On 2024-05-05 15:02, Cordell Bloor wrote:
There is nothing that can be done in rocm-hipamd to address this bug,
aside from removing arm64 from the rocm-hipamd architecture list. The
incompatibility is not with the HIP runtime, but with the HIP
language. This is a disagreement that glibc
to upstream than the
previously rejected patch. If not, it's up to glibc or LLVM to find a
solution. If they cannot, then we will have to drop arm64 support for
the HIP language.
Sincerely,
Cory Bloor
[3]: https://gcc.gnu.org/wiki/Offloading
From: Cordell Bloor
Date: Wed, 10 Apr 2024 16:49:24 -0600
Hi Timo,
On Sat, 2 Mar 2024 09:21:43 +0100 Timo =?utf-8?Q?R=C3=B6hling?=
wrote:
>
> On Sun, 25 Feb 2024 20:28:53 +0100 Lucas Nussbaum
> wrote:
> > > /usr/include/thrust/detail/type_traits.h:736:1: error: expected
> > > type-specifier before ‘template’
>
> This bug is caused by a #ifdef
Control: severity 1067956 important
The rocalution package has never successfully built for armhf, so I
don't think this qualifies as release-critical.
It's great to see that the rocalution package gets all the way into the
tests before failing, though. The upstream project only officially
Ah. That makes sense. Thanks, Christian!
On 2024-04-04 04:30, Christian Kastner wrote:
I just rebuilt rocfft to 6.0.2 but the issue is still present. But that
was naive, there are other < 6.0 components in the stack that could
affect this.
The problem appeared in rocfft 5.5.1 when rocm-hipamd
I tried to reproduce the rocfft callback bug with a W6800 (gfx1030). I
used a Debian Unstable docker container on an Ubuntu Noble host, but the
tests all passed. This made me realize that the test failure pattern on
the CI is that all the qemu-based workers are failing and all the
podman-based
Package: librocfft0
Version: 5.7.1-1
Severity: normal
X-Debbugs-Cc: c...@slerp.xyz
Dear Maintainer,
The rocfft callback tests were passing with rocfft 5.5.1 on hip 5.2.3,
but began failing when hip was updated to 5.7.1. These failures are
specific to gfx900 and gfx1030 to gfx1036. The failures
Control: reassign 1067356 libamdhip64-dev 5.7.1-1
Control: affects 1067356 hipsolver
Control: fixed 1067356 5.7.1-2
On 2024-03-20 15:00, Lucas Nussbaum wrote:
During a rebuild of all packages in sid, your package failed to build
on amd64.
Relevant part (hopefully):
make[1]: Entering
On Mon, 04 Mar 2024 04:35:50 + Cordell Bloor wrote:
> Many tests began failing for the gfx1035 ISA on the Debian ROCm CI upon
> the update to libhsa-runtime64-1 (5.7.1-1). The failure is an assertion:
>
> ./src/image/addrlib/src/gfx10/gfx10addrlib.cpp:1083: virtual
rocr::Addr
Hi Petter,
On 2024-03-15 02:39, Petter Reinholdtsen wrote:
[Cordell Bloor]
Thanks Petter. After inspecting the code and reviewing both your report
and the buildd logs, my conclusion is that this issue was fixed by
upstream and included in 5.7.1-1.
But your original report claimed
On Sun, 10 Mar 2024 06:56:56 +0100 Petter Reinholdtsen
wrote:
>
> In my sid chroot, on a laptop with no AMD GPU, I get this:
>
> root@minerva:/# rocminfo
> ROCk module is NOT loaded, possibly no GPU devices
> root@minerva:/# rocm_agent_enumerator
> gfx000
> root@minerva:/#
Thanks Petter. After
On Sat, 09 Mar 2024 22:57:24 +0100 Petter Reinholdtsen
wrote:
>
> Picking the correct endian test is quite complex, if we should believe
> https://stackoverflow.com/questions/8978935/detecting-endianness >.
