Bug#994833: OpenCL programs abort when intel-opencl-icd is installed

2022-01-31 Thread Andreas Beckmann
On 22/09/2021 15.59, Giuseppe Bilotta wrote: On Wed, Sep 22, 2021 at 8:32 AM Andreas Beckmann wrote: So if it does matter abi-wise for intel-opencl-icd which version of llvm libigfoo1 was compiled against, we should model this in the dependency chain. To be able to confirm this, I would need

Bug#994833: OpenCL programs abort when intel-opencl-icd is installed

2021-11-15 Thread Yuri D'Elia
Is there any progress on the FTBS? libigc is now at 1.0.8744-2 which breaks 21.32.20609-1, making the icd uninstallable.

Bug#994833: OpenCL programs abort when intel-opencl-icd is installed

2021-10-29 Thread Andreas Beckmann
Control: block -1 with 998058 Control: fixed -1 intel-graphics-compiler/1.0.8744-2 On 29/10/2021 16.33, Jonas Smedegaard wrote: Reopening, since seemingly the detail about coordinated upload is still missing: Currently (in sid) intel-opencl-icd is flagged breaking its own dependencies and

Bug#994833: OpenCL programs abort when intel-opencl-icd is installed

2021-10-29 Thread Jonas Smedegaard
Control: reopen -1 Andreas Beckmann wrote: > Commits are pushed to both packages in git, they need a coordinated > upload due to the Depends/Breaks added. Reopening, since seemingly the detail about coordinated upload is still missing: Currently (in sid) intel-opencl-icd is flagged breaking

Bug#994833: OpenCL programs abort when intel-opencl-icd is installed

2021-10-14 Thread Andreas Beckmann
Control: tag -1 pending On 22/09/2021 15.59, Giuseppe Bilotta wrote: The virtual ABI package sounds like the best choice. I've tried to implement this in src:intel-graphics-compiler and src:intel-compute-runtime. Couldn't test the first one due to the gcc 11 ftbfs, but the second one does

Bug#994833: OpenCL programs abort when intel-opencl-icd is installed

2021-09-22 Thread Giuseppe Bilotta
On Wed, Sep 22, 2021 at 8:32 AM Andreas Beckmann wrote: > So if it does matter abi-wise for intel-opencl-icd which version of llvm > libigfoo1 was compiled against, we should model this in the dependency > chain. To be able to confirm this, I would need a version of all the libig* stuff

Bug#994833: [Pkg-opencl-devel] Bug#994833: OpenCL programs abort when intel-opencl-icd is installed

2021-09-22 Thread Rebecca N. Palmer
On 22/09/2021 07:32, Andreas Beckmann wrote: We will proably still run into problems if different ICDs built against different LLVM versions are going to be loaded at the same time (e.g. pocl/llvm9 and intel/llvm1x) because the different llvm versions seem to stomp on each others internal

Bug#994833: OpenCL programs abort when intel-opencl-icd is installed

2021-09-22 Thread Andreas Beckmann
On 21/09/2021 19.21, Giuseppe Bilotta wrote: So I'm guessing the main issue at play here is that libigdfcl.so.1 is _allegedly_ recompiled against LLVM 12 (per the package requires), but for some reason the library is still looking for version 11 things. reassign -1 libigdfcl1 1.0.8279-1 (As a

Bug#994833: OpenCL programs abort when intel-opencl-icd is installed

2021-09-21 Thread Giuseppe Bilotta
Package: intel-opencl-icd Followup-For: Bug #994833 Downgrading ONLY intel-opencl-icd to version 20.44.18297-1 leads to a different error: Abort was called at 41 line in file: /build/intel-compute-runtime-7vSeZ9/intel-compute-runtime-20.44.18297/shared/source/built_ins/built_ins.cpp Aborted

Bug#994833: OpenCL programs abort when intel-opencl-icd is installed

2021-09-21 Thread Giuseppe Bilotta
Package: intel-opencl-icd Version: 21.32.20609-1 Severity: critical Running any OpenCL-enabled program with this ICD installed leads to: Abort was called at 42 line in file: ./shared/source/built_ins/built_ins.cpp Aborted The relevant backtrace is: #0 __GI_raise (sig=sig@entry=6) at