Package: udev
Version: 257.5-2
Severity: important
Hi!
We tested a bookworm to trixie update on one of our systems. The host
did not come back online after reboot, and the issue seems to be caused
by the 252 -> 257 systemd update.
The main network interface on the system was named "enp1s0" by sy
-build-options
Thanks!
Emanuele
>From f48d9086372b33b0ec2df93d48a054356699a54e Mon Sep 17 00:00:00 2001
From: Emanuele Rocca
Date: Tue, 6 May 2025 14:55:59 +0200
Subject: [PATCH 2/2] Support parallel builds
---
debian/rules | 9 +++--
1 file changed, 7 insertions(+), 2 deletions(-)
d
://wiki.debian.org/ToolChain/PACBTI#Packages_overriding_the_default_build_flags
From: Emanuele Rocca
Date: Tue, 6 May 2025 14:41:12 +0200
Subject: [PATCH] Do not override FFLAGS
---
debian/macro.f95_debian | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/debian/macro.f95_debian b/debian
https://wiki.debian.org/ToolChain/PACBTI#Packages_overriding_the_default_build_flags
From: Emanuele Rocca
Date: Tue, 6 May 2025 13:38:12 +0200
Subject: [PATCH] Honor CC and CFLAGS environment variables
---
src/makefile.unx | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/src/makefile.unx
attached.
Emanuele
From: Emanuele Rocca
Date: Tue, 6 May 2025 10:59:33 +0200
Subject: [PATCH] Use default CFLAGS and CXXFLAGS set by dpkg
---
CMakeLists.txt | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/CMakeLists.txt b/CMakeLists.txt
index 837b203..40310af 100644
--- a
Source: aespipe
Version: 2.4h-1
Severity: wishlist
User: debian-...@lists.debian.org
Usertags: pac-bti
There is a new upstream version of aespipe available, 2.4i. It includes
support for PAC/BTI on arm64, and it would be nice to have it in Debian
at some point.
Thanks!
Emanuele
pstream? Where can we
find out more about it?
Thanks,
Emanuele
Hello Helmut,
On 2025-04-22 10:19, Helmut Grohne wrote:
> +override_dh_auto_build:
> + dh_auto_build -- 'BCCFLAGS=$(CPPFLAGS_FOR_BUIL) $(CFLAGS_FOR_BUILD)'
It looks like there's a typo here, this should be CPPFLAGS_FOR_BUILD.
Hello Niels,
On 2024-12-28 01:06, Niels Thykier wrote:
> Please review attached as an example of how to fix this problem.
>
> Note: Untested, since I was doing my testing on amd64.
LGTM. I applied your patch and built the package with a regular user as
follows:
$ dpkg-buildpackage -us -uc -b -r
ta Revision: 3
TPM Vendor ID: IFX
Vendor Specific data: 042203f2 0274706d 3338 ff
TPM Version: 0101
Manufacturer Info: 49465800
Finally, I see that you're booting your kernel with pci=noaer. Any
specific reasons for that?
Thanks,
Emanuele
r performing some specific tpm operations?
- Have you by any chance upgraded/downgraded your coreboot firmware?
- Please share the full output from dmesg.
- Please try a more recent kernel from bookworm-backports to see if the
problem can still be reproduced. See
https://backports.debian.org/Instructions/
Thanks,
Emanuele
Source: binutils
Version: 2.44-3
Severity: normal
Hi,
while building binutils with DEB_BUILD_OPTIONS=nocheck, I noticed that
at least some of the upstream tests were still being executed:
[...]
Schedule of variations:
unix
Running target unix
Using /usr/share/dejagnu/baseboards/unix.exp as
Hi,
On 2025-03-21 09:59, Bastian Germann wrote:
> Please remove desmume from arm64 and ppc64el to let the pkg migrate to
> trixie. The build fails on these archs and it is unlikely that somone is
> going to run the program there.
