amdgpu_proc_options_ioctl(9)
π NAME π
amdgpu_proc_options_ioctl β set per-fd user options. Bestie, that's literally the whole vibe: this is the ioctl that sets per-fd user options, and it came from the Kernel API Manual β¨
π¦ SYNOPSIS π¦
int amdgpu_proc_options_ioctl (struct drm_device *dev , void *data ,
struct drm_file *filp );
π ARGUMENTS π
devβ drm dev pointer, the main-characterdrm_devicein question π«Άdataβ pointer tostruct drm_amdgpu_proc_optionsβ that's where the options info lives β¨filpβ drm file, aka thedrm_filebestie in this story π
πΈ DESCRIPTION πΈ
Okay so this function sets options stored on the per-file amdgpu_fpriv. That's the assignment, and it ate π
Currently the only supported option is AMDGPU_PROC_OPTIONS_OP_KFD_SIGBUS_DELAY, and no cap, it controls how KFD delivers SIGBUS for poison/RAS events to the calling process β immediate, suppressed, or delayed by N milliseconds. Periodt. π₯
β SEE ALSO β
- Kernel file:
./drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c amdgpu_driver_unload_kms(9)amdgpu_driver_load_kms(9)amdgpu_info_ioctl(9)amdgpu_driver_open_kms(9)amdgpu_driver_postclose_kms(9)amdgpu_get_vblank_counter_kms(9)amdgpu_enable_vblank_kms(9)amdgpu_disable_vblank_kms(9)
Bye bestie πβ¨ this cutie's source is ./drivers/gpu/drm/amd/amdgpu, it's dated October 2026, and yes, the page is amdgpu_proc_options_ioctl(9) β slay and goodbye ππ¦
π¦ where this page comes from
- nixpkgs unstable: linux-manual 7.3.1
- π girlified by deepseek-ai/DeepSeek-V4.1-Flash-1017 on : a language model's rewrite of the original, under its terms