summaryrefslogtreecommitdiffstats
path: root/src/mesa/drivers/dri/i965/brw_schedule_instructions.cpp
Commit message (Expand)AuthorAgeFilesLines
* i965/ir: Update several stale comments.Francisco Jerez2016-09-141-4/+4
* i965/fs: Misc simplification.Francisco Jerez2016-09-141-2/+2
* i965/fs: Replace fs_inst::regs_written with ::size_written field in bytes.Francisco Jerez2016-09-141-3/+3
* i965/vec4: Add wrapper functions for vec4_instruction::regs_read and ::regs_w...Francisco Jerez2016-09-141-4/+4
* i965/fs: Add wrapper functions for fs_inst::regs_read and ::regs_written.Francisco Jerez2016-09-141-15/+15
* i965/fs: Replace fs_reg::reg_offset with fs_reg::offset expressed in bytes.Francisco Jerez2016-09-141-5/+10
* intel: s/brw_device_info/gen_device_info/Jason Ekstrand2016-09-031-2/+2
* i965/fs: Remove special casing of framebuffer writes in scheduler code.Francisco Jerez2016-08-251-2/+1
* i965/sched: Simplify work done by add_barrier_deps().Matt Turner2016-08-191-0/+9
* i965/sched: Change the scheduling heuristics to favor early program termination.Francisco Jerez2016-08-181-3/+16
* i965/sched: Assign a preferred exit node to each node of the dependency graph.Francisco Jerez2016-08-181-0/+59
* i965/sched: Calculate the critical path of scheduling nodes non-recursively.Francisco Jerez2016-08-181-13/+12
* i965/fs: Keep track of flag dependencies with byte granularity during schedul...Francisco Jerez2016-05-271-10/+31
* i965/fs: Add missing get_latency_gen7() cases for the Gen7 pull constant opco...Francisco Jerez2016-05-271-0/+2
* i965/fs: Rename Gen4 physical varying pull constant load opcode.Francisco Jerez2016-05-271-1/+1
* i965: Add infrastucture for sample lod-zero operations.Matt Turner2016-05-191-0/+2
* i965: Don't add barrier deps for FB write messages.Matt Turner2016-03-301-1/+2
* i965: Add and use is_scheduling_barrier() function.Matt Turner2016-03-301-4/+17
* i965: Remove NOP insertion kludge in scheduler.Matt Turner2016-03-301-20/+5
* i965: Relax restriction on scheduling last instruction.Matt Turner2016-03-301-20/+3
* Revert "i965: Don't add barrier deps for FB write messages."Matt Turner2016-03-301-4/+3
* i965: Simplify full scheduling-barrier conditions.Matt Turner2016-03-301-27/+8
* i965: Remove incorrect cycle estimates.Matt Turner2016-03-301-10/+0
* i965: Use foreach_in_list_reverse_safe() macro.Matt Turner2016-03-121-12/+2
* i965: Don't add barrier deps for FB write messages.Kenneth Graunke2016-02-081-3/+4
* i965: Clean up #includes in the compiler.Matt Turner2015-11-241-2/+0
* i965: Replace HW_REG with ARF/FIXED_GRF.Matt Turner2015-11-131-38/+15
* i965: Rename GRF to VGRF.Matt Turner2015-11-131-13/+13
* i965: Use brw_reg's nr field to store register number.Matt Turner2015-11-131-31/+31
* i965: Remove fixed_hw_reg field from backend_reg.Matt Turner2015-11-131-25/+25
* i965/fs: Use regs_read/written for post-RA scheduling in calculate_depsJason Ekstrand2015-11-071-11/+4
* i965/sched: don't calculate live intervals for post-RA schedulingConnor Abbott2015-10-311-1/+2
* i965/sched: use liveness analysis for computing register pressureConnor Abbott2015-10-301-56/+229
* i965: dump scheduling cycle estimatesConnor Abbott2015-10-301-0/+20
* i965/sched: write-after-read dependencies are freeConnor Abbott2015-10-301-5/+5
* i965: fix cycle estimates when there's a pipeline stallConnor Abbott2015-10-301-7/+8
* i965: Test fixed_hw_reg.file against BRW_IMMEDIATE_VALUE, not IMM.Matt Turner2015-10-291-4/+4
* nir: remove dependency on glslRob Clark2015-10-161-1/+1
* i965: Turn BRW_MAX_MRF into a macro that accepts a hardware generationIago Toral Quiroga2015-09-211-2/+2
* i965/fs: Use exec_size instead of dst.width for computing component sizeJason Ekstrand2015-06-301-2/+2
* i965: Rename backend_visitor to backend_shaderJason Ekstrand2015-05-281-9/+9
* i965: Add typed surface access opcodes.Francisco Jerez2015-05-041-0/+3
* i965: Add untyped surface write opcode.Francisco Jerez2015-05-041-0/+1
* i965: Use device_info instead of the context in instruction schedulingJason Ekstrand2015-04-221-11/+10
* i965: Make scheduler cycle estimates use the proper stage name.Kenneth Graunke2015-02-191-5/+6
* i965/vec4: Don't infer MRF dependencies for send from GRF instructions.Francisco Jerez2015-02-101-14/+18
* i965/vec4: Fix the scheduler to take into account reads and writes of multipl...Francisco Jerez2015-02-101-5/+10
* i965/fs: Remove dependency of fs_inst on the visitor class.Francisco Jerez2015-02-101-4/+4
* i965: Factor out virtual GRF allocation to a separate object.Francisco Jerez2015-02-101-5/+5
* i965: Don't make instructions with a null dest a barrier to scheduling.Matt Turner2015-01-231-4/+8
generated by cgit v1.1 at 2024-05-15 10:39:46 +0000