[Bug target/71318] Can't disable ISAs via function attribute

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71318

Andrew Pinski  changed:

   What|Removed |Added

 Status|UNCONFIRMED |RESOLVED
   Target Milestone|--- |11.0
 Resolution|--- |FIXED

--- Comment #4 from Andrew Pinski  ---
This is definitely fixed for GCC 11 because
testsuite/gcc.target/i386/pr96744-1.c is the same as one of your testcases in
the patch.

[Bug c++/101904] Wrong result of decltype during instantiation of std::result_of

[redi at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101904

Jonathan Wakely  changed:

   What|Removed |Added

 Status|UNCONFIRMED |RESOLVED
 Resolution|--- |INVALID

--- Comment #1 from Jonathan Wakely  ---
Not a bug. The call to zzz.foo had to do overload resolution to decide which
foo to call. That means both result_of types get instantiated. That means the
lambda gets instantiated with an argument of int* as well as with an argument
of const int*. Without an explicit return type the body of the lambda is
instantiated, which fails the static_assert for the int* case

[Bug target/70973] x86: Can the __atomic_*() operations be made to list the LOCK prefixes?

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70973

Andrew Pinski  changed:

   What|Removed |Added

   Severity|normal  |enhancement

[Bug target/70797] gcc.exe: Internal compiler error: (program as)

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=70797

Andrew Pinski  changed:

   What|Removed |Added

 Status|UNCONFIRMED |RESOLVED
 Resolution|--- |INVALID

--- Comment #3 from Andrew Pinski  ---
The assembler is crash, GCC project does not include the assembler.

[Bug target/64749] [4.8 Regression] "truncating" instructions generated instead of a load one using SSE & AVX2 intrinsics

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64749

Andrew Pinski  changed:

   What|Removed |Added

 Status|UNCONFIRMED |RESOLVED
 Resolution|--- |FIXED
   Target Milestone|--- |4.9.0

--- Comment #2 from Andrew Pinski  ---
regmove was completely removed from GCC 4.9.0+ in r0-126146.
So nobody fixed it just removed it fully.

[Bug target/64749] [4.8 Regression] "truncating" instructions generated instead of a load one using SSE & AVX2 intrinsics

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64749

--- Comment #1 from Andrew Pinski  ---
Regmove takes:
(insn 23 21 24 4 (set (reg:V8HI 69 [ D.8600 ])
(mem:V8HI (plus:SI (mult:SI (reg/v:SI 67 [ i ])
(const_int 4 [0x4]))
(reg/v/f:SI 83 [ a ])) [0 *_7+0 S16 A128]))
/opt/compiler-explorer/gcc-4.8.5/lib/gcc/x86_64-linux-gnu/4.8.5/include/emmintrin.h:1376
1142 {*movv8hi_internal}
 (nil))
(insn 24 23 27 4 (set (reg:V8HI 88)
(truncate:V8HI (lshiftrt:V8SI (mult:V8SI (zero_extend:V8SI (reg:V8HI 69
[ D.8600 ]))
(zero_extend:V8SI (mem:V8HI (plus:SI (mult:SI (reg/v:SI 67
[ i ])
(const_int 4 [0x4]))
(reg/v/f:SI 85 [ yprime ])) [0 *_13+0 S16
A128])))
(const_int 16 [0x10]
/opt/compiler-explorer/gcc-4.8.5/lib/gcc/x86_64-linux-gnu/4.8.5/include/emmintrin.h:1376
1500 {*umulv8hi3_highpart}
 (nil))
(insn 27 24 29 4 (set (reg:V8SI 89 [ D.8600 ])
(zero_extend:V8SI (reg:V8HI 69 [ D.8600 ])))
/opt/compiler-explorer/gcc-4.8.5/lib/gcc/x86_64-linux-gnu/4.8.5/include/avx2intrin.h:499
1763 {avx2_zero_extendv8hiv8si2}
 (expr_list:REG_DEAD (reg:V8HI 69 [ D.8600 ])
(nil)))

And converts it to:
(insn 23 21 24 4 (set (reg:V8SI 69 [ D.8600 ])
(zero_extend:V8SI (mem:V8HI (plus:SI (mult:SI (reg/v:SI 67 [ i ])
(const_int 4 [0x4]))
(reg/v/f:SI 83 [ a ])) [0 *_7+0 S16 A128])))
/opt/compiler-explorer/gcc-4.8.5/lib/gcc/x86_64-linux-gnu/4.8.5/include/emmintrin.h:1376
1763 {avx2_zero_extendv8hiv8si2}
 (nil))
(insn 24 23 27 4 (set (reg:V8HI 88)
(truncate:V8HI (lshiftrt:V8SI (mult:V8SI (zero_extend:V8SI (subreg:V8HI
(reg:V8SI 69 [ D.8600 ]) 0))
(zero_extend:V8SI (mem:V8HI (plus:SI (mult:SI (reg/v:SI 67
[ i ])
(const_int 4 [0x4]))
(reg/v/f:SI 85 [ yprime ])) [0 *_13+0 S16
A128])))
(const_int 16 [0x10]
/opt/compiler-explorer/gcc-4.8.5/lib/gcc/x86_64-linux-gnu/4.8.5/include/emmintrin.h:1376
1500 {*umulv8hi3_highpart}
 (nil))
(insn 27 24 29 4 (set (reg:V8SI 89 [ D.8600 ])
(reg:V8SI 69 [ D.8600 ]))
/opt/compiler-explorer/gcc-4.8.5/lib/gcc/x86_64-linux-gnu/4.8.5/include/avx2intrin.h:499
1143 {*movv8si_internal}
 (expr_list:REG_DEAD (reg:V8SI 69 [ D.8600 ])
(nil)))

[Bug target/59062] poor code generated at -Os (affecting all gcc versions): 10+ sec vs. 1 sec at -O0

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=59062

Andrew Pinski  changed:

   What|Removed |Added

 Resolution|--- |FIXED
   Target Milestone|--- |5.0
 Status|NEW |RESOLVED

--- Comment #2 from Andrew Pinski  ---
Fixed in GCC 5+.
LIM is able to handle these loops now at -Os.

[Bug target/59968] Unused BT patterns

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=59968

--- Comment #20 from Andrew Pinski  ---
I think most of these have been fixed.  For an example comment #19 was fixed in
r12-1525

[Bug target/53228] [4.6/4.7/4.8 Regression] target attributes in libcpp/lex.c cause illegal instructions to be used elsewhere

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=53228

Andrew Pinski  changed:

   What|Removed |Added

 CC||michael at talosis dot ca

--- Comment #13 from Andrew Pinski  ---
*** Bug 48326 has been marked as a duplicate of this bug. ***

[Bug target/48326] Target attribute leaks from function pointers

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=48326

Andrew Pinski  changed:

   What|Removed |Added

 Resolution|FIXED   |DUPLICATE

--- Comment #8 from Andrew Pinski  ---
Dup of bug 53228.

*** This bug has been marked as a duplicate of bug 53228 ***

[Bug target/53228] [4.6/4.7/4.8 Regression] target attributes in libcpp/lex.c cause illegal instructions to be used elsewhere

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=53228

Andrew Pinski  changed:

   What|Removed |Added

 CC||aivchenk at gmail dot com

--- Comment #12 from Andrew Pinski  ---
*** Bug 55816 has been marked as a duplicate of this bug. ***

[Bug target/55816] Options from command line are added to target arch attribute eventhough they are in contradiction with that target

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55816

Andrew Pinski  changed:

   What|Removed |Added

 Resolution|--- |DUPLICATE
 Status|UNCONFIRMED |RESOLVED

--- Comment #1 from Andrew Pinski  ---
Dup of bug 53228.

