mirror of
https://github.com/tinygrad/tinygrad.git
synced 2026-06-24 02:14:17 +00:00
Compare commits
merge into: mirrors:master
mirrors:master
mirrors:gh-pages
mirrors:update_benchmark
mirrors:codegen_try_2
mirrors:codegen2
mirrors:clean_load
mirrors:remove_define_var
mirrors:stack_0
mirrors:port_nir
mirrors:more_ren_clean
mirrors:shrink_in_render
mirrors:python_prg
mirrors:slice_like_shrink
mirrors:fix_slice_store
mirrors:ps_clean
mirrors:index_to_slice
mirrors:all_bv_no_idx
mirrors:no_count_renderer
mirrors:minigen
mirrors:qcom_mmapeak
mirrors:broadcast_shape_2
mirrors:novecimg
mirrors:earlier_gater
mirrors:fix_src_spec
mirrors:mlperf_training_v6.0
mirrors:image_no_vec
mirrors:move_gates_to_load_store
mirrors:gate-on-load-store
mirrors:dtype_shape_2
mirrors:image_refactors_stack
mirrors:image_idx_xy
mirrors:contig_on_after
mirrors:remove_ck_contig
mirrors:dtype_shape
mirrors:renderers_use_shapes_2
mirrors:remove_vec_3
mirrors:remove_vec_2
mirrors:new_expander
mirrors:olmo3_support
mirrors:param_call_is_function
mirrors:move_schedule
mirrors:gemma4_gpt
mirrors:abstractions_4
mirrors:x86_moves
mirrors:rdna4_gemm
mirrors:fancy_usb
mirrors:deepsek
mirrors:precompile_backward
mirrors:tuplegtuple
mirrors:more_uop_mm
mirrors:assign_try_3
mirrors:no_assign_2
mirrors:ranged_call
mirrors:default_late_allreduce
mirrors:cat_mop
mirrors:only_assign_buffers
mirrors:fix_fa_fwd_clang_22
mirrors:late_allreduce
mirrors:sym_llm
mirrors:sym_work
mirrors:callify
mirrors:more_folding_rules
mirrors:assign_copy
mirrors:viz_cbroadcast
mirrors:mixins7
mirrors:remove_cast_hack
mirrors:revert-14734-tload
mirrors:amd_isel
mirrors:emu_cdna
mirrors:no_tuplize
mirrors:all_sink
mirrors:call_inline
mirrors:fun_w_egraphs
mirrors:call_is_kernels
mirrors:kernel_is_call
mirrors:new_ck
mirrors:test_unit
mirrors:mac_pytest
mirrors:user_failure
mirrors:test_amd_crash
mirrors:pyremu_try
mirrors:glm_flash
mirrors:amd_refac
mirrors:locals_have_device
mirrors:lds_double_buffering
mirrors:amd_sqtt
mirrors:rdna4_asm
mirrors:more_pcode
mirrors:gen_pdf_fast
mirrors:only_reg_emu2
mirrors:only_reg_emu
mirrors:rdna3_vibes
mirrors:hwtest_fixes
mirrors:more_early_comps
mirrors:remove_programspec
mirrors:one_compiler
mirrors:typed_checks
mirrors:execution_unit
mirrors:fix_llvm_cast
mirrors:pad_mixin
mirrors:symbolic-chunk-split
mirrors:more_apps
mirrors:q4k_support
mirrors:fast_llm
mirrors:heur_group
mirrors:llm_qwen
mirrors:revert-13583-rocr_desc_2
mirrors:sched_cache
mirrors:gate_rewrite
mirrors:llama_trainer
mirrors:topoprop
mirrors:cache_reshape
mirrors:bt_ei
mirrors:python_speed
mirrors:string_ranges
mirrors:scan_assign
mirrors:more_vmap
mirrors:scan_op_8
mirrors:print_special_ops
