Lines Matching +full:render +full:- +full:max
1 /* amdgpu_drm.h -- Public header for the amdgpu driver -*- linux-c -*-
82 * GPU's virtual address space via gart. Gart memory linearizes non-contiguous
89 * %AMDGPU_GEM_DOMAIN_GDS Global on-chip data storage used to share data
166 /* Flag that BO should be coherent across devices when using device-level
174 /* Set PTE.D and recompress during GTT->VRAM moves according to TILING flags. */
267 #define AMDGPU_CTX_PRIORITY_UNSET -2048
268 #define AMDGPU_CTX_PRIORITY_VERY_LOW -1023
269 #define AMDGPU_CTX_PRIORITY_LOW -512
378 /* SI-CI-VI: */
397 /* GFX9 - GFX11: */
480 /** BO status: 0 - BO is idle, 1 - BO is busy */
506 /** CS status: 0 - CS completed, 1 - CS still busy */
573 /* Default MTYPE. Pre-AI must use this. Recommended for newer ASICs. */
612 * From VCN4, AMDGPU_HW_IP_VCN_ENC is re-used to support
929 /* Query the max number of IBs per gang per submission */
1032 /** max. physical memory */
1035 /** Theoretical max. available memory in the given heap */
1047 * Theoretical possible max. size of buffer which
1107 /** Render backend pipe mask. One render backend is CB+DB. */
1120 /** Page table entry - fragment size */
1157 /* max gs wavefront per vgt*/
1210 /** Max handles as supported by firmware for UVD */
1296 * Conversion matrix with 3x4 dimensions in S31.32 sign-magnitude