Hello,
# TCWG CI
- Implemented increasing testsuite timeout for each testsuite try in GDB
check jobs. Sent a Gerrit review request for it but then found out that
it makes armhf jobs take a lot longer, so abandoned it.
# Community
- Finished reviewing SME patches for GDB and gdbserver.
--
Thiago
Progress:
* UM-2 [QEMU upstream maintainership]
- some bits and pieces for release
- code review, notably:
+ raspberry pi 4 support patchseries (a big 44-patch set)
+ Xilinx Versal CFI support patches
* QEMU-530 [QEMU ARM v9.4 Baseline CPU for TCG]
- more work on FEAT_MOPS: tested my memset implementation, and
fixed various bugs. Still no MTE tag checking support.
-- PMM
Hello,
# TCWG CI
- CI Babysitting: Worked on two regressions detected at the end of last
week.
- Enabled precommit testing for GDB patches.
# Misc
- Reviewing SME patches for GDB and gdbserver.
--
Thiago
Progress:
* UM-2 [QEMU upstream maintainership]
- release related stuff: handling merge requests, etc (now passed
back to RTH again)
- sent some patches for a few easy coverity issues
- worked through some code review (in particular some patches
from new contributors)
- created a gitlab "bite sized task" issue that better explains
and has more detail on the "convert from malloc to g_malloc"
suggested task for new contributors
* QEMU-530 [QEMU ARM v9.4 Baseline CPU for TCG]
- a little more progress with FEAT_MOPS. I now have an untested
implementation of the memset insns which I think is complete
except for MTE tag checking
-- PMM
Hello,
# TCWG CI
- Babysat the CI: Investigated 9 regressions in GCC, 2 in GDB, 1 in LLVM
and 1 in glibc. Reported 4 in GCC (GNU-855, GNU-857, GNU-858 and
GNU-859). 3 of which are fixed and another has a tentative fix.
- Reviewed a couple of Gerrit requests.
--
Thiago
Progress:
* UM-2 [QEMU upstream maintainership]
- fixed some places in our RTC device models where we were
putting a time_t or time_t offset into a 32-bit variable
- took over from RTH briefly for pullreq merge handling
- sent some patches fixing minor Coverity issues
- investigated a couple of "test case fails on big-endian host"
errors: sent patches for Sparc FPRS handling and for
the Arm SMMUv3 model
- Investigated report that our MPS2/3 M-profile models don't
have the same number of MPU regions configured as the real FPGA
images, and so you can't run the same Zephyr binary on both.
Have some preliminary patches to correct the number of regions.
* QEMU-530 [QEMU ARM v9.4 Baseline CPU for TCG]
- finally got back to FEAT_MOPS work; wrote a skeleton of code
for the memset operations that's about right up to the point
where it needs to actually do the memset...
-- PMM
Hi everyone,
I'm not 100% sure whether this is a bot misconfiguration issue, but couldn't find any other source of this failure:
```
******************** TEST 'test-suite :: MultiSource/Applications/ClamAV/clamscan.test' FAILED ********************
Executable '/home/tcwg-buildbot/worker/clang-aarch64-sve-vla/test/sandbox/build/MultiSource/Applications/ClamAV/clamscan' is missing
********************
```
Link: https://lab.llvm.org/buildbot/#/builders/197/builds/8209
Would anyone be able to take a look?
Thank you :)
-Andrzej
IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you.
Progress:
* UM-2 [QEMU upstream maintainership]
- investigated and fixed a bug in page-table-walk handling of
debug accesses that was introduced by the FEAT_RME changes
- that prompted me to look at a cleanup of some of the ptw code
so that we consistently look at "which space is this walk for"
and don't carry around a parallel boolean "secure/non-secure"
that duplicates that information
- found and fixed a few other ptw bugs along the way
-- PMM