*** This bug has been marked as a duplicate of bug 53228 ***

[Bug target/56298] wmmintrin.h aborts compilation on the machines without AES

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=56298

Andrew Pinski  changed:

   What|Removed |Added

 Status|NEW |RESOLVED
   Target Milestone|--- |4.9.0
 Resolution|--- |FIXED

--- Comment #11 from Andrew Pinski  ---
Fixed in r0-124016.

[Bug libstdc++/56538] no way to disable shared_ptr's use of atomics

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=56538

Andrew Pinski  changed:

   What|Removed |Added

  Component|target  |libstdc++
Summary|No option to disable slow   |no way to disable
   |'lock' instr. one does not  |shared_ptr's use of atomics
   |need|

--- Comment #1 from Andrew Pinski  ---
I really doubt we want to have this ability these days. But I will let the
libstdc++ folks decide.

[Bug target/55803] SSE2 double negation less efficient than explicit xor

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55803

Andrew Pinski  changed:

   What|Removed |Added

 Status|UNCONFIRMED |NEW
   Last reconfirmed||2021-08-15
 Ever confirmed|0   |1
   Severity|normal  |enhancement

--- Comment #3 from Andrew Pinski  ---
ICC does this trick while clang and MSVC don't.

[Bug target/58044] -mno-see2avx does not seems to work

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=58044

Andrew Pinski  changed:

   What|Removed |Added

 Status|UNCONFIRMED |RESOLVED
 Resolution|--- |INVALID

--- Comment #2 from Andrew Pinski  ---

-msse2avx
-mno-sse2avx
Specify that the assembler should encode SSE instructions with VEX prefix. The
option -mavx turns this on by default.

And:
GCC depresses SSEx instructions when -mavx is used. Instead, it generates new
AVX instructions or AVX equivalence for all SSEx instructions when needed.

So this is documented this way.

[Bug target/41202] -mno-mmx doesn't turn off sse*

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=41202

Andrew Pinski  changed:

   What|Removed |Added

   Keywords||documentation

--- Comment #2 from Andrew Pinski  ---
I think this is just a documentation issue. sse does not really depend on mmx
any more.

[Bug target/53133] XOR AL,AL to zero lower 8 bits of EAX/RAX causes partial register stall (Intel Core 2)

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=53133

Andrew Pinski  changed:

   What|Removed |Added

  Known to fail||5.5.0
 Status|NEW |RESOLVED
  Known to work||6.1.0
   Target Milestone|--- |6.0
   Keywords||missed-optimization
 Resolution|--- |FIXED

--- Comment #5 from Andrew Pinski  ---
.L2:
movlmem(%rip), %eax
shrq$2, %rax
andq%rcx, %rax
addq%rax, %rsi
subl$1, %edx
jne .L2

So both versions now match up because of r6-3841.

So closing as fixed.

[Bug target/49839] Use constants in registers preferably to inline constants (-Os)

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=49839

Andrew Pinski  changed:

   What|Removed |Added

   Severity|minor   |enhancement
 Ever confirmed|0   |1
   Keywords||missed-optimization
   Last reconfirmed||2021-08-15
 Status|UNCONFIRMED |NEW

--- Comment #2 from Andrew Pinski  ---
Trunk produces (after r12-1712):

xorl%eax, %eax
movl%eax, b
movl%eax, a
xorl%eax, %eax

So almost there.

[Bug target/48970] Inaccurate comments for processor_costs

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=48970

--- Comment #2 from Andrew Pinski  ---
  const int sse_load[5];/* cost of loading SSE register
   in 32bit, 64bit, 128bit, 256bit and 512bit
*/
  const int sse_store[5];   /* cost of storing SSE register
   in SImode, DImode and TImode.  */





(In reply to H.J. Lu from comment #1)
> Also
> 
>  const int mmxsse_to_integer;  /* cost of moving mmxsse register to
>integer and vice versa.  */

This changed with PR 90877.

[Bug target/28808] Alignment problem in __gthread_once_t in vxWorks

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=28808

Andrew Pinski  changed:

   What|Removed |Added

   Target Milestone|--- |4.5.0
 Status|NEW |RESOLVED
 Resolution|--- |FIXED

--- Comment #4 from Andrew Pinski  ---
Fixed in r0-92956 .

[Bug target/33587] INIT_SECTION_ASM_OP defined twice in tm.texi

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=33587

Andrew Pinski  changed:

   What|Removed |Added

 Ever confirmed|0   |1
   Last reconfirmed||2021-08-15
 Status|UNCONFIRMED |NEW

--- Comment #2 from Andrew Pinski  ---
Confirmed, it has been there since tm.texi was added to the source control back
in 1996.

[Bug target/34967] internal compiler error: in frv_print_operand, at config/frv/frv.c:3057

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=34967

Andrew Pinski  changed:

   What|Removed |Added

 Ever confirmed|0   |1
   Last reconfirmed||2021-08-15
 Status|UNCONFIRMED |NEW

--- Comment #4 from Andrew Pinski  ---
Confirmed, 

instead of fatal_insn, output_operand_lossage should be used.

[Bug target/32097] m32c: ICE in expand_builtin_apply, at builtins.c:1525 building libobjc

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=32097

Andrew Pinski  changed:

   What|Removed |Added

 Resolution|--- |FIXED
 Status|UNCONFIRMED |RESOLVED
   Target Milestone|--- |4.4.0

--- Comment #3 from Andrew Pinski  ---
Fixed at r0-87561.

[Bug target/27055] Structures are copied byte by byte into function arguments

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=27055

Andrew Pinski  changed:

   What|Removed |Added

   Keywords||missed-optimization
 Status|NEW |RESOLVED
   Target Milestone|--- |4.4.0
 Resolution|--- |FIXED

--- Comment #6 from Andrew Pinski  ---
Been fixed at least since 4.4.0:
movl%esp, %edi
rep movsl

Maybe even earlier.

[Bug target/86084] trunk/gcc/config/m32c/m32c.c:4368: suspicious comparison ?

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=86084

Andrew Pinski  changed:

   What|Removed |Added

 Status|UNCONFIRMED |NEW
 Ever confirmed|0   |1
   Last reconfirmed||2021-08-15

--- Comment #1 from Andrew Pinski  ---
Still there as of today.

[Bug target/90103] ICE building Glibc's e_atan2f.c with -O2 -mcpu=hs38_linux -frounding-math

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90103

Andrew Pinski  changed:

   What|Removed |Added

 Resolution|--- |FIXED
 Status|UNCONFIRMED |RESOLVED
   Keywords||ice-on-valid-code
   Target Milestone|--- |9.0

--- Comment #3 from Andrew Pinski  ---
Fixed in GCC 9, 8.x is no longer supported.

[Bug target/91591] Arc: ICE in trunc_int_for_mode, at explow.c:60

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91591

Andrew Pinski  changed:

   What|Removed |Added

 Ever confirmed|1   |0
 Status|WAITING |UNCONFIRMED

[Bug target/25690] error: unrecognizable insn - failed to assign &a[1] if a is a struct

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=25690

Andrew Pinski  changed:

   What|Removed |Added

 CC||r.slater at freescale dot com

--- Comment #7 from Andrew Pinski  ---
*** Bug 31702 has been marked as a duplicate of this bug. ***

[Bug target/31702] Pointer Arithmetic Crash When Optimizing

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=31702

Andrew Pinski  changed:

   What|Removed |Added

 Resolution|--- |DUPLICATE
 Status|UNCONFIRMED |RESOLVED

--- Comment #2 from Andrew Pinski  ---
Dup of bug 25690.

