- Original Message -
> From: "Adam D. Barratt"
> To: "Timothy Pearson" , "1061652"
> <1061...@bugs.debian.org>
> Sent: Monday, January 29, 2024 11:21:44 AM
> Subject: Re: Bug#1061652: bookworm-pu: package rss-glx/0.9.1-6.4+deb12u1
&
0.0 -0600
@@ -1,3 +1,12 @@
+rss-glx (0.9.1-6.4~deb12u1) bookworm; urgency=medium
+
+ * Non-maintainer upload.
+ * debian/patches/glfinish.patch: Call GLFinish() prior to glXSwapBuffers()
+(Closes: #1061507)
+ * Install screensavers into /usr/libexec/xscreensaver (Closes: #979490)
+
+
Is this still a problem? If so, the package to install will be rss-glx-dbgsym,
and a "thread apply all bt" in GDB would give the needed debug information once
the flux process hangs.
.
+ * debian/patches/glfinish.patch: Call GLFinish() prior to glXSwapBuffers()
+(Closes: #1061507)
+ * Install screensavers into /usr/libexec/xscreensaver (Closes: #979490)
+
+ -- Timothy Pearson Sat, 27 Jan 2024 08:41:00 -0600
+
rss-glx (0.9.1-6.3) unstable; urgency=medium
* Non-maintainer
Would it be possible for someone that can reproduce this bug to do the
following:
* Install the rss-glx-dbgsym package
* Provoke the bug
* Attach gdb to the running process with 'gdb --pid '
* Grab a backtrace and upload here?
If I can see where the process is stuck I'm willing to take a
Package: rss-glx
Version: 0.9.1-6.3+b1
Tags: patch
On certain hardware and with specific X11 windows (such as the TDE lock
screen), the RSS-GLX hacks no longer display after upgrading to Bookworm from
Bullseye. This was traced to a missing glFinish() call prior to
glXSwapBuffers(); likely the
Root cause found, merge request here:
https://salsa.debian.org/kernel-team/linux/-/merge_requests/917
I have traced this bug to a missing memory barrier in the powerpc IPI handling
code. io_uring uses task_work_add() to schedule I/O worker creation, which in
turn issues an IPI, and when precise timing conditions are met the inconsistent
state between the two CPU cores can lead to corruption of
I have traced this to a regression in the Linux kernel. The issue appears to
be a type of data race that is more likely to occur on ppc64el than on other
architectures, but is also likely to affect other architectures. The issue
remains in the latest GIT version of the Linux kernel, and I am
For 16 to work we'll need the Debian clang team to include this patchset:
https://reviews.llvm.org/D158066
Any chance of that happening?
- Original Message -
> From: "Andres Salomon"
> To: "Leandro Cunha" , 1051...@bugs.debian.org
> Cc: "Timothy Pear
We've started hitting this on a busy server after upgrading to Bookworm:
2023-09-07 17:00:31 0 [Warning] You need to use --log-bin to make
--expire-logs-days or --binlog-expire-logs-seconds work.
2023-09-07 17:00:31 0 [Note] Server socket created on IP: '127.0.0.1'.