mirrors:revert-13311-prof_link_sqtt
mirrors:parse_sqtt_more
mirrors:grad_outerworld
mirrors:sched_lin
mirrors:min_outer_range
mirrors:tiny_scan
mirrors:remove_assign
mirrors:algebraic_upat
mirrors:outer_range
mirrors:simple_pool
mirrors:er_prioity
mirrors:simple_priority
mirrors:no_sip_usbgpu
mirrors:self_type
mirrors:reshape_trait
mirrors:revert-13046-better_cleanup_arange_buffers
mirrors:amd_uop_cleanups
mirrors:amd_uop
mirrors:uop_prg
mirrors:late_add_load
mirrors:demote_op
mirrors:replace_if_with_range
mirrors:if_range_cleanups
mirrors:better_spec
mirrors:ignore_oob
mirrors:simp_spec
mirrors:spec_cleanup
mirrors:real_pyrender
mirrors:no_late_simplify_marg
mirrors:spec_3
mirrors:relu_pattern
mirrors:full_spec
mirrors:kitten_matmul
mirrors:late_Locals
mirrors:simpler_end
mirrors:no_merge_ends
mirrors:clean_spec
mirrors:multioutput
mirrors:num_batches_tracked_shape
mirrors:revert-12813-after_cleanups
mirrors:after_qol
mirrors:two_stage_remove
mirrors:test_fa
mirrors:no_decimals
mirrors:multi_range
mirrors:revert-12731-jitlink_init_check
mirrors:sd_load_simple
mirrors:new_shape
mirrors:remove_slow_tests
mirrors:remove_forced_re
mirrors:remove_pylint
mirrors:outerworld_work
mirrors:mt_typing
mirrors:typing_work
mirrors:min_new_lin
mirrors:ttomsa-new_linearizer
mirrors:new_linearizer
mirrors:group_div_rules
mirrors:more_uop_gc_test
mirrors:remove_on_stack
mirrors:end_all_ranges
mirrors:no_tsink_base
mirrors:delete_slow_rangeify
mirrors:rtoposort
mirrors:clone_tg
mirrors:more_rangeify_pm_tests
mirrors:spec_ren
mirrors:fix_rc
mirrors:add_local_buffer
mirrors:split_sub
mirrors:reprocess_node_experiments
mirrors:recurse_sub
mirrors:support_opts_in_contig
mirrors:lil_dg_change
mirrors:no_one_ranges
mirrors:fix_rangeify_tests
mirrors:fix_range_merging_sd
mirrors:buf_heur
mirrors:revert-12308-bufferize_early
mirrors:qcom_830
mirrors:set_testpath
mirrors:rangeify_const_folding
mirrors:check_clsetarg
mirrors:failing_assign
mirrors:revert-12104-simplify_valid_in_symbolic
mirrors:one_hcopt
mirrors:fix_sched_copy
mirrors:heur_4096
mirrors:delete_kernel_py
mirrors:emulate_ctx
mirrors:tc_is_heur
mirrors:test_from_postopt
mirrors:working_postopt
mirrors:simpler_postrange
mirrors:postrange_hcopts
mirrors:move_expander
mirrors:revert-11876-memplan_vbig
mirrors:qol_string
mirrors:split_buf_idx
mirrors:simple_gidxs
mirrors:late_codegen
mirrors:revert-11811-better_div_nesting
mirrors:revert-11809-no_ki_in_gpudims
mirrors:gl_dims
mirrors:r_new_opt
mirrors:rangeify_2
mirrors:r_cleanups
mirrors:no_buf_0
mirrors:disable_rules
mirrors:upd_test4
mirrors:bump_nv_2404
mirrors:revert-11715-revert-11711-amd_llvm_tc_cast
mirrors:better_syn
mirrors:new_test
mirrors:lil_rangeify_td
mirrors:no_merge_views
