Issues**: https://github.com/zephyrproject-rtos/zephyr/issues
* **Security Issues**: Email vulnerabilities@zephyrproject.org to report
security issues; also see our `Security`_ documentation. Security issues are
tracked separately at https://zephyrprojectsec.atlassian.net.
* **Zephyr Project Website**: https://zephyrproject.org
.. _Slack Invite: https://tinyurl.com/y5glwylp
.. _supported boards: http://docs.zephyrproject.org/latest/boards/index.html
.. _Zephyr Documentation: http://docs.zephyrproject.org
.. _Introduction to Zephyr: http://docs.zephyrproject.org/latest/introduction/index.html
.. _Getting Started Guide: http://docs.zephyrproject.org/latest/getting_started/index.html
.. _Contribution Guide: http://docs.zephyrproject.org/latest/contribute/index.html
.. _Zephyr GitHub wiki: https://github.com/zephyrproject-rtos/zephyr/wiki
.. _Zephyr Development mailing list: https://lists.zephyrproject.org/g/devel
.. _Zephyr mailing list subgroups: https://lists.zephyrproject.org/g/main/subgroups
.. _Sample and Demo Code Examples: http://docs.zephyrproject.org/latest/samples/index.html
.. _Security: http://docs.zephyrproject.org/latest/security/index.html
.. _Asking for Help Tips: https://docs.zephyrproject.org/latest/guides/getting-help.html

View File

@@ -1,5 +0,0 @@
VERSION_MAJOR = 2
VERSION_MINOR = 3
PATCHLEVEL = 0
VERSION_TWEAK = 0
EXTRAVERSION =

View File

@@ -1,11 +0,0 @@
# SPDX-License-Identifier: Apache-2.0
add_definitions(-D__ZEPHYR_SUPERVISOR__)
include_directories(
${ZEPHYR_BASE}/kernel/include
${ZEPHYR_BASE}/arch/${ARCH}/include
)
add_subdirectory(common)
add_subdirectory(${ARCH_DIR}/${ARCH} arch/${ARCH})

View File

@@ -1,590 +1,112 @@
# General architecture configuration options
# Kconfig - general architecture configuration options
#
# Copyright (c) 2014-2015 Wind River Systems, Inc.
# Copyright (c) 2015 Intel Corporation
# Copyright (c) 2016 Cadence Design Systems, Inc.
# SPDX-License-Identifier: Apache-2.0
# Include these first so that any properties (e.g. defaults) below can be
# overridden (by defining symbols in multiple locations)
# Note: $ARCH might be a glob pattern
source "$(ARCH_DIR)/$(ARCH)/Kconfig"
# Architecture symbols
#
# Should be 'select'ed by low-level symbols like SOC_SERIES_* or, lacking that,
# by SOC_*.
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#
choice
prompt "Architecture"
default X86
config ARC
bool
select ARCH_IS_SET
select HAS_DTS
help
ARC architecture
bool "ARC architecture"
config ARM
bool
select ARCH_IS_SET
select HAS_DTS
# FIXME: current state of the code for all ARM requires this, but
# is really only necessary for Cortex-M with ARM MPU!
select GEN_PRIV_STACKS
help
ARM architecture
bool "ARM architecture"
config X86
bool
select ARCH_IS_SET
bool "x86 architecture"
select NANOKERNEL_TICKLESS_IDLE_SUPPORTED
select ATOMIC_OPERATIONS_BUILTIN
select HAS_DTS
select ARCH_HAS_CUSTOM_SWAP_TO_MAIN if !X86_64
help
x86 architecture
config NIOS2
bool
select ARCH_IS_SET
bool "Nios II Gen 2 architecture"
select ATOMIC_OPERATIONS_C
select HAS_DTS
help
Nios II Gen 2 architecture
config RISCV
bool
select ARCH_IS_SET
select HAS_DTS
help
RISCV architecture
config XTENSA
bool
select ARCH_IS_SET
select HAS_DTS
select USE_SWITCH
select USE_SWITCH_SUPPORTED
help
Xtensa architecture
config ARCH_POSIX
bool
select ARCH_IS_SET
select HAS_DTS
select ATOMIC_OPERATIONS_BUILTIN
select ARCH_HAS_CUSTOM_SWAP_TO_MAIN
select ARCH_HAS_CUSTOM_BUSY_WAIT
select ARCH_HAS_THREAD_ABORT
select NATIVE_APPLICATION
select HAS_COVERAGE_SUPPORT
help
POSIX (native) architecture
config ARCH_IS_SET
bool
help
Helper symbol to detect SoCs forgetting to select one of the arch
symbols above. See the top-level CMakeLists.txt.
menu "General Architecture Options"
module = ARCH
module-str = arch
source "subsys/logging/Kconfig.template.log_config"
module = MPU
module-str = mpu
source "subsys/logging/Kconfig.template.log_config"
config BIG_ENDIAN
bool
help
This option tells the build system that the target system is big-endian.
Little-endian architecture is the default and should leave this option
unselected. This option is selected by arch/$ARCH/Kconfig,
soc/**/Kconfig, or boards/**/Kconfig and the user should generally avoid
modifying it. The option is used to select linker script OUTPUT_FORMAT
and command line option for gen_isr_tables.py.
config 64BIT
bool
help
This option tells the build system that the target system is
using a 64-bit address space, meaning that pointer and long types
are 64 bits wide. This option is selected by arch/$ARCH/Kconfig,
soc/**/Kconfig, or boards/**/Kconfig and the user should generally
avoid modifying it.
# Workaround for not being able to have commas in macro arguments
DT_CHOSEN_Z_SRAM := zephyr,sram
config SRAM_SIZE
int "SRAM Size in kB"
default $(dt_chosen_reg_size_int,$(DT_CHOSEN_Z_SRAM),0,K)
help
The SRAM size in kB. The default value comes from /chosen/zephyr,sram in
devicetree. The user should generally avoid changing it via menuconfig or
in configuration files.
config SRAM_BASE_ADDRESS
hex "SRAM Base Address"
default $(dt_chosen_reg_addr_hex,$(DT_CHOSEN_Z_SRAM))
help
The SRAM base address. The default value comes from from
/chosen/zephyr,sram in devicetree. The user should generally avoid
changing it via menuconfig or in configuration files.
if ARC || ARM || NIOS2 || X86
# Workaround for not being able to have commas in macro arguments
DT_CHOSEN_Z_FLASH := zephyr,flash
config FLASH_SIZE
int "Flash Size in kB"
default $(dt_chosen_reg_size_int,$(DT_CHOSEN_Z_FLASH),0,K) if (XIP && ARM) || !ARM
help
This option specifies the size of the flash in kB. It is normally set by
the board's defconfig file and the user should generally avoid modifying
it via the menu configuration.
config FLASH_BASE_ADDRESS
hex "Flash Base Address"
default $(dt_chosen_reg_addr_hex,$(DT_CHOSEN_Z_FLASH)) if (XIP && ARM) || !ARM
help
This option specifies the base address of the flash on the board. It is
normally set by the board's defconfig file and the user should generally
avoid modifying it via the menu configuration.
endif # ARM || ARC || NIOS2 || X86
if ARCH_HAS_TRUSTED_EXECUTION
config TRUSTED_EXECUTION_SECURE
bool "Trusted Execution: Secure firmware image"
help
Select this option to enable building a Secure firmware
image for a platform that supports Trusted Execution. A
Secure firmware image will execute in Secure state. It may
allow the CPU to execute in Non-Secure (Normal) state.
Therefore, a Secure firmware image shall be able to
configure security attributions of CPU resources (memory
areas, peripherals, interrupts, etc.) as well as to handle
faults, related to security violations. It may optionally
allow certain functions to be called from the Non-Secure
(Normal) domain.
config TRUSTED_EXECUTION_NONSECURE
depends on !TRUSTED_EXECUTION_SECURE
bool "Trusted Execution: Non-Secure firmware image"
help
Select this option to enable building a Non-Secure
firmware image for a platform that supports Trusted
Execution. A Non-Secure firmware image will execute
in Non-Secure (Normal) state. Therefore, it shall not
access CPU resources (memory areas, peripherals,
interrupts etc.) belonging to the Secure domain.
endif # ARCH_HAS_TRUSTED_EXECUTION
config HW_STACK_PROTECTION
bool "Hardware Stack Protection"
depends on ARCH_HAS_STACK_PROTECTION
help
Select this option to enable hardware-based platform features to
catch stack overflows when the system is running in privileged
mode. If CONFIG_USERSPACE is not enabled, the system is always
running in privileged mode.
Note that this does not necessarily prevent corruption and assertions
about the overall system state when a fault is triggered cannot be
made.
config USERSPACE
bool "User mode threads"
depends on ARCH_HAS_USERSPACE
depends on RUNTIME_ERROR_CHECKS
help
When enabled, threads may be created or dropped down to user mode,
which has significantly restricted permissions and must interact
with the kernel via system calls. See Zephyr documentation for more
details about this feature.
If a user thread overflows its stack, this will be caught and the
kernel itself will be shielded from harm. Enabling this option
may or may not catch stack overflows when the system is in
privileged mode or handling a system call; to ensure these are always
caught, enable CONFIG_HW_STACK_PROTECTION.
config PRIVILEGED_STACK_SIZE
int "Size of privileged stack"
default 1024
depends on ARCH_HAS_USERSPACE
help
This option sets the privileged stack region size that will be used
in addition to the user mode thread stack. During normal execution,
this region will be inaccessible from user mode. During system calls,
this region will be utilized by the system call. This value must be
a multiple of the minimum stack alignment.
config KOBJECT_TEXT_AREA
int "Size if kobject text area"
default 512 if COVERAGE_GCOV
default 512 if NO_OPTIMIZATIONS
default 256
depends on ARCH_HAS_USERSPACE
help
Size of kernel object text area. Used in linker script.
config GEN_PRIV_STACKS
bool
help
Selected if the architecture requires that privilege elevation stacks
be allocated in a separate memory area. This is typical of arches
whose MPUs require regions to be power-of-two aligned/sized.
FIXME: This should be removed and replaced with checks against
CONFIG_MPU_REQUIRES_POWER_OF_TWO_ALIGNMENT, but both ARM and ARC
changes will be necessary for this.
config STACK_GROWS_UP
bool "Stack grows towards higher memory addresses"
help
Select this option if the architecture has upward growing thread
stacks. This is not common.
config NO_UNUSED_STACK_INSPECTION
bool
help
Selected if the architecture will generate a fault if unused stack
memory is examined, which is the region between the current stack
pointer and the deepest available address in the current stack
region.
config MAX_THREAD_BYTES
int "Bytes to use when tracking object thread permissions"
default 2
depends on USERSPACE
help
Every kernel object will have an associated bitfield to store
thread permissions for that object. This controls the size of the
bitfield (in bytes) and imposes a limit on how many threads can
be created in the system.
config DYNAMIC_OBJECTS
bool "Allow kernel objects to be allocated at runtime"
depends on USERSPACE
help
Enabling this option allows for kernel objects to be requested from
the calling thread's resource pool, at a slight cost in performance
due to the supplemental run-time tables required to validate such
objects.
Objects allocated in this way can be freed with a supervisor-only
API call, or when the number of references to that object drops to
zero.
config NOCACHE_MEMORY
bool "Support for uncached memory"
depends on ARCH_HAS_NOCACHE_MEMORY_SUPPORT
help
Add a "nocache" read-write memory section that is configured to
not be cached. This memory section can be used to perform DMA
transfers when cache coherence issues are not optimal or can not
be solved using cache maintenance operations.
menu "Interrupt Configuration"
config DYNAMIC_INTERRUPTS
bool "Enable installation of IRQs at runtime"
help
Enable installation of interrupts at runtime, which will move some
interrupt-related data structures to RAM instead of ROM, and
on some architectures increase code size.
config GEN_ISR_TABLES
bool "Use generated IRQ tables"
help
This option controls whether a platform uses the gen_isr_tables
script to generate its interrupt tables. This mechanism will create
an appropriate hardware vector table and/or software IRQ table.
config GEN_IRQ_VECTOR_TABLE
bool "Generate an interrupt vector table"
default y
depends on GEN_ISR_TABLES
help
This option controls whether a platform using gen_isr_tables
needs an interrupt vector table created. Only disable this if the
platform does not use a vector table at all, or requires the vector
table to be in a format that is not an array of function pointers
indexed by IRQ line. In the latter case, the vector table must be
supplied by the application or architecture code.
config GEN_SW_ISR_TABLE
bool "Generate a software ISR table"
default y
depends on GEN_ISR_TABLES
help
This option controls whether a platform using gen_isr_tables
needs a software ISR table table created. This is an array of struct
_isr_table_entry containing the interrupt service routine and supplied
parameter.
config ARCH_SW_ISR_TABLE_ALIGN
int "Alignment size of a software ISR table"
default 0
depends on GEN_SW_ISR_TABLE
help
This option controls alignment size of generated
_sw_isr_table. Some architecture needs a software ISR table
to be aligned to architecture specific size. The default
size is 0 for no alignment.
config GEN_IRQ_START_VECTOR
int
default 0
depends on GEN_ISR_TABLES
help
On some architectures, part of the vector table may be reserved for
system exceptions and is declared separately from the tables
created by gen_isr_tables.py. When creating these tables, this value
will be subtracted from CONFIG_NUM_IRQS to properly size them.
This is a hidden option which needs to be set per architecture and
left alone.
config IRQ_OFFLOAD
bool "Enable IRQ offload"
depends on TEST
help
Enable irq_offload() API which allows functions to be synchronously
run in interrupt context. Only useful for test cases that need
to validate the correctness of kernel objects in IRQ context.
endmenu # Interrupt configuration
endmenu
#
# Architecture Capabilities
#
config ARCH_HAS_TRUSTED_EXECUTION
bool
config ARCH_HAS_STACK_PROTECTION
bool
config ARCH_HAS_USERSPACE
bool
config ARCH_HAS_EXECUTABLE_PAGE_BIT
bool
config ARCH_HAS_NOCACHE_MEMORY_SUPPORT
bool
config ARCH_HAS_RAMFUNC_SUPPORT
bool
config ARCH_HAS_NESTED_EXCEPTION_DETECTION
bool
#
# Other architecture related options
#
config ARCH_HAS_THREAD_ABORT
bool
endchoice
#
# Hidden PM feature configs which are to be selected by
# individual SoC.
#
config HAS_SYS_POWER_STATE_SLEEP_1
config SYS_POWER_LOW_POWER_STATE_SUPPORTED
# Hidden
bool
default n
help
This option signifies that the target supports the SYS_POWER_STATE_SLEEP_1
configuration option.
This option signifies that the target supports the SYS_POWER_LOW_POWER_STATE
configuration option.
config HAS_SYS_POWER_STATE_SLEEP_2
config SYS_POWER_DEEP_SLEEP_SUPPORTED
# Hidden
bool
default n
help
This option signifies that the target supports the SYS_POWER_STATE_SLEEP_2
configuration option.
config HAS_SYS_POWER_STATE_SLEEP_3
bool
help
This option signifies that the target supports the SYS_POWER_STATE_SLEEP_3
configuration option.
config HAS_SYS_POWER_STATE_DEEP_SLEEP_1
bool
help
This option signifies that the target supports the SYS_POWER_STATE_DEEP_SLEEP_1
configuration option.
config HAS_SYS_POWER_STATE_DEEP_SLEEP_2
bool
help
This option signifies that the target supports the SYS_POWER_STATE_DEEP_SLEEP_2
configuration option.
config HAS_SYS_POWER_STATE_DEEP_SLEEP_3
bool
help
This option signifies that the target supports the SYS_POWER_STATE_DEEP_SLEEP_3
configuration option.
This option signifies that the target supports the SYS_POWER_DEEP_SLEEP
configuration option.
config BOOTLOADER_CONTEXT_RESTORE_SUPPORTED
# Hidden
bool
default n
help
This option signifies that the target has options of bootloaders
that support context restore upon resume from deep sleep
This option signifies that the target has options of bootloaders
that support context restore upon resume from deep sleep
#
# Hidden CPU family configs
# End hidden PM feature configs
#
config CPU_HAS_TEE
bool
help
This option is enabled when the CPU has support for Trusted
Execution Environment (e.g. when it has a security attribution
unit).
config CPU_HAS_DCLS
bool
help
This option is enabled when the processor hardware is configured in
Dual-redundant Core Lock-step (DCLS) topology.
config CPU_HAS_FPU
bool
help
This option is enabled when the CPU has hardware floating point
unit.
config CPU_HAS_FPU_DOUBLE_PRECISION
bool
select CPU_HAS_FPU
help
When enabled, this indicates that the CPU has a double floating point
precision unit.
config CPU_HAS_MPU
bool
help
This option is enabled when the CPU has a Memory Protection Unit (MPU).
config MEMORY_PROTECTION
bool
help
This option is enabled when Memory Protection features are supported.
Memory protection support is currently available on ARC, ARM, and x86
architectures.
config MPU_REQUIRES_POWER_OF_TWO_ALIGNMENT
bool
help
This option is enabled when the MPU requires a power of two alignment
and size for MPU regions.
config MPU_REQUIRES_NON_OVERLAPPING_REGIONS
bool
help
This option is enabled when the MPU requires the active (i.e. enabled)
MPU regions to be non-overlapping with each other.
config MPU_GAP_FILLING
bool "Force MPU to be filling in background memory regions"
depends on MPU_REQUIRES_NON_OVERLAPPING_REGIONS
default y if !USERSPACE
help
This Kconfig option instructs the MPU driver to enforce
a full kernel SRAM partitioning, when it programs the
dynamic MPU regions (user thread stack, PRIV stack guard
and application memory domains) during context-switch. We
allow this to be a configurable option, in order to be able
to switch the option off and have an increased number of MPU
regions available for application memory domain programming.
Notes:
An increased number of MPU regions should only be required,
when building with USERSPACE support. As a result, when we
build without USERSPACE support, gap filling should always
be required.
When the option is switched off, access to memory areas not
covered by explicit MPU regions is restricted to privileged
code on an ARCH-specific basis. Refer to ARCH-specific
documentation for more information on how this option is
used.
menu "Floating Point Options"
config FPU
bool "Enable floating point unit (FPU)"
depends on CPU_HAS_FPU
depends on ARC || ARM || RISCV || X86
help
This option enables the hardware Floating Point Unit (FPU), in order to
support using the floating point registers and instructions.
When this option is enabled, by default, threads may use the floating
point registers only in an exclusive manner, and this usually means that
only one thread may perform floating point operations.
If it is necessary for multiple threads to perform concurrent floating
point operations, the "FPU register sharing" option must be enabled to
preserve the floating point registers across context switches.
Note that this option cannot be selected for the platforms that do not
include a hardware floating point unit; the floating point support for
those platforms is dependent on the availability of the toolchain-
provided software floating point library.
config FPU_SHARING
bool "FPU register sharing"
depends on FPU
help
This option enables preservation of the hardware floating point registers
across context switches to allow multiple threads to perform concurrent
floating point operations.
endmenu
config ARCH
string
help
System architecture string.
System architecture string.
config SOC
string
help
SoC name which can be found under soc/<arch>/<soc name>.
This option holds the directory name used by the build system to locate
the correct linker and header files for the SoC.
SoC name which can be found under arch/<arch>/soc/<soc name>.
This option holds the directory name used by the build system to locate
the correct linker and header files for the SoC. This option will go away
once all SoCs are using family/series structure.
config SOC_SERIES
string
help
SoC series name which can be found under soc/<arch>/<family>/<series>.
This option holds the directory name used by the build system to locate
the correct linker and header files.
SoC series name which can be found under arch/<arch>/soc/<family>/<series>.
This option holds the directory name used by the build system to locate
the correct linker and header files.
config SOC_FAMILY
string
help
SoC family name which can be found under soc/<arch>/<family>.
This option holds the directory name used by the build system to locate
the correct linker and header files.
SoC family name which can be found under arch/<arch>/soc/<family>.
This option holds the directory name used by the build system to locate
the correct linker and header files.
config BOARD
string
help
This option holds the name of the board and is used to locate the files
related to the board in the source tree (under boards/).
The Board is the first location where we search for a linker.ld file,
if not found we look for the linker file in
soc/<arch>/<family>/<series>
This option holds the name of the board and is used to located the files
related to the board in the source tree (under boards/).
The Board is the first location where we search for a linker.ld file,
if not found we look for the linker file in
arch/<arch>/soc/<family>/<series>
source "arch/*/Kconfig"
source "boards/Kconfig"

1
arch/Makefile Normal file
View File

@@ -0,0 +1 @@
obj-y += $(ARCH)/

View File

@@ -1,15 +0,0 @@
# SPDX-License-Identifier: Apache-2.0
# Enable debug support in mdb
# Dwarf version 2 can be recognized by mdb
# The default dwarf version in gdb is not recognized by mdb
zephyr_cc_option(-g3 -gdwarf-2)
# Without this (poorly named) option, compiler may generate undefined
# references to abort().
# See https://gcc.gnu.org/bugzilla/show_bug.cgi?id=63691
zephyr_cc_option(-fno-delete-null-pointer-checks)
zephyr_cc_option_ifdef(CONFIG_ARC_USE_UNALIGNED_MEM_ACCESS -munaligned-access)
add_subdirectory(core)

6
arch/arc/Kbuild Normal file
View File

@@ -0,0 +1,6 @@
subdir-ccflags-y +=-I$(srctree)/include/drivers
subdir-ccflags-y +=-I$(srctree)/drivers
subdir-asflags-y += $(subdir-ccflags-y)
obj-y += soc/$(SOC_PATH)/
obj-y += core/

View File

@@ -1,7 +1,28 @@
# ARC options
# ARC EM4 options
#
# Copyright (c) 2014 Wind River Systems, Inc.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#
choice
prompt "ARC SoC Selection"
depends on ARC
source "arch/arc/soc/*/Kconfig.soc"
endchoice
# Copyright (c) 2014, 2019 Wind River Systems, Inc.
# SPDX-License-Identifier: Apache-2.0
menu "ARC Options"
depends on ARC
@@ -9,70 +30,53 @@ menu "ARC Options"
config ARCH
default "arc"
choice
prompt "ARC core family"
default CPU_ARCEM
config ARCH_DEFCONFIG
string
default "arch/arc/defconfig"
config CPU_ARCEM
bool "ARC EM cores"
menu "ARC EM4 processor options"
config CPU_ARCEM4
bool
default y
select CPU_ARCV2
select ATOMIC_OPERATIONS_C
help
This option signifies the use of an ARC EM CPU
This option signifies the use of an ARC EM4 CPU
config CPU_ARCHS
bool "ARC HS cores"
select CPU_ARCV2
select ATOMIC_OPERATIONS_BUILTIN
help
This option signifies the use of an ARC HS CPU
endchoice
config CPU_EM4
bool
help
If y, the SoC uses an ARC EM4 CPU
config CPU_EM4_DMIPS
bool
help
If y, the SoC uses an ARC EM4 DMIPS CPU
config CPU_EM4_FPUS
bool
help
If y, the SoC uses an ARC EM4 DMIPS CPU with the single-precision
floating-point extension
config CPU_EM4_FPUDA
bool
help
If y, the SoC uses an ARC EM4 DMIPS CPU with single-precision
floating-point and double assist instructions
config CPU_EM6
bool
help
If y, the SoC uses an ARC EM6 CPU
config FP_FPU_DA
bool
endmenu
menu "ARCv2 Family Options"
config CPU_ARCV2
config CPU_ARCV2
bool
select ARCH_HAS_STACK_PROTECTION if ARC_HAS_STACK_CHECKING || ARC_MPU
select ARCH_HAS_USERSPACE if ARC_MPU
select USE_SWITCH
select USE_SWITCH_SUPPORTED
default y
select NANOKERNEL_TICKLESS_IDLE_SUPPORTED
help
This option signifies the use of a CPU of the ARCv2 family.
config NUM_IRQ_PRIO_LEVELS
int "Number of supported interrupt priority levels"
config NSIM
prompt "Running on the MetaWare nSIM simulator"
bool
default n
help
For running on nSIM simulator.
a) Uses non-XIP to run in RAM.
b) Linked at address 0x4000 with 0x4000 of RAM so that it works with
a pc_size of 16 (default).
config DATA_ENDIANNESS_LITTLE
bool
default y
help
This is driven by the processor implementation, since it is fixed in
hardware. The BSP should set this value to 'n' if the data is
implemented as big endian.
config NUM_IRQ_PRIO_LEVELS
int
prompt "Number of supported interrupt priority levels"
range 1 16
help
Interrupt priorities available will be 0 to NUM_IRQ_PRIO_LEVELS-1.
@@ -80,8 +84,9 @@ config NUM_IRQ_PRIO_LEVELS
The BSP must provide a valid default for proper operation.
config NUM_IRQS
int "Upper limit of interrupt numbers/IDs used"
config NUM_IRQS
int
prompt "Upper limit of interrupt numbers/IDs used"
range 17 256
help
Interrupts available will be 0 to NUM_IRQS-1.
@@ -91,244 +96,216 @@ config NUM_IRQS
The BSP must provide a valid default. This drives the size of the
vector table.
config RGF_NUM_BANKS
int "Number of General Purpose Register Banks"
config RGF_NUM_BANKS
int
prompt "Number of General Purpose Register Banks"
depends on CPU_ARCV2
range 1 2
default 2
help
The ARC CPU can be configured to have more than one register
bank. If fast interrupts are supported (FIRQ), the 2nd
register bank, in the set, will be used by FIRQ interrupts.
If fast interrupts are supported but there is only 1
register bank, the fast interrupt handler must save
and restore general purpose registers.
The ARC CPU can be configured to have more than one register
bank. If fast interrupts are supported (FIRQ), the 2nd
register bank, in the set, will be used by FIRQ interrupts.
If fast interrupts are supported but there is only 1
register bank, the fast interrupt handler must save
and restore general purpose regsiters.
config ARC_FIRQ
bool "FIRQ enable"
default y
help
Fast interrupts are supported (FIRQ). If FIRQ enabled, for interrupts
with highest priority, status32 and pc will be saved in aux regs,
other regs will be saved according to the number of register bank;
If FIRQ is disabled, the handle of interrupts with highest priority
will be same with other interrupts.
config ARC_FIRQ_STACK
bool "Enable separate firq stack"
depends on ARC_FIRQ && RGF_NUM_BANKS > 1
help
Use separate stack for FIRQ handing. When the fast irq is also a direct
irq, this will get the minimal interrupt latency.
config ARC_FIRQ_STACK_SIZE
int "FIRQ stack size"
depends on ARC_FIRQ_STACK
config FIRQ_STACK_SIZE
int
prompt "Size of stack for FIRQs (in bytes)"
depends on CPU_ARCV2
default 1024
help
The size of firq stack.
FIRQs and regular IRQs have different stacks so that a FIRQ can start
running without doing stack switching in software.
config ARC_HAS_STACK_CHECKING
bool "ARC has STACK_CHECKING"
default y
config ARC_STACK_CHECKING
bool "Enable Stack Checking"
depends on CPU_ARCV2
default n
help
ARC is configured with STACK_CHECKING which is a mechanism for
checking stack accesses and raising an exception when a stack
overflow or underflow is detected.
ARCV2 has a special feature allowing to check stack overflows. This
enables code that allows using this debug feature
config ARC_CONNECT
bool "ARC has ARC connect"
select SCHED_IPI_SUPPORTED
help
ARC is configured with ARC CONNECT which is a hardware for connecting
multi cores.
config ARC_STACK_CHECKING
bool
select NO_UNUSED_STACK_INSPECTION
help
Use ARC STACK_CHECKING to do stack protection
config ARC_STACK_PROTECTION
bool
default y if HW_STACK_PROTECTION
select ARC_STACK_CHECKING if ARC_HAS_STACK_CHECKING
select MPU_STACK_GUARD if (!ARC_STACK_CHECKING && ARC_MPU && ARC_MPU_VER !=2)
select THREAD_STACK_INFO
help
This option enables either:
- The ARC stack checking, or
- the MPU-based stack guard
to cause a system fatal error
if the bounds of the current process stack are overflowed.
The two stack guard options are mutually exclusive. The
selection of the ARC stack checking is
prioritized over the MPU-based stack guard.
config ARC_USE_UNALIGNED_MEM_ACCESS
bool "Enable unaligned access in HW"
default y if CPU_ARCHS
depends on (CPU_ARCEM && !ARC_HAS_SECURE) || CPU_ARCHS
help
ARC EM cores w/o secure shield 2+2 mode support might be configured
to support unaligned memory access which is then disabled by default.
Enable unaligned access in hardware and make software to use it.
config FAULT_DUMP
int "Fault dump level"
config FAULT_DUMP
int
prompt "Fault dump level"
default 2
range 0 2
help
Different levels for display information when a fault occurs.
Different levels for display information when a fault occurs.
2: The default. Display specific and verbose information. Consumes
2: The default. Display specific and verbose information. Consumes
the most memory (long strings).
1: Display general and short information. Consumes less memory
1: Display general and short information. Consumes less memory
(short strings).
0: Off.
0: Off.
config XIP
default y if !UART_NSIM
config IRQ_OFFLOAD
bool "Enable IRQ offload"
default n
help
Enable irq_offload() API which allows functions to be synchronously
run in interrupt context. Uses one entry in the IDT. Mainly useful
for test cases.
config GEN_ISR_TABLES
config XIP
default n if NSIM
default y
config GEN_IRQ_START_VECTOR
default 16
config HARVARD
bool "Harvard Architecture"
prompt "Harvard Architecture"
bool
default n
help
The ARC CPU can be configured to have two busses;
one for instruction fetching and another that serves as a data bus.
config CODE_DENSITY
bool "Code Density Option"
config ICCM_SIZE
int "ICCM Size in kB"
help
Enable code density option to get better code density
This option specifies the size of the ICCM in kB. It is normally set by
the board's defconfig file and the user should generally avoid modifying
it via the menu configuration.
config ARC_HAS_ACCL_REGS
bool "Reg Pair ACCL:ACCH (FPU and/or MPY > 6)"
default y if FPU
config ICCM_BASE_ADDRESS
hex "ICCM Base Address"
help
Depending on the configuration, CPU can contain accumulator reg-pair
(also referred to as r58:r59). These can also be used by gcc as GPR so
kernel needs to save/restore per process
This option specifies the base address of the ICCM on the board. It is
normally set by the board's defconfig file and the user should generally
avoid modifying it via the menu configuration.
config ARC_HAS_SECURE
bool "ARC has SecureShield"
select CPU_HAS_TEE
select ARCH_HAS_TRUSTED_EXECUTION
config DCCM_SIZE
int "DCCM Size in kB"
help
This option is enabled when ARC core supports secure mode
This option specifies the size of the DCCM in kB. It is normally set by
the board's defconfig file and the user should generally avoid modifying
it via the menu configuration.
config SJLI_TABLE_SIZE
int "SJLI table size"
depends on ARC_SECURE_FIRMWARE
default 8
config DCCM_BASE_ADDRESS
hex "DCCM Base Address"
help
The size of sjli (Secure Jump and Link Indexed) table. The
code in normal mode call secure services in secure mode through
sjli instruction.
This option specifies the base address of the DCCM on the board. It is
normally set by the board's defconfig file and the user should generally
avoid modifying it via the menu configuration.
config ARC_SECURE_FIRMWARE
bool "Generate Secure Firmware"
depends on ARC_HAS_SECURE
default y if TRUSTED_EXECUTION_SECURE
config SRAM_SIZE
int "SRAM Size in kB"
help
This option indicates that we are building a Zephyr image that
is intended to execute in secure mode. The option is only
applicable to ARC processors that implement the SecureShield.
This option specifies the size of the SRAM in kB. It is normally set by
the board's defconfig file and the user should generally avoid modifying
it via the menu configuration.
This option enables Zephyr to include code that executes in
secure mode, as well as to exclude code that is designed to
execute only in normal mode.
Code executing in secure mode has access to both the secure
and normal resources of the ARC processors.
config ARC_NORMAL_FIRMWARE
bool "Generate Normal Firmware"
depends on !ARC_SECURE_FIRMWARE
depends on ARC_HAS_SECURE
default y if TRUSTED_EXECUTION_NONSECURE
config SRAM_BASE_ADDRESS
hex "SRAM Base Address"
help
This option indicates that we are building a Zephyr image that
is intended to execute in normal mode. Execution of this
image is triggered by secure firmware that executes in secure
mode. The option is only applicable to ARC processors that
implement the SecureShield.
This option specifies the base address of the SRAM on the board. It is
normally set by the board's defconfig file and the user should generally
avoid modifying it via the menu configuration.
This option enables Zephyr to include code that executes in
normal mode only, as well as to exclude code that is
designed to execute only in secure mode.
Code executing in normal mode has no access to secure
resources of the ARC processors, and, therefore, it shall avoid
accessing them.
menu "ARC MPU Options"
depends on CPU_HAS_MPU
config ARC_MPU_ENABLE
bool "Enable MPU"
select ARC_MPU
config FLASH_SIZE
int "Flash Size in kB"
help
Enable MPU
This option specifies the size of the flash in kB. It is normally set by
the board's defconfig file and the user should generally avoid modifying
it via the menu configuration.
source "arch/arc/core/mpu/Kconfig"
config FLASH_BASE_ADDRESS
hex "Flash Base Address"
help
This option specifies the base address of the flash on the board. It is
normally set by the board's defconfig file and the user should generally
avoid modifying it via the menu configuration.
endmenu
config SW_ISR_TABLE
bool
prompt "Enable software interrupt handler table"
default y
help
Enable an interrupt handler table implemented in software. This
table, unlike ISRs connected directly in the vector table, allow
a parameter to be passed to the interrupt handlers. Also, invoking
the exeception/interrupt exit stub is automatically done.
config IRQ_VECTOR_TABLE_CUSTOM
bool
prompt "Projects provide a custom static IRQ part of vector table"
depends on !SW_ISR_TABLE
default n
help
Projects, not the BSP, provide the IRQ part of the vector table.
This is the table of interrupt handlers with the best potential
performance, but is the less flexible.
The ISRs are installed directly in the vector table, thus are
directly called by the CPU when an interrupt is taken. This adds
the least overhead when handling an interrupt.
Downsides:
- ISRs cannot have a parameter
- ISRs cannot be connected at runtime
- ISRs must notify the kernel manually by invoking _ExcExit() when
then are about to return.
config IRQ_VECTOR_TABLE_BSP
bool
# omit prompt to signify a "hidden" option
depends on SW_ISR_TABLE || !IRQ_VECTOR_TABLE_CUSTOM
default y
help
Not user-selectable, helps build system logic.
config ARCH_HAS_TASK_ABORT
bool
# omit prompt to signify a "hidden" option
default n
config ARCH_HAS_NANO_FIBER_ABORT
bool
# omit prompt to signify a "hidden" option
default n
config CACHE_LINE_SIZE_DETECT
bool "Detect d-cache line size at runtime"
bool
prompt "Detect d-cache line size at runtime"
default n
help
This option enables querying the d-cache build register for finding
the d-cache line size at the expense of taking more memory and code
and a slightly increased boot time.
This option enables querying the d-cache build register for finding
the d-cache line size at the expense of taking more memory and code
and a slightly increased boot time.
If the CPU's d-cache line size is known in advance, disable this
option and manually enter the value for CACHE_LINE_SIZE.
If the CPU's d-cache line size is known in advance, disable this
option and manually enter the value for CACHE_LINE_SIZE.
config CACHE_LINE_SIZE
int "Cache line size" if !CACHE_LINE_SIZE_DETECT
int
prompt "Cache line size" if !CACHE_LINE_SIZE_DETECT
default 32
help
Size in bytes of a CPU d-cache line.
Size in bytes of a CPU d-cache line.
Detect automatically at runtime by selecting CACHE_LINE_SIZE_DETECT.
Detect automatically at runtime by selecting CACHE_LINE_SIZE_DETECT.
config ARCH_CACHE_FLUSH_DETECT
bool
default n
config CACHE_FLUSHING
bool "Enable d-cache flushing mechanism"
help
This links in the sys_cache_flush() function, which provides a
way to flush multiple lines of the d-cache.
If the d-cache is present, set this to y.
If the d-cache is NOT present, set this to n.
config ARC_EXCEPTION_STACK_SIZE
int "ARC exception handling stack size"
default 768
help
Size in bytes of exception handling stack which is at the top of
interrupt stack to get smaller memory footprint because exception
is not frequent. To reduce the impact on interrupt handling,
especially nested interrupt, it cannot be too large.
endmenu
config ARC_EXCEPTION_DEBUG
bool "Unhandled exception debugging information"
bool
default n
depends on PRINTK || LOG
prompt "Enable d-cache flushing mechanism"
help
Print human-readable information about exception vectors, cause codes,
and parameters, at a cost of code/data size for the human-readable
strings.
This links in the sys_cache_flush() function, which provides a
way to flush multiple lines of the d-cache.
If the d-cache is present, set this to y.
If the d-cache is NOT present, set this to n.
endmenu
source "arch/arc/soc/*/Kconfig"
endmenu

15
arch/arc/Makefile Normal file
View File

@@ -0,0 +1,15 @@
cflags-y += $(call cc-option,-ffunction-sections,) $(call cc-option,-fdata-sections,)
cflags-$(CONFIG_ARC_STACK_CHECKING) = $(call cc-option,-fomit-frame-pointer)
cflags-$(CONFIG_LTO) = $(call cc-option,-flto,)
include $(srctree)/arch/$(ARCH)/soc/$(SOC_PATH)/Makefile
KBUILD_CFLAGS += $(cflags-y)
KBUILD_CXXFLAGS += $(cflags-y)
soc-cxxflags ?= $(soc-cflags)
soc-aflags ?= $(soc-cflags)
KBUILD_CFLAGS += $(soc-cflags)
KBUILD_CXXFLAGS += $(soc-cxxflags)
KBUILD_AFLAGS += $(soc-aflags)

View File

@@ -1,34 +0,0 @@
# SPDX-License-Identifier: Apache-2.0
zephyr_library()
zephyr_library_sources(
thread.c
thread_entry_wrapper.S
cpu_idle.S
fatal.c
fault.c
fault_s.S
irq_manage.c
timestamp.c
isr_wrapper.S
regular_irq.S
switch.S
prep_c.c
reset.S
vector_table.c
)
zephyr_library_sources_ifdef(CONFIG_CACHE_FLUSHING cache.c)
zephyr_library_sources_ifdef(CONFIG_ARC_FIRQ fast_irq.S)
zephyr_library_sources_if_kconfig(irq_offload.c)
zephyr_library_sources_ifdef(CONFIG_USERSPACE userspace.S)
zephyr_library_sources_ifdef(CONFIG_ARC_CONNECT arc_connect.c)
zephyr_library_sources_ifdef(CONFIG_ARC_CONNECT arc_smp.c)
add_subdirectory_ifdef(CONFIG_ARC_CORE_MPU mpu)
add_subdirectory_ifdef(CONFIG_ARC_SECURE_FIRMWARE secureshield)
zephyr_linker_sources(ROM_START SORT_KEY 0x0vectors vector_table.ld)

