projects / platform / upstream / gcc.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 2318f3b) | patch
re PR target/83008 ([performance] Is it better to avoid extra instructions in data...
authorSergey Shalnov <Sergey.Shalnov@intel.com>
Thu, 8 Feb 2018 22:31:15 +0000 (23:31 +0100)
committerUros Bizjak <uros@gcc.gnu.org>
Thu, 8 Feb 2018 22:31:15 +0000 (23:31 +0100)
commit001e73373e6d2e7c756141e0d7ac8e24ae1574ad
tree9309b5c9e55c0ac73b256026a8e60660f2741c0ctree | snapshot
parent2318f3b46ed94cde0bd04d82dd1eccd483c23e5dcommit | diff
re PR target/83008 ([performance] Is it better to avoid extra instructions in data passing between loops?)

PR target/83008
* config/i386/x86-tune-costs.h (skylake_cost): Fix cost of
storing integer register in SImode.  Fix cost of 256 and 512
byte aligned SSE register store.

* config/i386/i386.c (ix86_multiplication_cost): Fix
multiplication cost for TARGET_AVX512DQ.

testsuite/ChangeLog:

PR target/83008
* gcc.target/i386/pr83008.c: New test.

From-SVN: r257505
gcc/ChangeLog diff | blob | history
gcc/config/i386/i386.c diff | blob | history
gcc/config/i386/x86-tune-costs.h diff | blob | history
gcc/testsuite/ChangeLog diff | blob | history
gcc/testsuite/gcc.target/i386/pr83008.c [new file with mode: 0644] blob
Domain: System / Toolchain;