mirrors:changes_from_rangify
mirrors:cast_vec_as_cast
mirrors:split_decomp_pass
mirrors:simpler_fusion
mirrors:revert-11547-fix_mismatch_red
mirrors:view_after_const
mirrors:tmr
mirrors:load_view_late
mirrors:minor_views
mirrors:view_in_codegen
mirrors:fixup_fix_kernel
mirrors:swizzle_in_kernel
mirrors:late_view
mirrors:moveleftright
mirrors:fix_fuse
mirrors:kernelless
mirrors:mega_lowerer
mirrors:local_cache
mirrors:check_dims
mirrors:gldims
mirrors:kernel4
mirrors:move_simp_views
mirrors:lowerer_hacks
mirrors:store_is_void
mirrors:unused_permute_arg
mirrors:simpler_reduce
mirrors:less_wmma_arg
mirrors:delete_ones
mirrors:id_store
mirrors:before2
mirrors:continue_mat
mirrors:manual_kernel_3
mirrors:endrange
mirrors:warp_fun
mirrors:update_benchmark_search
mirrors:update_mlperf
mirrors:resnet-ci
mirrors:mlperf_training_v5.0
mirrors:dsp_search_merged
mirrors:dsp_search
mirrors:verify_sink_dims_clean
mirrors:revert-9438-simple_matmul_np_rand
mirrors:simpler-sd-compile
mirrors:wgpu-f16-packed
mirrors:mlperf_training_v4.1
mirrors:v0.13.0
mirrors:v0.12.0
mirrors:v0.11.0
mirrors:v0.10.3
mirrors:v0.10.2
mirrors:v0.10.1
mirrors:v0.10.0
mirrors:v0.9.2
mirrors:v0.9.1
mirrors:v0.9.0
mirrors:v0.8.0
mirrors:v0.7.0
mirrors:v0.6.0
mirrors:v0.5.0
mirrors:v0.4.0
...
pull from: mirrors:pyremu_try
mirrors:gh-pages
mirrors:master
mirrors:update_benchmark
mirrors:codegen_try_2
mirrors:codegen2
mirrors:clean_load
mirrors:remove_define_var
mirrors:stack_0
mirrors:port_nir
mirrors:more_ren_clean
mirrors:shrink_in_render
mirrors:python_prg
mirrors:slice_like_shrink
mirrors:fix_slice_store
mirrors:ps_clean
mirrors:index_to_slice
mirrors:all_bv_no_idx
mirrors:no_count_renderer
mirrors:minigen
mirrors:qcom_mmapeak
mirrors:broadcast_shape_2
mirrors:novecimg
mirrors:earlier_gater
mirrors:fix_src_spec
mirrors:mlperf_training_v6.0
mirrors:image_no_vec
mirrors:move_gates_to_load_store
mirrors:gate-on-load-store
mirrors:dtype_shape_2
mirrors:image_refactors_stack
mirrors:image_idx_xy
mirrors:contig_on_after
mirrors:remove_ck_contig
mirrors:dtype_shape
mirrors:renderers_use_shapes_2
mirrors:remove_vec_3
mirrors:remove_vec_2
mirrors:new_expander
mirrors:olmo3_support
mirrors:param_call_is_function
mirrors:move_schedule
mirrors:gemma4_gpt
mirrors:abstractions_4
mirrors:x86_moves
mirrors:rdna4_gemm
mirrors:fancy_usb
mirrors:deepsek
mirrors:precompile_backward
mirrors:tuplegtuple
mirrors:more_uop_mm
mirrors:assign_try_3
mirrors:no_assign_2
mirrors:ranged_call
mirrors:default_late_allreduce
mirrors:cat_mop
mirrors:only_assign_buffers
mirrors:fix_fa_fwd_clang_22
mirrors:late_allreduce
mirrors:sym_llm
mirrors:sym_work
mirrors:callify
mirrors:more_folding_rules
mirrors:assign_copy
