* [Bug middle-end/66747] [5/6 Regression] The commit r225260 broke the builds of the mips-{mti,img}-linux-gnu tool chains.
2015-07-03 1:04 [Bug c/66747] New: The commit r225260 broke the builds of the mips-{mti,img}-linux-gnu tool chains doug.gilmore at imgtec dot com
@ 2015-07-03 8:00 ` rguenth at gcc dot gnu.org
2015-07-03 8:21 ` [Bug middle-end/66747] [6 " edlinger at gcc dot gnu.org
` (7 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: rguenth at gcc dot gnu.org @ 2015-07-03 8:00 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66747
Richard Biener <rguenth at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Target Milestone|--- |5.2
Summary|The commit r225260 broke |[5/6 Regression] The commit
|the builds of the |r225260 broke the builds of
|mips-{mti,img}-linux-gnu |the
|tool chains. |mips-{mti,img}-linux-gnu
| |tool chains.
^ permalink raw reply [flat|nested] 10+ messages in thread
* [Bug middle-end/66747] [6 Regression] The commit r225260 broke the builds of the mips-{mti,img}-linux-gnu tool chains.
2015-07-03 1:04 [Bug c/66747] New: The commit r225260 broke the builds of the mips-{mti,img}-linux-gnu tool chains doug.gilmore at imgtec dot com
2015-07-03 8:00 ` [Bug middle-end/66747] [5/6 Regression] " rguenth at gcc dot gnu.org
@ 2015-07-03 8:21 ` edlinger at gcc dot gnu.org
2015-07-03 10:06 ` edlinger at gcc dot gnu.org
` (6 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: edlinger at gcc dot gnu.org @ 2015-07-03 8:21 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66747
Bernd Edlinger <edlinger at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |NEW
Last reconfirmed| |2015-07-03
Version|5.2.0 |6.0
Target Milestone|5.2 |6.0
Summary|[5/6 Regression] The commit |[6 Regression] The commit
|r225260 broke the builds of |r225260 broke the builds of
|the |the
|mips-{mti,img}-linux-gnu |mips-{mti,img}-linux-gnu
|tool chains. |tool chains.
Ever confirmed|0 |1
--- Comment #1 from Bernd Edlinger <edlinger at gcc dot gnu.org> ---
confirmed. but it is a gcc-6 regression only.
With that patch we call mips_initial_elimination_offset
from rtx_addr_can_trap_p_1. And it falls over insn 103:
(insn 103 58 62 (sequence [
(call_insn 59 58 61 (parallel [
(call (mem:SI (reg/f:SI 25 $25 [orig:235 D.3533 ]
[235]) [0 *_13 S4 A32])
(const_int 0 [0]))
(clobber (reg:SI 31 $31))
]) ../../../../gcc-6-20150614/libgcc/crtstuff.c:707 645
{call_internal}
(expr_list:REG_DEAD (reg/f:SI 25 $25 [orig:235 D.3533 ] [235])
(expr_list:REG_CALL_DECL (nil)
(nil)))
(expr_list (use (reg:SI 79 $fakec))
(nil)))
(insn 61 59 62 (set (reg/v/f:SI 16 $16 [orig:234 pD.3516 ] [234])
(plus:SI (reg/v/f:SI 16 $16 [orig:234 pD.3516 ] [234])
(const_int -4 [0xfffffffffffffffc]))) 13 {*addsi3}
(nil))
]) ../../../../gcc-6-20150614/libgcc/crtstuff.c:707 -1
(nil))
unrecognizable, in mips_insn_has_flexible_gp_ref_p for whatever reason.
