west.yml: update zephyr to fb01c781b6c #10312

tmleman · 2025-10-17T11:41:36Z

Total of 2816 commits.

Changes include:

45bfe066c74 ipc: intel_adsp_host_ipc: clear tx_ack_pending on PM resume
6930d71447e west.yml: update hal_nxp
d04f91bce42 west: update Xtensa HAL to latest SHA
b25a2187388 west.yml: Update NXP HAL to get fix for Build failure
74b51136a45 soc: nxp: imx93: drop SOF-specific region mapping
ff9835b8b1c boards: nxp: imx95_evk: drop rimage support
0d4d0cf8b73 modules: hal_nxp: Pull in SDK cmc, spc, vbat, wuu drivers
0d86ebb83ba llext: load: memcpy section header
a7be1ffb466 boards: intel_adsp: add ace40 sim .dts
b3a7eeaa792 soc: intel_adsp: winstream only if not sim
7f42d4c1c22 boards: intel_adsp/ace: remove kconfig to skip 2nd core flow
e8a14b2494d soc: intel_adsp: send fwready msg for sim
38d49efdac8 kernel: mem_domain: keep track of threads only if needed
5ec79bf556b soc: intel_adsp/ace: allows more spin relax loop per CPU
3c2ecf79f53 west: update mipi-sys-t to lastest SHA
b6c66b78582 west: update hal_nxp to mcux sdk 25.09.00
95d260e77ed xtensa: mmu/ptables: rename flags to attrs under arch_mem_map()
84ade183ed1 mmu: cosmetic changes to page table variable names
e709cbe3c89 xtensa: mmu: fix __arch_mem_map assert message
1f07b5b1903 xtensa: userspace: save PS to thread struct later
4709e3feb9a xtensa: rename xtensa_asm2_s.h to xtensa_asm2.inc.S
6b082061cc1 xtensa: move saved FPU and HIFI registers to end of BSA
6877bc8644c xtensa: userspace: handle DTLB multihit exception
d3a126cb5da xtensa: userspace: handle load/store ring exception
9f0f59649bf xtensa: save EXCCAUSE/EXCVADDR in BSA during exception
c76b338ec46 xtensa: mmu: properly restore PTE attributes via reset_region()
f2cf818f67b xtensa: fatal: use named macros for PS register bits...
74d715c5ef9 pm: Remove "power state consistency" check
11a2ec4b139 west.yml: depend on hal_nxp pr 605
ca641d84107 llext: only build fs_loader.c if file-systems are enabled
54545430e46 drivers: sai: Fix compile time error with clang
96d654bbc86 west.yml: Update hal_xtensa revision
cf7e2e63c17 soc: intel_adsp: rework host IPC using IPC service
d910306fd00 soc: intel_adsp: remove IDC dt default for CONFIG_INTEL_ADSP_IPC
0b3b6f30a3d ipm: remove the Intel Audio DSP CAVS host IPM driver

Copilot

Pull Request Overview

Updates the Zephyr revision in the west.yml manifest file to incorporate 1464 commits of upstream changes, including hardware abstraction layer updates, architecture improvements, and driver fixes.

Updates Zephyr revision from e876dee6f2e to d99cddd4ec4
Incorporates HAL updates for NXP (MCUX SDK 25.09.00) and Xtensa platforms
Includes Xtensa architecture improvements for MMU, userspace, and exception handling

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

tmleman · 2025-10-17T11:43:21Z

I can't reproduce the previous D3 failures on LNL. Let's see how it looks in CI.

lgirdwood · 2025-10-17T15:50:46Z

I can't reproduce the previous D3 failures on LNL. Let's see how it looks in CI.

The CI looks OK on LNL and others with Jenkins. Still some CI pending though.

kv2019i · 2025-10-20T10:03:37Z

@tmleman Seems quickbuild tests around D3 entry are failing. I think you saw this with the earlier attempt so seems to repeat. Tests with Linux seem to work and no failures with D3 entry seen.

tmleman · 2025-10-20T14:34:10Z

@tmleman Seems quickbuild tests around D3 entry are failing. I think you saw this with the earlier attempt so seems to repeat. Tests with Linux seem to work and no failures with D3 entry seen.

