| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | nir: Teach nir_opt_algebraic about adding and subtracting the same thing | Kristian Høgsberg Kristensen | 2015-12-29 | 1 | -0/+4 |
| * | nir: Optimize useless comparisons against true/false. | Matt Turner | 2015-12-08 | 1 | -2/+4 |
| * | nir: Add opcodes for saturated vector math. | Eric Anholt | 2015-10-23 | 1 | -0/+6 |
| * | nir: Add fdph and fdph_replicated opcodes | Jason Ekstrand | 2015-09-22 | 1 | -0/+1 |
| * | nir: add lowering for ffract | Rob Clark | 2015-09-16 | 1 | -0/+1 |
| * | nir: Add a fdot instruction that replicates the result to a vec4 | Jason Ekstrand | 2015-09-15 | 1 | -0/+3 |
| * | nir: Simplify feq(fneg(a), a)) -> feq(a, 0.0) | Thomas Helland | 2015-08-18 | 1 | -0/+1 |
| * | nir: Simplify fne(fneg(a), a) -> fne(a, 0.0) | Thomas Helland | 2015-08-18 | 1 | -0/+1 |
| * | nir: Add algebraic opt for no-op iand. | Eric Anholt | 2015-08-04 | 1 | -0/+1 |
| * | nir: Recognize max(min(a, 1.0), 0.0) as fsat(a). | Kenneth Graunke | 2015-06-25 | 1 | -0/+1 |
| * | nir: Remove sRGB colorspace conversion round-trip. | Matt Turner | 2015-05-22 | 1 | -0/+2 |
| * | nir: Delete all traces of nir_op_flog | Ian Romanick | 2015-05-08 | 1 | -6/+0 |
| * | nir: Delete all traces of nir_op_fexp | Ian Romanick | 2015-05-08 | 1 | -7/+0 |
| * | nir: Allow feq/fne/ieq/ine to be optimized with inot. | Matt Turner | 2015-05-07 | 1 | -0/+4 |
| * | nir: Recognize (a < c || b < c) as min(a, b) < c. | Matt Turner | 2015-05-07 | 1 | -0/+2 |
| * | nir: Recognize trivial min/max. | Matt Turner | 2015-05-07 | 1 | -0/+6 |
| * | nir: Recognize i2b(b2i(x)) as x. | Matt Turner | 2015-05-07 | 1 | -0/+1 |
| * | nir: Recognize imul(b2i(a), b2i(b)) as a logical AND. | Matt Turner | 2015-05-07 | 1 | -0/+1 |
| * | nir: Transform pow(x, 4) into (x*x)*(x*x). | Matt Turner | 2015-04-24 | 1 | -0/+1 |
| * | nir: Fix typo in "ushr by 0" algebraic replacement | Ian Romanick | 2015-04-14 | 1 | -1/+1 |
| * | nir: split out lower_sub from lower_negate | Rob Clark | 2015-04-11 | 1 | -2/+2 |
| * | nir: add option to lower slt/sge/seq/sne | Rob Clark | 2015-04-05 | 1 | -0/+4 |
| * | nir: Remove useless ftrunc inside f2i/f2u. | Matt Turner | 2015-04-01 | 1 | -0/+4 |
| * | nir: Recognize (a < b || a < c) as a < max(b, c). | Matt Turner | 2015-04-01 | 1 | -0/+2 |
| * | nir: Add addition/multiplication identities of exp/log. | Matt Turner | 2015-04-01 | 1 | -0/+6 |
| * | nir: Add identities for the log function. | Matt Turner | 2015-04-01 | 1 | -0/+8 |
| * | nir: Add identities for the exponential function. | Matt Turner | 2015-04-01 | 1 | -0/+6 |
| * | nir: Recognize another open coded lrp. | Matt Turner | 2015-04-01 | 1 | -0/+1 |
| * | nir: Recognize open coded lrp. | Matt Turner | 2015-04-01 | 1 | -0/+1 |
| * | nir: Move the compare-with-zero optimizations to the late section | Jason Ekstrand | 2015-04-01 | 1 | -4/+4 |
| * | nir/algebraic: Add a seperate section for "late" optimizations | Jason Ekstrand | 2015-04-01 | 1 | -0/+9 |
| * | nir/algebraic: Remove a duplicate optimization | Jason Ekstrand | 2015-04-01 | 1 | -3/+0 |
| * | nir: Recognize a pattern of bool frobbing from TGSI KILL_IF. | Eric Anholt | 2015-04-01 | 1 | -0/+2 |
| * | nir: Recognize a pattern for doing b2f without the opcode. | Eric Anholt | 2015-04-01 | 1 | -0/+1 |
| * | nir: Lower subtraction to add with negation when !lower_negate. | Kenneth Graunke | 2015-03-27 | 1 | -0/+2 |
| * | nir: Add optional lowering of flrp. | Eric Anholt | 2015-03-27 | 1 | -0/+1 |
| * | nir: Recognize sat(add(b2f(a), b2f(b))) as a logical OR. | Matt Turner | 2015-03-24 | 1 | -0/+1 |
| * | nir: Recognize mul(b2f(a), b2f(b)) as a logical AND. | Matt Turner | 2015-03-24 | 1 | -0/+2 |
| * | nir: Optimize a + neg(a) | Thomas Helland | 2015-03-11 | 1 | -0/+2 |
| * | nir: Optimize (a*b)+(a*c) -> a*(b+c) | Thomas Helland | 2015-03-11 | 1 | -0/+2 |
| * | nir: Generalize the optimization of subs of subs from 0. | Eric Anholt | 2015-02-21 | 1 | -2/+2 |
| * | nir: Collapse repeated bcsels on the same argument. | Eric Anholt | 2015-02-21 | 1 | -0/+1 |
| * | nir: When faced with a csel on !condition, just flip the arguments. | Eric Anholt | 2015-02-21 | 1 | -0/+1 |
| * | nir: Add a couple of simplifications of csel operations. | Eric Anholt | 2015-02-21 | 1 | -0/+3 |
| * | nir: Recognize and reduce duplicated fsats. | Eric Anholt | 2015-02-18 | 1 | -0/+2 |
| * | nir: Add a flag for lowering fsat. | Eric Anholt | 2015-02-18 | 1 | -1/+2 |
| * | nir: Add a flag for lowering ffma. | Eric Anholt | 2015-02-18 | 1 | -1/+2 |
| * | nir: Add a flag for lowering fneg/ineg. | Eric Anholt | 2015-02-18 | 1 | -0/+10 |
| * | nir: Add a flag for lowering fsqrt(x) to frcp(frsqrt(x)). | Eric Anholt | 2015-02-18 | 1 | -1/+2 |
| * | nir: Add lowering of POW instructions if the lower flag is set. | Eric Anholt | 2015-02-18 | 1 | -0/+1 |
