Control: tags -1 wontfix
GCC-13 works as expected. Turns out to be a UB case in highway source code.
Closing
Total Test time (real) = 26.65 sec
The following tests FAILED:
446 - HwyWidenMulTestGroup/HwyWidenMulTest.TestAllSatWidenMulPairwiseAdd/EMU128
# GetParam() = 2305843009213693952 (Subprocess aborted)
452 - HwyWidenMulTestGroup/HwyWidenMulTest.TestAllSumOfMulQuadAccumulate/EMU128
# GetParam() = 2
Package: gcc-13
Version: 13.2.0-3
highway does not seem to work on ia64 with LTO (see #1051769).
On yttrium with gcc-13:
% /usr/bin/g++-13 -g -O2 -ffile-prefix-map=/home/malat/highway-1.0.7=.
-flto=auto -ffat-lto-objects -specs=/usr/share/dpkg/pie-compile.specs
-Wformat -Werror=format-security -
Package: gcc-12
Version: 12.2.0-12
highway does not seems to work on ia64 with LTO:
https://buildd.debian.org/status/fetch.php?pkg=highway&arch=ia64&ver=1.0.7-3&stamp=1694507301&raw=0
The fun part is that even gdb crash on the generated exe:
% gdb tests/copy_test
GNU gdb (Debian 10.1-2) 10.1.9
Package: gcc-13
Version: 13.2.0-3
Forwarded: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110812
Affects: src:highway
src:highway fails to compile on riscv64 with LTO. Confirmed upstream.
On Sun, Sep 3, 2023 at 9:23 AM Mathieu Malaterre wrote:
>
> On Sat, Sep 2, 2023 at 1:31 PM Matthias Klose wrote:
> > upstream asks for a self-contained test case. Not sure if that's something
> > you
> > tried in https://bugs.debian.org/1050415
>
> Current
Version: 12.2.0-18
malat@barriere ~ % apt-cache policy gcc-12
gcc-12:
Installed: 12.2.0-18
Candidate: 12.2.0-18
Version table:
*** 12.2.0-18 100
1 https://deb.debian.org/debian experimental/main i386 Packages
100 /var/lib/dpkg/status
12.2.0-14 500
500 https://
On Sat, Sep 2, 2023 at 1:31 PM Matthias Klose wrote:
> upstream asks for a self-contained test case. Not sure if that's something you
> tried in https://bugs.debian.org/1050415
Currently working on PR/111231. cresult is difficult to work with as
it default to aggressive renaming. I've switch to c
Package: g++-13
Version: 13.2.0-2
Forwarded: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=111231
Affects: src:highway
I am getting some odd behavior for unit test of highway. I believe
there is some wrong-code generation using g++ + -O2 on armhf. I also
believe this is different from Debian bug #1
Package: g++-13
Version: 13.2.0-2
Forwarded: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110643
Affects: src:highway
I am getting some odd behavior for unit test of highway. I believe
there is some wrong-code generation using g++ + -O1.
Source: creduce
Version: 2.11.0~20230819-1
Could someone please document where creduce homepage is located nowadays.
http://embed.cs.utah.edu/creduce/ seems to be gone.
I am not clear what to do with reports such as:
===< pass_clang_binsrch :: replace-function-def-with-decl >===
Segmentation fa
On Thu, Aug 24, 2023 at 7:03 PM Matthias Klose wrote:
>
> Control: tags -1 - moreinfo
>
> On 24.08.23 15:15, Mathieu Malaterre wrote:
> > On Thu, Aug 24, 2023 at 2:21 PM Matthias Klose wrote:
> >>
> >> Control: tags -1 + moreinfo
> >>
&g
On Thu, Aug 24, 2023 at 2:21 PM Matthias Klose wrote:
>
> Control: tags -1 + moreinfo
>
> On 24.08.23 11:54, Mathieu Malaterre wrote:
> > Package: cvise
> > Version: 2.8.0-1
> >
> > I cannot run cvise in Debian/sid:
>
> [...]