Please don't remove the arm64 version just yet. The software has ex
On 2025-03-19 09:45, Aurelien Jarno wrote:
> So far my finding are quite limited. It seems the new GCC generates more
> optimized code in general, and especially for the memmem() function, and
> that changes the location of the memmove() function located just after in
> the binary. Aligning it to 6
On 2025-03-19 08:01, Bastian Blank wrote:
> On Mon, Mar 17, 2025 at 03:47:11PM +0100, Emanuele Rocca wrote:
> > Makes sense. Also noteworthy, CONFIG_KALLSYMS_ALL is enabled on arm64 on
> > at least Fedora/RHEL/SuSE too.
>
> Please provide a justification.
- Needed for perf
Hi,
On 2025-03-16 01:20, Hastalavista_debian wrote:
> I wish to use LKRG (https://lkrg.org) on my RPi, but it reported error "LKRG
> requires
> CONFIG_KALLSYMS_ALL" when building, which is strange to me because I have
> never
> run
> into such problems on amd64 laptops. I checked config file of a
ate->pattern,
> | ^~
> bing_probes.c:31:1: note: include ‘’ or provide a declaration of
> ‘memcpy’
The fix suggested by the compiler (first part) works fine. See attached
patch.
Emanuele
>From ad2111765318727b9ce8ff40ddd25ae7ec728ce4 Mon Sep 17 00:00:0
gelog blahtexml-0.9/debian/changelog
--- blahtexml-0.9/debian/changelog 2012-04-20 12:59:48.0 +0200
+++ blahtexml-0.9/debian/changelog 2025-03-14 13:58:43.0 +0100
@@ -1,3 +1,15 @@
+blahtexml (0.9-1.2) unstable; urgency=medium
+
+ [ Emanuele Rocca ]
+ * Non-maintainer upload.
+ * D
rom ab969b5f8620be60c2ad6aef144b99e1d254f5e6 Mon Sep 17 00:00:00 2001
From: Emanuele Rocca
Date: Fri, 14 Mar 2025 13:58:22 +0100
Subject: [PATCH 1/2] Do not override CFLAGS, use CXXFLAGS for cpp
---
makefile | 12 +---
1 file changed, 5 insertions(+), 7 deletions(-)
diff --git a/makefile b/makefile
index 3249
the
appropriate flags. This is necessary, among other things, to enable
PAC/BTI on arm64: https://wiki.debian.org/ToolChain/PACBTI
Thanks,
Emanuele
>From 9c782b057222cd4b1da4499029ead6609013a765 Mon Sep 17 00:00:00 2001
From: Emanuele Rocca
Date: Fri, 14 Mar 2025 11:02:37 +
Subject: [PA
snippet in d/rules should do the trick:
# Support parallel builds
ifneq (,$(filter parallel=%,$(DEB_BUILD_OPTIONS)))
NJOBS := -j $(patsubst parallel=%,%,$(filter parallel=%,$(DEB_BUILD_OPTIONS)))
endif
And then pass the variable to make, something like: $(MAKE) $(NJOBS)
Thanks!
Emanuele
"Usage in debian/rules" in dpkg-buildflags(1).
Thanks,
Emanuele
should do the trick:
set(CMAKE_CXX_FLAGS "$ENV{CXXFLAGS} -Wall -Wextra -Werror=switch")
Thanks,
Emanuele
Hi,
I've prepared an NMU for android-platform-external-libunwind (versioned
as 10.0.0+r36-4.1). Please find the diff attached.
commit a2061b471eb85b27f67be83e0febd657e01d35ca
Author: Emanuele Rocca
Date: Wed Mar 5 13:38:39 2025 +0100
Update path to 7zip to fix FTBFS #1060969
diff --
Source: alevt
Version: 1:1.8.2-2
User: debian-...@lists.debian.org
Usertags: pac-bti
Hi,
alevt currently hardcodes CFLAGS in the Makefile:
https://sources.debian.org/src/alevt/1%3A1.8.2-2/Makefile/#L17
Please consider appending to CFLAGS instead, to make alevt use the
default CFLAGS set by dpkg-
variable should similarly be honored to allow users to build the
program with a different compiler. This can be useful for instance to
test building with an older compiler when investigating compiler
regressions.