20
arch/arc/core/Makefile Normal file
View File

@@ -0,0 +1,20 @@
ccflags-y += -I$(srctree)/kernel/unified/include
ccflags-y +=-I$(srctree)/arch/$(ARCH)/include
obj-y += thread.o thread_entry_wrapper.o \
cpu_idle.o fast_irq.o fatal.o fault.o \
fault_s.o irq_manage.o cache.o \
isr_wrapper.o regular_irq.o swap_macros.h swap.o \
sys_fatal_error_handler.o
obj-y += prep_c.o \
reset.o \
vector_table.o
obj-$(CONFIG_IRQ_OFFLOAD) += irq_offload.o
# Some ARC cores like the EM4 lack the atomic LLOCK/SCOND and
# can't use these.
obj-$(CONFIG_ATOMIC_OPERATIONS_CUSTOM) += atomic.o
obj-$(CONFIG_IRQ_VECTOR_TABLE_BSP) += irq_vector_table.o
obj-$(CONFIG_SW_ISR_TABLE) += sw_isr_table.o

View File

@@ -1,449 +0,0 @@
/*
* Copyright (c) 2019 Synopsys.
*
* SPDX-License-Identifier: Apache-2.0
*/
/**
* @file
* @brief ARCv2 ARC CONNECT driver
*
*/
#include <kernel.h>
#include <arch/cpu.h>
#include <spinlock.h>
static struct k_spinlock arc_connect_spinlock;
#define LOCKED(lck) for (k_spinlock_key_t __i = {}, \
__key = k_spin_lock(lck); \
!__i.key; \
k_spin_unlock(lck, __key), __i.key = 1)
/* Generate an inter-core interrupt to the target core */
void z_arc_connect_ici_generate(u32_t core)
{
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd(ARC_CONNECT_CMD_INTRPT_GENERATE_IRQ, core);
}
}
/* Acknowledge the inter-core interrupt raised by core */
void z_arc_connect_ici_ack(u32_t core)
{
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd(ARC_CONNECT_CMD_INTRPT_GENERATE_ACK, core);
}
}
/* Read inter-core interrupt status */
u32_t z_arc_connect_ici_read_status(u32_t core)
{
u32_t ret = 0;
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd(ARC_CONNECT_CMD_INTRPT_READ_STATUS, core);
ret = z_arc_connect_cmd_readback();
}
return ret;
}
/* Check the source of inter-core interrupt */
u32_t z_arc_connect_ici_check_src(void)
{
u32_t ret = 0;
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd(ARC_CONNECT_CMD_INTRPT_CHECK_SOURCE, 0);
ret = z_arc_connect_cmd_readback();
}
return ret;
}
/* Clear the inter-core interrupt */
void z_arc_connect_ici_clear(void)
{
u32_t cpu, c;
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd(ARC_CONNECT_CMD_INTRPT_CHECK_SOURCE, 0);
cpu = z_arc_connect_cmd_readback(); /* 1,2,4,8... */
/*
* In rare case, multiple concurrent ICIs sent to same target can
* possibly be coalesced by MCIP into 1 asserted IRQ, so @cpu can be
* "vectored" (multiple bits sets) as opposed to typical single bit
*/
while (cpu) {
c = find_lsb_set(cpu) - 1;
z_arc_connect_cmd(
ARC_CONNECT_CMD_INTRPT_GENERATE_ACK, c);
cpu &= ~(1U << c);
}
}
}
/* Reset the cores in core_mask */
void z_arc_connect_debug_reset(u32_t core_mask)
{
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd_data(ARC_CONNECT_CMD_DEBUG_RESET,
0, core_mask);
}
}
/* Halt the cores in core_mask */
void z_arc_connect_debug_halt(u32_t core_mask)
{
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd_data(ARC_CONNECT_CMD_DEBUG_HALT,
0, core_mask);
}
}
/* Run the cores in core_mask */
void z_arc_connect_debug_run(u32_t core_mask)
{
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd_data(ARC_CONNECT_CMD_DEBUG_RUN,
0, core_mask);
}
}
/* Set core mask */
void z_arc_connect_debug_mask_set(u32_t core_mask, u32_t mask)
{
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd_data(ARC_CONNECT_CMD_DEBUG_SET_MASK,
mask, core_mask);
}
}
/* Read core mask */
u32_t z_arc_connect_debug_mask_read(u32_t core_mask)
{
u32_t ret = 0;
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd_data(ARC_CONNECT_CMD_DEBUG_READ_MASK,
0, core_mask);
ret = z_arc_connect_cmd_readback();
}
return ret;
}
/*
* Select cores that should be halted if the core issuing the command is halted
*/
void z_arc_connect_debug_select_set(u32_t core_mask)
{
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd_data(ARC_CONNECT_CMD_DEBUG_SET_SELECT,
0, core_mask);
}
}
/* Read the select value */
u32_t z_arc_connect_debug_select_read(void)
{
u32_t ret = 0;
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd(ARC_CONNECT_CMD_DEBUG_READ_SELECT, 0);
ret = z_arc_connect_cmd_readback();
}
return ret;
}
/* Read the status, halt or run of all cores in the system */
u32_t z_arc_connect_debug_en_read(void)
{
u32_t ret = 0;
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd(ARC_CONNECT_CMD_DEBUG_READ_EN, 0);
ret = z_arc_connect_cmd_readback();
}
return ret;
}
/* Read the last command sent */
u32_t z_arc_connect_debug_cmd_read(void)
{
u32_t ret = 0;
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd(ARC_CONNECT_CMD_DEBUG_READ_CMD, 0);
ret = z_arc_connect_cmd_readback();
}
return ret;
}
/* Read the value of internal MCD_CORE register */
u32_t z_arc_connect_debug_core_read(void)
{
u32_t ret = 0;
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd(ARC_CONNECT_CMD_DEBUG_READ_CORE, 0);
ret = z_arc_connect_cmd_readback();
}
return ret;
}
/* Clear global free running counter */
void z_arc_connect_gfrc_clear(void)
{
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd(ARC_CONNECT_CMD_GFRC_CLEAR, 0);
}
}
/* Read total 64 bits of global free running counter */
u64_t z_arc_connect_gfrc_read(void)
{
u32_t low;
u32_t high;
u32_t key;
/*
* each core has its own arc connect interface, i.e.,
* CMD/READBACK. So several concurrent commands to ARC
* connect are of if they are trying to access different
* sub-components. For GFRC, HW allows simultaneously accessing to
* counters. So an irq lock is enough.
*/
key = arch_irq_lock();
z_arc_connect_cmd(ARC_CONNECT_CMD_GFRC_READ_LO, 0);
low = z_arc_connect_cmd_readback();
z_arc_connect_cmd(ARC_CONNECT_CMD_GFRC_READ_HI, 0);
high = z_arc_connect_cmd_readback();
arch_irq_unlock(key);
return (((u64_t)high) << 32) | low;
}
/* Enable global free running counter */
void z_arc_connect_gfrc_enable(void)
{
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd(ARC_CONNECT_CMD_GFRC_ENABLE, 0);
}
}
/* Disable global free running counter */
void z_arc_connect_gfrc_disable(void)
{
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd(ARC_CONNECT_CMD_GFRC_DISABLE, 0);
}
}
/* Disable global free running counter */
void z_arc_connect_gfrc_core_set(u32_t core_mask)
{
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd_data(ARC_CONNECT_CMD_GFRC_SET_CORE,
0, core_mask);
}
}
/* Set the relevant cores to halt global free running counter */
u32_t z_arc_connect_gfrc_halt_read(void)
{
u32_t ret = 0;
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd(ARC_CONNECT_CMD_GFRC_READ_HALT, 0);
ret = z_arc_connect_cmd_readback();
}
return ret;
}
/* Read the internal CORE register */
u32_t z_arc_connect_gfrc_core_read(void)
{
u32_t ret = 0;
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd(ARC_CONNECT_CMD_GFRC_READ_CORE, 0);
ret = z_arc_connect_cmd_readback();
}
return ret;
}
/* Enable interrupt distribute unit */
void z_arc_connect_idu_enable(void)
{
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd(ARC_CONNECT_CMD_IDU_ENABLE, 0);
}
}
/* Disable interrupt distribute unit */
void z_arc_connect_idu_disable(void)
{
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd(ARC_CONNECT_CMD_IDU_DISABLE, 0);
}
}
/* Read enable status of interrupt distribute unit */
u32_t z_arc_connect_idu_read_enable(void)
{
u32_t ret = 0;
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd(ARC_CONNECT_CMD_IDU_READ_ENABLE, 0);
ret = z_arc_connect_cmd_readback();
}
return ret;
}
/*
* Set the triggering mode and distribution mode for the specified common
* interrupt
*/
void z_arc_connect_idu_set_mode(u32_t irq_num,
u16_t trigger_mode, u16_t distri_mode)
{
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd_data(ARC_CONNECT_CMD_IDU_SET_MODE,
irq_num, (distri_mode | (trigger_mode << 4)));
}
}
/* Read the internal MODE register of the specified common interrupt */
u32_t z_arc_connect_idu_read_mode(u32_t irq_num)
{
u32_t ret = 0;
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd(ARC_CONNECT_CMD_IDU_READ_MODE, irq_num);
ret = z_arc_connect_cmd_readback();
}
return ret;
}
/*
* Set the target cores to receive the specified common interrupt
* when it is triggered
*/
void z_arc_connect_idu_set_dest(u32_t irq_num, u32_t core_mask)
{
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd_data(ARC_CONNECT_CMD_IDU_SET_DEST,
irq_num, core_mask);
}
}
/* Read the internal DEST register of the specified common interrupt */
u32_t z_arc_connect_idu_read_dest(u32_t irq_num)
{
u32_t ret = 0;
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd(ARC_CONNECT_CMD_IDU_READ_DEST, irq_num);
ret = z_arc_connect_cmd_readback();
}
return ret;
}
/* Assert the specified common interrupt */
void z_arc_connect_idu_gen_cirq(u32_t irq_num)
{
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd(ARC_CONNECT_CMD_IDU_GEN_CIRQ, irq_num);
}
}
/* Acknowledge the specified common interrupt */
void z_arc_connect_idu_ack_cirq(u32_t irq_num)
{
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd(ARC_CONNECT_CMD_IDU_ACK_CIRQ, irq_num);
}
}
/* Read the internal STATUS register of the specified common interrupt */
u32_t z_arc_connect_idu_check_status(u32_t irq_num)
{
u32_t ret = 0;
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd(ARC_CONNECT_CMD_IDU_CHECK_STATUS, irq_num);
ret = z_arc_connect_cmd_readback();
}
return ret;
}
/* Read the internal SOURCE register of the specified common interrupt */
u32_t z_arc_connect_idu_check_source(u32_t irq_num)
{
u32_t ret = 0;
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd(ARC_CONNECT_CMD_IDU_CHECK_SOURCE, irq_num);
ret = z_arc_connect_cmd_readback();
}
return ret;
}
/* Mask or unmask the specified common interrupt */
void z_arc_connect_idu_set_mask(u32_t irq_num, u32_t mask)
{
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd_data(ARC_CONNECT_CMD_IDU_SET_MASK,
irq_num, mask);
}
}
/* Read the internal MASK register of the specified common interrupt */
u32_t z_arc_connect_idu_read_mask(u32_t irq_num)
{
u32_t ret = 0;
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd(ARC_CONNECT_CMD_IDU_READ_MASK, irq_num);
ret = z_arc_connect_cmd_readback();
}
return ret;
}
/*
* Check if it is the first-acknowledging core to the common interrupt
* if IDU is programmed in the first-acknowledged mode
*/
u32_t z_arc_connect_idu_check_first(u32_t irq_num)
{
u32_t ret = 0;
LOCKED(&arc_connect_spinlock) {
z_arc_connect_cmd(ARC_CONNECT_CMD_IDU_CHECK_FIRST, irq_num);
ret = z_arc_connect_cmd_readback();
}
return ret;
}

View File

@@ -1,151 +0,0 @@
/*
* Copyright (c) 2019 Synopsys.
*
* SPDX-License-Identifier: Apache-2.0
*/
/**
* @file
* @brief codes required for ARC multicore and Zephyr smp support
*
*/
#include <device.h>
#include <kernel.h>
#include <kernel_structs.h>
#include <ksched.h>
#include <soc.h>
#include <init.h>
#ifndef IRQ_ICI
#define IRQ_ICI 19
#endif
#define ARCV2_ICI_IRQ_PRIORITY 1
volatile struct {
arch_cpustart_t fn;
void *arg;
} arc_cpu_init[CONFIG_MP_NUM_CPUS];
/*
* arc_cpu_wake_flag is used to sync up master core and slave cores
* Slave core will spin for arc_cpu_wake_flag until master core sets
* it to the core id of slave core. Then, slave core clears it to notify
* master core that it's waken
*
*/
volatile u32_t arc_cpu_wake_flag;
volatile char *arc_cpu_sp;
/*
* _curr_cpu is used to record the struct of _cpu_t of each cpu.
* for efficient usage in assembly
*/
volatile _cpu_t *_curr_cpu[CONFIG_MP_NUM_CPUS];
/* Called from Zephyr initialization */
void arch_start_cpu(int cpu_num, k_thread_stack_t *stack, int sz,
arch_cpustart_t fn, void *arg)
{
_curr_cpu[cpu_num] = &(_kernel.cpus[cpu_num]);
arc_cpu_init[cpu_num].fn = fn;
arc_cpu_init[cpu_num].arg = arg;
/* set the initial sp of target sp through arc_cpu_sp
* arc_cpu_wake_flag will protect arc_cpu_sp that
* only one slave cpu can read it per time
*/
arc_cpu_sp = Z_THREAD_STACK_BUFFER(stack) + sz;
arc_cpu_wake_flag = cpu_num;
/* wait slave cpu to start */
while (arc_cpu_wake_flag != 0) {
;
}
}
/* the C entry of slave cores */
void z_arc_slave_start(int cpu_num)
{
arch_cpustart_t fn;
#ifdef CONFIG_SMP
z_icache_setup();
z_irq_setup();
z_arc_connect_ici_clear();
z_irq_priority_set(IRQ_ICI, ARCV2_ICI_IRQ_PRIORITY, 0);
irq_enable(IRQ_ICI);
#endif
/* call the function set by arch_start_cpu */
fn = arc_cpu_init[cpu_num].fn;
fn(arc_cpu_init[cpu_num].arg);
}
#ifdef CONFIG_SMP
static void sched_ipi_handler(void *unused)
{
ARG_UNUSED(unused);
z_arc_connect_ici_clear();
z_sched_ipi();
}
/* arch implementation of sched_ipi */
void arch_sched_ipi(void)
{
u32_t i;
/* broadcast sched_ipi request to other cores
* if the target is current core, hardware will ignore it
*/
for (i = 0; i < CONFIG_MP_NUM_CPUS; i++) {
z_arc_connect_ici_generate(i);
}
}
static int arc_smp_init(struct device *dev)
{
ARG_UNUSED(dev);
struct arc_connect_bcr bcr;
/* necessary master core init */
_curr_cpu[0] = &(_kernel.cpus[0]);
bcr.val = z_arc_v2_aux_reg_read(_ARC_V2_CONNECT_BCR);
if (bcr.ipi) {
/* register ici interrupt, just need master core to register once */
z_arc_connect_ici_clear();
IRQ_CONNECT(IRQ_ICI, ARCV2_ICI_IRQ_PRIORITY,
sched_ipi_handler, NULL, 0);
irq_enable(IRQ_ICI);
} else {
__ASSERT(0,
"ARC connect has no inter-core interrupt\n");
return -ENODEV;
}
if (bcr.gfrc) {
/* global free running count init */
z_arc_connect_gfrc_enable();
/* when all cores halt, gfrc halt */
z_arc_connect_gfrc_core_set((1 << CONFIG_MP_NUM_CPUS) - 1);
z_arc_connect_gfrc_clear();
} else {
__ASSERT(0,
"ARC connect has no global free running counter\n");
return -ENODEV;
}
return 0;
}
SYS_INIT(arc_smp_init, PRE_KERNEL_1, CONFIG_KERNEL_INIT_PRIORITY_DEFAULT);
#endif

434
arch/arc/core/atomic.S Normal file
View File

@@ -0,0 +1,434 @@
/*
* Copyright (c) 2014 Wind River Systems, Inc.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
* @file
* @brief ARC atomic operations library
*
* This library provides routines to perform a number of atomic operations
* on a memory location: add, subtract, increment, decrement, bitwise OR,
* bitwise NOR, bitwise AND, bitwise NAND, set, clear and compare-and-swap.
*
* This requires the processor to support LLOCK and SCOND instructions,
* where they are not supported on ARC EM family processors.
*/
#define _ASMLANGUAGE
#include <toolchain.h>
#include <sections.h>
/* exports */
GTEXT(atomic_set)
GTEXT(atomic_get)
GTEXT(atomic_add)
GTEXT(atomic_nand)
GTEXT(atomic_and)
GTEXT(atomic_or)
GTEXT(atomic_xor)
GTEXT(atomic_clear)
GTEXT(atomic_dec)
GTEXT(atomic_inc)
GTEXT(atomic_sub)
GTEXT(atomic_cas)
.section .TEXT._Atomic, "ax"
.balign 2
/**
*
* @brief Atomically clear a memory location
*
* This routine atomically clears the contents of <target> and returns the old
* value that was in <target>.
*
* This routine can be used from both task and interrupt level.
*
* @return Contents of <target> before the atomic operation
*
* ERRNO: N/A
*
* atomic_val_t atomic_clear
* (
* atomic_t *target /@ memory location to clear @/
* )
*/
SECTION_SUBSEC_FUNC(TEXT, atomic_clear_set, atomic_clear)
mov_s r1, 0
/* fall through into atomic_set */
/**
*
* @brief Atomically set a memory location
*
* This routine atomically sets the contents of <target> to <value> and returns
* the old value that was in <target>.
*
* This routine can be used from both task and interrupt level.
*
* @return Contents of <target> before the atomic operation
*
* ERRNO: N/A
*
* atomic_val_t atomic_set
* (
* atomic_t *target, /@ memory location to set @/
* atomic_val_t value /@ set with this value @/
* )
*
*/
SECTION_SUBSEC_FUNC(TEXT, atomic_clear_set, atomic_set)
ex r1, [r0] /* swap new value with old value */
j_s.d [blink]
mov_s r0, r1 /* return old value */
/**
*
* @brief Get the value of a shared memory atomically
*
* This routine atomically retrieves the value in *target
*
* atomic_val_t atomic_get
* (
* atomic_t *target /@ address of atom to be retrieved @/
* )
*
* RETURN: value read from address target.
*
*/
SECTION_FUNC(TEXT, atomic_get)
ld_s r0, [r0, 0]
j_s [blink]
/**
*
* @brief Atomically increment a memory location
*
* This routine atomically increments the value in <target>. The operation is
* done using unsigned integer arithmetic. Various CPU architectures may impose
* restrictions with regards to the alignment and cache attributes of the
* atomic_t type.
*
* This routine can be used from both task and interrupt level.
*
* @return Contents of <target> before the atomic operation
*
* ERRNO: N/A
*
* atomic_val_t atomic_inc
* (
* atomic_t *target, /@ memory location to increment @/
* )
*
*/
SECTION_SUBSEC_FUNC(TEXT, atomic_inc_add, atomic_inc)
mov_s r1, 1
/* fall through into atomic_add */
/**
*
* @brief Atomically add a value to a memory location
*
* This routine atomically adds the contents of <target> and <value>, placing
* the result in <target>. The operation is done using signed integer arithmetic.
* Various CPU architectures may impose restrictions with regards to the
* alignment and cache attributes of the atomic_t type.
*
* This routine can be used from both task and interrupt level.
*
* @return Contents of <target> before the atomic operation
*
* ERRNO: N/A
*
* atomic_val_t atomic_add
* (
* atomic_t *target, /@ memory location to add to @/
* atomic_val_t value /@ value to add @/
* )
*/
SECTION_SUBSEC_FUNC(TEXT, atomic_inc_add, atomic_add)
llock r2, [r0] /* load old value and mark exclusive access */
add_s r3, r1, r2
scond r3, [r0] /* try to store new value */
/* STATUS32.Z = 1 if successful */
bne_s atomic_add /* if store is not successful, retry */
j_s.d [blink]
mov_s r0, r2 /* return old value */
/**
*
* @brief Atomically decrement a memory location
*
* This routine atomically decrements the value in <target>. The operation is
* done using unsigned integer arithmetic. Various CPU architectures may impose
* restrictions with regards to the alignment and cache attributes of the
* atomic_t type.
*
* This routine can be used from both task and interrupt level.
*
* @return Contents of <target> before the atomic operation
*
* ERRNO: N/A
*
* atomic_val_t atomic_dec
* (
* atomic_t *target, /@ memory location to decrement @/
* )
*
*/
SECTION_SUBSEC_FUNC(TEXT, atomic_dec_sub, atomic_dec)
mov_s r1, 1
/* fall through into atomic_sub */
/**
*
* @brief Atomically subtract a value from a memory location
*
* This routine atomically subtracts <value> from the contents of <target>,
* placing the result in <target>. The operation is done using signed integer
* arithmetic. Various CPU architectures may impose restrictions with regards to
* the alignment and cache attributes of the atomic_t type.
*
* This routine can be used from both task and interrupt level.
*
* @return Contents of <target> before the atomic operation
*
* ERRNO: N/A
*
* atomic_val_t atomic_sub
* (
* atomic_t *target, /@ memory location to subtract from @/
* atomic_val_t value /@ value to subtract @/
* )
*
*/
SECTION_SUBSEC_FUNC(TEXT, atomic_dec_sub, atomic_sub)
llock r2, [r0] /* load old value and mark exclusive access */
sub r3, r2, r1
scond r3, [r0] /* try to store new value */
/* STATUS32.Z = 1 if successful */
bne_s atomic_sub /* if store is not successful, retry */
j_s.d [blink]
mov_s r0, r2 /* return old value */
/**
*
* @brief Atomically perform a bitwise NAND on a memory location
*
* This routine atomically performs a bitwise NAND operation of the contents of
* <target> and <value>, placing the result in <target>.
* Various CPU architectures may impose restrictions with regards to the
* alignment and cache attributes of the atomic_t type.
*
* This routine can be used from both task and interrupt level.
*
* @return Contents of <target> before the atomic operation
*
* ERRNO: N/A
*
* atomic_val_t atomic_nand
* (
* atomic_t *target, /@ memory location to NAND @/
* atomic_val_t value /@ NAND with this value @/
* )
*
*/
SECTION_FUNC(TEXT, atomic_nand)
llock r2, [r0] /* load old value and mark exclusive access */
and r3, r1, r2
not r3, r3
scond r3, [r0] /* try to store new value */
/* STATUS32.Z = 1 if successful */
bne_s atomic_nand /* if store is not successful, retry */
j_s.d [blink]
mov_s r0, r2 /* return old value */
/**
*
* @brief Atomically perform a bitwise AND on a memory location
*
* This routine atomically performs a bitwise AND operation of the contents of
* <target> and <value>, placing the result in <target>.
* Various CPU architectures may impose restrictions with regards to the
* alignment and cache attributes of the atomic_t type.
*
* This routine can be used from both task and interrupt level.
*
* @return Contents of <target> before the atomic operation
*
* ERRNO: N/A
*
* atomic_val_t atomic_and
* (
* atomic_t *target, /@ memory location to AND @/
* atomic_val_t value /@ AND with this value @/
* )
*
*/
SECTION_FUNC(TEXT, atomic_and)
llock r2, [r0] /* load old value and mark exclusive access */
and r3, r1, r2
scond r3, [r0] /* try to store new value */
/* STATUS32.Z = 1 if successful */
bne_s atomic_and /* if store is not successful, retry */
j_s.d [blink]
mov_s r0, r2 /* return old value */
/**
*
* @brief Atomically perform a bitwise OR on memory location
*
* This routine atomically performs a bitwise OR operation of the contents of
* <target> and <value>, placing the result in <target>.
* Various CPU architectures may impose restrictions with regards to the
* alignment and cache attributes of the atomic_t type.
*
* This routine can be used from both task and interrupt level.
*
* @return Contents of <target> before the atomic operation
*
* ERRNO: N/A
*
* atomic_val_t atomic_or
* (
* atomic_t *target, /@ memory location to OR @/
* atomic_val_t value /@ OR with this value @/
* )
*
*/
SECTION_FUNC(TEXT, atomic_or)
llock r2, [r0] /* load old value and mark exclusive access */
or r3, r1, r2
scond r3, [r0] /* try to store new value */
/* STATUS32.Z = 1 if successful */
bne_s atomic_or /* if store is not successful, retry */
j_s.d [blink]
mov_s r0, r2 /* return old value */
/**
*
* @brief Atomically perform a bitwise XOR on a memory location
*
* This routine atomically performs a bitwise XOR operation of the contents of
* <target> and <value>, placing the result in <target>.
* Various CPU architectures may impose restrictions with regards to the
* alignment and cache attributes of the atomic_t type.
*
* This routine can be used from both task and interrupt level.
*
* @return Contents of <target> before the atomic operation
*
* ERRNO: N/A
*
* atomic_val_t atomic_xor
* (
* atomic_t *target, /@ memory location to XOR @/
* atomic_val_t value /@ XOR with this value @/
* )
*
*/
SECTION_FUNC(TEXT, atomic_xor)
llock r2, [r0] /* load old value and mark exclusive access */
xor r3, r1, r2
scond r3, [r0] /* try to store new value */
/* STATUS32.Z = 1 if successful */
bne_s atomic_xor /* if store is not successful, retry */
j_s.d [blink]
mov_s r0, r2 /* return old value */
/**
*
* @brief Atomically compare-and-swap the contents of a memory location
*
* This routine performs an atomic compare-and-swap. testing that the contents of
* <target> contains <oldValue>, and if it does, setting the value of <target>
* to <newValue>. Various CPU architectures may impose restrictions with regards
* to the alignment and cache attributes of the atomic_t type.
*
* This routine can be used from both task and interrupt level.
*
* @return 1 if the swap is actually executed, 0 otherwise.
*
* ERRNO: N/A
*
* int atomic_cas
* (
* atomic_t *target, /@ memory location to compare-and-swap @/
* atomic_val_t oldValue, /@ compare to this value @/
* atomic_val_t newValue, /@ swap with this value @/
* )
*
*/
SECTION_FUNC(TEXT, atomic_cas)
llock r3, [r0] /* load old value and mark exclusive access */
cmp_s r1, r3
bne_s nanoAtomicCas_fail
scond r2, [r0] /* try to store new value */
/* STATUS32.Z = 1 if successful */
bne_s atomic_cas /* if store is not successful, retry */
j_s.d [blink]
mov_s r0, 1 /* return TRUE */
/* failed comparison */
nanoAtomicCas_fail:
scond r1, [r0] /* write old value to clear the access lock */
j_s.d [blink]
mov_s r0, 0 /* return FALSE */

View File

@@ -3,7 +3,17 @@
/*
* Copyright (c) 2016 Synopsys, Inc. All rights reserved.
*
* SPDX-License-Identifier: Apache-2.0
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
@@ -13,17 +23,18 @@
* This module contains functions for manipulation of the d-cache.
*/
#include <kernel.h>
#include <nanokernel.h>
#include <arch/cpu.h>
#include <sys/util.h>
#include <misc/util.h>
#include <toolchain.h>
#include <cache.h>
#include <linker/linker-defs.h>
#include <linker-defs.h>
#include <arch/arc/v2/aux_regs.h>
#include <kernel_internal.h>
#include <sys/__assert.h>
#include <nano_internal.h>
#include <misc/__assert.h>
#include <init.h>
#include <stdbool.h>
#if defined(CONFIG_CACHE_FLUSHING)
#if (CONFIG_CACHE_LINE_SIZE == 0) && !defined(CONFIG_CACHE_LINE_SIZE_DETECT)
#error Cannot use this implementation with a cache line size of 0
@@ -47,19 +58,19 @@
#define DC_CTRL_OP_SUCCEEDED 0x4 /* d-cache operation succeeded */
static bool dcache_available(void)
static int dcache_available(void)
{
unsigned long val = z_arc_v2_aux_reg_read(_ARC_V2_D_CACHE_BUILD);
unsigned long val = _arc_v2_aux_reg_read(_ARC_V2_D_CACHE_BUILD);
val &= 0xff; /* extract version */
return (val == 0) ? false : true;
return (val == 0)?0:1;
}
static void dcache_dc_ctrl(u32_t dcache_en_mask)
static void dcache_dc_ctrl(uint32_t dcache_en_mask)
{
if (dcache_available()) {
z_arc_v2_aux_reg_write(_ARC_V2_DC_CTRL, dcache_en_mask);
}
if (!dcache_available())
return;
_arc_v2_aux_reg_write(_ARC_V2_DC_CTRL, dcache_en_mask);
}
static void dcache_enable(void)
@@ -84,36 +95,35 @@ static void dcache_enable(void)
*
* @return N/A
*/
static void dcache_flush_mlines(u32_t start_addr, u32_t size)
static void dcache_flush_mlines(uint32_t start_addr, uint32_t size)
{
u32_t end_addr;
uint32_t end_addr;
unsigned int key;
if (!dcache_available() || (size == 0U)) {
if (!dcache_available() || (size == 0)) {
return;
}
end_addr = start_addr + size - 1;
start_addr &= (u32_t)(~(DCACHE_LINE_SIZE - 1));
start_addr &= (uint32_t)(~(DCACHE_LINE_SIZE - 1));
key = arch_irq_lock(); /* --enter critical section-- */
key = irq_lock(); /* --enter critical section-- */
do {
z_arc_v2_aux_reg_write(_ARC_V2_DC_FLDL, start_addr);
_arc_v2_aux_reg_write(_ARC_V2_DC_FLDL, start_addr);
__asm__ volatile("nop_s");
__asm__ volatile("nop_s");
__asm__ volatile("nop_s");
/* wait for flush completion */
do {
if ((z_arc_v2_aux_reg_read(_ARC_V2_DC_CTRL) &
DC_CTRL_FLUSH_STATUS) == 0) {
if ((_arc_v2_aux_reg_read(_ARC_V2_DC_CTRL) &
DC_CTRL_FLUSH_STATUS) == 0)
break;
}
} while (1);
start_addr += DCACHE_LINE_SIZE;
} while (start_addr <= end_addr);
arch_irq_unlock(key); /* --exit critical section-- */
irq_unlock(key); /* --exit critical section-- */
}
@@ -137,7 +147,7 @@ static void dcache_flush_mlines(u32_t start_addr, u32_t size)
void sys_cache_flush(vaddr_t start_addr, size_t size)
{
dcache_flush_mlines((u32_t)start_addr, (u32_t)size);
dcache_flush_mlines((uint32_t)start_addr, (uint32_t)size);
}
@@ -145,12 +155,12 @@ void sys_cache_flush(vaddr_t start_addr, size_t size)
size_t sys_cache_line_size;
static void init_dcache_line_size(void)
{
u32_t val;
uint32_t val;
val = z_arc_v2_aux_reg_read(_ARC_V2_D_CACHE_BUILD);
__ASSERT((val&0xff) != 0U, "d-cache is not present");
val = _arc_v2_aux_reg_read(_ARC_V2_D_CACHE_BUILD);
__ASSERT((val&0xff) != 0, "d-cache is not present");
val = ((val>>16) & 0xf) + 1;
val *= 16U;
val *= 16;
sys_cache_line_size = (size_t) val;
}
#endif
@@ -169,3 +179,6 @@ static int init_dcache(struct device *unused)
}
SYS_INIT(init_dcache, PRE_KERNEL_1, CONFIG_KERNEL_INIT_PRIORITY_DEFAULT);
#endif /* CONFIG_CACHE_FLUSHING */

View File

@@ -1,7 +1,17 @@
/*
* Copyright (c) 2014 Wind River Systems, Inc.
*
* SPDX-License-Identifier: Apache-2.0
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
@@ -11,18 +21,19 @@
* CPU power management routines.
*/
#define _ASMLANGUAGE
#include <kernel_structs.h>
#include <offsets_short.h>
#include <toolchain.h>
#include <linker/sections.h>
#include <sections.h>
#include <arch/cpu.h>
GTEXT(arch_cpu_idle)
GTEXT(arch_cpu_atomic_idle)
GDATA(z_arc_cpu_sleep_mode)
GTEXT(nano_cpu_idle)
GTEXT(nano_cpu_atomic_idle)
GDATA(nano_cpu_sleep_mode)
SECTION_VAR(BSS, z_arc_cpu_sleep_mode)
.balign 4
SECTION_VAR(BSS, nano_cpu_sleep_mode)
.word 0
/*
@@ -33,30 +44,17 @@ SECTION_VAR(BSS, z_arc_cpu_sleep_mode)
* void nanCpuIdle(void)
*/
SECTION_FUNC(TEXT, arch_cpu_idle)
SECTION_FUNC(TEXT, nano_cpu_idle)
#ifdef CONFIG_TRACING
#ifdef CONFIG_KERNEL_EVENT_LOGGER_SLEEP
push_s blink
jl sys_trace_idle
jl _sys_k_event_logger_enter_sleep
pop_s blink
#endif
ld r1, [z_arc_cpu_sleep_mode]
ld r1, [nano_cpu_sleep_mode]
or r1, r1, (1 << 4) /* set IRQ-enabled bit */
/*
* It's found that (in nsim_hs_smp), when cpu
* is sleeping, no response to inter-processor interrupt
* although it's pending and interrupts are enabled.
* (Here fire SNPS JIRA issue P10019563-41294 to trace)
* here is a workround
*/
#if defined(CONFIG_SOC_NSIM) && defined(CONFIG_SMP)
seti r1
_z_arc_idle_loop:
b _z_arc_idle_loop
#else
sleep r1
#endif
j_s [blink]
nop
@@ -65,17 +63,17 @@ _z_arc_idle_loop:
*
* This function exits with interrupts restored to <key>.
*
* void arch_cpu_atomic_idle(unsigned int key)
* void nano_cpu_atomic_idle(unsigned int key)
*/
SECTION_FUNC(TEXT, arch_cpu_atomic_idle)
SECTION_FUNC(TEXT, nano_cpu_atomic_idle)
#ifdef CONFIG_TRACING
#ifdef CONFIG_KERNEL_EVENT_LOGGER_SLEEP
push_s blink
jl sys_trace_idle
jl _sys_k_event_logger_enter_sleep
pop_s blink
#endif
ld r1, [z_arc_cpu_sleep_mode]
ld r1, [nano_cpu_sleep_mode]
or r1, r1, (1 << 4) /* set IRQ-enabled bit */
sleep r1
j_s.d [blink]