>
> Note, there are middle endian architetures in addition to big and little
> endian
Package: rocminfo
Version: 5.7.1-1
Severity: normal
X-Debbugs-Cc: c...@slerp.xyz
Dear Maintainer,
On systems, the rocm_agent_enumerator command may crash with an error:
Traceback (most recent call last):
File "/usr/bin/rocm_agent_enumerator", line 260, in
main()
File
Package: libhsa-runtime-dev
Version: 5.7.1-1
Severity: normal
X-Debbugs-Cc: c...@slerp.xyz
Dear Maintainer,
The endianness detection logic in hsa/hsa.h fails on arm64 and ppc64el,
which leads to build failures in rccl on those platforms [1].
The current logic is:
// Try to detect CPU
Package: dpkg-dev
Version: 1.22.5
Severity: wishlist
X-Debbugs-Cc: c...@slerp.xyz, debian...@lists.debian.org
Dear Maintainer,
When packaging the AMD ROCm GPU libraries for Debian, we are currently
using CXX=hipcc or CXX=clang++ to build libraries written in HIP as if
they were written in C++.
Package: libhsa-runtime64-1
Version: 5.7.1-1
Severity: normal
X-Debbugs-Cc: c...@slerp.xyz
Dear Maintainer,
Many tests began failing for the gfx1035 ISA on the Debian ROCm CI upon
the update to libhsa-runtime64-1 (5.7.1-1). The failure is an assertion:
Package: rocminfo
Version: 5.7.1-1
Severity: normal
X-Debbugs-Cc: c...@slerp.xyz
Dear Maintainer,
There are warnings emitted on stderr when running rocm_agent_enumerator:
# rocm_agent_enumerator
/usr/bin/rocm_agent_enumerator:152: SyntaxWarning: invalid escape sequence '\A'
line_search_term =
Package: clang-17
Version: 1:17.0.6-5
Severity: minor
X-Debbugs-Cc: c...@slerp.xyz, debian...@lists.debian.org
Dear Maintainer,
I noticed a build failure in rocmprim on ppc64el that seems to have been
introduced between clang-15 and clang-17. I'm not certain that this is a
compiler issue,
This segfault does seem to be caused by mixing clang-15 and clang-17 in
the HIP RTC codepath. When libamdhip64 from ROCm 5.6.1 (built with the
same clang-17 as rocm-compilersupport 6.0+git20231212.4510c28+dfsg-1) is
used, the segfault disappeared [1].
Sincerely,
Cory Bloor
[1]:
Hi Christian,
On 2024-02-26 00:54, Christian Kastner wrote:
On gfx1031/gfx1032/gfx1034, there are numerous occurrences of
HIPSPARSE_STATUS_INTERNAL_ERROR, see [1] for a full log. Interestingly,
only some of them lead to test failures (some examples below), and
sometimes there is more than one
The segfault on the very first rocfft test (and in no other library) is
probably a good indication that HIP runtime compilation (RTC) is broken.
That is a feature that is used for every rocFFT function, but not used
by any other ROCm library.