Thanks!
Emanuele
[1]
https://www.gnu.org/software/make/manual/html_node/Conditional
-decls -c dis_relay.c -o dis_relay.o
Please ensure that acm honors CFLAGS when building dis_relay too.
Thanks!
Emanuele
-deb-build-options
Thanks!
Emanuele
commit a748bbfa58911ed3bd90d2318a459140b5aebf46
Author: Emanuele Rocca
Date: Thu Feb 27 10:28:47 2025 +
Support parallel builds
diff --git a/debian/rules b/debian/rules
index 5e37786..91053f8 100755
--- a/debian/rules
+++ b/debian/rules
@@ -15,6
actually build with hardening
flags, and enable branch protection.
Thanks!
Emanuele
diff --git a/debian/rules b/debian/rules
index a1da33b..5e37786 100755
--- a/debian/rules
+++ b/debian/rules
@@ -9,11 +9,10 @@
# Copyright © 2010-2014 Johnny Willemsen
#export DEB_BUILD_MAINT_OPTIONS = hardening
possible according to Policy 4.9:
https://www.debian.org/doc/debian-policy/ch-source.html#main-building-script-debian-rules
Thanks,
Emanuele
[1] https://manpages.debian.org/unstable/dpkg-dev/dpkg-buildflags.1.en.html
From: Emanuele Rocca
Date: Wed, 26 Feb 2025 14:51:46 +0100
X-Dgit-Generated
users wishing to do so can compile the program with a compiler other
than gcc.
Thanks,
Emanuele
[1] https://manpages.debian.org/unstable/dpkg-dev/dpkg-buildflags.1.en.html
Index: 3d-ascii-viewer-c-1.4.0+git20240503+ds/Makefile
Hi Ard,
On 2017-10-19 02:41, Ard Biesheuvel wrote:
> When booting an .iso image via HTTP boot from UEFI, the .iso image will
> be exposed to the OS as a ramdisk via the ACPI NFIT table, and will be
> picked up by the existing NFIT code in the kernel, which will expose it
> as a /dev/pmemXXX device
station the build shows no test
failure.
I tried however to build in a VM with only one CPU core available, ie:
passing the "-smp 1" argument to qemu. The build fails. Interestingly,
this is the case also with -smp 2. With -smp 3 the build seems fine.
Emanuele
https://people.debian.org/~ema/valgrind_armhf-2024-12-17T09:07:39Z.build
https://people.debian.org/~ema/valgrind_i386-2024-12-17T09:04:47Z.build
Can you please double-check once again?
Thanks,
Emanuele
to fail the build due to doxygen, and
> one of those is this one.
The version of doxygen currently in unstable is 1.9.8+ds-2, and toulbar2
builds fine with it.
Downgrading severity, please feel free to raise it again once doxygen
1.12.0 enters sid.
Thanks,
Emanuele
On 2024-12-06 05:25, Emanuele Rocca wrote:
> DEB_BUILD_MAINT_OPTIONS=-stackprotector
Apologies, I meant:
DEB_BUILD_MAINT_OPTIONS=hardening=-stackprotector
Hi,
On 2024-12-03 12:27, Andres Salomon wrote:
> Currently waiting on #1088699 to fix this.
On arm64 I tried building chromium 131.0.6778.85-1~deb12u1 without stack
protector, and the resulting build does not crash:
DEB_BUILD_MAINT_OPTIONS=-stackprotector
Not ideal, but better than a chromium
Hi Andres,
On 2024-12-03 12:27, Andres Salomon wrote:
> Currently waiting on #1088699 to fix this.
