Commit History

Author SHA1 Message Date
  Jeff Bolz 2711d0215f vulkan: Catch pipeline creation failure and print an error message (#11436) 1 year ago
  Jeff Bolz 4a75d19376 vulkan: compile shaders on-demand (#11406) 1 year ago
  amd-dwang 955a6c2d91 Vulkan-run-test: fix mmq_wg_denoms (#11343) 1 year ago
  Jeff Bolz 5245729e33 vulkan: fix diag_mask_inf (#11323) 1 year ago
  Jeff Bolz aea8ddd516 vulkan: fix coopmat2 validation failures (#11284) 1 year ago
  Jeff Bolz 44e18ef939 vulkan: fix coopmat2 flash attention for non-contiguous inputs (#11281) 1 year ago
  Jeff Bolz bd38ddea01 vulkan: support copy from f32 to q4_0/q4_1/q5_0/q5_1/q8_0/iq4_nl (#11166) 1 year ago
  0cc4m c3f9d25706 Vulkan: Fix float16 use on devices without float16 support + fix subgroup_size_control validation error (#11161) 1 year ago
  Molly Sophia ee7136c6d1 llama: add support for QRWKV6 model architecture (#11001) 1 year ago
  Mathieu Baudier 02f0430141 Disable GL_KHR_cooperative_matrix Vulkan extension if not available. (#11117) 1 year ago
  0cc4m b56f079e28 Vulkan: Add device-specific blacklist for coopmat for the AMD proprietary driver (#11074) 1 year ago
  Jeff Bolz 716bd6dec3 vulkan: optimize mul_mat for small values of N (#10991) 1 year ago
  Jeff Bolz a813badbbd vulkan: im2col and matmul optimizations for stable diffusion (#10942) 1 year ago
  Jeff Bolz fdd2188912 vulkan: Use push constant offset to handle misaligned descriptors (#10987) 1 year ago
  Eve d79d8f39b4 vulkan: multi-row k quants (#10846) 1 year ago
  Jeff Bolz ebdee9478c vulkan: build fixes for 32b (#10927) 1 year ago
  Eve 7b1ec53f56 vulkan: bugfixes for small subgroup size systems + llvmpipe test (#10809) 1 year ago
  Zhiyuan Li 160bc039c8 rwkv6: add wkv6 support for Vulkan backend (#10829) 1 year ago
  HimariO ba1cb19cdd llama : add Qwen2VL support + multimodal RoPE (#10361) 1 year ago
  Eve 64ae065511 vulkan: small mul_mat_vec optimizations (#10665) 1 year ago
  0cc4m dc5301d565 Vulkan: Add VK_EXT_subgroup_size_control support to ensure full subgroups for coopmats (#10721) 1 year ago
  Jeff Bolz b685daf386 vulkan: request round-to-even for fp16 in im2col/rope_head (#10767) 1 year ago
  Eve dafae66cc2 vulkan: dynamic subgroup size for the remaining k quants (#10745) 1 year ago
  Jeff Bolz a05e2afcc2 vulkan: disable spirv-opt for coopmat shaders (#10763) 1 year ago
  Jeff Bolz 3d98b4cb22 vulkan: fix compile warnings (#10731) 1 year ago
  Jeff Bolz ecc93d0558 vulkan: compile a test shader in cmake to check for coopmat2 support (#10713) 1 year ago
  0cc4m 3df784b305 Vulkan: VK_KHR_cooperative_matrix support to speed up prompt processing (#10597) 1 year ago
  Jeff Bolz c9c6e01dae vulkan: Add VK_NV_cooperative_matrix2 support for mul_mat and flash attention (#10206) 1 year ago
  Jeff Bolz 2759916d86 vulkan: Implement "fast divide" (mul+shift) for unary ops like copy (#10642) 1 year ago
  Jeff Bolz cc98896db8 vulkan: optimize and reenable split_k (#10637) 1 year ago