Whether this bug is because the HIP stack is
Package: libhipsparse0-tests
Version: 5.5.1-1
Severity: normal
X-Debbugs-Cc: c...@slerp.xyz
Dear Maintainer,
The update of rocsparse 5.5.1 to 5.7.1 seems to have caused a regression
in hipsparse. Although, it's also possible that this problem was because
rocsparse was therefore rebuilt with the
Package: libamd-comgr2
Version: 6.0+git20231212.4510c28+dfsg-1~exp2
Severity: important
X-Debbugs-Cc: c...@slerp.xyz
Dear Maintainer,
The rocfft tests began segfaulting on all architectures when
rocm-compilersupport 6.0+git20231212.4510c28+dfsg-1~exp2 was uploaded to
unstable. You can see it
Package: wnpp
Severity: wishlist
Owner: Cordell Bloor
X-Debbugs-Cc: debian-de...@lists.debian.org, c...@slerp.xyz,
debian...@lists.debian.org
* Package name: rocm-tensile
Version : 6.0.2
* URL : https://github.com/ROCm/Tensile
* License : Expat
Programming
Package: wnpp
Severity: wishlist
X-Debbugs-Cc: c...@slerp.xyz, debian...@lists.debian.org
* Package name: hipblaslt
Version : 6.0.2
* URL : https://github.com/ROCm/hipBLASLt
* License : Expat
Programming Lang: C, C++, HIP
Description : portable interface
Package: wnpp
Severity: wishlist
X-Debbugs-Cc: c...@slerp.xyz, debian...@lists.debian.org
* Package name: cxxheaderparser
Version : 1.3.1
Upstream Contact: RobotPy Development Team
* URL : https://github.com/robotpy/cxxheaderparser
* License : BSD-3-Clause
Package: clang-17
Version: 1:17.0.6-5
Severity: wishlist
X-Debbugs-Cc: c...@slerp.xyz, debian...@lists.debian.org
Dear Maintainer,
LLVM upstream added support for HIP in /usr, but it incorrectly reports
HIP as being installed in /usr/local [1]. I have not verified whether
this bug affects Debian
Package: wnpp
Severity: wishlist
Owner: Cordell Bloor
X-Debbugs-Cc: debian-de...@lists.debian.org, debian...@lists.debian.org,
c...@slerp.xyz
* Package name: hipify
Version : 6.0.2
* URL : https://github.com/ROCm/HIPIFY
* License : Expat
Programming Lang: C
Hi Etienne,
Could we add 'Recommends: liboam-dev' to librocm-smi-dev until the CMake
config in the latter is modified to make liboam-dev optional? There has
been no progress on this issue for some time, so I think it may be worth
applying that mitigation.
I wouldn't consider the addition of
found 1059607 6.5.13-1
found 1059607 6.6.9-1
thanks
On 2024-01-06 06:24, Salvatore Bonaccorso wrote:
And given this commit was backported to 4.19.298, 5.10.200, 6.1.62,
6.5.11 and 6.6.1. Can you confirm the problem is as well present in
the current available 6.5.13-1 and 6.6.9-1?
I can
I was discussing this bug with Christian Kastner the other day. He
suggested:
On 2024-01-02 14:55, Christian Kastner wrote:
The hint in the offending comment regarding the dwc3 driver made me
curious. It seems that it is not enabled in the Debian kernel, so this
is probably something
b25a2f247083389e44a1c7ffb9375c7cf20fabb5 is the first bad commit
commit b25a2f247083389e44a1c7ffb9375c7cf20fabb5
Author: Vicki Pfau
Date: Wed Sep 27 13:22:12 2023 -0700
PCI: Prevent xHCI driver from claiming AMD VanGogh USB3 DRD device
commit 7e6f3b6d2c352b5fde37ce3fed83bdf6172eebd4
found 1059607 6.1.66-1
not-found 1059607 6.1.55-1
not-found 1059607 6.1.52-1
not-found 1059607 6.1.38-1
thanks
On 2023-12-29 06:22, Diederik de Haas wrote:
On Friday, 29 December 2023 13:50:34 CET Salvatore Bonaccorso wrote:
I realize this is asking much, but do you think you would be able to
Package: src:linux
Version: 6.1.67-1
Severity: important
X-Debbugs-Cc: c...@slerp.xyz
Dear Maintainer,
When Debian 12 is installed on the Steam Deck, the process goes smoothly until
the system is rebooted. When the system reboots, it switches to the 6.1.0-16
kernel package. At this point, power
The test failures are reproducible on my Radeon VII (gfx906) workstation
when the failing seed is specified. So, the problem is not specific to
gfx1032:
$ ROCFFT_LAYER=1 /usr/libexec/rocm/librocfft0-tests/rocfft-test
Package: libhsa-runtime-dev
Version: 5.2.3-6
Severity: normal
Tags: fixed-upstream
X-Debbugs-Cc: c...@slerp.xyz, jonathanchesterfi...@gmail.com
Dear Maintainer,
As reported by Jon Chesterfield [1],
> HSA 1.2 introduces an API call hsa_amd_memory_async_copy_on_engine
>
>
Package: librocfft0-tests
Version: 5.5.0-6
Severity: normal
Dear Maintainer,
The rocfft tests passed then failed on amd64+gfx1032 with an identical set of
dependencies. The failing log contained:
55s Random seed: 190206186
<...>
14657s [ RUN ]
cessfully.