../../../../gcc-6-20150614/libgcc/crtstuff.c:708:1: internal compiler error: in
get_attr_got, at config/mips/mips.md:685
0xa9505a _fatal_insn(char const*, rtx_def const*, char const*, int, char
const*)
../../gcc-6-20150614/gcc/rtl-error.c:110
0xa95089 _fatal_insn_not_found(rtx_def const*, char const*, int, char const*)
../../gcc-6-20150614/gcc/rtl-error.c:118
0xe1beb5 get_attr_got
../../gcc-6-20150614/gcc/config/mips/mips.md:685
0xe1beb5 get_attr_got
../../gcc-6-20150614/gcc/config/mips/mips.md:662
0xe07c8a mips_insn_has_flexible_gp_ref_p
../../gcc-6-20150614/gcc/config/mips/mips.c:9905
0xe017f9 mips_find_gp_ref
../../gcc-6-20150614/gcc/config/mips/mips.c:9836
0xe020ed mips_global_pointer
../../gcc-6-20150614/gcc/config/mips/mips.c:9957
0xe020ed mips_compute_frame_info
../../gcc-6-20150614/gcc/config/mips/mips.c:10327
0xe0e8d6 mips_initial_elimination_offset(int, int)
../../gcc-6-20150614/gcc/config/mips/mips.c:10534
0xa9764d rtx_addr_can_trap_p_1
../../gcc-6-20150614/gcc/rtlanal.c:627
0xa97335 rtx_addr_can_trap_p_1
../../gcc-6-20150614/gcc/rtlanal.c:697
0xa99f87 may_trap_p_1(rtx_def const*, unsigned int)
../../gcc-6-20150614/gcc/rtlanal.c:2836
0xa8d7f7 fill_slots_from_thread
../../gcc-6-20150614/gcc/reorg.c:2521
0xa902dd fill_eager_delay_slots
../../gcc-6-20150614/gcc/reorg.c:2958
0xa902dd dbr_schedule
../../gcc-6-20150614/gcc/reorg.c:3805
0xa902dd rest_of_handle_delay_slots
../../gcc-6-20150614/gcc/reorg.c:3944
0xa902dd execute
../../gcc-6-20150614/gcc/reorg.c:3975
^ permalink raw reply [flat|nested] 10+ messages in thread
* [Bug middle-end/66747] [6 Regression] The commit r225260 broke the builds of the mips-{mti,img}-linux-gnu tool chains.
2015-07-03 1:04 [Bug c/66747] New: The commit r225260 broke the builds of the mips-{mti,img}-linux-gnu tool chains doug.gilmore at imgtec dot com
2015-07-03 8:00 ` [Bug middle-end/66747] [5/6 Regression] " rguenth at gcc dot gnu.org
2015-07-03 8:21 ` [Bug middle-end/66747] [6 " edlinger at gcc dot gnu.org
@ 2015-07-03 10:06 ` edlinger at gcc dot gnu.org
2015-07-03 11:23 ` bernd.edlinger at hotmail dot de
` (5 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: edlinger at gcc dot gnu.org @ 2015-07-03 10:06 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66747
--- Comment #2 from Bernd Edlinger <edlinger at gcc dot gnu.org> ---
get_attr_got fails when called with a SEQUENCE.
I am not sure if reg_overlap_mentioned_p works with SEQUENCE.
--- gcc/config/mips/mips.c.jj 2015-06-08 23:06:50.000000000 +0200
+++ gcc/config/mips/mips.c 2015-07-03 11:38:48.962336295 +0200
@@ -9902,7 +9902,7 @@ mips_cfun_has_inflexible_gp_ref_p (void)
static bool
mips_insn_has_flexible_gp_ref_p (rtx_insn *insn)
{
- return (get_attr_got (insn) != GOT_UNSET
+ return ((PATTERN (insn)->code != SEQUENCE && get_attr_got (insn) !=
GOT_UNSET)
|| mips_small_data_pattern_p (PATTERN (insn))
|| reg_overlap_mentioned_p (pic_offset_table_rtx, PATTERN (insn)));
}
^ permalink raw reply [flat|nested] 10+ messages in thread
* [Bug middle-end/66747] [6 Regression] The commit r225260 broke the builds of the mips-{mti,img}-linux-gnu tool chains.
2015-07-03 1:04 [Bug c/66747] New: The commit r225260 broke the builds of the mips-{mti,img}-linux-gnu tool chains doug.gilmore at imgtec dot com
` (2 preceding siblings ...)
2015-07-03 10:06 ` edlinger at gcc dot gnu.org
@ 2015-07-03 11:23 ` bernd.edlinger at hotmail dot de
2015-07-03 20:42 ` doug.gilmore at imgtec dot com
` (4 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: bernd.edlinger at hotmail dot de @ 2015-07-03 11:23 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66747
--- Comment #3 from Bernd Edlinger <bernd.edlinger at hotmail dot de> ---
Created attachment 35902
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=35902&action=edit
proposed patch
this slightly improved patch should fix it, could you give it a try?
Thanks,
Bernd.
^ permalink raw reply [flat|nested] 10+ messages in thread
* [Bug middle-end/66747] [6 Regression] The commit r225260 broke the builds of the mips-{mti,img}-linux-gnu tool chains.
2015-07-03 1:04 [Bug c/66747] New: The commit r225260 broke the builds of the mips-{mti,img}-linux-gnu tool chains doug.gilmore at imgtec dot com
` (3 preceding siblings ...)
2015-07-03 11:23 ` bernd.edlinger at hotmail dot de
@ 2015-07-03 20:42 ` doug.gilmore at imgtec dot com
2015-07-04 3:44 ` doug.gilmore at imgtec dot com
` (3 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: doug.gilmore at imgtec dot com @ 2015-07-03 20:42 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66747
--- Comment #4 from Doug Gilmore <doug.gilmore at imgtec dot com> ---
Thanks!
I started up a build with the patch and it got through
the initial_gcc build so that is a good sign.