True, these are the same failures I saw earlier. I was unable to reproduce them locally initially and thought the problem had disappeared. It turns out that the test sequence matters and the issue is a regression introduced by zephyrproject-rtos/zephyr#96233

lgirdwood · 2025-10-22T11:34:11Z

@tmleman Seems quickbuild tests around D3 entry are failing. I think you saw this with the earlier attempt so seems to repeat. Tests with Linux seem to work and no failures with D3 entry seen.

True, these are the same failures I saw earlier. I was unable to reproduce them locally initially and thought the problem had disappeared. It turns out that the test sequence matters and the issue is a regression introduced by zephyrproject-rtos/zephyr#96233

Ok, I guess you have a fix brewing @tmleman ?

tmleman · 2025-10-22T11:55:18Z

@tmleman Seems quickbuild tests around D3 entry are failing. I think you saw this with the earlier attempt so seems to repeat. Tests with Linux seem to work and no failures with D3 entry seen.

True, these are the same failures I saw earlier. I was unable to reproduce them locally initially and thought the problem had disappeared. It turns out that the test sequence matters and the issue is a regression introduced by zephyrproject-rtos/zephyr#96233

Ok, I guess you have a fix brewing @tmleman ?

I'm working on it.

tmleman · 2025-10-28T10:23:11Z

@lgirdwood @kv2019i We need to wait for this: zephyrproject-rtos/zephyr#98320, and then I will make another attempt with update.

kv2019i · 2025-10-29T19:09:52Z

zephyrproject-rtos/zephyr#98320 now merged, this PR can be attempted again.

Total of 2816 commits. Changes include: 45bfe066c74 ipc: intel_adsp_host_ipc: clear tx_ack_pending on PM resume 6930d71447e west.yml: update hal_nxp d04f91bce42 west: update Xtensa HAL to latest SHA b25a2187388 west.yml: Update NXP HAL to get fix for Build failure 74b51136a45 soc: nxp: imx93: drop SOF-specific region mapping ff9835b8b1c boards: nxp: imx95_evk: drop rimage support 0d4d0cf8b73 modules: hal_nxp: Pull in SDK cmc, spc, vbat, wuu drivers 0d86ebb83ba llext: load: memcpy section header a7be1ffb466 boards: intel_adsp: add ace40 sim .dts b3a7eeaa792 soc: intel_adsp: winstream only if not sim 7f42d4c1c22 boards: intel_adsp/ace: remove kconfig to skip 2nd core flow e8a14b2494d soc: intel_adsp: send fwready msg for sim 38d49efdac8 kernel: mem_domain: keep track of threads only if needed 5ec79bf556b soc: intel_adsp/ace: allows more spin relax loop per CPU 3c2ecf79f53 west: update mipi-sys-t to lastest SHA b6c66b78582 west: update hal_nxp to mcux sdk 25.09.00 95d260e77ed xtensa: mmu/ptables: rename flags to attrs under arch_mem_map() 84ade183ed1 mmu: cosmetic changes to page table variable names e709cbe3c89 xtensa: mmu: fix __arch_mem_map assert message 1f07b5b1903 xtensa: userspace: save PS to thread struct later 4709e3feb9a xtensa: rename xtensa_asm2_s.h to xtensa_asm2.inc.S 6b082061cc1 xtensa: move saved FPU and HIFI registers to end of BSA 6877bc8644c xtensa: userspace: handle DTLB multihit exception d3a126cb5da xtensa: userspace: handle load/store ring exception 9f0f59649bf xtensa: save EXCCAUSE/EXCVADDR in BSA during exception c76b338ec46 xtensa: mmu: properly restore PTE attributes via reset_region() f2cf818f67b xtensa: fatal: use named macros for PS register bits... 74d715c5ef9 pm: Remove "power state consistency" check 11a2ec4b139 west.yml: depend on hal_nxp pr 605 ca641d84107 llext: only build fs_loader.c if file-systems are enabled 54545430e46 drivers: sai: Fix compile time error with clang 96d654bbc86 west.yml: Update hal_xtensa revision cf7e2e63c17 soc: intel_adsp: rework host IPC using IPC service d910306fd00 soc: intel_adsp: remove IDC dt default for CONFIG_INTEL_ADSP_IPC 0b3b6f30a3d ipm: remove the Intel Audio DSP CAVS host IPM driver Signed-off-by: Tomasz Leman <tomasz.m.leman@intel.com>

tmleman · 2025-10-30T12:25:28Z

@dbaluta I would need help from someone at NXP now: https://github.com/thesofproject/sof/actions/runs/18936261883/job/54063751698?pr=10312