View File

@@ -1,7 +1,17 @@
/*
* Copyright (c) 2014 Wind River Systems, Inc.
*
* SPDX-License-Identifier: Apache-2.0
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
@@ -13,15 +23,35 @@
* See isr_wrapper.S for details.
*/
#define _ASMLANGUAGE
#include <kernel_structs.h>
#include <offsets_short.h>
#include <toolchain.h>
#include <linker/sections.h>
#include <arch/cpu.h>
#include <swap_macros.h>
#include "swap_macros.h"
GTEXT(_firq_enter)
GTEXT(_firq_exit)
GTEXT(_firq_stack_setup)
#if CONFIG_RGF_NUM_BANKS != 1
GDATA(_firq_stack)
GTEXT(_is_next_thread_current)
SECTION_VAR(NOINIT, _firq_stack)
.space CONFIG_FIRQ_STACK_SIZE
#else
GDATA(saved_r0)
#endif
.macro _firq_return
#if CONFIG_RGF_NUM_BANKS == 1
b _firq_no_reschedule
#else
rtie
#endif
.endm
/**
*
@@ -46,27 +76,24 @@ GTEXT(_firq_exit)
*/
SECTION_FUNC(TEXT, _firq_enter)
/*
* ATTENTION:
* If CONFIG_RGF_NUM_BANKS>1, firq uses a 2nd register bank so GPRs do
* not need to be saved.
* If CONFIG_RGF_NUM_BANKS==1, firq must use the stack to save registers.
* This has already been done by _isr_wrapper.
* This has already been done by _isr_enter.
*/
#ifdef CONFIG_ARC_STACK_CHECKING
#ifdef CONFIG_ARC_SECURE_FIRMWARE
lr r2, [_ARC_V2_SEC_STAT]
bclr r2, r2, _ARC_V2_SEC_STAT_SSC_BIT
sflag r2
#else
/* disable stack checking */
lr r2, [_ARC_V2_STATUS32]
bclr r2, r2, _ARC_V2_STATUS32_SC_BIT
kflag r2
#endif
#endif
#if CONFIG_RGF_NUM_BANKS != 1
#ifndef CONFIG_FIRQ_NO_LPCC
/*
* Save LP_START/LP_COUNT/LP_END because called handler might use.
* Save these in callee saved registers to avoid using memory.
@@ -76,60 +103,10 @@ SECTION_FUNC(TEXT, _firq_enter)
lr r24, [_ARC_V2_LP_START]
lr r25, [_ARC_V2_LP_END]
#endif
/* check whether irq stack is used */
_check_and_inc_int_nest_counter r0, r1
bne.d firq_nest
mov_s r0, sp
_get_curr_cpu_irq_stack sp
#if CONFIG_RGF_NUM_BANKS != 1
b firq_nest_1
firq_nest:
/*
* because firq and rirq share the same interrupt stack,
* switch back to original register bank to get correct sp.
* to get better firq latency, an approach is to prepare
* separate interrupt stack for firq and do not do thread
* switch in firq.
*/
lr r1, [_ARC_V2_STATUS32]
and r1, r1, ~_ARC_V2_STATUS32_RB(7)
kflag r1
/* here use _ARC_V2_USER_SP and ilink to exchange sp
* save original value of _ARC_V2_USER_SP and ilink into
* the stack of interrupted context first, then restore them later
*/
push ilink
PUSHAX ilink, _ARC_V2_USER_SP
/* sp here is the sp of interrupted context */
sr sp, [_ARC_V2_USER_SP]
/* here, bank 0 sp must go back to the value before push and
* PUSHAX as we will switch to bank1, the pop and POPAX later will
* change bank1's sp, not bank0's sp
*/
add sp, sp, 8
/* switch back to banked reg, only ilink can be used */
lr ilink, [_ARC_V2_STATUS32]
or ilink, ilink, _ARC_V2_STATUS32_RB(1)
kflag ilink
lr sp, [_ARC_V2_USER_SP]
POPAX ilink, _ARC_V2_USER_SP
pop ilink
firq_nest_1:
#else
firq_nest:
#endif
push_s r0
j @_isr_demux
/**
*
* @brief Work to be done exiting a FIRQ
@@ -140,90 +117,113 @@ firq_nest:
SECTION_FUNC(TEXT, _firq_exit)
#if CONFIG_RGF_NUM_BANKS != 1
#ifndef CONFIG_FIRQ_NO_LPCC
/* restore lp_count, lp_start, lp_end from r23-r25 */
mov lp_count,r23
sr r24, [_ARC_V2_LP_START]
sr r25, [_ARC_V2_LP_END]
#endif
_dec_int_nest_counter r0, r1
#endif
_check_nest_int_by_irq_act r0, r1
mov_s r1, _kernel
ld_s r2, [r1, _kernel_offset_to_current]
jne _firq_no_switch
#if CONFIG_NUM_IRQ_PRIO_LEVELS > 1
/* check if we're a nested interrupt: if so, let the interrupted
* interrupt handle the reschedule */
/* sp is struct k_thread **old of z_arc_switch_in_isr
* which is a wrapper of z_get_next_switch_handle.
* r0 contains the 1st thread in ready queue. if
* it equals _current(r2) ,then do swap, or no swap.
lr r3, [_ARC_V2_AUX_IRQ_ACT]
/* the OS on ARCv2 always runs in kernel mode, so assume bit31 [U] in
* AUX_IRQ_ACT is always 0: if the contents of AUX_IRQ_ACT is not 1, it
* means that another bit is set so an interrupt was interrupted.
*/
_get_next_switch_handle
/* restore interrupted context' sp */
pop sp
breq r3, 1, _firq_check_for_swap
cmp r0, r2
bne _firq_switch
/* fall to no switch */
_firq_return
#endif
.balign 4
_firq_no_switch:
_firq_check_for_swap:
/* coop thread ? do not schedule */
ld_s r0, [r2, _thread_offset_to_prio]
brlt r0, 0, _firq_no_reschedule
/* scheduler locked ? do not schedule */
ld_s r0, [r2, _thread_offset_to_sched_locked]
brgt r0, 0, _firq_no_reschedule
/* check if the current thread needs to be rescheduled */
push_s r2
push_s r1
push_s blink
jl _is_next_thread_current
pop_s blink
pop_s r1
pop_s r2
#if CONFIG_RGF_NUM_BANKS != 1
#ifndef CONFIG_FIRQ_NO_LPCC
/*
* restore lp_count, lp_start, lp_end from r23-r25 in case
* _is_next_thread_current() routine used them
*/
mov lp_count,r23
sr r24, [_ARC_V2_LP_START]
sr r25, [_ARC_V2_LP_END]
#endif
#endif
breq r0, 0, _firq_reschedule
/* fall to no rescheduling */
.balign 4
_firq_no_reschedule:
/*
* Keeping this code block close to those that use it allows using brxx
* instruction instead of a pair of cmp and bxx
*/
#if CONFIG_RGF_NUM_BANKS == 1
_pop_irq_stack_frame
add sp,sp,4 /* don't need r0 from stack */
pop_s r1
pop_s r2
pop_s r3
pop r4
pop r5
pop r6
pop r7
pop r8
pop r9
pop r10
pop r11
pop_s r12
pop_s r13
pop_s blink
pop_s r0
sr r0, [_ARC_V2_LP_END]
pop_s r0
sr r0, [_ARC_V2_LP_START]
pop_s r0
mov lp_count,r0
ld r0,[saved_r0]
add sp,sp,8 /* don't need ilink & status32_po from stack */
#endif
rtie
.balign 4
_firq_switch:
_firq_reschedule:
#if CONFIG_RGF_NUM_BANKS != 1
/*
* save r0, r2 in irq stack for a while, as they will be changed by register
* bank switch
*/
_get_curr_cpu_irq_stack r1
st r0, [r1, -4]
st r2, [r1, -8]
/*
* We know there is no interrupted interrupt of lower priority at this
* point, so when switching back to register bank 0, it will contain the
* registers from the interrupted thread.
*/
#if defined(CONFIG_USERSPACE)
/* when USERSPACE is configured, here need to consider the case where firq comes
* out in user mode, according to ARCv2 ISA and nsim, the following micro ops
* will be executed:
* sp<-reg bank1'sp
* switch between sp and _ARC_V2_USER_SP
* then:
* sp is the sp of kernel stack of interrupted thread
* _ARC_V2_USER_SP is reg bank1'sp
* the sp of user stack of interrupted thread is reg bank0'sp
* if firq comes out in kernel mode, the following micro ops will be executed:
* sp<-reg bank'sp
* so, sw needs to do necessary handling to set up the correct sp
*/
lr r0, [_ARC_V2_AUX_IRQ_ACT]
bbit0 r0, 31, _firq_from_kernel
aex sp, [_ARC_V2_USER_SP]
lr r0, [_ARC_V2_STATUS32]
and r0, r0, ~_ARC_V2_STATUS32_RB(7)
kflag r0
aex sp, [_ARC_V2_USER_SP]
b _firq_create_irq_stack_frame
_firq_from_kernel:
#endif
/* chose register bank #0 */
lr r0, [_ARC_V2_STATUS32]
and r0, r0, ~_ARC_V2_STATUS32_RB(7)
kflag r0
_firq_create_irq_stack_frame:
/* we're back on the outgoing thread's stack */
_create_irq_stack_frame
@@ -236,49 +236,79 @@ _firq_create_irq_stack_frame:
st_s r0, [sp, ___isf_t_status32_OFFSET]
st ilink, [sp, ___isf_t_pc_OFFSET] /* ilink into pc */
/*
* load r0, r2 from irq stack
*/
_get_curr_cpu_irq_stack r1
ld r0, [r1, -4]
ld r2, [r1, -8]
#endif
/* r2 is old thread */
_irq_store_old_thread_callee_regs
mov_s r1, _kernel
ld_s r2, [r1, _kernel_offset_to_current]
_save_callee_saved_regs
st _CAUSE_FIRQ, [r2, _thread_offset_to_relinquish_cause]
/* mov new thread (r0) to r2 */
/*
* Save needed registers to callee saved ones. It is faster than
* pushing them to stack. It is possible to do since program has
* just saved them and the calling routine will save them in turn
* if it uses them.
*/
mov_s r13, blink
mov_s r14, r1
jl _get_next_ready_thread
mov_s blink, r13
mov_s r1, r14
mov_s r2, r0
st_s r2, [r1, _kernel_offset_to_current]
mov r2, r0
_load_new_thread_callee_regs
#ifdef CONFIG_ARC_STACK_CHECKING
/* Use stack top and down registers from restored context */
add r3, r2, _K_THREAD_NO_FLOAT_SIZEOF
sr r3, [_ARC_V2_KSTACK_TOP]
ld_s r3, [r2, _thread_offset_to_stack_top]
sr r3, [_ARC_V2_KSTACK_BASE]
#endif
/*
* _load_callee_saved_regs expects incoming thread in r2.
* _load_callee_saved_regs restores the stack pointer.
*/
_load_callee_saved_regs
breq r3, _CAUSE_RIRQ, _firq_switch_from_rirq
nop_s
breq r3, _CAUSE_FIRQ, _firq_switch_from_firq
nop_s
ld_s r3, [r2, _thread_offset_to_relinquish_cause]
breq r3, _CAUSE_RIRQ, _firq_return_from_rirq
nop
breq r3, _CAUSE_FIRQ, _firq_return_from_firq
nop
/* fall through */
.balign 4
_firq_switch_from_coop:
_firq_return_from_coop:
_set_misc_regs_irq_switch_from_coop
ld_s r3, [r2, _thread_offset_to_intlock_key]
st 0, [r2, _thread_offset_to_intlock_key]
/* pc into ilink */
pop_s r0
mov_s ilink, r0
mov ilink, r0
pop_s r0 /* status32 into r0 */
/*
* There are only two interrupt lock states: locked and unlocked. When
* entering _Swap(), they are always locked, so the IE bit is unset in
* status32. If the incoming thread had them locked recursively, it
* means that the IE bit should stay unset. The only time the bit
* has to change is if they were not locked recursively.
*/
and.f r3, r3, (1 << 4)
or.nz r0, r0, _ARC_V2_STATUS32_IE
sr r0, [_ARC_V2_STATUS32_P0]
ld_s r0, [r2, _thread_offset_to_return_value]
rtie
.balign 4
_firq_switch_from_rirq:
_firq_switch_from_firq:
_set_misc_regs_irq_switch_from_irq
_firq_return_from_rirq:
_firq_return_from_firq:
_pop_irq_stack_frame
@@ -288,3 +318,33 @@ _firq_switch_from_firq:
/* LP registers are already restored, just switch back to bank 0 */
rtie
/**
*
* @brief Install the FIRQ stack in register bank 1 if CONFIG_RGF_NUM_BANK!=1
*
* @return N/A
*/
SECTION_FUNC(TEXT, _firq_stack_setup)
#if CONFIG_RGF_NUM_BANKS != 1
lr r0, [_ARC_V2_STATUS32]
and r0, r0, ~_ARC_V2_STATUS32_RB(7)
or r0, r0, _ARC_V2_STATUS32_RB(1)
kflag r0
mov sp, _firq_stack
add sp, sp, CONFIG_FIRQ_STACK_SIZE
/*
* We have to reload r0 here, because it is bank1 r0 which contains
* garbage, not bank0 r0 containing the previous value of status32.
*/
lr r0, [_ARC_V2_STATUS32]
and r0, r0, ~_ARC_V2_STATUS32_RB(7)
kflag r0
#endif
j_s [blink]

View File

@@ -1,7 +1,17 @@
/*
* Copyright (c) 2014 Wind River Systems, Inc.
*
* SPDX-License-Identifier: Apache-2.0
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
@@ -12,33 +22,74 @@
* ARCv2 CPUs.
*/
#include <kernel.h>
#include <kernel_structs.h>
#include <offsets_short.h>
#include <toolchain.h>
#include <arch/cpu.h>
#include <logging/log.h>
LOG_MODULE_DECLARE(os);
void z_arc_fatal_error(unsigned int reason, const z_arch_esf_t *esf)
#ifdef CONFIG_PRINTK
#include <misc/printk.h>
#define PR_EXC(...) printk(__VA_ARGS__)
#else
#define PR_EXC(...)
#endif /* CONFIG_PRINTK */
const NANO_ESF _default_esf = {
0xdeaddead, /* placeholder */
};
/**
*
* @brief Nanokernel fatal error handler
*
* This routine is called when fatal error conditions are detected by software
* and is responsible only for reporting the error. Once reported, it then
* invokes the user provided routine _SysFatalErrorHandler() which is
* responsible for implementing the error handling policy.
*
* The caller is expected to always provide a usable ESF. In the event that the
* fatal error does not have a hardware generated ESF, the caller should either
* create its own or use a pointer to the global default ESF <_default_esf>.
*
* @return This function does not return.
*/
FUNC_NORETURN void _NanoFatalErrorHandler(unsigned int reason,
const NANO_ESF *pEsf)
{
if (reason == K_ERR_CPU_EXCEPTION) {
LOG_ERR("Faulting instruction address = 0x%lx",
z_arc_v2_aux_reg_read(_ARC_V2_ERET));
switch (reason) {
case _NANO_ERR_INVALID_TASK_EXIT:
PR_EXC("***** Invalid Exit Software Error! *****\n");
break;
#if defined(CONFIG_STACK_CANARIES)
case _NANO_ERR_STACK_CHK_FAIL:
PR_EXC("***** Stack Check Fail! *****\n");
break;
#endif
case _NANO_ERR_ALLOCATION_FAIL:
PR_EXC("**** Kernel Allocation Failure! ****\n");
break;
default:
PR_EXC("**** Unknown Fatal Error %d! ****\n", reason);
break;
}
PR_EXC("Current thread ID = %p\n"
"Faulting instruction address = 0x%lx\n",
k_current_get(),
_arc_v2_aux_reg_read(_ARC_V2_ERET));
z_fatal_error(reason, esf);
}
FUNC_NORETURN void arch_syscall_oops(void *ssf_ptr)
{
z_arc_fatal_error(K_ERR_KERNEL_OOPS, ssf_ptr);
CODE_UNREACHABLE;
}
FUNC_NORETURN void arch_system_halt(unsigned int reason)
{
ARG_UNUSED(reason);
__asm__("brk");
CODE_UNREACHABLE;
/*
* Now that the error has been reported, call the user implemented
* policy
* to respond to the error. The decisions as to what responses are
* appropriate to the various errors are something the customer must
* decide.
*/
_SysFatalErrorHandler(reason, pEsf);
for (;;)
;
}

View File

@@ -1,7 +1,17 @@
/*
* Copyright (c) 2014 Wind River Systems, Inc.
*
* SPDX-License-Identifier: Apache-2.0
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
@@ -12,423 +22,70 @@
*/
#include <toolchain.h>
#include <linker/sections.h>
#include <sections.h>
#include <inttypes.h>
#include <kernel.h>
#include <kernel_internal.h>
#include <nanokernel.h>
#include <kernel_structs.h>
#include <exc_handle.h>
#include <logging/log.h>
LOG_MODULE_DECLARE(os);
#ifdef CONFIG_USERSPACE
Z_EXC_DECLARE(z_arc_user_string_nlen);
static const struct z_exc_handle exceptions[] = {
Z_EXC_HANDLE(z_arc_user_string_nlen)
};
#endif
#if defined(CONFIG_MPU_STACK_GUARD)
#define IS_MPU_GUARD_VIOLATION(guard_start, fault_addr, stack_ptr) \
((fault_addr >= guard_start) && \
(fault_addr < (guard_start + STACK_GUARD_SIZE)) && \
(stack_ptr <= (guard_start + STACK_GUARD_SIZE)))
/**
* @brief Assess occurrence of current thread's stack corruption
*
* This function performs an assessment whether a memory fault (on a
* given memory address) is the result of stack memory corruption of
* the current thread.
*
* Thread stack corruption for supervisor threads or user threads in
* privilege mode (when User Space is supported) is reported upon an
* attempt to access the stack guard area (if MPU Stack Guard feature
* is supported). Additionally the current thread stack pointer
* must be pointing inside or below the guard area.
*
* Thread stack corruption for user threads in user mode is reported,
* if the current stack pointer is pointing below the start of the current
* thread's stack.
*
* Notes:
* - we assume a fully descending stack,
* - we assume a stacking error has occurred,
* - the function shall be called when handling MPU privilege violation
*
* If stack corruption is detected, the function returns the lowest
* allowed address where the Stack Pointer can safely point to, to
* prevent from errors when un-stacking the corrupted stack frame
* upon exception return.
*
* @param fault_addr memory address on which memory access violation
* has been reported.
* @param sp stack pointer when exception comes out
*
* @return The lowest allowed stack frame pointer, if error is a
* thread stack corruption, otherwise return 0.
*/
static u32_t z_check_thread_stack_fail(const u32_t fault_addr, u32_t sp)
{
const struct k_thread *thread = _current;
if (!thread) {
return 0;
}
#if defined(CONFIG_USERSPACE)
if (thread->arch.priv_stack_start) {
/* User thread */
if (z_arc_v2_aux_reg_read(_ARC_V2_ERSTATUS)
& _ARC_V2_STATUS32_U) {
/* Thread's user stack corruption */
#ifdef CONFIG_ARC_HAS_SECURE
sp = z_arc_v2_aux_reg_read(_ARC_V2_SEC_U_SP);
#ifdef CONFIG_PRINTK
#include <misc/printk.h>
#define PR_EXC(...) printk(__VA_ARGS__)
#else
sp = z_arc_v2_aux_reg_read(_ARC_V2_USER_SP);
#endif
if (sp <= (u32_t)thread->stack_obj) {
return (u32_t)thread->stack_obj;
}
} else {
/* User thread in privilege mode */
if (IS_MPU_GUARD_VIOLATION(
thread->arch.priv_stack_start - STACK_GUARD_SIZE,
fault_addr, sp)) {
/* Thread's privilege stack corruption */
return thread->arch.priv_stack_start;
}
}
} else {
/* Supervisor thread */
if (IS_MPU_GUARD_VIOLATION((u32_t)thread->stack_obj,
fault_addr, sp)) {
/* Supervisor thread stack corruption */
return (u32_t)thread->stack_obj + STACK_GUARD_SIZE;
}
}
#else /* CONFIG_USERSPACE */
if (IS_MPU_GUARD_VIOLATION(thread->stack_info.start,
fault_addr, sp)) {
/* Thread stack corruption */
return thread->stack_info.start + STACK_GUARD_SIZE;
}
#endif /* CONFIG_USERSPACE */
return 0;
}
#define PR_EXC(...)
#endif /* CONFIG_PRINTK */
#if (CONFIG_FAULT_DUMP > 0)
#define FAULT_DUMP(esf, fault) _FaultDump(esf, fault)
#else
#define FAULT_DUMP(esf, fault) \
do { \
(void) esf; \
(void) fault; \
} while ((0))
#endif
#ifdef CONFIG_ARC_EXCEPTION_DEBUG
/* For EV_ProtV, the numbering/semantics of the parameter are consistent across
* several codes, although not all combination will be reported.
#if (CONFIG_FAULT_DUMP > 0)
/*
* @brief Dump information regarding fault (FAULT_DUMP > 0)
*
* These codes and parameters do not have associated* names in
* the technical manual, just switch on the values in Table 6-5
* Dump information regarding the fault when CONFIG_FAULT_DUMP is set to 1
* (short form).
*
* @return N/A
*/
static const char *get_protv_access_err(u32_t parameter)
void _FaultDump(const NANO_ESF *esf, int fault)
{
switch (parameter) {
case 0x1:
return "code protection scheme";
case 0x2:
return "stack checking scheme";
case 0x4:
return "MPU";
case 0x8:
return "MMU";
case 0x10:
return "NVM";
case 0x24:
return "Secure MPU";
case 0x44:
return "Secure MPU with SID mismatch";
default:
return "unknown";
}
}
ARG_UNUSED(esf);
#ifdef CONFIG_PRINTK
uint32_t exc_addr = _arc_v2_aux_reg_read(_ARC_V2_EFA);
uint32_t ecr = _arc_v2_aux_reg_read(_ARC_V2_ECR);
static void dump_protv_exception(u32_t cause, u32_t parameter)
{
switch (cause) {
case 0x0:
LOG_ERR("Instruction fetch violation (%s)",
get_protv_access_err(parameter));
break;
case 0x1:
LOG_ERR("Memory read protection violation (%s)",
get_protv_access_err(parameter));
break;
case 0x2:
LOG_ERR("Memory write protection violation (%s)",
get_protv_access_err(parameter));
break;
case 0x3:
LOG_ERR("Memory read-modify-write violation (%s)",
get_protv_access_err(parameter));
break;
case 0x10:
LOG_ERR("Normal vector table in secure memory");
break;
case 0x11:
LOG_ERR("NS handler code located in S memory");
break;
case 0x12:
LOG_ERR("NSC Table Range Violation");
break;
default:
LOG_ERR("unknown");
break;
}
PR_EXC("Exception vector: 0x%" PRIx32 ", cause code: 0x%" PRIx32
", parameter 0x%" PRIx32 "\n",
_ARC_V2_ECR_VECTOR(ecr),
_ARC_V2_ECR_CODE(ecr),
_ARC_V2_ECR_PARAMETER(ecr));
PR_EXC("Address 0x%" PRIx32 "\n", exc_addr);
#endif
}
static void dump_machine_check_exception(u32_t cause, u32_t parameter)
{
switch (cause) {
case 0x0:
LOG_ERR("double fault");
break;
case 0x1:
LOG_ERR("overlapping TLB entries");
break;
case 0x2:
LOG_ERR("fatal TLB error");
break;
case 0x3:
LOG_ERR("fatal cache error");
break;
case 0x4:
LOG_ERR("internal memory error on instruction fetch");
break;
case 0x5:
LOG_ERR("internal memory error on data fetch");
break;
case 0x6:
LOG_ERR("illegal overlapping MPU entries");
if (parameter == 0x1) {
LOG_ERR(" - jump and branch target");
}
break;
case 0x10:
LOG_ERR("secure vector table not located in secure memory");
break;
case 0x11:
LOG_ERR("NSC jump table not located in secure memory");
break;
case 0x12:
LOG_ERR("secure handler code not located in secure memory");
break;
case 0x13:
LOG_ERR("NSC target address not located in secure memory");
break;
case 0x80:
LOG_ERR("uncorrectable ECC or parity error in vector memory");
break;
default:
LOG_ERR("unknown");
break;
}
}
static void dump_privilege_exception(u32_t cause, u32_t parameter)
{
switch (cause) {
case 0x0:
LOG_ERR("Privilege violation");
break;
case 0x1:
LOG_ERR("disabled extension");
break;
case 0x2:
LOG_ERR("action point hit");
break;
case 0x10:
switch (parameter) {
case 0x1:
LOG_ERR("N to S return using incorrect return mechanism");
break;
case 0x2:
LOG_ERR("N to S return with incorrect operating mode");
break;
case 0x3:
LOG_ERR("IRQ/exception return fetch from wrong mode");
break;
case 0x4:
LOG_ERR("attempt to halt secure processor in NS mode");
break;
case 0x20:
LOG_ERR("attempt to access secure resource from normal mode");
break;
case 0x40:
LOG_ERR("SID violation on resource access (APEX/UAUX/key NVM)");
break;
default:
LOG_ERR("unknown");
break;
}
break;
case 0x13:
switch (parameter) {
case 0x20:
LOG_ERR("attempt to access secure APEX feature from NS mode");
break;
case 0x40:
LOG_ERR("SID violation on access to APEX feature");
break;
default:
LOG_ERR("unknown");
break;
}
break;
default:
LOG_ERR("unknown");
break;
}
}
static void dump_exception_info(u32_t vector, u32_t cause, u32_t parameter)
{
if (vector >= 0x10 && vector <= 0xFF) {
LOG_ERR("interrupt %u", vector);
return;
}
/* Names are exactly as they appear in Designware ARCv2 ISA
* Programmer's reference manual for easy searching
*/
switch (vector) {
case ARC_EV_RESET:
LOG_ERR("Reset");
break;
case ARC_EV_MEM_ERROR:
LOG_ERR("Memory Error");
break;
case ARC_EV_INS_ERROR:
LOG_ERR("Instruction Error");
break;
case ARC_EV_MACHINE_CHECK:
LOG_ERR("EV_MachineCheck");
dump_machine_check_exception(cause, parameter);
break;
case ARC_EV_TLB_MISS_I:
LOG_ERR("EV_TLBMissI");
break;
case ARC_EV_TLB_MISS_D:
LOG_ERR("EV_TLBMissD");
break;
case ARC_EV_PROT_V:
LOG_ERR("EV_ProtV");
dump_protv_exception(cause, parameter);
break;
case ARC_EV_PRIVILEGE_V:
LOG_ERR("EV_PrivilegeV");
dump_privilege_exception(cause, parameter);
break;
case ARC_EV_SWI:
LOG_ERR("EV_SWI");
break;
case ARC_EV_TRAP:
LOG_ERR("EV_Trap");
break;
case ARC_EV_EXTENSION:
LOG_ERR("EV_Extension");
break;
case ARC_EV_DIV_ZERO:
LOG_ERR("EV_DivZero");
break;
case ARC_EV_DC_ERROR:
LOG_ERR("EV_DCError");
break;
case ARC_EV_MISALIGNED:
LOG_ERR("EV_Misaligned");
break;
case ARC_EV_VEC_UNIT:
LOG_ERR("EV_VecUnit");
break;
default:
LOG_ERR("unknown");
break;
}
}
#endif /* CONFIG_ARC_EXCEPTION_DEBUG */
#endif /* CONFIG_FAULT_DUMP */
/*
* @brief Fault handler
*
* This routine is called when fatal error conditions are detected by hardware
* and is responsible only for reporting the error. Once reported, it then
* invokes the user provided routine k_sys_fatal_error_handler() which is
* invokes the user provided routine _SysFatalErrorHandler() which is
* responsible for implementing the error handling policy.
*
* @return This function does not return.
*/
void _Fault(z_arch_esf_t *esf, u32_t old_sp)
void _Fault(void)
{
u32_t vector, cause, parameter;
u32_t exc_addr = z_arc_v2_aux_reg_read(_ARC_V2_EFA);
u32_t ecr = z_arc_v2_aux_reg_read(_ARC_V2_ECR);
uint32_t ecr = _arc_v2_aux_reg_read(_ARC_V2_ECR);
#ifdef CONFIG_USERSPACE
for (int i = 0; i < ARRAY_SIZE(exceptions); i++) {
u32_t start = (u32_t)exceptions[i].start;
u32_t end = (u32_t)exceptions[i].end;
FAULT_DUMP(&_default_esf, ecr);
if (esf->pc >= start && esf->pc < end) {
esf->pc = (u32_t)(exceptions[i].fixup);
return;
}
}
#endif
vector = Z_ARC_V2_ECR_VECTOR(ecr);
cause = Z_ARC_V2_ECR_CODE(ecr);
parameter = Z_ARC_V2_ECR_PARAMETER(ecr);
/* exception raised by kernel */
if (vector == ARC_EV_TRAP && parameter == _TRAP_S_CALL_RUNTIME_EXCEPT) {
/*
* in user mode software-triggered system fatal exceptions only allow
* K_ERR_KERNEL_OOPS and K_ERR_STACK_CHK_FAIL
*/
#ifdef CONFIG_USERSPACE
if ((esf->status32 & _ARC_V2_STATUS32_U) &&
esf->r0 != K_ERR_STACK_CHK_FAIL) {
esf->r0 = K_ERR_KERNEL_OOPS;
}
#endif
z_arc_fatal_error(esf->r0, esf);
return;
}
LOG_ERR("***** Exception vector: 0x%x, cause code: 0x%x, parameter 0x%x",
vector, cause, parameter);
LOG_ERR("Address 0x%x", exc_addr);
#ifdef CONFIG_ARC_EXCEPTION_DEBUG
dump_exception_info(vector, cause, parameter);
#endif
#ifdef CONFIG_ARC_STACK_CHECKING
/* Vector 6 = EV_ProV. Regardless of cause, parameter 2 means stack
* check violation
* stack check and mpu violation can come out together, then
* parameter = 0x2 | [0x4 | 0x8 | 0x1]
*/
if (vector == ARC_EV_PROT_V && parameter & 0x2) {
z_arc_fatal_error(K_ERR_STACK_CHK_FAIL, esf);
return;
}
#endif
#ifdef CONFIG_MPU_STACK_GUARD
if (vector == ARC_EV_PROT_V && ((parameter == 0x4) ||
(parameter == 0x24))) {
if (z_check_thread_stack_fail(exc_addr, old_sp)) {
z_arc_fatal_error(K_ERR_STACK_CHK_FAIL, esf);
return;
}
}
#endif
z_arc_fatal_error(K_ERR_CPU_EXCEPTION, esf);
_SysFatalErrorHandler(_NANO_ERR_HW_EXCEPTION, &_default_esf);
}

View File

@@ -1,8 +1,17 @@
/*
* Copyright (c) 2014 Wind River Systems, Inc.
* Copyright (c) 2018 Synopsys.
*
* SPDX-License-Identifier: Apache-2.0
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
@@ -12,13 +21,15 @@
* Fault handlers for ARCv2 processors.
*/
#define _ASMLANGUAGE
#include <toolchain.h>
#include <linker/sections.h>
#include <sections.h>
#include <arch/cpu.h>
#include <swap_macros.h>
#include <syscall.h>
#include "swap_macros.h"
GTEXT(_Fault)
GTEXT(__reset)
GTEXT(__memory_error)
GTEXT(__instruction_error)
@@ -33,31 +44,17 @@ GTEXT(__ev_extension)
GTEXT(__ev_div_zero)
GTEXT(__ev_dc_error)
GTEXT(__ev_maligned)
#ifdef CONFIG_IRQ_OFFLOAD
GTEXT(z_irq_do_offload);
#endif
.macro _save_exc_regs_into_stack
#ifdef CONFIG_ARC_HAS_SECURE
/* ERSEC_STAT is IOW/RAZ in normal mode */
lr r0,[_ARC_V2_ERSEC_STAT]
st_s r0, [sp, ___isf_t_sec_stat_OFFSET]
#endif
lr r0,[_ARC_V2_ERET]
st_s r0, [sp, ___isf_t_pc_OFFSET]
lr r0,[_ARC_V2_ERSTATUS]
st_s r0, [sp, ___isf_t_status32_OFFSET]
.endm
SECTION_VAR(BSS, saved_stack_pointer)
.word 0
/*
* The exception handling will use top part of interrupt stack to
* get smaller memory footprint, because exception is not frequent.
* To reduce the impact on interrupt handling, especially nested interrupt
* the top part of interrupt stack cannot be too large, so add a check
* here
*/
#if CONFIG_ARC_EXCEPTION_STACK_SIZE > (CONFIG_ISR_STACK_SIZE >> 1)
#error "interrupt stack size is too small"
#if CONFIG_RGF_NUM_BANKS == 1
GDATA(_exception_stack)
SECTION_VAR(NOINIT, _exception_stack)
.space 512
/* note: QUARK_SE_C1000_SS can't afford 512B */
#else
GDATA(_firq_stack)
#endif
/*
@@ -72,196 +69,153 @@ SECTION_SUBSEC_FUNC(TEXT,__fault,__ev_tlb_miss_d)
SECTION_SUBSEC_FUNC(TEXT,__fault,__ev_prot_v)
SECTION_SUBSEC_FUNC(TEXT,__fault,__ev_privilege_v)
SECTION_SUBSEC_FUNC(TEXT,__fault,__ev_swi)
#ifndef CONFIG_IRQ_OFFLOAD
SECTION_SUBSEC_FUNC(TEXT,__fault,__ev_trap)
#endif
SECTION_SUBSEC_FUNC(TEXT,__fault,__ev_extension)
SECTION_SUBSEC_FUNC(TEXT,__fault,__ev_div_zero)
SECTION_SUBSEC_FUNC(TEXT,__fault,__ev_dc_error)
SECTION_SUBSEC_FUNC(TEXT,__fault,__ev_maligned)
_exc_entry:
/*
* re-use the top part of interrupt stack as exception
* stack. If this top part is used by interrupt handling,
* and exception is raised, then here it's guaranteed that
* exception handling has necessary stack to use
*/
mov_s ilink, sp
_get_curr_cpu_irq_stack sp
sub sp, sp, (CONFIG_ISR_STACK_SIZE - CONFIG_ARC_EXCEPTION_STACK_SIZE)
/*
* save caller saved registers
* this stack frame is set up in exception stack,
* not in the original sp (thread stack or interrupt stack).
* Because the exception may be raised by stack checking or
* mpu protect violation related to stack. If this stack frame
* is setup in original sp, double exception may be raised during
* _create_irq_stack_frame, which is unrecoverable.
*/
_create_irq_stack_frame
_save_exc_regs_into_stack
/* sp is parameter of _Fault */
mov_s r0, sp
/* ilink is the thread's original sp */
mov_s r1, ilink
jl _Fault
_exc_return:
/* the exception cause must be fixed in exception handler when exception returns
* directly, or exception will be repeated.
*
* If thread switch is raised in exception handler, the context of old thread will
* not be saved, i.e., it cannot be recovered, because we don't know where the
* exception comes out, thread context?irq_context?nest irq context?
*/
_get_next_switch_handle
breq r0, r2, _exc_return_from_exc
mov_s r2, r0
#ifdef CONFIG_ARC_SECURE_FIRMWARE
/*
* sync up the ERSEC_STAT.ERM and SEC_STAT.IRM.
* use a fake interrupt return to simulate an exception turn.
* ERM and IRM record which mode the cpu should return, 1: secure
* 0: normal
*/
lr r3,[_ARC_V2_ERSEC_STAT]
btst r3, 31
bset.nz r3, r3, _ARC_V2_SEC_STAT_IRM_BIT
bclr.z r3, r3, _ARC_V2_SEC_STAT_IRM_BIT
sflag r3
#endif
/* clear AE bit to forget this was an exception, and go to
* register bank0 (if exception is raised in firq with 2 reg
* banks, then we may be bank1)
*/
#if defined(CONFIG_ARC_FIRQ) && CONFIG_RGF_NUM_BANKS != 1
/* save r2 in ilink because of the possible following reg
* bank switch
*/
mov_s ilink, r2
#endif
lr r3, [_ARC_V2_STATUS32]
and r3,r3,(~(_ARC_V2_STATUS32_AE | _ARC_V2_STATUS32_RB(7)))
kflag r3
/* pretend lowest priority interrupt happened to use common handler
* if exception is raised in irq, i.e., _ARC_V2_AUX_IRQ_ACT !=0,
* ignore irq handling, we cannot return to irq handling which may
* raise exception again. The ignored interrupts will be re-triggered
* if not cleared, or re-triggered by interrupt sources, or just missed
* Before invoking exception handler, the kernel switches to an exception
* stack, to save the faulting thread's registers.
* The exception is fatal and all the kernel can do is just print
* a diagnostic message and halt.
*/
#ifdef CONFIG_ARC_SECURE_FIRMWARE
mov_s r3, (1 << (ARC_N_IRQ_START_LEVEL - 1))
#else
mov_s r3, (1 << (CONFIG_NUM_IRQ_PRIO_LEVELS - 1))
#endif
#ifdef CONFIG_ARC_NORMAL_FIRMWARE
#ifdef CONFIG_ARC_STACK_CHECKING
push_s r2
mov_s r0, _ARC_V2_AUX_IRQ_ACT
mov_s r1, r3
mov_s r6, ARC_S_CALL_AUX_WRITE
sjli SJLI_CALL_ARC_SECURE
/* disable stack checking */
lr r2, [_ARC_V2_STATUS32]
bclr r2, r2, _ARC_V2_STATUS32_SC_BIT
kflag r2
pop_s r2
#endif
st sp, [saved_stack_pointer]
#if CONFIG_RGF_NUM_BANKS == 1
mov_s sp, _exception_stack
add sp, sp, 512
#else
sr r3, [_ARC_V2_AUX_IRQ_ACT]
mov_s sp, _firq_stack
add sp, sp, CONFIG_FIRQ_STACK_SIZE
#endif
#if defined(CONFIG_ARC_FIRQ) && CONFIG_RGF_NUM_BANKS != 1
mov r2, ilink
#endif
/* Assumption: r2 has next thread */
b _rirq_newthread_switch
_exc_return_from_exc:
/* exception handler may change return address.
* reload it
*/
ld_s r0, [sp, ___isf_t_pc_OFFSET]
sr r0, [_ARC_V2_ERET]
_pop_irq_stack_frame
mov_s sp, ilink
rtie
/* separated entry for trap which may be used by irq_offload, USERPSACE */
SECTION_SUBSEC_FUNC(TEXT,__fault,__ev_trap)
/* get the id of trap_s */
lr ilink, [_ARC_V2_ECR]
and ilink, ilink, 0x3f
#ifdef CONFIG_USERSPACE
cmp ilink, _TRAP_S_CALL_SYSTEM_CALL
bne _do_non_syscall_trap
/* do sys_call */
mov_s ilink, K_SYSCALL_LIMIT
cmp r6, ilink
blo valid_syscall_id
mov_s r0, r6
mov_s r6, K_SYSCALL_BAD
valid_syscall_id:
/* create a sys call frame
* caller regs (r0 - 12) are saved in _create_irq_stack_frame
* ok to use them later
*/
_create_irq_stack_frame
_save_exc_regs_into_stack
/* exc return and do sys call in kernel mode,
* so need to clear U bit, r0 is already loaded
* with ERSTATUS in _save_exc_regs_into_stack
*/
bclr r0, r0, _ARC_V2_STATUS32_U_BIT
sr r0, [_ARC_V2_ERSTATUS]
mov_s r0, _arc_do_syscall
sr r0, [_ARC_V2_ERET]
rtie
_do_non_syscall_trap:
#endif /* CONFIG_USERSPACE */
#ifdef CONFIG_IRQ_OFFLOAD
/*
* IRQ_OFFLOAD is to simulate interrupt handling through exception,
* so its entry is different with normal exception handling, it is
* handled in isr stack
*/
cmp ilink, _TRAP_S_SCALL_IRQ_OFFLOAD
bne _exc_entry
/* save caller saved registers */
_create_irq_stack_frame
_save_exc_regs_into_stack
lr r0,[_ARC_V2_ERSTATUS]
st_s r0, [sp, ___isf_t_status32_OFFSET]
lr r0,[_ARC_V2_ERET]
st_s r0, [sp, ___isf_t_pc_OFFSET] /* eret into pc */
/* check whether irq stack is used */
_check_and_inc_int_nest_counter r0, r1
bne.d exc_nest_handle
mov_s r0, sp
_get_curr_cpu_irq_stack sp
exc_nest_handle:
push_s r0
jl z_irq_do_offload
pop sp
_dec_int_nest_counter r0, r1
jl _Fault
/* if _Fault returns, restore the registers */
_pop_irq_stack_frame
/* ERSTATUS, ERET are not changed, so ok to rtie */
/* now restore the stack */
ld sp,[saved_stack_pointer]
rtie
#ifdef CONFIG_IRQ_OFFLOAD
GTEXT(_irq_do_offload);
SECTION_SUBSEC_FUNC(TEXT,__fault,__ev_trap)
/*
* Before invoking exception handler, the kernel switches to an exception
* stack to save the faulting thread's registers.
* The exception is fatal and all the kernel can do is just print
* a diagnostic message and halt.
*/
#ifdef CONFIG_ARC_STACK_CHECKING
push_s r2
/* disable stack checking */
lr r2, [_ARC_V2_STATUS32]
bclr r2, r2, _ARC_V2_STATUS32_SC_BIT
kflag r2
pop_s r2
#endif
/* save caller saved registers */
_create_irq_stack_frame
lr r0,[_ARC_V2_ERSTATUS]
st_s r0, [sp, ___isf_t_status32_OFFSET]
lr r0,[_ARC_V2_ERET]
st_s r0, [sp, ___isf_t_pc_OFFSET] /* eret into pc */
jl _irq_do_offload
mov_s r1, _kernel
ld_s r2, [r1, _kernel_offset_to_current]
/* check if we're a nested interrupt: if so, let the
* interrupted interrupt handle the reschedule
*/
lr r3, [_ARC_V2_AUX_IRQ_ACT]
/* the OS on ARCv2 always runs in kernel mode, so assume bit31 [U] in
* AUX_IRQ_ACT is always 0: if the contents of AUX_IRQ_ACT is 0, it
* means trap was taken from outside an interrupt handler.
* But if it was inside, let that handler do the swap.
*/
breq r3, 0, _trap_check_for_swap
_trap_return:
_pop_irq_stack_frame
rtie
.balign 4
_trap_check_for_swap:
/* coop thread ? do not schedule */
ld_s r0, [r2, _thread_offset_to_prio]
brlt r0, 0, _trap_return
/* scheduler locked ? do not schedule */
ld_s r0, [r2, _thread_offset_to_sched_locked]
brgt r0, 0, _trap_return
/* check if the current thread needs to be rescheduled */
push_s r2
push_s r1
push_s blink
jl _is_next_thread_current
pop_s blink
pop_s r1
pop_s r2
brne r0, 0, _trap_return
_save_callee_saved_regs
st _CAUSE_RIRQ, [r2, _thread_offset_to_relinquish_cause]
/* note: Ok to use _CAUSE_RIRQ since everything is saved */
/*
* Save needed registers to callee saved ones. It is faster than
* pushing registers to stack. It is possible to do since program has
* just saved them and the calling routine will save them in turn
* if it uses them.
*/
mov_s r13, blink
mov_s r14, r0
mov_s r15, r1
jl _get_next_ready_thread
mov_s r2, r0
mov_s r1, r15
mov_s r0, r14
mov_s blink, r13
st_s r2, [r1, _kernel_offset_to_current]
/* clear AE bit to forget this was an exception */
lr r3, [_ARC_V2_STATUS32]
and r3,r3,(~_ARC_V2_STATUS32_AE)
kflag r3
/* pretend lowest priority interrupt happened to use common handler */
lr r3, [_ARC_V2_AUX_IRQ_ACT]
or r3,r3,(1<<(CONFIG_NUM_IRQ_PRIO_LEVELS-1)) /* use lowest */
sr r3, [_ARC_V2_AUX_IRQ_ACT]
/* Assumption: r2 has current thread */
b _rirq_common_interrupt_swap
#endif /* CONFIG_IRQ_OFFLOAD */
b _exc_entry