>
> > with:
> &g
For reference, creduce seems to be happy with the exact same settings:
% creduce check.sh testcase.i
===< 150190 >===
running 3 interestingness tests in parallel
===< pass_unifdef :: 0 >===
===< pass_comments :: 0 >===
===< pass_ifs :: 0 >===
===< pass_includes :: 0 >===
===< pass_line_markers ::
Package: cvise
Version: 2.8.0-1
I cannot run cvise in Debian/sid:
% cvise check.sh testcase.i
00:00:07 INFO ===< 150150 >===
00:00:07 INFO running 4 interestingness tests in parallel
00:00:07 INFO INITIAL PASSES
00:00:07 INFO ===< IncludesPass >===
Traceback (most recent call last):
File "/usr/
Package: g++-13
Version: 13.1.0-7
Forwarded: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110560
Affects: src:highway
src:highway fails to compile on riscv64. Confirmed upstream. Already
fixed in GCC14, backported to GCC13:
* https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110560#c2
Package: g++-13
Version: 13.1.0-6
Forwarded: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110264
Affects: src:highway
src:highway fails to compile on riscv64. Confirmed upstream. Fixed in GCC14
Package: g++-13
Version: 13.1.0-6
Forwarded: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110280
Affects: src:highway
src:highway fails to compile on arm64. Confirmed upstream
Version: 12.2.0-8
malat@barriere ~ % apt-cache policy gcc-12
gcc-12:
Installed: 12.2.0-18
Candidate: 12.2.0-18
Version table:
*** 12.2.0-18 100
1 https://deb.debian.org/debian experimental/main i386 Packages
100 /var/lib/dpkg/status
12.2.0-14 500
500 https://d
Source: gcc-13
Version: 13-20230411-1
I cannot compile highway on riscv/rv64gcv1p0 with gcc-13.
Compilation fais with:
% /usr/bin/g++-13 -freport-bug -DHWY_SHARED_DEFINE
-I"/home/malat/highway-1.0.4~git20230317.8681eb8" -g -O2
-ffile-prefix-map=/home/malat/highway-1.0.4~git20230317.8681eb8=.
-f
Control: reopen -1
> the upstream fix went into 12.2.0-2.
Would be so kind as to demonstrate how ?
Here is what I see on my side:
% ssh barriere.debian.org
% sessionid=$(schroot -b -c sid_i386-dchroot)
% dd-schroot-cmd -c $sessionid apt-get update
% dd-schroot-cmd -c $sessionid apt-get upgrade
Control: severity -1 grave
By default packages on Debian are build with -O2 since I see that it
affects release arches:
* armel
* i386
* mips64el
* mipsel
Would it be possible to include fix:
https://gcc.gnu.org/g:9f532fec01d6651cc3cc136073f044a7953d8560
Thanks,
Control: forwarded -1 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=108927
Control: tags -1 confirmed upstream patch
[PATCH] RISC-V: Remove void_type_node of void_args for vsetvlmax intrinsic
https://gcc.gnu.org/pipermail/gcc-patches/2023-February/612858.html
Source: gcc-snapshot
riscv64 intrisincs are currently broken.
% cat t.c
#include
int main()
{
size_t vl = __riscv_vsetvlmax_e8mf8();
return vl;
}
Gives:
g++ -march=rv64gcv1p0 t.c
t.c: In function 'int main()':
t.c:5:39: error: too few arguments to function 'long unsigned int
__riscv_
Control: reassign -1 libc6.1-dev 2.36-5
Looks like the issue is not fixed on ia64 / sparc64.
Steps:
% cat p.cxx
#include
int main() { return 0; }
Lead to:
malat@yttrium ~ % /usr/lib/gcc-snapshot/bin/g++ -v p.cxx
Using built-in specs.