* The installation package was faulty and contained
"/usr/lib/llvm-17/lib/cmake/clang/ClangTargets.cmake"
but not all the files it references.
```
The error can be reproduced with this minimal CMakeLists.txt:
```
cmake_minimum_required(VERSION 3.22)
project(exampl
Hi Christian,
I think the path forward to closing this bug is to patch rocr-runtime
and remove this message. My conclusion from this investigation is that
the message is merely informational. It's normal for HSA_AMD_SVM to be
disabled and for HMM features to therefore be unavailable.
We
Hi Christian,
On 2023-11-24 03:26, Christian Kastner wrote:
On 2023-11-23 08:35, Cordell Bloor wrote:
On 2023-11-22 03:19, Christian Kastner wrote:
The Linux kernel on Debian is built without HSA_AMD_SVM enabled. That is
the KConfig for "Enable HMM-based shared virtual memory manager&qu
Hi Christian,
On 2023-11-22 03:19, Christian Kastner wrote:
The Linux kernel on Debian is built without HSA_AMD_SVM enabled. That is
the KConfig for "Enable HMM-based shared virtual memory manager", which
is required for xnack+ operation. The xnack feature allows some AMD GPUs
to retry memory
I've included some additional information on the ticket and have been
discussing this with the upstream developers. I'll summarize the
information here.
On 2023-11-18 00:39, Cordell Bloor wrote:
Each time a HIP application is executed, the rocr-runtime prints the message:
KFD does
This message is not observed when running Debian Sid in a docker
container on the Ubuntu 20.04 kernel (5.15.0) with the AMD-provided
amdgpu-dkms from ROCm 5.6.0.
Sincerely,
Cory Bloor
I have badly misdiagnosed this problem.
On 2023-11-18 00:52, Cordell Bloor wrote:
The rocblas-test executable sets a five-second alarm signal before it
executes some tests. If the alarm goes off before the test completes,
rocblas-test will abort, under the assumption that there was deadlock
Package: librocprim-tests
Version: 5.5.1-2
Severity: normal
X-Debbugs-Cc: c...@slerp.xyz
Dear Maintainer,
RocprimDeviceRadixSort.SortKeysOver4G fails when executed on GPUs with
8 GB of memory or less [1]. To ensure that tests are useful on a wide
variety of hardware, the appropriate behaviour
Package: librocsparse0-tests
Version: 5.5.1-2
Severity: minor
X-Debbugs-Cc: c...@slerp.xyz
Dear Maintainer,
There is a test failing with gfx1030 on the CI:
[ RUN ]
quick/bsrmv.level2/f32_c_500_842_1_1_n1_n0p5_row_2_NT_0b_rand_t2
clients/common/rocsparse_check.cpp:287: Failure
Package: librocprim-tests
Version: 5.5.1-2
Severity: important
X-Debbugs-Cc: c...@slerp.xyz
Dear Maintainer,
In ROCm 5.5, the rocprim library added functionality that branched on
the detected GPU hardware to select the appropriate implementation of
the library functionality. This does not
Package: librocblas0-tests
Version: 5.5.1+dfsg-3
Severity: important
X-Debbugs-Cc: c...@slerp.xyz
Dear Maintainer,
The rocblas-test executable sets a five-second alarm signal before it
executes some tests. If the alarm goes off before the test completes,
rocblas-test will abort, under the
Package: libhsa-runtime64-1
Version: 5.2.3-5
Severity: normal
X-Debbugs-Cc: c...@slerp.xyz
Dear Maintainer,
Each time a HIP application is executed, the rocr-runtime prints the message:
KFD does not support xnack mode query.
ROCr must assume xnack is disabled.
It is unclear to me
HI BogDan,
On 2023-11-14 09:47, BogDan Vatra wrote:
ROCm debian packages are very old, any [chance] to upgrade it to 5.7.1?
