All,
During Connect the suggestion was made that each working group should have
its own IRC Channel for discussions and topics relating to the group in
particular (as opposed to #linaro which is 'generic' Linaro conversations).
Therefore I have just set up #linaro-tcwg on Freenode for the Toolchain
Working Group.
This channel is public and open to anyone who wants to talk with the TCWG
group about anything toolchain related.
Thanks,
Matt
--
Matthew Gretton-Dann
Toolchain Working Group, Linaro
Hi,
We are having some serious problems after we upgraded from C++14 to C++17
on an Jetson TX2 ARM device. Our system tests started to behave differently
and fail.
It seems that when our application uses a library (also developed by us)
some data gets corrupted when delivered to a class constructor. For
example, the .second of and std::pair<float> appears to be the .first and
the .second is garbage. This is deterministic, but different tests are
failing depending on the combination: library C++17/C++14 <-> application
C++14/C++17.
This is on Ubuntu 18.04 and gcc version 7.5.0 (Ubuntu 7.5.0-3ubuntu1~18.04).
Nothing like this happens on Intel.
So:
ARM, C++14: OK
Intel, C++14: OK
ARM, C++17: FAIL
Intel, C++17: OK
Any ideas what could cause this? I know this is a bit vague, but this a
commercial, closed-source application so I cannot yet give any other
information.
BR,
Jussi Lind
[VIRT-349 # QEMU SVE2 Support ]
Posted the first incremental patch set
for review, as requested by Qualcomm.
[VIRT-327 # Richard's upstream QEMU work ]
Patch review, much of it 5.0 related,
but also v6 of the riscv vector patch set.
Revise the PIE and linkage patch set.
r~
== Progress ==
* GCC upstream validation:
- reported a couple of failures/regressions
* GCC:
- comparing testsuite results for cortex-m3 and cortex-m33 with
cortex-a9 to check what cortex-M problem there are. Fixed a few
testcases. Managed to use qemu-system-arm to run cortex-m33 code with
CMSE features. (Thanks to Peter Maydell for the help!)
* misc:
- infra fixes / troubleshooting / reviews
== Next ==
* FDPIC GDB
* GCC/cortex-M
Progress:
* VIRT-65 [QEMU upstream maintainership]
- Softfreeze this week, so a lot of pull request handling, and
one last small arm pullreq to put together and send out
- Our Coverity Scan setup was broken (the person who usually does
the builds updated his machine to a newer Fedora which broke the tool).
I pulled up an old patchset I had from a year ago which scripts things
so you can do a build-and-upload automatically via Docker so that
we're not dependent on one person for this. This took longer than I
expected as there were a bunch of things that needed fixing both in
the scripting and in bits of the codebase that the tools warned about.
Patch series sent.
- When we did get a Coverity run through, it reported 112 new issues;
spent some time going through and triaging them (lots of false positives
but also lots of real bugs) and submitting patches for some of them.
* VIRT-364 [QEMU support for ARMv8.1-M extensions]
- sketched out the work required and created some 'story' level
cards to go under the epic. Still open questions:
+ should we model Cortex-M55 or a more generic 'max' CPU?
+ what board model do we want to use?
+ does LITE have any guest code using v8.1M features for testing?
Random notes:
* my new desk chair and standing desk arrived this week
(the latter still needs assembling...) so I am gradually
improving the rough edges of my WFH setup.
thanks
-- PMM
== Progress ==
* Out of office 1 day
* More Morello
- Patch set for the 'disassemble' command in review
- Working on the final part of the capability formatting
== Plan ==
* More Morello
Progress:
[VIRT-349 # QEMU SVE2 Support ]
10 of N insn groups implemented.
Annoyingly, there doesn't seem to be a summary of new insns, so confirming that
all of the insns have been implemented.
[VIRT-327 # Richard's upstream QEMU work ]
Patch review. Mostly Phil's arm --disable-tcg set.
Cleanup for some of the new Coverity reports.
Flush tcg patch queue.
[GCC]
Two revisions improving TImode comparisons for aarch64 (PR94174).
https://gcc.gnu.org/pipermail/gcc-patches/2020-March/542447.html
[Kernel]
Follow-up on a discussion a few weeks ago re __range_ok. The gcc cmpti patch
set is in support of that, but I also found a much better solution for constant
sized checks.
http://lists.infradead.org/pipermail/linux-arm-kernel/2020-March/719754.html
r~
VirtIO Related Work ([VIRT-366])
================================
- Whole bunch of scoping work on tasks for this cycle
- Briefed the TSC Strategic Meeting on work plan
- Feedback on Friday call seemed positive, hopefully will get the
votes
[VIRT-366] <https://projects.linaro.org/browse/VIRT-366>
QEMU Tooling ([VIRT-252])
=========================
Extend gdbstub for SVE ([VIRT-281])
- posted [PATCH v1 00/28 for 5.0] testing and gdbstub Message-Id:
<20200316172155.971-1-alex.bennee(a)linaro.org>
- now merged and parent cards closed out including
<https://projects.linaro.org/browse/VIRT-26>
Upstream Work ([VIRT-109])
==========================
- posted [PATCH v1 0/5] testing updates (docker and acceptance CI
tweaks) Message-Id: <20200309202318.3481-1-alex.bennee(a)linaro.org>
- spent a fair time fighting with check-acceptance tests weirdness
- fielded questions from potential students for GSoC cache modelling
project
- more time tracking down "check-acceptance" issues
- posted [RFC PATCH for 5.0] configure: disable MTTCG for MIPS
guests Message-Id: <20200320114522.16273-1-alex.bennee(a)linaro.org>
Completed Reviews [2/2]
=======================
[PATCH v10 0/3] Acceptance test: Add "boot_linux" acceptance test
Message-Id: <20200317141654.29355-3-crosa(a)redhat.com>
[PATCH 0/5] QEMU Gating CI
Message-Id: <20200312193616.438922-5-crosa(a)redhat.com>
Absences
========
- Splitting time between work and home-schooling
Current Review Queue
====================
* [PATCH 0/2] thread: add lock guard macros
Message-Id: <20200311123624.277221-1-stefanha(a)redhat.com>
Added: <2020-03-11 Wed>
* [PATCH v8 0/9] qemu-binfmt-conf.sh
Message-Id: <20200307170251.GA7@dd5f6ec33fb0>
Added: <2020-03-08 Sun>
* [RFC][PATCH 0/3] IVSHMEM version 2 device for QEMU
Message-Id: <cover.1573477032.git.jan.kiszka(a)siemens.com>
Added: <2020-02-14 Fri>
* [RFC 0/9] Add an interVM memory sharing device
Message-Id: <1580815851-28887-1-git-send-email-i.kotrasinsk(a)partner.samsung.com>
Added: <2020-02-05 Wed>
--
Alex Bennée
== Progress ==
* BFD Linker:
- GNU-629: non-contiguous memory support: minor follow-ups
* GCC upstream validation:
- reported a couple of failures/regressions
- PR90378 working on reproducing a non-functional code problem with spec2006
on arm with LTO and -Os. Looks like a stack corruption issue
* GCC:
- comparing testsuite results for cortex-m3 and cortex-m33 with
cortex-a9 to check what cortex-M problem there are.
* misc:
- infra fixes / troubleshooting / reviews
== Next ==
* spec2006 / LTO / Os / aarch32
* FDPIC GDB
* GCC/cortex-M