2023-09-07 17:00:31 0 [Note]
Package: libclang-common-15-dev
Version: 1:15.0.7-8
Severity: important
Tags: upstream
Forwarded: https://github.com/llvm/llvm-project/issues/64664
After upgrading from clang 14 to clang 15 debian's chromium package fails to
compile with SIMD errors:
In my opinion, as a maintainer and user of Chromium (as distinct from Chrome),
we absolutely need to ship this with the code removed / disabled. This is a
deliberate attempt to lock out the open Web, to force the use of a
Google-approved (and Google-locked) software, firmware, and hardware
Package: kotlin
Version: 1.3.31
When attempting to package Jitsi for Debian, I ran into a blocking issue due to
Kotlin 1.3.31 not providing the typeOf() reflection function, which was
introduced as experimental in Kotlin 1.3.40. Since the lift to move from one
minor version to another is
Earlier merge request was redundant, after futher tracing I believe this merge
request directly addresses the root of the remaining issues:
https://salsa.debian.org/installer-team/debian-installer/-/merge_requests/34
- Original Message -
> From: "Timothy Pearson"
> T
Issue traced and merge request opened here:
https://salsa.debian.org/kernel-team/linux/-/merge_requests/743
- Original Message -
> From: "Timothy Pearson"
> To: "990016" <990...@bugs.debian.org>
> Cc: "Salvatore Bonaccorso"
> Sent: Mo
Unfortunately it appears the bug persists. On the latest ppc64el netinst
weekly build there is no trace of the ast module:
~ # modprobe ast
modprobe: FATAL: Module ast not found in directory
/lib/modules/6.1.0-9-powerpc64le
Package: gox
Version: 0.3.0-8+b5
When attempting to build a package with the bundled gox version in Debian, it
doesn't seem to understand linux/ppc64le and silently exits. Installing the
upstream version directly with "go install" yields a working gox binary. Issue
first discovered with the
Merge request to fix this bug here:
https://salsa.debian.org/mozilla-team/thunderbird/-/merge_requests/7
Package: thunderbird
Version: 1:102.9.0-1~deb11u1
Severity: important
Thunderbird uses an incorrect SQLite endianness when built on a little endian
ppc64 (ppc64el) system. As a result, it is unable to initialize and load its
databases, leading to a blank / unusable application.
There is some
- Original Message -
> From: "Andres Salomon"
> To: "Soren Stoutner"
> Cc: 1020...@bugs.debian.org, "Dmitry Shachnev" , "Agustin
> Martin" , "Debian
> Qt/KDE Maintainers" , "Roland
> Rosenfeld" , "Re
Package: libsndfile1
Version: 1.1.0-3
Severity: important
Tags: sid
X-Debbugs-CC: dilin...@queued.net
When attempting to build an application that uses both libsndfile and libFLAC,
the application fails to link with "/usr/bin/ld: warning: libFLAC.so.8, needed
by /lib/-linux-gnu/libsndfile.so.1,
affects chroot
After investigation, this appears to be the same bug as #1021109. Installing
the locales package and adding a default locale prior to upgrading to sid
appears to work around the issue.
- Original Message -
> From: "Debian Bug Tracking System"
> To:
Package: bash
Version: 5.2-1
When attempting to enter a sid chroot on ppc64el, bash causes a SIGSEGV.
Confirmed to be in the bash package by successfully installing a bookworm
chroot, then confirming the crash occurs after upgrading only bash to the
version in sid.
The areas I'm specifically interested in helping with are the security updates
and the ungoogled Chromium patchset -- I'm already maintaining the ungoogled
patches on top of the ppc64el patches, and I'd like to get both of those in
Debian proper.
I'm willing to help out, provided I can also work toward enabling ppc64el
support in the official packages. I'm already on the biweekly pull / build
schedule as a result of needing to patch + rebuild the ppc64el binaries, so am
familiar with the workload involved.
On my Talos II workstation
As a quick update, I'm starting another push (small at first) to try to get
some of the more downstream-invasive changes upstream. Starting with the build
config generators, as downstream distros can only patch in the entire generated
config which gets messy.
- Original Message -
> From: "Andres Salomon"
> To: "1005083" <1005...@bugs.debian.org>
> Cc: "Timothy Pearson"
> Sent: Tuesday, February 15, 2022 4:21:27 PM
> Subject: Re: Chromium 99 update
>
> Hi,
>
> So the good ne
- Original Message -
> From: "Salvatore Bonaccorso"
> To: "Timothy Pearson" , "990279"
> <990...@bugs.debian.org>
> Cc: "Christian König" , "Xi Ruoyao"
> , "Alex Deucher"
>
> Sent: Wednesday, Fe
- Original Message -
> From: "Salvatore Bonaccorso"
> To: "Timothy Pearson" , "990279"
> <990...@bugs.debian.org>
> Cc: "Christian König" , "Xi Ruoyao"
> , "Alex Deucher"
>
> Sent: Tuesday, Fe
I can confirm Bullseye is still affected by this bug.
Is there any chance of this fix being applied to the Bullseye stable kernels?
