Bugzilla – Bug 1224459
WARNING at apply_retpolines when using NVIDIA proprietary driver - 470 series
Last modified: 2024-07-18 12:25:15 UTC
Upstream NVIDIA Bug Ticket: https://forums.developer.nvidia.com/t/warning-at-apply-retpolines-when-using-nvidia-proprietary-driver-470-series/292638 Downstream NVIDIA Bug Ticket: [ 5.884307] ------------[ cut here ]------------ [ 5.884310] WARNING: CPU: 13 PID: 714 at arch/x86/kernel/alternative.c:654 apply_retpolines+0x350/0x400 [ 5.884316] Modules linked in: xfs(+) mac80211 ac97_bus snd_pcm_dmaengine snd_hda_codec snd_rpl_pci_acp6x snd_acp_pci snd_hda_core snd_acp_legacy_common kvm snd_pci_acp6x snd_hwdep snd_pci_acp5x snd_pcm asus_nb_wmi cfg80211 snd_rn_pci_acp3x snd_timer rapl wmi_bmof snd_acp_config pcspkr libarc4 snd snd_soc_acpi k10temp soundcore snd_pci_acp3x i2c_piix4 asus_wireless amd_pmc joydev mac_hid nvme_fabrics efi_pstore nfnetlink dmi_sysfs ip_tables x_tables hid_asus asus_wmi sparse_keymap platform_profile usbkbd usbhid amdgpu crct10dif_pclmul crc32_pclmul polyval_clmulni amdxcp i2c_algo_bit polyval_generic drm_ttm_helper ghash_clmulni_intel nvme drm_exec sha256_ssse3 sha1_ssse3 gpu_sched hid_multitouch aesni_intel xhci_pci nvme_core drm_suballoc_helper hid_generic xhci_pci_renesas ucsi_acpi crypto_simd drm_buddy cryptd xhci_hcd typec_ucsi drm_display_helper ccp nvme_auth sp5100_tco typec i2c_hid_acpi video i2c_hid wmi hid btrfs blake2b_generic libcrc32c xor raid6_pq msr autofs4 [ 5.884393] CPU: 13 PID: 714 Comm: modprobe Tainted: P O 6.9.0-rc7-x64v1+ #3 openSUNLIGHT Rolling [ 5.884396] Hardware name: ASUSTeK COMPUTER INC. ROG Zephyrus G14 GA401QM_GA401QM/GA401QM, BIOS GA401QM.415 08/11/2023 [ 5.884397] RIP: 0010:apply_retpolines+0x350/0x400 [ 5.884400] Code: 40 80 fe e9 0f 85 96 00 00 00 39 ca 0f 8e 40 ff ff ff 8d 71 01 48 63 c9 c6 44 0d c0 cc e9 51 ff ff ff 41 b8 e0 ff ff ff eb bb <0f> 0b e9 94 fd ff ff 0f 0b 0f b6 8d 69 ff ff ff 89 ce 83 e6 f0 40 [ 5.884402] RSP: 0018:ffffa644c0af7950 EFLAGS: 00010212 [ 5.884404] RAX: 0000000001867640 RBX: ffffffffc2160654 RCX: 0000000000000005 [ 5.884406] RDX: 0000000000000005 RSI: 0000000000000000 RDI: 0000000000000000 [ 5.884407] RBP: ffffa644c0af7a20 R08: 0000000000000000 R09: 0000000000000000 [ 5.884408] R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000 [ 5.884409] R13: ffffa644c0af7970 R14: ffffffffc434cdb4 R15: ffffffffc434cd60 [ 5.884411] FS: 00007cb6ad272740(0000) GS:ffff8c404e880000(0000) knlGS:0000000000000000 [ 5.884413] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 5.884414] CR2: 0000557df1daa410 CR3: 00000001104c6000 CR4: 0000000000f50ef0 [ 5.884416] PKRU: 55555554 [ 5.884417] Call Trace: [ 5.884418] <TASK> [ 5.884421] ? show_regs.cold+0x19/0x20 [ 5.884426] ? apply_retpolines+0x350/0x400 [ 5.884428] ? __warn.cold+0xc3/0x11d [ 5.884431] ? apply_retpolines+0x350/0x400 [ 5.884433] ? report_bug+0xed/0x160 [ 5.884438] ? handle_bug+0x51/0xa0 [ 5.884441] ? exc_invalid_op+0x18/0x80 [ 5.884444] ? asm_exc_invalid_op+0x1b/0x20 [ 5.884449] ? apply_retpolines+0x350/0x400 [ 5.884452] ? srso_alias_return_thunk+0x5/0xfbef5 [ 5.884457] module_finalize+0x1b9/0x330 [ 5.884461] ? add_kallsyms+0x2bb/0x350 [ 5.884466] load_module+0x1734/0x1dd0 [ 5.884469] ? srso_alias_return_thunk+0x5/0xfbef5 [ 5.884474] ? vfree.part.0+0xf0/0x280 [ 5.884477] ? srso_alias_return_thunk+0x5/0xfbef5 [ 5.884481] ? kfree+0x2a2/0x2f0 [ 5.884488] init_module_from_file+0x96/0x100 [ 5.884491] ? srso_alias_return_thunk+0x5/0xfbef5 [ 5.884494] ? init_module_from_file+0x96/0x100 [ 5.884500] idempotent_init_module+0x11c/0x2b0 [ 5.884504] __x64_sys_finit_module+0x64/0xd0 [ 5.884507] x64_sys_call+0x1d6e/0x25c0 [ 