mirrors:viz_cbroadcast
mirrors:mixins7
mirrors:remove_cast_hack
mirrors:revert-14734-tload
mirrors:amd_isel
mirrors:emu_cdna
mirrors:no_tuplize
mirrors:all_sink
mirrors:call_inline
mirrors:fun_w_egraphs
mirrors:call_is_kernels
mirrors:kernel_is_call
mirrors:new_ck
mirrors:test_unit
mirrors:mac_pytest
mirrors:user_failure
mirrors:test_amd_crash
mirrors:pyremu_try
mirrors:glm_flash
mirrors:amd_refac
mirrors:locals_have_device
mirrors:lds_double_buffering
mirrors:amd_sqtt
mirrors:rdna4_asm
mirrors:more_pcode
mirrors:gen_pdf_fast
mirrors:only_reg_emu2
mirrors:only_reg_emu
mirrors:rdna3_vibes
mirrors:hwtest_fixes
mirrors:more_early_comps
mirrors:remove_programspec
mirrors:one_compiler
mirrors:typed_checks
mirrors:execution_unit
mirrors:fix_llvm_cast
mirrors:pad_mixin
mirrors:symbolic-chunk-split
mirrors:more_apps
mirrors:q4k_support
mirrors:fast_llm
mirrors:heur_group
mirrors:llm_qwen
mirrors:revert-13583-rocr_desc_2
mirrors:sched_cache
mirrors:gate_rewrite
mirrors:llama_trainer
mirrors:topoprop
mirrors:cache_reshape
mirrors:bt_ei
mirrors:python_speed
mirrors:string_ranges
mirrors:scan_assign
mirrors:more_vmap
mirrors:scan_op_8
mirrors:print_special_ops
mirrors:revert-13311-prof_link_sqtt
mirrors:parse_sqtt_more
mirrors:grad_outerworld
mirrors:sched_lin
mirrors:min_outer_range
mirrors:tiny_scan
mirrors:remove_assign
mirrors:algebraic_upat
mirrors:outer_range
mirrors:simple_pool
mirrors:er_prioity
mirrors:simple_priority
mirrors:no_sip_usbgpu
mirrors:self_type
mirrors:reshape_trait
mirrors:revert-13046-better_cleanup_arange_buffers
mirrors:amd_uop_cleanups
mirrors:amd_uop
mirrors:uop_prg
mirrors:late_add_load
mirrors:demote_op
mirrors:replace_if_with_range
mirrors:if_range_cleanups
mirrors:better_spec
mirrors:ignore_oob
mirrors:simp_spec
mirrors:spec_cleanup
mirrors:real_pyrender
mirrors:no_late_simplify_marg
mirrors:spec_3
mirrors:relu_pattern
mirrors:full_spec
mirrors:kitten_matmul
mirrors:late_Locals
mirrors:simpler_end
mirrors:no_merge_ends
mirrors:clean_spec
mirrors:multioutput
mirrors:num_batches_tracked_shape
mirrors:revert-12813-after_cleanups
mirrors:after_qol
mirrors:two_stage_remove
mirrors:test_fa
mirrors:no_decimals
mirrors:multi_range
mirrors:revert-12731-jitlink_init_check
mirrors:sd_load_simple
mirrors:new_shape
mirrors:remove_slow_tests
mirrors:remove_forced_re
mirrors:remove_pylint
mirrors:outerworld_work
mirrors:mt_typing
mirrors:typing_work
mirrors:min_new_lin
mirrors:ttomsa-new_linearizer
mirrors:new_linearizer
mirrors:group_div_rules
mirrors:more_uop_gc_test
mirrors:remove_on_stack
mirrors:end_all_ranges
mirrors:no_tsink_base
mirrors:delete_slow_rangeify
mirrors:rtoposort
mirrors:clone_tg
mirrors:more_rangeify_pm_tests
mirrors:spec_ren
mirrors:fix_rc
