* [Bug ipa/97335] [11 Regression] 525.x264_r fails with -Ofast -g -march=znver2 -flto=16 since r11-3641-gc34db4b6f8a5d803
2020-10-08 11:54 [Bug ipa/97335] New: [11 Regression] 525.x264_r fails with -Ofast -g -march=znver2 -flto=16 since r11-3641-gc34db4b6f8a5d803 marxin at gcc dot gnu.org
@ 2020-10-08 11:55 ` marxin at gcc dot gnu.org
2020-10-08 12:55 ` marxin at gcc dot gnu.org
` (11 subsequent siblings)
12 siblings, 0 replies; 14+ messages in thread
From: marxin at gcc dot gnu.org @ 2020-10-08 11:55 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97335
Martin Liška <marxin at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
See Also| |https://gcc.gnu.org/bugzill
| |a/show_bug.cgi?id=97292
Status|UNCONFIRMED |NEW
Known to work| |10.2.0
Blocks| |26163
Known to fail| |11.0
Last reconfirmed| |2020-10-08
Ever confirmed|0 |1
Target Milestone|--- |11.0
Referenced Bugs:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=26163
[Bug 26163] [meta-bug] missed optimization in SPEC (2k17, 2k and 2k6 and 95)
^ permalink raw reply [flat|nested] 14+ messages in thread* [Bug ipa/97335] [11 Regression] 525.x264_r fails with -Ofast -g -march=znver2 -flto=16 since r11-3641-gc34db4b6f8a5d803
2020-10-08 11:54 [Bug ipa/97335] New: [11 Regression] 525.x264_r fails with -Ofast -g -march=znver2 -flto=16 since r11-3641-gc34db4b6f8a5d803 marxin at gcc dot gnu.org
2020-10-08 11:55 ` [Bug ipa/97335] " marxin at gcc dot gnu.org
@ 2020-10-08 12:55 ` marxin at gcc dot gnu.org
2020-10-08 15:31 ` marxin at gcc dot gnu.org
` (10 subsequent siblings)
12 siblings, 0 replies; 14+ messages in thread
From: marxin at gcc dot gnu.org @ 2020-10-08 12:55 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97335
--- Comment #1 from Martin Liška <marxin at gcc dot gnu.org> ---
With -O3 -g -flto=16 --param=lto-partitions=1, the first bad binary is with
-fdbg-cnt=ipa_mod_ref:1046.
It seems a misc tool fails with:
./ldecod_r -i BuckBunny.264 -o BuckBunny.yuv
Setting Default Parameters...
Parsing Configfile decoder.cfg
----------------------------- JM 17.1 (FRExt) -----------------------------
--------------------------------------------------------------------------
Input H.264 bitstream : BuckBunny.264
Output decoded YUV : BuckBunny.yuv
Input reference file : test_rec.yuv
--------------------------------------------------------------------------
POC must = frame# or field# for SNRs to be correct
--------------------------------------------------------------------------
Frame POC Pic# QP SnrY SnrU SnrV Y:U:V Time(ms)
--------------------------------------------------------------------------
Input reference file : test_rec.yuv does not exist
SNR values are not available
00000(IDR) 0 0 28 4:2:0 0
00002( P ) 4 1 31 4:2:0 0
00001( b ) 2 2 31 4:2:0 0
00003( I ) 6 2 28 4:2:0 0
00006( P ) 12 4 28 4:2:0 0
mb_qp_delta is out of range
PRE dump files:
https://drive.google.com/file/d/1GzqkYP4PRtXVkVbvGGJUukz8-9ay-uMy/view?usp=sharing
^ permalink raw reply [flat|nested] 14+ messages in thread* [Bug ipa/97335] [11 Regression] 525.x264_r fails with -Ofast -g -march=znver2 -flto=16 since r11-3641-gc34db4b6f8a5d803
2020-10-08 11:54 [Bug ipa/97335] New: [11 Regression] 525.x264_r fails with -Ofast -g -march=znver2 -flto=16 since r11-3641-gc34db4b6f8a5d803 marxin at gcc dot gnu.org
2020-10-08 11:55 ` [Bug ipa/97335] " marxin at gcc dot gnu.org
2020-10-08 12:55 ` marxin at gcc dot gnu.org
@ 2020-10-08 15:31 ` marxin at gcc dot gnu.org
2020-10-08 15:33 ` marxin at gcc dot gnu.org
` (9 subsequent siblings)
12 siblings, 0 replies; 14+ messages in thread
From: marxin at gcc dot gnu.org @ 2020-10-08 15:31 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97335
--- Comment #2 from Martin Liška <marxin at gcc dot gnu.org> ---
With the current master (g:3e1123e52f8eca2650efa0bc81768792d328961f) one needs
only one IPA modref transform to trigger the wrong code:
-fdbg-cnt=ipa_mod_ref:778-778
where one needs to copy the input file to
benchspec/CPU/525.x264_r/build/build_peak_gcc7-m64.0000 folder:
cp ../../../525.x264_r/data/test/input/BuckBunny.264 .