5.884510] do_syscall_64+0x7e/0x180 [ 5.884513] ? srso_alias_return_thunk+0x5/0xfbef5 [ 5.884516] ? srso_alias_return_thunk+0x5/0xfbef5 [ 5.884518] ? ksys_lseek+0x80/0xd0 [ 5.884524] ? srso_alias_return_thunk+0x5/0xfbef5 [ 5.884527] ? syscall_exit_to_user_mode+0x81/0x270 [ 5.884531] ? srso_alias_return_thunk+0x5/0xfbef5 [ 5.884534] ? do_syscall_64+0x8b/0x180 [ 5.884537] ? srso_alias_return_thunk+0x5/0xfbef5 [ 5.884539] ? do_user_addr_fault+0x339/0x660 [ 5.884541] ? srso_alias_return_thunk+0x5/0xfbef5 [ 5.884543] ? irqentry_exit_to_user_mode+0x76/0x270 [ 5.884546] ? srso_alias_return_thunk+0x5/0xfbef5 [ 5.884548] ? irqentry_exit+0x43/0x50 [ 5.884550] ? srso_alias_return_thunk+0x5/0xfbef5 [ 5.884552] ? exc_page_fault+0x96/0x1c0 [ 5.884555] entry_SYSCALL_64_after_hwframe+0x76/0x7e [ 5.884557] RIP: 0033:0x7cb6ac911bcd [ 5.884559] Code: 90 90 90 90 90 90 90 90 90 90 90 f3 0f 1e fa 66 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 1b d2 0d 00 f7 d8 64 89 01 48 [ 5.884561] RSP: 002b:00007ffe483a5418 EFLAGS: 00000246 ORIG_RAX: 0000000000000139 [ 5.884563] RAX: ffffffffffffffda RBX: 00005e3822c24170 RCX: 00007cb6ac911bcd [ 5.884564] RDX: 0000000000000004 RSI: 00005e3822c1e110 RDI: 0000000000000003 [ 5.884566] RBP: 00005e3822c1e110 R08: 00007cb6ac9efb20 R09: 00005e3822c241f0 [ 5.884567] R10: 0000000000000050 R11: 0000000000000246 R12: 0000000000040004 [ 5.884568] R13: 0000000000000000 R14: 00005e3822c1e170 R15: 00005e3822c1ddd0 [ 5.884572] </TASK> [ 5.884573] ---[ end trace 0000000000000000 ]--- Driver: nvidia-glG05-470.239.06-60.1 - 470 Series OS: TW 20240510 Kernel: 6.9.x stable version and mainline version(6.9.0+)
Hmm. I don't know anything about that. Is this related to boo#1212841 maybe?
Hi Stefan, Is not related to boo#1212841, it would be a problem related to the proprietary driver. After the new video driver from NVIDIA will be integrated, this ticket should be taken as a reference for the internal tests in openQA.
Ok. So this is a regression caused by updating to kernel 6.9.0+. Mabye our kernel guys have a clue about this. (In reply to Ionut Nechita from comment #2) > Is not related to boo#1212841, it would be a problem related to the > proprietary driver. > After the new video driver from NVIDIA will be integrated, this ticket > should be taken as a reference for the internal tests in openQA. I find this comment confusing and not helpful at all.
Reassigning to kernel guys in hope to receive some feedback. ;-)
Through a quick glance, I see no big code change in the relevant part triggering the WARNING. Might it be because of CONFIG_* renames which confuses Nvidia driver builds? e.g. CONFIG_RETPOLINE was renamed to CONFIG_MITIGATION_RETPOLINE.
Wow! Thanks for this hint. When I differ the previous driver with the current one I see: --- NVIDIA-Linux-x86_64-470.239.06/kernel/nvidia/nv.c 2024-02-03 07:26:18.000000000 +0100 +++ NVIDIA-Linux-x86_64-470.256.02/kernel/nvidia/nv.c 2024-05-02 17:16:35.000000000 +0200 [...] -#if !defined(CONFIG_RETPOLINE) +/* + * Commit aefb2f2e619b ("x86/bugs: Rename CONFIG_RETPOLINE => + * CONFIG_MITIGATION_RETPOLINE) in v6.8 renamed CONFIG_RETPOLINE. + */ +#if !defined(CONFIG_RETPOLINE) && !defined(CONFIG_MITIGATION_RETPOLINE) #include "nv-retpoline.h" #endif So I guess this issue has been fixed with current G05 driver.
@Ionut Please test again with 470.256.02 driver (current G05 package) and let me know whether this fixes the issue. Thanks!
@Ionut Any news on that one?
This is an nvidia issue, not the kernel. And hopefully fixed...
Still waiting for a response for more than a month now. Please reopen once you can provide the requested feedback. Thanks.