[Bug middle-end/121427] For some structs, GCC ignores the volatile qualifier

2025-08-06 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121427 Andrew Pinski changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution|---

[Bug middle-end/99258] volatile struct access optimized away

2025-08-06 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99258 Andrew Pinski changed: What|Removed |Added CC||pascal_cuoq at hotmail dot com --- Comme

[Bug tree-optimization/121422] [16 Regression] wrong code for proping zero incorrectly

2025-08-05 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121422 Andrew Pinski changed: What|Removed |Added Last reconfirmed||2025-08-06 Version|14.2.0

[Bug tree-optimization/121422] New: [16 Regression] wrong code for proping zero incorrectly

2025-08-05 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121422 Bug ID: 121422 Summary: [16 Regression] wrong code for proping zero incorrectly Product: gcc Version: 14.2.0 Status: UNCONFIRMED Keywords: wrong-code

[Bug c/121421] [ICE] internal compiler error: Segmentation fault

2025-08-05 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121421 Andrew Pinski changed: What|Removed |Added Status|UNCONFIRMED |ASSIGNED Assignee|unassigned

[Bug middle-end/116375] redundant copy not eliminated with CONSTRUCTOR

2025-08-05 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116375 --- Comment #6 from Andrew Pinski --- Created attachment 62064 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=62064&action=edit Current patch which does not work I have some new ideas on how to fix this. more related to https://gcc.gnu.or

[Bug middle-end/116375] redundant copy not eliminated with CONSTRUCTOR

2025-08-05 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116375 --- Comment #5 from Andrew Pinski --- So even with hacking the verifiers (for now) and getting this in the .optimized: ``` void g () { [local count: 1073741824]: f ({}); [tail call] return; } ``` We still get a memset and a memcpy. So t

[Bug target/121420] Failed to build BPF program accepted by Clang (error: cannot take address of bit-field 'mem_hops')

2025-08-05 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121420 --- Comment #8 from Andrew Pinski --- (In reply to Sam James from comment #7) > -> > https://lore.kernel.org/linux-perf-users/ > fea380fb0934d039d19821bba88130e632bbfe8d.1754438581.git@gentoo.org/T/#u > > MOVED then? Let's wait for the rep

[Bug target/121420] Failed to build BPF program accepted by Clang (error: cannot take address of bit-field 'mem_hops')

2025-08-05 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121420 --- Comment #6 from Andrew Pinski --- s/FIELD_EXISTENCE/BPF_FIELD_EXISTS/ But you get the idea.

[Bug target/121420] Failed to build BPF program accepted by Clang (error: cannot take address of bit-field 'mem_hops')

2025-08-05 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121420 --- Comment #5 from Andrew Pinski --- (In reply to Andrew Pinski from comment #4) > I think this patch fixes the issue: diff --git a/tools/perf/util/bpf_skel/sample_filter.bpf.c b/tools/perf/util/bpf_skel/sample_filter.bpf.c index b195e6efeb8b.

[Bug target/121420] Failed to build BPF program accepted by Clang (error: cannot take address of bit-field 'mem_hops')

2025-08-05 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121420 --- Comment #4 from Andrew Pinski --- I think this patch fixes the issue: ``` diff --git a/tools/perf/util/bpf_skel/sample_filter.bpf.c b/tools/perf/util/bpf_skel/sample_filter.bpf.c index b195e6efeb8b..f1ffef44f603 100644 --- a/tools/perf/util/

[Bug target/121420] Failed to build BPF program accepted by Clang (error: cannot take address of bit-field 'mem_hops')

2025-08-05 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121420 --- Comment #3 from Andrew Pinski --- https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=3a8b8fc3174891c4c12f5766d82184a82d4b2e3e introduced the brokenness after the Cupertino Miranda's patch. commit 3a8b8fc3174891c4c

[Bug target/121420] Failed to build BPF program accepted by Clang (error: cannot take address of bit-field 'mem_hops')

2025-08-05 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121420 --- Comment #2 from Andrew Pinski --- This looks like it is doing the old way. because GCC documentation says it should be just done as: __builtin_preserve_field_info (arg->y, FIELD_BYTE_OFFSET); https://gcc.gnu.org/onlinedocs/gcc/BPF-Built-i