I think the main difficulty with rocm-device-libs in particular is that
the sources are closely coupled with LLVM and the binary package is only
guaranteed to be
Hello Maintainer,
Would you be open to a patch or merge request that enabled HIP support
to the blender package?
HIP has been available in Debian for over a year now and the packaging
is a bit more mature. It is now possible to build for HIP just by adding
hipcc to the build dependencies
Thanks Lucas,
On 2023-07-25 14:56, Lucas Nussbaum wrote:
# Writing Kernels...
Generating kernels: Launching 8 threads...
Traceback (most recent call last):
File "/<>/tensile/Tensile/Parallel.py", line 54, in
apply_print_exception
return func(*args)
^^^
File
Package: libamdhip64-dev
Version: 5.2.3-10
Severity: normal
X-Debbugs-Cc: c...@slerp.xyz
Dear Maintainer,
The rocprim-tests package FTBFS on ppc64el in part due to a failure to
link compiler_rt for half-float operations [1]. This results in the
errors:
/usr/bin/ld:
Package: clang-15
Version: 1:15.0.7-6
Severity: normal
X-Debbugs-Cc: c...@slerp.xyz, debian...@lists.debian.org
Dear Maintainer,
On ppc64el, the clang compiler crashed on buildd [1] during the
compilation of test_block_sort.cpp in the rocprim package. We're only
compiling HIP programs for
Package: wnpp
Severity: wishlist
Owner: Cordell Bloor
X-Debbugs-Cc: debian-de...@lists.debian.org, c...@slerp.xyz,
debian...@lists.debian.org
* Package name: rocm-validation-suite
Version : 5.6.0
* URL : https://github.com/ROCm-Developer-Tools/ROCmValidationSuite
Package: wnpp
Severity: wishlist
Owner: Cordell Bloor
X-Debbugs-Cc: debian-de...@lists.debian.org, c...@slerp.xyz,
debian...@lists.debian.org
* Package name: rocdbgapi
Version : 5.6.0
* URL : https://github.com/ROCm-Developer-Tools/ROCdbgapi
* License : Expat
Package: ftp.debian.org
Severity: normal
User: ftp.debian@packages.debian.org
Usertags: remove
X-Debbugs-Cc: rocm-hip...@packages.debian.org
Control: affects -1 + src:rocm-hipamd
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA512
Dear FTP team,
Please remove the ppc64 and s390x binaries of
Package: clang-15
Version: 1:15.0.7-4
Severity: wishlist
X-Debbugs-Cc: c...@slerp.xyz, debian...@lists.debian.org
Dear Maintainer,
I would like to update HIP to use clang-16, but that would require
building the rocm-device-libs with clang-16. Once I do that, they may no
longer be compatible with
Package: wnpp
Severity: wishlist
Owner: Cordell Bloor
X-Debbugs-Cc: debian-de...@lists.debian.org, c...@slerp.xyz,
debian...@lists.debian.org
* Package name: miopen
Version : 5.5.0
Upstream Author : Advanced Micro Devices, Inc.
* URL : https://github.com
Package: clang-15
Version: 1:15.0.7-4
Severity: wishlist
X-Debbugs-Cc: c...@slerp.xyz, debian...@lists.debian.org
Dear Maintainer,
The upstream LLVM project checks /etc/os-release to determine if it
should search for HIP in /usr and /usr/local. This is unfortunate
because it introduces
tag 1031799 patch
thanks
Brad King has mostly fixed this problem upstream [1]. I've opened a pull
request [2] that backports his patch series and adds another small fix
required to close this bug.