View File

@@ -1,7 +1,17 @@
/*
* Copyright (c) 2014 Wind River Systems, Inc.
*
* SPDX-License-Identifier: Apache-2.0
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
@@ -17,71 +27,13 @@
* number from 16 to last IRQ number on the platform.
*/
#include <kernel.h>
#include <nanokernel.h>
#include <arch/cpu.h>
#include <sys/__assert.h>
#include <misc/__assert.h>
#include <toolchain.h>
#include <linker/sections.h>
#include <sections.h>
#include <sw_isr_table.h>
#include <irq.h>
#include <sys/printk.h>
/*
* storage space for the interrupt stack of fast_irq
*/
#if defined(CONFIG_ARC_FIRQ_STACK)
#if defined(CONFIG_SMP)
K_THREAD_STACK_ARRAY_DEFINE(_firq_interrupt_stack, CONFIG_MP_NUM_CPUS,
CONFIG_ARC_FIRQ_STACK_SIZE);
#else
K_THREAD_STACK_DEFINE(_firq_interrupt_stack, CONFIG_ARC_FIRQ_STACK_SIZE);
#endif
/*
* @brief Set the stack pointer for firq handling
*
* @return N/A
*/
void z_arc_firq_stack_set(void)
{
#ifdef CONFIG_SMP
char *firq_sp = Z_THREAD_STACK_BUFFER(
_firq_interrupt_stack[z_arc_v2_core_id()]) +
CONFIG_ARC_FIRQ_STACK_SIZE;
#else
char *firq_sp = Z_THREAD_STACK_BUFFER(_firq_interrupt_stack) +
CONFIG_ARC_FIRQ_STACK_SIZE;
#endif
/* the z_arc_firq_stack_set must be called when irq diasbled, as
* it can be called not only in the init phase but also other places
*/
unsigned int key = arch_irq_lock();
__asm__ volatile (
/* only ilink will not be banked, so use ilink as channel
* between 2 banks
*/
"mov ilink, %0 \n\t"
"lr %0, [%1] \n\t"
"or %0, %0, %2 \n\t"
"kflag %0 \n\t"
"mov sp, ilink \n\t"
/* switch back to bank0, use ilink to avoid the pollution of
* bank1's gp regs.
*/
"lr ilink, [%1] \n\t"
"and ilink, ilink, %3 \n\t"
"kflag ilink \n\t"
:
: "r"(firq_sp), "i"(_ARC_V2_STATUS32),
"i"(_ARC_V2_STATUS32_RB(1)),
"i"(~_ARC_V2_STATUS32_RB(7))
);
arch_irq_unlock(key);
}
#endif
/*
* @brief Enable an interrupt line
@@ -93,9 +45,12 @@ void z_arc_firq_stack_set(void)
* @return N/A
*/
void arch_irq_enable(unsigned int irq)
void _arch_irq_enable(unsigned int irq)
{
z_arc_v2_irq_unit_int_enable(irq);
int key = irq_lock();
_arc_v2_irq_unit_int_enable(irq);
irq_unlock(key);
}
/*
@@ -107,20 +62,12 @@ void arch_irq_enable(unsigned int irq)
* @return N/A
*/
void arch_irq_disable(unsigned int irq)
void _arch_irq_disable(unsigned int irq)
{
z_arc_v2_irq_unit_int_disable(irq);
}
int key = irq_lock();
/**
* @brief Return IRQ enable state
*
* @param irq IRQ line
* @return interrupt enable state, true or false
*/
int arch_irq_is_enabled(unsigned int irq)
{
return z_arc_v2_irq_unit_int_enabled(irq);
_arc_v2_irq_unit_int_disable(irq);
irq_unlock(key);
}
/*
@@ -137,23 +84,16 @@ int arch_irq_is_enabled(unsigned int irq)
* @return N/A
*/
void z_irq_priority_set(unsigned int irq, unsigned int prio, u32_t flags)
void _irq_priority_set(unsigned int irq, unsigned int prio, uint32_t flags)
{
ARG_UNUSED(flags);
int key = irq_lock();
__ASSERT(prio < CONFIG_NUM_IRQ_PRIO_LEVELS,
"invalid priority %d for irq %d", prio, irq);
/* 0 -> CONFIG_NUM_IRQ_PRIO_LEVELS allocted to secure world
* left prio levels allocated to normal world
*/
#if defined(CONFIG_ARC_SECURE_FIRMWARE)
prio = prio < ARC_N_IRQ_START_LEVEL ?
prio : (ARC_N_IRQ_START_LEVEL - 1);
#elif defined(CONFIG_ARC_NORMAL_FIRMWARE)
prio = prio < ARC_N_IRQ_START_LEVEL ?
ARC_N_IRQ_START_LEVEL : prio;
#endif
z_arc_v2_irq_unit_prio_set(irq, prio);
_arc_v2_irq_unit_prio_set(irq, prio);
irq_unlock(key);
}
/*
@@ -165,19 +105,12 @@ void z_irq_priority_set(unsigned int irq, unsigned int prio, u32_t flags)
* @return N/A
*/
void z_irq_spurious(void *unused)
#include <misc/printk.h>
void _irq_spurious(void *unused)
{
ARG_UNUSED(unused);
z_fatal_error(K_ERR_SPURIOUS_IRQ, NULL);
printk("_irq_spurious(). Spinning...\n");
for (;;)
;
}
#ifdef CONFIG_DYNAMIC_INTERRUPTS
int arch_irq_connect_dynamic(unsigned int irq, unsigned int priority,
void (*routine)(void *parameter), void *parameter,
u32_t flags)
{
z_isr_install(irq, routine, parameter);
z_irq_priority_set(irq, priority, flags);
return irq;
}
#endif /* CONFIG_DYNAMIC_INTERRUPTS */

View File

@@ -1,33 +1,45 @@
/*
* Copyright (c) 2015 Intel corporation
*
* SPDX-License-Identifier: Apache-2.0
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
* @file Software interrupts utility code - ARC implementation
*/
#include <kernel.h>
#include <nanokernel.h>
#include <irq_offload.h>
static irq_offload_routine_t offload_routine;
static void *offload_param;
/* Called by trap_s exception handler */
void z_irq_do_offload(void)
void _irq_do_offload(void)
{
offload_routine(offload_param);
}
void arch_irq_offload(irq_offload_routine_t routine, void *parameter)
void irq_offload(irq_offload_routine_t routine, void *parameter)
{
int key;
key = irq_lock();
offload_routine = routine;
offload_param = parameter;
__asm__ volatile ("trap_s %[id]"
:
: [id] "i"(_TRAP_S_SCALL_IRQ_OFFLOAD) : );
__asm__ volatile ("trap_s 0");
irq_unlock(key);
}

View File

@@ -0,0 +1,53 @@
/*
* Copyright (c) 2014 Wind River Systems, Inc.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
* @brief IRQ part of vector table for Quark SE Sensor Subsystem
*
* This file contains the IRQ part of the vector table. It is meant to be used
* for one of two cases:
*
* a) When software-managed ISRs (SW_ISR_TABLE) is enabled, and in that case it
* binds _IsrWrapper() to all the IRQ entries in the vector table.
*
* b) When the BSP is written so that device ISRs are installed directly in the
* vector table, they are enumerated here.
*
*/
#include <toolchain.h>
#include <sections.h>
extern void _isr_enter(void);
typedef void (*vth)(void); /* Vector Table Handler */
#if defined(CONFIG_SW_ISR_TABLE)
vth __irq_vector_table _irq_vector_table[CONFIG_NUM_IRQS - 16] = {
[0 ...(CONFIG_NUM_IRQS - 17)] = _isr_enter
};
#elif !defined(CONFIG_IRQ_VECTOR_TABLE_CUSTOM)
extern void _SpuriousIRQ(void);
/* placeholders: fill with real ISRs */
vth __irq_vector_table _irq_vector_table[CONFIG_NUM_IRQS - 16] = {
[0 ...(CONFIG_NUM_IRQS - 17)] = _SpuriousIRQ
};
#endif /* CONFIG_SW_ISR_TABLE */

View File

@@ -1,7 +1,17 @@
/*
* Copyright (c) 2014-2015 Wind River Systems, Inc.
*
* SPDX-License-Identifier: Apache-2.0
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
@@ -13,19 +23,27 @@
* a parameter.
*/
#define _ASMLANGUAGE
#include <offsets_short.h>
#include <toolchain.h>
#include <linker/sections.h>
#include <sections.h>
#include <sw_isr_table.h>
#include <kernel_structs.h>
#include <arch/cpu.h>
#include <swap_macros.h>
GTEXT(_isr_wrapper)
GTEXT(_isr_enter)
GTEXT(_isr_demux)
#if CONFIG_RGF_NUM_BANKS == 1
GDATA(saved_r0)
SECTION_VAR(BSS, saved_r0)
.word 0
#endif
#if defined(CONFIG_SYS_POWER_MANAGEMENT)
GTEXT(z_sys_power_save_idle_exit)
GTEXT(_sys_power_save_idle_exit)
#endif
/*
@@ -34,39 +52,48 @@ _rirq_enter/_firq_enter: they are jump points.
The flow is the following:
ISR -> _isr_wrapper -- + -> _rirq_enter -> _isr_demux -> ISR -> _rirq_exit
|
+ -> _firq_enter -> _isr_demux -> ISR -> _firq_exit
ISR -> _isr_enter -- + -> _rirq_enter -> _isr_demux -> ISR -> _rirq_exit
|
+ -> _firq_enter -> _isr_demux -> ISR -> _firq_exit
Context switch explanation:
The context switch code is spread in these files:
isr_wrapper.s, switch.s, swap_macros.h, fast_irq.s, regular_irq.s
isr_wrapper.s, swap.s, swap_macros.s, fast_irq.s, regular_irq.s
IRQ stack frame layout:
high address
high address
status32
pc
lp_count
lp_start
lp_end
blink
r13
...
sp -> r0
status32
pc
lp_count
lp_start
lp_end
blink
r13
...
sp -> r0
low address
low address
Registers not taken into account in the current implementation.
jli_base
ldi_base
ei_base
accl
acch
The context switch code adopts this standard so that it is easier to follow:
- r2 contains _kernel.current ASAP, and the incoming thread when we
transition from outgoing thread to incoming thread
- r1 contains _kernel ASAP and is not overwritten over the lifespan of
the functions.
- r2 contains _kernel.current ASAP, and the incoming thread when we
transition from outgoing thread to incoming thread
Not loading _kernel into r0 allows loading _kernel without stomping on
the parameter in r0 in arch_switch().
the parameter in r0 in _Swap().
ARCv2 processors have two kinds of interrupts: fast (FIRQ) and regular. The
@@ -92,55 +119,53 @@ registers (to avoid stack accesses). It is possible to register a FIRQ
handler that operates outside of the kernel, but care must be taken to only
use instructions that only use the banked registers.
The kernel is able to handle transitions to and from FIRQ, RIRQ and threads.
The contexts are saved 'lazily': the minimum amount of work is
The kernel is able to handle transitions to and from FIRQ, RIRQ and threads
(fibers/task). The contexts are saved 'lazily': the minimum amount of work is
done upfront, and the rest is done when needed:
o RIRQ
All needed registers to run C code in the ISR are saved automatically
on the outgoing thread's stack: loop, status32, pc, and the caller-
saved GPRs. That stack frame layout is pre-determined. If returning
to a thread, the stack is popped and no registers have to be saved by
the kernel. If a context switch is required, the callee-saved GPRs
are then saved in the thread's stack.
All needed regisers to run C code in the ISR are saved automatically
on the outgoing thread's stack: loop, status32, pc, and the caller-
saved GPRs. That stack frame layout is pre-determined. If returning
to a fiber, the stack is popped and no registers have to be saved by
the kernel. If a context switch is required, the callee-saved GPRs
are then saved in the thread control structure (TCS).
o FIRQ
First, a FIRQ can be interrupting a lower-priority RIRQ: if this is
the case, the FIRQ does not take a scheduling decision and leaves it
the RIRQ to handle. This limits the amount of code that has to run at
interrupt-level.
First, a FIRQ can be interrupting a lower-priority RIRQ: if this is the case,
the FIRQ does not take a scheduling decision and leaves it the RIRQ to
handle. This limits the amount of code that has to run at interrupt-level.
CONFIG_RGF_NUM_BANKS==1 case:
Registers are saved on the stack frame just as they are for RIRQ.
Context switch can happen just as it does in the RIRQ case, however,
if the FIRQ interrupted a RIRQ, the FIRQ will return from interrupt
and let the RIRQ do the context switch. At entry, one register is
needed in order to have code to save other registers. r0 is saved
first in the stack and restored later
CONFIG_RGF_NUM_BANKS==1 case:
Registers are saved on the stack frame just as they are for RIRQ.
Context switch can happen just as it does in the RIRQ case, however,
if the FIRQ interrupted a RIRQ, the FIRQ will return from interrupt and
let the RIRQ do the context switch. At entry, one register is needed in order
to have code to save other registers. r0 is saved first in a global called
saved_r0.
CONFIG_RGF_NUM_BANKS!=1 case:
During early initialization, the sp in the 2nd register bank is made to
refer to _firq_stack. This allows for the FIRQ handler to use its own
stack. GPRs are banked, loop registers are saved in unused callee saved
regs upon interrupt entry. If returning to a thread, loop registers are
restored and the CPU switches back to bank 0 for the GPRs. If a context
switch is needed, at this point only are all the registers saved.
First, a stack frame with the same layout as the automatic RIRQ one is
created and then the callee-saved GPRs are saved in the stack.
status32_p0 and ilink are saved in this case, not status32 and pc.
To create the stack frame, the FIRQ handling code must first go back to
using bank0 of registers, since that is where the registers containing
the exiting thread are saved. Care must be taken not to touch any
register before saving them: the only one usable at that point is the
stack pointer.
CONFIG_RGF_NUM_BANKS!=1 case:
During early initialization, the sp in the 2nd register bank is made to
refer to _firq_stack. This allows for the FIRQ handler to use its own stack.
GPRs are banked, loop registers are saved in unused callee saved regs upon
interrupt entry. If returning to a fiber, loop registers are restored and the
CPU switches back to bank 0 for the GPRs. If a context switch is
needed, at this point only are all the registers saved. First, a
stack frame with the same layout as the automatic RIRQ one is created
and then the callee-saved GPRs are saved in the TCS. status32_p0 and
ilink are saved in this case, not status32 and pc.
To create the stack frame, the FIRQ handling code must first go back to using
bank0 of registers, since that is where the registers containing the exiting
thread are saved. Care must be taken not to touch any register before saving
them: the only one usable at that point is the stack pointer.
o coop
When a coop context switch is done, the callee-saved registers are
saved in the stack. The other GPRs do not need to be saved, since the
compiler has already placed them on the stack.
When a coop context switch is done, the callee-saved registers are
saved in the TCS. The other GPRs do not need to be saved, since the
compiler has already placed them on the stack.
For restoring the contexts, there are six cases. In all cases, the
callee-saved registers of the incoming thread have to be restored. Then, there
@@ -148,87 +173,102 @@ are specifics for each case:
From coop:
o to coop
o to coop
Do a normal function call return.
Restore interrupt lock level and do a normal function call return.
o to any irq
o to any irq
The incoming interrupted thread has an IRQ stack frame containing the
caller-saved registers that has to be popped. status32 has to be
restored, then we jump to the interrupted instruction.
The incoming interrupted thread has an IRQ stack frame containing the
caller-saved registers that has to be popped. status32 has to be restored,
then we jump to the interrupted instruction.
From FIRQ:
When CONFIG_RGF_NUM_BANKS==1, context switch is done as it is for RIRQ.
When CONFIG_RGF_NUM_BANKS!=1, the processor is put back to using bank0,
not bank1 anymore, because it had to save the outgoing context from
bank0, and now has to load the incoming one into bank0.
When CONFIG_RGF_NUM_BANKS==1, context switch is done as it is for RIRQ.
When CONFIG_RGF_NUM_BANKS!=1, the processor is put back to using bank0,
not bank1 anymore, because it had to save the outgoing context from bank0,
and now has to load the incoming one
into bank0.
o to coop
o to coop
The address of the returning instruction from arch_switch() is loaded
in ilink and the saved status32 in status32_p0.
The address of the returning instruction from _Swap() is loaded in ilink and
the saved status32 in status32_p0, taking care to adjust the interrupt lock
state desired in status32_p0. The return value is put in r0.
o to any irq
o to any irq
The IRQ has saved the caller-saved registers in a stack frame, which
must be popped, and status32 and pc loaded in status32_p0 and ilink.
The IRQ has saved the caller-saved registers in a stack frame, which must be
popped, and statu32 and pc loaded in status32_p0 and ilink.
From RIRQ:
o to coop
o to coop
The interrupt return mechanism in the processor expects a stack frame,
but the outgoing context did not create one. A fake one is created
here, with only the relevant values filled in: pc, status32.
The interrupt return mechanism in the processor expects a stack frame, but
the outgoing context did not create one. A fake one is created here, with
only the relevant values filled in: pc, status32 and the return value in r0.
There is a discrepancy between the ABI from the ARCv2 docs,
including the way the processor pushes GPRs in pairs in the IRQ stack
frame, and the ABI GCC uses. r13 should be a callee-saved register,
but GCC treats it as caller-saved. This means that the processor pushes
it in the stack frame along with r12, but the compiler does not save it
before entering a function. So, it is saved as part of the callee-saved
registers, and restored there, but the processor restores it _a second
time_ when popping the IRQ stack frame. Thus, the correct value must
also be put in the fake stack frame when returning to a thread that
context switched out cooperatively.
There is a discrepancy between the ABI from the ARCv2 docs, including the
way the processor pushes GPRs in pairs in the IRQ stack frame, and the ABI
GCC uses. r13 should be a callee-saved register, but GCC treats it as
caller-saved. This means that the processor pushes it in the stack frame
along with r12, but the compiler does not save it before entering a
function. So, it is saved as part of the callee-saved registers, and
restored there, but the processor restores it _a second time_ when popping
the IRQ stack frame. Thus, the correct value must also be put in the fake
stack frame when returning to a thread that context switched out
cooperatively.
o to any irq
o to any irq
Both types of IRQs already have an IRQ stack frame: simply return from
interrupt.
Both types of IRQs already have an IRQ stack frame: simply return from
interrupt.
*/
SECTION_FUNC(TEXT, _isr_wrapper)
#if defined(CONFIG_ARC_FIRQ)
SECTION_FUNC(TEXT, _isr_enter)
#if CONFIG_RGF_NUM_BANKS == 1
/* free r0 here, use r0 to check whether irq is firq.
* for rirq, as sp will not change and r0 already saved, this action
* in fact is an action like nop.
* for firq, r0 will be restored later
*/
st_s r0, [sp]
st r0,[saved_r0]
#endif
lr r0, [_ARC_V2_AUX_IRQ_ACT]
ffs r0, r0
cmp r0, 0
#if CONFIG_RGF_NUM_BANKS == 1
bnz rirq_path
ld_s r0, [sp]
/* 1-register bank FIRQ handling must save registers on stack */
_create_irq_stack_frame
lr r0, [_ARC_V2_STATUS32_P0]
st_s r0, [sp, ___isf_t_status32_OFFSET]
lr r0, [_ARC_V2_ERET]
st_s r0, [sp, ___isf_t_pc_OFFSET]
mov_s r3, _firq_exit
mov_s r2, _firq_enter
lr r0,[_ARC_V2_STATUS32_P0]
push_s r0
mov r0,ilink
push_s r0
mov r0,lp_count
push_s r0
lr r0, [_ARC_V2_LP_START]
push_s r0
lr r0, [_ARC_V2_LP_END]
push_s r0
push_s blink
push_s r13
push_s r12
push r11
push r10
push r9
push r8
push r7
push r6
push r5
push r4
push_s r3
push_s r2
push_s r1
ld r0,[saved_r0]
push_s r0
mov r3, _firq_exit
mov r2, _firq_enter
j_s [r2]
rirq_path:
mov_s r3, _rirq_exit
mov_s r2, _rirq_enter
mov r3, _rirq_exit
mov r2, _rirq_enter
j_s [r2]
#else
mov.z r3, _firq_exit
@@ -237,25 +277,55 @@ rirq_path:
mov.nz r2, _rirq_enter
j_s [r2]
#endif
#ifdef CONFIG_KERNEL_EVENT_LOGGER_SLEEP
GTEXT(_sys_k_event_logger_exit_sleep)
.macro log_sleep_k_event
clri r0 /* do not interrupt event logger operations */
push_s r0
push_s blink
jl _sys_k_event_logger_exit_sleep
pop_s blink
pop_s r0
seti r0
.endm
#else
mov_s r3, _rirq_exit
mov_s r2, _rirq_enter
j_s [r2]
#define log_sleep_k_event
#endif
#if defined(CONFIG_KERNEL_EVENT_LOGGER_INTERRUPT)
GTEXT(_sys_k_event_logger_interrupt)
.macro log_interrupt_k_event
clri r0 /* do not interrupt event logger operations */
push_s r0
push_s blink
jl _sys_k_event_logger_interrupt
pop_s blink
pop_s r0
seti r0
.endm
#else
#define log_interrupt_k_event
#endif
#if defined(CONFIG_SYS_POWER_MANAGEMENT)
.macro exit_tickless_idle
clri r0 /* do not interrupt exiting tickless idle operations */
push_s r1
push_s r0
mov_s r1, _kernel
ld_s r0, [r1, _kernel_offset_to_idle] /* requested idle duration */
breq r0, 0, _skip_sys_power_save_idle_exit
st 0, [r1, _kernel_offset_to_idle] /* zero idle duration */
push_s blink
jl z_sys_power_save_idle_exit
jl _sys_power_save_idle_exit
pop_s blink
_skip_sys_power_save_idle_exit:
pop_s r0
pop_s r1
seti r0
.endm
#else
@@ -266,63 +336,23 @@ _skip_sys_power_save_idle_exit:
SECTION_FUNC(TEXT, _isr_demux)
push_s r3
/* according to ARCv2 ISA, r25, r30, r58, r59 are caller-saved
* scratch registers, possibly used by interrupt handlers
*/
push r25
push r30
#ifdef CONFIG_ARC_HAS_ACCL_REGS
push r58
push r59
#endif
#ifdef CONFIG_EXECUTION_BENCHMARKING
bl read_timer_start_of_isr
#endif
#if defined(CONFIG_TRACING)
bl sys_trace_isr_enter
#endif
/* cannot be done before this point because we must be able to run C */
/* r0 is available to be stomped here, and exit_tickless_idle uses it */
exit_tickless_idle
log_interrupt_k_event
log_sleep_k_event
lr r0, [_ARC_V2_ICAUSE]
/* handle software triggered interrupt */
lr r3, [_ARC_V2_AUX_IRQ_HINT]
brne r3, r0, irq_hint_handled
sr 0, [_ARC_V2_AUX_IRQ_HINT]
irq_hint_handled:
sub r0, r0, 16
mov_s r1, _sw_isr_table
mov r1, _sw_isr_table
add3 r0, r1, r0 /* table entries are 8-bytes wide */
ld_s r1, [r0, 4] /* ISR into r1 */
#ifdef CONFIG_EXECUTION_BENCHMARKING
push_s r0
push_s r1
bl read_timer_end_of_isr
pop_s r1
pop_s r0
#endif
jl_s.d [r1]
ld_s r0, [r0] /* delay slot: ISR parameter into r0 */
#ifdef CONFIG_TRACING_ISR
bl sys_trace_isr_exit
#endif
#ifdef CONFIG_ARC_HAS_ACCL_REGS
pop r59
pop r58
#endif
pop r30
pop r25
/* back from ISR, jump to exit stub */
pop_s r3
j_s [r3]
nop_s
nop

View File

@@ -1,6 +0,0 @@
# SPDX-License-Identifier: Apache-2.0
zephyr_library()
zephyr_library_sources_if_kconfig(arc_core_mpu.c)
zephyr_library_sources_if_kconfig(arc_mpu.c)

View File

@@ -1,37 +0,0 @@
# Memory Protection Unit (MPU) configuration options
# Copyright (c) 2017 Synopsys
# SPDX-License-Identifier: Apache-2.0
config ARC_MPU_VER
int "ARC MPU version"
range 2 4
default 2
help
ARC MPU has several versions. For MPU v2, the minimum region is 2048 bytes;
For MPU v3, the minimum region is 32 bytes
config ARC_CORE_MPU
bool "ARC Core MPU functionalities"
help
ARC core MPU functionalities
config MPU_STACK_GUARD
bool "Thread Stack Guards"
depends on ARC_CORE_MPU && ARC_MPU_VER !=2
help
Enable thread stack guards via MPU. ARC supports built-in stack protection.
If your core supports that, it is preferred over MPU stack guard.
For ARC_MPU_VER == 2, it requires 2048 extra bytes and a strong start address
alignment, this will bring big waste of memory, so no support for it.
config ARC_MPU
bool "ARC MPU Support"
select ARC_CORE_MPU
select THREAD_STACK_INFO
select MEMORY_PROTECTION
select GEN_PRIV_STACKS if ARC_MPU_VER = 2
select MPU_REQUIRES_POWER_OF_TWO_ALIGNMENT if ARC_MPU_VER = 2
select MPU_REQUIRES_NON_OVERLAPPING_REGIONS if ARC_MPU_VER = 3
help
Target has ARC MPU (currently only works for EMSK 2.2/2.3 ARCEM7D)

View File

@@ -1,101 +0,0 @@
/*
* Copyright (c) 2017 Synopsys.
*
* SPDX-License-Identifier: Apache-2.0
*/
#include <device.h>
#include <init.h>
#include <kernel.h>
#include <soc.h>
#include <arch/arc/v2/mpu/arc_core_mpu.h>
#include <kernel_structs.h>
/*
* @brief Configure MPU for the thread
*
* This function configures per thread memory map reprogramming the MPU.
*
* @param thread thread info data structure.
*/
void configure_mpu_thread(struct k_thread *thread)
{
arc_core_mpu_disable();
arc_core_mpu_configure_thread(thread);
arc_core_mpu_enable();
}
#if defined(CONFIG_USERSPACE)
int arch_mem_domain_max_partitions_get(void)
{
return arc_core_mpu_get_max_domain_partition_regions();
}
/*
* Reset MPU region for a single memory partition
*/
void arch_mem_domain_partition_remove(struct k_mem_domain *domain,
u32_t partition_id)
{
if (_current->mem_domain_info.mem_domain != domain) {
return;
}
arc_core_mpu_disable();
arc_core_mpu_remove_mem_partition(domain, partition_id);
arc_core_mpu_enable();
}
/*
* Configure MPU memory domain
*/
void arch_mem_domain_thread_add(struct k_thread *thread)
{
if (_current != thread) {
return;
}
arc_core_mpu_disable();
arc_core_mpu_configure_mem_domain(thread);
arc_core_mpu_enable();
}
/*
* Destroy MPU regions for the mem domain
*/
void arch_mem_domain_destroy(struct k_mem_domain *domain)
{
if (_current->mem_domain_info.mem_domain != domain) {
return;
}
arc_core_mpu_disable();
arc_core_mpu_remove_mem_domain(domain);
arc_core_mpu_enable();
}
void arch_mem_domain_partition_add(struct k_mem_domain *domain,
u32_t partition_id)
{
/* No-op on this architecture */
}
void arch_mem_domain_thread_remove(struct k_thread *thread)
{
if (_current != thread) {
return;
}
arch_mem_domain_destroy(thread->mem_domain_info.mem_domain);
}
/*
* Validate the given buffer is user accessible or not
*/
int arch_buffer_validate(void *addr, size_t size, int write)
{
return arc_core_mpu_buffer_validate(addr, size, write);
}
#endif

View File

@@ -1,59 +0,0 @@
/*
* Copyright (c) 2019 Synopsys.
*
* SPDX-License-Identifier: Apache-2.0
*/
#include <device.h>
#include <init.h>
#include <kernel.h>
#include <soc.h>
#include <arch/arc/v2/aux_regs.h>
#include <arch/arc/v2/mpu/arc_mpu.h>
#include <arch/arc/v2/mpu/arc_core_mpu.h>
#include <linker/linker-defs.h>
#define LOG_LEVEL CONFIG_MPU_LOG_LEVEL
#include <logging/log.h>
LOG_MODULE_REGISTER(mpu);
/**
* @brief Get the number of supported MPU regions
*
*/
static inline u8_t get_num_regions(void)
{
u32_t num = z_arc_v2_aux_reg_read(_ARC_V2_MPU_BUILD);
num = (num & 0xFF00U) >> 8U;
return (u8_t)num;
}
/**
* This internal function is utilized by the MPU driver to parse the intent
* type (i.e. THREAD_STACK_REGION) and return the correct parameter set.
*/
static inline u32_t get_region_attr_by_type(u32_t type)
{
switch (type) {
case THREAD_STACK_USER_REGION:
return REGION_RAM_ATTR;
case THREAD_STACK_REGION:
return AUX_MPU_ATTR_KW | AUX_MPU_ATTR_KR;
case THREAD_APP_DATA_REGION:
return REGION_RAM_ATTR;
case THREAD_STACK_GUARD_REGION:
/* no Write and Execute to guard region */
return AUX_MPU_ATTR_UR | AUX_MPU_ATTR_KR;
default:
/* unknown type */
return 0;
}
}
#if CONFIG_ARC_MPU_VER == 2
#include "arc_mpu_v2_internal.h"
#elif CONFIG_ARC_MPU_VER == 3
#include "arc_mpu_v3_internal.h"
#endif

View File

@@ -1,433 +0,0 @@
/*
* Copyright (c) 2019 Synopsys.
*
* SPDX-License-Identifier: Apache-2.0
*/
#ifndef ZEPHYR_ARCH_ARC_CORE_MPU_ARC_MPU_V2_INTERNAL_H_
#define ZEPHYR_ARCH_ARC_CORE_MPU_ARC_MPU_V2_INTERNAL_H_
#define AUX_MPU_RDB_VALID_MASK (0x1)
#define AUX_MPU_EN_ENABLE (0x40000000)
#define AUX_MPU_EN_DISABLE (0xBFFFFFFF)
#define AUX_MPU_RDP_REGION_SIZE(bits) \
(((bits - 1) & 0x3) | (((bits - 1) & 0x1C) << 7))
#define AUX_MPU_RDP_ATTR_MASK (0x1FC)
#define AUX_MPU_RDP_SIZE_MASK (0xE03)
#define _ARC_V2_MPU_EN (0x409)
#define _ARC_V2_MPU_RDB0 (0x422)
#define _ARC_V2_MPU_RDP0 (0x423)
/* For MPU version 2, the minimum protection region size is 2048 bytes */
#define ARC_FEATURE_MPU_ALIGNMENT_BITS 11
/**
* This internal function initializes a MPU region
*/
static inline void _region_init(u32_t index, u32_t region_addr, u32_t size,
u32_t region_attr)
{
u8_t bits = find_msb_set(size) - 1;
index = index * 2U;
if (bits < ARC_FEATURE_MPU_ALIGNMENT_BITS) {
bits = ARC_FEATURE_MPU_ALIGNMENT_BITS;
}
if ((1 << bits) < size) {
bits++;
}
if (size > 0) {
region_attr &= ~(AUX_MPU_RDP_SIZE_MASK);
region_attr |= AUX_MPU_RDP_REGION_SIZE(bits);
region_addr |= AUX_MPU_RDB_VALID_MASK;
} else {
region_addr = 0U;
}
z_arc_v2_aux_reg_write(_ARC_V2_MPU_RDP0 + index, region_attr);
z_arc_v2_aux_reg_write(_ARC_V2_MPU_RDB0 + index, region_addr);
}
/**
* This internal function is utilized by the MPU driver to parse the intent
* type (i.e. THREAD_STACK_REGION) and return the correct region index.
*/
static inline int get_region_index_by_type(u32_t type)
{
/*
* The new MPU regions are allocated per type after the statically
* configured regions. The type is one-indexed rather than
* zero-indexed.
*
* For ARC MPU v2, the smaller index has higher priority, so the
* index is allocated in reverse order. Static regions start from
* the biggest index, then thread related regions.
*
*/
switch (type) {
case THREAD_STACK_USER_REGION:
return get_num_regions() - mpu_config.num_regions
- THREAD_STACK_REGION;
case THREAD_STACK_REGION:
case THREAD_APP_DATA_REGION:
case THREAD_DOMAIN_PARTITION_REGION:
/*
* Start domain partition region from stack guard region
* since stack guard is not supported.
*/
return get_num_regions() - mpu_config.num_regions - type + 1;
default:
__ASSERT(0, "Unsupported type");
return -EINVAL;
}
}
/**
* This internal function checks if region is enabled or not
*/
static inline bool _is_enabled_region(u32_t r_index)
{
return ((z_arc_v2_aux_reg_read(_ARC_V2_MPU_RDB0 + r_index * 2U)
& AUX_MPU_RDB_VALID_MASK) == AUX_MPU_RDB_VALID_MASK);
}
/**
* This internal function check if the given buffer in in the region
*/
static inline bool _is_in_region(u32_t r_index, u32_t start, u32_t size)
{
u32_t r_addr_start;
u32_t r_addr_end;
u32_t r_size_lshift;
r_addr_start = z_arc_v2_aux_reg_read(_ARC_V2_MPU_RDB0 + r_index * 2U)
& (~AUX_MPU_RDB_VALID_MASK);
r_size_lshift = z_arc_v2_aux_reg_read(_ARC_V2_MPU_RDP0 + r_index * 2U)
& AUX_MPU_RDP_SIZE_MASK;
r_size_lshift = (r_size_lshift & 0x3) | ((r_size_lshift >> 7) & 0x1C);
r_addr_end = r_addr_start + (1 << (r_size_lshift + 1));
if (start >= r_addr_start && (start + size) <= r_addr_end) {
return 1;
}
return 0;
}
/**
* This internal function check if the region is user accessible or not
*/
static inline bool _is_user_accessible_region(u32_t r_index, int write)
{
u32_t r_ap;
r_ap = z_arc_v2_aux_reg_read(_ARC_V2_MPU_RDP0 + r_index * 2U);
r_ap &= AUX_MPU_RDP_ATTR_MASK;
if (write) {
return ((r_ap & (AUX_MPU_ATTR_UW | AUX_MPU_ATTR_KW)) ==
(AUX_MPU_ATTR_UW | AUX_MPU_ATTR_KW));
}
return ((r_ap & (AUX_MPU_ATTR_UR | AUX_MPU_ATTR_KR)) ==
(AUX_MPU_ATTR_UR | AUX_MPU_ATTR_KR));
}
/**
* @brief configure the base address and size for an MPU region
*
* @param type MPU region type
* @param base base address in RAM
* @param size size of the region
*/
static inline int _mpu_configure(u8_t type, u32_t base, u32_t size)
{
s32_t region_index = get_region_index_by_type(type);
u32_t region_attr = get_region_attr_by_type(type);
LOG_DBG("Region info: 0x%x 0x%x", base, size);
if (region_attr == 0U || region_index < 0) {
return -EINVAL;
}
/*
* For ARC MPU v2, MPU regions can be overlapped, smaller
* region index has higher priority.
*/
_region_init(region_index, base, size, region_attr);
return 0;
}
/* ARC Core MPU Driver API Implementation for ARC MPUv2 */
/**
* @brief enable the MPU
*/
void arc_core_mpu_enable(void)
{
/* Enable MPU */
z_arc_v2_aux_reg_write(_ARC_V2_MPU_EN,
z_arc_v2_aux_reg_read(_ARC_V2_MPU_EN) | AUX_MPU_EN_ENABLE);
}
/**
* @brief disable the MPU
*/
void arc_core_mpu_disable(void)
{
/* Disable MPU */
z_arc_v2_aux_reg_write(_ARC_V2_MPU_EN,
z_arc_v2_aux_reg_read(_ARC_V2_MPU_EN) & AUX_MPU_EN_DISABLE);
}
/**
* @brief configure the thread's MPU regions
*
* @param thread the target thread
*/
void arc_core_mpu_configure_thread(struct k_thread *thread)
{
#if defined(CONFIG_USERSPACE)
/* configure stack region of user thread */
if (thread->base.user_options & K_USER) {
LOG_DBG("configure user thread %p's stack", thread);
if (_mpu_configure(THREAD_STACK_USER_REGION,
(u32_t)thread->stack_obj, thread->stack_info.size) < 0) {
LOG_ERR("user thread %p's stack failed", thread);
return;
}
}
LOG_DBG("configure thread %p's domain", thread);
arc_core_mpu_configure_mem_domain(thread);
#endif
}
/**
* @brief configure the default region
*
* @param region_attr region attribute of default region
*/
void arc_core_mpu_default(u32_t region_attr)
{
u32_t val = z_arc_v2_aux_reg_read(_ARC_V2_MPU_EN) &
(~AUX_MPU_RDP_ATTR_MASK);
region_attr &= AUX_MPU_RDP_ATTR_MASK;
z_arc_v2_aux_reg_write(_ARC_V2_MPU_EN, region_attr | val);
}
/**
* @brief configure the MPU region
*
* @param index MPU region index
* @param base base address
* @param region_attr region attribute
*/
int arc_core_mpu_region(u32_t index, u32_t base, u32_t size,
u32_t region_attr)
{
if (index >= get_num_regions()) {
return -EINVAL;
}
region_attr &= AUX_MPU_RDP_ATTR_MASK;
_region_init(index, base, size, region_attr);
return 0;
}
#if defined(CONFIG_USERSPACE)
/**
* @brief configure MPU regions for the memory partitions of the memory domain
*
* @param thread the thread which has memory domain
*/
void arc_core_mpu_configure_mem_domain(struct k_thread *thread)
{
int region_index =
get_region_index_by_type(THREAD_DOMAIN_PARTITION_REGION);
u32_t num_partitions;
struct k_mem_partition *pparts;
struct k_mem_domain *mem_domain = NULL;
if (thread) {
mem_domain = thread->mem_domain_info.mem_domain;
}
if (mem_domain) {
LOG_DBG("configure domain: %p", mem_domain);
num_partitions = mem_domain->num_partitions;
pparts = mem_domain->partitions;
} else {
LOG_DBG("disable domain partition regions");
num_partitions = 0U;
pparts = NULL;
}
for (; region_index >= 0; region_index--) {
if (num_partitions) {
LOG_DBG("set region 0x%x 0x%lx 0x%x",
region_index, pparts->start, pparts->size);
_region_init(region_index, pparts->start,
pparts->size, pparts->attr);
num_partitions--;
} else {
/* clear the left mpu entries */
_region_init(region_index, 0, 0, 0);
}
pparts++;
}
}
/**
* @brief remove MPU regions for the memory partitions of the memory domain
*
* @param mem_domain the target memory domain
*/
void arc_core_mpu_remove_mem_domain(struct k_mem_domain *mem_domain)
{
ARG_UNUSED(mem_domain);
int region_index =
get_region_index_by_type(THREAD_DOMAIN_PARTITION_REGION);
for (; region_index >= 0; region_index--) {
_region_init(region_index, 0, 0, 0);
}
}
/**
* @brief reset MPU region for a single memory partition
*
* @param domain the target memory domain
* @param partition_id memory partition id
*/
void arc_core_mpu_remove_mem_partition(struct k_mem_domain *domain,
u32_t part_id)
{
ARG_UNUSED(domain);
int region_index =
get_region_index_by_type(THREAD_DOMAIN_PARTITION_REGION);
LOG_DBG("disable region 0x%x", region_index + part_id);
/* Disable region */
_region_init(region_index + part_id, 0, 0, 0);
}
/**
* @brief get the maximum number of free regions for memory domain partitions
*/
int arc_core_mpu_get_max_domain_partition_regions(void)
{
return get_region_index_by_type(THREAD_DOMAIN_PARTITION_REGION) + 1;
}
/**
* @brief validate the given buffer is user accessible or not
*/
int arc_core_mpu_buffer_validate(void *addr, size_t size, int write)
{
int r_index;
/*
* For ARC MPU v2, smaller region number takes priority.
* we can stop the iteration immediately once we find the
* matched region that grants permission or denies access.
*
*/
for (r_index = 0; r_index < get_num_regions(); r_index++) {
if (!_is_enabled_region(r_index) ||
!_is_in_region(r_index, (u32_t)addr, size)) {
continue;
}
if (_is_user_accessible_region(r_index, write)) {
return 0;
} else {
return -EPERM;
}
}
return -EPERM;
}
#endif /* CONFIG_USERSPACE */
/* ARC MPU Driver Initial Setup */
/*
* @brief MPU default initialization and configuration
*
* This function provides the default configuration mechanism for the Memory
* Protection Unit (MPU).
*/
static int arc_mpu_init(struct device *arg)
{
ARG_UNUSED(arg);
u32_t num_regions;
u32_t i;
num_regions = get_num_regions();
/* ARC MPU supports up to 16 Regions */
if (mpu_config.num_regions > num_regions) {
__ASSERT(0,
"Request to configure: %u regions (supported: %u)\n",
mpu_config.num_regions, num_regions);
return -EINVAL;
}
/* Disable MPU */
arc_core_mpu_disable();
int r_index;
/*
* the MPU regions are filled in the reverse order.
* According to ARCv2 ISA, the MPU region with smaller
* index has higher priority. The static background MPU
* regions in mpu_config will be in the bottom. Then
* the special type regions will be above.
*
*/
r_index = num_regions - mpu_config.num_regions;
/* clear all the regions first */
for (i = 0U; i < r_index; i++) {
_region_init(i, 0, 0, 0);
}
/* configure the static regions */
for (i = 0U; i < mpu_config.num_regions; i++) {
_region_init(r_index,
mpu_config.mpu_regions[i].base,
mpu_config.mpu_regions[i].size,
mpu_config.mpu_regions[i].attr);
r_index++;
}
/* default region: no read, write and execute */
arc_core_mpu_default(0);
/* Enable MPU */
arc_core_mpu_enable();
return 0;
}
SYS_INIT(arc_mpu_init, PRE_KERNEL_1,
CONFIG_KERNEL_INIT_PRIORITY_DEFAULT);
#endif /* ZEPHYR_ARCH_ARC_CORE_MPU_ARC_MPU_V2_INTERNAL_H_ */