[Bug tree-optimization/93507] Missed abstraction removal in VN

2025-08-05 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93507 Andrew Pinski changed: What|Removed |Added See Also||https://gcc.gnu.org/bugzill

[Bug target/121419] __clzdi2 emitted when building udev-hid-bpf

2025-08-05 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121419 --- Comment #5 from Andrew Pinski --- LLVM's compiler-rt even depends on it always being expanded; https://github.com/llvm/llvm-project/issues/12035 . :)

[Bug target/121419] __clzdi2 emitted when building udev-hid-bpf

2025-08-05 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121419 --- Comment #4 from Andrew Pinski --- https://gitlab.freedesktop.org/libevdev/udev-hid-bpf/-/blob/main/src/bpf/testing/0010-WALTOP__Batteryless-Tablet.bpf.c?ref_type=heads

[Bug target/121419] __clzdi2 emitted when building udev-hid-bpf

2025-08-05 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121419 --- Comment #3 from Andrew Pinski --- There might be a better way of implementing scaled_log2 rather than using __builtin_clzg here.

[Bug target/121419] __clzdi2 emitted when building udev-hid-bpf

2025-08-05 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121419 Andrew Pinski changed: What|Removed |Added Target||bpf --- Comment #2 from Andrew Pinski

[Bug c++/121408] ICE when using dynamic_cast inside a lambda in a template parameter with C++20/C++23

2025-08-05 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121408 Andrew Pinski changed: What|Removed |Added Keywords||error-recovery, |

[Bug tree-optimization/121418] New: Missed copy prop for aggregate after a memcpy

2025-08-05 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121418 Bug ID: 121418 Summary: Missed copy prop for aggregate after a memcpy Product: gcc Version: 14.2.0 Status: UNCONFIRMED Keywords: missed-optimization Severity: enhancemen

[Bug target/121415] aarch64: Failure to handle PSTATE.SM & ZA for tlsdesc calls

2025-08-05 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121415 Andrew Pinski changed: What|Removed |Added See Also||https://github.com/llvm/llv

[Bug target/121415] aarch64: Failure to handle PSTATE.SM & ZA for tlsdesc calls

2025-08-05 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121415 Andrew Pinski changed: What|Removed |Added Ever confirmed|0 |1 Last reconfirmed|

[Bug c++/121408] ICE when using dynamic_cast inside a lambda in a template parameter with C++20/C++23

2025-08-05 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121408 Andrew Pinski changed: What|Removed |Added Last reconfirmed||2025-08-05 Status|UNCONFIRM

[Bug target/111231] [13/14/15/16 regression] armhf: Miscompilation with -O2/-fno-exceptions level (-fno-tree-vectorize is working)

2025-08-05 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111231 --- Comment #40 from Andrew Pinski --- (In reply to Andrew Pinski from comment #39) > (In reply to Richard Earnshaw from comment #34) > > To be honest, I'm more concerned that we aren't eliminating a lot of these > > copies during the gimple opt

[Bug tree-optimization/121418] Missed copy prop for aggregate after a memcpy

2025-08-05 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121418 Andrew Pinski changed: What|Removed |Added Blocks||121364 Last reconfirmed|

[Bug tree-optimization/121417] New: Missed copy prop for aggregate with memcpy

2025-08-05 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121417 Bug ID: 121417 Summary: Missed copy prop for aggregate with memcpy Product: gcc Version: 14.2.0 Status: UNCONFIRMED Keywords: missed-optimization Severity: enhancement

[Bug tree-optimization/121417] Missed copy prop for aggregate with memcpy

2025-08-05 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121417 Andrew Pinski changed: What|Removed |Added Last reconfirmed||2025-08-05 Ever confirmed|0

[Bug target/111231] [13/14/15/16 regression] armhf: Miscompilation with -O2/-fno-exceptions level (-fno-tree-vectorize is working)

2025-08-05 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111231 --- Comment #39 from Andrew Pinski --- (In reply to Richard Earnshaw from comment #34) > To be honest, I'm more concerned that we aren't eliminating a lot of these > copies during the gimple optimization phase. The memcpy is really a type > pun

[Bug target/121343] [avr] Use hard-register constraints instead of explicit hard-regs in avr.md