One can see the difference in PRE dump file.
^ permalink raw reply [flat|nested] 14+ messages in thread* [Bug ipa/97335] [11 Regression] 525.x264_r fails with -Ofast -g -march=znver2 -flto=16 since r11-3641-gc34db4b6f8a5d803
2020-10-08 11:54 [Bug ipa/97335] New: [11 Regression] 525.x264_r fails with -Ofast -g -march=znver2 -flto=16 since r11-3641-gc34db4b6f8a5d803 marxin at gcc dot gnu.org
` (2 preceding siblings ...)
2020-10-08 15:31 ` marxin at gcc dot gnu.org
@ 2020-10-08 15:33 ` marxin at gcc dot gnu.org
2020-10-09 8:27 ` marxin at gcc dot gnu.org
` (8 subsequent siblings)
12 siblings, 0 replies; 14+ messages in thread
From: marxin at gcc dot gnu.org @ 2020-10-08 15:33 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97335
--- Comment #3 from Martin Liška <marxin at gcc dot gnu.org> ---
One needs -flto here.
^ permalink raw reply [flat|nested] 14+ messages in thread* [Bug ipa/97335] [11 Regression] 525.x264_r fails with -Ofast -g -march=znver2 -flto=16 since r11-3641-gc34db4b6f8a5d803
2020-10-08 11:54 [Bug ipa/97335] New: [11 Regression] 525.x264_r fails with -Ofast -g -march=znver2 -flto=16 since r11-3641-gc34db4b6f8a5d803 marxin at gcc dot gnu.org
` (3 preceding siblings ...)
2020-10-08 15:33 ` marxin at gcc dot gnu.org
@ 2020-10-09 8:27 ` marxin at gcc dot gnu.org
2020-10-09 8:31 ` marxin at gcc dot gnu.org
` (7 subsequent siblings)
12 siblings, 0 replies; 14+ messages in thread
From: marxin at gcc dot gnu.org @ 2020-10-09 8:27 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97335
--- Comment #4 from Martin Liška <marxin at gcc dot gnu.org> ---
This case is about:
<bb 28> [local count: 177167400]:
+ pretmp_327 = MEM[(int *)dep_dp_76(D) + 8B];
+ goto <bb 46>; [100.00%]
+
+ <bb 29> [local count: 177167400]:
ctx_144 = _44 + 4;
+ pretmp_335 = MEM[(int *)dep_dp_76(D) + 8B];
where dep_dp is of type:
typedef struct
{
unsigned int Drange;
unsigned int Dvalue;
int DbitsLeft;
byte *Dcodestrm;
int *Dcodestrm_len;
} DecodingEnvironment;
and so dep_dp_76(D) + 8B is 'Dvalue' field of the struct.
Looking at the code, there are really places where one takes pointer of the
field:
benchspec/CPU/525.x264_r/src/ldecod_src/biaridecod.c: unsigned int *value =
&dep->Dvalue;
benchspec/CPU/525.x264_r/src/ldecod_src/biaridecod.c: unsigned int *value =
&dep->Dvalue;
But the usage seems fine to me, no casting and direct usage of the 'unsigned
int *'.
^ permalink raw reply [flat|nested] 14+ messages in thread* [Bug ipa/97335] [11 Regression] 525.x264_r fails with -Ofast -g -march=znver2 -flto=16 since r11-3641-gc34db4b6f8a5d803
2020-10-08 11:54 [Bug ipa/97335] New: [11 Regression] 525.x264_r fails with -Ofast -g -march=znver2 -flto=16 since r11-3641-gc34db4b6f8a5d803 marxin at gcc dot gnu.org
` (4 preceding siblings ...)