mirrors:add_local_buffer
mirrors:split_sub
mirrors:reprocess_node_experiments
mirrors:recurse_sub
mirrors:support_opts_in_contig
mirrors:lil_dg_change
mirrors:no_one_ranges
mirrors:fix_rangeify_tests
mirrors:fix_range_merging_sd
mirrors:buf_heur
mirrors:revert-12308-bufferize_early
mirrors:qcom_830
mirrors:set_testpath
mirrors:rangeify_const_folding
mirrors:check_clsetarg
mirrors:failing_assign
mirrors:revert-12104-simplify_valid_in_symbolic
mirrors:one_hcopt
mirrors:fix_sched_copy
mirrors:heur_4096
mirrors:delete_kernel_py
mirrors:emulate_ctx
mirrors:tc_is_heur
mirrors:test_from_postopt
mirrors:working_postopt
mirrors:simpler_postrange
mirrors:postrange_hcopts
mirrors:move_expander
mirrors:revert-11876-memplan_vbig
mirrors:qol_string
mirrors:split_buf_idx
mirrors:simple_gidxs
mirrors:late_codegen
mirrors:revert-11811-better_div_nesting
mirrors:revert-11809-no_ki_in_gpudims
mirrors:gl_dims
mirrors:r_new_opt
mirrors:rangeify_2
mirrors:r_cleanups
mirrors:no_buf_0
mirrors:disable_rules
mirrors:upd_test4
mirrors:bump_nv_2404
mirrors:revert-11715-revert-11711-amd_llvm_tc_cast
mirrors:better_syn
mirrors:new_test
mirrors:lil_rangeify_td
mirrors:no_merge_views
mirrors:changes_from_rangify
mirrors:cast_vec_as_cast
mirrors:split_decomp_pass
mirrors:simpler_fusion
mirrors:revert-11547-fix_mismatch_red
mirrors:view_after_const
mirrors:tmr
mirrors:load_view_late
mirrors:minor_views
mirrors:view_in_codegen
mirrors:fixup_fix_kernel
mirrors:swizzle_in_kernel
mirrors:late_view
mirrors:moveleftright
mirrors:fix_fuse
mirrors:kernelless
mirrors:mega_lowerer
mirrors:local_cache
mirrors:check_dims
mirrors:gldims
mirrors:kernel4
mirrors:move_simp_views
mirrors:lowerer_hacks
mirrors:store_is_void
mirrors:unused_permute_arg
mirrors:simpler_reduce
mirrors:less_wmma_arg
mirrors:delete_ones
mirrors:id_store
mirrors:before2
mirrors:continue_mat
mirrors:manual_kernel_3
mirrors:endrange
mirrors:warp_fun
mirrors:update_benchmark_search
mirrors:update_mlperf
mirrors:resnet-ci
mirrors:mlperf_training_v5.0
mirrors:dsp_search_merged
mirrors:dsp_search
mirrors:verify_sink_dims_clean
mirrors:revert-9438-simple_matmul_np_rand
mirrors:simpler-sd-compile
mirrors:wgpu-f16-packed
mirrors:mlperf_training_v4.1
mirrors:v0.13.0
mirrors:v0.12.0
mirrors:v0.11.0
mirrors:v0.10.3
mirrors:v0.10.2
mirrors:v0.10.1
mirrors:v0.10.0
mirrors:v0.9.2
mirrors:v0.9.1
mirrors:v0.9.0
mirrors:v0.8.0
mirrors:v0.7.0
mirrors:v0.6.0
mirrors:v0.5.0
mirrors:v0.4.0
3 commits
master
...
pyremu_try
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
|
|
ef3c614b4d |
Merge branch 'master' into pyremu_try | ||
|
|
d8d2ad92e7 |
Merge branch 'master' into pyremu_try | ||
|
|
b7031aa381 | assembly/amd: set PYTHON_REMU=1 |