View File

@@ -1,918 +0,0 @@
/*
* Copyright (c) 2019 Synopsys.
*
* SPDX-License-Identifier: Apache-2.0
*/
#ifndef ZEPHYR_ARCH_ARC_CORE_MPU_ARC_MPU_V3_INTERNAL_H_
#define ZEPHYR_ARCH_ARC_CORE_MPU_ARC_MPU_V3_INTERNAL_H_
#define AUX_MPU_RPER_SID1 0x10000
/* valid mask: SID1+secure+valid */
#define AUX_MPU_RPER_VALID_MASK ((0x1) | AUX_MPU_RPER_SID1 | AUX_MPU_ATTR_S)
#define AUX_MPU_RPER_ATTR_MASK (0x1FF)
#define _ARC_V2_MPU_EN (0x409)
/* aux regs added in MPU version 3 */
#define _ARC_V2_MPU_INDEX (0x448) /* MPU index */
#define _ARC_V2_MPU_RSTART (0x449) /* MPU region start address */
#define _ARC_V2_MPU_REND (0x44A) /* MPU region end address */
#define _ARC_V2_MPU_RPER (0x44B) /* MPU region permission register */
#define _ARC_V2_MPU_PROBE (0x44C) /* MPU probe register */
/* For MPU version 3, the minimum protection region size is 32 bytes */
#define ARC_FEATURE_MPU_ALIGNMENT_BITS 5
#define CALC_REGION_END_ADDR(start, size) \
(start + size - (1 << ARC_FEATURE_MPU_ALIGNMENT_BITS))
/* ARC MPU version 3 does not support mpu region overlap in hardware
* so if we want to allocate MPU region dynamically, e.g. thread stack,
* memory domain from a background region, a dynamic region splitting
* approach is designed. pls see comments in
* _dynamic_region_allocate_and_init
* But this approach has an impact on performance of thread switch.
* As a trade off, we can use the default mpu region as the background region
* to avoid the dynamic region splitting. This will give more privilege to
* codes in kernel mode which can access the memory region not covered by
* explicit mpu entry. Considering memory protection is mainly used to
* isolate malicious codes in user mode, it makes sense to get better
* thread switch performance through default mpu region.
* CONFIG_MPU_GAP_FILLING is used to turn this on/off.
*
*/
#if defined(CONFIG_MPU_GAP_FILLING)
#if defined(CONFIG_USERSPACE) && defined(CONFIG_MPU_STACK_GUARD)
/* 1 for stack guard , 1 for user thread, 1 for split */
#define MPU_REGION_NUM_FOR_THREAD 3
#elif defined(CONFIG_USERSPACE) || defined(CONFIG_MPU_STACK_GUARD)
/* 1 for stack guard or user thread stack , 1 for split */
#define MPU_REGION_NUM_FOR_THREAD 2
#else
#define MPU_REGION_NUM_FOR_THREAD 0
#endif
#define MPU_DYNAMIC_REGION_AREAS_NUM 2
/**
* @brief internal structure holding information of
* memory areas where dynamic MPU programming is allowed.
*/
struct dynamic_region_info {
u8_t index;
u32_t base;
u32_t size;
u32_t attr;
};
static u8_t dynamic_regions_num;
static u8_t dynamic_region_index;
/**
* Global array, holding the MPU region index of
* the memory region inside which dynamic memory
* regions may be configured.
*/
static struct dynamic_region_info dyn_reg_info[MPU_DYNAMIC_REGION_AREAS_NUM];
#endif /* CONFIG_MPU_GAP_FILLING */
static u8_t static_regions_num;
#ifdef CONFIG_ARC_NORMAL_FIRMWARE
/* \todo through secure service to access mpu */
static inline void _region_init(u32_t index, u32_t region_addr, u32_t size,
u32_t region_attr)
{
}
static inline void _region_set_attr(u32_t index, u32_t attr)
{
}
static inline u32_t _region_get_attr(u32_t index)
{
return 0;
}
static inline u32_t _region_get_start(u32_t index)
{
return 0;
}
static inline void _region_set_start(u32_t index, u32_t start)
{
}
static inline u32_t _region_get_end(u32_t index)
{
return 0;
}
static inline void _region_set_end(u32_t index, u32_t end)
{
}
/**
* This internal function probes the given addr's MPU index.if not
* in MPU, returns error
*/
static inline int _mpu_probe(u32_t addr)
{
return -EINVAL;
}
/**
* This internal function checks if MPU region is enabled or not
*/
static inline bool _is_enabled_region(u32_t r_index)
{
return false;
}
/**
* This internal function check if the region is user accessible or not
*/
static inline bool _is_user_accessible_region(u32_t r_index, int write)
{
return false;
}
#else /* CONFIG_ARC_NORMAL_FIRMWARE */
/* the following functions are prepared for SECURE_FRIMWARE */
static inline void _region_init(u32_t index, u32_t region_addr, u32_t size,
u32_t region_attr)
{
if (size < (1 << ARC_FEATURE_MPU_ALIGNMENT_BITS)) {
size = (1 << ARC_FEATURE_MPU_ALIGNMENT_BITS);
}
if (region_attr) {
region_attr &= AUX_MPU_RPER_ATTR_MASK;
region_attr |= AUX_MPU_RPER_VALID_MASK;
}
z_arc_v2_aux_reg_write(_ARC_V2_MPU_INDEX, index);
z_arc_v2_aux_reg_write(_ARC_V2_MPU_RSTART, region_addr);
z_arc_v2_aux_reg_write(_ARC_V2_MPU_REND,
CALC_REGION_END_ADDR(region_addr, size));
z_arc_v2_aux_reg_write(_ARC_V2_MPU_RPER, region_attr);
}
static inline void _region_set_attr(u32_t index, u32_t attr)
{
z_arc_v2_aux_reg_write(_ARC_V2_MPU_INDEX, index);
z_arc_v2_aux_reg_write(_ARC_V2_MPU_RPER, attr |
AUX_MPU_RPER_VALID_MASK);
}
static inline u32_t _region_get_attr(u32_t index)
{
z_arc_v2_aux_reg_write(_ARC_V2_MPU_INDEX, index);
return z_arc_v2_aux_reg_read(_ARC_V2_MPU_RPER);
}
static inline u32_t _region_get_start(u32_t index)
{
z_arc_v2_aux_reg_write(_ARC_V2_MPU_INDEX, index);
return z_arc_v2_aux_reg_read(_ARC_V2_MPU_RSTART);
}
static inline void _region_set_start(u32_t index, u32_t start)
{
z_arc_v2_aux_reg_write(_ARC_V2_MPU_INDEX, index);
z_arc_v2_aux_reg_write(_ARC_V2_MPU_RSTART, start);
}
static inline u32_t _region_get_end(u32_t index)
{
z_arc_v2_aux_reg_write(_ARC_V2_MPU_INDEX, index);
return z_arc_v2_aux_reg_read(_ARC_V2_MPU_REND) +
(1 << ARC_FEATURE_MPU_ALIGNMENT_BITS);
}
static inline void _region_set_end(u32_t index, u32_t end)
{
z_arc_v2_aux_reg_write(_ARC_V2_MPU_INDEX, index);
z_arc_v2_aux_reg_write(_ARC_V2_MPU_REND, end -
(1 << ARC_FEATURE_MPU_ALIGNMENT_BITS));
}
/**
* This internal function probes the given addr's MPU index.if not
* in MPU, returns error
*/
static inline int _mpu_probe(u32_t addr)
{
u32_t val;
z_arc_v2_aux_reg_write(_ARC_V2_MPU_PROBE, addr);
val = z_arc_v2_aux_reg_read(_ARC_V2_MPU_INDEX);
/* if no match or multiple regions match, return error */
if (val & 0xC0000000) {
return -EINVAL;
} else {
return val;
}
}
/**
* This internal function checks if MPU region is enabled or not
*/
static inline bool _is_enabled_region(u32_t r_index)
{
z_arc_v2_aux_reg_write(_ARC_V2_MPU_INDEX, r_index);
return ((z_arc_v2_aux_reg_read(_ARC_V2_MPU_RPER) &
AUX_MPU_RPER_VALID_MASK) == AUX_MPU_RPER_VALID_MASK);
}
/**
* This internal function check if the region is user accessible or not
*/
static inline bool _is_user_accessible_region(u32_t r_index, int write)
{
u32_t r_ap;
z_arc_v2_aux_reg_write(_ARC_V2_MPU_INDEX, r_index);
r_ap = z_arc_v2_aux_reg_read(_ARC_V2_MPU_RPER);
r_ap &= AUX_MPU_RPER_ATTR_MASK;
if (write) {
return ((r_ap & (AUX_MPU_ATTR_UW | AUX_MPU_ATTR_KW)) ==
(AUX_MPU_ATTR_UW | AUX_MPU_ATTR_KW));
}
return ((r_ap & (AUX_MPU_ATTR_UR | AUX_MPU_ATTR_KR)) ==
(AUX_MPU_ATTR_UR | AUX_MPU_ATTR_KR));
}
#endif /* CONFIG_ARC_NORMAL_FIRMWARE */
/**
* This internal function checks the area given by (start, size)
* and returns the index if the area match one MPU entry
*/
static inline int _get_region_index(u32_t start, u32_t size)
{
int index = _mpu_probe(start);
if (index > 0 && index == _mpu_probe(start + size - 1)) {
return index;
}
return -EINVAL;
}
#if defined(CONFIG_MPU_GAP_FILLING)
/**
* This internal function allocates a dynamic MPU region and returns
* the index or error
*/
static inline int _dynamic_region_allocate_index(void)
{
if (dynamic_region_index >= get_num_regions()) {
LOG_ERR("no enough mpu entries %d", dynamic_region_index);
return -EINVAL;
}
return dynamic_region_index++;
}
/* @brief allocate and init a dynamic MPU region
*
* This internal function performs the allocation and initialization of
* a dynamic MPU region
*
* @param base region base
* @param size region size
* @param attr region attribute
* @return <0 failure, >0 allocated dynamic region index
*/
static int _dynamic_region_allocate_and_init(u32_t base, u32_t size,
u32_t attr)
{
int u_region_index = _get_region_index(base, size);
int region_index;
LOG_DBG("Region info: base 0x%x size 0x%x attr 0x%x", base, size, attr);
if (u_region_index == -EINVAL) {
/* no underlying region */
region_index = _dynamic_region_allocate_index();
if (region_index > 0) {
/* a new region */
_region_init(region_index, base, size, attr);
}
return region_index;
}
/*
* The new memory region is to be placed inside the underlying
* region, possibly splitting the underlying region into two.
*/
u32_t u_region_start = _region_get_start(u_region_index);
u32_t u_region_end = _region_get_end(u_region_index);
u32_t u_region_attr = _region_get_attr(u_region_index);
u32_t end = base + size;
if ((base == u_region_start) && (end == u_region_end)) {
/* The new region overlaps entirely with the
* underlying region. In this case we simply
* update the partition attributes of the
* underlying region with those of the new
* region.
*/
_region_init(u_region_index, base, size, attr);
region_index = u_region_index;
} else if (base == u_region_start) {
/* The new region starts exactly at the start of the
* underlying region; the start of the underlying
* region needs to be set to the end of the new region.
*/
_region_set_start(u_region_index, base + size);
_region_set_attr(u_region_index, u_region_attr);
region_index = _dynamic_region_allocate_index();
if (region_index > 0) {
_region_init(region_index, base, size, attr);
}
} else if (end == u_region_end) {
/* The new region ends exactly at the end of the
* underlying region; the end of the underlying
* region needs to be set to the start of the
* new region.
*/
_region_set_end(u_region_index, base);
_region_set_attr(u_region_index, u_region_attr);
region_index = _dynamic_region_allocate_index();
if (region_index > 0) {
_region_init(region_index, base, size, attr);
}
} else {
/* The new region lies strictly inside the
* underlying region, which needs to split
* into two regions.
*/
_region_set_end(u_region_index, base);
_region_set_attr(u_region_index, u_region_attr);
region_index = _dynamic_region_allocate_index();
if (region_index > 0) {
_region_init(region_index, base, size, attr);
region_index = _dynamic_region_allocate_index();
if (region_index > 0) {
_region_init(region_index, base + size,
u_region_end - end, u_region_attr);
}
}
}
return region_index;
}
/* @brief reset the dynamic MPU regions
*
* This internal function performs the reset of dynamic MPU regions
*/
static void _mpu_reset_dynamic_regions(void)
{
u32_t i;
u32_t num_regions = get_num_regions();
for (i = static_regions_num; i < num_regions; i++) {
_region_init(i, 0, 0, 0);
}
for (i = 0U; i < dynamic_regions_num; i++) {
_region_init(
dyn_reg_info[i].index,
dyn_reg_info[i].base,
dyn_reg_info[i].size,
dyn_reg_info[i].attr);
}
/* dynamic regions are after static regions */
dynamic_region_index = static_regions_num;
}
/**
* @brief configure the base address and size for an MPU region
*
* @param type MPU region type
* @param base base address in RAM
* @param size size of the region
*/
static inline int _mpu_configure(u8_t type, u32_t base, u32_t size)
{
u32_t region_attr = get_region_attr_by_type(type);
return _dynamic_region_allocate_and_init(base, size, region_attr);
}
#else
/**
* This internal function is utilized by the MPU driver to parse the intent
* type (i.e. THREAD_STACK_REGION) and return the correct region index.
*/
static inline int get_region_index_by_type(u32_t type)
{
/*
* The new MPU regions are allocated per type after the statically
* configured regions. The type is one-indexed rather than
* zero-indexed.
*
* For ARC MPU v2, the smaller index has higher priority, so the
* index is allocated in reverse order. Static regions start from
* the biggest index, then thread related regions.
*
*/
switch (type) {
case THREAD_STACK_USER_REGION:
return static_regions_num + THREAD_STACK_REGION;
case THREAD_STACK_REGION:
case THREAD_APP_DATA_REGION:
case THREAD_STACK_GUARD_REGION:
return static_regions_num + type;
case THREAD_DOMAIN_PARTITION_REGION:
#if defined(CONFIG_MPU_STACK_GUARD)
return static_regions_num + type;
#else
/*
* Start domain partition region from stack guard region
* since stack guard is not enabled.
*/
return static_regions_num + type - 1;
#endif
default:
__ASSERT(0, "Unsupported type");
return -EINVAL;
}
}
/**
* @brief configure the base address and size for an MPU region
*
* @param type MPU region type
* @param base base address in RAM
* @param size size of the region
*/
static inline int _mpu_configure(u8_t type, u32_t base, u32_t size)
{
int region_index = get_region_index_by_type(type);
u32_t region_attr = get_region_attr_by_type(type);
LOG_DBG("Region info: 0x%x 0x%x", base, size);
if (region_attr == 0U || region_index < 0) {
return -EINVAL;
}
_region_init(region_index, base, size, region_attr);
return 0;
}
#endif
/* ARC Core MPU Driver API Implementation for ARC MPUv3 */
/**
* @brief enable the MPU
*/
void arc_core_mpu_enable(void)
{
#ifdef CONFIG_ARC_SECURE_FIRMWARE
/* the default region:
* secure:0x8000, SID:0x10000, KW:0x100 KR:0x80
*/
#define MPU_ENABLE_ATTR 0x18180
#else
#define MPU_ENABLE_ATTR 0
#endif
arc_core_mpu_default(MPU_ENABLE_ATTR);
}
/**
* @brief disable the MPU
*/
void arc_core_mpu_disable(void)
{
/* MPU is always enabled, use default region to
* simulate MPU disable
*/
arc_core_mpu_default(REGION_ALL_ATTR | AUX_MPU_ATTR_S |
AUX_MPU_RPER_SID1);
}
/**
* @brief configure the thread's mpu regions
*
* @param thread the target thread
*/
void arc_core_mpu_configure_thread(struct k_thread *thread)
{
#if defined(CONFIG_MPU_GAP_FILLING)
/* the mpu entries of ARC MPUv3 are divided into 2 parts:
* static entries: global mpu entries, not changed in context switch
* dynamic entries: MPU entries changed in context switch and
* memory domain configure, including:
* MPU entries for user thread stack
* MPU entries for stack guard
* MPU entries for mem domain
* MPU entries for other thread specific regions
* before configuring thread specific mpu entries, need to reset dynamic
* entries
*/
_mpu_reset_dynamic_regions();
#endif
#if defined(CONFIG_MPU_STACK_GUARD)
#if defined(CONFIG_USERSPACE)
if ((thread->base.user_options & K_USER) != 0U) {
/* the areas before and after the user stack of thread is
* kernel only. These area can be used as stack guard.
* -----------------------
* | kernel only area |
* |---------------------|
* | user stack |
* |---------------------|
* |privilege stack guard|
* |---------------------|
* | privilege stack |
* -----------------------
*/
if (_mpu_configure(THREAD_STACK_GUARD_REGION,
thread->arch.priv_stack_start - STACK_GUARD_SIZE,
STACK_GUARD_SIZE) < 0) {
LOG_ERR("thread %p's stack guard failed", thread);
return;
}
} else {
if (_mpu_configure(THREAD_STACK_GUARD_REGION,
thread->stack_info.start - STACK_GUARD_SIZE,
STACK_GUARD_SIZE) < 0) {
LOG_ERR("thread %p's stack guard failed", thread);
return;
}
}
#else
if (_mpu_configure(THREAD_STACK_GUARD_REGION,
thread->stack_info.start - STACK_GUARD_SIZE,
STACK_GUARD_SIZE) < 0) {
LOG_ERR("thread %p's stack guard failed", thread);
return;
}
#endif
#endif
#if defined(CONFIG_USERSPACE)
/* configure stack region of user thread */
if (thread->base.user_options & K_USER) {
LOG_DBG("configure user thread %p's stack", thread);
if (_mpu_configure(THREAD_STACK_USER_REGION,
(u32_t)thread->stack_obj, thread->stack_info.size) < 0) {
LOG_ERR("thread %p's stack failed", thread);
return;
}
}
#if defined(CONFIG_MPU_GAP_FILLING)
u32_t num_partitions;
struct k_mem_partition *pparts;
struct k_mem_domain *mem_domain = thread->mem_domain_info.mem_domain;
/* configure thread's memory domain */
if (mem_domain) {
LOG_DBG("configure thread %p's domain: %p",
thread, mem_domain);
num_partitions = mem_domain->num_partitions;
pparts = mem_domain->partitions;
} else {
num_partitions = 0U;
pparts = NULL;
}
for (u32_t i = 0; i < num_partitions; i++) {
if (pparts->size) {
if (_dynamic_region_allocate_and_init(pparts->start,
pparts->size, pparts->attr) < 0) {
LOG_ERR(
"thread %p's mem region: %p failed",
thread, pparts);
return;
}
}
pparts++;
}
#else
arc_core_mpu_configure_mem_domain(thread);
#endif
#endif
}
/**
* @brief configure the default region
*
* @param region_attr region attribute of default region
*/
void arc_core_mpu_default(u32_t region_attr)
{
#ifdef CONFIG_ARC_NORMAL_FIRMWARE
/* \todo through secure service to access mpu */
#else
z_arc_v2_aux_reg_write(_ARC_V2_MPU_EN, region_attr);
#endif
}
/**
* @brief configure the MPU region
*
* @param index MPU region index
* @param base base address
* @param size region size
* @param region_attr region attribute
*/
int arc_core_mpu_region(u32_t index, u32_t base, u32_t size,
u32_t region_attr)
{
if (index >= get_num_regions()) {
return -EINVAL;
}
region_attr &= AUX_MPU_RPER_ATTR_MASK;
_region_init(index, base, size, region_attr);
return 0;
}
#if defined(CONFIG_USERSPACE)
/**
* @brief configure MPU regions for the memory partitions of the memory domain
*
* @param thread the thread which has memory domain
*/
#if defined(CONFIG_MPU_GAP_FILLING)
void arc_core_mpu_configure_mem_domain(struct k_thread *thread)
{
arc_core_mpu_configure_thread(thread);
}
#else
void arc_core_mpu_configure_mem_domain(struct k_thread *thread)
{
u32_t region_index;
u32_t num_partitions;
u32_t num_regions;
struct k_mem_partition *pparts;
struct k_mem_domain *mem_domain = NULL;
if (thread) {
mem_domain = thread->mem_domain_info.mem_domain;
}
if (mem_domain) {
LOG_DBG("configure domain: %p", mem_domain);
num_partitions = mem_domain->num_partitions;
pparts = mem_domain->partitions;
} else {
LOG_DBG("disable domain partition regions");
num_partitions = 0U;
pparts = NULL;
}
num_regions = get_num_regions();
region_index = get_region_index_by_type(THREAD_DOMAIN_PARTITION_REGION);
while (num_partitions && region_index < num_regions) {
if (pparts->size > 0) {
LOG_DBG("set region 0x%x 0x%lx 0x%x",
region_index, pparts->start, pparts->size);
_region_init(region_index, pparts->start,
pparts->size, pparts->attr);
region_index++;
}
pparts++;
num_partitions--;
}
while (region_index < num_regions) {
/* clear the left mpu entries */
_region_init(region_index, 0, 0, 0);
region_index++;
}
}
#endif
/**
* @brief remove MPU regions for the memory partitions of the memory domain
*
* @param mem_domain the target memory domain
*/
void arc_core_mpu_remove_mem_domain(struct k_mem_domain *mem_domain)
{
u32_t num_partitions;
struct k_mem_partition *pparts;
int index;
if (mem_domain) {
LOG_DBG("configure domain: %p", mem_domain);
num_partitions = mem_domain->num_partitions;
pparts = mem_domain->partitions;
} else {
LOG_DBG("disable domain partition regions");
num_partitions = 0U;
pparts = NULL;
}
for (u32_t i = 0; i < num_partitions; i++) {
if (pparts->size) {
index = _get_region_index(pparts->start,
pparts->size);
if (index > 0) {
#if defined(CONFIG_MPU_GAP_FILLING)
_region_set_attr(index,
REGION_KERNEL_RAM_ATTR);
#else
_region_init(index, 0, 0, 0);
#endif
}
}
pparts++;
}
}
/**
* @brief reset MPU region for a single memory partition
*
* @param partition_id memory partition id
*/
void arc_core_mpu_remove_mem_partition(struct k_mem_domain *domain,
u32_t partition_id)
{
struct k_mem_partition *partition = &domain->partitions[partition_id];
int region_index = _get_region_index(partition->start,
partition->size);
if (region_index < 0) {
return;
}
LOG_DBG("remove region 0x%x", region_index);
#if defined(CONFIG_MPU_GAP_FILLING)
_region_set_attr(region_index, REGION_KERNEL_RAM_ATTR);
#else
_region_init(region_index, 0, 0, 0);
#endif
}
/**
* @brief get the maximum number of free regions for memory domain partitions
*/
int arc_core_mpu_get_max_domain_partition_regions(void)
{
#if defined(CONFIG_MPU_GAP_FILLING)
/* consider the worst case: each partition requires split */
return (get_num_regions() - MPU_REGION_NUM_FOR_THREAD) / 2;
#else
return get_num_regions() -
get_region_index_by_type(THREAD_DOMAIN_PARTITION_REGION) - 1;
#endif
}
/**
* @brief validate the given buffer is user accessible or not
*/
int arc_core_mpu_buffer_validate(void *addr, size_t size, int write)
{
int r_index;
int key = arch_irq_lock();
/*
* For ARC MPU v3, overlapping is not supported.
* we can stop the iteration immediately once we find the
* matched region that grants permission or denies access.
*/
r_index = _mpu_probe((u32_t)addr);
/* match and the area is in one region */
if (r_index >= 0 && r_index == _mpu_probe((u32_t)addr + (size - 1))) {
if (_is_user_accessible_region(r_index, write)) {
r_index = 0;
} else {
r_index = -EPERM;
}
} else {
r_index = -EPERM;
}
arch_irq_unlock(key);
return r_index;
}
#endif /* CONFIG_USERSPACE */
/* ARC MPU Driver Initial Setup */
/*
* @brief MPU default initialization and configuration
*
* This function provides the default configuration mechanism for the Memory
* Protection Unit (MPU).
*/
static int arc_mpu_init(struct device *arg)
{
ARG_UNUSED(arg);
u32_t num_regions;
u32_t i;
num_regions = get_num_regions();
/* ARC MPU supports up to 16 Regions */
if (mpu_config.num_regions > num_regions) {
__ASSERT(0,
"Request to configure: %u regions (supported: %u)\n",
mpu_config.num_regions, num_regions);
return -EINVAL;
}
static_regions_num = 0;
/* Disable MPU */
arc_core_mpu_disable();
for (i = 0U; i < mpu_config.num_regions; i++) {
/* skip empty region */
if (mpu_config.mpu_regions[i].size == 0) {
continue;
}
#if defined(CONFIG_MPU_GAP_FILLING)
_region_init(static_regions_num,
mpu_config.mpu_regions[i].base,
mpu_config.mpu_regions[i].size,
mpu_config.mpu_regions[i].attr);
/* record the static region which can be split */
if (mpu_config.mpu_regions[i].attr & REGION_DYNAMIC) {
if (dynamic_regions_num >=
MPU_DYNAMIC_REGION_AREAS_NUM) {
LOG_ERR("not enough dynamic regions %d",
dynamic_regions_num);
return -EINVAL;
}
dyn_reg_info[dynamic_regions_num].index = i;
dyn_reg_info[dynamic_regions_num].base =
mpu_config.mpu_regions[i].base;
dyn_reg_info[dynamic_regions_num].size =
mpu_config.mpu_regions[i].size;
dyn_reg_info[dynamic_regions_num].attr =
mpu_config.mpu_regions[i].attr;
dynamic_regions_num++;
}
static_regions_num++;
#else
/* dynamic region will be covered by default mpu setting
* no need to configure
*/
if (!(mpu_config.mpu_regions[i].attr & REGION_DYNAMIC)) {
_region_init(static_regions_num,
mpu_config.mpu_regions[i].base,
mpu_config.mpu_regions[i].size,
mpu_config.mpu_regions[i].attr);
static_regions_num++;
}
#endif
}
for (i = static_regions_num; i < num_regions; i++) {
_region_init(i, 0, 0, 0);
}
/* Enable MPU */
arc_core_mpu_enable();
return 0;
}
SYS_INIT(arc_mpu_init, PRE_KERNEL_1,
CONFIG_KERNEL_INIT_PRIORITY_DEFAULT);
#endif /* ZEPHYR_ARCH_ARC_CORE_MPU_ARC_MPU_V3_INTERNAL_H_ */

View File

@@ -1,19 +1,30 @@
/*
* Copyright (c) 2014 Wind River Systems, Inc.
*
* SPDX-License-Identifier: Apache-2.0
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
* @file
* @brief ARCv2 kernel structure member offset definition file
* @brief ARCv2 nano kernel structure member offset definition file
*
* This module is responsible for the generation of the absolute symbols whose
* value represents the member offsets for various ARCv2 kernel structures.
* value represents the member offsets for various ARCv2 nanokernel
* structures.
*
* All of the absolute symbols defined by this module will be present in the
* final kernel ELF image (due to the linker's reference to the _OffsetAbsSyms
* symbol).
* final microkernel or nanokernel ELF image (due to the linker's reference to
* the _OffsetAbsSyms symbol).
*
* INTERNAL
* It is NOT necessary to define the offset for every member of a structure.
@@ -22,25 +33,16 @@
* completeness.
*/
#include <kernel.h>
#include <kernel_arch_data.h>
#include <gen_offset.h>
#include <kernel_structs.h>
#include <kernel_offsets.h>
GEN_OFFSET_SYM(_thread_arch_t, intlock_key);
GEN_OFFSET_SYM(_thread_arch_t, relinquish_cause);
GEN_OFFSET_SYM(_thread_arch_t, return_value);
#ifdef CONFIG_ARC_STACK_CHECKING
GEN_OFFSET_SYM(_thread_arch_t, k_stack_base);
GEN_OFFSET_SYM(_thread_arch_t, k_stack_top);
#ifdef CONFIG_USERSPACE
GEN_OFFSET_SYM(_thread_arch_t, u_stack_base);
GEN_OFFSET_SYM(_thread_arch_t, u_stack_top);
GEN_OFFSET_SYM(_thread_arch_t, stack_top);
#endif
#endif
#ifdef CONFIG_USERSPACE
GEN_OFFSET_SYM(_thread_arch_t, priv_stack_start);
#endif
/* ARCv2-specific IRQ stack frame structure member offsets */
GEN_OFFSET_SYM(_isf_t, r0);
@@ -61,15 +63,7 @@ GEN_OFFSET_SYM(_isf_t, blink);
GEN_OFFSET_SYM(_isf_t, lp_end);
GEN_OFFSET_SYM(_isf_t, lp_start);
GEN_OFFSET_SYM(_isf_t, lp_count);
#ifdef CONFIG_CODE_DENSITY
GEN_OFFSET_SYM(_isf_t, ei_base);
GEN_OFFSET_SYM(_isf_t, ldi_base);
GEN_OFFSET_SYM(_isf_t, jli_base);
#endif
GEN_OFFSET_SYM(_isf_t, pc);
#ifdef CONFIG_ARC_HAS_SECURE
GEN_OFFSET_SYM(_isf_t, sec_stat);
#endif
GEN_OFFSET_SYM(_isf_t, status32);
GEN_ABSOLUTE_SYM(___isf_t_SIZEOF, sizeof(_isf_t));
@@ -91,30 +85,7 @@ GEN_OFFSET_SYM(_callee_saved_stack_t, r24);
GEN_OFFSET_SYM(_callee_saved_stack_t, r25);
GEN_OFFSET_SYM(_callee_saved_stack_t, r26);
GEN_OFFSET_SYM(_callee_saved_stack_t, fp);
#ifdef CONFIG_USERSPACE
#ifdef CONFIG_ARC_HAS_SECURE
GEN_OFFSET_SYM(_callee_saved_stack_t, kernel_sp);
GEN_OFFSET_SYM(_callee_saved_stack_t, user_sp);
#else
GEN_OFFSET_SYM(_callee_saved_stack_t, user_sp);
#endif
#endif
GEN_OFFSET_SYM(_callee_saved_stack_t, r30);
#ifdef CONFIG_ARC_HAS_ACCL_REGS
GEN_OFFSET_SYM(_callee_saved_stack_t, r58);
GEN_OFFSET_SYM(_callee_saved_stack_t, r59);
#endif
#ifdef CONFIG_FPU_SHARING
GEN_OFFSET_SYM(_callee_saved_stack_t, fpu_status);
GEN_OFFSET_SYM(_callee_saved_stack_t, fpu_ctrl);
#ifdef CONFIG_FP_FPU_DA
GEN_OFFSET_SYM(_callee_saved_stack_t, dpfp2h);
GEN_OFFSET_SYM(_callee_saved_stack_t, dpfp2l);
GEN_OFFSET_SYM(_callee_saved_stack_t, dpfp1h);
GEN_OFFSET_SYM(_callee_saved_stack_t, dpfp1l);
#endif
#endif
GEN_ABSOLUTE_SYM(___callee_saved_stack_t_SIZEOF, sizeof(_callee_saved_stack_t));
GEN_ABSOLUTE_SYM(_K_THREAD_NO_FLOAT_SIZEOF, sizeof(struct k_thread));

View File

@@ -1,7 +1,17 @@
/*
* Copyright (c) 2014 Wind River Systems, Inc.
*
* SPDX-License-Identifier: Apache-2.0
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
@@ -10,22 +20,20 @@
*
*
* Initialization of full C support: zero the .bss, copy the .data if XIP,
* call z_cstart().
* call _Cstart().
*
* Stack is available in this module, but not the global data/bss until their
* initialization is performed.
*/
#include <zephyr/types.h>
#include <stdint.h>
#include <toolchain.h>
#include <linker/linker-defs.h>
#include <linker-defs.h>
#include <arch/arc/v2/aux_regs.h>
#include <kernel_structs.h>
#include <kernel_internal.h>
#include <nano_internal.h>
/* XXX - keep for future use in full-featured cache APIs */
#if 0
/**
*
* @brief Disable the i-cache if present
@@ -40,14 +48,14 @@ static void disable_icache(void)
{
unsigned int val;
val = z_arc_v2_aux_reg_read(_ARC_V2_I_CACHE_BUILD);
val = _arc_v2_aux_reg_read(_ARC_V2_I_CACHE_BUILD);
val &= 0xff; /* version field */
if (val == 0) {
return; /* skip if i-cache is not present */
}
z_arc_v2_aux_reg_write(_ARC_V2_IC_IVIC, 0);
_arc_v2_aux_reg_write(_ARC_V2_IC_IVIC, 0);
__asm__ __volatile__ ("nop");
z_arc_v2_aux_reg_write(_ARC_V2_IC_CTRL, 1);
_arc_v2_aux_reg_write(_ARC_V2_IC_CTRL, 1);
}
/**
@@ -64,16 +72,44 @@ static void invalidate_dcache(void)
{
unsigned int val;
val = z_arc_v2_aux_reg_read(_ARC_V2_D_CACHE_BUILD);
val = _arc_v2_aux_reg_read(_ARC_V2_D_CACHE_BUILD);
val &= 0xff; /* version field */
if (val == 0) {
return; /* skip if d-cache is not present */
}
z_arc_v2_aux_reg_write(_ARC_V2_DC_IVDC, 1);
_arc_v2_aux_reg_write(_ARC_V2_DC_IVDC, 1);
}
#endif
extern FUNC_NORETURN void z_cstart(void);
/**
*
* @brief Adjust the vector table base
*
* Set the vector table base if the value found in the
* _ARC_V2_IRQ_VECT_BASE auxiliary register is different from the
* _VectorTable known by software. It is important to do this very early
* so that exception vectors can be handled.
*
* @return N/A
*/
static void adjust_vector_table_base(void)
{
extern struct vector_table _VectorTable;
unsigned int vbr;
/* if the compiled-in vector table is different
* from the base address known by the ARC CPU,
* set the vector base to the compiled-in address.
*/
vbr = _arc_v2_aux_reg_read(_ARC_V2_IRQ_VECT_BASE);
vbr &= 0xfffffc00;
if (vbr != (unsigned int)&_VectorTable) {
_arc_v2_aux_reg_write(_ARC_V2_IRQ_VECT_BASE,
(unsigned int)&_VectorTable);
}
}
extern FUNC_NORETURN void _Cstart(void);
/**
*
* @brief Prepare to and run C code
@@ -85,9 +121,11 @@ extern FUNC_NORETURN void z_cstart(void);
void _PrepC(void)
{
z_icache_setup();
z_bss_zero();
z_data_copy();
z_cstart();
disable_icache();
invalidate_dcache();
adjust_vector_table_base();
_bss_zero();
_data_copy();
_Cstart();
CODE_UNREACHABLE;
}