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121343 Andrew Pinski changed: What|Removed |Added Severity|normal |enhancement

[Bug bootstrap/121407] Patches to fix building docs under makeinfo 4.8

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121407 --- Comment #3 from Andrew Pinski --- (In reply to Kirill A. Korinsky from comment #2) > Andrew, not sure. > > A hunk for intrinsic.texi and gfortran.texi probably new. Ok, looks like only part is fixed. That bug is still open it seems. So let

[Bug middle-end/121389] [15/16 regression] Failed musttail with -fsanitize=address -O0 (error: cannot tail-call: return value changed after call)

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121389 Andrew Pinski changed: What|Removed |Added Ever confirmed|0 |1 Status|UNCONFIRMED

[Bug middle-end/121389] [15/16 regression] Failed musttail with -fsanitize=address -O0 (error: cannot tail-call: return value changed after call)

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121389 --- Comment #3 from Andrew Pinski --- Created attachment 62051 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=62051&action=edit Reduced testcase I added back some code because the loop became not happening.

[Bug middle-end/121389] [15/16 regression] Failed musttail with -fsanitize=address -O0 (error: cannot tail-call: return value changed after call)

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121389 --- Comment #2 from Andrew Pinski --- Reducing ...

[Bug middle-end/121206] [16 regression] Hang in ranger when building gromacs-2025.2 since r16-1645-g309dbcea2cabb3

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121206 --- Comment #9 from Andrew Pinski --- (In reply to Sam James from comment #8) > r16-1645-g309dbcea2cabb3 That might have just exposed it.

[Bug tree-optimization/121405] [13/14/15/16 Regression] Another missed VN via a copy (but via an int copy)

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121405 Andrew Pinski changed: What|Removed |Added Attachment #62046|0 |1 is obsolete|

[Bug middle-end/120544] [15/16 Regression] extend.texi causes a build failure for makeinfo 4.13 (and before)

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=120544 Andrew Pinski changed: What|Removed |Added CC||kirill at korins dot ky --- Comment #12

[Bug bootstrap/121407] Patches to fix building docs under makeinfo 4.8

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121407 Andrew Pinski changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution|---

[Bug c/121406] Type compatibility checks between old-style function definition and prototyped function type ignore parameters

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121406 Andrew Pinski changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution|---

[Bug c/121406] Type compatibility checks between old-style function definition and prototyped function type ignore parameters

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121406 Andrew Pinski changed: What|Removed |Added Keywords||diagnostic --- Comment #1 from Andrew P

[Bug tree-optimization/121405] [13/14/15/16 Regression] Another missed VN via a copy (but via an int copy)

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121405 --- Comment #3 from Andrew Pinski --- Created attachment 62047 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=62047&action=edit Reduced using int rather than char Combine in GCC 15+ is not able to do the combine in this case. Instead we g

[Bug tree-optimization/121362] SCCVN (FRE) sometimes does not read through aggregate copies

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121362 Andrew Pinski changed: What|Removed |Added See Also||https://gcc.gnu.org/bugzill

[Bug tree-optimization/121405] [13/14/15/16 Regression] Another missed VN via a copy (but via an int copy)

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121405 --- Comment #2 from Andrew Pinski --- This is the .optimized from highway where I saw this: ``` MEM [(struct Vec128 *)&D.741237] = _4040; MEM [(struct Vec128 *)&D.741237 + 1B] = _4047; MEM [(struct Vec128 *)&D.741237 + 2B] = _4054; ME

[Bug tree-optimization/121405] New: Another missed VN via a copy (but via an int copy)

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121405 Bug ID: 121405 Summary: Another missed VN via a copy (but via an int copy) Product: gcc Version: 14.2.0 Status: UNCONFIRMED Keywords: missed-optimization Severity: norma

[Bug tree-optimization/121405] [13/14/15/16 Regression] Another missed VN via a copy (but via an int copy)

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121405 Andrew Pinski changed: What|Removed |Added Target Milestone|--- |13.5 Known to fail|

[Bug tree-optimization/121405] Another missed VN via a copy (but via an int copy)

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121405 Andrew Pinski changed: What|Removed |Added Severity|normal |enhancement Version|14.2.0