COLLECT_GCC=/usr/lib/gcc-snapshot/bin/g++
COLLECT_LTO_WRAP
Package: gcc-snapshot
Version: 1:20230108-1
on riscv64, here is what I see:
$ cat /tmp/bla.c
int main() {}
$ /usr/lib/gcc-snapshot/bin/gcc /tmp/bla.c
/usr/bin/ld: cannot find -lgcc_s: No such file or directory
collect2: error: ld returned 1 exit status
Thanks
Control: forwarded -1 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=78971
Here is the current state for compiling webkit on mipsel (1)
# Lower memory requirements on architectures with only 2 GB address space
ifneq (,$(filter $(DEB_HOST_ARCH),mips mipsel sh4))
CFLAGS := $(CFLAGS:-g1=-g0)
CFLAGS :=
Control: found 1016903 12.2.0-9
See i386:
*
https://buildd.debian.org/status/fetch.php?pkg=highway&arch=i386&ver=1.0.3%7Egit20221102.4899d11-2&stamp=1667806545&raw=0
Control: found 1016903 12.2.0-7
For example ppc64:
*
https://buildd.debian.org/status/fetch.php?pkg=highway&arch=ppc64&ver=1.0.3%7Egit20221102.4899d11-1&stamp=1667470496&raw=0
Control: reopen -1
> -- Forwarded message --
> From: Matthias Klose
> To: 1016903-d...@bugs.debian.org, 1016...@bugs.debian.org
> Cc:
> Bcc:
> Date: Thu, 3 Nov 2022 09:28:54 +0100
> Subject: Fixed in 12.2.0-2 and later.
> Version: 12.2.0-2
>
> Fixed in 12.2.0-2 and later.
Nope. A
Control: affects -1 src:jpeg-xl
Just a wild guess:
*
https://buildd.debian.org/status/fetch.php?pkg=jpeg-xl&arch=armhf&ver=0.7.0-2&stamp=1664702466&raw=0
For reference:
malat@amdahl /tmp % apt-cache policy libc6-dev
libc6-dev:
Installed: 2.35-1
Candidate: 2.35-1
Version table:
*** 2.35-1 500
500 https://deb.debian.org/debian sid/main armhf Packages
100 /var/lib/dpkg/status
Source: gcc-snapshot
Version: 1:20220920-1
Severity: grave
Per original reference:
--- Comment #1 from Andrew Pinski ---
Is this a packaging issue?
> ignoring nonexistent directory
> "/usr/lib/gcc-snapshot/lib/gcc/arm-linux-gnueabihf/13/include-fixed/arm-linux-gnueabihf"
ignoring nonexistent di
Control: fixed -1 1:20220920-1
06187
--- Comment #55 from Richard Earnshaw ---
(In reply to Mathieu Malaterre from comment #53)
>
> gcc-12 seems to be generating wrong-code for a different unit-test:
I've just pushed my patch to the gcc-12 branch, could you try that please?
Package: g++-12
Version: 12.1.0-7
Forwarded: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=106322
Affects: src:highway
I am getting some odd behavior for unit test of highway. I believe
there is some wrong-code generation using g++ + -O2 (tree-vectorize is
default now).
Package: gcc-11
Version: 11.3.0-3
Forwarded: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=106187
g++-11 -02 cannot be used to compile highway on armhf. It seems to be
fixed in g++-12.
Package: gcc-11
Version: 11.3.0-3
This is a followup to bug #928224. It turns out that gcc-11 is now
generating neon instructions, which contradicts our definition of
armhf.
Reference:
$ ssh abel.debian.org
[...]
malat@abel ~ % gcc-11 --verbose
Using built-in specs.
