index
:
~currojerez/mesa
atomic-counters
clover-1.2-headers
clover-compile-program
clover-internal-ref-counting
clover-link-program
clover-next
for-felix
for-iago
for-kayden
for-mark
for-samuel
for-valtteri
i965-const-cache
i965-discard-rework
i965-fb-fetch
i965-ir-analysis
i965-ir-builder
i965-ir-byte-units
i965-l3-partitioning
i965-late-simd-lowering
i965-oes-image-atomic
i965-packed-ra
i965-sched-discard
i965-scratch-non-coherent
i965-simd-lowering
i965-simd32-cs
i965-simd32-fs
i965-spilling-fixes
i965-ssbo
i965-unified-visitor
image-load-store
image-load-store-es31
image-load-store-flush
image-load-store-l3
image-load-store-lower
image-load-store-nir
image-load-store-scalar
image-load-store-ssbo
iris-cache-tracker
jenkins
jenkins-vk
master
mesa-next
ralloc-new
ralloc-new-less-fancy
remove-znew
Unnamed repository; edit this file to name it for gitweb.
currojerez
summary
refs
log
tree
commit
diff
log msg
author
committer
range
Age
Commit message (
Expand
)
Author
Files
Lines
2016-09-12
i965/vec4: Assert that pull constant load offsets are 16B-aligned.
i965-ir-byte-units
Francisco Jerez
1
-0
/
+1
2016-09-12
i965/vec4: Assert that ATTR regions are register-aligned.
Francisco Jerez
1
-0
/
+1
2016-09-12
i965/vec4: Don't spill non-GRF-aligned register regions.
Francisco Jerez
2
-2
/
+5
2016-09-12
i965/vec4: Fix copy propagation for non-register-aligned regions.
Francisco Jerez
1
-3
/
+6
2016-09-12
i965/vec4: Compare full register offsets in cmod propagation.
Francisco Jerez
1
-1
/
+1
2016-09-12
i965/vec4: Assign correct destination offset to rewritten instruction in regi...
Francisco Jerez
1
-2
/
+1
2016-09-12
i965/vec4: Don't coalesce registers with overlapping writes not matching the ...
Francisco Jerez
1
-4
/
+6
2016-09-12
i965/vec4: Compare full register offsets in opt_register_coalesce nop move ch...
Francisco Jerez
1
-1
/
+1
2016-09-12
i965/vec4: Check that the write offsets match when setting dependency controls.
Francisco Jerez
1
-0
/
+2
2016-09-12
i965/vec4: Change opt_vector_float to keep track of the last offset seen in b...
Francisco Jerez
1
-3
/
+3
2016-09-12
i965/vec4: Simplify src/dst_reg to brw_reg conversion by using byte_offset().
Francisco Jerez
1
-7
/
+8
2016-09-12
i965/ir: Update several stale comments.
Francisco Jerez
5
-26
/
+22
2016-09-12
i965/ir: Don't print ARF subnr values twice.
Francisco Jerez
2
-8
/
+0
2016-09-12
i965/vec4: Print src/dst_reg::offset field consistently for all register files.
Francisco Jerez
1
-6
/
+15
2016-09-12
i965/fs: Print fs_reg::offset field consistently for all register files.
Francisco Jerez
1
-16
/
+22
2016-09-12
i965/fs: Misc simplification.
Francisco Jerez
4
-6
/
+6
2016-09-12
i965/fs: Get rid of fs_inst::set_smear().
Francisco Jerez
3
-32
/
+15
2016-09-12
i965/fs: Use region_contained_in() in compute-to-mrf coalescing pass.
Francisco Jerez
1
-3
/
+2
2016-09-12
i965/fs: Move region_contained_in to the IR header and fix for non-VGRF files.
Francisco Jerez
2
-14
/
+13
2016-09-12
i965/fs: Change region_contained_in() to use byte units.
Francisco Jerez
1
-15
/
+10
2016-09-12
i965/fs: Simplify copy propagation LOAD_PAYLOAD ACP setup.
Francisco Jerez
1
-3
/
+2
2016-09-12
i965/fs: Simplify a bunch of fs_inst::size_written calculations by using comp...