[Bug tree-optimization/121362] SCCVN (FRE) sometimes does not read through aggregate copies

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121362 --- Comment #7 from Andrew Pinski --- (In reply to Richard Biener from comment #6) > Fixed. I'd be interested in cases we're still missing. This fixed the cases that I saw inside highway which was a store of 2 null pointers to a struct and the

[Bug bootstrap/121400] Gcc fails to bootstrap on OpenBSD/aarch64 due to missed #ifdef MD_ARCH_FRAME_STATE_T

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121400 --- Comment #2 from Andrew Pinski --- (In reply to Kirill A. Korinsky from comment #1) > /build/pobj/gcc-15.1.0/gcc-15.1.0/libgcc/config/aarch64/__arm_tpidr2_restore. > S: Assembler messages: > /build/pobj/gcc-15.1.0/gcc-15.1.0/libgcc/config/aar

[Bug target/121388] Invalid cble produced with -march=armv9-a+cmpbr while building bid_binarydecimal

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121388 --- Comment #1 from Andrew Pinski --- Created attachment 62044 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=62044&action=edit Semi reduced This as reduced as I can get it. The problem looks related to how long branches work. So we get

[Bug target/121394] [16 Regression][gcn] Since r16-2595-gf1c80147641783: link-time error: libm_a-e_atan2.o):(.rodata.cst32): SHF_MERGE section size (56) must be a multiple of sh_entsize (32)

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121394 --- Comment #9 from Andrew Pinski --- (In reply to Andrew Pinski from comment #8) > So now where does `.align` get emitted. This is due to r0-46282-gf41115930523b3; the current code does: ``` /* Make sure all constants in SECTION_MERGE and no

[Bug target/121394] [16 Regression][gcn] Since r16-2595-gf1c80147641783: link-time error: libm_a-e_atan2.o):(.rodata.cst32): SHF_MERGE section size (56) must be a multiple of sh_entsize (32)

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121394 Andrew Pinski changed: What|Removed |Added Assignee|unassigned at gcc dot gnu.org |pinskia at gcc dot gnu.org

[Bug target/121394] [16 Regression][gcn] Since r16-2595-gf1c80147641783: link-time error: libm_a-e_atan2.o):(.rodata.cst32): SHF_MERGE section size (56) must be a multiple of sh_entsize (32)

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121394 Andrew Pinski changed: What|Removed |Added Last reconfirmed||2025-08-04 Ever confirmed|0

[Bug target/121394] [16 Regression][gcn] Since r16-2595-gf1c80147641783: link-time error: libm_a-e_atan2.o):(.rodata.cst32): SHF_MERGE section size (56) must be a multiple of sh_entsize (32)

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121394 --- Comment #8 from Andrew Pinski --- (In reply to Andrew Pinski from comment #6) > (In reply to Andrew Pinski from comment #5) > > Oh yes: > > https://www.sourceware.org/ml/binutils/2002-11/msg00604.html > > > > Now I need to find out the conc

[Bug target/121394] [16 Regression][gcn] Since r16-2595-gf1c80147641783: link-time error: libm_a-e_atan2.o):(.rodata.cst32): SHF_MERGE section size (56) must be a multiple of sh_entsize (32)

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121394 Andrew Pinski changed: What|Removed |Added Keywords||link-failure --- Comment #6 from Andrew

[Bug target/121394] [16 Regression][gcn] Since r16-2595-gf1c80147641783: link-time error: libm_a-e_atan2.o):(.rodata.cst32): SHF_MERGE section size (56) must be a multiple of sh_entsize (32)

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121394 --- Comment #5 from Andrew Pinski --- Oh yes: https://www.sourceware.org/ml/binutils/2002-11/msg00604.html Now I need to find out the conclusion of that

[Bug libgcc/121397] [15/16 Regression] build fails in libgcc for aarch64-openbsd due to unused static function in enable-execute-stack-mprotect.c

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121397 Andrew Pinski changed: What|Removed |Added Target Milestone|--- |15.2 Summary|build fails in

[Bug libgcc/121397] build fails in libgcc for aarch64-openbsd due to unused static function in enable-execute-stack-mprotect.c

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121397 Andrew Pinski changed: What|Removed |Added Target||aarch64*-*-openbsd Summary|B

[Bug bootstrap/121397] Bootstrap by clang 19.1.7 fails due to unused static function

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121397 Andrew Pinski changed: What|Removed |Added Last reconfirmed||2025-08-04 Ever confirmed|0

[Bug target/121394] [16 Regression][gcn] Since r16-2595-gf1c80147641783: link-time error: libm_a-e_atan2.o):(.rodata.cst32): SHF_MERGE section size (56) must be a multiple of sh_entsize (32)

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121394 --- Comment #2 from Andrew Pinski --- This is using clang as the assembler? I am suspecting it is a bug there. The section is not being padding to the alignment of the section.