We're having to maintain kernel builds in our own repositories to fix this
regression, and that introduces some lag when e.g. security updates are pushed
to
. That means ppc64el, arm64,
riscv, sparc64 in practice.
- Original Message -
> From: "Jochen Sprickerhof"
> To: "Timothy Pearson" , 990...@bugs.debian.org
> Sent: Saturday, July 10, 2021 1:09:14 PM
> Subject: Re: Bug#990016: Debian installer images missing ASpeed v
pers. If this is due to lack of hardware access, we may be
able to assist.
- Original Message -
> From: "Timothy Pearson"
> To: "990279" <990...@bugs.debian.org>
> Sent: Saturday, July 24, 2021 11:42:02 PM
> Subject: Also affects stable...
> Just h
Just hit this upgrading a stable box from 4.19.0-6-powerpc64le to
4.19.0-17-powerpc64le. AMD RX480, works perfectly again after downgrading to
the older kernel.
- Original Message -
> From: "Timothy Pearson"
> To: "Steve McIntyre"
> Cc: "990017" <990...@bugs.debian.org>
> Sent: Thursday, June 17, 2021 7:56:27 PM
> Subject: Re: Bug#990017: [REGRESSION] Bullseye CD installer images fail to
>
- Original Message -
> From: "Steve McIntyre"
> To: "Timothy Pearson" , 990...@bugs.debian.org
> Sent: Thursday, June 17, 2021 7:43:15 PM
> Subject: Re: Bug#990017: [REGRESSION] Bullseye CD installer images fail to
> install GRUB on OpenPOWER ma
Package: debian-cd
Severity: Grave
Attempting to use the latest Bullseye RC2 CD installer on an OpenPOWER machine
results in a fatal error during bootloader installation:
Jun 17 21:14:45 grub-installer: info: Installing grub on '/dev/sda1'
Jun 17 21:14:45 grub-installer: info: grub-install does
Package: debian-cd
Severity: Important
The current Bullseye installer images don't include the "ast" video driver,
making it impossible to install Debian on systems that don't have a fallback
VGA BIOS (anything non-x86, e.g. ARM/OpenPOWER).
Ubuntu fixed this issue almost 5 years ago:
Package: debian-cd
Despite having a wide deployment of desktop-class systems, the ppc64le port
does not have a live CD/DVD available. This makes installation needlessly
difficult on desktop machines that may or may not have external BMC or serial
access, given that the installer images also
- Original Message -
> From: "殷啟聰 | Kai-Chung Yan"
> To: 960...@bugs.debian.org, "Timothy Pearson"
> Sent: Sunday, May 17, 2020 11:13:32 PM
> Subject: Re: Update android-platform-external-libunwind and build for ppc64el
>> According to the sour
Package: android-platform-external-libunwind
According to the source trees at both
https://github.com/Unity-Technologies/android-libunwind and
https://android.googlesource.com/platform/external/libunwind/, ppc64[le]
systems are supported. However, the version in Debian, even Sid, explicitly
Package: libpng16-16:ppc64le
Version: 1.6.36-6
libpng is built without VSX support on POWER systems. This breaks assumptions
in other software, such as node optipng (log below).
I can work around it with this, but it is not ideal:
CFLAGS="-DPNG_POWERPC_VSX_OPT=0" npm install optipng-bin
npm
Package: bind9
Version: 1:9.11.5.P4+dfsg-5.1
Latest Debian Buster ppc64le, bind9 keeps crashing every few days with an
assert:
named[412]: ../../../lib/dns/rbtdb.c:10330: INSIST(((void
*)((header)->link.prev) != (void *)(-1))) failed, back trace
named[412]: #0 0x10609cee8 in ??
named[412]: #1
Package: reptyr
Version: 0.6.2-1.2
Upstream reptyr 0.70 now has support for PowerPC. We tested it earlier today
to great success, however the older 0.6.2 version in Debian lacks this support
and we had to build from GIT.
Please update to the latest reptyr release and enable ppc64le builds.
