https://gcc.gnu.org/bugzilla/show_bug.cgi?id=59401
Oleg Endo changed:
What|Removed |Added
Status|UNCONFIRMED |RESOLVED
Resolution|---
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=59401
--- Comment #13 from Oleg Endo ---
Author: olegendo
Date: Thu Oct 16 12:21:29 2014
New Revision: 216314
URL: https://gcc.gnu.org/viewcvs?rev=216314&root=gcc&view=rev
Log:
gcc/
PR target/59401
* config/sh/sh.h (CALL_REALLY_USED_REGISTERS)
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=59401
--- Comment #12 from Kazumoto Kojima ---
(In reply to Oleg Endo from comment #11)
> Kaz, could you please add the attached patch to your test run?
No new failures for the top level "make -k check" on sh4-unknown-linux-gnu.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=59401
--- Comment #11 from Oleg Endo ---
Created attachment 33723
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=33723&action=edit
Make GBR call preserved by default
(In reply to Oleg Endo from comment #10)
> (In reply to Kazumoto Kojima from co
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=59401
--- Comment #10 from Oleg Endo ---
(In reply to Kazumoto Kojima from comment #9)
> (In reply to Oleg Endo from comment #8)
> > change the
> > value for gbr in sh.h CALL_USED_REGISTERS from '1' to '0' and confirm that
> > everything is still OK?
>
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=59401
--- Comment #9 from Kazumoto Kojima ---
(In reply to Oleg Endo from comment #8)
> change the
> value for gbr in sh.h CALL_USED_REGISTERS from '1' to '0' and confirm that
> everything is still OK?
The comment and document about CALL_USED_REGISTER
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=59401
--- Comment #8 from Oleg Endo ---
(In reply to Kazumoto Kojima from comment #7)
> (In reply to Oleg Endo from comment #6)
> > Kaz, what's your opinion on making GBR to be call preserved by default?
>
> Looks OK to me for 5.0. It's clearly an AB
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=59401
--- Comment #7 from Kazumoto Kojima ---
(In reply to Oleg Endo from comment #6)
> Kaz, what's your opinion on making GBR to be call preserved by default?
Looks OK to me for 5.0. It's clearly an ABI change but a change to
the more robust directi
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=59401
--- Comment #6 from Oleg Endo ---
(In reply to Oleg Endo from comment #5)
> Ugh, there's actually another problem with this thing, I think:
>
> ...
>
> By default, GBR is marked as call-clobbered. Currently, if the GBR mem
> optimization thing
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=59401
--- Comment #5 from Oleg Endo ---
Ugh, there's actually another problem with this thing, I think:
void other (void);
int test0 (void)
{
int x = ((int*)__builtin_thread_pointer ())[2];
other ();
return ((int*)__builtin_thread_pointer ())
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=59401
--- Comment #4 from Kazumoto Kojima ---
(In reply to Oleg Endo from comment #3)
> Do you have an opinion on that?
Looks the latter is enough for the released branches. I'm OK with eather way,
though.
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=59401
Oleg Endo changed:
What|Removed |Added
CC||kkojima at gcc dot gnu.org
--- Comment #3 fr
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=59401
--- Comment #2 from Oleg Endo ---
Author: olegendo
Date: Sun Oct 12 23:14:07 2014
New Revision: 216128
URL: https://gcc.gnu.org/viewcvs?rev=216128&root=gcc&view=rev
Log:
gcc/
PR target/59401
* config/sh/sh-protos (sh_find_equiv_gbr_addr)
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59401
--- Comment #1 from Oleg Endo ---
An example where the base address is retrieved from the GBR in one basic block,
but used in different basic blocks:
struct tcb_t
{
int x, y, z, w;
};
int test_01 (int a, tcb_t* b, int c)
{
tcb_t* tcb = (tcb_
14 matches
Mail list logo