[Bug target/120718] ICE (unrecognizable insn) with const_poly_int in v2si vector

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=120718 Andrew Pinski changed: What|Removed |Added CC||jakub at gcc dot gnu.org --- Comment #4

[Bug target/121358] [15/16 Regression] SVE ICE compiling highway since r15-4235

2025-08-04 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121358 Andrew Pinski changed: What|Removed |Added Status|NEW |RESOLVED Resolution|---

[Bug target/121388] New: Invalid cble produced with -march=armv9-a+cmpbr while building bid_binarydecimal

2025-08-03 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121388 Bug ID: 121388 Summary: Invalid cble produced with -march=armv9-a+cmpbr while building bid_binarydecimal Product: gcc Version: 14.2.0 Status: UNCONFIRMED Keywo

[Bug target/121385] libgcc build failure with -march=armv9-a+cmpbr

2025-08-03 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121385 --- Comment #6 from Andrew Pinski --- (In reply to Andrew Pinski from comment #5) > Or easier disable aarch64_tbzltdi1 if TARGET_CMPBR like: > diff --git a/gcc/config/aarch64/aarch64.md b/gcc/config/aarch64/aarch64.md > index a4ae6859da0..7c87b0

[Bug target/121385] libgcc build failure with -march=armv9-a+cmpbr

2025-08-03 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121385 --- Comment #5 from Andrew Pinski --- Both can match: (jump_insn 297 296 298 35 (set (pc) (if_then_else (lt (reg:DI 170 [ _94 ]) (const_int 0 [0])) (label_ref 303) (pc))) "../../../comb/libgcc/libg

[Bug target/121385] libgcc build failure with -march=armv9-a+cmpbr

2025-08-03 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121385 --- Comment #4 from Andrew Pinski --- So there is a conflict between aarch64_tbzltdi1 and aarch64_cbltdi

[Bug bootstrap/121386] [14/15 Regression] Can't build on OpenBSD -current/amd64 with the prebuilt flex sources

2025-08-03 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121386 --- Comment #6 from Andrew Pinski --- There has definitely improvements to flex in this area too: e.g. https://github.com/westes/flex/commit/5574881ff4c9b08c73b6acc0087d6046c2b48de5 and https://github.com/westes/flex/commit/8ab4ea7364641bf498

[Bug bootstrap/121386] [14/15 Regression] Can't build on OpenBSD -current/amd64 with the prebuilt flex sources

2025-08-03 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121386 --- Comment #3 from Andrew Pinski --- Maybe there is min version of flex that is needed to build the sources from that makes it usable on openbsd.

[Bug bootstrap/121386] [14/15 Regression] Can't build on OpenBSD -current/amd64 with the prebuilt flex sources

2025-08-03 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121386 Andrew Pinski changed: What|Removed |Added Summary|Can't build on OpenBSD |[14/15 Regression] Can't

[Bug bootstrap/121386] Can't build on OpenBSD -current/amd64

2025-08-03 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121386 Andrew Pinski changed: What|Removed |Added Status|UNCONFIRMED |NEW Last reconfirmed|

[Bug target/121385] libgcc build failure with -march=armv9-a+cmpbr

2025-08-03 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121385 Andrew Pinski changed: What|Removed |Added Status|UNCONFIRMED |NEW Last reconfirmed|

[Bug target/121385] libgcc build failure with -march=armv9-a+cmpbr

2025-08-03 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121385 --- Comment #3 from Andrew Pinski --- Created attachment 62039 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=62039&action=edit Reduced testcase

