gmc_v9_0_flu...pu_tlb_pasid(9)
β¨ gmc_v9_0_flush_gpu_tlb_pasid(9) β¨
okay so bestie, welcome to the Kernel API Manual π
this is the gmc_v9_0_flush_gpu_tlb_pasid page and honestly it's a short one but it ate, so let's get into it π«Ά
πΈ NAME πΈ
gmc_v9_0_flush_gpu_tlb_pasid β tlb flush via pasid β¨
like, that's literally the whole vibe: we're doing a TLB flush, and we're doing it via pasid. periodt π
π¦ SYNOPSIS π¦
void gmc_v9_0_flush_gpu_tlb_pasid (struct amdgpu_device *adev , uint16_t
pasid , uint32_t flush_type , bool all_hub , uint32_t inst );
no cap, that's the exact signature β returns void, so don't go looking for a return value, she gives you nothing back and that's iconic of her π
π ARGUMENTS π
adevβ amdgpu_device pointer π« the main character device struct, we love herpasidβ pasid to be flush π yes that's the actual wording, the pasid that's getting flushedflush_typeβ the flush type β it's the type. of the flush. iykykall_hubβ flush all hubs π₯ all of them. every hub. no hub left behind fr frinstβ is used to select which instance of KIQ to use for the invalidation π§ so like, this one picks which KIQ instance handles the invalidation, okay? not me thinking that's lowkey the most interesting argument here π
π DESCRIPTION π
Flush the TLB for the requested pasid. π―
that's it. that's the whole description. she's concise, she's efficient, she's giving main character energy with zero filler π β¨
ngl bestie, short and sweet is a valid aesthetic β just don't forget it's a TLB flush for the requested pasid specifically, okay? π«Ά
π SEE ALSO π
Kernel file ./drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c π β go read the source, it's all in there:
gmc_v9_0_use_invalidate_semaphore(9)gmc_v9_0_flush_gpu_tlb(9)gmc_v9_0_mc_init(9)gmc_v9_0_save_registers(9)gmc_v9_0_restore_registers(9)gmc_v9_0_gart_enable(9)gmc_v9_0_gart_disable(9)
the whole gmc_v9_0 friend group ππ go say hi to all of them
aaaand that's the page!! π¦ mwah, you understood it, i'm obsessed with you π«Ά so anyway, this little beauty lives over in ./drivers/gpu/drm/amd/amdgpu, dated October 2026, and she answers to gmc_v9_0_flu...pu_tlb_pasid(9) β yes the name is truncated in the header, no we're not fixing it, she's mysterious like that πβ¨
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