*** This bug has been marked as a duplicate of bug 25690 ***

[Bug middle-end/91883] Division by a constant could be optimized for known variables value range

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=91883

Andrew Pinski  changed:

   What|Removed |Added

   Severity|normal  |enhancement
   Last reconfirmed|2019-09-25 00:00:00 |2021-8-14

[Bug middle-end/92574] Inefficient code for multidimensional array assess

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92574

Andrew Pinski  changed:

   What|Removed |Added

 Depends on||85017

--- Comment #3 from Andrew Pinski  ---
Basically PR 85017


Referenced Bugs:

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85017
[Bug 85017] Missed constant propagation to index of array reference

[Bug tree-optimization/85017] Missed constant propagation to index of array reference

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85017

Andrew Pinski  changed:

   What|Removed |Added

 CC||ubizjak at gmail dot com

--- Comment #2 from Andrew Pinski  ---
*** Bug 87055 has been marked as a duplicate of this bug. ***

[Bug middle-end/87055] Unoptimal address calculation

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=87055

Andrew Pinski  changed:

   What|Removed |Added

 Resolution|--- |DUPLICATE
 Status|NEW |RESOLVED

--- Comment #2 from Andrew Pinski  ---
Dup of bug 85017

*** This bug has been marked as a duplicate of bug 85017 ***

[Bug middle-end/87055] Unoptimal address calculation

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=87055

Andrew Pinski  changed:

   What|Removed |Added

   Last reconfirmed|2018-08-22 00:00:00 |2021-8-14
   Severity|normal  |enhancement

[Bug middle-end/82153] missed optimization: double truncating

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82153

Andrew Pinski  changed:

   What|Removed |Added

   Last reconfirmed|2017-09-12 00:00:00 |2021-8-14
   Severity|normal  |enhancement

[Bug middle-end/80342] useless outermost conversions not fully elided by genmatch generated code

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=80342

Andrew Pinski  changed:

   What|Removed |Added

   Last reconfirmed|2017-04-06 00:00:00 |2021-8-14
 CC||pinskia at gcc dot gnu.org

--- Comment #2 from Andrew Pinski  ---
Yes I noticed this too and I had to worked around this in phiopt too:
https://gcc.gnu.org/pipermail/gcc-patches/2021-July/574783.html

[Bug tree-optimization/96912] Failure to optimize pblendvb pattern

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96912

Andrew Pinski  changed:

   What|Removed |Added

   Last reconfirmed|2020-09-03 00:00:00 |2021-8-14
   Severity|normal  |enhancement

[Bug tree-optimization/97464] Missed redundant store optimization opportunity

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97464

Andrew Pinski  changed:

   What|Removed |Added

 Depends on||88854
   Last reconfirmed|2020-10-19 00:00:00 |2021-8-14
   Severity|normal  |enhancement

--- Comment #2 from Andrew Pinski  ---
(In reply to Richard Biener from comment #1)
> And I believe we have a duplicate report somewhere.

Yes I suspect it is the same as PR 88854.


Referenced Bugs:

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88854
[Bug 88854] redundant store after load that would makes aliasing UB

[Bug tree-optimization/95929] Failure to optimize tautological comparisons of comparisons to a single one

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95929

Andrew Pinski  changed:

   What|Removed |Added

 Status|UNCONFIRMED |NEW
   Last reconfirmed||2021-08-15
   Severity|normal  |enhancement
 Ever confirmed|0   |1

--- Comment #2 from Andrew Pinski  ---
Confirmed.

[Bug tree-optimization/95926] Failure to optimize xor pattern when using temporary variable

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95926

Andrew Pinski  changed:

   What|Removed |Added

   Last reconfirmed|2020-06-29 00:00:00 |2021-8-14
   Severity|normal  |enhancement

[Bug tree-optimization/95424] Failure to optimize division with numerator of 1

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95424

Andrew Pinski  changed:

   What|Removed |Added

   Severity|normal  |enhancement
   Last reconfirmed|2020-06-02 00:00:00 |2021-8-14

[Bug tree-optimization/94920] Failure to optimize abs pattern from arithmetic with selected operands based on comparisons with 0

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94920

Andrew Pinski  changed:

   What|Removed |Added

   Last reconfirmed|2020-05-04 00:00:00 |2021-8-14
   Severity|normal  |enhancement

--- Comment #1 from Andrew Pinski  ---
  if (x_4(D) <= 0)
goto ; [INV]
  else
goto ; [INV]

   :
  iftmp.0_6 = -x_4(D);

   :
  # iftmp.0_3 = PHI 

x <= 0 ? -x : 0 isn't this MAX<-x, 0>?

So there are two issues here, first is the above one, the second one is
MAX + MAX<-a,0> is not optimized to just MAX<-a,a> or just ABS.

E.g.:
unsigned f1(int x)
{
   int t = (x >= 0 ? x : 0) ;
   int tt = -x;
   int t1 = (tt >= 0 ? tt : 0) ;
return t + t1;
}

[Bug tree-optimization/94914] Failure to optimize check of high part of 64-bit result of 32 by 32 multiplication into overflow check

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94914

Andrew Pinski  changed:

   What|Removed |Added

   Keywords||missed-optimization
   Last reconfirmed|2020-05-04 00:00:00 |2021-8-14
   Severity|normal  |enhancement

[Bug middle-end/19987] [meta-bug] fold missing optimizations in general

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=19987
Bug 19987 depends on bug 94877, which changed state.

Bug 94877 Summary: Failure to simplify ~(x + 1) to -2 - x
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94877

   What|Removed |Added

 Status|NEW |RESOLVED
 Resolution|--- |FIXED

[Bug tree-optimization/94877] Failure to simplify ~(x + 1) to -2 - x

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94877

Andrew Pinski  changed:

   What|Removed |Added

 Depends on||96685
   Target Milestone|--- |11.0
 Status|NEW |RESOLVED
 Resolution|--- |FIXED

--- Comment #5 from Andrew Pinski  ---
Fixed in GCC 11+ by r11-5958 aka PR 96685.


Referenced Bugs:

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96685
[Bug 96685] Failure to optimize not+sub to add+not

[Bug tree-optimization/89749] Very odd vector constructor

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=89749

Andrew Pinski  changed:

   What|Removed |Added

 Status|NEW |ASSIGNED
   Assignee|unassigned at gcc dot gnu.org  |pinskia at gcc dot 
gnu.org

--- Comment #4 from Andrew Pinski  ---
Mine for GCC 13.

[Bug tree-optimization/58497] SLP vectorizes identical operations

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=58497

Andrew Pinski  changed:

   What|Removed |Added

   Assignee|unassigned at gcc dot gnu.org  |pinskia at gcc dot 
gnu.org
 Status|NEW |ASSIGNED

--- Comment #13 from Andrew Pinski  ---
Mine for GCC 13, I have patches which turn:

  W_6 = BIT_INSERT_EXPR ;
  W_7 = BIT_INSERT_EXPR ;
  W_8 = BIT_INSERT_EXPR ;
  W_9 = BIT_INSERT_EXPR ;
Into:
W_9 = {_2,_2,_2,_2};

This improvement deals with bitfields but vectors have a similar issue with
Bit_inserts so I deal with it there.

[Bug tree-optimization/89749] Very odd vector constructor

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=89749

Andrew Pinski  changed:

   What|Removed |Added

   Severity|normal  |enhancement
  Known to fail||10.3.0

--- Comment #3 from Andrew Pinski  ---
SLP is able to figure this out in GCC 11+.

[Bug tree-optimization/88605] vector extensions: Widening or conversion generates inefficient or scalar code.

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88605

Andrew Pinski  changed:

   What|Removed |Added

   Severity|normal  |enhancement

[Bug middle-end/85559] [meta-bug] Improve conditional move

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=85559
Bug 85559 depends on bug 65492, which changed state.