[Bug target/121385] libgcc build failure with -march=armv9-a+cmpbr

2025-08-03 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121385 --- Comment #2 from Andrew Pinski --- ``` ../../gcc/cc1 _mulvdi3.i -O2 -march=armv9-a+cmpbr vprintf getchar fgetc_unlocked getc_unlocked getchar_unlocked putchar fputc_unlocked putc_unlocked putchar_unlocked getline feof_unlocked ferror_unlock

[Bug target/121385] libgcc build failure with -march=armv9-a+cmpbr

2025-08-03 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121385 --- Comment #1 from Andrew Pinski --- Created attachment 62038 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=62038&action=edit preprocessed source

[Bug middle-end/23782] SRA pessimizes passing structures by value at -Os (+22% code size)

2025-08-03 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=23782 --- Comment #11 from Andrew Pinski --- After my patch, we get: ``` void drawbutton (struct coord upperleft, struct coord lowerright, unsigned int upperleftcolor, unsigned int lowerrightrcolor, unsigned int fillcolor, unsigned int drawbackground)

[Bug tree-optimization/121382] [15/16 Regression] wrong code at -O2 on x86_64-linux-gnu

2025-08-03 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121382 Andrew Pinski changed: What|Removed |Added Last reconfirmed||2025-08-03 Ever confirmed|0

[Bug tree-optimization/121382] [15/16 Regression] wrong code at -O2 on x86_64-linux-gnu

2025-08-03 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121382 --- Comment #2 from Andrew Pinski --- Created attachment 62037 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=62037&action=edit Slighty better testcase This testcase does NOT depend on IPA VRP to produce the needed range to cause the issu

[Bug tree-optimization/121382] [15/16 Regression] wrong code at -O2 on x86_64-linux-gnu

2025-08-03 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121382 Andrew Pinski changed: What|Removed |Added Target Milestone|--- |15.2

[Bug tree-optimization/121383] if-convert lacks support for && conditions

2025-08-03 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121383 Andrew Pinski changed: What|Removed |Added Status|NEW |ASSIGNED

[Bug tree-optimization/121383] if-convert lacks support for && conditions

2025-08-03 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121383 Andrew Pinski changed: What|Removed |Added Ever confirmed|0 |1 Status|UNCONFIRMED

[Bug libstdc++/121348] `vector::resize(n, value)` incorrectly requires the element type to be assignable

2025-08-02 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121348 --- Comment #5 from Andrew Pinski --- (In reply to Andrew Pinski from comment #4) > I suspect the issue is we use std::ill rather than std::uninitialized_fill . s/std::ill/std::fill/. Sorry snout the typo

[Bug libstdc++/121348] `vector::resize(n, value)` incorrectly requires the element type to be assignable

2025-08-02 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121348 --- Comment #4 from Andrew Pinski --- I suspect the issue is we use std::ill rather than std::uninitialized_fill .

[Bug c/121376] Objects with temporary lifetime do not work correctly in c11+

2025-08-02 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121376 --- Comment #1 from Andrew Pinski --- https://www.open-std.org/jtc1/sc22/wg14/www/docs/n1285.htm

[Bug tree-optimization/121280] [15/16 Regression] False positive array-bounds warning with O3 and std::vector.back() of a local vector copy

2025-08-02 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121280 --- Comment #1 from Andrew Pinski --- Note this code is undefined if vect.empty() is true due to front being undefined in that case. Adding that check makes the warning go away too; either for `vect_copy.empty()`or `vect.empty()`. So this is de

[Bug target/121336] cris-elf fails to build with binutils-2.45 due to ambiguous --em option

2025-08-02 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121336 Andrew Pinski changed: What|Removed |Added Last reconfirmed||2025-08-02 Status|UNCONFIRM

[Bug libstdc++/121348] `vector::resize(n, value)` incorrectly requires the element type to be assignable

2025-08-02 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121348 --- Comment #2 from Andrew Pinski --- https://cplusplus.github.io/LWG/lwg-defects.html#2323 Change it to just: equires: T shall be CopyInsertable into *this. So this looks like it was changed for C++20.

