On Tue, Jul 22, 2025 at 11:06:26PM +0530, Naresh Kamboju wrote:
Regressions while building allyesconfig build for arm64 and x86 with the toolchain clang-nightly version 22.0.0.
This is a "regression" in the sense that the warnings are new but it is from the toolchain side, not the kernel side.
## Build log mm/ksm.c:3674:11: error: variable 'output' is used uninitialized whenever 'if' condition is false [-Werror,-Wsometimes-uninitialized] 3674 | else if (ksm_advisor == KSM_ADVISOR_SCAN_TIME) | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Fix pending in Andrew's tree:
https://git.kernel.org/akpm/mm/c/153ad566724fe6f57b14f66e9726d295d22e576d
drivers/media/platform/samsung/s5p-mfc/s5p_mfc_cmd_v6.c:45:7: error: variable 'h2r_args' is uninitialized when passed as a const pointer argument here [-Werror,-Wuninitialized-const-pointer] 45 | &h2r_args); | ^~~~~~~~
drivers/media/platform/samsung/s5p-mfc/s5p_mfc_cmd_v6.c:133:7: error: variable 'h2r_args' is uninitialized when passed as a const pointer argument here [-Werror,-Wuninitialized-const-pointer] 133 | &h2r_args); | ^~~~~~~~
drivers/media/platform/samsung/s5p-mfc/s5p_mfc_cmd_v6.c:148:7: error: variable 'h2r_args' is uninitialized when passed as a const pointer argument here [-Werror,-Wuninitialized-const-pointer] 148 | &h2r_args); | ^~~~~~~~
https://lore.kernel.org/all/20250715-media-s5p-mfc-fix-uninit-const-pointer-...
drivers/net/wireless/mediatek/mt76/mt7996/mcu.c:3365:21: error: variable 'hdr' is uninitialized when passed as a const pointer argument here [-Werror,-Wuninitialized-const-pointer] 3365 | skb_put_data(skb, &hdr, sizeof(hdr)); | ^~~
drivers/net/wireless/mediatek/mt76/mt7996/mcu.c:1875:21: error: variable 'hdr' is uninitialized when passed as a const pointer argument here [-Werror,-Wuninitialized-const-pointer] 1875 | skb_put_data(skb, &hdr, sizeof(hdr)); | ^~~
https://lore.kernel.org/all/20250715-mt7996-fix-uninit-const-pointer-v1-1-b5...
drivers/net/wireless/broadcom/brcm80211/brcmsmac/phy/phy_lcn.c:2728:13: error: variable 'diq_start' is uninitialized when passed as a const pointer argument here [-Werror,-Wuninitialized-const-pointer] 2728 | &diq_start, 1, 16, 69); | ^~~~~~~~~
Pending in the wireless tree:
https://git.kernel.org/wireless/wireless-next/c/81284e86bf8849f8e98e8ead3ff5...
drivers/usb/atm/cxacru.c:1104:6: error: variable 'bp' is used uninitialized whenever 'if' condition is false [-Werror,-Wsometimes-uninitialized] 1104 | if (instance->modem_type->boot_rom_patch) { | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
I need to send a v2 for this based on Greg's feedback:
https://lore.kernel.org/all/20250716172903.GA4010969@ax162/
drivers/gpu/drm/msm/disp/dpu1/dpu_plane.c:1066:6: error: variable 'crtc_state' is used uninitialized whenever 'if' condition is false [-Werror,-Wsometimes-uninitialized] 1066 | if (plane_state->crtc) | ^~~~~~~~~~~~~~~~~
https://lore.kernel.org/all/20250715-drm-msm-fix-const-uninit-warning-v1-1-d...
drivers/gpu/drm/amd/amdgpu/imu_v12_0.c:374:30: error: variable 'data' is uninitialized when used here [-Werror,-Wuninitialized] 374 | program_imu_rlc_ram(adev, data, (const u32)size); | ^~~~
https://lore.kernel.org/all/20250715-drm-amdgpu-fix-const-uninit-warning-v1-...
Alex said he picked it up but I have not seen it appear in his tree yet:
https://lore.kernel.org/all/BL1PR12MB5144E40FC1D87BABD459A0D6F750A@BL1PR12MB...
Cheers, Nathan