I'll send an update once the build is done.
^ permalink raw reply [flat|nested] 10+ messages in thread
* [Bug middle-end/66747] [6 Regression] The commit r225260 broke the builds of the mips-{mti,img}-linux-gnu tool chains.
2015-07-03 1:04 [Bug c/66747] New: The commit r225260 broke the builds of the mips-{mti,img}-linux-gnu tool chains doug.gilmore at imgtec dot com
` (4 preceding siblings ...)
2015-07-03 20:42 ` doug.gilmore at imgtec dot com
@ 2015-07-04 3:44 ` doug.gilmore at imgtec dot com
2015-07-04 7:50 ` [Bug target/66747] " edlinger at gcc dot gnu.org
` (2 subsequent siblings)
8 siblings, 0 replies; 10+ messages in thread
From: doug.gilmore at imgtec dot com @ 2015-07-04 3:44 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66747
Doug Gilmore <doug.gilmore at imgtec dot com> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |matthew.fortune at imgtec dot com
--- Comment #5 from Doug Gilmore <doug.gilmore at imgtec dot com> ---
The build succeeded and the regression test run
showed no regressions.
Bernd: could you send the patch to the list for approval?
Thanks!
^ permalink raw reply [flat|nested] 10+ messages in thread
* [Bug target/66747] [6 Regression] The commit r225260 broke the builds of the mips-{mti,img}-linux-gnu tool chains.
2015-07-03 1:04 [Bug c/66747] New: The commit r225260 broke the builds of the mips-{mti,img}-linux-gnu tool chains doug.gilmore at imgtec dot com
` (5 preceding siblings ...)
2015-07-04 3:44 ` doug.gilmore at imgtec dot com
@ 2015-07-04 7:50 ` edlinger at gcc dot gnu.org
2015-07-04 19:02 ` edlinger at gcc dot gnu.org
2015-07-06 20:32 ` doug.gilmore at imgtec dot com
8 siblings, 0 replies; 10+ messages in thread
From: edlinger at gcc dot gnu.org @ 2015-07-04 7:50 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66747
Bernd Edlinger <edlinger at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Component|middle-end |target
Assignee|unassigned at gcc dot gnu.org |edlinger at gcc dot gnu.org
--- Comment #6 from Bernd Edlinger <edlinger at gcc dot gnu.org> ---
patch was posted here: https://gcc.gnu.org/ml/gcc-patches/2015-07/msg00238.html
^ permalink raw reply [flat|nested] 10+ messages in thread
* [Bug target/66747] [6 Regression] The commit r225260 broke the builds of the mips-{mti,img}-linux-gnu tool chains.
2015-07-03 1:04 [Bug c/66747] New: The commit r225260 broke the builds of the mips-{mti,img}-linux-gnu tool chains doug.gilmore at imgtec dot com
` (6 preceding siblings ...)
2015-07-04 7:50 ` [Bug target/66747] " edlinger at gcc dot gnu.org
@ 2015-07-04 19:02 ` edlinger at gcc dot gnu.org
2015-07-06 20:32 ` doug.gilmore at imgtec dot com
8 siblings, 0 replies; 10+ messages in thread
From: edlinger at gcc dot gnu.org @ 2015-07-04 19:02 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66747
--- Comment #7 from Bernd Edlinger <edlinger at gcc dot gnu.org> ---
Author: edlinger
Date: Sat Jul 4 19:02:18 2015
New Revision: 225416
URL: https://gcc.gnu.org/viewcvs?rev=225416&root=gcc&view=rev
Log:
2015-07-04 Bernd Edlinger <bernd.edlinger@hotmail.de>
PR target/66747
* config/mips/mips.c (mips_find_gp_ref): Handle instruction sequences.
Modified:
trunk/gcc/ChangeLog
trunk/gcc/config/mips/mips.c
^ permalink raw reply [flat|nested] 10+ messages in thread
* [Bug target/66747] [6 Regression] The commit r225260 broke the builds of the mips-{mti,img}-linux-gnu tool chains.
2015-07-03 1:04 [Bug c/66747] New: The commit r225260 broke the builds of the mips-{mti,img}-linux-gnu tool chains doug.gilmore at imgtec dot com
` (7 preceding siblings ...)
2015-07-04 19:02 ` edlinger at gcc dot gnu.org
@ 2015-07-06 20:32 ` doug.gilmore at imgtec dot com
8 siblings, 0 replies; 10+ messages in thread
From: doug.gilmore at imgtec dot com @ 2015-07-06 20:32 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66747
--- Comment #9 from Doug Gilmore <doug.gilmore at imgtec dot com> ---
Our nightly builds are now clean with this patch.
Thanks!
^ permalink raw reply [flat|nested] 10+ messages in thread