MCBP (INT)(9)
π mcbp (int) π
okay so we're in the Kernel API Manual bestie, and this little cutie is called mcbp (int) π
it's a kernel parameter and honestly it's giving "small but powerful main character energy" β¨
πΈ what it does πΈ
It is used to enable mid command buffer preemption. π«Ά
The values are literally:
0= disabled π1= enabled π₯-1auto (default) π
so yes, default is -1, and yes, it's auto, she's mysterious like that π§ no cap.
it lives in the kernel file ./drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c β that's the address, bestie, don't lose it π
π SEE ALSO π
okay the SEE ALSO list is long, like long long, but every single one of these is iconic in her own way, so we're listing them all, periodt π
vramlimit (int)(9)π¦vis_vramlimit (int)(9)π¦gartsize (uint)(9)π¦gttsize (int)(9)π¦moverate (int)(9)π¦audio (int)(9)π¦disp_priority (int)(9)π¦hw_i2c (int)(9)π¦pcie_gen2 (int)(9)π¦msi (int)(9)π¦svm_default_granularity (uint)(9)π¦lockup_timeout (string)(9)π¦dpm (int)(9)π¦fw_load_type (int)(9)π¦aspm (int)(9)π¦runpm (int)(9)π¦ip_block_mask (uint)(9)π¦bapm (int)(9)π¦deep_color (int)(9)π¦vm_size (int)(9)π¦vm_fragment_size (int)(9)π¦vm_block_size (int)(9)π¦vm_fault_stop (int)(9)π¦vm_update_mode (int)(9)π¦exp_hw_support (int)(9)π¦dc (int)(9)π¦sched_jobs (int)(9)π¦sched_hw_submission (int)(9)π¦ppfeaturemask (hexint)(9)π¦forcelongtraining (uint)(9)π¦pcie_gen_cap (uint)(9)π¦pcie_lane_cap (uint)(9)π¦cg_mask (ullong)(9)π¦pg_mask (uint)(9)π¦sdma_phase_quantum (uint)(9)π¦disable_cu (charp)(9)π¦virtual_display (charp)(9)π¦lbpw (int)(9)π¦gpu_recovery (int)(9)π¦emu_mode (int)(9)π¦ras_enable (int)(9)π¦ras_mask (uint)(9)π¦timeout_fatal_disable (bool)(9)π¦timeout_period (uint)(9)π¦si_support (int)(9)π¦cik_support (int)(9)π¦smu_memory_pool_size (uint)(9)π¦async_gfx_ring (int)(9)π¦discovery (int)(9)π¦mes_log_enable (int)(9)π¦uni_mes (int)(9)π¦noretry (int)(9)π¦force_asic_type (int)(9)π¦use_xgmi_p2p (int)(9)π¦sched_policy (int)(9)π¦hws_max_conc_proc (int)(9)π¦cwsr_enable (int)(9)π¦max_num_of_queues_per_device (int)(9)π¦send_sigterm (int)(9)π¦halt_if_hws_hang (int)(9)π¦hws_gws_support(bool)(9)π¦queue_preemption_timeout_ms (int)(9)π¦debug_evictions(bool)(9)π¦no_system_mem_limit(bool)(9)π¦no_queue_eviction_on_vm_fault (int)(9)π¦mtype_local (int)(9)π¦pcie_p2p (bool)(9)π¦dcfeaturemask (uint)(9)π¦dcdebugmask (uint)(9)π¦abmlevel (uint)(9)π¦damageclips (int)(9)π¦tmz (int)(9)π¦freesync_video (uint)(9)π¦reset_method (int)(9)π¦bad_page_threshold (int) Bad page threshold is specifies the(9)π¦vcnfw_log (int)(9)π¦sg_display (int)(9)π¦umsch_mm (int)(9)π¦umsch_mm_fwlog (int)(9)π¦smu_pptable_id (int)(9)π¦partition_mode (int)(9)π¦enforce_isolation (int)(9)π¦modeset (int)(9)π¦seamless (int)(9)π¦debug_mask (uint)(9)π¦agp (int)(9)π¦wbrf (int)(9)π¦rebar (int)(9)π¦user_queue (int)(9)π¦ptl (int)(9)π¦
and that's the whole vibe, bestie π this page is from the ./drivers/gpu/drm/amd/amdgpu source and it's dated October 2026 π mwah, slay, go read your kernel params 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