Package: syslinux-utils
Version: 3:6.04~git20190206.bf6db5b4+dfsg1-1
There are two versions of the hybrid ISO utility, a native C implementation and
a Perl version. Neither of these should be in the architecture-dependent
(x86-only) syslinux-utils package, but especially the Perl version of
Package: systemd
Version: 241-5
Ever since the switch to systemd we have been experiencing significant problems
with our diskless nodes, where if the NFS connection is dropped for any reason
(NFS server reboot, network router state reset, etc.) there is a high chance
the diskless nodes will
On 03/06/2019 05:40 AM, Paul Slootman wrote:
> On Wed 06 Mar 2019, Timothy Pearson wrote:
>
>> Package: rsync
>> Version: 0.9.7-10+b1
>
> This is either a really ancient version, or a version not related to
> rsync. Please check this. What does 'rsync --version' sh
Package: rsync
Version: 0.9.7-10+b1
rsync occasionally crashes with a SIGSEGV when run from a cron job and
the remote host closes the connection unexpectedly.
Backtrace:
> Core was generated by `rsync -av user@test-host:/first/path/file
> /second/path/file'.
> Program terminated with signal
Just wanted to see if this was something we could get applied to the
kernel. Right now people are needing to rebuild their kernels locally
to update firmware, or use a more complex out of band update mechanism,
which is not ideal.
zaj...@gmail.com>
To: David Woodhouse ,
Brian Norris ,
Boris Brezillon ,
Marek Vasut , Richard Weinberger
Cc: Benjamin Herrenschmidt , Timothy Pearson
, linux-...@lists.infradead.org,
Michael Ellerman , Miquel Raynal
, =?utf-8?b?UmFmYcWCIE1pxYJlY2tp?=
, Paul Mackerras ,
linuxppc-...@lists.ozl
This was actually a bug in mesa. See
https://bugs.freedesktop.org/show_bug.cgi?id=107012
This is actually a bug in mesa. See
https://bugs.freedesktop.org/show_bug.cgi?id=107012 .
Package: mesa-utils
Version: 8.4.0-1
glxgears crashes immediately on startup using the amdgpu drivers. Other
3D applications, including complex ones like rss-glx, openarena, and
supertux, work fine.
Package xserver-xorg-video-amdgpu is version 18.0.1-1. Kernel is 4.18
staging.
Backtrace:
>
Package: alien-arena
Version: 7.66+dfsg-5
When alien-arena loads a map, right before the gameplay starts, we get a
SIGSEGV and the game quits:
> Thread 1 "alienarena" received signal SIGSEGV, Segmentation fault.
> 0x71ba5ec4 in llvm::Value::getContext() const () from
>
Package: coreutils
Version: 8.26-3
"uname -p" returns "unknown" on ppc64el. It should return a valid string,
such as ppc64le.
Package: linux-image-powerpc64le
Version:
For an unknown (historical?) reason, the AMDGPU kernel module is not built
on ppcl64el systems. I have rebuilt the kernel and verified that it does
work on POWER9 with Linux 4.14 and an AMD Vega 64 card, so the module
build should be enabled for Debian.
Package: linux-image-4.6.0-1-powerpc64le
Version: 4.6.4-1
Severity: important
The PowerNV MTD module is required for checkstop diagnostics on POWER8 and
higher machines, but it is not enabled in the kernel build.
Please enable it as a module with the following kernel .config lines:
CONFIG_MTD=m
Package: linux-image-4.6.0-1-powerpc64le
Version: 4.6.4-1
Severity: important
The opal-prd module is required for checkstop diagnostics on POWER8 and
higher machines, but it is not enabled in the kernel build.
Please enable it as a module with the following kernel .config line:
CONFIG_OPAL_PRD=m
Package: libpam-pkcs11
Version: 0.6.8-4
Severity: important
This package is out of date by at least a year; since the last udate there
have been numerous fixes that allow this module to integrate properly into
modern card based environments.
Please update this package to the latest upstream GIT
Good catch! The version was wrong, and in fact was pulled four days
before my commit that fixed this bug last year. Apologies for the noise,
I had no idea the version in Stretch was that old.
Package: libhx509-5-heimdal
Version: 1.7~git20160703+dfsg-1
Severity: grave
Heimdal's X509 PAM authentication plugin crashes with the following
backtrace when attempting to use a cryptographic card to log in:
Program received signal SIGSEGV, Segmentation fault.