[Bug libstdc++/121348] `vector::resize(n, value)` incorrectly requires the element type to be assignable

2025-08-02 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121348 --- Comment #1 from Andrew Pinski --- https://eel.is/c++draft/vector.capacity Preconditions: T is Cpp17CopyInsertable into vector. https://eel.is/c++draft/container.alloc.reqmts#def:Cpp17CopyInsertable_into_X T is Cpp17CopyInsertable into X m

[Bug target/121358] [15/16 Regression] SVE ICE compiling highway since r15-4235

2025-08-02 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121358 --- Comment #5 from Andrew Pinski --- This patch fixes this issue: ``` diff --git a/gcc/simplify-rtx.cc b/gcc/simplify-rtx.cc index c723a07f06b..c5f49ff2645 100644 --- a/gcc/simplify-rtx.cc +++ b/gcc/simplify-rtx.cc @@ -8460,7 +8460,8 @@ simplif

[Bug target/121358] [15/16 Regression] SVE ICE compiling highway since r15-4235

2025-08-02 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121358 --- Comment #4 from Andrew Pinski --- So I think the problem is force_subreg expected simplify_gen_subreg would simplify the subreg of the const_vector into a const_int but in this case we don't as we don't have a way to simplify const_vectors w

[Bug libstdc++/121374] wrong value std::numeric_limits<__float128>::max_digits10

2025-08-02 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121374 Andrew Pinski changed: What|Removed |Added Status|UNCONFIRMED |NEW Ever confirmed|0

[Bug libstdc++/121373] std::byteswap is not available when using import std

2025-08-02 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121373 --- Comment #3 from Andrew Pinski --- (In reply to printfne from comment #2) > (In reply to Andrew Pinski from comment #1) > > > For some reasons, I couldn't find this patch in the short term. > > > > Found it: > > > > https://inbox.sourceware

[Bug tree-optimization/121372] [15/16 regression] miscompilation in OpenBLAS since r15-571-g1e0ae1f52741f7

2025-08-02 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121372 --- Comment #9 from Andrew Pinski --- r8-3818-g806aa9b2f24aa2 was when GCC documentation added that inline-asm example. https://inbox.sourceware.org/gcc-patches/20170821012323.gc3...@bubble.grove.modra.org/

[Bug tree-optimization/121372] [15/16 regression] miscompilation in OpenBLAS since r15-571-g1e0ae1f52741f7

2025-08-02 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121372 Andrew Pinski changed: What|Removed |Added Resolution|--- |MOVED Status|UNCONFIRMED

[Bug tree-optimization/121370] [13/14/15/16 regression] wrong code at -O{s,2,3} on x86_64-linux-gnu

2025-08-02 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121370 Andrew Pinski changed: What|Removed |Added Ever confirmed|0 |1 Status|UNCONFIRMED

[Bug libstdc++/121373] std::byte_swap is not available when using import std

2025-08-02 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121373 Andrew Pinski changed: What|Removed |Added Ever confirmed|0 |1 Keywords|

[Bug middle-end/121371] Misoptimize when signed integer is used as index and increase repeatedly

2025-08-02 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121371 --- Comment #2 from Andrew Pinski --- If you used size_t, then it would just work.

[Bug c++/121368] [[nodiscard]] warning not triggered when discarding return value from function pointer call

2025-08-02 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121368 Andrew Pinski changed: What|Removed |Added See Also||https://github.com/llvm/llv

[Bug c++/121368] [[nodiscard]] warning not triggered when discarding return value from function pointer call

2025-08-02 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121368 Andrew Pinski changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution|---

[Bug target/121358] [15/16 Regression] SVE ICE compiling highway since r15-4235

2025-08-02 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121358 --- Comment #3 from Andrew Pinski --- Note normally the backend supports this const_vector correctly but the subreg around looks like it is causing issues. I might take a look this afternoon.

[Bug c/121369] Error instead of warning for -Wincompatible-pointer-types?

2025-08-02 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=121369 Andrew Pinski changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution|---

[Bug c++/84476] [[nodiscard]] ignored on virtual functions accessed through pointer

2025-08-02 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84476 Andrew Pinski changed: What|Removed |Added CC||vladimir.krivopalov at gmail dot c

[Bug c++/95773] [[nodiscard]] attribute is ignored for calls to overridden functions

2025-08-02 Thread pinskia at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95773 Andrew Pinski changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution|---

  1   2   3   4   5   6   7   8   9   10   >