View File

@@ -1,7 +1,17 @@
/*
* Copyright (c) 2014 Wind River Systems, Inc.
*
* SPDX-License-Identifier: Apache-2.0
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
@@ -14,179 +24,37 @@
* See isr_wrapper.S for details.
*/
#define _ASMLANGUAGE
#include <kernel_structs.h>
#include <offsets_short.h>
#include <toolchain.h>
#include <linker/sections.h>
#include <arch/cpu.h>
#include <swap_macros.h>
#include "swap_macros.h"
GTEXT(_rirq_enter)
GTEXT(_rirq_exit)
GTEXT(_rirq_newthread_switch)
GTEXT(_rirq_common_interrupt_swap)
GTEXT(_is_next_thread_current)
#if 0 /* TODO: when FIRQ is not present, all would be regular */
#define NUM_REGULAR_IRQ_PRIO_LEVELS CONFIG_NUM_IRQ_PRIO_LEVELS
#else
#define NUM_REGULAR_IRQ_PRIO_LEVELS (CONFIG_NUM_IRQ_PRIO_LEVELS-1)
#endif
/* note: the above define assumes that prio 0 IRQ is for FIRQ, and
* that all others are regular interrupts.
* TODO: Revist this if FIRQ becomes configurable.
*/
#if NUM_REGULAR_IRQ_PRIO_LEVELS > 1
#error "nested regular interrupts are not supported."
/*
* Nesting of Regularing interrupts is not yet supported.
* Set CONFIG_NUM_IRQ_PRIO_LEVELS to 2 even if SOC supports more.
*/
#endif
===========================================================
RETURN FROM INTERRUPT TO COOPERATIVE THREAD
===========================================================
That's a special case because:
1. We return from IRQ handler to a cooperative thread
2. During IRQ handling context switch did happen
3. Returning to a thread which previously gave control
to another thread because of:
- Calling k_sleep()
- Explicitly yielding
- Bumping into locked sync primitive etc
What (3) means is before passing control to another thread our thread
in question:
a. Stashed all precious caller-saved registers on its stack
b. Pushed return address to the top of the stack as well
That's how thread's stack looks like right before jumping to another thread:
----------------------------->8---------------------------------
PRE-CONTEXT-SWITCH STACK
lower_addr, let's say: 0x1000
--------------------------------------
SP -> | Return address; PC (Program Counter), in fact value taken from
| BLINK register in arch_switch()
--------------------------------------
| STATUS32 value, we explicitly save it here for later usage, read-on
--------------------------------------
| Caller-saved registers: some of R0-R12
--------------------------------------
|...
|...
higher_addr, let's say: 0x2000
----------------------------->8---------------------------------
When context gets switched the kernel saves callee-saved registers in the
thread's stack right on top of pre-switch contents so that's what we have:
----------------------------->8---------------------------------
POST-CONTEXT-SWITCH STACK
lower_addr, let's say: 0x1000
--------------------------------------
SP -> | Callee-saved registers: see struct _callee_saved_stack{}
| |- R13
| |- R14
| | ...
| \- FP
| ...
--------------------------------------
| Return address; PC (Program Counter)
--------------------------------------
| STATUS32 value
--------------------------------------
| Caller-saved registers: some of R0-R12
--------------------------------------
|...
|...
higher_addr, let's say: 0x2000
----------------------------->8---------------------------------
So how do we return in such a complex scenario.
First we restore callee-saved regs with help of _load_callee_saved_regs().
Now we're back to PRE-CONTEXT-SWITCH STACK (see above).
Logically our next step is to load return address from the top of the stack
and jump to that address to continue execution of the desired thread, but
we're still in interrupt handling mode and the only way to return to normal
execution mode is to execute "rtie" instruction. And here we need to deal
with peculiarities of return from IRQ on ARCv2 cores.
Instead of simple jump to a return address stored in the tip of thread's stack
(with subsequent interrupt enable) ARCv2 core additionally automatically
restores some registers from stack. Most important ones are
PC ("Program Counter") which holds address of the next instruction to execute
and STATUS32 which holds imortant flags including global interrupt enable,
zero, carry etc.
To make things worse depending on ARC core configuration and run-time setup
of certain features different set of registers will be restored.
Typically those same registers are automatically saved on stack on entry to
an interrupt, but remember we're returning to the thread which was
not interrupted by interrupt and so on its stack there're no automatically
saved registers, still inevitably on RTIE execution register restoration
will happen. So if we do nothing special we'll end-up with that:
----------------------------->8---------------------------------
lower_addr, let's say: 0x1000
--------------------------------------
# | Return address; PC (Program Counter)
| --------------------------------------
| | STATUS32 value
| --------------------------------------
|
sizeof(_irq_stack_frame)
|
| | Caller-saved registers: R0-R12
V --------------------------------------
|...
SP -> | < Some data on thread's stack>
|...
higher_addr, let's say: 0x2000
----------------------------->8---------------------------------
I.e. we'll go much deeper down the stack over needed return address, read
some value from unexpected location in stack and will try to jump there.
Nobody knows were we end-up then.
To work-around that problem we need to mimic existance of IRQ stack frame
of which we really only need return address obviously to return where we
need to. For that we just shift SP so that it points sizeof(_irq_stack_frame)
above like that:
----------------------------->8---------------------------------
lower_addr, let's say: 0x1000
SP -> |
A | < Some unrelated data >
| |
|
sizeof(_irq_stack_frame)
|
| --------------------------------------
| | Return address; PC (Program Counter)
| --------------------------------------
# | STATUS32 value
--------------------------------------
| Caller-saved registers: R0-R12
--------------------------------------
|...
| < Some data on thread's stack>
|...
higher_addr, let's say: 0x2000
----------------------------->8---------------------------------
Indeed R0-R13 "restored" from IRQ stack frame will contain garbage but
it makes no difference because we're returning to execution of code as if
we're returning from yet another function call and so we will restore
all needed registers from the stack.
One other important remark here is R13.
CPU hardware automatically save/restore registers in pairs and since we
wanted to save/restore R12 in IRQ stack frame as a caller-saved register we
just happen to do that for R13 as well. But given compiler treats it as
a callee-saved register we save/restore it separately in _callee_saved_stack
structure. And when we restore callee-saved registers from stack we among
other registers recover R13. But later on return from IRQ with RTIE
instruction, R13 will be "restored" again from fake IRQ stack frame and
if we don't copy correct R13 value to fake IRQ stack frame R13 value
will be corrupted.
*/
/**
*
@@ -203,26 +71,20 @@ will be corrupted.
SECTION_FUNC(TEXT, _rirq_enter)
/* the ISR will be handled in separate interrupt stack,
* so stack checking must be diabled, or exception will
* be caused
*/
_disable_stack_checking r2
clri
/* check whether irq stack is used, if
* not switch to isr stack
*/
_check_and_inc_int_nest_counter r0, r1
bne.d rirq_nest
mov_s r0, sp
_get_curr_cpu_irq_stack sp
rirq_nest:
push_s r0
seti
mov r1, _kernel
#ifdef CONFIG_ARC_STACK_CHECKING
/* disable stack checking */
lr r2, [_ARC_V2_STATUS32]
bclr r2, r2, _ARC_V2_STATUS32_SC_BIT
kflag r2
#endif
ld_s r2, [r1, _kernel_offset_to_current]
#if NUM_REGULAR_IRQ_PRIO_LEVELS == 1
st sp, [r2, _thread_offset_to_sp]
ld sp, [r1, _kernel_offset_to_irq_stack]
#else
#error regular irq nesting is not implemented
#endif
j _isr_demux
@@ -234,72 +96,142 @@ rirq_nest:
*/
SECTION_FUNC(TEXT, _rirq_exit)
mov r1, _kernel
ld_s r2, [r1, _kernel_offset_to_current]
/*
* Lock interrupts to ensure kernel queues do not change from this
* point on until return from interrupt.
*/
clri
pop sp
#if NUM_REGULAR_IRQ_PRIO_LEVELS > 1
/* check if we're a nested interrupt: if so, let the interrupted interrupt
* handle the reschedule */
_dec_int_nest_counter r0, r1
lr r3, [_ARC_V2_AUX_IRQ_ACT]
ffs r0, r3
_check_nest_int_by_irq_act r0, r1
asl r0, 1, r0
jne _rirq_no_switch
/* sp is struct k_thread **old of z_arc_switch_in_isr
* which is a wrapper of z_get_next_switch_handle.
* r0 contains the 1st thread in ready queue. if
* it equals _current(r2) ,then do swap, or no swap.
/* the OS on ARCv2 always runs in kernel mode, so assume bit31 [U] in
* AUX_IRQ_ACT is always 0: if the contents of AUX_IRQ_ACT is greater
* than FFS(AUX_IRQ_ACT), it means that another bit is set so an
* interrupt was interrupted.
*/
_get_next_switch_handle
cmp r0, r2
beq _rirq_no_switch
#ifdef CONFIG_ARC_SECURE_FIRMWARE
/* here need to remember SEC_STAT.IRM bit */
lr r3, [_ARC_V2_SEC_STAT]
push_s r3
cmp r0, r3
brgt _rirq_return_from_rirq
ld sp, [r2, _thread_offset_to_sp]
#endif
/* r2 is old thread */
_irq_store_old_thread_callee_regs
/*
* Both (a)reschedule and (b)non-reschedule cases need to load the current
* thread's stack, but don't have to use it until the decision is taken:
* load the delay slots with the 'load stack pointer' instruction.
*
* a) needs to load it to save outgoing context.
* b) needs to load it to restore the interrupted context.
*/
/* coop thread ? do not schedule */
ld_s r0, [r2, _thread_offset_to_prio]
cmp_s r0, 0
blt.d _rirq_no_reschedule
ld sp, [r2, _thread_offset_to_sp]
/* scheduler locked ? do not schedule */
ld_s r0, [r2, _thread_offset_to_sched_locked]
brgt.d r0, 0, _rirq_no_reschedule
ld sp, [r2, _thread_offset_to_sp]
/* check if the current thread needs to be rescheduled */
push_s r2
push_s r1
push_s blink
jl _is_next_thread_current
pop_s blink
pop_s r1
pop_s r2
brne.d r0, 0, _rirq_no_reschedule
ld sp, [r2, _thread_offset_to_sp]
/*
* Get the next scheduled thread. On _get_next_ready_thread
* return it is stored in r0.
*/
push_s r2
push_s r1
push_s blink
jl _get_next_ready_thread
pop_s blink
pop_s r1
pop_s r2
.balign 4
_rirq_reschedule:
/* _save_callee_saved_regs expects outgoing thread in r2 */
_save_callee_saved_regs
st _CAUSE_RIRQ, [r2, _thread_offset_to_relinquish_cause]
/* mov new thread (r0) to r2 */
/* incoming thread is in r0: it becomes the new 'current' */
mov r2, r0
st_s r2, [r1, _kernel_offset_to_current]
/* _rirq_newthread_switch required by exception handling */
.balign 4
_rirq_newthread_switch:
_rirq_common_interrupt_swap:
/* r2 contains pointer to new thread */
_load_new_thread_callee_regs
#ifdef CONFIG_ARC_STACK_CHECKING
/* Use stack top and down registers from restored context */
add r3, r2, _K_THREAD_NO_FLOAT_SIZEOF
sr r3, [_ARC_V2_KSTACK_TOP]
ld_s r3, [r2, _thread_offset_to_stack_top]
sr r3, [_ARC_V2_KSTACK_BASE]
#endif
/*
* _load_callee_saved_regs expects incoming thread in r2.
* _load_callee_saved_regs restores the stack pointer.
*/
_load_callee_saved_regs
breq r3, _CAUSE_RIRQ, _rirq_switch_from_rirq
nop_s
breq r3, _CAUSE_FIRQ, _rirq_switch_from_firq
nop_s
ld_s r3, [r2, _thread_offset_to_relinquish_cause]
breq r3, _CAUSE_RIRQ, _rirq_return_from_rirq
nop
breq r3, _CAUSE_FIRQ, _rirq_return_from_firq
nop
/* fall through */
.balign 4
_rirq_switch_from_coop:
_rirq_return_from_coop:
/* for a cooperative switch, it's not in irq, so
* need to set some regs for irq return
*/
_set_misc_regs_irq_switch_from_coop
/* status32 and pc (blink) are already on the stack in the right order */
/*
* See verbose explanation of
* RETURN FROM INTERRUPT TO COOPERATIVE THREAD above
*/
/* update status32.ie (explanation in firq_exit:_firq_return_from_coop) */
ld_s r0, [sp, 4]
ld_s r3, [r2, _thread_offset_to_intlock_key]
st 0, [r2, _thread_offset_to_intlock_key]
cmp r3, 0
or.ne r0, r0, _ARC_V2_STATUS32_IE
st_s r0, [sp, 4]
/* carve fake stack */
sub sp, sp, ___isf_t_pc_OFFSET
/*
* a) status32/pc are already on the stack
* b) a real value will be pushed in r0
*/
sub sp, sp, (___isf_t_SIZEOF - 12)
/* reset zero-overhead loops */
st 0, [sp, ___isf_t_lp_end_OFFSET]
/* push return value on stack */
ld_s r0, [r2, _thread_offset_to_return_value]
push_s r0
/*
* r13 is part of both the callee and caller-saved register sets because
@@ -309,15 +241,19 @@ _rirq_switch_from_coop:
*/
st_s r13, [sp, ___isf_t_r13_OFFSET]
/* stack now has the IRQ stack frame layout, pointing to sp */
/* rtie will pop the rest from the stack */
rtie
/* stack now has the IRQ stack frame layout, pointing to r0 */
/* fall through to rtie instruction */
.balign 4
_rirq_switch_from_firq:
_rirq_switch_from_rirq:
_rirq_return_from_firq:
_rirq_return_from_rirq:
_set_misc_regs_irq_switch_from_irq
/* rtie will pop the rest from the stack */
/* fall through to rtie instruction */
.balign 4
_rirq_no_reschedule:
_rirq_no_switch:
rtie

View File

@@ -1,7 +1,17 @@
/*
* Copyright (c) 2014 Wind River Systems, Inc.
*
* SPDX-License-Identifier: Apache-2.0
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
@@ -11,20 +21,34 @@
* Reset handler that prepares the system for running C code.
*/
#include <toolchain.h>
#include <linker/sections.h>
#include <arch/cpu.h>
#include <swap_macros.h>
#define _ASMLANGUAGE
GDATA(z_interrupt_stacks)
GDATA(z_main_stack)
GDATA(_VectorTable)
// #include <board.h>
#include <toolchain.h>
#include <sections.h>
#include <arch/cpu.h>
GDATA(_interrupt_stack)
GDATA(_firq_stack)
GDATA(_main_stack)
/* use one of the available interrupt stacks during init */
/* FIRQ only ? */
#if CONFIG_NUM_IRQ_PRIO_LEVELS == 1
#define INIT_STACK z_interrupt_stacks
#define INIT_STACK_SIZE CONFIG_ISR_STACK_SIZE
/* FIRQ, but uses _interrupt_stack ? */
#if CONFIG_RGF_NUM_BANKS == 1
#define INIT_STACK _interrupt_stack
#define INIT_STACK_SIZE CONFIG_ISR_STACK_SIZE
#else
#define INIT_STACK _firq_stack
#define INIT_STACK_SIZE CONFIG_FIRQ_STACK_SIZE
#endif
#else
#define INIT_STACK _interrupt_stack
#define INIT_STACK_SIZE CONFIG_ISR_STACK_SIZE
#endif
GTEXT(__reset)
GTEXT(__start)
@@ -45,114 +69,9 @@ GTEXT(__start)
SECTION_FUNC(TEXT,__reset)
SECTION_FUNC(TEXT,__start)
/* lock interrupts: will get unlocked when switch to main task
* also make sure the processor in the correct status
*/
mov_s r0, 0
kflag r0
#ifdef CONFIG_ARC_SECURE_FIRMWARE
sflag r0
#endif
#if defined(CONFIG_BOOT_TIME_MEASUREMENT) && defined(CONFIG_ARCV2_TIMER)
/*
* ARCV2 timer (timer0) is a free run timer, let it start to count
* here.
*/
mov_s r0, 0xffffffff
sr r0, [_ARC_V2_TMR0_LIMIT]
mov_s r0, 0
sr r0, [_ARC_V2_TMR0_COUNT]
#endif
/* interrupt related init */
#ifndef CONFIG_ARC_NORMAL_FIRMWARE
/* IRQ_ACT and IRQ_CTRL should be initialized and set in secure mode */
sr r0, [_ARC_V2_AUX_IRQ_ACT]
sr r0, [_ARC_V2_AUX_IRQ_CTRL]
#endif
sr r0, [_ARC_V2_AUX_IRQ_HINT]
/* set the vector table base early,
* so that exception vectors can be handled.
*/
mov_s r0, _VectorTable
#ifdef CONFIG_ARC_SECURE_FIRMWARE
sr r0, [_ARC_V2_IRQ_VECT_BASE_S]
#else
sr r0, [_ARC_V2_IRQ_VECT_BASE]
#endif
#if defined(CONFIG_USERSPACE)
lr r0, [_ARC_V2_STATUS32]
bset r0, r0, _ARC_V2_STATUS32_US_BIT
kflag r0
#endif
#ifdef CONFIG_ARC_USE_UNALIGNED_MEM_ACCESS
lr r0, [_ARC_V2_STATUS32]
bset r0, r0, _ARC_V2_STATUS32_AD_BIT
kflag r0
#endif
mov_s r1, 1
invalidate_and_disable_icache:
lr r0, [_ARC_V2_I_CACHE_BUILD]
and.f r0, r0, 0xff
bz.nd invalidate_dcache
mov_s r2, 0
sr r2, [_ARC_V2_IC_IVIC]
/* writing to IC_IVIC needs 3 NOPs */
nop_s
nop_s
nop_s
sr r1, [_ARC_V2_IC_CTRL]
invalidate_dcache:
lr r3, [_ARC_V2_D_CACHE_BUILD]
and.f r3, r3, 0xff
bz.nd done_cache_invalidate
sr r1, [_ARC_V2_DC_IVDC]
done_cache_invalidate:
#if defined(CONFIG_SYS_POWER_DEEP_SLEEP_STATES) && \
!defined(CONFIG_BOOTLOADER_CONTEXT_RESTORE)
jl @_sys_resume_from_deep_sleep
#endif
#if defined(CONFIG_SMP) || CONFIG_MP_NUM_CPUS > 1
_get_cpu_id r0
breq r0, 0, _master_core_startup
/*
* Non-masters wait for master core (core 0) to boot enough
*/
_slave_core_wait:
#if CONFIG_MP_NUM_CPUS == 1
kflag 1
#endif
ld r1, [arc_cpu_wake_flag]
brne r0, r1, _slave_core_wait
ld sp, [arc_cpu_sp]
/* signal master core that slave core runs */
st 0, [arc_cpu_wake_flag]
#if defined(CONFIG_ARC_FIRQ_STACK)
push r0
jl z_arc_firq_stack_set
pop r0
#endif
j z_arc_slave_start
_master_core_startup:
#endif
/* lock interrupts: will get unlocked when switch to main task */
clri
#ifdef CONFIG_INIT_STACKS
/*
@@ -160,21 +79,24 @@ _master_core_startup:
* FIRQ stack when CONFIG_INIT_STACKS is enabled before switching to
* one of them for the rest of the early boot
*/
mov_s sp, z_main_stack
mov sp, _main_stack
add sp, sp, CONFIG_MAIN_STACK_SIZE
mov_s r0, z_interrupt_stacks
mov_s r0, _interrupt_stack
mov_s r1, 0xaa
mov_s r2, CONFIG_ISR_STACK_SIZE
jl memset
#endif /* CONFIG_INIT_STACKS */
mov_s sp, INIT_STACK
add sp, sp, INIT_STACK_SIZE
#if defined(CONFIG_ARC_FIRQ_STACK)
jl z_arc_firq_stack_set
#if CONFIG_RGF_NUM_BANKS != 1
mov_s r0, _firq_stack
mov_s r1, 0xaa
mov_s r2, CONFIG_FIRQ_STACK_SIZE
jl memset
#endif
#endif /* CONFIG_INIT_STACKS */
mov sp, INIT_STACK
add sp, sp, INIT_STACK_SIZE
j @_PrepC

View File

@@ -1,12 +0,0 @@
#
# Copyright (c) 2018 Synopsys, Inc. All rights reserved.
#
# SPDX-License-Identifier: Apache-2.0
zephyr_library()
zephyr_library_sources(
arc_sjli.c
arc_secure.S
secure_sys_services.c
)

View File

@@ -1,122 +0,0 @@
/*
* Copyright (c) 2018 Synopsys.
*
* SPDX-License-Identifier: Apache-2.0
*/
#include <toolchain.h>
#include <linker/sections.h>
#include <arch/cpu.h>
.macro clear_scratch_regs
mov r1, 0
mov r2, 0
mov r3, 0
mov r4, 0
mov r5, 0
mov r6, 0
mov r7, 0
mov r8, 0
mov r9, 0
mov r10, 0
mov r11, 0
mov r12, 0
.endm
.macro clear_callee_regs
mov r25, 0
mov r24, 0
mov r23, 0
mov r22, 0
mov r21, 0
mov r20, 0
mov r19, 0
mov r18, 0
mov r17, 0
mov r16, 0
mov r15, 0
mov r14, 0
mov r13, 0
.endm
GTEXT(arc_go_to_normal)
GTEXT(_arc_do_secure_call)
GDATA(arc_s_call_table)
SECTION_FUNC(TEXT, _arc_do_secure_call)
/* r0-r5: arg1-arg6, r6 is call id */
/* the call id should be checked */
/* disable normal interrupt happened when processor in secure mode ? */
/* seti (0x30 | (ARC_N_IRQ_START_LEVEL-1)) */
breq r6, ARC_S_CALL_CLRI, _s_clri
breq r6, ARC_S_CALL_SETI, _s_seti
push_s blink
mov blink, arc_s_call_table
ld.as r6, [blink, r6]
jl [r6]
/*
* no need to clear callee regs, as they will be saved and restored
* automatically
*/
clear_scratch_regs
mov r29, 0
mov r30, 0
_arc_do_secure_call_exit:
pop_s blink
j [blink]
/* enable normal interrupt */
/*
* j.d [blink]
* seti (0x30 | (CONFIG_NUM_IRQ_PRIO_LEVELS - 1))
*/
_s_clri:
lr r0, [_ARC_V2_STATUS32]
and r0, r0, 0x1e
asr r0, r0
or r0, r0, 0x30
mov r6, (0x30 | (ARC_N_IRQ_START_LEVEL-1))
j.d [blink]
seti r6
_s_seti:
btst r0, 4
jnz __seti_0
mov r0, (CONFIG_NUM_IRQ_PRIO_LEVELS - 1)
lr r6, [_ARC_V2_STATUS32]
and r6, r6, 0x1e
asr r6, r6
cmp r0, r6
mov.hs r0, r6
__seti_0:
and r0, r0, 0xf
brhs r0, ARC_N_IRQ_START_LEVEL, __seti_1
mov r0, ARC_N_IRQ_START_LEVEL
__seti_1:
or r0, r0, 0x30
j.d [blink]
seti r0
SECTION_FUNC(TEXT, arc_go_to_normal)
clear_callee_regs
clear_scratch_regs
mov fp, 0
mov r29, 0
mov r30, 0
mov blink, 0
jl [r0]
/* should not come here */
kflag 1

View File

@@ -1,68 +0,0 @@
/*
* Copyright (c) 2018 Synopsys, Inc. All rights reserved.
*
* SPDX-License-Identifier: Apache-2.0
*/
#include <device.h>
#include <kernel.h>
#include <errno.h>
#include <zephyr/types.h>
#include <init.h>
#include <toolchain.h>
#include <arch/arc/v2/secureshield/arc_secure.h>
static void _default_sjli_entry(void);
/*
* sjli vector table must be in instruction space
* \todo: how to let user to install customized sjli entry easily, e.g.
* through macros or with the help of compiler?
*/
const static u32_t _sjli_vector_table[CONFIG_SJLI_TABLE_SIZE] = {
[0] = (u32_t)_arc_do_secure_call,
[1 ... (CONFIG_SJLI_TABLE_SIZE - 1)] = (u32_t)_default_sjli_entry,
};
/*
* @brief default entry of sjli call
*
*/
static void _default_sjli_entry(void)
{
printk("default sjli entry\n");
}
/*
* @brief initializaiton of sjli related functions
*
*/
static void sjli_table_init(void)
{
/* install SJLI table */
z_arc_v2_aux_reg_write(_ARC_V2_NSC_TABLE_BASE, _sjli_vector_table);
z_arc_v2_aux_reg_write(_ARC_V2_NSC_TABLE_TOP,
(_sjli_vector_table + CONFIG_SJLI_TABLE_SIZE));
}
/*
* @brief initializaiton of secureshield related functions.
*/
static int arc_secureshield_init(struct device *arg)
{
sjli_table_init();
/* set nic bit to enable seti/clri and
* sleep/wevt in normal mode.
* If not set, direct call of seti/clri etc. will raise exception.
* Then, these seti/clri instructions should be replaced with secure
* secure services (sjli call)
*
*/
__asm__ volatile("sflag 0x20");
return 0;
}
SYS_INIT(arc_secureshield_init, PRE_KERNEL_1,
CONFIG_KERNEL_INIT_PRIORITY_DEFAULT);

View File

@@ -1,89 +0,0 @@
/*
* Copyright (c) 2018 Synopsys, Inc. All rights reserved.
*
* SPDX-License-Identifier: Apache-2.0
*/
#include <errno.h>
#include <kernel.h>
#include <arch/cpu.h>
#include <zephyr/types.h>
#include <soc.h>
#include <toolchain.h>
#include <arch/arc/v2/secureshield/arc_secure.h>
#define IRQ_PRIO_MASK (0xffff << ARC_N_IRQ_START_LEVEL)
/*
* @brief read secure auxiliary regs on behalf of normal mode
*
* @param aux_reg address of aux reg
*
* Some aux regs require secure privilege, this function implements
* an secure service to access secure aux regs. Check should be done
* to decide whether the access is valid.
*/
static s32_t arc_s_aux_read(u32_t aux_reg)
{
return -1;
}
/*
* @brief write secure auxiliary regs on behalf of normal mode
*
* @param aux_reg address of aux reg
* @param val, the val to write
*
* Some aux regs require secure privilege, this function implements
* an secure service to access secure aux regs. Check should be done
* to decide whether the access is valid.
*/
static s32_t arc_s_aux_write(u32_t aux_reg, u32_t val)
{
if (aux_reg == _ARC_V2_AUX_IRQ_ACT) {
/* 0 -> CONFIG_NUM_IRQ_PRIO_LEVELS allocated to secure world
* left prio levels allocated to normal world
*/
val &= IRQ_PRIO_MASK;
z_arc_v2_aux_reg_write(_ARC_V2_AUX_IRQ_ACT, val |
(z_arc_v2_aux_reg_read(_ARC_V2_AUX_IRQ_ACT) &
(~IRQ_PRIO_MASK)));
return 0;
}
return -1;
}
/*
* @brief allocate interrupt for normal world
*
* @param intno, the interrupt to be allocated to normal world
*
* By default, most interrupts are configured to be secure in initialization.
* If normal world wants to use an interrupt, through this secure service to
* apply one. Necessary check should be done to decide whether the apply is
* valid
*/
static s32_t arc_s_irq_alloc(u32_t intno)
{
z_arc_v2_irq_uinit_secure_set(intno, 0);
return 0;
}
/*
* \todo, to access MPU from normal mode, secure mpu service should be
* created. In the secure mpu service, the parameters should be checked
* (e.g., not overwrite the mpu regions for secure world)that operations
* are valid
*/
/*
* \todo, how to add secure service easily
*/
const _arc_s_call_handler_t arc_s_call_table[ARC_S_CALL_LIMIT] = {
[ARC_S_CALL_AUX_READ] = (_arc_s_call_handler_t)arc_s_aux_read,
[ARC_S_CALL_AUX_WRITE] = (_arc_s_call_handler_t)arc_s_aux_write,
[ARC_S_CALL_IRQ_ALLOC] = (_arc_s_call_handler_t)arc_s_irq_alloc,
};

View File

@@ -0,0 +1,64 @@
/* sw_isr_table.S - ISR table for static ISR declarations for ARC */
/*
* Copyright (c) 2015 Intel Corporation
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
#define _ASMLANGUAGE
#include <toolchain.h>
#include <sections.h>
#include <arch/cpu.h>
/*
* enable preprocessor features, such
* as %expr - evaluate the expression and use it as a string
*/
.altmacro
/*
* Define an ISR table entry
* Define symbol as weak and give the section .gnu.linkonce
* prefix. This allows linker overload the symbol and the
* whole section by the one defined by a device driver
*/
.macro _isr_table_entry_declare index
WDATA(_isr_irq\index)
.section .gnu.linkonce.isr_irq\index
_isr_irq\index: .word 0xABAD1DEA, _irq_spurious
.endm
/*
* Declare the ISR table
*/
.macro _isr_table_declare from, to
counter = \from
.rept (\to - \from)
_isr_table_entry_declare %counter
counter = counter + 1
.endr
.endm
GTEXT(_irq_spurious)
GDATA(_sw_isr_table)
.section .isr_irq16
.align
_sw_isr_table:
/*In ARC architecture, IRQ 0-15 are reserved for the system and are not
assignable by the user, for that reason the isr table linker section
start at IRQ 16*/
_isr_table_declare 16 CONFIG_NUM_IRQS

198
arch/arc/core/swap.S Normal file
View File

@@ -0,0 +1,198 @@
/*
* Copyright (c) 2014-2015 Wind River Systems, Inc.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
* @file
* @brief Thread context switching
*
* This module implements the routines necessary for thread context switching
* on ARCv2 CPUs.
*
* See isr_wrapper.S for details.
*/
#define _ASMLANGUAGE
#include <kernel_structs.h>
#include <offsets_short.h>
#include <toolchain.h>
#include <arch/cpu.h>
#include <v2/irq.h>
#include "swap_macros.h"
GTEXT(_Swap)
GTEXT(_get_next_ready_thread)
GDATA(_k_neg_eagain)
GDATA(_kernel)
/**
*
* @brief Initiate a cooperative context switch
*
* The _Swap() routine is invoked by various nanokernel services to effect
* a cooperative context switch. Prior to invoking _Swap(), the caller
* disables interrupts via nanoCpuIntLock() and the return 'key' is passed as a
* parameter to _Swap(). The key is in fact the value stored in the register
* operand of a CLRI instruction.
*
* It stores the intlock key parameter into current->intlock_key.
* Given that _Swap() is called to effect a cooperative context switch,
* the caller-saved integer registers are saved on the stack by the function
* call preamble to _Swap(). This creates a custom stack frame that will be
* popped when returning from _Swap(), but is not suitable for handling a return
* from an exception. Thus, the fact that the thread is pending because of a
* cooperative call to _Swap() has to be recorded via the _CAUSE_COOP code in
* the relinquish_cause of the thread's k_thread structure. The
* _IrqExit()/_FirqExit() code will take care of doing the right thing to
* restore the thread status.
*
* When _Swap() is invoked, we know the decision to perform a context switch or
* not has already been taken and a context switch must happen.
*
* @return may contain a return value setup by a call to fiberRtnValueSet()
*
* C function prototype:
*
* unsigned int _Swap (unsigned int key);
*
*/
SECTION_FUNC(TEXT, _Swap)
/* interrupts are locked, interrupt key is in r0 */
mov r1, _kernel
ld_s r2, [r1, _kernel_offset_to_current]
/* save intlock key */
st_s r0, [r2, _thread_offset_to_intlock_key]
st _CAUSE_COOP, [r2, _thread_offset_to_relinquish_cause]
/*
* Carve space for the return value. Setting it to a defafult of
* -EAGAIN eliminates the need for the timeout code to set it.
* If another value is ever needed, it can be modified with
* fiberRtnValueSet().
*/
ld r3, [_k_neg_eagain]
st_s r3, [r2, _thread_offset_to_return_value]
/*
* Save status32 and blink on the stack before the callee-saved registers.
* This is the same layout as the start of an IRQ stack frame.
*/
lr r3, [_ARC_V2_STATUS32]
push_s r3
#ifdef CONFIG_ARC_STACK_CHECKING
/* disable stack checking during swap */
bclr r3, r3, _ARC_V2_STATUS32_SC_BIT
kflag r3
#endif
push_s blink
_save_callee_saved_regs
/* find out incoming thread (fiber or task) */
/*
* Save needed registers to callee saved ones. It is faster than
* pushing them to stack. It is possible to do since program has
* just saved them and the calling routine will save them in turn
* if it uses them.
*/
mov_s r13, blink
mov_s r14, r0
mov_s r15, r1
jl _get_next_ready_thread
mov_s r2, r0
mov_s r1, r15
mov_s r0, r14
mov_s blink, r13
/* entering here, r2 contains the new current thread */
#ifdef CONFIG_ARC_STACK_CHECKING
/* Use stack top and down registers from restored context */
add r3, r2, _K_THREAD_NO_FLOAT_SIZEOF
sr r3, [_ARC_V2_KSTACK_TOP]
ld_s r3, [r2, _thread_offset_to_stack_top]
sr r3, [_ARC_V2_KSTACK_BASE]
#endif
/* XXX - can be moved to delay slot of _CAUSE_RIRQ ? */
st_s r2, [r1, _kernel_offset_to_current]
_load_callee_saved_regs
ld_s r3, [r2, _thread_offset_to_relinquish_cause]
breq r3, _CAUSE_RIRQ, _swap_return_from_rirq
nop
breq r3, _CAUSE_FIRQ, _swap_return_from_firq
nop
/* fall through to _swap_return_from_coop */
.balign 4
_swap_return_from_coop:
ld_s r1, [r2, _thread_offset_to_intlock_key]
st 0, [r2, _thread_offset_to_intlock_key]
ld_s r0, [r2, _thread_offset_to_return_value]
lr ilink, [_ARC_V2_STATUS32]
bbit1 ilink, _ARC_V2_STATUS32_AE_BIT, _return_from_exc
pop_s blink /* pc into blink */
pop_s r3 /* status32 into r3 */
kflag r3 /* write status32 */
j_s.d [blink] /* always execute delay slot */
seti r1 /* delay slot */
.balign 4
_swap_return_from_rirq:
_swap_return_from_firq:
lr r3, [_ARC_V2_STATUS32]
bbit1 r3, _ARC_V2_STATUS32_AE_BIT, _return_from_exc_irq
/* pretend interrupt happened to use rtie instruction */
lr r3, [_ARC_V2_AUX_IRQ_ACT]
brne r3,0,_swap_already_in_irq
or r3,r3,(1<<(CONFIG_NUM_IRQ_PRIO_LEVELS-1)) /* use lowest */
sr r3, [_ARC_V2_AUX_IRQ_ACT]
_swap_already_in_irq:
rtie
.balign 4
_return_from_exc_irq:
_pop_irq_stack_frame
sub_s sp, sp, 8
_return_from_exc:
/* put the return address to eret */
ld ilink, [sp] /* pc into ilink */
sr ilink, [_ARC_V2_ERET]
/* put status32 into estatus */
ld ilink, [sp, 4] /* status32 into ilink */
sr ilink, [_ARC_V2_ERSTATUS]
add_s sp, sp, 8
rtie

171
arch/arc/core/swap_macros.h Normal file
View File

@@ -0,0 +1,171 @@
/* swap_macros.h - helper macros for context switch */
/*
* Copyright (c) 2014 Wind River Systems, Inc.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
#ifndef _SWAP_MACROS__H_
#define _SWAP_MACROS__H_
#include <kernel_structs.h>
#include <offsets_short.h>
#include <toolchain.h>
#include <arch/cpu.h>
#ifdef __cplusplus
extern "C" {
#endif
#ifdef _ASMLANGUAGE
/* entering this macro, current is in r2 */
.macro _save_callee_saved_regs
sub_s sp, sp, ___callee_saved_stack_t_SIZEOF
/* save regs on stack */
st_s r13, [sp, ___callee_saved_stack_t_r13_OFFSET]
st_s r14, [sp, ___callee_saved_stack_t_r14_OFFSET]
st_s r15, [sp, ___callee_saved_stack_t_r15_OFFSET]
st r16, [sp, ___callee_saved_stack_t_r16_OFFSET]
st r17, [sp, ___callee_saved_stack_t_r17_OFFSET]
st r18, [sp, ___callee_saved_stack_t_r18_OFFSET]
st r19, [sp, ___callee_saved_stack_t_r19_OFFSET]
st r20, [sp, ___callee_saved_stack_t_r20_OFFSET]
st r21, [sp, ___callee_saved_stack_t_r21_OFFSET]
st r22, [sp, ___callee_saved_stack_t_r22_OFFSET]
st r23, [sp, ___callee_saved_stack_t_r23_OFFSET]
st r24, [sp, ___callee_saved_stack_t_r24_OFFSET]
st r25, [sp, ___callee_saved_stack_t_r25_OFFSET]
st r26, [sp, ___callee_saved_stack_t_r26_OFFSET]
st fp, [sp, ___callee_saved_stack_t_fp_OFFSET]
st r30, [sp, ___callee_saved_stack_t_r30_OFFSET]
/* save stack pointer in struct tcs */
st sp, [r2, _thread_offset_to_sp]
.endm
/* entering this macro, current is in r2 */
.macro _load_callee_saved_regs
/* restore stack pointer from struct tcs */
ld sp, [r2, _thread_offset_to_sp]
ld_s r13, [sp, ___callee_saved_stack_t_r13_OFFSET]
ld_s r14, [sp, ___callee_saved_stack_t_r14_OFFSET]
ld_s r15, [sp, ___callee_saved_stack_t_r15_OFFSET]
ld r16, [sp, ___callee_saved_stack_t_r16_OFFSET]
ld r17, [sp, ___callee_saved_stack_t_r17_OFFSET]
ld r18, [sp, ___callee_saved_stack_t_r18_OFFSET]
ld r19, [sp, ___callee_saved_stack_t_r19_OFFSET]
ld r20, [sp, ___callee_saved_stack_t_r20_OFFSET]
ld r21, [sp, ___callee_saved_stack_t_r21_OFFSET]
ld r22, [sp, ___callee_saved_stack_t_r22_OFFSET]
ld r23, [sp, ___callee_saved_stack_t_r23_OFFSET]
ld r24, [sp, ___callee_saved_stack_t_r24_OFFSET]
ld r25, [sp, ___callee_saved_stack_t_r25_OFFSET]
ld r26, [sp, ___callee_saved_stack_t_r26_OFFSET]
ld fp, [sp, ___callee_saved_stack_t_fp_OFFSET]
ld r30, [sp, ___callee_saved_stack_t_r30_OFFSET]
add_s sp, sp, ___callee_saved_stack_t_SIZEOF
.endm
/*
* Must be called with interrupts locked or in P0.
* Upon exit, sp will be pointing to the stack frame.
*/
.macro _create_irq_stack_frame
sub_s sp, sp, ___isf_t_SIZEOF
st blink, [sp, ___isf_t_blink_OFFSET]
/* store these right away so we can use them if needed */
st_s r13, [sp, ___isf_t_r13_OFFSET]
st_s r12, [sp, ___isf_t_r12_OFFSET]
st r11, [sp, ___isf_t_r11_OFFSET]
st r10, [sp, ___isf_t_r10_OFFSET]
st r9, [sp, ___isf_t_r9_OFFSET]
st r8, [sp, ___isf_t_r8_OFFSET]
st r7, [sp, ___isf_t_r7_OFFSET]
st r6, [sp, ___isf_t_r6_OFFSET]
st r5, [sp, ___isf_t_r5_OFFSET]
st r4, [sp, ___isf_t_r4_OFFSET]
st_s r3, [sp, ___isf_t_r3_OFFSET]
st_s r2, [sp, ___isf_t_r2_OFFSET]
st_s r1, [sp, ___isf_t_r1_OFFSET]
st_s r0, [sp, ___isf_t_r0_OFFSET]
mov r0, lp_count
st_s r0, [sp, ___isf_t_lp_count_OFFSET]
lr r1, [_ARC_V2_LP_START]
lr r0, [_ARC_V2_LP_END]
st_s r1, [sp, ___isf_t_lp_start_OFFSET]
st_s r0, [sp, ___isf_t_lp_end_OFFSET]
.endm
/*
* Must be called with interrupts locked or in P0.
* sp must be pointing the to stack frame.
*/
.macro _pop_irq_stack_frame
ld blink, [sp, ___isf_t_blink_OFFSET]
ld_s r0, [sp, ___isf_t_lp_count_OFFSET]
mov lp_count, r0
ld_s r1, [sp, ___isf_t_lp_start_OFFSET]
ld_s r0, [sp, ___isf_t_lp_end_OFFSET]
sr r1, [_ARC_V2_LP_START]
sr r0, [_ARC_V2_LP_END]
ld_s r13, [sp, ___isf_t_r13_OFFSET]
ld_s r12, [sp, ___isf_t_r12_OFFSET]
ld r11, [sp, ___isf_t_r11_OFFSET]
ld r10, [sp, ___isf_t_r10_OFFSET]
ld r9, [sp, ___isf_t_r9_OFFSET]
ld r8, [sp, ___isf_t_r8_OFFSET]
ld r7, [sp, ___isf_t_r7_OFFSET]
ld r6, [sp, ___isf_t_r6_OFFSET]
ld r5, [sp, ___isf_t_r5_OFFSET]
ld r4, [sp, ___isf_t_r4_OFFSET]
ld_s r3, [sp, ___isf_t_r3_OFFSET]
ld_s r2, [sp, ___isf_t_r2_OFFSET]
ld_s r1, [sp, ___isf_t_r1_OFFSET]
ld_s r0, [sp, ___isf_t_r0_OFFSET]
/*
* All gprs have been reloaded, the only one that is still usable is
* ilink.
*
* The pc and status32 values will still be on the stack. We cannot
* pop them yet because the callers of _pop_irq_stack_frame must reload
* status32 differently depending on the execution context they are running
* in (_Swap(), firq or exception).
*/
add_s sp, sp, ___isf_t_SIZEOF
.endm
#endif /* _ASMLANGUAGE */
#ifdef __cplusplus
}
#endif
#endif /* _SWAP_MACROS__H_ */