I tried building the affected chromium version using clang from
bookworm-proposed-updates with CC=clang-19 and CXX=clang++-19, and with
the following build-depends bumped:
lld-19, clang-19, cla
Package: sbuild
Version: 0.86.0
Severity: important
Hi!
Starting with version 0.86.0, sbuild using the unshare backend does not
build gcc packages successfully, at least on arm64 and amd64. The
version immediately before that, 0.85.11, works. Similar failures have
been seen on other architectures
Package: chromium
Version: 131.0.6778.85-1~deb12u1
Severity: serious
User: debian-...@lists.debian.org
Usertags: arm64
On arm64 systems, the version of chromium currently in bookworm-security
(131.0.6778.85-1~deb12u1) fails to start with several "stack smashing detected"
errors. The bookworm versi
Source: halide
Version: 18.0.0-8
Severity: normal
Tags: trixie sid ftbfs
User: debian-...@lists.debian.org
Usertags: arm64, sve
On arm64 systems with SVE support halide fails to build with the following
error:
[2087/4204] cd
/<>-build/build/stage-0/halide/python_bindings/apps &&
/<>-build/bui
h as:
curl http://localhost:8000/./testpackage_1.0.0-1_all.deb
In the specific case of curl, the --path-as-is option can be used to
skip normalization and reproduce.
Thanks,
Emanuele
Source: tiledarray
Version: 1.0.0-1
Severity: serious
Justification: FTBFS
Tags: trixie sid ftbfs
During a rebuild of all packages in sid, tiledarray failed to build from source
on both amd64 and arm64 with the following error message:
-- Performing Test EIGEN3_COMPILES - Success
CMake Error at /
Source: singularity-container
Version: 4.1.5+ds3-1
Severity: serious
Justification: FTBFS
Tags: trixie sid ftbfs
During a rebuild of all packages in sid, singularity-container failed to
build from source on both amd64 and arm64 with the following error
message:
github.com/sylabs/singularity/vendo
Source: nthash
Version: 2.3.0+dfsg-1
Severity: serious
Justification: FTBFS
Tags: trixie sid ftbfs
User: debian-...@lists.debian.org
Usertags: arm64
During a rebuild of all packages in sid, this package failed to build on
arm64. On amd64 it builds correctly.
Here is the error message:
[ 75%] Lin
Source: seqkit
Version: 2.8.2+ds-1
Severity: serious
Justification: FTBFS
Tags: trixie sid ftbfs
During a rebuild of all packages in sid, seqkit failed to build from
source on both amd64 and arm64 with the following error message:
github.com/shenwei356/seqkit/seqkit/cmd
# github.com/shenwei356/se
Source: fiat-ecmwf
Version: 1.4.0-1
Severity: serious
Justification: FTBFS
Tags: trixie sid ftbfs
User: debian-...@lists.debian.org
Usertags: arm64
During a rebuild of all packages in sid, this package failed to build on
arm64. On amd64 it builds correctly.
It seems that the build uses /usr/lib/x
Source: dbcsr
Version: 2.6.0-2
Severity: serious
Justification: FTBFS
Tags: trixie sid ftbfs
User: debian-...@lists.debian.org
Usertags: arm64
During a rebuild of all packages in sid, this package failed to build on
arm64. On amd64 it builds correctly.
It seems that the build uses /usr/lib/x86_64
Source: coccinelle
Version: 1.2.deb-1
Severity: serious
Justification: FTBFS
Tags: trixie sid ftbfs
During a rebuild of all packages in sid, coccinelle failed to build on
arm64 and amd64 with the following error.