0x7f5c0f30e38d in
For now, since klibc does not have NFSV4 support, the attached file works
around the problem (place in /usr/share/initramfs-tools/hooks/nfsv4).
nfsv4
Description: Binary data
What is the status of this bug report? NFSv3 is becoming obsolete, and
this bug report is over *9 years* old now!
> As I understand your first mail in this report, this requires a fix
> somewhere first? (I wouldn't want to apply a "temporary" patch in
> Debian.)
Unfortunately it doesn't look like upstream is going to patch their code.
There is a protracted "blame game" going on between mesa and gcc:
Is there any chance this bug report could be handled soon? We keep having
to recompile mesa to enable OpenCL for our POWER machines.
Thanks!
> On 2016-07-21 19:46:17, Timothy Pearson wrote:
> Would it be also possible to simple switch powerpc to any-powerpc?
I see no obvious reason for that not to work. I only have access to
POWER8+ machines for testing, however AFAIK the AltiVec parts of the
codebase should function correctly
Package: handbrake
Version: 0.10.5+ds1-2
Severity: important
Handbrake should be built for ppc64el. It has been tested on POWER8 and
works fine. The only required change is to add ppc64el to the
debian/control file.
Package: nvidia-texture-tools
Version: 2.0.8-1+dfsg-8
Tags: patch
nvidia-texture-tools is not built for ppc64el due to a failure to build
from source on that platform. This patch fixes the build failure:
https://github.com/castano/nvidia-texture-tools/pull/238
Please apply the patch to the
Apologies; one of the options given above was incorrect.
The following options need to be set on ppc64el to enable PCI VFIO
functionality:
CONFIG_SPAPR_TCE_IOMMU=y
CONFIG_VFIO=m
CONFIG_VFIO_PCI=m
Package: linux-image-4.5.0-2-powerpc64le
Version: 4.5.5-1
Severity: important
VFIO PCI passthrough is not available on POWER8 systems due to several
missing kernel configuration options. PCI passthrough is an important
feature of POWER8 machines, works well under at least Linux 4.5, and
should
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA224
Package: linux-image-4.3.0-1-powerpc64le
Version: 4.3.5-1
When upgrading to the latest 4.3 testing kernel on a Mellanox MT25208
controller, all IPoIB links showed a severe performance regression as
compared with kernel 3.16. The bandwidth of each
A similar crash also occurs when outputs are hotplugged. Revised patch
attached.
Note that at least on my test machine with an HDMI display attached Xorg
starts but the display remains disabled. It is currently unknown whether
this is a general bug in the radeon driver or something pcc64el
The attached patch fixes the crash and Xorg appears to start normally.--- xorg-server-1.18.0/randr/randr.c 2015-11-06 08:07:41.0 -0600
+++ xorg-server-1.18.0/randr/randr.c 2016-02-15 12:49:06.469645421 -0600
@@ -559,6 +559,9 @@
mastersp = pScrPriv;
}
+if (!mastersp)
+
Package: xserver-xorg-core
Version: 2:1.18.0-3
On a POWER8 machine under both Debian Jessie and Debian Stretch Xorg
crashes on startup when using the radeon driver and a Radeon R9 290X.
Kernel modesetting outside of X appears to function normally.
crash:
Program received signal SIGSEGV,
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA224
Package: linux-image-4.3.0-1-powerpc64
Version: 4.3.3-7
Severity: normal
The POWER8 processors use VMX to accelerate cryptographic functions.
Linux supports the use of VMX for cryptographic acceleration, however
Debian does not build this module,
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA224
> On Wed, Jan 13, 2016 at 16:27:02 -0600, Timothy Pearson wrote:
>
>> Package: mesa
>> Version: 11.0.8-1
>> Severity: normal
>> Tags: +patch
>>
>> The mesa package currently relies on an autoge
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA224
I'm still seeing this issue with Debian Jessie on POWER8 hardware, trying
to boot the FreeBSD 10.2 ISO image. The error is almost identical, with a
slight change to the kernel address:
Booting [/boot/kernel/kernel]...