Sincerely,
Cory Bloor
[1]: https://gitlab.kitware.com/cmake/cmake/-/merge_requests/8525
[2]:
Package: libamdhip64-dev
Version: 5.2.3-8
Severity: normal
X-Debbugs-Cc: c...@slerp.xyz
Dear Maintainer,
The libamdhip64-dev package requires the CXX language to be enabled in
order for find_package(hip) to succeed. This is because there is an
unquoted check against the CXX compiler within
Package: libamdhip64-dev
Version: 5.2.3-8
Severity: normal
X-Debbugs-Cc: c...@slerp.xyz
Dear Maintainer,
The libamdhip64-dev package requires hipcc in order for
find_package(hip) to succeed. This is because within hip-config.cmake,
the variable hip_HIPCC_EXECUTABLE is set with the path to hipcc
Package: libhipsparse-dev
Version: 5.3.3+dfsg-1+b1
Severity: normal
X-Debbugs-Cc: c...@slerp.xyz
Dear Maintainer,
The libhipsparse-dev package is missing a dependency on libamdhip64-dev.
This can be illustrated with this sample:
apt-get -y update
apt-get -y install build-essential cmake
Package: libhiprand-dev
Version: 5.3.3-3
Severity: normal
X-Debbugs-Cc: c...@slerp.xyz
Dear Maintainer,
The libhiprand-dev package is missing a dependency on libamdhip64-dev
and librocrand-dev. This can be illustrated with this sample:
apt-get -y update
apt-get -y install
Package: librocrand-dev
Version: 5.3.3-1
Severity: normal
X-Debbugs-Cc: c...@slerp.xyz
Dear Maintainer,
The librocrand-dev package is missing a dependency on libamdhip64-dev.
# What was done?
In a fresh Bookworm container,
apt-get -y update
apt-get -y install build-essential cmake
Package: clang-15
Version: 1:15.0.7-4
Severity: normal
X-Debbugs-Cc: c...@slerp.xyz, debian...@lists.debian.org
Dear Maintainer,
I'm having trouble reproducing this problem, but I figured I should
report it anyway in case other users had seen similar behaviour.
While compiling the rocthrust
Package: hipcc
Version: 5.2.3-6
Severity: normal
X-Debbugs-Cc: c...@slerp.xyz, debian...@lists.debian.org
Dear Maintainer,
I installed the hipcc package and ran
hipconfig --check
and saw a long string of output that ended with
Check system installation:
check hipconfig in PATH...
Package: clang-16
Version: 1:16.0.0-1~exp5
Severity: normal
X-Debbugs-Cc: c...@slerp.xyz, debian...@lists.debian.org
Dear Maintainer,
Thank you for packaging amdgpu-arch as part of clang-tools-16! I see
that it can correctly detect the architecture of my installed GPU when
libhsa-runtime-dev is
I have been discussing this issue with the upstream developers and have
opened a merge request with a fix [1].
On 2/22/23 15:34, Cordell Bloor wrote:
It is not possible to use CMake's HIP language support together with
the Debian package for HIP. Consider this sample project:
CMakeLists.txt
I went to submit my patch for this bug upstream and found that a fix is
already scheduled for ROCm 5.5. The upstream patch is not publicly
available yet, but I was happy to at least confirm that upstream
considers this behaviour to be a bug.
There is a related issue once finding hip-lang-config.cmake is handled. The
next problem is this:
CMake Error at
/usr/share/cmake-3.25/Modules/CMakeFindDependencyMacro.cmake:47
(find_package):
By not providing "Findamd_comgr.cmake" in CMAKE_MODULE_PATH this
project
has asked
Package: clang-15
Version: 1:15.0.7-1
Severity: normal
X-Debbugs-Cc: c...@slerp.xyz, debian...@lists.debian.org
Dear Maintainer,
Clang does not search the system paths when looking for the ROCm
components that it requires for building HIP langauge code. Consider the
following sample program:
Package: cmake-data
Version: 3.25.1-1
Severity: normal
X-Debbugs-Cc: c...@slerp.xyz, debian...@lists.debian.org
Dear Maintainer,
It is not possible to use CMake's HIP language support together with
the Debian package for HIP. Consider this sample project:
CMakeLists.txt:
Package: wnpp
Severity: wishlist
Owner: Cordell Bloor
X-Debbugs-Cc: debian-de...@lists.debian.org, c...@slerp.xyz,
debian...@lists.debian.org
* Package name: hipfort
Version : 5.4.3
Upstream Author : Advanced Micro Devices, Inc.