/__w/sof/sof/modules/hal/nxp/mcux/mcux-sdk-ng/drivers/sai/fsl_sai.c: In function 'SAI_TransferTxCreateHandle':
/__w/sof/sof/modules/hal/nxp/mcux/mcux-sdk-ng/drivers/sai/fsl_sai.c:1951:16: error: array subscript 4 is above array bounds of 'sai_handle_t *[4][2]' {aka 'struct _sai_handle *[4][2]'} [-Werror=array-bounds]
 1951 |     s_saiHandle[SAI_GetInstance(base)][0] = handle;
      |     ~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~
/__w/sof/sof/modules/hal/nxp/mcux/mcux-sdk-ng/drivers/sai/fsl_sai.c:157:22: note: while referencing 's_saiHandle'
  157 | static sai_handle_t *s_saiHandle[ARRAY_SIZE(s_saiBases)][2];
      |                      ^~~~~~~~~~~
/__w/sof/sof/modules/hal/nxp/mcux/mcux-sdk-ng/drivers/sai/fsl_sai.c: In function 'SAI_TransferRxCreateHandle':
/__w/sof/sof/modules/hal/nxp/mcux/mcux-sdk-ng/drivers/sai/fsl_sai.c:1982:16: error: array subscript 4 is above array bounds of 'sai_handle_t *[4][2]' {aka 'struct _sai_handle *[4][2]'} [-Werror=array-bounds]
 1982 |     s_saiHandle[SAI_GetInstance(base)][1] = handle;
      |     ~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~
/__w/sof/sof/modules/hal/nxp/mcux/mcux-sdk-ng/drivers/sai/fsl_sai.c:157:22: note: while referencing 's_saiHandle'
  157 | static sai_handle_t *s_saiHandle[ARRAY_SIZE(s_saiBases)][2];
      |                      ^~~~~~~~~~~
cc1: all warnings being treated as errors

Generating files from /__w/sof/sof/build-imx8/zephyr/zephyr.elf for board: imx8qm_mek
[346/346] west sign --if-tool-available --tool rimage ...
Signing with tool /__w/sof/sof/build-rimage/rimage
Signing for SOC target imx8
/__w/sof/sof/build-rimage/rimage -o /__w/sof/sof/build-imx8/zephyr/zephyr.ri -k 'key param ignored by imx8' -f 2.14.99 -b 1 -e -c /__w/sof/sof/sof/tools/rimage/config/imx8.toml /__w/sof/sof/build-imx8/zephyr/zephyr.elf
error: 'alias_mask' not found

iuliana-prodan · 2025-10-30T18:06:56Z

@dbaluta I would need help from someone at NXP now: https://github.com/thesofproject/sof/actions/runs/18936261883/job/54063751698?pr=10312

/__w/sof/sof/modules/hal/nxp/mcux/mcux-sdk-ng/drivers/sai/fsl_sai.c: In function 'SAI_TransferTxCreateHandle':
/__w/sof/sof/modules/hal/nxp/mcux/mcux-sdk-ng/drivers/sai/fsl_sai.c:1951:16: error: array subscript 4 is above array bounds of 'sai_handle_t *[4][2]' {aka 'struct _sai_handle *[4][2]'} [-Werror=array-bounds]
 1951 |     s_saiHandle[SAI_GetInstance(base)][0] = handle;
      |     ~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~
/__w/sof/sof/modules/hal/nxp/mcux/mcux-sdk-ng/drivers/sai/fsl_sai.c:157:22: note: while referencing 's_saiHandle'
  157 | static sai_handle_t *s_saiHandle[ARRAY_SIZE(s_saiBases)][2];
      |                      ^~~~~~~~~~~
/__w/sof/sof/modules/hal/nxp/mcux/mcux-sdk-ng/drivers/sai/fsl_sai.c: In function 'SAI_TransferRxCreateHandle':
/__w/sof/sof/modules/hal/nxp/mcux/mcux-sdk-ng/drivers/sai/fsl_sai.c:1982:16: error: array subscript 4 is above array bounds of 'sai_handle_t *[4][2]' {aka 'struct _sai_handle *[4][2]'} [-Werror=array-bounds]
 1982 |     s_saiHandle[SAI_GetInstance(base)][1] = handle;
      |     ~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~
/__w/sof/sof/modules/hal/nxp/mcux/mcux-sdk-ng/drivers/sai/fsl_sai.c:157:22: note: while referencing 's_saiHandle'
  157 | static sai_handle_t *s_saiHandle[ARRAY_SIZE(s_saiBases)][2];
      |                      ^~~~~~~~~~~
cc1: all warnings being treated as errors