View File

@@ -1,164 +0,0 @@
/*
* Copyright (c) 2019 Synopsys.
*
* SPDX-License-Identifier: Apache-2.0
*/
/**
* @file
* @brief Thread context switching
*
* This module implements the routines necessary for thread context switching
* on ARCv2 CPUs.
*
* See isr_wrapper.S for details.
*/
#include <kernel_structs.h>
#include <offsets_short.h>
#include <toolchain.h>
#include <linker/sections.h>
#include <arch/cpu.h>
#include <v2/irq.h>
#include <swap_macros.h>
GTEXT(arch_switch)
/**
*
* @brief Initiate a cooperative context switch
*
* The arch_switch routine is invoked by various kernel services to effect
* a cooperative context switch. Prior to invoking arch_switch, the caller
* disables interrupts via irq_lock()
* Given that arch_switch() is called to effect a cooperative context switch,
* the caller-saved integer registers are saved on the stack by the function
* call preamble to arch_switch. This creates a custom stack frame that will
* be popped when returning from arch_switch, but is not suitable for handling
* a return from an exception. Thus, the fact that the thread is pending because
* of a cooperative call to arch_switch() has to be recorded via the
* _CAUSE_COOP code in the relinquish_cause of the thread's k_thread structure.
* The _rirq_exit()/_firq_exit() code will take care of doing the right thing
* to restore the thread status.
*
* When arch_switch() is invoked, we know the decision to perform a context
* switch or not has already been taken and a context switch must happen.
*
*
* C function prototype:
*
* void arch_switch(void *switch_to, void **switched_from);
*
*/
SECTION_FUNC(TEXT, arch_switch)
#ifdef CONFIG_EXECUTION_BENCHMARKING
push_s r0
push_s r1
push_s blink
bl read_timer_start_of_swap
pop_s blink
pop_s r1
pop_s r0
#endif
/*
* r0 = new_thread->switch_handle = switch_to thread,
* r1 = &old_thread->switch_handle
* get old_thread from r1
*/
sub r2, r1, ___thread_t_switch_handle_OFFSET
st _CAUSE_COOP, [r2, _thread_offset_to_relinquish_cause]
/*
* Save status32 and blink on the stack before the callee-saved registers.
* This is the same layout as the start of an IRQ stack frame.
*/
lr r3, [_ARC_V2_STATUS32]
push_s r3
#ifdef CONFIG_ARC_HAS_SECURE
#ifdef CONFIG_ARC_SECURE_FIRMWARE
lr r3, [_ARC_V2_SEC_STAT]
#else
mov_s r3, 0
#endif
push_s r3
#endif
push_s blink
_store_old_thread_callee_regs
/* disable stack checking here, as sp will be changed to target
* thread'sp
*/
_disable_stack_checking r3
mov_s r2, r0
_load_new_thread_callee_regs
breq r3, _CAUSE_RIRQ, _switch_return_from_rirq
nop_s
breq r3, _CAUSE_FIRQ, _switch_return_from_firq
nop_s
/* fall through to _switch_return_from_coop */
.balign 4
_switch_return_from_coop:
pop_s blink /* pc into blink */
#ifdef CONFIG_ARC_HAS_SECURE
pop_s r3 /* pop SEC_STAT */
#ifdef CONFIG_ARC_SECURE_FIRMWARE
sflag r3
#endif
#endif
pop_s r3 /* status32 into r3 */
kflag r3 /* write status32 */
#ifdef CONFIG_EXECUTION_BENCHMARKING
push_s blink
bl read_timer_end_of_swap
pop_s blink
#endif /* CONFIG_EXECUTION_BENCHMARKING */
j_s [blink]
.balign 4
_switch_return_from_rirq:
_switch_return_from_firq:
_set_misc_regs_irq_switch_from_irq
/* use lowest interrupt priority to simulate
* a interrupt return to load left regs of new
* thread
*/
lr r3, [_ARC_V2_AUX_IRQ_ACT]
#ifdef CONFIG_ARC_SECURE_FIRMWARE
or r3, r3, (1 << (ARC_N_IRQ_START_LEVEL - 1))
#else
or r3, r3, (1 << (CONFIG_NUM_IRQ_PRIO_LEVELS - 1))
#endif
#ifdef CONFIG_ARC_NORMAL_FIRMWARE
mov_s r0, _ARC_V2_AUX_IRQ_ACT
mov_s r1, r3
mov_s r6, ARC_S_CALL_AUX_WRITE
sjli SJLI_CALL_ARC_SECURE
#else
sr r3, [_ARC_V2_AUX_IRQ_ACT]
#endif
rtie

View File

@@ -0,0 +1,66 @@
/*
* Copyright (c) 2014 Wind River Systems, Inc.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
* @file
* @brief ARCv2 system fatal error handler
*
* This module provides the _SysFatalErrorHandler() routine for ARCv2 BSPs.
*/
#include <nanokernel.h>
#include <toolchain.h>
#include <sections.h>
#include <kernel_structs.h>
#include <misc/printk.h>
/**
*
* @brief Fatal error handler
*
* This routine implements the corrective action to be taken when the system
* detects a fatal error.
*
* This sample implementation attempts to abort the current thread and allow
* the system to continue executing, which may permit the system to continue
* functioning with degraded capabilities.
*
* System designers may wish to enhance or substitute this sample
* implementation to take other actions, such as logging error (or debug)
* information to a persistent repository and/or rebooting the system.
*
* @param reason the fatal error reason
* @param pEsf pointer to exception stack frame
*
* @return N/A
*/
FUNC_NORETURN void _SysFatalErrorHandler(unsigned int reason,
const NANO_ESF *pEsf)
{
ARG_UNUSED(reason);
ARG_UNUSED(pEsf);
if (k_is_in_isr() || _is_thread_essential()) {
printk("Fatal fault in %s! Spinning...\n",
k_is_in_isr() ? "ISR" : "essential thread");
for (;;)
; /* spin forever */
}
printk("Fatal fault in thread! Aborting.\n");
k_thread_abort(_current);
CODE_UNREACHABLE;
}

View File

@@ -1,38 +1,64 @@
/*
* Copyright (c) 2014 Wind River Systems, Inc.
*
* SPDX-License-Identifier: Apache-2.0
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
* @file
* @brief New thread creation for ARCv2
*
* Core thread related primitives for the ARCv2 processor architecture.
* Core nanokernel fiber related primitives for the ARCv2 processor
* architecture.
*/
#include <kernel.h>
#include <ksched.h>
#include <nanokernel.h>
#include <toolchain.h>
#include <kernel_structs.h>
#include <offsets_short.h>
#include <wait_q.h>
#ifdef CONFIG_USERSPACE
#include <arch/arc/v2/mpu/arc_core_mpu.h>
#endif
#ifdef CONFIG_INIT_STACKS
#include <string.h>
#endif /* CONFIG_INIT_STACKS */
/* initial stack frame */
struct init_stack_frame {
u32_t pc;
#ifdef CONFIG_ARC_HAS_SECURE
u32_t sec_stat;
#endif
u32_t status32;
u32_t r3;
u32_t r2;
u32_t r1;
u32_t r0;
uint32_t pc;
uint32_t status32;
uint32_t r3;
uint32_t r2;
uint32_t r1;
uint32_t r0;
};
#if defined(CONFIG_THREAD_MONITOR)
/*
* Add a thread to the kernel's list of active threads.
*/
static ALWAYS_INLINE void thread_monitor_init(struct k_thread *thread)
{
unsigned int key;
key = irq_lock();
thread->next_thread = _kernel.threads;
_kernel.threads = thread;
irq_unlock(key);
}
#else
#define thread_monitor_init(thread) \
do {/* do nothing */ \
} while ((0))
#endif /* CONFIG_THREAD_MONITOR */
/*
* @brief Initialize a new thread from its stack space
*
@@ -42,7 +68,7 @@ struct init_stack_frame {
* needed anymore.
*
* The initial context is a basic stack frame that contains arguments for
* z_thread_entry() return address, that points at z_thread_entry()
* _thread_entry() return address, that points at _thread_entry()
* and status register.
*
* <options> is currently unused.
@@ -58,241 +84,78 @@ struct init_stack_frame {
*
* @return N/A
*/
void arch_new_thread(struct k_thread *thread, k_thread_stack_t *stack,
size_t stackSize, k_thread_entry_t pEntry,
void *parameter1, void *parameter2, void *parameter3,
int priority, unsigned int options)
void _new_thread(char *pStackMem, size_t stackSize,
_thread_entry_t pEntry,
void *parameter1, void *parameter2, void *parameter3,
int priority, unsigned options)
{
char *pStackMem = Z_THREAD_STACK_BUFFER(stack);
_ASSERT_VALID_PRIO(priority, pEntry);
char *stackEnd;
char *priv_stack_end;
char *stackEnd = pStackMem + stackSize;
struct init_stack_frame *pInitCtx;
#ifdef CONFIG_USERSPACE
struct k_thread *thread = (struct k_thread *) pStackMem;
size_t stackAdjSize;
size_t offset = 0;
stackAdjSize = Z_ARC_MPU_SIZE_ALIGN(stackSize);
stackEnd = pStackMem + stackAdjSize;
#ifdef CONFIG_STACK_POINTER_RANDOM
offset = stackAdjSize - stackSize;
#ifdef CONFIG_INIT_STACKS
memset(pStackMem, 0xaa, stackSize);
#endif
if (options & K_USER) {
#ifdef CONFIG_GEN_PRIV_STACKS
thread->arch.priv_stack_start =
(u32_t)z_priv_stack_find(thread->stack_obj);
/* carve the thread entry struct from the "base" of the stack */
pInitCtx = (struct init_stack_frame *)(STACK_ROUND_DOWN(stackEnd) -
sizeof(struct init_stack_frame));
pInitCtx->pc = ((uint32_t)_thread_entry_wrapper);
pInitCtx->r0 = (uint32_t)pEntry;
pInitCtx->r1 = (uint32_t)parameter1;
pInitCtx->r2 = (uint32_t)parameter2;
pInitCtx->r3 = (uint32_t)parameter3;
/*
* For now set the interrupt priority to 15
* we can leave interrupt enable flag set to 0 as
* seti instruction in the end of the _Swap() will
* enable the interrupts based on intlock_key
* value.
*/
#ifdef CONFIG_ARC_STACK_CHECKING
pInitCtx->status32 = _ARC_V2_STATUS32_SC | _ARC_V2_STATUS32_E(_ARC_V2_DEF_IRQ_LEVEL);
thread->arch.stack_top = (uint32_t) stackEnd;
#else
thread->arch.priv_stack_start =
(u32_t)(stackEnd + STACK_GUARD_SIZE);
pInitCtx->status32 = _ARC_V2_STATUS32_E(_ARC_V2_DEF_IRQ_LEVEL);
#endif
priv_stack_end = (char *)Z_STACK_PTR_ALIGN(
thread->arch.priv_stack_start +
CONFIG_PRIVILEGED_STACK_SIZE);
_init_thread_base(&thread->base, priority, K_PRESTART, options);
/* reserve 4 bytes for the start of user sp */
priv_stack_end -= 4;
(*(u32_t *)priv_stack_end) = Z_STACK_PTR_ALIGN(
(u32_t)stackEnd - offset);
/* static threads overwrite them afterwards with real values */
thread->init_data = NULL;
thread->fn_abort = NULL;
#ifdef CONFIG_THREAD_USERSPACE_LOCAL_DATA
/* reserve stack space for the userspace local data struct */
thread->userspace_local_data =
(struct _thread_userspace_local_data *)
Z_STACK_PTR_ALIGN(stackEnd -
sizeof(*thread->userspace_local_data) - offset);
/* update the start of user sp */
(*(u32_t *)priv_stack_end) =
(u32_t) thread->userspace_local_data;
#ifdef CONFIG_THREAD_CUSTOM_DATA
/* Initialize custom data field (value is opaque to kernel) */
thread->custom_data = NULL;
#endif
} else {
pStackMem += STACK_GUARD_SIZE;
stackEnd += STACK_GUARD_SIZE;
thread->arch.priv_stack_start = 0;
#ifdef CONFIG_THREAD_USERSPACE_LOCAL_DATA
/* reserve stack space for the userspace local data struct */
priv_stack_end = (char *)Z_STACK_PTR_ALIGN(stackEnd
- sizeof(*thread->userspace_local_data) - offset);
thread->userspace_local_data =
(struct _thread_userspace_local_data *)priv_stack_end;
#else
priv_stack_end = (char *)Z_STACK_PTR_ALIGN(stackEnd - offset);
#ifdef CONFIG_THREAD_MONITOR
/*
* In debug mode thread->entry give direct access to the thread entry
* and the corresponding parameters.
*/
thread->entry = (struct __thread_entry *)(pInitCtx);
#endif
}
z_new_thread_init(thread, pStackMem, stackAdjSize);
/* carve the thread entry struct from the "base" of
the privileged stack */
pInitCtx = (struct init_stack_frame *)(
priv_stack_end - sizeof(struct init_stack_frame));
/* fill init context */
pInitCtx->status32 = 0U;
if (options & K_USER) {
pInitCtx->pc = ((u32_t)z_user_thread_entry_wrapper);
} else {
pInitCtx->pc = ((u32_t)z_thread_entry_wrapper);
}
/*
* enable US bit, US is read as zero in user mode. This will allow use
* mode sleep instructions, and it enables a form of denial-of-service
* attack by putting the processor in sleep mode, but since interrupt
* level/mask can't be set from user space that's not worse than
* executing a loop without yielding.
* intlock_key is constructed based on ARCv2 ISA Programmer's
* Reference Manual CLRI instruction description:
* dst[31:6] dst[5] dst[4] dst[3:0]
* 26'd0 1 STATUS32.IE STATUS32.E[3:0]
*/
pInitCtx->status32 |= _ARC_V2_STATUS32_US;
#else /* For no USERSPACE feature */
pStackMem += STACK_GUARD_SIZE;
stackEnd = pStackMem + stackSize;
z_new_thread_init(thread, pStackMem, stackSize);
priv_stack_end = stackEnd;
pInitCtx = (struct init_stack_frame *)(
Z_STACK_PTR_ALIGN(priv_stack_end) -
sizeof(struct init_stack_frame));
pInitCtx->status32 = 0U;
pInitCtx->pc = ((u32_t)z_thread_entry_wrapper);
#endif
#ifdef CONFIG_ARC_SECURE_FIRMWARE
pInitCtx->sec_stat = z_arc_v2_aux_reg_read(_ARC_V2_SEC_STAT);
#endif
pInitCtx->r0 = (u32_t)pEntry;
pInitCtx->r1 = (u32_t)parameter1;
pInitCtx->r2 = (u32_t)parameter2;
pInitCtx->r3 = (u32_t)parameter3;
/* stack check configuration */
#ifdef CONFIG_ARC_STACK_CHECKING
#ifdef CONFIG_ARC_SECURE_FIRMWARE
pInitCtx->sec_stat |= _ARC_V2_SEC_STAT_SSC;
#else
pInitCtx->status32 |= _ARC_V2_STATUS32_SC;
#endif
#ifdef CONFIG_USERSPACE
if (options & K_USER) {
thread->arch.u_stack_top = (u32_t)pStackMem;
thread->arch.u_stack_base = (u32_t)stackEnd;
thread->arch.k_stack_top =
(u32_t)(thread->arch.priv_stack_start);
thread->arch.k_stack_base = (u32_t)
(thread->arch.priv_stack_start + CONFIG_PRIVILEGED_STACK_SIZE);
} else {
thread->arch.k_stack_top = (u32_t)pStackMem;
thread->arch.k_stack_base = (u32_t)stackEnd;
thread->arch.u_stack_top = 0;
thread->arch.u_stack_base = 0;
}
#else
thread->arch.k_stack_top = (u32_t) pStackMem;
thread->arch.k_stack_base = (u32_t) stackEnd;
#endif
#endif
#ifdef CONFIG_ARC_USE_UNALIGNED_MEM_ACCESS
pInitCtx->status32 |= _ARC_V2_STATUS32_AD;
#endif
thread->switch_handle = thread;
thread->arch.intlock_key = 0x3F;
thread->arch.relinquish_cause = _CAUSE_COOP;
thread->callee_saved.sp =
(u32_t)pInitCtx - ___callee_saved_stack_t_SIZEOF;
(uint32_t)pInitCtx - ___callee_saved_stack_t_SIZEOF;
/* initial values in all other regs/k_thread entries are irrelevant */
thread_monitor_init(thread);
}
void *z_arch_get_next_switch_handle(struct k_thread **old_thread)
{
*old_thread = _current;
return z_get_next_switch_handle(*old_thread);
}
#ifdef CONFIG_USERSPACE
FUNC_NORETURN void arch_user_mode_enter(k_thread_entry_t user_entry,
void *p1, void *p2, void *p3)
{
_current->stack_info.start = (u32_t)_current->stack_obj;
#ifdef CONFIG_GEN_PRIV_STACKS
_current->arch.priv_stack_start =
(u32_t)z_priv_stack_find(_current->stack_obj);
#else
_current->arch.priv_stack_start =
(u32_t)(_current->stack_info.start +
_current->stack_info.size + STACK_GUARD_SIZE);
#endif
#ifdef CONFIG_ARC_STACK_CHECKING
_current->arch.k_stack_top = _current->arch.priv_stack_start;
_current->arch.k_stack_base = _current->arch.priv_stack_start +
CONFIG_PRIVILEGED_STACK_SIZE;
_current->arch.u_stack_top = _current->stack_info.start;
_current->arch.u_stack_base = _current->stack_info.start +
_current->stack_info.size;
#endif
/* possible optimizaiton: no need to load mem domain anymore */
/* need to lock cpu here ? */
configure_mpu_thread(_current);
z_arc_userspace_enter(user_entry, p1, p2, p3,
(u32_t)_current->stack_obj,
_current->stack_info.size, _current);
CODE_UNREACHABLE;
}
#endif
#if defined(CONFIG_FPU) && defined(CONFIG_FPU_SHARING)
int arch_float_disable(struct k_thread *thread)
{
unsigned int key;
/* Ensure a preemptive context switch does not occur */
key = irq_lock();
/* Disable all floating point capabilities for the thread */
thread->base.user_options &= ~K_FP_REGS;
irq_unlock(key);
return 0;
}
int arch_float_enable(struct k_thread *thread)
{
unsigned int key;
/* Ensure a preemptive context switch does not occur */
key = irq_lock();
/* Enable all floating point capabilities for the thread */
thread->base.user_options |= K_FP_REGS;
irq_unlock(key);
return 0;
}
#endif /* CONFIG_FPU && CONFIG_FPU_SHARING */

View File

@@ -1,38 +1,48 @@
/*
* Copyright (c) 2014-2015 Wind River Systems, Inc.
*
* SPDX-License-Identifier: Apache-2.0
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
* @file
* @brief Wrapper for z_thread_entry
* @brief Wrapper for _thread_entry
*
* Wrapper for z_thread_entry routine when called from the initial context.
* Wrapper for _thread_entry routine when called from the initial context.
*/
#include <toolchain.h>
#include <linker/sections.h>
#include <v2/irq.h>
#define _ASMLANGUAGE
GTEXT(z_thread_entry_wrapper)
GTEXT(z_thread_entry_wrapper1)
#include <toolchain.h>
#include <sections.h>
GTEXT(_thread_entry_wrapper)
GTEXT(_thread_entry)
/*
* @brief Wrapper for z_thread_entry
* @brief Wrapper for _thread_entry
*
* The routine pops parameters for the z_thread_entry from stack frame, prepared
* by the arch_new_thread() routine.
* The routine pops parameters for the _thread_entry from stack frame, prepared
* by the _new_thread() routine.
*
* @return N/A
*/
SECTION_FUNC(TEXT, z_thread_entry_wrapper)
seti _ARC_V2_INIT_IRQ_LOCK_KEY
z_thread_entry_wrapper1:
SECTION_FUNC(TEXT, _thread_entry_wrapper)
pop_s r3
pop_s r2
pop_s r1
pop_s r0
j z_thread_entry
j _thread_entry
nop

View File

@@ -1,39 +0,0 @@
/*
* Copyright (c) 2017 Synopsys, Inc.
*
* SPDX-License-Identifier: Apache-2.0
*/
/**
* @file
* @brief Time Stamp API for ARCv2
*
* Provide 64-bit time stamp API
*/
#include <kernel.h>
#include <toolchain.h>
#include <kernel_structs.h>
/*
* @brief Read 64-bit timestamp value
*
* This function returns a 64-bit bit time stamp value that is clocked
* at the same frequency as the CPU.
*
* @return 64-bit time stamp value
*/
u64_t z_tsc_read(void)
{
unsigned int key;
u64_t t;
u32_t count;
key = arch_irq_lock();
t = (u64_t)z_tick_get();
count = z_arc_v2_aux_reg_read(_ARC_V2_TMR0_COUNT);
arch_irq_unlock(key);
t *= k_ticks_to_cyc_floor64(1);
t += (u64_t)count;
return t;
}

View File

@@ -1,294 +0,0 @@
/*
* Copyright (c) 2018 Synopsys.
*
* SPDX-License-Identifier: Apache-2.0
*/
#include <offsets_short.h>
#include <toolchain.h>
#include <linker/sections.h>
#include <kernel_structs.h>
#include <arch/cpu.h>
#include <syscall.h>
#include <swap_macros.h>
#include <v2/irq.h>
.macro clear_scratch_regs
mov_s r1, 0
mov_s r2, 0
mov_s r3, 0
mov_s r4, 0
mov_s r5, 0
mov_s r6, 0
mov_s r7, 0
mov_s r8, 0
mov_s r9, 0
mov_s r10, 0
mov_s r11, 0
mov_s r12, 0
.endm
.macro clear_callee_regs
mov_s r25, 0
mov_s r24, 0
mov_s r23, 0
mov_s r22, 0
mov_s r21, 0
mov_s r20, 0
mov_s r19, 0
mov_s r18, 0
mov_s r17, 0
mov_s r16, 0
mov_s r15, 0
mov_s r14, 0
mov_s r13, 0
.endm
GTEXT(z_arc_userspace_enter)
GTEXT(_arc_do_syscall)
GTEXT(z_user_thread_entry_wrapper)
GTEXT(arch_user_string_nlen)
GTEXT(z_arc_user_string_nlen_fault_start)
GTEXT(z_arc_user_string_nlen_fault_end)
GTEXT(z_arc_user_string_nlen_fixup)
/*
* @brief Wrapper for z_thread_entry in the case of user thread
* The init parameters are in privileged stack
*
* @return N/A
*/
SECTION_FUNC(TEXT, z_user_thread_entry_wrapper)
seti _ARC_V2_INIT_IRQ_LOCK_KEY
pop_s r3
pop_s r2
pop_s r1
pop_s r0
/* the start of user sp is in r5 */
pop r5
/* start of privilege stack in blink */
mov_s blink, sp
st.aw r0, [r5, -4]
st.aw r1, [r5, -4]
st.aw r2, [r5, -4]
st.aw r3, [r5, -4]
/*
* when CONFIG_INIT_STACKS is enable, stack will be initialized
* in z_new_thread_init.
*/
j _arc_go_to_user_space
/**
*
* User space entry function
*
* This function is the entry point to user mode from privileged execution.
* The conversion is one way, and threads which transition to user mode do
* not transition back later, unless they are doing system calls.
*
*/
SECTION_FUNC(TEXT, z_arc_userspace_enter)
/*
* In ARCv2, the U bit can only be set through exception return
*/
/* disable stack checking as the stack should be initialized */
_disable_stack_checking blink
/* the end of user stack in r5 */
add r5, r4, r5
/* get start of privilege stack, r6 points to current thread */
ld blink, [r6, _thread_offset_to_priv_stack_start]
add blink, blink, CONFIG_PRIVILEGED_STACK_SIZE
mov_s sp, r5
push_s r0
push_s r1
push_s r2
push_s r3
mov r5, sp /* skip r0, r1, r2, r3 */
/* to avoid the leakage of kernel info, the thread stack needs to be
* re-initialized
*/
#ifdef CONFIG_INIT_STACKS
mov_s r0, 0xaaaaaaaa
#else
mov_s r0, 0x0
#endif
_clear_user_stack:
st.ab r0, [r4, 4]
cmp r4, r5
jlt _clear_user_stack
/* reload the stack checking regs as the original kernel stack
* becomes user stack
*/
#ifdef CONFIG_ARC_STACK_CHECKING
/* current thread in r6, SMP case is also considered */
mov r2, r6
_load_stack_check_regs
_enable_stack_checking r0
#endif
/* the following codes are used to switch from kernel mode
* to user mode by fake exception, because U bit can only be set
* by exception
*/
_arc_go_to_user_space:
lr r0, [_ARC_V2_STATUS32]
bset r0, r0, _ARC_V2_STATUS32_U_BIT
mov_s r1, z_thread_entry_wrapper1
sr r0, [_ARC_V2_ERSTATUS]
sr r1, [_ARC_V2_ERET]
/* fake exception return */
lr r0, [_ARC_V2_STATUS32]
bset r0, r0, _ARC_V2_STATUS32_AE_BIT
kflag r0
/* when exception returns from kernel to user, sp and _ARC_V2_USER_SP
* /_ARC_V2_SECU_SP will be switched
*/
#if defined(CONFIG_ARC_HAS_SECURE) && defined(CONFIG_ARC_SECURE_FIRMWARE)
lr r0, [_ARC_V2_SEC_STAT]
/* the mode returns from exception return is secure mode */
bset r0, r0, 31
sr r0, [_ARC_V2_ERSEC_STAT]
sr r5, [_ARC_V2_SEC_U_SP]
#else
sr r5, [_ARC_V2_USER_SP]
#endif
mov_s sp, blink
mov_s r0, 0
clear_callee_regs
clear_scratch_regs
mov_s fp, 0
mov_s r29, 0
mov_s r30, 0
mov_s blink, 0
#ifdef CONFIG_EXECUTION_BENCHMARKING
bl read_timer_end_of_userspace_enter
#endif
rtie
/**
*
* Userspace system call function
*
* This function is used to do system calls from unprivileged code. This
* function is responsible for the following:
* 1) Dispatching the system call
* 2) Restoring stack and calling back to the caller of the system call
*
*/
SECTION_FUNC(TEXT, _arc_do_syscall)
/*
* r0-r5: arg1-arg6, r6 is call id which is already checked in
* trap_s handler, r7 is the system call stack frame pointer
* need to recover r0, r1, r2 because they will be modified in
* _create_irq_stack_frame. If a specific syscall frame (different
* with irq stack frame) is defined, the cover of r0, r1, r2 can be
* optimized.
*/
ld_s r0, [sp, ___isf_t_r0_OFFSET]
ld_s r1, [sp, ___isf_t_r1_OFFSET]
ld_s r2, [sp, ___isf_t_r2_OFFSET]
mov r7, sp
mov_s blink, _k_syscall_table
ld.as r6, [blink, r6]
jl [r6]
/* save return value */
st_s r0, [sp, ___isf_t_r0_OFFSET]
mov_s r29, 0
mov_s r30, 0
/* through fake exception return, go back to the caller */
lr r0, [_ARC_V2_STATUS32]
bset r0, r0, _ARC_V2_STATUS32_AE_BIT
kflag r0
#ifdef CONFIG_ARC_SECURE_FIRMWARE
ld_s r0, [sp, ___isf_t_sec_stat_OFFSET]
sr r0,[_ARC_V2_ERSEC_STAT]
#endif
ld_s r0, [sp, ___isf_t_status32_OFFSET]
sr r0,[_ARC_V2_ERSTATUS]
ld_s r0, [sp, ___isf_t_pc_OFFSET] /* eret into pc */
sr r0,[_ARC_V2_ERET]
_pop_irq_stack_frame
rtie
/*
* size_t arch_user_string_nlen(const char *s, size_t maxsize, int *err_arg)
*/
SECTION_FUNC(TEXT, arch_user_string_nlen)
/* int err; */
sub_s sp,sp,0x4
/* Initial error value (-1 failure), store at [sp,0] */
mov_s r3, -1
st_s r3, [sp, 0]
/* Loop setup.
* r12 (position locator) = s - 1
* r0 (length counter return value)) = 0
* lp_count = maxsize + 1
* */
sub r12, r0, 0x1
mov_s r0, 0
add_s r1, r1, 1
mov lp_count, r1
strlen_loop:
z_arc_user_string_nlen_fault_start:
/* is the byte at ++r12 a NULL? if so, we're done. Might fault! */
ldb.aw r1, [r12, 1]
z_arc_user_string_nlen_fault_end:
brne_s r1, 0, not_null
strlen_done:
/* Success, set err to 0 */
mov_s r1, 0
st_s r1, [sp, 0]
z_arc_user_string_nlen_fixup:
/* *err_arg = err; Pop stack and return */
ld_s r1, [sp, 0]
add_s sp, sp, 4
j_s.d [blink]
st_s r1, [r2, 0]
not_null:
/* check if we've hit the maximum, if so we're done. */
brne.d.nt lp_count, 0x1, inc_len
sub lp_count, lp_count, 0x1
b_s strlen_done
inc_len:
/* increment length measurement, loop again */
add_s r0, r0, 1
b_s strlen_loop

View File

@@ -1,7 +1,17 @@
/*
* Copyright (c) 2014 Wind River Systems, Inc.
*
* SPDX-License-Identifier: Apache-2.0
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
@@ -23,44 +33,45 @@
* swapped.
*/
#include <zephyr/types.h>
#include <stdint.h>
#include <toolchain.h>
#include "vector_table.h"
struct vector_table {
u32_t reset;
u32_t memory_error;
u32_t instruction_error;
u32_t ev_machine_check;
u32_t ev_tlb_miss_i;
u32_t ev_tlb_miss_d;
u32_t ev_prot_v;
u32_t ev_privilege_v;
u32_t ev_swi;
u32_t ev_trap;
u32_t ev_extension;
u32_t ev_div_zero;
u32_t ev_dc_error;
u32_t ev_maligned;
u32_t unused_1;
u32_t unused_2;
uint32_t reset;
uint32_t memory_error;
uint32_t instruction_error;
uint32_t ev_machine_check;
uint32_t ev_tlb_miss_i;
uint32_t ev_tlb_miss_d;
uint32_t ev_prot_v;
uint32_t ev_privilege_v;
uint32_t ev_swi;
uint32_t ev_trap;
uint32_t ev_extension;
uint32_t ev_div_zero;
uint32_t ev_dc_error;
uint32_t ev_maligned;
uint32_t unused_1;
uint32_t unused_2;
};
struct vector_table _VectorTable Z_GENERIC_SECTION(.exc_vector_table) = {
(u32_t)__reset,
(u32_t)__memory_error,
(u32_t)__instruction_error,
(u32_t)__ev_machine_check,
(u32_t)__ev_tlb_miss_i,
(u32_t)__ev_tlb_miss_d,
(u32_t)__ev_prot_v,
(u32_t)__ev_privilege_v,
(u32_t)__ev_swi,
(u32_t)__ev_trap,
(u32_t)__ev_extension,
(u32_t)__ev_div_zero,
(u32_t)__ev_dc_error,
(u32_t)__ev_maligned,
struct vector_table _VectorTable _GENERIC_SECTION(.exc_vector_table) = {
(uint32_t)__reset,
(uint32_t)__memory_error,
(uint32_t)__instruction_error,
(uint32_t)__ev_machine_check,
(uint32_t)__ev_tlb_miss_i,
(uint32_t)__ev_tlb_miss_d,
(uint32_t)__ev_prot_v,
(uint32_t)__ev_privilege_v,
(uint32_t)__ev_swi,
(uint32_t)__ev_trap,
(uint32_t)__ev_extension,
(uint32_t)__ev_div_zero,
(uint32_t)__ev_dc_error,
(uint32_t)__ev_maligned,
0,
0
};

View File

@@ -1,11 +0,0 @@
/*
* Copyright (c) 2019 Nordic Semiconductor ASA
*
* SPDX-License-Identifier: Apache-2.0
*/
/* when !XIP, .text is in RAM, and vector table must be at its very start */
KEEP(*(.exc_vector_table))
KEEP(*(".exc_vector_table.*"))
KEEP(*(_IRQ_VECTOR_TABLE_SECTION_NAME))

6
arch/arc/defconfig Normal file
View File

@@ -0,0 +1,6 @@
CONFIG_ARC=y
CONFIG_SYS_CLOCK_HW_CYCLES_PER_SEC=32000000
CONFIG_CPU_ARCEM4=y
CONFIG_CPU_ARCV2=y
CONFIG_ARCV2_INTERRUPT_UNIT=y
CONFIG_ARCV2_TIMER=y

View File

@@ -1,7 +1,17 @@
/*
* Copyright (c) 2014-2016 Wind River Systems, Inc.
*
* SPDX-License-Identifier: Apache-2.0
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
@@ -17,143 +27,97 @@
* symbols" in the offsets.o module.
*/
#ifndef ZEPHYR_ARCH_ARC_INCLUDE_KERNEL_ARCH_DATA_H_
#define ZEPHYR_ARCH_ARC_INCLUDE_KERNEL_ARCH_DATA_H_
#include <toolchain.h>
#include <linker/sections.h>
#include <arch/cpu.h>
#include <vector_table.h>
#ifndef _ASMLANGUAGE
#include <kernel.h>
#include <zephyr/types.h>
#include <sys/util.h>
#include <sys/dlist.h>
#ifndef _kernel_arch_data__h_
#define _kernel_arch_data__h_
#ifdef __cplusplus
extern "C" {
#endif
#ifdef CONFIG_ARC_HAS_SECURE
#include <toolchain.h>
#include <sections.h>
#include <arch/cpu.h>
#include <vector_table.h>
#ifndef _ASMLANGUAGE
#include <kernel.h>
#include <nano_internal.h>
#include <stdint.h>
#include <misc/util.h>
#include <misc/dlist.h>
#endif
#ifndef _ASMLANGUAGE
struct _caller_saved {
/*
* Saved on the stack as part of handling a regular IRQ or by the
* kernel when calling the FIRQ return code.
*/
};
typedef struct _caller_saved _caller_saved_t;
struct _irq_stack_frame {
u32_t lp_end;
u32_t lp_start;
u32_t lp_count;
uint32_t r0;
uint32_t r1;
uint32_t r2;
uint32_t r3;
uint32_t r4;
uint32_t r5;
uint32_t r6;
uint32_t r7;
uint32_t r8;
uint32_t r9;
uint32_t r10;
uint32_t r11;
uint32_t r12;
uint32_t r13;
uint32_t blink;
uint32_t lp_end;
uint32_t lp_start;
uint32_t lp_count;
#ifdef CONFIG_CODE_DENSITY
/*
* Currently unsupported. This is where those registers are
* automatically pushed on the stack by the CPU when taking a regular
* IRQ.
*/
u32_t ei_base;
u32_t ldi_base;
u32_t jli_base;
uint32_t ei_base;
uint32_t ldi_base;
uint32_t jli_base;
#endif
u32_t r0;
u32_t r1;
u32_t r2;
u32_t r3;
u32_t r4;
u32_t r5;
u32_t r6;
u32_t r7;
u32_t r8;
u32_t r9;
u32_t r10;
u32_t r11;
u32_t r12;
u32_t r13;
u32_t blink;
u32_t pc;
u32_t sec_stat;
u32_t status32;
uint32_t pc;
uint32_t status32;
};
#else
struct _irq_stack_frame {
u32_t r0;
u32_t r1;
u32_t r2;
u32_t r3;
u32_t r4;
u32_t r5;
u32_t r6;
u32_t r7;
u32_t r8;
u32_t r9;
u32_t r10;
u32_t r11;
u32_t r12;
u32_t r13;
u32_t blink;
u32_t lp_end;
u32_t lp_start;
u32_t lp_count;
#ifdef CONFIG_CODE_DENSITY
/*
* Currently unsupported. This is where those registers are
* automatically pushed on the stack by the CPU when taking a regular
* IRQ.
*/
u32_t ei_base;
u32_t ldi_base;
u32_t jli_base;
#endif
u32_t pc;
u32_t status32;
};
#endif
typedef struct _irq_stack_frame _isf_t;
struct _callee_saved {
uint32_t sp; /* r28 */
};
typedef struct _callee_saved _callee_saved_t;
/* callee-saved registers pushed on the stack, not in k_thread */
struct _callee_saved_stack {
u32_t r13;
u32_t r14;
u32_t r15;
u32_t r16;
u32_t r17;
u32_t r18;
u32_t r19;
u32_t r20;
u32_t r21;
u32_t r22;
u32_t r23;
u32_t r24;
u32_t r25;
u32_t r26;
u32_t fp; /* r27 */
#ifdef CONFIG_USERSPACE
#ifdef CONFIG_ARC_HAS_SECURE
u32_t user_sp;
u32_t kernel_sp;
#else
u32_t user_sp;
#endif
#endif
uint32_t r13;
uint32_t r14;
uint32_t r15;
uint32_t r16;
uint32_t r17;
uint32_t r18;
uint32_t r19;
uint32_t r20;
uint32_t r21;
uint32_t r22;
uint32_t r23;
uint32_t r24;
uint32_t r25;
uint32_t r26;
uint32_t fp; /* r27 */
/* r28 is the stack pointer and saved separately */
/* r29 is ILINK and does not need to be saved */
u32_t r30;
#ifdef CONFIG_ARC_HAS_ACCL_REGS
u32_t r58;
u32_t r59;
#endif
#ifdef CONFIG_FPU_SHARING
u32_t fpu_status;
u32_t fpu_ctrl;
#ifdef CONFIG_FP_FPU_DA
u32_t dpfp2h;
u32_t dpfp2l;
u32_t dpfp1h;
u32_t dpfp1l;
#endif
#endif
uint32_t r30;
/*
* No need to save r31 (blink), it's either alread pushed as the pc or
* blink on an irq stack frame.
@@ -162,10 +126,62 @@ struct _callee_saved_stack {
typedef struct _callee_saved_stack _callee_saved_stack_t;
#endif /* _ASMLANGUAGE */
/* stacks */
#define STACK_ALIGN_SIZE 4
#define STACK_ROUND_UP(x) ROUND_UP(x, STACK_ALIGN_SIZE)
#define STACK_ROUND_DOWN(x) ROUND_DOWN(x, STACK_ALIGN_SIZE)
/*
* Reason a thread has relinquished control: fibers can only be in the NONE
* or COOP state, tasks can be one in the four.
*/
#define _CAUSE_NONE 0
#define _CAUSE_COOP 1
#define _CAUSE_RIRQ 2
#define _CAUSE_FIRQ 3
#ifndef _ASMLANGUAGE
struct _thread_arch {
/* interrupt key when relinquishing control */
uint32_t intlock_key;
/* one of the _CAUSE_xxxx definitions above */
int relinquish_cause;
/* return value from _Swap */
unsigned int return_value;
#ifdef CONFIG_ARC_STACK_CHECKING
/* top of stack for hardware stack checking */
uint32_t stack_top;
#endif
};
typedef struct _thread_arch _thread_arch_t;
struct _kernel_arch {
char *rirq_sp; /* regular IRQ stack pointer base */
/*
* FIRQ stack pointer is installed once in the second bank's SP, so
* there is no need to track it in _kernel.
*/
};
typedef struct _kernel_arch _kernel_arch_t;
#endif /* _ASMLANGUAGE */
#ifdef __cplusplus
}
#endif
#endif /* _ASMLANGUAGE */
#endif /* ZEPHYR_ARCH_ARC_INCLUDE_KERNEL_ARCH_DATA_H_ */
#endif /* _kernel_arch_data__h_ */