[7]
Underfull \hbox (badness 1) in paragraph at lines 76--82
! Missing \cr inse
Source: xserver-xorg-video-trident
Version: 1:1.4.0-1
Severity: serious
Justification: FTBFS
Tags: trixie sid ftbfs
User: debian-...@lists.debian.org
Usertags: arm64
During a rebuild of all packages in sid, xserver-xorg-video-trident failed
to build on arm64 with the following error. On amd64 the
Source: xserver-xorg-video-tdfx
Version: 1:1.5.0-5
Severity: serious
Justification: FTBFS
Tags: trixie sid ftbfs
User: debian-...@lists.debian.org
Usertags: arm64
During a rebuild of all packages in sid, xserver-xorg-video-tdfx failed
to build on arm64 with the following error. On amd64 the packag
Source: superlu-dist
Version: 8.2.1+dfsg1-5
Severity: serious
Justification: FTBFS
Tags: trixie sid ftbfs
User: debian-...@lists.debian.org
Usertags: arm64
During a rebuild of all packages in sid, superlu-dist failed to build on
arm64 with the following error. On amd64 the package builds correctly
Source: rocsparse
Version: 5.7.1-5
Severity: serious
Justification: FTBFS
Tags: trixie sid ftbfs
During a rebuild of all packages in sid, rocsparse failed to build
with the following error.
Traceback (most recent call last):
File "/usr/lib/python3/dist-packages/sphinx/cmd/build.py", line 337,
Source: mfem
Version: 4.7+ds-1
Severity: serious
Justification: FTBFS
Tags: trixie sid ftbfs
User: debian-...@lists.debian.org
Usertags: arm64
During a rebuild of all packages in sid, mfem failed to build on arm64
with the following error. On amd64 the package builds correctly.
/usr/bin/ld: warni
Source: elkcode
Version: 9.6.8-1
Severity: serious
Justification: FTBFS
Tags: trixie sid ftbfs
User: debian-...@lists.debian.org
Usertags: arm64
During a rebuild of all packages in sid, this package failed to build on
arm64. On amd64 it builds correctly.
It seems that the build uses /usr/lib/x86_
Source: groonga
Version: 14.0.5+dfsg-2
Severity: serious
Justification: FTBFS
Tags: trixie sid ftbfs
During a rebuild of all packages in sid, groonga failed to build from
source on both amd64 and arm64 with the following error message:
/usr/bin/install: cannot stat './html/_static/webpack-macros.
Source: kf6-ktexttemplate
Version: 6.6.0-1
Severity: wishlist
Hi,
gcc-14 14.2.0-8 includes a fix for #1080974. I've just double-checked
that kf6-ktexttemplate 6.6.0-1 now builds properly on arm64 without the
need for -mno-late-ldp-fusion.
Please drop it from d/rules.
Thanks!
Emanuele
Control: tags -1 + fixed-upstream
On 2024-10-29 08:35, Emanuele Rocca wrote:
> Meanwhile the issue has been fixed upstream in GCC (trunk), a backport
> to 14 should be coming soon:
> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116783#c6
Here is the backport, merged in the gcc-14 bran
Source: coreboot
Version: 4.15~dfsg-4
Severity: serious
Justification: FTBFS
Tags: trixie sid ftbfs
During a rebuild of all packages in sid, this package failed to build on arm64
with the following error. On amd64 the package builds correctly.
accessors/cmos-hw-unix.c: In function ‘cmos_hal_read’
On 2024-09-11 06:52, Aurélien COUDERC wrote:
> I have the change committed locally and will upload it with KF6 6.6
> which release is at the door and should be out on Sep 13th.
Thanks.
Meanwhile the issue has been fixed upstream in GCC (trunk), a backport
to 14 should be coming soon:
https://gcc.
Hi,
On 2024-10-23 11:12, Paul Gevers wrote:
> On 23-10-2024 10:40, Emanuele Rocca wrote:
> > Indeed the simplest repro is just running the following in a sid armhf
> > schroot running on a amd64 host with binfmt qemu-user-static emulation
> > on:
>
> For avoidance of
On 2024-10-22 11:04, Étienne Mollier wrote:
> I just saw the answer from Paul in parallel, and apparently the crash
> condition also occurs in armhf on top of amd64.
