On 26 January 2018 at 16:13, Richard Biener <richard.guent...@gmail.com> wrote: > On Fri, Jan 26, 2018 at 3:57 PM, Christophe Lyon > <christophe.l...@linaro.org> wrote: >> On 26 January 2018 at 11:25, Richard Biener <rguent...@suse.de> wrote: >>> On Thu, 25 Jan 2018, Richard Biener wrote: >>> >>>> On Thu, 25 Jan 2018, Marc Glisse wrote: >>>> >>>> > On Thu, 25 Jan 2018, Richard Biener wrote: >>>> > >>>> > > --- gcc/match.pd (revision 257047) >>>> > > +++ gcc/match.pd (working copy) >>>> > > @@ -1939,6 +1939,37 @@ DEFINE_INT_AND_FLOAT_ROUND_FN (RINT) >>>> > > (minus (convert (view_convert:stype @1)) >>>> > > (convert (view_convert:stype @2))))))) >>>> > > >>>> > > +/* (A * C) +- (B * C) -> (A+-B) * C and (A * C) +- A -> A * (C+-1). >>>> > > + Modeled after fold_plusminus_mult_expr. */ >>>> > > +(if (!TYPE_SATURATING (type) >>>> > > + && (!FLOAT_TYPE_P (type) || flag_associative_math)) >>>> > > + (for plusminus (plus minus) >>>> > > + (simplify >>>> > > + (plusminus (mult:s @0 @1) (mult:cs @0 @2)) >>>> > >>>> > No :c on the first mult, so we don't actually handle A*C+B*C? >>>> >>>> Hmm, I somehow convinced myself that it's only necessary on one >>>> of the mults... but you are of course right. Will re-test with >>>> that fixed. >>> >>> This is what I have applied. Note I had to XFAIL a minor part of >>> gcc.dg/tree-ssa/loop-15.c, namely simplifying the final value >>> replacement down to n * n. While the patch should enable this >>> the place where the transform could trigger with enough information >>> about n is VRP but that doesn't end up folding all stmts - and >>> that's something I'm not willing to change right now. >>> >>> Bootstrapped and tested on x86_64-unknown-linux-gnu, applied on trunk. >>> >> >> Hi, >> >> >> I think this patch caused regressions on aarch64: >> FAIL: gcc.dg/wmul-1.c scan-tree-dump-not widening_mul "WIDEN_MULT_PLUS_EXPR" > > Late forwprop does > > @@ -75,7 +21,7 @@ > _1 = (long unsigned int) Idx_6(D); > _2 = _1 * 40; > _12 = _1 * 4; > - _17 = _2 + _12; > + _17 = _1 * 44; > _13 = Arr_7(D) + _17; > MEM[(int[10] *)_13] = 1; > _4 = _2 + 400; > _18 = _4 + _12; > _16 = Arr_7(D) + _18; > MEM[(int[10] *)_16] = 2; > > > which I'm not sure ends up profitable at the end. It makes _12 > dead so the total number of multiplications stays the same. > For this we then apply the WIDEN_MULT_PLUS_EXPR two > times given _2 is no longer used multiple times. > > The reason the above transform happens is the match.pd :s > behavior which "ignores" :s in case the resulting expression > is "simple". > > Can you open a bugreport? >
Sure, this is: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84067 > Richard. > >> Thanks, >> >> Christophe >> >>> Richard. >>> >>> 2018-01-26 Richard Biener <rguent...@suse.de> >>> >>> PR tree-optimization/81082 >>> * fold-const.c (fold_plusminus_mult_expr): Do not perform the >>> association if it requires casting to unsigned. >>> * match.pd ((A * C) +- (B * C) -> (A+-B)): New patterns derived >>> from fold_plusminus_mult_expr to catch important cases late when >>> range info is available. >>> >>> * gcc.dg/vect/pr81082.c: New testcase. >>> * gcc.dg/tree-ssa/loop-15.c: XFAIL the (int)((unsigned)n + -1U) * n >>> + n >>> simplification to n * n. >>> >>> Index: gcc/fold-const.c >>> =================================================================== >>> --- gcc/fold-const.c (revision 257047) >>> +++ gcc/fold-const.c (working copy) >>> @@ -7097,7 +7097,7 @@ fold_plusminus_mult_expr (location_t loc >>> >>> /* Same may be zero and thus the operation 'code' may overflow. Likewise >>> same may be minus one and thus the multiplication may overflow. >>> Perform >>> - the operations in an unsigned type. */ >>> + the sum operation in an unsigned type. */ >>> tree