COLLECT_GCC=gcc-11
COLLECT_LT
Source: gcc-11
Version: 11.2.0-13
At least on hppa combination of __asm__ .global + .symver results in
undefined symbols in the shared library:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1003088
https://buildd.debian.org/status/fetch.php?pkg=libkcapi&arch=hppa&ver=1.3.1-1&stamp=1641233787
Package: cvise
Version: 2.1.0-1
Severity: normal
Tags: patch
Dear Maintainer,
Could you please add the missing line in d/control:
Recommends: colordiff
Currently cvise tries and fails if not found withg:
/bin/sh: 1: colordiff: not found
-- System Information:
Debian Release: 11.1
APT prefer
On Thu, Nov 15, 2018 at 8:07 AM Mathieu Malaterre wrote:
>
> On Wed, Nov 14, 2018 at 3:35 PM YunQiang Su wrote:
> >
> > YunQiang Su 于2018年11月14日周三 下午10:29写道：
> > >
> > > With read as(1), I see
> > >
> > > --reduce-memory-overheads
> &
e went from:
as: out of memory allocating 4072 bytes after a total of 546328576 bytes
to
as: out of memory allocating 4072 bytes after a total of 557305856 bytes
So I would even be tempted to report a bug to gcc/gas...
> > Mathieu Malaterre 于2018年11月9日周五 下午5:14写道：
> > >
> &
Hi James,
On Thu, Nov 23, 2017 at 1:14 PM, James Cowgill wrote:
> On 23/11/17 11:42, Mathieu Malaterre wrote:
>> YunQiang,
>>
>> Do you know of any drawbacks ?
>
> This has been raised before:
> https://lists.debian.org/debian-mips/2016/10/msg00049.html
>
>
YunQiang,
Do you know of any drawbacks ?
On Thu, Nov 23, 2017 at 11:29 AM, YunQiang Su wrote:
> I guess we should set this param for mips/mipsel by default?
>
> On Thu, Nov 23, 2017 at 5:20 AM, Mathieu Malaterre wrote:
>> On Wed, Nov 22, 2017 at 2:24 PM, Mathieu Malaterre wrote
Hi,
On Tue, Aug 22, 2017 at 4:11 PM, Matthias Klose wrote:
> Control: tags -1 + wontfix
>
> On 22.08.2017 10:10, Aaron Marcher wrote:
>> Package: gcc-multilib
>> Version: 4:6.3.0-4
>>
>> The package gcc-multilib conflicts with any GCC cross toolchains (for
>> example "gcc-aarch64-linux-gnu"). As
On Sun, Feb 26, 2017 at 7:33 PM, John Paul Adrian Glaubitz
wrote:
[...]
> Since PIE is already enabled for ppc64el, I think we should also enable
> it for ppc64 to mitigate this problem.
[...]
Would it be possible to also activate PIE on powerpc ? The test suite
is also failing for the exact same
On Sat, Nov 26, 2016 at 3:48 PM, Matthias Klose wrote:
> Control: severity -1 important
> Control: reassign -1 gcc-6
>
> Mathieu, these issues are not RC issues, gcc doesn't become unusable.
I managed to skrew up the `reassign` and missed the original
`severity` report. Sorry for the mess :(
Package: src:gcc-6
Version: 6.2.0-4
Severity: important
[...]
Observed that boot arguments (passed as CONFIG_CMDLINE) are not being
picked up by kernel while using gcc-ppc64-linux-gnu v5.4.0 and v6.1.1.
While it works as expected with v5.3.1 .
Found that in init/main.c in setup_command_line() t
Control: forwarded -1 https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71777
On Tue, Jul 5, 2016 at 10:09 PM, Adam Borowski wrote:
> On Tue, Jul 05, 2016 at 08:31:24AM +0200, Mathieu Malaterre wrote:
>> I cannot see the builtin mentionned anywhere other than on the X86 page:
>>
>&g
Package: gcc,libboost-program-options-dev
Version: gcc/4:4.7.2-1
Version: libboost-program-options-dev/1.49.0.1
Severity: important
For some reason I cannot compile and link to a boost lib using -flto
-Ofast (everything is fine when each is taken separatly). I am
assigning to both packages, feel f
Control: severity 754349 important
On Thu, Jul 10, 2014 at 12:49 PM, Emilio Pozuelo Monfort
wrote:
[...]