Indeed the simplest repro is just running the following in a sid armhf
schroot running on a amd64 host with binfmt qemu-user-static
Hi,
On 2024-10-21 11:53, Étienne Mollier wrote:
> The issue looks very specific to this way of running Xvfb :5,
> other ways I know of within autopkgtests seem to be running fine
> in the same configuration. Quick lookup of open xvfb issues
> suggest that #1084230 could be relevant to the problem
consider a stable update for this change.
Emanuele
Hi Simon,
On 2024-09-24 09:16, Simon McVittie wrote:
> In a recent upload this failed on armhf, and only armhf: the actual
> tests appear to have all succeeded, but then xvfb-run exited with status
> 1 anyway. I am able to reproduce this on the armhf porterbox amdahl.
>
> A simplified reproducer
Hey Aurélien,
On 2024-09-06 04:37, Emanuele Rocca wrote:
> I thus tried to build the package with -O2 -mno-late-ldp-fusion, and
> indeed it builds fine
For the time being you can, if you want, drop the -O1 and build with
-mno-late-ldp-fusion instead:
export DEB_CXXFLAGS_MAINT_APPEND +
ne in order to test the functionality?
Thanks,
Emanuele
[1] https://sourceware.org/gdb/current/onlinedocs/gdb.html/AMD-GPU.html
Source: mir
Version: 2.14.1-7
Severity: serious
Tags: sid trixie ftbfs
User: debian-...@lists.debian.org
Usertags: arm64 armel armhf
mir fails to build with the following error on arm64, armel, armhf,
alpha, hppa, powerpc and sparc64. The problem can be reproduced with
both GCC 14 and 13, hence it
Hi,
On 2024-08-15 10:46, Andreas Beckmann wrote:
> /usr/lib/llvm-17/include/clang/Basic/TokenKinds.def:23:26: error: pasting
> "kw_" and "[" does not give a valid preprocessing token
>23 | #define KEYWORD(X,Y) TOK(kw_ ## X)
This is a LLVM 17 / GCC 14 regression:
https://bugs.debian.org/10770
On 2024-09-06 10:04, Emanuele Rocca wrote:
> The problem seems to be a GCC 14 regression, given that with GCC 13 and
> optimization level -O2 the build is fine.
The issue seems related to the new optimization pass to fuse loads and
stores to adjacent memory locations into load and stor
Source: kf6-ktexttemplate
Version: 6.5.0-2
Tags: sid trixie
User: debian-...@lists.debian.org
Usertags: arm64
X-Debbugs-Cc: Aurélien COUDERC
When building kf6-ktexttemplate on arm64 with gcc-14 and -O2, the build
fails due to the following build-time tests segfaulting:
2 - testbuiltins
On 2024-07-25 08:52, NoisyCoil wrote:
> llvm-toolchain-17 fails to build from source with gcc-14 on arm64 due to this
> upstream bug:
> https://github.com/llvm/llvm-project/issues/78691. A fix was merged to v18
> (https://github.com/llvm/llvm-project/pull/78704), but AFAICS the patch was
> not ba
Bonjour Cyril,
On 2024-08-20 11:45, Cyril Brulebois wrote:
> I couldn't replicate the FTBFS within a devel sid chroot that has tons of
> extra packages, including python3-packaging, and python3-platformdirs, but
> not python3-jaraco.text.
Could it be that your sid chroot has an older version of
Package: python3-pkg-resources
Version: 72.2.0-1
Severity: serious
X-Debbugs-CC: debian-b...@lists.debian.org
[debian-boot added to CC as this issue breaks d-i builds]
Importing pkg_resources fails with the following error:
(sid-amd64-sbuild)ema@ariel:~$ python3
Python 3.12.5 (main, Aug 7 2
Hi Bernhard,
On 2024-08-17 05:49, Bernhard Übelacker wrote:
> If pointer authentication is not part of the arm64 baseline,
> maybe the package isa-support should provide a package
> which rr can depend on, like it does to sse2-support at i386?
Basic pointer authentication instructions are defined
ainer upload.