2020-10-09 8:27 ` marxin at gcc dot gnu.org
@ 2020-10-09 8:31 ` marxin at gcc dot gnu.org
2020-10-09 8:47 ` marxin at gcc dot gnu.org
` (6 subsequent siblings)
12 siblings, 0 replies; 14+ messages in thread
From: marxin at gcc dot gnu.org @ 2020-10-09 8:31 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97335
--- Comment #5 from Martin Liška <marxin at gcc dot gnu.org> ---
(In reply to Martin Liška from comment #4)
> This case is about:
>
> <bb 28> [local count: 177167400]:
> + pretmp_327 = MEM[(int *)dep_dp_76(D) + 8B];
> + goto <bb 46>; [100.00%]
> +
> + <bb 29> [local count: 177167400]:
> ctx_144 = _44 + 4;
> + pretmp_335 = MEM[(int *)dep_dp_76(D) + 8B];
>
> where dep_dp is of type:
>
> typedef struct
> {
> unsigned int Drange;
> unsigned int Dvalue;
> int DbitsLeft;
> byte *Dcodestrm;
> int *Dcodestrm_len;
> } DecodingEnvironment;
>
Sorry, dep_dp_76(D) + 8B corresponds to DbitsLeft. But again, both places where
address is taken don't cast the pointer or so:
benchspec/CPU/525.x264_r/src/ldecod_src/biaridecod.c: int *DbitsLeft =
&dep->DbitsLeft;
benchspec/CPU/525.x264_r/src/ldecod_src/biaridecod.c: int *DbitsLeft =
&dep->DbitsLeft;
^ permalink raw reply [flat|nested] 14+ messages in thread* [Bug ipa/97335] [11 Regression] 525.x264_r fails with -Ofast -g -march=znver2 -flto=16 since r11-3641-gc34db4b6f8a5d803
2020-10-08 11:54 [Bug ipa/97335] New: [11 Regression] 525.x264_r fails with -Ofast -g -march=znver2 -flto=16 since r11-3641-gc34db4b6f8a5d803 marxin at gcc dot gnu.org
` (5 preceding siblings ...)
2020-10-09 8:31 ` marxin at gcc dot gnu.org
@ 2020-10-09 8:47 ` marxin at gcc dot gnu.org
2020-10-09 8:48 ` marxin at gcc dot gnu.org
` (5 subsequent siblings)
12 siblings, 0 replies; 14+ messages in thread
From: marxin at gcc dot gnu.org @ 2020-10-09 8:47 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97335
--- Comment #6 from Martin Liška <marxin at gcc dot gnu.org> ---
PRE dump files:
https://drive.google.com/file/d/1pWbEPfAqIThlTdEYWJ_g2yvDowqUSa-Q/view?usp=sharing
https://drive.google.com/file/d/1pWbEPfAqIThlTdEYWJ_g2yvDowqUSa-Q/view?usp=sharing
^ permalink raw reply [flat|nested] 14+ messages in thread* [Bug ipa/97335] [11 Regression] 525.x264_r fails with -Ofast -g -march=znver2 -flto=16 since r11-3641-gc34db4b6f8a5d803
2020-10-08 11:54 [Bug ipa/97335] New: [11 Regression] 525.x264_r fails with -Ofast -g -march=znver2 -flto=16 since r11-3641-gc34db4b6f8a5d803 marxin at gcc dot gnu.org
` (6 preceding siblings ...)
2020-10-09 8:47 ` marxin at gcc dot gnu.org
@ 2020-10-09 8:48 ` marxin at gcc dot gnu.org
2020-10-09 8:54 ` marxin at gcc dot gnu.org
` (4 subsequent siblings)
12 siblings, 0 replies; 14+ messages in thread
From: marxin at gcc dot gnu.org @ 2020-10-09 8:48 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97335
--- Comment #7 from Martin Liška <marxin at gcc dot gnu.org> ---
(In reply to Martin Liška from comment #6)
> PRE dump files:
>
> https://drive.google.com/file/d/1pWbEPfAqIThlTdEYWJ_g2yvDowqUSa-Q/
> view?usp=sharing
>
> https://drive.google.com/file/d/1pWbEPfAqIThlTdEYWJ_g2yvDowqUSa-Q/
> view?usp=sharing
The second one is:
https://drive.google.com/file/d/11qmfbZCgWrkkpuTxRxFE6gq1Bcd3GkHf/view?usp=sharing
^ permalink raw reply [flat|nested] 14+ messages in thread* [Bug ipa/97335] [11 Regression] 525.x264_r fails with -Ofast -g -march=znver2 -flto=16 since r11-3641-gc34db4b6f8a5d803
2020-10-08 11:54 [Bug ipa/97335] New: [11 Regression] 525.x264_r fails with -Ofast -g -march=znver2 -flto=16 since r11-3641-gc34db4b6f8a5d803 marxin at gcc dot gnu.org
` (7 preceding siblings ...)
2020-10-09 8:48 ` marxin at gcc dot gnu.org
@ 2020-10-09 8:54 ` marxin at gcc dot gnu.org
2020-10-09 9:17 ` marxin at gcc dot gnu.org
` (3 subsequent siblings)
12 siblings, 0 replies; 14+ messages in thread
From: marxin at gcc dot gnu.org @ 2020-10-09 8:54 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97335
--- Comment #8 from Martin Liška <marxin at gcc dot gnu.org> ---
The change is:
ipa-modref: in read_MVD_CABAC/1761, call to exp_golomb_decode_eq_prob/1803 does
not clobber 3->3
^ permalink raw reply [flat|nested] 14+ messages in thread* [Bug ipa/97335] [11 Regression] 525.x264_r fails with -Ofast -g -march=znver2 -flto=16 since r11-3641-gc34db4b6f8a5d803
2020-10-08 11:54 [Bug ipa/97335] New: [11 Regression] 525.x264_r fails with -Ofast -g -march=znver2 -flto=16 since r11-3641-gc34db4b6f8a5d803 marxin at gcc dot gnu.org
` (8 preceding siblings ...)