> I scheduled rebuilds on kfreebsd and sparc and the rebuilds succeeded as they
> now picked ecj & ecj-gcj 3.10.0+3.9.0-3 rather than 3.10.0-1. There is still a
> bug somewhere as the package
On Wed, Jul 9, 2014 at 11:35 AM, Matthias Klose wrote:
> Please tell the versions used for both ecj and ecj-gcj.
ecj (3.10.0+3.9.0-3)
ecj-gcj (3.10.0+3.9.0-3)
--
To UNSUBSCRIBE, email to debian-gcc-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debi
On Wed, Jul 9, 2014 at 11:36 AM, Matthias Klose wrote:
> Please tell the versions used for both ecj and ecj-gcj.
ecj (3.10.0+3.9.0-3)
ecj-gcj (3.10.0+3.9.0-3)
--
To UNSUBSCRIBE, email to debian-gcc-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debi
Package: gcc-4.9
Version: 4.9.0-10
Severity: serious
Affects: src:vtk-dicom
Looks like java compiler from latest gcc 4.9 still does not work:
/usr/bin/javac -source 1.5 -target 1.5 -classpath
/usr/share/java/vtk.jar:/usr/lib/vtk-5.8/java -sourcepath
/usr/lib/vtk-5.8/java/vtk/ -d /«PKGBUILDDIR»/ob
Package: gcc-4.9
Version: 4.9.0-9
Severity: serious
Affects: src:gdcm
GDCM FTBFS on multiple (exotic) platforms namely hurd/kfreebsd and
sparc. All of which are using the java compiler from gcc. So I am
reporting the issue here as serious:
-- Java version 1.5.0 configured successfully!
Exception
Package: gcc-multilib
Version: 4:4.7.2-1
Severity: important
For some reason I can no longuer use my 32bits schroot system to compile it
fails with:
$ echo "int main(){}" > t.c
$ gcc -v t.c
Using built-in specs.
COLLECT_GCC=gcc
COLLECT_LTO_WRAPPER=/usr/lib/gcc/i486-linux-gnu/4.7/lto-wrapper
Targ
found 697842 4.7.2-17
notfound 697842 4.7.2-5
thanks
My system contained clang++ from experimental which pulled some extra
deps. I do not think this is relevant to report bug on experimental.
Feel free to close.
Thanks
--
To UNSUBSCRIBE, email to debian-gcc-requ...@lists.debian.org
with a subj
severity 696320 important
severity 696854 important
thanks
downgrading severity since there are workarounds (using LD_LIBRARY_PATH)
--
To UNSUBSCRIBE, email to debian-gcc-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Archive:
http://list
Here is the output for libstdc++6-4.7-dbg: (debian sid):
$ gdb bla
(gdb) set verbose on
(gdb) r
Starting program: /tmp/bla
Reading symbols from /lib64/ld-linux-x86-64.so.2...(no debugging
symbols found)...done.
Loaded symbols for /lib64/ld-linux-x86-64.so.2
Reading symbols from system-supplied DSO
On Thu, Dec 20, 2012 at 9:31 PM, Andrey Rahmatullin wrote:
>
> /usr/lib/x86_64-linux-gnu/debug/libstdc++.so.6 instead of
> /usr/lib/debug/usr/lib/x86_64-linux-gnu/libstdc++.so.6 sounds wrong to me.
> Also, if it's indeed wrong then other versions of libstdc++ are affected
> too.
I am not clear wh
Package: libstdc++6-4.4-doc
Version: 4.4.5-8
Severity: minor
Tags: patch
There is a minor typo in the documentation:
$ grep gdbint /usr/share/doc/gcc-4.4-base/libstdc++/html/manual/debug.html
or put into a .gdbint file to establish default debugging
It should read as .gdbinit
-- System Infor
Forgot the file.
bla.cxx
Description: Binary data
Package: libstdc++6-4.4-dbg
Version: 4.4.5-8
Severity: grave
Dear packager,
I believe the current -dbg package is not working as it should. Here is what I
did:
$ g++ -o bla -g bla.cxx
$ gdb bla
GNU gdb (GDB) 7.3-debian
Copyright (C) 2011 Free Software Foundation, Inc.