This appears to be a compilation error related to i.MX93, which is not supported by the build script: https://github.com/thesofproject/sof/blob/main/scripts/xtensa-build-zephyr.py#L201.
Why is the CI attempting to compile for this target?
For ARM64, we only support the i.MX95 platform.

Generating files from /__w/sof/sof/build-imx8/zephyr/zephyr.elf for board: imx8qm_mek
[346/346] west sign --if-tool-available --tool rimage ...
Signing with tool /__w/sof/sof/build-rimage/rimage
Signing for SOC target imx8
/__w/sof/sof/build-rimage/rimage -o /__w/sof/sof/build-imx8/zephyr/zephyr.ri -k 'key param ignored by imx8' -f 2.14.99 -b 1 -e -c /__w/sof/sof/sof/tools/rimage/config/imx8.toml /__w/sof/sof/build-imx8/zephyr/zephyr.elf
error: 'alias_mask' not found

I’ve tested this locally, and it works fine.
With Zephyr updated to commit fb01c781b6c, I was able to compile successfully for imx8 (specifically the imx8qm_mek board).
I used the following command:
$ ./scripts/xtensa-build-zephyr.py -u -p imx8

LaurentiuM1234 · 2025-10-30T18:51:39Z

@dbaluta I would need help from someone at NXP now: https://github.com/thesofproject/sof/actions/runs/18936261883/job/54063751698?pr=10312

/__w/sof/sof/modules/hal/nxp/mcux/mcux-sdk-ng/drivers/sai/fsl_sai.c: In function 'SAI_TransferTxCreateHandle':
/__w/sof/sof/modules/hal/nxp/mcux/mcux-sdk-ng/drivers/sai/fsl_sai.c:1951:16: error: array subscript 4 is above array bounds of 'sai_handle_t *[4][2]' {aka 'struct _sai_handle *[4][2]'} [-Werror=array-bounds]
 1951 |     s_saiHandle[SAI_GetInstance(base)][0] = handle;
      |     ~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~
/__w/sof/sof/modules/hal/nxp/mcux/mcux-sdk-ng/drivers/sai/fsl_sai.c:157:22: note: while referencing 's_saiHandle'
  157 | static sai_handle_t *s_saiHandle[ARRAY_SIZE(s_saiBases)][2];
      |                      ^~~~~~~~~~~
/__w/sof/sof/modules/hal/nxp/mcux/mcux-sdk-ng/drivers/sai/fsl_sai.c: In function 'SAI_TransferRxCreateHandle':
/__w/sof/sof/modules/hal/nxp/mcux/mcux-sdk-ng/drivers/sai/fsl_sai.c:1982:16: error: array subscript 4 is above array bounds of 'sai_handle_t *[4][2]' {aka 'struct _sai_handle *[4][2]'} [-Werror=array-bounds]
 1982 |     s_saiHandle[SAI_GetInstance(base)][1] = handle;
      |     ~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~
/__w/sof/sof/modules/hal/nxp/mcux/mcux-sdk-ng/drivers/sai/fsl_sai.c:157:22: note: while referencing 's_saiHandle'
  157 | static sai_handle_t *s_saiHandle[ARRAY_SIZE(s_saiBases)][2];
      |                      ^~~~~~~~~~~
cc1: all warnings being treated as errors