utype = unsigned_type_for (type); >>> tree tem = fold_build2_loc (loc, code, utype, >>> fold_convert_loc (loc, utype, alt0), >>> @@ -7110,9 +7110,9 @@ fold_plusminus_mult_expr (location_t loc >>> return fold_build2_loc (loc, MULT_EXPR, type, >>> fold_convert (type, tem), same); >>> >>> - return fold_convert_loc (loc, type, >>> - fold_build2_loc (loc, MULT_EXPR, utype, tem, >>> - fold_convert_loc (loc, utype, >>> same))); >>> + /* Do not resort to unsigned multiplication because >>> + we lose the no-overflow property of the expression. */ >>> + return NULL_TREE; >>> } >>> >>> /* Subroutine of native_encode_expr. Encode the INTEGER_CST >>> Index: gcc/match.pd >>> =================================================================== >>> --- gcc/match.pd (revision 257047) >>> +++ gcc/match.pd (working copy) >>> @@ -1939,6 +1939,37 @@ DEFINE_INT_AND_FLOAT_ROUND_FN (RINT) >>> (minus (convert (view_convert:stype @1)) >>> (convert (view_convert:stype @2))))))) >>> >>> +/* (A * C) +- (B * C) -> (A+-B) * C and (A * C) +- A -> A * (C+-1). >>> + Modeled after fold_plusminus_mult_expr. */ >>> +(if (!TYPE_SATURATING (type) >>> + && (!FLOAT_TYPE_P (type) || flag_associative_math)) >>> + (for plusminus (plus minus) >>> + (simplify >>> + (plusminus (mult:cs @0 @1) (mult:cs @0 @2)) >>> + (if (!ANY_INTEGRAL_TYPE_P (type) >>> + || TYPE_OVERFLOW_WRAPS (type) >>> + || (INTEGRAL_TYPE_P (type) >>> + && tree_expr_nonzero_p (@0) >>> + && expr_not_equal_to (@0, wi::minus_one (TYPE_PRECISION >>> (type))))) >>> + (mult (plusminus @1 @2) @0))) >>> + /* We cannot generate constant 1 for fract. */ >>> + (if (!ALL_FRACT_MODE_P (TYPE_MODE (type))) >>> + (simplify >>> + (plusminus @0 (mult:cs @0 @2)) >>> + (if (!ANY_INTEGRAL_TYPE_P (type) >>> + || TYPE_OVERFLOW_WRAPS (type) >>> + || (INTEGRAL_TYPE_P (type) >>> + && tree_expr_nonzero_p (@0) >>> + && expr_not_equal_to (@0, wi::minus_one (TYPE_PRECISION >>> (type))))) >>> + (mult (plusminus { build_one_cst (type); } @2) @0))) >>> + (simplify >>> + (plusminus (mult:cs @0 @2) @0) >>> + (if (!ANY_INTEGRAL_TYPE_P (type) >>> + || TYPE_OVERFLOW_WRAPS (type) >>> + || (INTEGRAL_TYPE_P (type) >>> + && tree_expr_nonzero_p (@0) >>> + && expr_not_equal_to (@0, wi::minus_one (TYPE_PRECISION >>> (type))))) >>> + (mult (plusminus @2 { build_one_cst (type); }) @0)))))) >>> >>> /* Simplifications of MIN_EXPR, MAX_EXPR, fmin() and fmax(). */ >>> >>> Index: gcc/testsuite/gcc.dg/vect/pr81082.c >>> =================================================================== >>> --- gcc/testsuite/gcc.dg/vect/pr81082.c (revision 0) >>> +++ gcc/testsuite/gcc.dg/vect/pr81082.c (working copy) >>> @@ -0,0 +1,15 @@ >>> +/* { dg-do compile } */ >>> +/* { dg-require-effective-target vect_int } */ >>> + >>> +int >>> +f (int *x, int b1, int b2, int b3) >>> +{ >>> + int foo = 0; >>> + for (int i1 = 0; i1 < b1; ++i1) >>> + for (int i2 = 0; i2 < b2; ++i2) >>> + for (int i3 = 0; i3 < b3; ++i3) >>> + foo += x[i1 * b2 * b3 + i2 * b3 + (i3 - 1)]; >>> + return foo; >>> +} >>> + >>> +/* { dg-final { scan-tree-dump "vectorized 1 loops in function" "vect" } } >>> */ >>> Index: gcc/testsuite/gcc.dg/tree-ssa/loop-15.c >>> =================================================================== >>> --- gcc/testsuite/gcc.dg/tree-ssa/loop-15.c (revision 257048) >>> +++ gcc/testsuite/gcc.dg/tree-ssa/loop-15.c (working copy) >>> @@ -19,7 +19,7 @@ int bla(void) >>> } >>> >>> /* Since the loop is removed, there should be no addition. */ >>> -/* { dg-final { scan-tree-dump-times " \\+ " 0 "optimized" } } */ >>> +/* { dg-final { scan-tree-dump-times " \\+ " 0 "optimized" { xfail *-*-* } >>> } } */ >>> /* { dg-final { scan-tree-dump-times " \\* " 1 "optimized" } } */ >>> >>> /* The if from the loop header copying remains in the code. */
