aboutsummaryrefslogtreecommitdiffstats
path: root/src/mesa/drivers/dri/i965/brw_vec4_cse.cpp
Commit message (Expand)AuthorAgeFilesLines
* i965/vec4: teach CSE about exec_size, group and doublesIago Toral Quiroga2017-01-031-7/+20
* i965/vec4: handle 32 and 64 bit channels in liveness analysisJuan A. Suarez Romero2017-01-031-1/+1
* i965/vec4: use byte_offset() instead of offset()Iago Toral Quiroga2016-10-271-4/+6
* i965/vec4: Replace vec4_instruction::regs_written with ::size_written field i...Francisco Jerez2016-09-141-1/+1
* i965/vec4: Add wrapper functions for vec4_instruction::regs_read and ::regs_w...Francisco Jerez2016-09-141-3/+3
* i965/vec4: Ignore swizzle of VGRF for use by var_range_end().Matt Turner2016-08-191-1/+1
* i965: Add tessellation control shaders.Kenneth Graunke2015-12-221-0/+2
* i965: Replace HW_REG with ARF/FIXED_GRF.Matt Turner2015-11-131-1/+2
* i965: Rename GRF to VGRF.Matt Turner2015-11-131-2/+2
* i965: Use brw_reg's nr field to store register number.Matt Turner2015-11-131-1/+1
* i965: Define virtual instruction to calculate the high 32 bits of a multiply.Francisco Jerez2015-08-061-0/+1
* i965/vec4: Take into account all instruction fields in CSE instructions_match().Francisco Jerez2015-06-091-0/+8
* i965: Perform basic optimizations on the FIND_LIVE_CHANNEL opcode.Francisco Jerez2015-05-041-0/+1
* i965: Perform basic optimizations on the BROADCAST opcode.Francisco Jerez2015-05-041-0/+1
* i965/vec4: Define helpers to calculate the common live interval of a range of...Francisco Jerez2015-03-231-7/+1
* i965/vec4: Define helper functions to convert a register to a variable index.Francisco Jerez2015-03-231-0/+1
* i965/vec4: Don't lose the force_writemask_all flag during CSE.Francisco Jerez2015-03-231-0/+3
* i965/vec4: Fix handling of multiple register reads and writes in opt_cse().Francisco Jerez2015-03-231-8/+17
* i965: De-duplicate is_expression_commutative() functions.Kenneth Graunke2015-03-151-23/+1
* i965: Factor out virtual GRF allocation to a separate object.Francisco Jerez2015-02-101-1/+1
* i965: Don't consider null dst instructions as matching non-null dst.Matt Turner2015-01-151-1/+2
* i965: Consider SEL.{GE,L} to be commutative operations.Matt Turner2015-01-081-6/+16
* i965/vec4: Perform CSE on MOV ..., VF instructions.Matt Turner2014-12-291-5/+11
* i965/vec4: Allow CSE on uniform-vec4 expansion MOVs.Matt Turner2014-12-051-0/+1
* i965/vec4: Perform CSE on MAD instructions with final arguments switched.Matt Turner2014-10-291-1/+5
* i965: Make instruction lists local to the bblocks.Matt Turner2014-09-241-6/+0
* i965: Remove cfg-invalidating parameter from invalidate_live_intervals.Matt Turner2014-09-241-1/+1
* i965/vec4: Only examine virtual_grf_end for GRF sourcesIan Romanick2014-09-111-8/+12
* i965: Use basic-block aware insertion/removal functions.Matt Turner2014-08-221-4/+4
* i965: Add and use foreach_block macro.Matt Turner2014-08-181-3/+1
* i965: Revert part of f5cc3fdcf1680b116612fac7c39f1bd79f5e555e.Kenneth Graunke2014-08-111-1/+1
* i965/cse: Don't eliminate instructions with side-effectsJason Ekstrand2014-08-111-1/+1
* i965/vec4: Use foreach_inst_in_block a couple more places.Matt Turner2014-07-211-4/+1
* i965: Replace cfg instances with calls to calculate_cfg().Matt Turner2014-07-211-4/+2
* i965/vec4: Invalidate live intervals in opt_cse, not _local.Matt Turner2014-07-141-3/+3
* i965/vec4: Move aeb list into opt_cse_local.Matt Turner2014-07-141-6/+6
* i965/vec4: Perform CSE on CMP(N) instructions.Matt Turner2014-07-061-1/+16
* i965/vec4: Don't emit null MOVs in CSE.Matt Turner2014-07-061-5/+7
* i965/vec4: Improve CSE performance by expiring some available expressions.Matt Turner2014-07-061-0/+20
* i965/vec4: Add basic common subexpression elimination.Kenneth Graunke2014-07-061-0/+232