Bug 65492 Summary: Bad optimization in -O3 due to if-conversion and/or unrolling
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65492

   What|Removed |Added

 Status|NEW |RESOLVED
 Resolution|--- |FIXED

[Bug tree-optimization/65492] Bad optimization in -O3 due to if-conversion and/or unrolling

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65492

Andrew Pinski  changed:

   What|Removed |Added

   Target Milestone|--- |6.0
 Resolution|--- |FIXED
 Status|NEW |RESOLVED
  Known to work||6.1.0
   See Also||https://gcc.gnu.org/bugzill
   ||a/show_bug.cgi?id=66002

--- Comment #13 from Andrew Pinski  ---
This is fixed in GCC 6, we produce:

  _76 = MIN_EXPR ;
  offset_75 = MAX_EXPR <_76, 1>;
...

  _104 = offset_75 + -1;

Rather than:

  offset_30 = _52 <= 254 ? offset_67 : 1;
  prephitmp_119 = _52 <= 254 ? pretmp_118 : 0;
  _17 = offset_67 <= 255;
  offset_69 = _17 ? offset_30 : 255;
  prephitmp_109 = _17 ? prephitmp_119 : 254;

This was fixed by r6-528.  PR 66002 is describing almost the same issue even.

[Bug tree-optimization/79958] Missed tree DSE

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=79958

Andrew Pinski  changed:

   What|Removed |Added

   Last reconfirmed|2017-03-10 00:00:00 |2021-8-14
   Severity|normal  |enhancement

--- Comment #4 from Andrew Pinski  ---
With the reduced testcase in comment #2 and -O2 -fno-tree-dce we still get:
  x.i = 1;
  if (b_3(D) != 0)
goto ; [0.04%]
  else
goto ; [99.96%]

   [local count: 429496]:
  bar ();
  __builtin_abort ();

   [local count: 1073312329]:
  bar (); [tail call]
  x ={v} {CLOBBER};

As Richi mentioned tree level DCE can figure it out for the reduced testcase
while DSE cannot.

[Bug tree-optimization/84712] Missed evaluating to constant at tree level

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84712

Andrew Pinski  changed:

   What|Removed |Added

 Resolution|--- |FIXED
 Status|NEW |RESOLVED
   Target Milestone|--- |10.0

--- Comment #3 from Andrew Pinski  ---
Fixed by r10-1420 which added the late FRE even.

[Bug tree-optimization/83544] Missed optimization opportunity for constant folding

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=83544

Andrew Pinski  changed:

   What|Removed |Added

 Status|NEW |RESOLVED
   Target Milestone|--- |11.0
 Resolution|--- |FIXED

--- Comment #2 from Andrew Pinski  ---
Fixed 100% in 11+ by r11-6100, there was some other VRP patch back in GCC 9
which also fixed this too but I can't find it.

[Bug middle-end/101913] -Wstrict-overflow -O3 false alarm on tzdb localtime.c

[eggert at cs dot ucla.edu via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101913

--- Comment #3 from eggert at cs dot ucla.edu ---
(In reply to Andrew Pinski from comment #1)
> >-1L << 63 is LONG_MIN
> No it is undefined and has an overflow bit on it.
> You want (long)(-1UL << 63) for it be correct.
> But the warning is still there.

I updated the attachment to avoid any undefined behavior, by replacing -1L<<63
with -1-9223372036854775807L. As you say, the bug is still there.

> I thought -fsanitize=undefined enabled -fwrapv too ...

It doesn't. (Perhaps you meant -ftrapv?)

[Bug tree-optimization/94802] Failure to recognize identities with __builtin_clz

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=94802

Andrew Pinski  changed:

   What|Removed |Added

   Target Milestone|--- |11.0

[Bug tree-optimization/97964] Missed optimization opportunity for VRP

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=97964

Andrew Pinski  changed:

   What|Removed |Added

   Target Milestone|--- |11.0

[Bug tree-optimization/83544] Missed optimization opportunity for constant folding

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=83544

Andrew Pinski  changed:

   What|Removed |Added

   Severity|normal  |enhancement
  Known to fail||8.5.0
  Known to work||9.1.0, 9.4.0

[Bug target/99783] relocation truncated to fit: R_OR1K_GOT16 on OpenRISC, building libgeos

[cvs-commit at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99783

--- Comment #4 from CVS Commits  ---
The master branch has been updated by Stafford Horne :

https://gcc.gnu.org/g:da8a9d695b3c4b9397b5d9a27660bfa48af8d707

commit r12-2910-gda8a9d695b3c4b9397b5d9a27660bfa48af8d707
Author: Stafford Horne 
Date:   Wed Apr 21 05:33:15 2021 +0900

or1k: Use cmodel=large when building crtstuff

When linking gcc runtime objects into large binaries the link may fail
with the below errors.  This will happen even if we are building with
-mcmodel=large.

   
/home/shorne/work/openrisc/output/host/lib/gcc/or1k-buildroot-linux-uclibc/10.3.0/crtbeginS.o:
in function `deregister_tm_clones':
crtstuff.c:(.text+0x3c): relocation truncated to fit: R_OR1K_GOT16
against undefined symbol `_ITM_deregisterTMCloneTable'
   
/home/shorne/work/openrisc/output/host/lib/gcc/or1k-buildroot-linux-uclibc/10.3.0/crtbeginS.o:
in function `register_tm_clones':
crtstuff.c:(.text+0xc0): relocation truncated to fit: R_OR1K_GOT16
against undefined symbol `_ITM_registerTMCloneTable'

This patch builds the gcc crtstuff binaries always with the
-mcmodel=large option to ensure they can be linked into large binaries.

libgcc/ChangeLog:

PR target/99783
* config.host (or1k-*, tmake_file): Add or1k/t-crtstuff.
* config/or1k/t-crtstuff: New file.

[Bug target/99783] relocation truncated to fit: R_OR1K_GOT16 on OpenRISC, building libgeos

[cvs-commit at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99783

--- Comment #3 from CVS Commits  ---
The master branch has been updated by Stafford Horne :

https://gcc.gnu.org/g:eff8110674ef193481d3657456a262beeb9951ff

commit r12-2909-geff8110674ef193481d3657456a262beeb9951ff
Author: Stafford Horne 
Date:   Tue Apr 6 05:47:17 2021 +0900

or1k: Add mcmodel option to handle large GOTs

When building libgeos we get an error with:

linux-uclibc/9.3.0/crtbeginS.o: in function `__do_global_dtors_aux':
crtstuff.c:(.text+0x118): relocation truncated to fit: R_OR1K_GOT16
against symbol `__cxa_finalize' defined in .text section in
   
/home/shorne/work/openrisc/3eb9f9d0f6d8274b2d19753c006bd83f7d536e3c/output/host/or1k-buildroot-linux-uclibc/sysroot/lib/libc.so.

This is caused by GOT code having a limit of 64k.  In OpenRISC this
looks to be the only relocation code pattern to be limited to 64k.

This patch allows specifying a new option -mcmodel=large which can be
used to generate 2 more instructions to construct 32-bit addresses for
up to 4G GOTs.

gcc/ChangeLog:

PR target/99783
* config/or1k/or1k-opts.h: New file.
* config/or1k/or1k.c (or1k_legitimize_address_1, print_reloc):
Support generating gotha relocations if -mcmodel=large is
specified.
* config/or1k/or1k.h (TARGET_CMODEL_SMALL, TARGET_CMODEL_LARGE):
New macros.
* config/or1k/or1k.opt (mcmodel=): New option.
* doc/invoke.texi (OpenRISC Options): Document mcmodel.