Kernel entry at 0x101f00 ...
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA224
Package: mesa
Version: 11.0.8-1
Severity: normal
Tags: +patch
The mesa package currently relies on an autogenerated debug symbol package
for osmesa. The attached patch switches this to a control file-defined
debug symbol package.
-BEGIN PGP
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA224
Package: mesa
Version: 11.0.8-1
Severity: important
OpenCL was disabled on ppc64el due to an altivec-related build failure.
The attached patch fixes the build failure and reenable OpenCL on ppc64el.
Note that an upstream bug report was filed
I went ahead and filed a bug report upstream at:
https://sourceware.org/bugzilla/show_bug.cgi?id=19421
Hopefully this helps prod some action on this serious bug...
P,S. I completely agree with the previous commenter; this bug is unreal.
This is 2015, not 1998...core libraries breaking modern
Package: linux-image-4.2.0-1-powerpc64le
Version: 4.2.6-3
Severity: normal
Many server- and workstation-class boards use the ASpeed Technologies BMC
device as the primary VGA output. Linux supports initialization of this
device via the AST DRM module, however Debian does not build this module,
Package: linux-image-4.2.0-1-powerpc64le
Version: 4.2.6-3
Severity: grave
Using the latest debian-installer netboot kernel to install on an
OpenPOWER server; install proceeds normally until disk detection when it
fails stating no disks can be found. dmesg shows a slew of errors similar
to the
An additional patch is required to correct a second crash when using
syncrepl and slapi plugins. This crash has been reported in upstream
ITS#7641
(http://www.openldap.org/its/index.cgi/Incoming?id=7641;selectid=7641);
the requisite patch is available in that report.
--
To UNSUBSCRIBE, email
Upstream has generated a corrected patch that has been verified as
working. This patch should be applied to the Debian OpenLDAP package
instead.
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Package: openldap
Version: 2.4.31-1+nmu2
OpenLDAP crashes when syncrepl has been enabled and plugins are in use.
Full details (along with a patch to fix this problem) are available in
upstream Bug 7636
(http://www.openldap.org/its/index.cgi/Incoming?id=7636).
The patch in that report is
Package: slapd
Version: 2.4.31-1+nmu2
OpenLDAP crashes when syncrepl has been enabled and plugins are in use.
Full details (along with a patch to fix this problem) are available in
upstream Bug 7636
(http://www.openldap.org/its/index.cgi/Incoming?id=7636).
The patch in that report is applicable
Control: tags -moreinfo +fixed
Initial tests of cups 1.6.2 indicate that this bug is fixed, and that
Kerberos authentication now works properly. I will reopen this report if
any glitches are noted during testing over the next few weeks.
Thanks!
--
To UNSUBSCRIBE, email to
Control: tags -1 +moreinfo
Hi Timothy, and thanks for your bugreport,
Can you confirm that this still applies to CUPS 1.6.2 (as uploaded to
unstable)?
Thanks in advance, cheers,
OdyX
I will see what I can do in the next couple of weeks. The machine in
question is slated for deployment
Package: cups
Version: 1.5.3-2.16
When using the Web interface to CUPS in a Kerberized environment, CUPS
fails to pass Kerberos authentication information to the local IPP socket.
This manifests as access being denied to any pages which access printer
information, even though the user logged in
The attached patch resolves the issue.--- ruby1.9.1-1.9.3.194/debian/rules
+++ ruby1.9.1-1.9.3.194/debian/rules
@@ -206,7 +206,8 @@
dh_movefiles -p$(cdbs_curpkg) \
usr/lib/libruby-$(ruby_ver).so \
usr/lib/libruby-$(ruby_ver)-static.a \
- $(ruby_libdir)/mkmf.rb
+
Package: ruby1.9.1-dev
Version: 1.9.3.194-1
The ruby.pc (or ruby-1.9.pc, depending on exactly which Ruby sources you
look at) pkg-config information file that comes as part of the upstream
Ruby sources is not part of the ruby1.9.1-dev package. This makes it
difficult for third party build
91 matches
Mail list logo
