This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug other/48696] Horrible bitfield code generation on x86
- From: "torvalds at linux-foundation dot org" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Wed, 20 Apr 2011 04:28:19 +0000
- Subject: [Bug other/48696] Horrible bitfield code generation on x86
- Auto-submitted: auto-generated
- References: <bug-48696-4@http.gcc.gnu.org/bugzilla/>
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48696
--- Comment #1 from Linus Torvalds <torvalds@linux-foundation.org> 2011-04-20 04:28:18 UTC ---
Side note, this is talked about in the Intel optimization manual
3.6.4 Store Forwarding
Under "General Optimization Guidelines" -> "Optimizing Memory Accesses" ->
"Store Forwarding".
I'm sure people are aware of it, but I'm not certain people realize how
horrible the performance hit can be. I have to admit to having been a bit
surprised myself just _how_ clearly it shows up in profiles.