https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81501
H.J. Lu changed:
What|Removed |Added
Target Milestone|--- |16.0
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81501
H.J. Lu changed:
What|Removed |Added
Attachment #61788|0 |1
is obsolete|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81501
H.J. Lu changed:
What|Removed |Added
Attachment #60535|0 |1
is obsolete|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81501
Sam James changed:
What|Removed |Added
Keywords||patch
--- Comment #12 from Sam James ---
ht
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81501
H.J. Lu changed:
What|Removed |Added
Attachment #53473|0 |1
is obsolete|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81501
Sam James changed:
What|Removed |Added
See Also||https://github.com/llvm/llv
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81501
--- Comment #9 from Roy Jacobson ---
Thanks for the patch! I wonder if it would handle coroutines correctly. Clang
has this open bug https://github.com/llvm/llvm-project/issues/47179 that is
related to this optimization.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81501
--- Comment #8 from H.J. Lu ---
Created attachment 53473
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=53473&action=edit
A patch
This patch uses a single UNSPEC_TLS_LD_BASE in the whole function.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81501
--- Comment #7 from Yann Droneaud ---
(In reply to Roy Jacobson from comment #6)
> We recently upgraded our toolchain from GCC9 to GCC11, and we're seeing
> __tls_get_addr take up to 10% of total runtime under some workloads, where
> it was 1-2%
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81501
Roy Jacobson changed:
What|Removed |Added
CC||roi.jacobson1 at gmail dot com
--- Commen
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81501
Andrew Pinski changed:
What|Removed |Added
Summary|Unneccessary calls to |mulitple calls to
|__t
11 matches
Mail list logo