Generating files from /__w/sof/sof/build-imx8/zephyr/zephyr.elf for board: imx8qm_mek
[346/346] west sign --if-tool-available --tool rimage ...
Signing with tool /__w/sof/sof/build-rimage/rimage
Signing for SOC target imx8
/__w/sof/sof/build-rimage/rimage -o /__w/sof/sof/build-imx8/zephyr/zephyr.ri -k 'key param ignored by imx8' -f 2.14.99 -b 1 -e -c /__w/sof/sof/sof/tools/rimage/config/imx8.toml /__w/sof/sof/build-imx8/zephyr/zephyr.elf
error: 'alias_mask' not found

please ignore the MX93-related failure. This was fixed at some point and re-introduced by some recent NXP HAL update on Zephyr side.

@lgirdwood @kv2019i do you see any value in keeping the build for MX93 in the CI? The only reason for keeping it that I can think of is that it's the only ARM64 platform so it might offer us extra coverage.

If not, then we can remove it. The MX93 support never became "official" and, internally, we don't test this platform.

as for the MX95-related issues: you'd need #10319 to go alongside the manifest update.

kv2019i · 2025-10-31T08:01:13Z

Thanks @LaurentiuM1234 for the update. Given Zephyr will soon get the CONFIG_SOF removal, it is great if we can now merge this and catch up with upstream Zephyr before next set of changes. I'll file a bug for imx93 fail. I think it's valuable to have it (given it is so different platform), but that of course depends on whether you have the resourcing to keep it running (even if at a slower pace). I just removed some Intel CI tests that had been failing for multiple months -- ideally we'd like to keep these, but in practise we had to face resourcing constraints and remove them for now.

Copilot AI review requested due to automatic review settings October 17, 2025 11:41

tmleman requested review from dbaluta, kv2019i, lbetlej, lgirdwood, mmaka1 and plbossart as code owners October 17, 2025 11:41

Copilot AI reviewed Oct 17, 2025

View reviewed changes

tmleman requested review from abonislawski, serhiy-katsyuba-intel and softwarecki October 17, 2025 13:50

serhiy-katsyuba-intel approved these changes Oct 17, 2025

View reviewed changes

lgirdwood approved these changes Oct 17, 2025

View reviewed changes

kv2019i mentioned this pull request Oct 29, 2025

zephyr: CMakeLists.txt: add rimage support for the imx95 target #10319

Merged

tmleman force-pushed the topic/upstream/pr/zephyr/upgrade/25ww41 branch from ffd41dc to 43ade03 Compare October 30, 2025 09:38

tmleman changed the title ~~west.yml: update zephyr to d99cddd4ec4~~ west.yml: update zephyr to fb01c781b6c Oct 30, 2025

abonislawski approved these changes Oct 31, 2025

View reviewed changes

iuliana-prodan approved these changes Oct 31, 2025

View reviewed changes

kv2019i merged commit 203a2f9 into thesofproject:main Oct 31, 2025
36 of 42 checks passed

kv2019i mentioned this pull request Oct 31, 2025

[BUG] fix imx93 github CI build test after 31th Oct Zephyr update in SOF #10344

Open

west.yml: update zephyr to fb01c781b6c #10312

west.yml: update zephyr to fb01c781b6c #10312

Uh oh!

Conversation

tmleman commented Oct 17, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull Request Overview

Uh oh!

tmleman commented Oct 17, 2025

Uh oh!

lgirdwood commented Oct 17, 2025

Uh oh!

kv2019i commented Oct 20, 2025

Uh oh!

tmleman commented Oct 20, 2025

Uh oh!

lgirdwood commented Oct 22, 2025

Uh oh!

tmleman commented Oct 22, 2025

Uh oh!

tmleman commented Oct 28, 2025

Uh oh!

kv2019i commented Oct 29, 2025

Uh oh!

tmleman commented Oct 30, 2025

Uh oh!

iuliana-prodan commented Oct 30, 2025

Uh oh!

LaurentiuM1234 commented Oct 30, 2025

Uh oh!

kv2019i commented Oct 31, 2025

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

7 participants

tmleman commented Oct 17, 2025 •

edited

Loading