Francisco Jerez
4
-15
/
+19
2016-09-12
i965/fs: Simplify result_live calculation in dead_code_eliminate().
Francisco Jerez
1
-9
/
+3
2016-09-12
i965/fs: Simplify and fix buggy stride/offset calculations using subscript().
Francisco Jerez
2
-59
/
+17
2016-09-12
i965/fs: Simplify get_fpu_lowered_simd_width() by using inequalities instead ...
Francisco Jerez
1
-2
/
+2
2016-09-12
i965/fs: Simplify byte_offset().
Francisco Jerez
1
-8
/
+4
2016-09-12
i965/fs: Fix signedness of the return value of fs_inst::size_read().
Francisco Jerez
2
-2
/
+2
2016-09-12
i965/fs: Switch mask_relative_to() used in compute-to-mrf to byte units.
Francisco Jerez
1
-10
/
+10
2016-09-12
i965/fs: Fix bogus sub-MRF offset calculation in compute-to-mrf.
Francisco Jerez
1
-6
/
+6
2016-09-12
i965/fs: Take into account copy register offset during compute-to-mrf.
Francisco Jerez
1
-1
/
+1
2016-09-12
i965/vec4: Drop backend_reg::in_range() in favor of regions_overlap().
Francisco Jerez
5
-20
/
+12
2016-09-12
i965/vec4: Port regions_overlap() to the vec4 IR.
Francisco Jerez
1
-4
/
+58
2016-09-12
i965/fs: Stop using fs_reg::in_range() in favor of regions_overlap().
Francisco Jerez
1
-1
/
+3
2016-09-12
i965/fs: Drop fs_inst::overwrites_reg() in favor of regions_overlap().
Francisco Jerez
7
-14
/
+15
2016-09-12
i965/fs: Fix LOAD_PAYLOAD handling in register coalesce is_nop_mov().
Francisco Jerez
1
-1
/
+3
2016-09-12
i965/fs: Fix can_propagate_from() source/destination overlap check.
Francisco Jerez
1
-2
/
+2
2016-09-12
i965/fs: Compare full register offsets in cmod propagation pass.
Francisco Jerez
1
-2
/
+1
2016-09-12
i965/fs: Don't consider LOAD_PAYLOAD with stride > 1 source to behave like a ...
Francisco Jerez
1
-1
/
+1
2016-09-12
i965/fs: Don't consider LOAD_PAYLOAD with sub-GRF offset to behave like a raw...
Francisco Jerez
1
-1
/
+1
2016-09-12
i965/vec4: Take into account misalignment in regs_written() and regs_read().
Francisco Jerez
1
-4
/
+6
2016-09-12
i965/fs: Take into account misalignment in regs_written() and regs_read().
Francisco Jerez
2
-29
/
+5
2016-09-12
i965/fs: Take into account trailing padding in regs_written() and regs_read().
Francisco Jerez
1
-2
/
+20
2016-09-12
i965/fs: Handle fixed HW GRF subnr in reg_offset().
Francisco Jerez
1
-1
/
+2
2016-09-12
i965/fs: Handle arbitrary offsets in brw_reg_from_fs_reg for MRF/VGRF registers.
Francisco Jerez
1
-3
/
+2
2016-09-12
i965/fs: Return more accurate read size for LINTERP from fs_inst::size_read.
Francisco Jerez
1
-1
/
+1
2016-09-12
i965/fs: Return more accurate read size from fs_inst::size_read for IMM and U...
Francisco Jerez
1
-1
/
+1
2016-09-12
i965/vec4: Replace vec4_instruction::regs_read with ::size_read using byte un...
Francisco Jerez
3
-13
/
+25
2016-09-12
i965/fs: Replace fs_inst::regs_read with ::size_read using byte units.
Francisco Jerez
3
-32
/
+34
2016-09-12
i965/ir: Drop backend_instruction::regs_written field.
Francisco Jerez
1
-1
/
+0
2016-09-12
i965/vec4: Replace vec4_instruction::regs_written with ::size_written field i...
Francisco Jerez
9
-11
/
+12
[next]