* URL : https://github.com
I've prepared the rocr-runtime package for upload. I tested it with
several of the math libraries on Debian Bookworm and Ubuntu Lunar. This
patch has also gone through significant testing upstream.
I think this is ready. Would a team upload be appropriate for this sort
of change?
Sincerely,
Package: libamdhip64-dev
Version: 5.2.3-4
Severity: normal
X-Debbugs-Cc: c...@slerp.xyz
Dear Maintainer,
The CMake support for the HIP language is badly broken. Consider a
sample project:
CMakeLists.txt
CMakeLists.txt
cmake_minimum_required(VERSION 3.22)
project(example LANGUAGES
Package: libamdhip64-dev
Version: 5.2.3-4
Severity: normal
X-Debbugs-Cc: c...@slerp.xyz
Dear Maintainer,
The dependencies for libhipamd64-dev appear to be underspecified.
main.cpp:
#include
#include
#define CHECK_HIP(expr) do { \
hipError_t result =
Package: rocminfo
Version: 5.2.3-2
Severity: normal
X-Debbugs-Cc: c...@slerp.xyz
Dear Maintainer,
When a user invokes rocm_agent_enumerator but does not have sufficient
permissions to read /sys/class/kfd/kfd/topology/nodes/*/properties,
the program will crash with the error
Traceback (most
Package: libamdhip64-dev
Version: 5.2.3-4
Severity: normal
X-Debbugs-Cc: c...@slerp.xyz
Dear Maintainer,
This package provides a HIP version file so that Clang can detect the
version of HIP that is available. That file is installed at:
/usr/share/hip/version/.hipVersion
but the path that
On 2023-02-13 17:22, Santiago Vila wrote:
[ 8%] Linking CXX shared library libhipsparse.so
cd /<>/obj-x86_64-linux-gnu/library && /usr/bin/cmake -E
cmake_link_script CMakeFiles/hipsparse.dir/link.txt --verbose=1
/usr/bin/c++ -fPIC -g -O2 -ffile-prefix-map=/<>=.
-fstack-protector-strong
On 1/18/23 15:03, Étienne Mollier wrote:
2. The symbol tracking needs to be reviewed by somebody more experienced
than me. I think that anything in the rocrand::detail or
rocrand_host::detail namespace should be marked optional, as those symbols
are not intended for use by library users.
Hi Étienne,
I've updated the rocrand package sources on Salsa to rocrand 5.3.3 and
transformed it into a MUT package. I've confirmed that the resulting
library works correctly using it to configure rocfft 5.4.2 (which was
how I discovered this bug originally).
rocrand 5.3.3-1 just needs
Thanks for taking a look at this, Étienne.
On 1/15/23 11:16, Étienne Mollier wrote:
Not sure it will help, but I suppose I could first bump the
rocrand package up to 5.2.3 to be consistent with the build
chain. It should still be easy to do while we're not in the
next stage of the freeze on
Package: librocrand-dev
Version: 5.0.0-2
Severity: normal
X-Debbugs-Cc: c...@slerp.xyz
Dear Maintainer,
The CMake integration for the librocrand-dev package appears to be
broken. A call to find_package(rocrand) will fail with an error like:
CMake Error at
Package: rocminfo
Version: 5.2.3-1
Severity: normal
X-Debbugs-Cc: c...@slerp.xyz
Dear Maintainer,
After installing rocminfo on a fresh new Debian image, I ran the command
and saw the error:
sh: 1: lsmod: not found
ROCk module is NOT loaded, possibly no GPU devices
After running `apt
Package: libhsa-runtime64-1
Version: 5.2.3-1
Severity: minor
X-Debbugs-Cc: c...@slerp.xyz
Dear Maintainer,
During the initialization of libhsa-runtime64, the library will attempt
to load libhsa-amd-aqlprofile64.so. The aqlprofile library is an
optional proprietary extension for profiling and is
Package: wnpp
Severity: wishlist
Owner: Cordell Bloor
X-Debbugs-Cc: debian-de...@lists.debian.org, c...@slerp.xyz,
debian...@lists.debian.org
* Package name: rocalution
Version : 5.3.0
Upstream Author : Advanced Micro Devices, Inc.
