public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug target/95435] New: bad builtin memcpy performance with znver1/znver2 and 32bit
@ 2020-05-30 10:12 jan at jki dot io
  2020-05-30 10:12 ` [Bug target/95435] " jan at jki dot io
                   ` (13 more replies)
  0 siblings, 14 replies; 15+ messages in thread
From: jan at jki dot io @ 2020-05-30 10:12 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95435

            Bug ID: 95435
           Summary: bad builtin memcpy performance with znver1/znver2 and
                    32bit
           Product: gcc
           Version: 10.1.1
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: target
          Assignee: unassigned at gcc dot gnu.org
          Reporter: jan at jki dot io
  Target Milestone: ---

Created attachment 48641
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=48641&action=edit
gcc -g -m32  -march=znver2 -O1 -s testmem_modified.c -o tm32

Hi,
I have found a regression with znver2 and 32bit compiles regarding builtin
memcpy.
Test program is this one:
https://github.com/level1wendell/memcpy_sse

results:
gcc -g -m32  -march=znver2 -O1 -s testmem_modified.c -o tm32
32 MB = 2.462717 ms

gcc -g -m32  -march=skylake -O1 -s testmem_modified.c -o tm32
32 MB = 1.135762 ms

gcc -fno-builtin-memcpy -g -m32  -march=znver2 -O1 -s testmem_modified.c -o
tm32
32 MB = 1.138656 ms

I have attached the generated assembler code for the first 2 cases.

^ permalink raw reply	[flat|nested] 15+ messages in thread
end of thread, other threads:[~2020-07-30  8:44 UTC | newest]

Thread overview: 15+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-05-30 10:12 [Bug target/95435] New: bad builtin memcpy performance with znver1/znver2 and 32bit jan at jki dot io
2020-05-30 10:12 ` [Bug target/95435] " jan at jki dot io
2020-05-30 10:16 ` jan at jki dot io
2020-05-30 10:18 ` glisse at gcc dot gnu.org
2020-05-30 10:23 ` jan at jki dot io
2020-05-30 16:43 ` amonakov at gcc dot gnu.org
2020-06-01  9:27 ` marxin at gcc dot gnu.org
2020-06-01  9:33 ` marxin at gcc dot gnu.org
2020-06-01  9:48 ` ubizjak at gmail dot com
2020-06-01 10:24 ` amonakov at gcc dot gnu.org
2020-06-01 10:48 ` ubizjak at gmail dot com
2020-06-01 10:58 ` marxin at gcc dot gnu.org
2020-07-30  8:32 ` cvs-commit at gcc dot gnu.org
2020-07-30  8:39 ` cvs-commit at gcc dot gnu.org
2020-07-30  8:44 ` marxin at gcc dot gnu.org

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).