[Bug middle-end/101913] -Wstrict-overflow -O3 false alarm on tzdb localtime.c

[eggert at cs dot ucla.edu via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101913

eggert at cs dot ucla.edu changed:

   What|Removed |Added

  Attachment #51304|0   |1
is obsolete||

--- Comment #2 from eggert at cs dot ucla.edu ---
Created attachment 51305
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=51305&action=edit
-Wstrict-overflow -O3 false alarm example (v2)

This updates the example by replacing -1L<<63 with -1-9223372036854775807L, to
avoid any undefined behavior in evaluating that constant expression.

[Bug tree-optimization/54345] jump threading leaks e->aux heap memory

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=54345

--- Comment #6 from Andrew Pinski  ---
There has been so many changes to this code since GCC 4.8.0 dealing with the
allocations of this pointer.  I can't tell if this has been fixed really.

[Bug c/101916] New: SH4 ICE: Segmentation fault signal terminated program cc1

[giulio.benetti at benettiengineering dot com via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101916

Bug ID: 101916
   Summary: SH4 ICE: Segmentation fault signal terminated program
cc1
   Product: gcc
   Version: 11.1.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: c
  Assignee: unassigned at gcc dot gnu.org
  Reporter: giulio.benetti at benettiengineering dot com
  Target Milestone: ---

When building package heirloom-mailx on buildroot we get this error:
'''
/home/giuliobenetti/br_reproduce/911f5c024834741754102ff1bbb05c4a64c54a0b/output/host/bin/sh4-linux-gcc
-D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64  -O2   -fPIC
-D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64
-DMAILRC='"/etc/nail.rc"' -DMAILSPOOL='"/var/mail"'
-DSENDMAIL='"/usr/lib/sendmail"'-c strings.c
sh4-linux-gcc.br_real: internal compiler error: Segmentation fault signal
terminated program cc1
Please submit a full bug report,
with preprocessed source if appropriate.
See  for instructions.
make[2]: *** [Makefile:80: mime.o] Error 4
make[2]: *** Waiting for unfinished jobs
make[1]: *** [package/pkg-generic.mk:250:
/home/giuliobenetti/br_reproduce/911f5c024834741754102ff1bbb05c4a64c54a0b/output/build/heirloom-mailx-12.5/.stamp_built]
Error 2
make: *** [Makefile:23: _all] Error 2
'''

To reproduce it:
'''
# git clone git://git.busybox.net/buildroot
# wget https://git.busybox.net/buildroot-test/tree/utils/br-reproduce-build

- modify BASE_GIT=... with your buildroot path in br-reproduce-build then:
# chmod a+x br-reproduce-build
# ./br-reproduce-build 911f5c024834741754102ff1bbb05c4a64c54a0b
'''

[Bug target/49064] [x86/x64]: broken alias analysis leads vectorizer to emit poor code

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=49064

Andrew Pinski  changed:

   What|Removed |Added

   Keywords||alias, missed-optimization
 Status|UNCONFIRMED |RESOLVED
  Component|tree-optimization   |target
 Resolution|--- |INVALID

--- Comment #5 from Andrew Pinski  ---
All compilers I could test that support __m128i and _mm_xor_si128 cause an
aliasing issue. So this is invalid.

[Bug lto/48475] gcc.dg/initpri3.c FAILs with -flto -O2 --param partial-inlining-entry-probability=100

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=48475

Andrew Pinski  changed:

   What|Removed |Added

 CC||marxin at gcc dot gnu.org
  Known to work||4.7.3
  Component|tree-optimization   |lto

--- Comment #1 from Andrew Pinski  ---
It seems to be working in GCC 4.7.3+.
I can reproduce the failure in GCC 4.6.4.

Someone would need to go back and figure out what fixed this if it is a still
latent bug or not.

[Bug rtl-optimization/45026] struct passed as argument in memory compiles to dead stores

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=45026

Andrew Pinski  changed:

   What|Removed |Added

   See Also||https://gcc.gnu.org/bugzill
   ||a/show_bug.cgi?id=83827

--- Comment #3 from Andrew Pinski  ---
There are around 5 or 6 more of these issues floating around about argument
passing and return values.  PR 83827 is just one.

[Bug rtl-optimization/45026] struct passed as argument in memory compiles to dead stores

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=45026

Andrew Pinski  changed:

   What|Removed |Added

   Keywords||missed-optimization
   Severity|normal  |enhancement

[Bug rtl-optimization/83827] vector load/store with struct in registers

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=83827

Andrew Pinski  changed:

   What|Removed |Added

   Severity|normal  |enhancement

[Bug tree-optimization/95731] Failure to optimize a >= 0 && b >= 0 to (a | b) >= 0

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95731

Andrew Pinski  changed:

   What|Removed |Added

 CC||bugzilla@poradnik-webmaster
   ||a.com

--- Comment #8 from Andrew Pinski  ---
*** Bug 83123 has been marked as a duplicate of this bug. ***

[Bug rtl-optimization/83123] Int compare - different asm code for different return type

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=83123

Andrew Pinski  changed:

   What|Removed |Added

 Resolution|--- |DUPLICATE
 Status|NEW |RESOLVED

--- Comment #4 from Andrew Pinski  ---
Handled by 95731.

*** This bug has been marked as a duplicate of bug 95731 ***

[Bug tree-optimization/95731] Failure to optimize a >= 0 && b >= 0 to (a | b) >= 0

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95731

Andrew Pinski  changed:

   What|Removed |Added

   Target Milestone|--- |11.0

[Bug rtl-optimization/83123] Int compare - different asm code for different return type

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=83123

--- Comment #3 from Andrew Pinski  ---
This is now optimized at the gimple level to just:

  _7 = a_3(D) & b_4(D);
  _8 = _7 < 0;

[Bug rtl-optimization/83123] Int compare - different asm code for different return type

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=83123

Andrew Pinski  changed:

   What|Removed |Added

  Known to fail||10.3.0
  Known to work||11.1.0
   Severity|normal  |enhancement
   Target Milestone|--- |11.0

[Bug rtl-optimization/82454] Possible future performance regression in x86 for 64-bit constant expansions

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82454

Andrew Pinski  changed:

   What|Removed |Added

   Severity|normal  |enhancement

[Bug rtl-optimization/82729] adjacent small objects can be initialized with a single store (but aren't for char a[] = "a")

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82729

Andrew Pinski  changed:

   What|Removed |Added

   Severity|normal  |enhancement

--- Comment #6 from Andrew Pinski  ---
This is what we produce on the gimple level:

  abc = "abc";
  ab = "ab";
  a = "a";


Interesting no compiler I see does this optimization for this code.