License GPLv3+: GNU GPL ve
Package: gcj-4.6-jre-headless
Version: 4.6.2-2
Severity: important
hi,
for some reason .java-gcj-4.6.jinfo is installed in inproper location:
/usr/lib/jvm/java-1.5.0-gcj-4.6/.java-gcj-4.6.jinfo
This breaks update-java-alternatives behavior
Thanks
-- System Information:
Debian Release: 6
Package: gcj-jdk
Version: 4.6.1-9
Severity: serious
Justification: FTBFS
Hi,
one of my package does not build properly on kfreebsd-amd64:
https://buildd.debian.org/status/fetch.php?pkg=gdcm&arch=kfreebsd-amd64&ver=2.0.18-5&stamp=1318325853
[ 49%] javac *.java; jar cvf -> gdcm.jar
cd
/bu
Package: g++-4.4
Version: 4.4.5-8
Severity: normal
File: /usr/bin/g++-4.4
For some reason I cannot compile anymore using my default g++ compiler. Trying
to compile the GDCM project leads to:
cd
/home/mathieu/debian/debian-med/trunk/packages/gdcm/trunk/gdcm-2.0.18/debian/build-python2.6/Source/
On Tue, Apr 26, 2011 at 5:31 PM, Konstantinos Margaritis
wrote:
> On 26 April 2011 18:03, Matthias Klose wrote:
>> I'll make GCC 4.6 the default after the release of
>> GCC 4.5.3, expected later this week, at least on amd64, armel, i386 and
>> powerpc.
>
> Could you include armhf in the list as w
Source: gcc-4.4
Version: 4.4.4-8
Severity: serious
Justification: FTBFS
g++ 4.4.4-8 fails to compile ParaView on armel. See:
https://buildd.debian.org/fetch.cgi?pkg=paraview&arch=armel&ver=3.8.1-1&stamp=1282234558&file=log&as=raw
[ 33%] Building CXX object
VTK/VolumeRendering/CMakeFiles/vtkVolum
On Sun, Aug 30, 2009 at 9:19 AM, Matthias Klose wrote:
> On 07.07.2009 18:08, Mathieu Malaterre wrote:
>>
>> Package: gcc-snapshot
>> Version: 20090620-1
>> Severity: normal
>>
>>
>> It would be nice if the package would provide a proper javac symlink
javac
Description: Binary data
Package: gcc-snapshot
Version: 20090620-1
Severity: normal
It would be nice if the package would provide a proper javac symlink just like
gcj package is doing.
See bug #528084 for reference.
Thank you
-- System Information:
Debian Release: 5.0.2
APT prefers stable
APT policy: (500, 'stab
Thanks Falk, you can close the bug. Updating binutils from unstable did
the trick
$ sudo apt-get install -t unstable binutils
Mathieu
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
$ /usr/lib/gcc-snapshot/bin/g++ -o RegularExpression.o -Dvtksys_EXPORTS
-g
-I/home/mathieu/Dashboards/MyTests/VTK-gcc-snapshot/Utilities
-DVTK_IN_VTK -DKWSYS_NAMESPACE=vtksys -c
/home/mathieu/Dashboards/MyTests/ParaView/VTK/Utilities/kwsys/RegularExpression.cxx
/tmp/ccRIu5Mp.s: Assemble
Package: gcc-snapshot
Version: 20051124-1
Severity: important
I cannot build VTK (http://vtk.org) using gcc-snapshot and
$ (cd Utilities/kwsys && make)
Building dependencies. cmake.depends...
Building object file Directory.o...
/tmp/ccFywWsc.s: Assembler messages:
/tmp/ccFywWsc.s:2650: Error: u
Package: gcc-4.0
Version: 4.0.1-2
Severity: important
Hi,
Description:
I am running a program that is outputing a file. The very same program
when run with LD_PRELOAD set to libGL.so, is segfaulting. The segfault
does not occur when using gcc2.95, gcc3.2, gcc3.3 and gcc3.4 (debian
package). The s
75 matches
Mail list logo
