Ping. Hopefully this is easier to review/test now that we fix the AArch64 bug first and deliberately regress code quality so that the impact of the combine patch can be measured.
On 15/10/2020 09:59, Alex Coplan via Gcc-patches wrote: > Currently, make_extraction() identifies where we can emit an ASHIFT of > an extend in place of an extraction, but fails to make the corresponding > canonicalization/simplification when presented with a MULT by a power of > two. Such a representation is canonical when representing a left-shifted > address inside a MEM. > > This patch remedies this situation: after the patch, make_extraction() > now also identifies RTXs such as: > > (mult:DI (subreg:DI (reg:SI r)) (const_int 2^n)) > > and rewrites this as: > > (mult:DI (sign_extend:DI (reg:SI r)) (const_int 2^n)) > > instead of using a sign_extract. > > (This patch also fixes up a comment in expand_compound_operation() which > appears to have suffered from bitrot.) > > This fixes several quality regressions on AArch64 after removing support for > addresses represented as sign_extract insns (1/2). > > In particular, after the fix for PR96998, for the relevant testcase, we > have: > > .L2: > sxtw x0, w0 // 8 [c=4 l=4] *extendsidi2_aarch64/0 > add x0, x19, x0, lsl 2 // 39 [c=8 l=4] *add_lsl_di > bl h // 11 [c=4 l=4] *call_value_insn/1 > b .L2 // 54 [c=4 l=4] jump > > and after this patch, we have: > > .L2: > add x0, x19, w0, sxtw 2 // 39 [c=8 l=4] > *add_extendsi_shft_di > bl h // 11 [c=4 l=4] *call_value_insn/1 > b .L2 // 54 [c=4 l=4] jump > > which restores the original optimal sequence we saw before the AArch64 > patch. > > Testing: > * Bootstrapped and regtested on aarch64-linux-gnu, arm-linux-gnueabihf, > and x86_64-linux-gnu. > > OK for trunk? > > Thanks, > Alex > > --- > > gcc/ChangeLog: > > * combine.c (expand_compound_operation): Tweak variable name in > comment to match source. > (make_extraction): Handle mult by power of two in addition to > ashift. > > --- > gcc/combine.c | 37 ++++++++++++++++++++++--------------- > 1 file changed, 22 insertions(+), 15 deletions(-)