[Bug c/101915] New: Microblaze ICE: in extract_insn, at recog.c:2770

[giulio.benetti at benettiengineering dot com via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101915

Bug ID: 101915
   Summary: Microblaze ICE: in extract_insn, at recog.c:2770
   Product: gcc
   Version: unknown
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: c
  Assignee: unassigned at gcc dot gnu.org
  Reporter: giulio.benetti at benettiengineering dot com
  Target Milestone: ---

When building package lmbench on buildroot we get this error:
'''
/home/giuliobenetti/git/upstream/output-microblaze-lmbench-gcc11/host/bin/microblazeel-buildroot-linux-uclibc-gcc
-D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64  -O2 -g0 
-I/home/giuliobenetti/git/upstream/output-microblaze-lmbench-gcc11/host/bin/../microblazeel-buildroot-linux-uclibc/sysroot/usr/include/tirpc
 -fno-tree-loop-distribute-patterns -DHAVE_off64_t -DHAVE_lseek64 -DRUSAGE
-DHAVE_uint=1 -DHAVE_int64_t=1 -DHAVE_pmap_clnt_h -DHAVE_socklen_t
-DHAVE_DRAND48 -DHAVE_SCHED_SETAFFINITY=1   -o ../bin/microblazeel/stream
stream.c ../bin/microblazeel/lmbench.a
-L/home/giuliobenetti/git/upstream/output-microblaze-lmbench-gcc11/host/bin/../microblazeel-buildroot-linux-uclibc/sysroot/usr/lib
-ltirpc 
-L/home/giuliobenetti/git/upstream/output-microblaze-lmbench-gcc11/host/bin/../microblazeel-buildroot-linux-uclibc/sysroot/usr/lib
-ltirpc -lm
par_ops.c: In function ‘int64_bit_2’:
par_ops.c:94:1: error: unrecognizable insn:
   94 | }
  | ^
par_ops.c:99:9: note: in expansion of macro ‘BENCHMARK’
   99 | BENCHMARK(benchmark, 2, REPEAT_2)  
\
  | ^
par_ops.c:215:1: note: in expansion of macro ‘PARALLEL_BENCHMARKS’
  215 | PARALLEL_BENCHMARKS(int64_bit)
  | ^~~
(insn 353 354 352 3 (set (reg/v:DI 24 r24 [orig:74 i0 ] [74])
(plus:DI (reg:DI 26 r26 [orig:48 _9 ] [48])
(const_int -1 [0x]))) -1
 (nil))
during RTL pass: cprop_hardreg
par_ops.c:94:1: internal compiler error: in extract_insn, at recog.c:2770
   94 | }
  | ^
par_ops.c:99:9: note: in expansion of macro ‘BENCHMARK’
   99 | BENCHMARK(benchmark, 2, REPEAT_2)  
\
  | ^
par_ops.c:215:1: note: in expansion of macro ‘PARALLEL_BENCHMARKS’
  215 | PARALLEL_BENCHMARKS(int64_bit)
  | ^~~
0x122170e diagnostic_impl(rich_location*, diagnostic_metadata const*, int, char
const*, __va_list_tag (*) [1], diagnostic_t)
???:0
0x12225ce internal_error(char const*, ...)
???:0
0x57a818 fancy_abort(char const*, int, char const*)
???:0
0x55c92f _fatal_insn(char const*, rtx_def const*, char const*, int, char
const*)
???:0
0x55c951 _fatal_insn_not_found(rtx_def const*, char const*, int, char const*)
???:0
0x55bc9e extract_insn(rtx_insn*) [clone .cold]
???:0
0xa24fec extract_constrain_insn(rtx_insn*)
???:0
0xa292c7 copyprop_hardreg_forward_1(basic_block_def*, value_data*) [clone
.isra.0]
???:0
0xa29fbd (anonymous namespace)::pass_cprop_hardreg::execute(function*)
???:0
Please submit a full bug report,
with preprocessed source if appropriate.
Please include the complete backtrace with any bug report.
See  for instructions.
make[3]: *** [Makefile:453: ../bin/microblazeel/par_ops] Error 1
make[2]: *** [Makefile:114: lmbench] Error 2
make[1]: *** [package/pkg-generic.mk:270:
/home/giuliobenetti/git/upstream/output-microblaze-lmbench-gcc11/build/lmbench-3.0-a9/.stamp_built]
Error 2
make: *** [Makefile:23: _all] Error 2
'''

To reproduce the bug:
'''
# git clone git://git.busybox.net/buildroot
# wget https://git.busybox.net/buildroot-test/tree/utils/br-reproduce-build

- modify BASE_GIT=... with your buildroot path in br-reproduce-build then:
# chmod a+x br-reproduce-build
# ./br-reproduce-build c6849f0f45b32e1f484fa5442b9d313fff4b2e8c
'''

I've found that it can be worked around disabling the optimization passing -O0
to CFLAGS.

[Bug rtl-optimization/82356] auto-vectorizing pack of 16->8 has a redundant AND after a shift

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=82356

Andrew Pinski  changed:

   What|Removed |Added

  Component|tree-optimization   |rtl-optimization
   Severity|normal  |enhancement
   Last reconfirmed|2017-09-29 00:00:00 |2021-8-14

[Bug rtl-optimization/52781] duplicate pass names, rtl dce

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=52781

--- Comment #2 from Andrew Pinski  ---
4.7.0+, uses ud_dce and rtl_dce as the name of the RTL dumps now.

[Bug rtl-optimization/52781] duplicate pass names, rtl dce

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=52781

Andrew Pinski  changed:

   What|Removed |Added

 Status|UNCONFIRMED |RESOLVED
 Resolution|--- |FIXED
  Known to work|4.6.0   |4.7.0
  Known to fail|4.7.0   |4.6.0
   Target Milestone|--- |4.7.0

--- Comment #1 from Andrew Pinski  ---
t.c.116t.cdce  t.c.281r.ud_dce  t.c.316r.rtl_dce

r0-96801
r0-109214 is the full on fix.

4.6.0 was broken while 4.7.0 is fixed.

[Bug target/99783] relocation truncated to fit: R_OR1K_GOT16 on OpenRISC, building libgeos

[giulio.benetti at benettiengineering dot com via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99783

Giulio Benetti  changed:

   What|Removed |Added

 CC||giulio.benetti@benettiengin
   ||eering.com

--- Comment #2 from Giulio Benetti  ---
This bug is fixed by this patchset and passing -mcmodel=large to CFLAGS:
https://www.mail-archive.com/gcc-patches@gcc.gnu.org/msg259303.html

[Bug c++/101914] New: internal compiler error: in tsubst, at cp/pt.c:15553

[accounts at prantare dot xyz via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101914

Bug ID: 101914
   Summary: internal compiler error: in tsubst, at cp/pt.c:15553
   Product: gcc
   Version: 12.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: c++
  Assignee: unassigned at gcc dot gnu.org
  Reporter: accounts at prantare dot xyz
  Target Milestone: ---

The following (preprocessed) code

---
# 0 "main.cpp"
# 0 ""
# 0 ""
# 1 "/usr/include/stdc-predef.h" 1 3 4
# 0 "" 2
# 1 "main.cpp"
template
struct A {
int x;
};

template
A(T) -> A;

template
using B = decltype(A{v});

template
auto x = A>{};

int main()
{
x<0>;
}
---

yields the following error message:

---
main.cpp: In instantiation of ‘auto x<0>’:
main.cpp:17:5:   required from here
main.cpp:13:6: internal compiler error: in tsubst, at cp/pt.c:15553
   13 | auto x = A>{};
  |  ^
0x6eef60 tsubst(tree_node*, tree_node*, int, tree_node*)
../../gcc/gcc/cp/pt.c:15553
0xaeb4f4 tsubst_copy_and_build(tree_node*, tree_node*, int, tree_node*, bool,
bool)
../../gcc/gcc/cp/pt.c:20630
0xb08797 tsubst(tree_node*, tree_node*, int, tree_node*)
../../gcc/gcc/cp/pt.c:16094
0xb0b174 tsubst_template_args(tree_node*, tree_node*, int, tree_node*)
../../gcc/gcc/cp/pt.c:13355
0xb14683 tsubst_aggr_type
../../gcc/gcc/cp/pt.c:13559
0xaeb4f4 tsubst_copy_and_build(tree_node*, tree_node*, int, tree_node*, bool,
bool)
../../gcc/gcc/cp/pt.c:20630
0xaf62a7 tsubst_expr(tree_node*, tree_node*, int, tree_node*, bool)
../../gcc/gcc/cp/pt.c:19094
0xafd231 tsubst_expr(tree_node*, tree_node*, int, tree_node*, bool)
../../gcc/gcc/cp/pt.c:18108
0xafd231 tsubst_init
../../gcc/gcc/cp/pt.c:16452
0xafe252 regenerate_decl_from_template
../../gcc/gcc/cp/pt.c:25490
0xafe252 instantiate_body
../../gcc/gcc/cp/pt.c:25813
0xb001ea instantiate_decl(tree_node*, bool, bool)
../../gcc/gcc/cp/pt.c:26171
0x9e07aa maybe_instantiate_decl(tree_node*)
../../gcc/gcc/cp/decl2.c:5495
0x9e1739 maybe_instantiate_decl(tree_node*)
../../gcc/gcc/cp/decl2.c:5698
0x9e1739 mark_used(tree_node*, int)
../../gcc/gcc/cp/decl2.c:5724
0xb3f7b8 finish_id_expression_1
../../gcc/gcc/cp/semantics.c:4132
0xb3fbb7 finish_id_expression(tree_node*, tree_node*, tree_node*, cp_id_kind*,
bool, bool, bool*, bool, bool, bool, bool, char const**, unsigned int)
../../gcc/gcc/cp/semantics.c:4261
0xa9e152 cp_parser_primary_expression
../../gcc/gcc/cp/parser.c:6008
0xaa0aed cp_parser_postfix_expression
../../gcc/gcc/cp/parser.c:7566
0xa8a1aa cp_parser_binary_expression
../../gcc/gcc/cp/parser.c:9896
Please submit a full bug report,
with preprocessed source if appropriate.
Please include the complete backtrace with any bug report.
See  for instructions.
---

The command that triggers the bug: g++ main.cpp -std=c++20

Version et.c.:

---
$ g++ -v:

Using built-in specs.
COLLECT_GCC=g++
COLLECT_LTO_WRAPPER=/usr/local/libexec/gcc/x86_64-pc-linux-gnu/12.0.0/lto-wrapper
Target: x86_64-pc-linux-gnu
Configured with: ../gcc/configure --disable-multilib
Thread model: posix
Supported LTO compression algorithms: zlib
gcc version 12.0.0 20210813 (experimental) (GCC)
---

The bug appears in 11.1.0 too, on two different machines.
It compiles successfully with clang 14.

[Bug target/45434] x86 missed optimization: use high register (ah, bh, ch, dh) when available to make comparisons

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=45434

Andrew Pinski  changed:

   What|Removed |Added

  Component|rtl-optimization|target
 Target||i?86-*-* x86_64-*-*
   Severity|minor   |enhancement
   Keywords||missed-optimization

[Bug tree-optimization/49028] Missed optimization of pointer arithmetic

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=49028

Andrew Pinski  changed:

   What|Removed |Added

   Severity|minor   |enhancement
  Component|rtl-optimization|tree-optimization
   Last reconfirmed|2011-05-18 09:40:28 |2021-8-14

--- Comment #4 from Andrew Pinski  ---

  _3 = _2 + 8;
  _4 = _3 - _1;
  _5 = _4 /[ex] 8;
  _6 = _5 % 16;
  _7 = (long unsigned int) _6;
  _8 = _7 * 8;
  _9 = _1 + _8;

I Noticed only MSVC actually does this optimization.

[Bug middle-end/101913] -Wstrict-overflow -O3 false alarm on tzdb localtime.c

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101913

--- Comment #1 from Andrew Pinski  ---
>-1L << 63 is LONG_MIN
No it is undefined and has an overflow bit on it.
You want (long)(-1UL << 63) for it be correct.
But the warning is still there.

I thought -fsanitize=undefined enabled -fwrapv too ...

[Bug c/101913] New: -Wstrict-overflow -O3 false alarm on tzdb localtime.c

[eggert at cs dot ucla.edu via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101913

Bug ID: 101913
   Summary: -Wstrict-overflow -O3 false alarm on tzdb localtime.c
   Product: gcc
   Version: 11.2.1
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: c
  Assignee: unassigned at gcc dot gnu.org
  Reporter: eggert at cs dot ucla.edu
  Target Milestone: ---

Created attachment 51304
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=51304&action=edit
-Wstrict-overflow -O3 false alarms

I ran into this problem when compiling tzdb localtime.c with gcc (GCC) 11.2.1
20210728 (Red Hat 11.2.1-1) on x86-64. To reproduce, compile the attached
program v.i with:

gcc -Wstrict-overflow -O3 -fsanitize=undefined -S v.i

The output is:

v.i: In function ‘tzloadbody’:
v.i:22:1: warning: assuming signed overflow does not occur when simplifying
con\
ditional to constant [-Wstrict-overflow]
   22 | tzloadbody (struct state *sp, int leapcnt0, int timecnt0)
  | ^~
v.i:29:25: warning: assuming signed overflow does not occur when simplifying
co\
nditional to constant [-Wstrict-overflow]
   29 |   sp->types[i] = at <= 0xu;
  |  ~~~^~

1. These diagnostics make no sense, since none of the underlined code involves
any operations that can overflow.

2. I looked through the code carefully, and none of the arithmetic operations
can possibly overflow. In (result << 8), 'result' is at most 2**32 - 1 and this
fits comfortably into 'long'. -1L << 63 is LONG_MIN. "++i" is executed only
when i is less than some other int. timecnt++ and leapcnt++ are executed at
most INT_MAX times. prevcorr + 1 is evaluated only when prevcorr < corr, and
prevcorr - 1 is evaluated only when prevcorr > corr.

3. The diagnostics vanish if you change MYSTERY from 5 to 4 in the first line.
This suggests that GCC is somehow getting confused about whether 'long' has 8
bytes (which it does on this platform) or 4 bytes.

[Bug c++/101911] Type cannot be defined inside of the lambda in unevaluated context

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101911

Andrew Pinski  changed:

   What|Removed |Added

   Last reconfirmed||2021-08-14
 Status|UNCONFIRMED |NEW
 Ever confirmed|0   |1

--- Comment #1 from Andrew Pinski  ---
Confirmed.

[Bug c/22249] GCC does not reject an incompatible type declaration

[muecker at gwdg dot de via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=22249

Martin Uecker  changed:

   What|Removed |Added

 CC||muecker at gwdg dot de

--- Comment #2 from Martin Uecker  ---
This is diagnosed in GCC 11.

https://godbolt.org/z/GcE19jqbK

[Bug c/29970] mixing ({...}) with VLA leads to massive breakage

[muecker at gwdg dot de via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=29970

--- Comment #12 from Martin Uecker  ---
Another version of the patch:

https://gcc.gnu.org/pipermail/gcc-patches/2021-August/577402.html

[Bug middle-end/100406] bogus/missing -Wmismatched-new-delete

[msebor at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100406
Bug 100406 depends on bug 101791, which changed state.

Bug 101791 Summary: missing warning on a mismatch between scalar and array 
forms of new and delete
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101791

   What|Removed |Added

 Status|ASSIGNED|RESOLVED
 Resolution|--- |FIXED

[Bug middle-end/101791] missing warning on a mismatch between scalar and array forms of new and delete

[msebor at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101791

Martin Sebor  changed:

   What|Removed |Added

 Status|ASSIGNED|RESOLVED
 Resolution|--- |FIXED

--- Comment #2 from Martin Sebor  ---
Fixed in r12-2908.

[Bug middle-end/101791] missing warning on a mismatch between scalar and array forms of new and delete

[cvs-commit at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101791

--- Comment #1 from CVS Commits  ---
The master branch has been updated by Martin Sebor :

https://gcc.gnu.org/g:96194a07bdbc57dd9733892a791d87dbe25f0802

commit r12-2908-g96194a07bdbc57dd9733892a791d87dbe25f0802
Author: Martin Sebor 
Date:   Sat Aug 14 13:25:41 2021 -0600

Diagnose mismatches between array and scalar new and delete [PR101791].

