| Commit message (Expand) | Author | Age | Files | Lines |
... | |
* | nvc0/ir: skip back edges when determining latest sched value | Christoph Bumiller | 2013-03-12 | 1 | -3/+4 |
* | nvc0/ir: use large issue delay after RET, too | Christoph Bumiller | 2013-03-12 | 1 | -1/+1 |
* | nv50/ir: fix size adjustment for sched info for multiple functions | Christoph Bumiller | 2013-03-12 | 1 | -6/+11 |
* | nv50/ir: print function inputs and outputs | Christoph Bumiller | 2013-03-12 | 1 | -1/+22 |
* | nv50/ir/ssa: add a few comments regarding RenamePass | Christoph Bumiller | 2013-03-12 | 1 | -0/+19 |
* | nv50/ir/tgsi: Exclude local declarations from function prototypes. | Francisco Jerez | 2013-03-12 | 1 | -5/+28 |
* | nv50/ir/opt: try to make use of SUCLAMP addend | Christoph Bumiller | 2013-03-12 | 1 | -0/+45 |
* | nv50/ir: don't assert on type in Modifier.applyTo if it is 0 | Christoph Bumiller | 2013-03-12 | 1 | -0/+2 |
* | nv50/ir: add support for barriers | Christoph Bumiller | 2013-03-12 | 7 | -15/+161 |
* | nv50/ir/tgsi: add support for atomics | Christoph Bumiller | 2013-03-12 | 1 | -0/+89 |
* | nv50/ir/tgsi: handle TGSI_OPCODE_LOAD,STORE | Christoph Bumiller | 2013-03-12 | 7 | -30/+303 |
* | nvc0/ir: don't replace load from input in COMPUTE progs with VFETCH | Christoph Bumiller | 2013-03-12 | 1 | -2/+7 |
* | nvc0/ir: implement lowering of surface ops for nve4 | Christoph Bumiller | 2013-03-12 | 8 | -16/+429 |
* | nvc0/ir: add formatted surface load lib code, move to extra header | Christoph Bumiller | 2013-03-12 | 6 | -149/+1309 |
* | nv50/ir: extend moveSources for delta < 0 | Christoph Bumiller | 2013-03-12 | 2 | -16/+31 |
* | nvc0/ir: lower atomics in s[] | Christoph Bumiller | 2013-03-12 | 1 | -0/+33 |
* | nvc0/ir/emit: implement INSBF, EXTBF, PERMT and ATOM | Christoph Bumiller | 2013-03-12 | 2 | -1/+133 |
* | nv50/ir/emit: handle OP_ATOM | Christoph Bumiller | 2013-03-12 | 1 | -0/+41 |
* | nvc0/ir/target: some ops can't be predicated, e.g. CALL | Christoph Bumiller | 2013-03-12 | 1 | -0/+8 |
* | nv50/ir/opt: CALLs cannot load | Christoph Bumiller | 2013-03-12 | 1 | -0/+3 |
* | nv50/ir: add support for indirect BRA,CALL | Christoph Bumiller | 2013-03-12 | 5 | -6/+29 |
* | nvc0/ir/emit: implement move to and logic ops on predicates | Christoph Bumiller | 2013-03-12 | 1 | -0/+45 |
* | nvc0/ir/emit: implement surface related ops | Christoph Bumiller | 2013-03-12 | 2 | -0/+301 |
* | nv50/ir: initialize CodeEmitters' specialized target fields | Christoph Bumiller | 2013-03-12 | 3 | -9/+10 |
* | nv50/ir/opt: make optimization aware of atomics, barriers, surface ops | Christoph Bumiller | 2013-03-12 | 2 | -1/+28 |
* | nv50/ir: add various new OPs that will be needed for compute | Christoph Bumiller | 2013-03-12 | 9 | -48/+179 |
* | nv50/ir: Rename "mkLoad" to "mkLoadv" for consistency. | Francisco Jerez | 2013-03-12 | 4 | -12/+21 |
* | nv50/ir: fix comparison of system values | Christoph Bumiller | 2013-03-12 | 1 | -0/+3 |
* | nv50/ir/tgsi: Translate grid-related system parameters. | Francisco Jerez | 2013-03-12 | 1 | -0/+4 |
* | nv50/ir/tgsi: Accept COMPUTE programs. | Francisco Jerez | 2013-03-12 | 1 | -0/+1 |
* | nv50/ir/ra: make sure all used function inputs get assigned a reg | Christoph Bumiller | 2013-03-12 | 1 | -0/+7 |
* | nv50/ir/ra: also add pre-existing MERGE,SPLIT to constraint list | Christoph Bumiller | 2013-03-12 | 1 | -1/+3 |
* | nv50/ir/ra: fix confusion with conditional RegisterSet::occupy | Christoph Bumiller | 2013-03-12 | 2 | -12/+32 |
* | nv50/ir/ra: swap copyCompound args if src is compound and dst isn't | Christoph Bumiller | 2013-03-12 | 1 | -0/+9 |
* | nv50/ir/ra: Fix maxGPR calculation for programs with multiple functions. | Francisco Jerez | 2013-03-12 | 1 | -1/+1 |
* | nv50/ir/ra: Fix traversal before the beginning of the active list in buildRIG. | Francisco Jerez | 2013-03-12 | 1 | -6/+5 |
* | nv50/ir/ra: Fix RegisterSet::occupy(const Value *v). | Francisco Jerez | 2013-03-12 | 1 | -1/+1 |
* | nv50/ir/ra: Fix argument const-ness in RegisterSet::idToUnits and idToBytes | Francisco Jerez | 2013-03-12 | 1 | -2/+2 |
* | nv50/ir/opt: Fix tryPropagateBranch for BBs with several exit branches. | Francisco Jerez | 2013-03-12 | 1 | -28/+32 |
* | nv50/ir: Clean up references to function values before destroying them. | Francisco Jerez | 2013-03-12 | 1 | -0/+4 |
* | nouveau: Bail out from nouveau_fence_wait if flushing the pushbuf fails. | Francisco Jerez | 2013-03-12 | 1 | -2/+4 |
* | mesa: Use correct functions for enum conversion. | Vinson Lee | 2013-03-11 | 1 | -2/+2 |
* | freedreno: gallium driver for adreno | Rob Clark | 2013-03-11 | 49 | -1/+9973 |
* | d3d1x: Remove. | José Fonseca | 2013-03-12 | 137 | -27307/+1 |
* | nv50: Remove nv0_ir_from_sm4.* | José Fonseca | 2013-03-12 | 2 | -2512/+0 |
* | gallivm: clean up passing derivatives around | Roland Scheidegger | 2013-03-12 | 6 | -249/+196 |
* | i965: Fix typo in doxygen hyperlink | Chad Versace | 2013-03-11 | 1 | -1/+1 |
* | mesa: Reduce memory usage for reg alloc with many graph nodes (part 2). | Eric Anholt | 2013-03-11 | 1 | -4/+8 |
* | mesa: Reduce the memory usage for reg alloc with many graph nodes (part 1) | Eric Anholt | 2013-03-11 | 1 | -1/+13 |
* | i965/fs: Improve CSE performance by expiring some available expressions. | Eric Anholt | 2013-03-11 | 1 | -1/+19 |