* URL : https://github.com
Package: wnpp
Severity: wishlist
Owner: Cordell Bloor
X-Debbugs-Cc: debian-de...@lists.debian.org, c...@slerp.xyz,
debian...@lists.debian.org
* Package name: hipsolver
Version : 5.3.0
Upstream Author : Advanced Micro Devices, Inc.
* URL : https://github.com
Package: wnpp
Severity: wishlist
Owner: Cordell Bloor
X-Debbugs-Cc: debian-de...@lists.debian.org, c...@slerp.xyz,
debian...@lists.debian.org
* Package name: hipsparse
Version : 5.3.0
Upstream Author : Advanced Micro Devices, Inc.
* URL : https://github.com
Package: wnpp
Severity: wishlist
Owner: Cordell Bloor
X-Debbugs-Cc: debian-de...@lists.debian.org, c...@slerp.xyz,
debian...@lists.debian.org
* Package name: hipfft
Version : 5.3.0
Upstream Author : Advanced Micro Devices, Inc.
* URL : https://github.com
Package: wnpp
Severity: wishlist
Owner: Cordell Bloor
X-Debbugs-Cc: debian-de...@lists.debian.org, c...@slerp.xyz,
debian...@lists.debian.org
* Package name: hipcub
Version : 5.3.0
Upstream Author : Advanced Micro Devices, Inc.
* URL : https://github.com
Package: wnpp
Severity: wishlist
Owner: Cordell Bloor
X-Debbugs-Cc: debian-de...@lists.debian.org, c...@slerp.xyz,
debian...@lists.debian.org
* Package name: hipblas
Version : 5.3.0
Upstream Author : Advanced Micro Devices, Inc.
* URL : https://github.com
Package: wnpp
Severity: wishlist
Owner: Cordell Bloor
X-Debbugs-Cc: debian-de...@lists.debian.org, debian...@lists.debian.org
* Package name: rocsolver
Version : 5.3.0
Upstream Author : Advanced Micro Devices, Inc.
* URL : https://github.com/ROCmSoftwarePlatform
Package: wnpp
Severity: wishlist
Owner: Cordell Bloor
X-Debbugs-Cc: debian-de...@lists.debian.org, debian...@lists.debian.org
* Package name: rocsparse
Version : 5.3.0
Upstream Author : Advanced Micro Devices, Inc.
* URL : https://github.com/ROCmSoftwarePlatform
Package: wnpp
Severity: wishlist
Owner: Cordell Bloor
X-Debbugs-Cc: debian-de...@lists.debian.org, c...@slerp.xyz,
debian...@lists.debian.org
* Package name: rccl
Version : 5.3.0
Upstream Author : Advanced Micro Devices, Inc.
* URL : https://github.com
Package: wnpp
Severity: wishlist
Owner: Cordell Bloor
X-Debbugs-Cc: debian-de...@lists.debian.org, c...@slerp.xyz,
debian...@lists.debian.org
* Package name: rocblas
Version : 5.3.0
Upstream Author : Advanced Micro Devices, Inc.
* URL : https://github.com
The correct value for the URL field should be:
https://github.com/ROCmSoftwarePlatform/rocFFT
Apologies for the mistake. I didn't have my mail client set up correctly
the first time I submitted and the original report was lost into the
ether. It seems that I rushed through filling out the
Package: wnpp
Severity: wishlist
Owner: Cordell Bloor
X-Debbugs-Cc: debian-de...@lists.debian.org, debian...@lists.debian.org
* Package name: rocfft
Version : 5.3.0
Upstream Author : Advanced Micro Devices, Inc.
* URL : https://www.example.org/
* License
Hi Étienne,
On 2022-10-26 12:44, Étienne Mollier wrote:
I'm not too sure yet of the implementation to not export the oam
target; this looks to have been autogenerated by cmake. About
the dependency, I'm thinking twice before making that a hard
one, since it is possible to make use of the
1 - 100 of 110 matches
Mail list logo