+ * Call dh_autoreconf to update config.{sub,guess} and aclocal.m4.
+Closes: #727343
+
+ -- Emanuele Rocca Mon, 19 Aug 2024 14:26:02 +
+
chise-base (0.3.0-2.4) unstable; urgency=medium
* Non-maintainer upload.
diff -Nru chise-base-0.3.0/debian/rules chise-base-0.3.0/d
Source: pytorch
Version: 2.1.2+dfsg-4
Severity: serious
Justification: FTBFS
Tags: sid ftbfs
User: lu...@debian.org
Usertags: ftbfs-impfuncdef
Dear Maintainer,
pytorch fails to build from source with the following errors:
/<>/torch/csrc/dynamo/eval_frame.c: In function ‘eval_custom_code’:
/<>/to
Package: python3-torch
Version: 2.1.2+dfsg-4
Severity: serious
Hi,
python3-torch cannot be installed in sid:
(sid-amd64-sbuild)ema@ariel:/$ sudo apt install python3-torch
Some packages could not be installed. This may mean that you have
requested an impossible situation or if you are using
On 2024-06-22 08:48, Salvatore Bonaccorso wrote:
> This one will be to late for 12.6 bookworm point release, but maybe we
> can try to have it for the august bookworm point release? Or are there
> concerns about regressions?
No specific concerns. The changes are all in the upstream branch
releases
2:26:30 xxx named[3063010]: loading configuration: failure
Jul 25 12:26:30 xxx named[3063010]: exiting (due to fatal error)
Jul 25 12:26:30 xxx systemd[1]: named.service: Main process exited,
code=exited, status=1/FAILURE
--
Emanuele Leuzzi
Reti & Sistemi
SSIS - Società Servizi Informati
64 bit architectures,
i686-w64-mingw32-objcopy on 32 bits.
* Do not override DEB_HOST_GNU_CPU in d/strip_debug.sh, use a new variable
instead.
I have implemented the changes he suggested, please find a patch
attached.
Emanuele
diff -Nru dxvk-2.4/debian/changelog dxvk-2.4/debian/changelog
Hi,
On 2024-07-18 11:45, Aurelien Jarno wrote:
> With glibc 2.39, the revert is not possible anymore, therefore I
> rocm-hipamd FTBFS again. I have therefore upgraded the severity to
> serious.
>
> Emanuele Rocca is working on a solution on the upstream LLVM side. In
> the mean
!
Emanuele
From: Emanuele Rocca
Date: Fri, 19 Jul 2024 09:25:38 +0200
Subject: [PATCH] [Clang] make SVE types known to device targets too (#99446)
For the purpose of preprocessing and declarations in header files,
ensure clang accepts SVE types for both device and host targets.
Co-authored-by
Hi,
I could reproduce the problem with GCC 12 on aarch64, but not with GCC
13 and 14.
Using the following test program:
#include
float test(_Float16 x){
return x;
}
int main() {
printf("%f\n", test(1.0));
}
This is the error message you get on aarch64 with g++-12:
$
come up
with a fix. It seems that they will be shipping it as part of the
upcoming 2.4.2 release.
To see if the proposed patch actually works for us, I've quilt-imported
[2], added python3-skimage to Build-Depends, and verified that indeed
the package builds fine on arm64 again.
.
Thanks,
Emanuele
Hi,
On 2024-02-09 11:36, Emanuele Rocca wrote:
> In order to properly support PAC/BTI in Debian we need first GCC to
> enable support for the feature, and that has not happened yet.
PAC/BTI support is now turned on in GCC starting with 13.3.0-2. I have tried a
glibc rebuild in sid wi
On 2024-07-01 03:12, Emanuele Rocca wrote:
> I haven't tested it on cross-builds of the native compiler as that
> currently fails due to unsatisfied dependencies in sid. That case
> should work fine too though.
Double-checked today, cross-builds of the native compiler work as well
Hi Matthias,
you can find an updated patch attached, it allows to disable the flag by
building with DEB_BUILD_MAINT_OPTIONS=hardening=-branch.