2020-10-09 8:54 ` marxin at gcc dot gnu.org
@ 2020-10-09 9:17 ` marxin at gcc dot gnu.org
2020-10-09 9:25 ` hubicka at gcc dot gnu.org
` (2 subsequent siblings)
12 siblings, 0 replies; 14+ messages in thread
From: marxin at gcc dot gnu.org @ 2020-10-09 9:17 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97335
--- Comment #9 from Martin Liška <marxin at gcc dot gnu.org> ---
So adding only -fno-ipa-cp fixes the problem!
^ permalink raw reply [flat|nested] 14+ messages in thread* [Bug ipa/97335] [11 Regression] 525.x264_r fails with -Ofast -g -march=znver2 -flto=16 since r11-3641-gc34db4b6f8a5d803
2020-10-08 11:54 [Bug ipa/97335] New: [11 Regression] 525.x264_r fails with -Ofast -g -march=znver2 -flto=16 since r11-3641-gc34db4b6f8a5d803 marxin at gcc dot gnu.org
` (9 preceding siblings ...)
2020-10-09 9:17 ` marxin at gcc dot gnu.org
@ 2020-10-09 9:25 ` hubicka at gcc dot gnu.org
2020-10-09 10:27 ` cvs-commit at gcc dot gnu.org
2020-10-09 10:28 ` marxin at gcc dot gnu.org
12 siblings, 0 replies; 14+ messages in thread
From: hubicka at gcc dot gnu.org @ 2020-10-09 9:25 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97335
--- Comment #10 from Jan Hubicka <hubicka at gcc dot gnu.org> ---
Created attachment 49338
--> https://gcc.gnu.org/bugzilla/attachment.cgi?id=49338&action=edit
disable param tracking on clones
Perhaps this has chance to help.
^ permalink raw reply [flat|nested] 14+ messages in thread* [Bug ipa/97335] [11 Regression] 525.x264_r fails with -Ofast -g -march=znver2 -flto=16 since r11-3641-gc34db4b6f8a5d803
2020-10-08 11:54 [Bug ipa/97335] New: [11 Regression] 525.x264_r fails with -Ofast -g -march=znver2 -flto=16 since r11-3641-gc34db4b6f8a5d803 marxin at gcc dot gnu.org
` (10 preceding siblings ...)
2020-10-09 9:25 ` hubicka at gcc dot gnu.org
@ 2020-10-09 10:27 ` cvs-commit at gcc dot gnu.org
2020-10-09 10:28 ` marxin at gcc dot gnu.org
12 siblings, 0 replies; 14+ messages in thread
From: cvs-commit at gcc dot gnu.org @ 2020-10-09 10:27 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97335
--- Comment #11 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Martin Liska <marxin@gcc.gnu.org>:
https://gcc.gnu.org/g:ffe8baa996486fa0313aa804a064a58b0b161f07
commit r11-3748-gffe8baa996486fa0313aa804a064a58b0b161f07
Author: Jan Hubicka <jh@suse.cz>
Date: Fri Oct 9 11:29:58 2020 +0200
IPA modref: fix miscompilation in clone when IPA modref is used
gcc/ChangeLog:
PR ipa/97292
PR ipa/97335
* ipa-modref-tree.h (copy_from): Drop summary in a
clone.
^ permalink raw reply [flat|nested] 14+ messages in thread* [Bug ipa/97335] [11 Regression] 525.x264_r fails with -Ofast -g -march=znver2 -flto=16 since r11-3641-gc34db4b6f8a5d803
2020-10-08 11:54 [Bug ipa/97335] New: [11 Regression] 525.x264_r fails with -Ofast -g -march=znver2 -flto=16 since r11-3641-gc34db4b6f8a5d803 marxin at gcc dot gnu.org
` (11 preceding siblings ...)
2020-10-09 10:27 ` cvs-commit at gcc dot gnu.org
@ 2020-10-09 10:28 ` marxin at gcc dot gnu.org
12 siblings, 0 replies; 14+ messages in thread
From: marxin at gcc dot gnu.org @ 2020-10-09 10:28 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97335
Martin Liška <marxin at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Resolution|--- |FIXED
Status|NEW |RESOLVED
--- Comment #12 from Martin Liška <marxin at gcc dot gnu.org> ---
Fixed now.
^ permalink raw reply [flat|nested] 14+ messages in thread