Resolves:
PR middle-end/101791 - missing warning on a mismatch between scalar and
array forms of new and delete

gcc/ChangeLog:

PR middle-end/101791
* gimple-ssa-warn-access.cc (new_delete_mismatch_p): Use new
argument
to valid_new_delete_pair_p.
* tree.c (valid_new_delete_pair_p): Add argument.
* tree.h (valid_new_delete_pair_p): Same.

gcc/testsuite/ChangeLog:

PR middle-end/101791
* g++.dg/warn/Wmismatched-new-delete-6.C: New test.
* g++.dg/warn/Wmismatched-new-delete-7.C: New test.

[Bug tree-optimization/101912] New: -Wmaybe-uninitialized false alarm in tzdb localtime.c

[eggert at cs dot ucla.edu via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101912

Bug ID: 101912
   Summary: -Wmaybe-uninitialized false alarm in tzdb localtime.c
   Product: gcc
   Version: 11.2.1
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: tree-optimization
  Assignee: unassigned at gcc dot gnu.org
  Reporter: eggert at cs dot ucla.edu
  Target Milestone: ---

Created attachment 51303
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=51303&action=edit
-Wuninitialized false alarm

This is a simplified version of a problem I found while compiling tzdb's
localtime.c with gcc (GCC) 11.2.1 20210728 (Red Hat 11.2.1-1) on x86-64.
Compile the attached program w.i with:

gcc -Wmaybe-uninitialized -O2 -S w.i

The output is:

w.i: In function ‘tzloadbody’:
w.i:17:27: warning: ‘prevcorr’ may be used uninitialized in this function
[-Wmaybe-uninitialized]
   16 |: (corr == prevcorr
  |  ~
   17 |   || corr == prevcorr - 1)
  |   ^~~~


This is a false alarm, since the underlined code is executed only when leapcnt
is nonzero, and leapcnt is set to a nonzero value only after prevcorr is
initialized.

[Bug c/101838] sizeof of struct with array of unspecified bound folded to constant

[muecker at gwdg dot de via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101838

Martin Uecker  changed:

   What|Removed |Added

 Status|UNCONFIRMED |RESOLVED
 Resolution|--- |FIXED

--- Comment #3 from Martin Uecker  ---
Fixed on trunk.

[Bug c++/101911] New: Type cannot be defined inside of the lambda in unevaluated context

[eligorkadaf at gmail dot com via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101911

Bug ID: 101911
   Summary: Type cannot be defined inside of the lambda in
unevaluated context
   Product: gcc
   Version: 11.2.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: c++
  Assignee: unassigned at gcc dot gnu.org
  Reporter: eligorkadaf at gmail dot com
  Target Milestone: ---

The following code is rejected by gcc:

#include 

int main() {
   std::vector vec;
}
gcc claims that types cannot be defined inside of decltype. Hovewer, I did not
find anything in standard to confirm such a message in this context. So if you
need any proof, proof me please that there is a reason for gcc to claim that
first.

https://godbolt.org/z/8n6djencY

[Bug go/95389] Kubernetes build fails because of mangled PkgPath

[ulrich.teichert at gmx dot de via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95389

--- Comment #5 from Ulrich Teichert  ---
Thanks for the head-up, I'll give it a spin and report back. But building gcc
on my box takes a while, so please don't hold your breath ;-)

[Bug middle-end/101910] New: tsvc regressions for -O2 -ftree-loop-vectorize at zen hardware

[hubicka at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101910

Bug ID: 101910
   Summary: tsvc regressions for -O2 -ftree-loop-vectorize at zen
hardware
   Product: gcc
   Version: 12.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: middle-end
  Assignee: unassigned at gcc dot gnu.org
  Reporter: hubicka at gcc dot gnu.org
  Target Milestone: ---

https://lnt.opensuse.org/db_default/v4/CPP/latest_runs_report?younger_in_days=14&older_in_days=0&all_changes=on&min_percentage_change=0.001&revisions=e54acea9e5a821448af97c66e94a1e4c4f977d5d%2Ce87209a1269622017bf3d98bf71502dcb0f893aa%2C73474527aaa24d9236aca074c5494a07f40ce058&include_user_branches=on

shows large regressions on tsvc for benchmarks s253, s271, s318, s3110, s281,
s292, s443

Maybe generic model needs retuning here.

[Bug middle-end/101909] New: 73% regression on tfft benchmark for -O2 -ftree-loop-vectorize compared to -O2 on zen hardware

[hubicka at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101909

Bug ID: 101909
   Summary: 73% regression on tfft benchmark for -O2
-ftree-loop-vectorize compared to -O2 on zen hardware
   Product: gcc
   Version: 12.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: middle-end
  Assignee: unassigned at gcc dot gnu.org
  Reporter: hubicka at gcc dot gnu.org
  Target Milestone: ---

https://lnt.opensuse.org/db_default/v4/CPP/latest_runs_report?younger_in_days=14&older_in_days=0&all_changes=on&min_percentage_change=0.001&revisions=e54acea9e5a821448af97c66e94a1e4c4f977d5d%2Ce87209a1269622017bf3d98bf71502dcb0f893aa%2C73474527aaa24d9236aca074c5494a07f40ce058&include_user_branches=on

This regression seems to be limited to zenith tester and does not reproduce on
benzen

[Bug middle-end/101908] New: cray regression with -O2 -ftree-slp-vectorize compared to -O2

[hubicka at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101908

Bug ID: 101908
   Summary: cray regression with -O2 -ftree-slp-vectorize compared
to -O2
   Product: gcc
   Version: 12.0
Status: UNCONFIRMED
  Severity: normal
  Priority: P3
 Component: middle-end
  Assignee: unassigned at gcc dot gnu.org
  Reporter: hubicka at gcc dot gnu.org
  Target Milestone: ---

This should be easy to track since cray is stupid benchmark.

According to run
https://lnt.opensuse.org/db_default/v4/CPP/latest_runs_report?younger_in_days=14&older_in_days=0&all_changes=on&min_percentage_change=0.001&revisions=e54acea9e5a821448af97c66e94a1e4c4f977d5d%2Ce87209a1269622017bf3d98bf71502dcb0f893aa%2C73474527aaa24d9236aca074c5494a07f40ce058&include_user_branches=on

slp-vectorize seems to cause 101%-124% regression on zen and 96% on kabylake.ve

[Bug tree-optimization/92740] induct2 (from polyhedron) regresses 267% with -O2 -ftree-vectorize -ftree-slp-vectorize -fvect-cost-modes=dynamic or cheap compared to -O2

[hubicka at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92740

Jan Hubicka  changed:

   What|Removed |Added

Summary|induct2 (from polyhedron)   |induct2 (from polyhedron)
   |regresses 267% with -O2 |regresses 267% with -O2
   |-ftree-vectorize|-ftree-vectorize
   |-ftree-slp-vectorize|-ftree-slp-vectorize
   |-fvect-cost-modes=dynamic   |-fvect-cost-modes=dynamic
   |compared to -O2 |or cheap compared to -O2
 Status|WAITING |NEW

--- Comment #3 from Jan Hubicka  ---
I have run again compare with -O2 and -O2 -ftree-loop-vectorize
https://lnt.opensuse.org/db_default/v4/CPP/latest_runs_report?younger_in_days=14&older_in_days=0&all_changes=on&min_percentage_change=0.001&revisions=e54acea9e5a821448af97c66e94a1e4c4f977d5d%2Ce87209a1269622017bf3d98bf71502dcb0f893aa%2C73474527aaa24d9236aca074c5494a07f40ce058&include_user_branches=on

regression is 269% on zen 236 on kaby.