It does the right thing on libgcc-13-dev built on arm64 as well as
libgcc-13-dev-arm64-cross built on amd64. I haven't tested it on
cross-builds of the na
Control: tags -1 + patch
On 2024-06-14 10:42, Emanuele Rocca wrote:
> stderr: x86_64-w64-mingw32-gcc: error: unrecognized command-line option
> ‘-mbranch-protection=standard’
Please consider applying the attached patch to fix this FTBFS bug by
using CFLAGS as defined for amd64 instead of
Source: dxvk
Version: 2.3.1-1
Severity: serious
User: debian-...@lists.debian.org
Usertags: pac-bti
dxvk fails to build from source on arm64 due to the aarch64-specific
compiler flag -mbranch-protection=standard leaking into CFLAGS used by
x86_64-w64-mingw32-gcc:
meson.build:1:0: ERROR: Unable t
Source: elfutils
Version: 0.191-1
Severity: important
All test failures during make check are ignored, see the following line
in debian/rules:
-LC_CTYPE=C $(MAKE) check $(USE_JOBS) || { cat tests/test-suite.log; exit 1; }
Please drop the initial hyphen to ensure that the build fails in case of
Hello!
On 2024-06-06 12:48, Bo YU wrote:
> hmm, here I use libdw-dev instead of libunwind-dev as upstream
> suggested. I just have uploaded it to mentor because my mentor did not
> respond to me for a while. Could you help me to try to build it on
> arm64 or armhf again? I have tested it on qemu-u
tps://salsa.debian.org/debian/strace/-/commit/711a9970c553b8fb9a5b9eb3a4b80600eb8112c4
With the attached patch strace builds fine on arm64 and armhf.
Cheers,
Emanuele
diff -Nru strace-6.8/debian/changelog strace-6.8/debian/changelog
--- strace-6.8/debian/changelog 2024-05-10 06:53:53.0 +0
Hi,
On 2024-06-04 04:17, John wrote:
> CONFIG_ARCH_R9A09G057=y #RZ/V2H
Just a note to mention that this setting is being added in Linux 6.10,
we will have to wait for such version to be packaged before enabling it.
Hi,
On 2024-05-30 12:00, Thorsten Leemhuis wrote:
> On 29.05.24 23:12, Emanuele Rocca wrote:
> > On 2024-05-27 02:31, Thorsten Leemhuis wrote:
> >> Would also help a lot to know if this is a 6.8.y only thing, or happens
> >> with 6.9 and mainline as well, as 6.8.
sources.debian.org/src/siridb-server/2.0.51-3/debian/control/#L18
FTR I am tracking these problems with a usertag:
https://bugs.debian.org/cgi-bin/pkgreport.cgi?users=debian-...@lists.debian.org;tag=32bit-stackclash
Emanuele
Hey,
On 2023-11-05 04:12, Hervé Werner wrote:
> I faced this issue on real data but I struggled to find a reliable scenario
> to reproduce it. Here is what I just came up with:
> sudo mkfs -t ext4 -O fast_commit,inline_data /dev/sdb
> sudo mount /dev/sdb /mnt/
> sudo install -d -o myuser /m
e:
qemu-system-x86_64 -cpu host -machine accel=kvm -drive
file=sid.img,format=qcow2 -m 4G -smp `nproc`
Upgrade the kernel, and run the autopkgtest again as above.
Emanuele
turned on, see
attached patch.
Thanks,
Emanuele
[1] https://wiki.debian.org/ToolChain/PACBTI
diff -Nru xfsprogs-6.5.0/debian/rules xfsprogs-6.5.0/debian/rules
--- xfsprogs-6.5.0/debian/rules 2023-08-15 21:15:28.0 +0200
+++ xfsprogs-6.5.0/debian/rules 2024-02-12 15:51:07.0 +0100
1 - 100 of 619 matches
Mail list logo