View File

@@ -1,7 +1,17 @@
/*
* Copyright (c) 2014-2016 Wind River Systems, Inc.
*
* SPDX-License-Identifier: Apache-2.0
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
@@ -17,64 +27,72 @@
* symbols" in the offsets.o module.
*/
#ifndef ZEPHYR_ARCH_ARC_INCLUDE_KERNEL_ARCH_FUNC_H_
#define ZEPHYR_ARCH_ARC_INCLUDE_KERNEL_ARCH_FUNC_H_
#ifndef _kernel_arch_func__h_
#define _kernel_arch_func__h_
#ifdef __cplusplus
extern "C" {
#endif
#if !defined(_ASMLANGUAGE)
#include <kernel_arch_data.h>
#ifdef CONFIG_CPU_ARCV2
#include <v2/cache.h>
#include <v2/irq.h>
#endif
#ifdef __cplusplus
extern "C" {
#endif
static ALWAYS_INLINE void arch_kernel_init(void)
static ALWAYS_INLINE void nanoArchInit(void)
{
z_irq_setup();
_icache_setup();
_irq_setup();
}
static ALWAYS_INLINE void
_set_thread_return_value(struct k_thread *thread, unsigned int value)
{
thread->arch.return_value = value;
}
static ALWAYS_INLINE int _is_in_isr(void)
{
uint32_t act = _arc_v2_aux_reg_read(_ARC_V2_AUX_IRQ_ACT);
#if CONFIG_IRQ_OFFLOAD
/* Check if we're in a TRAP_S exception as well */
if (_arc_v2_aux_reg_read(_ARC_V2_STATUS32) & _ARC_V2_STATUS32_AE &&
_ARC_V2_ECR_VECTOR(_arc_v2_aux_reg_read(_ARC_V2_ECR)) == EXC_EV_TRAP
) {
return 1;
}
#endif
return ((act & 0xffff) != 0);
}
/**
*
* @brief Indicates the interrupt number of the highest priority
* @bried Indicates the interrupt number of the highest priority
* active interrupt
*
* @return IRQ number
*/
static ALWAYS_INLINE int Z_INTERRUPT_CAUSE(void)
static ALWAYS_INLINE int _INTERRUPT_CAUSE(void)
{
u32_t irq_num = z_arc_v2_aux_reg_read(_ARC_V2_ICAUSE);
uint32_t irq_num = _arc_v2_aux_reg_read(_ARC_V2_ICAUSE);
return irq_num;
}
static inline bool arch_is_in_isr(void)
extern void _thread_entry_wrapper(void);
static inline void _IntLibInit(void)
{
return z_arc_v2_irq_unit_is_in_isr();
/* nothing needed, here because the kernel requires it */
}
extern void z_thread_entry_wrapper(void);
extern void z_user_thread_entry_wrapper(void);
extern void z_arc_userspace_enter(k_thread_entry_t user_entry, void *p1,
void *p2, void *p3, u32_t stack, u32_t size,
struct k_thread *thread);
extern void arch_switch(void *switch_to, void **switched_from);
extern void z_arc_fatal_error(unsigned int reason, const z_arch_esf_t *esf);
extern void arch_sched_ipi(void);
#endif /* _ASMLANGUAGE */
#ifdef __cplusplus
}
#endif
#endif /* _ASMLANGUAGE */
#endif /* ZEPHYR_ARCH_ARC_INCLUDE_KERNEL_ARCH_FUNC_H_ */
#endif /* _kernel_arch_func__h_ */

View File

@@ -0,0 +1,46 @@
/*
* Copyright (c) 2016 Intel Corporation
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
* @file
* @brief Kernel event logger support for ARM
*/
#ifndef __KERNEL_EVENT_LOGGER_ARCH_H__
#define __KERNEL_EVENT_LOGGER_ARCH_H__
#ifdef __cplusplus
extern "C" {
#endif
/**
* @brief Get the identification of the current interrupt.
*
* This routine obtain the key of the interrupt that is currently processed
* if it is called from a ISR context.
*
* @return The key of the interrupt that is currently being processed.
*/
int _sys_current_irq_key_get(void)
{
return _INTERRUPT_CAUSE();
}
#ifdef __cplusplus
}
#endif
#endif /* __KERNEL_EVENT_LOGGER_ARCH_H__ */

View File

@@ -1,11 +1,21 @@
/*
* Copyright (c) 2016 Wind River Systems, Inc.
*
* SPDX-License-Identifier: Apache-2.0
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
#ifndef ZEPHYR_ARCH_ARC_INCLUDE_OFFSETS_SHORT_ARCH_H_
#define ZEPHYR_ARCH_ARC_INCLUDE_OFFSETS_SHORT_ARCH_H_
#ifndef _offsets_short_arch__h_
#define _offsets_short_arch__h_
#include <offsets.h>
@@ -17,27 +27,22 @@
/* threads */
#define _thread_offset_to_intlock_key \
(___thread_t_arch_OFFSET + ___thread_arch_t_intlock_key_OFFSET)
#define _thread_offset_to_relinquish_cause \
(___thread_t_arch_OFFSET + ___thread_arch_t_relinquish_cause_OFFSET)
#define _thread_offset_to_k_stack_base \
(___thread_t_arch_OFFSET + ___thread_arch_t_k_stack_base_OFFSET)
#define _thread_offset_to_return_value \
(___thread_t_arch_OFFSET + ___thread_arch_t_return_value_OFFSET)
#define _thread_offset_to_k_stack_top \
(___thread_t_arch_OFFSET + ___thread_arch_t_k_stack_top_OFFSET)
#define _thread_offset_to_u_stack_base \
(___thread_t_arch_OFFSET + ___thread_arch_t_u_stack_base_OFFSET)
#define _thread_offset_to_u_stack_top \
(___thread_t_arch_OFFSET + ___thread_arch_t_u_stack_top_OFFSET)
#define _thread_offset_to_priv_stack_start \
(___thread_t_arch_OFFSET + ___thread_arch_t_priv_stack_start_OFFSET)
#define _thread_offset_to_stack_top \
(___thread_t_arch_OFFSET + ___thread_arch_t_stack_top_OFFSET)
#define _thread_offset_to_sp \
(___thread_t_callee_saved_OFFSET + ___callee_saved_t_sp_OFFSET)
/* end - threads */
#endif /* ZEPHYR_ARCH_ARC_INCLUDE_OFFSETS_SHORT_ARCH_H_ */
#endif /* _offsets_short_arch__h_ */

View File

@@ -0,0 +1,48 @@
/*
* Copyright (c) 2016 Synopsys, Inc. All rights reserved.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
* @file
* @brief ARC nanokernel declarations to start a task
*
* ARC-specific parts of start_task().
*
* Currently empty, only here for abstraction.
*/
#ifndef _START_TASK_ARCH__H_
#define _START_TASK_ARCH__H_
#include <toolchain.h>
#include <sections.h>
#include <micro_private.h>
#include <kernel_structs.h>
#include <microkernel/task.h>
#ifdef __cplusplus
extern "C" {
#endif
#define _START_TASK_ARCH(task, opt_ptr) \
do {/* nothing */ \
} while ((0))
#ifdef __cplusplus
}
#endif
#endif /* _START_TASK_ARCH__H_ */

View File

@@ -1,524 +0,0 @@
/* swap_macros.h - helper macros for context switch */
/*
* Copyright (c) 2014 Wind River Systems, Inc.
*
* SPDX-License-Identifier: Apache-2.0
*/
#ifndef ZEPHYR_ARCH_ARC_INCLUDE_SWAP_MACROS_H_
#define ZEPHYR_ARCH_ARC_INCLUDE_SWAP_MACROS_H_
#include <kernel_structs.h>
#include <offsets_short.h>
#include <toolchain.h>
#include <arch/cpu.h>
#ifdef _ASMLANGUAGE
/* save callee regs of current thread in r2 */
.macro _save_callee_saved_regs
sub_s sp, sp, ___callee_saved_stack_t_SIZEOF
/* save regs on stack */
st_s r13, [sp, ___callee_saved_stack_t_r13_OFFSET]
st_s r14, [sp, ___callee_saved_stack_t_r14_OFFSET]
st_s r15, [sp, ___callee_saved_stack_t_r15_OFFSET]
st r16, [sp, ___callee_saved_stack_t_r16_OFFSET]
st r17, [sp, ___callee_saved_stack_t_r17_OFFSET]
st r18, [sp, ___callee_saved_stack_t_r18_OFFSET]
st r19, [sp, ___callee_saved_stack_t_r19_OFFSET]
st r20, [sp, ___callee_saved_stack_t_r20_OFFSET]
st r21, [sp, ___callee_saved_stack_t_r21_OFFSET]
st r22, [sp, ___callee_saved_stack_t_r22_OFFSET]
st r23, [sp, ___callee_saved_stack_t_r23_OFFSET]
st r24, [sp, ___callee_saved_stack_t_r24_OFFSET]
st r25, [sp, ___callee_saved_stack_t_r25_OFFSET]
st r26, [sp, ___callee_saved_stack_t_r26_OFFSET]
st fp, [sp, ___callee_saved_stack_t_fp_OFFSET]
#ifdef CONFIG_USERSPACE
#ifdef CONFIG_ARC_HAS_SECURE
#ifdef CONFIG_ARC_SECURE_FIRMWARE
lr r13, [_ARC_V2_SEC_U_SP]
st_s r13, [sp, ___callee_saved_stack_t_user_sp_OFFSET]
lr r13, [_ARC_V2_SEC_K_SP]
st_s r13, [sp, ___callee_saved_stack_t_kernel_sp_OFFSET]
#else
lr r13, [_ARC_V2_USER_SP]
st_s r13, [sp, ___callee_saved_stack_t_user_sp_OFFSET]
lr r13, [_ARC_V2_KERNEL_SP]
st_s r13, [sp, ___callee_saved_stack_t_kernel_sp_OFFSET]
#endif /* CONFIG_ARC_SECURE_FIRMWARE */
#else
lr r13, [_ARC_V2_USER_SP]
st_s r13, [sp, ___callee_saved_stack_t_user_sp_OFFSET]
#endif
#endif
st r30, [sp, ___callee_saved_stack_t_r30_OFFSET]
#ifdef CONFIG_ARC_HAS_ACCL_REGS
st r58, [sp, ___callee_saved_stack_t_r58_OFFSET]
st r59, [sp, ___callee_saved_stack_t_r59_OFFSET]
#endif
#ifdef CONFIG_FPU_SHARING
ld_s r13, [r2, ___thread_base_t_user_options_OFFSET]
/* K_FP_REGS is bit 1 */
bbit0 r13, 1, 1f
lr r13, [_ARC_V2_FPU_STATUS]
st_s r13, [sp, ___callee_saved_stack_t_fpu_status_OFFSET]
lr r13, [_ARC_V2_FPU_CTRL]
st_s r13, [sp, ___callee_saved_stack_t_fpu_ctrl_OFFSET]
#ifdef CONFIG_FP_FPU_DA
lr r13, [_ARC_V2_FPU_DPFP1L]
st_s r13, [sp, ___callee_saved_stack_t_dpfp1l_OFFSET]
lr r13, [_ARC_V2_FPU_DPFP1H]
st_s r13, [sp, ___callee_saved_stack_t_dpfp1h_OFFSET]
lr r13, [_ARC_V2_FPU_DPFP2L]
st_s r13, [sp, ___callee_saved_stack_t_dpfp2l_OFFSET]
lr r13, [_ARC_V2_FPU_DPFP2H]
st_s r13, [sp, ___callee_saved_stack_t_dpfp2h_OFFSET]
#endif
1 :
#endif
/* save stack pointer in struct k_thread */
st sp, [r2, _thread_offset_to_sp]
.endm
/* load the callee regs of thread (in r2)*/
.macro _load_callee_saved_regs
/* restore stack pointer from struct k_thread */
ld sp, [r2, _thread_offset_to_sp]
#ifdef CONFIG_ARC_HAS_ACCL_REGS
ld r58, [sp, ___callee_saved_stack_t_r58_OFFSET]
ld r59, [sp, ___callee_saved_stack_t_r59_OFFSET]
#endif
#ifdef CONFIG_FPU_SHARING
ld_s r13, [r2, ___thread_base_t_user_options_OFFSET]
/* K_FP_REGS is bit 1 */
bbit0 r13, 1, 2f
ld_s r13, [sp, ___callee_saved_stack_t_fpu_status_OFFSET]
sr r13, [_ARC_V2_FPU_STATUS]
ld_s r13, [sp, ___callee_saved_stack_t_fpu_ctrl_OFFSET]
sr r13, [_ARC_V2_FPU_CTRL]
#ifdef CONFIG_FP_FPU_DA
ld_s r13, [sp, ___callee_saved_stack_t_dpfp1l_OFFSET]
sr r13, [_ARC_V2_FPU_DPFP1L]
ld_s r13, [sp, ___callee_saved_stack_t_dpfp1h_OFFSET]
sr r13, [_ARC_V2_FPU_DPFP1H]
ld_s r13, [sp, ___callee_saved_stack_t_dpfp2l_OFFSET]
sr r13, [_ARC_V2_FPU_DPFP2L]
ld_s r13, [sp, ___callee_saved_stack_t_dpfp2h_OFFSET]
sr r13, [_ARC_V2_FPU_DPFP2H]
#endif
2 :
#endif
#ifdef CONFIG_USERSPACE
#ifdef CONFIG_ARC_HAS_SECURE
#ifdef CONFIG_ARC_SECURE_FIRMWARE
ld_s r13, [sp, ___callee_saved_stack_t_user_sp_OFFSET]
sr r13, [_ARC_V2_SEC_U_SP]
ld_s r13, [sp, ___callee_saved_stack_t_kernel_sp_OFFSET]
sr r13, [_ARC_V2_SEC_K_SP]
#else
ld_s r13, [sp, ___callee_saved_stack_t_user_sp_OFFSET]
sr r13, [_ARC_V2_USER_SP]
ld_s r13, [sp, ___callee_saved_stack_t_kernel_sp_OFFSET]
sr r13, [_ARC_V2_KERNEL_SP]
#endif /* CONFIG_ARC_SECURE_FIRMWARE */
#else
ld_s r13, [sp, ___callee_saved_stack_t_user_sp_OFFSET]
sr r13, [_ARC_V2_USER_SP]
#endif
#endif
ld_s r13, [sp, ___callee_saved_stack_t_r13_OFFSET]
ld_s r14, [sp, ___callee_saved_stack_t_r14_OFFSET]
ld_s r15, [sp, ___callee_saved_stack_t_r15_OFFSET]
ld r16, [sp, ___callee_saved_stack_t_r16_OFFSET]
ld r17, [sp, ___callee_saved_stack_t_r17_OFFSET]
ld r18, [sp, ___callee_saved_stack_t_r18_OFFSET]
ld r19, [sp, ___callee_saved_stack_t_r19_OFFSET]
ld r20, [sp, ___callee_saved_stack_t_r20_OFFSET]
ld r21, [sp, ___callee_saved_stack_t_r21_OFFSET]
ld r22, [sp, ___callee_saved_stack_t_r22_OFFSET]
ld r23, [sp, ___callee_saved_stack_t_r23_OFFSET]
ld r24, [sp, ___callee_saved_stack_t_r24_OFFSET]
ld r25, [sp, ___callee_saved_stack_t_r25_OFFSET]
ld r26, [sp, ___callee_saved_stack_t_r26_OFFSET]
ld fp, [sp, ___callee_saved_stack_t_fp_OFFSET]
ld r30, [sp, ___callee_saved_stack_t_r30_OFFSET]
add_s sp, sp, ___callee_saved_stack_t_SIZEOF
.endm
/* discard callee regs */
.macro _discard_callee_saved_regs
add_s sp, sp, ___callee_saved_stack_t_SIZEOF
.endm
/*
* Must be called with interrupts locked or in P0.
* Upon exit, sp will be pointing to the stack frame.
*/
.macro _create_irq_stack_frame
sub_s sp, sp, ___isf_t_SIZEOF
st blink, [sp, ___isf_t_blink_OFFSET]
/* store these right away so we can use them if needed */
st_s r13, [sp, ___isf_t_r13_OFFSET]
st_s r12, [sp, ___isf_t_r12_OFFSET]
st r11, [sp, ___isf_t_r11_OFFSET]
st r10, [sp, ___isf_t_r10_OFFSET]
st r9, [sp, ___isf_t_r9_OFFSET]
st r8, [sp, ___isf_t_r8_OFFSET]
st r7, [sp, ___isf_t_r7_OFFSET]
st r6, [sp, ___isf_t_r6_OFFSET]
st r5, [sp, ___isf_t_r5_OFFSET]
st r4, [sp, ___isf_t_r4_OFFSET]
st_s r3, [sp, ___isf_t_r3_OFFSET]
st_s r2, [sp, ___isf_t_r2_OFFSET]
st_s r1, [sp, ___isf_t_r1_OFFSET]
st_s r0, [sp, ___isf_t_r0_OFFSET]
mov r0, lp_count
st_s r0, [sp, ___isf_t_lp_count_OFFSET]
lr r1, [_ARC_V2_LP_START]
lr r0, [_ARC_V2_LP_END]
st_s r1, [sp, ___isf_t_lp_start_OFFSET]
st_s r0, [sp, ___isf_t_lp_end_OFFSET]
#ifdef CONFIG_CODE_DENSITY
lr r1, [_ARC_V2_JLI_BASE]
lr r0, [_ARC_V2_LDI_BASE]
lr r2, [_ARC_V2_EI_BASE]
st_s r1, [sp, ___isf_t_jli_base_OFFSET]
st_s r0, [sp, ___isf_t_ldi_base_OFFSET]
st_s r2, [sp, ___isf_t_ei_base_OFFSET]
#endif
.endm
/*
* Must be called with interrupts locked or in P0.
* sp must be pointing the to stack frame.
*/
.macro _pop_irq_stack_frame
ld blink, [sp, ___isf_t_blink_OFFSET]
#ifdef CONFIG_CODE_DENSITY
ld_s r1, [sp, ___isf_t_jli_base_OFFSET]
ld_s r0, [sp, ___isf_t_ldi_base_OFFSET]
ld_s r2, [sp, ___isf_t_ei_base_OFFSET]
sr r1, [_ARC_V2_JLI_BASE]
sr r0, [_ARC_V2_LDI_BASE]
sr r2, [_ARC_V2_EI_BASE]
#endif
ld_s r0, [sp, ___isf_t_lp_count_OFFSET]
mov lp_count, r0
ld_s r1, [sp, ___isf_t_lp_start_OFFSET]
ld_s r0, [sp, ___isf_t_lp_end_OFFSET]
sr r1, [_ARC_V2_LP_START]
sr r0, [_ARC_V2_LP_END]
ld_s r13, [sp, ___isf_t_r13_OFFSET]
ld_s r12, [sp, ___isf_t_r12_OFFSET]
ld r11, [sp, ___isf_t_r11_OFFSET]
ld r10, [sp, ___isf_t_r10_OFFSET]
ld r9, [sp, ___isf_t_r9_OFFSET]
ld r8, [sp, ___isf_t_r8_OFFSET]
ld r7, [sp, ___isf_t_r7_OFFSET]
ld r6, [sp, ___isf_t_r6_OFFSET]
ld r5, [sp, ___isf_t_r5_OFFSET]
ld r4, [sp, ___isf_t_r4_OFFSET]
ld_s r3, [sp, ___isf_t_r3_OFFSET]
ld_s r2, [sp, ___isf_t_r2_OFFSET]
ld_s r1, [sp, ___isf_t_r1_OFFSET]
ld_s r0, [sp, ___isf_t_r0_OFFSET]
/*
* All gprs have been reloaded, the only one that is still usable is
* ilink.
*
* The pc and status32 values will still be on the stack. We cannot
* pop them yet because the callers of _pop_irq_stack_frame must reload
* status32 differently depending on the execution context they are
* running in (arch_switch(), firq or exception).
*/
add_s sp, sp, ___isf_t_SIZEOF
.endm
/*
* To use this macro, r2 should have the value of thread struct pointer to
* _kernel.current. r3 is a scratch reg.
*/
.macro _load_stack_check_regs
#if defined(CONFIG_ARC_SECURE_FIRMWARE)
ld r3, [r2, _thread_offset_to_k_stack_base]
sr r3, [_ARC_V2_S_KSTACK_BASE]
ld r3, [r2, _thread_offset_to_k_stack_top]
sr r3, [_ARC_V2_S_KSTACK_TOP]
#ifdef CONFIG_USERSPACE
ld r3, [r2, _thread_offset_to_u_stack_base]
sr r3, [_ARC_V2_S_USTACK_BASE]
ld r3, [r2, _thread_offset_to_u_stack_top]
sr r3, [_ARC_V2_S_USTACK_TOP]
#endif
#else /* CONFIG_ARC_HAS_SECURE */
ld r3, [r2, _thread_offset_to_k_stack_base]
sr r3, [_ARC_V2_KSTACK_BASE]
ld r3, [r2, _thread_offset_to_k_stack_top]
sr r3, [_ARC_V2_KSTACK_TOP]
#ifdef CONFIG_USERSPACE
ld r3, [r2, _thread_offset_to_u_stack_base]
sr r3, [_ARC_V2_USTACK_BASE]
ld r3, [r2, _thread_offset_to_u_stack_top]
sr r3, [_ARC_V2_USTACK_TOP]
#endif
#endif /* CONFIG_ARC_SECURE_FIRMWARE */
.endm
/* check and increase the interrupt nest counter
* after increase, check whether nest counter == 1
* the result will be EQ bit of status32
* two temp regs are needed
*/
.macro _check_and_inc_int_nest_counter reg1 reg2
#ifdef CONFIG_SMP
_get_cpu_id \reg1
ld.as \reg1, [@_curr_cpu, \reg1]
ld \reg2, [\reg1, ___cpu_t_nested_OFFSET]
#else
mov \reg1, _kernel
ld \reg2, [\reg1, _kernel_offset_to_nested]
#endif
add \reg2, \reg2, 1
#ifdef CONFIG_SMP
st \reg2, [\reg1, ___cpu_t_nested_OFFSET]
#else
st \reg2, [\reg1, _kernel_offset_to_nested]
#endif
cmp \reg2, 1
.endm
/* decrease interrupt stack nest counter
* the counter > 0, interrupt stack is used, or
* not used
*/
.macro _dec_int_nest_counter reg1 reg2
#ifdef CONFIG_SMP
_get_cpu_id \reg1
ld.as \reg1, [@_curr_cpu, \reg1]
ld \reg2, [\reg1, ___cpu_t_nested_OFFSET]
#else
mov \reg1, _kernel
ld \reg2, [\reg1, _kernel_offset_to_nested]
#endif
sub \reg2, \reg2, 1
#ifdef CONFIG_SMP
st \reg2, [\reg1, ___cpu_t_nested_OFFSET]
#else
st \reg2, [\reg1, _kernel_offset_to_nested]
#endif
.endm
/* If multi bits in IRQ_ACT are set, i.e. last bit != fist bit, it's
* in nest interrupt. The result will be EQ bit of status32
* need two temp reg to do this
*/
.macro _check_nest_int_by_irq_act reg1, reg2
lr \reg1, [_ARC_V2_AUX_IRQ_ACT]
#ifdef CONFIG_ARC_SECURE_FIRMWARE
and \reg1, \reg1, ((1 << ARC_N_IRQ_START_LEVEL) - 1)
#else
and \reg1, \reg1, 0xffff
#endif
ffs \reg2, \reg1
fls \reg1, \reg1
cmp \reg1, \reg2
.endm
/* macro to get id of current cpu
* the result will be in reg (a reg)
*/
.macro _get_cpu_id reg
lr \reg, [_ARC_V2_IDENTITY]
xbfu \reg, \reg, 0xe8
.endm
/* macro to get the interrupt stack of current cpu
* the result will be in irq_sp (a reg)
*/
.macro _get_curr_cpu_irq_stack irq_sp
#ifdef CONFIG_SMP
_get_cpu_id \irq_sp
ld.as \irq_sp, [@_curr_cpu, \irq_sp]
ld \irq_sp, [\irq_sp, ___cpu_t_irq_stack_OFFSET]
#else
mov \irq_sp, _kernel
ld \irq_sp, [\irq_sp, _kernel_offset_to_irq_stack]
#endif
.endm
/* macro to push aux reg through reg */
.macro PUSHAX reg aux
lr \reg, [\aux]
st.a \reg, [sp, -4]
.endm
/* macro to pop aux reg through reg */
.macro POPAX reg aux
ld.ab \reg, [sp, 4]
sr \reg, [\aux]
.endm
/* macro to store old thread call regs */
.macro _store_old_thread_callee_regs
_save_callee_saved_regs
#ifdef CONFIG_SMP
/* save old thread into switch handle which is required by
* wait_for_switch
*/
st r2, [r2, ___thread_t_switch_handle_OFFSET]
#endif
.endm
/* macro to store old thread call regs in interrupt*/
.macro _irq_store_old_thread_callee_regs
#if defined(CONFIG_USERSPACE)
/*
* when USERSPACE is enabled, according to ARCv2 ISA, SP will be switched
* if interrupt comes out in user mode, and will be recorded in bit 31
* (U bit) of IRQ_ACT. when interrupt exits, SP will be switched back
* according to U bit.
*
* need to remember the user/kernel status of interrupted thread, will be
* restored when thread switched back
*
*/
lr r1, [_ARC_V2_AUX_IRQ_ACT]
and r3, r1, 0x80000000
push_s r3
bclr r1, r1, 31
sr r1, [_ARC_V2_AUX_IRQ_ACT]
#endif
_store_old_thread_callee_regs
.endm
/* macro to load new thread callee regs */
.macro _load_new_thread_callee_regs
#ifdef CONFIG_ARC_STACK_CHECKING
_load_stack_check_regs
#endif
/*
* _load_callee_saved_regs expects incoming thread in r2.
* _load_callee_saved_regs restores the stack pointer.
*/
_load_callee_saved_regs
#if defined(CONFIG_MPU_STACK_GUARD) || defined(CONFIG_USERSPACE)
push_s r2
bl configure_mpu_thread
pop_s r2
#endif
ld r3, [r2, _thread_offset_to_relinquish_cause]
.endm
/* when switch to thread caused by coop, some status regs need to set */
.macro _set_misc_regs_irq_switch_from_coop
#ifdef CONFIG_ARC_SECURE_FIRMWARE
/* must return to secure mode, so set IRM bit to 1 */
lr r0, [_ARC_V2_SEC_STAT]
bset r0, r0, _ARC_V2_SEC_STAT_IRM_BIT
sflag r0
#endif
.endm
/* when switch to thread caused by irq, some status regs need to set */
.macro _set_misc_regs_irq_switch_from_irq
#if defined(CONFIG_USERSPACE)
/*
* need to recover the user/kernel status of interrupted thread
*/
pop_s r3
lr r2, [_ARC_V2_AUX_IRQ_ACT]
or r2, r2, r3
sr r2, [_ARC_V2_AUX_IRQ_ACT]
#endif
#ifdef CONFIG_ARC_SECURE_FIRMWARE
/* here need to recover SEC_STAT.IRM bit */
pop_s r3
sflag r3
#endif
.endm
/* macro to get next switch handle in assembly */
.macro _get_next_switch_handle
push_s r2
mov r0, sp
bl z_arch_get_next_switch_handle
pop_s r2
.endm
/* macro to disable stack checking in assembly, need a GPR
* to do this
*/
.macro _disable_stack_checking reg
#ifdef CONFIG_ARC_STACK_CHECKING
#ifdef CONFIG_ARC_SECURE_FIRMWARE
lr \reg, [_ARC_V2_SEC_STAT]
bclr \reg, \reg, _ARC_V2_SEC_STAT_SSC_BIT
sflag \reg
#else
lr \reg, [_ARC_V2_STATUS32]
bclr \reg, \reg, _ARC_V2_STATUS32_SC_BIT
kflag \reg
#endif
#endif
.endm
/* macro to enable stack checking in assembly, need a GPR
* to do this
*/
.macro _enable_stack_checking reg
#ifdef CONFIG_ARC_STACK_CHECKING
#ifdef CONFIG_ARC_SECURE_FIRMWARE
lr \reg, [_ARC_V2_SEC_STAT]
bset \reg, \reg, _ARC_V2_SEC_STAT_SSC_BIT
sflag \reg
#else
lr \reg, [_ARC_V2_STATUS32]
bset \reg, \reg, _ARC_V2_STATUS32_SC_BIT
kflag \reg
#endif
#endif
.endm
#endif /* _ASMLANGUAGE */
#endif /* ZEPHYR_ARCH_ARC_INCLUDE_SWAP_MACROS_H_ */

View File

@@ -1,7 +1,17 @@
/*
* Copyright (c) 2014 Wind River Systems, Inc.
*
* SPDX-License-Identifier: Apache-2.0
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
@@ -12,17 +22,17 @@
* ARCv2 processor architecture.
*/
#ifndef ZEPHYR_ARCH_ARC_INCLUDE_V2_CACHE_H_
#define ZEPHYR_ARCH_ARC_INCLUDE_V2_CACHE_H_
#ifndef _ARCV2_CACHE__H_
#define _ARCV2_CACHE__H_
#include <arch/cpu.h>
#ifndef _ASMLANGUAGE
#ifdef __cplusplus
extern "C" {
#endif
#ifndef _ASMLANGUAGE
/* i-cache defines for IC_CTRL register */
#define IC_CACHE_ENABLE 0x00
#define IC_CACHE_DISABLE 0x01
@@ -34,26 +44,26 @@ extern "C" {
*
* Enables the i-cache and sets it to direct access mode.
*/
static ALWAYS_INLINE void z_icache_setup(void)
static ALWAYS_INLINE void _icache_setup(void)
{
u32_t icache_config = (
uint32_t icache_config = (
IC_CACHE_DIRECT | /* direct mapping (one-way assoc.) */
IC_CACHE_ENABLE /* i-cache enabled */
);
u32_t val;
uint32_t val;
val = z_arc_v2_aux_reg_read(_ARC_V2_I_CACHE_BUILD);
val = _arc_v2_aux_reg_read(_ARC_V2_I_CACHE_BUILD);
val &= 0xff;
if (val != 0U) { /* is i-cache present? */
if (val != 0) { /* is i-cache present? */
/* configure i-cache */
z_arc_v2_aux_reg_write(_ARC_V2_IC_CTRL, icache_config);
_arc_v2_aux_reg_write(_ARC_V2_IC_CTRL, icache_config);
}
}
#endif /* _ASMLANGUAGE */
#ifdef __cplusplus
}
#endif
#endif /* _ASMLANGUAGE */
#endif /* ZEPHYR_ARCH_ARC_INCLUDE_V2_CACHE_H_ */
#endif /* _ARCV2_CACHE__H_ */

View File

@@ -1,21 +1,29 @@
/*
* Copyright (c) 2014 Wind River Systems, Inc.
*
* SPDX-License-Identifier: Apache-2.0
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
* @file
* @brief Interrupt helper functions (ARC)
*
* This file contains private kernel structures definitions and various
* This file contains private nanokernel structures definitions and various
* other definitions for the ARCv2 processor architecture.
*/
#ifndef ZEPHYR_ARCH_ARC_INCLUDE_V2_IRQ_H_
#define ZEPHYR_ARCH_ARC_INCLUDE_V2_IRQ_H_
#include <arch/cpu.h>
#ifndef _ARCV2_IRQ__H_
#define _ARCV2_IRQ__H_
#ifdef __cplusplus
extern "C" {
@@ -23,53 +31,37 @@ extern "C" {
#define _ARC_V2_AUX_IRQ_CTRL_BLINK (1 << 9)
#define _ARC_V2_AUX_IRQ_CTRL_LOOP_REGS (1 << 10)
#define _ARC_V2_AUX_IRQ_CTRL_U (1 << 11)
#define _ARC_V2_AUX_IRQ_CTRL_LP (1 << 13)
#define _ARC_V2_AUX_IRQ_CTRL_14_REGS 7
#define _ARC_V2_AUX_IRQ_CTRL_16_REGS 8
#define _ARC_V2_AUX_IRQ_CTRL_32_REGS 16
#ifdef CONFIG_ARC_SECURE_FIRMWARE
#define _ARC_V2_DEF_IRQ_LEVEL (ARC_N_IRQ_START_LEVEL - 1)
#else
#define _ARC_V2_DEF_IRQ_LEVEL (CONFIG_NUM_IRQ_PRIO_LEVELS - 1)
#endif
#define _ARC_V2_DEF_IRQ_LEVEL (CONFIG_NUM_IRQ_PRIO_LEVELS-1)
#define _ARC_V2_WAKE_IRQ_LEVEL _ARC_V2_DEF_IRQ_LEVEL
/*
* INIT_IRQ_LOCK_KEY is init interrupt level setting of a thread.
* It's configured by seti instruction when a thread starts to run
*, i.e., z_thread_entry_wrapper and z_user_thread_entry_wrapper
*/
#define _ARC_V2_INIT_IRQ_LOCK_KEY (0x10 | _ARC_V2_DEF_IRQ_LEVEL)
#ifndef _ASMLANGUAGE
extern void _firq_stack_setup(void);
extern char _interrupt_stack[];
/*
* z_irq_setup
* _irq_setup
*
* Configures interrupt handling parameters
*/
static ALWAYS_INLINE void z_irq_setup(void)
static ALWAYS_INLINE void _irq_setup(void)
{
u32_t aux_irq_ctrl_value = (
uint32_t aux_irq_ctrl_value = (
_ARC_V2_AUX_IRQ_CTRL_LOOP_REGS | /* save lp_xxx registers */
#ifdef CONFIG_CODE_DENSITY
_ARC_V2_AUX_IRQ_CTRL_LP | /* save code density registers */
#endif
_ARC_V2_AUX_IRQ_CTRL_BLINK | /* save blink */
_ARC_V2_AUX_IRQ_CTRL_14_REGS /* save r0 -> r13 (caller-saved) */
);
z_arc_cpu_sleep_mode = _ARC_V2_WAKE_IRQ_LEVEL;
nano_cpu_sleep_mode = _ARC_V2_WAKE_IRQ_LEVEL;
_arc_v2_aux_reg_write(_ARC_V2_AUX_IRQ_CTRL, aux_irq_ctrl_value);
#ifdef CONFIG_ARC_NORMAL_FIRMWARE
/* normal mode cannot write irq_ctrl, ignore it */
aux_irq_ctrl_value = aux_irq_ctrl_value;
#else
z_arc_v2_aux_reg_write(_ARC_V2_AUX_IRQ_CTRL, aux_irq_ctrl_value);
#endif
_kernel.irq_stack = _interrupt_stack + CONFIG_ISR_STACK_SIZE;
_firq_stack_setup();
}
#endif /* _ASMLANGUAGE */
@@ -78,4 +70,4 @@ static ALWAYS_INLINE void z_irq_setup(void)
}
#endif
#endif /* ZEPHYR_ARCH_ARC_INCLUDE_V2_IRQ_H_ */
#endif /* _ARCV2_IRQ__H_ */

View File

@@ -1,7 +1,17 @@
/*
* Copyright (c) 2014 Wind River Systems, Inc.
*
* SPDX-License-Identifier: Apache-2.0
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/**
@@ -18,15 +28,20 @@
* Refer to the ARCv2 manual for an explanation of the exceptions.
*/
#ifndef ZEPHYR_ARCH_ARC_INCLUDE_VECTOR_TABLE_H_
#define ZEPHYR_ARCH_ARC_INCLUDE_VECTOR_TABLE_H_
#ifndef _VECTOR_TABLE__H_
#define _VECTOR_TABLE__H_
#ifdef __cplusplus
extern "C" {
#endif
#define EXC_EV_TRAP 0x9
#ifdef _ASMLANGUAGE
#include <board.h>
#include <toolchain.h>
#include <linker/sections.h>
#include <sections.h>
GTEXT(__start)
GTEXT(_VectorTable)
@@ -51,10 +66,6 @@ GTEXT(_isr_wrapper)
#else
#ifdef __cplusplus
extern "C" {
#endif
extern void __reset(void);
extern void __memory_error(void);
extern void __instruction_error(void);
@@ -70,10 +81,10 @@ extern void __ev_div_zero(void);
extern void __ev_dc_error(void);
extern void __ev_maligned(void);
#endif /* _ASMLANGUAGE */
#ifdef __cplusplus
}
#endif
#endif /* _ASMLANGUAGE */
#endif /* ZEPHYR_ARCH_ARC_INCLUDE_VECTOR_TABLE_H_ */
#endif /* _VECTOR_TABLE__H_ */

View File

@@ -0,0 +1,8 @@
ccflags-y +=-I$(srctree)/arch/arc/soc/
ccflags-y +=-I$(srctree)/include
ccflags-y +=-I$(srctree)/include/drivers
ccflags-y +=-I$(srctree)/drivers
asflags-y := ${ccflags-y}
obj-y = soc.o soc_config.o

Some files were not shown because too many files have changed in this diff Show More