# Progress #
o Upstream GDB
* More patch reviewing and answering questions.
o GDB:
* GNU-644 - [GDB, AArch64] gdb.base/step-over-syscalls.exp failures
- Managed to track this down all the way to the aarch64
single-stepping infrastructure in the kernel. Created reproduction steps
and sent an e-mail to linux-arm-kernel
- Also noticed a difference in the order in which SIGCHLD's are
delivered between x86_64 and aarch64. This confused GDB, which expects
the ordering it sees on x86_64. Working on a fix.
* Tweaked QEMU setup a little for SVE testing
* GNU-645 - gdbserver is not using SVE register descriptions properly
- Tracked down why gdbserver is not sending SVE register data back
to GDB. Working on a fix.
* GNU-170 - GDB BZ #21221 - gdb hangs while stepping an empty loop
- On hold for now.
# Plan #
o GDB
* GNU-644 - [GDB, AArch64] gdb.base/step-over-syscalls.exp failures
- Continue working on a fix.
* GNU-645 - gdbserver is not using SVE register descriptions properly
- Continue working on a fix.
[VIRT-262 # ARMv8.1-PAN Privileged Access Never]
Started, based on VHE patch set due to mmu_idx reorg therein.
Needs some minor re-work to handle Secure EL1.
[VIRT-273 # ARMv8.2-ATS1E1, AT S1E1R and AT S1E1W instruction variants ]
Started.
[VIRT-327 # Richard's upstream QEMU work ]
Some soft-freeze bug fixing of stuff that I broke this cycle.
Some patch review.
[Kernel]
Posted v7 of the ARMv8.5-RNG patch set. There's some significant
mis-communication going on between me and Mark Rutland; I have no
idea what he wants at this point...
[GCC]
Committed the base asm-flags patch set. Posted a follow-up
to un-break thumb1, and add tests for it.
r~
== This Week ==
* GCC
- PR89007: Posted patch upstream.
- GNU-583: Worked with Kugan, to reproduce LTO failure and verified
his patch fixes it.
- Bug 5479: Investigated with Maxim, most likely a glibc issue.
* Validation:
- Posted abe patch to add --gcc_patch_file option.
== Next Week ==
- Continue ongoing tasks.
Monday off
== Progress ==
* GCC:
- -mpure-code on v6m: answered questions about my patch. Almost OK
* Linker:
- started looking at implementing non-contiguous memory support
* GCC upstream validation:
- reported several issues
* misc:
- infra fixes / troubleshooting / reviews
== Next ==
* GCC: pure-code/v6m, handle feedback
* Binutils: support non-contiguous memory regions in linker
[Morello]
- Dynamic linking implementation complete and tests written based on
GNU ld output.
- Good progress on refactoring the work prior to breaking it down into
reviewable chunks. Have got a good idea of what I'd like the
implementation to look like.
TODO:
- Finish refactoring patches.
- Rebase into reviewable patches.
- Document the design decisions.
- Test on more than just the examples in the toolchain.
- Rebase on top of latest CUCL drop when merge is finished.
[Other]
LLVM-MC upstream review for the Linux Kernel.
QEMU Tooling ([VIRT-252])
=========================
QEMU plugin support ([VIRT-280])
- feature now merged in 4.2
- closed out a bunch of related JIRA cards
- API version now merged and this card is closed
Extend gdbstub for SVE ([VIRT-281])
- posted {RFC PATCH 00/11} gdbstub re-factor and SVE support
Message-Id: <20191115173000.21891-1-alex.bennee(a)linaro.org>
[VIRT-281] https://projects.linaro.org/browse/VIRT-281
[working prototype]
https://github.com/stsquad/qemu/tree/gdbstub/sve-registers
Upstream Work ([VIRT-109])
==========================
- posted {PULL 0/8} testing and tcg plugin api ver Message-Id:
<CAFEAcA_9AwoTE9zaKbiF6DkpN+O8LaEKGOct-m5S3yvFBHGK1g(a)mail.gmail.com>
- posted {PATCH for 4.2-rc2 v1 0/5} misc doc and testing fixes
Message-Id: <20191113115952.775-1-alex.bennee(a)linaro.org>
- posted {RFC PATCH} scripts/tap-driver: report "slow" tests (HACK)
Message-Id: <20191113142101.30280-1-alex.bennee(a)linaro.org>
Other Activities
================
- wrote [draft of KVM Forum conference report]
[draft of KVM Forum conference report]
https://collaborate.linaro.org/pages/resumedraft.action?draftId=128647720
Completed Reviews [5/5]
=======================
{PATCH 0/2} replace sysconf(_SC_PAGESIZE) with qemu_real_host_page_size
Message-Id: <20191015031350.4345-2-richardw.yang(a)linux.intel.com>
{PATCH v2 0/4} arm/aspeed: Watchdog and SDRAM fixes
Message-Id: <20191113005201.19005-1-joel(a)jms.id.au>
{kvm-unit-test PATCH 0/5} Improvements for the Travis CI
Message-Id: <20191113112649.14322-6-thuth(a)redhat.com>
{PATCH} target/arm: Clean up arm_cpu_vq_map_next_smaller asserts
Message-Id: <20191115131623.322-1-richard.henderson(a)linaro.org>
{PATCH v1 0/5} Make the qemu_logfile handle thread safe.
Message-Id: <20191112150105.2498-1-robert.foley(a)linaro.org>
Current Review Queue
====================
* {PATCH v2 0/6} Make the qemu_logfile handle thread safe.
Message-Id: <20191115131040.2834-1-robert.foley(a)linaro.org>
Added: <2019-11-15 Fri>
* {PATCH 0/1} tests/vm: Allow to set path to qemu-img
Message-Id: <20191114134246.12073-1-wainersm(a)redhat.com>
Added: <2019-11-14 Thu>
* {PATCH v7 0/8} Acceptance test: Add "boot_linux" acceptance test
Message-Id: <20191104151323.9883-1-crosa(a)redhat.com>
Added: <2019-11-04 Mon>
* {RFC 0/3} tests/vhost-user-fs-test: add vhost-user-fs test case
Message-Id: <20191025100152.6638-1-stefanha(a)redhat.com>
Added: <2019-10-25 Fri>
--
Alex Bennée
Progress:
* VIRT-65 [QEMU upstream maintainership]
- wrote up first draft of requirements for the better CI setup for
testing pull request merge builds
(https://wiki.qemu.org/Requirements/GatingCI)
- wrote up version of the QEMU Summit minutes that includes enough
context to make sense to people who weren't present; plan to
send to qemu-devel next Friday
- got 4.2 rc1 out of the door; this release seems to be pleasantly
quiet for a change.
thanks
-- PMM
== Progress ==
* Committed SVE-related fix for InstCombine
* Started triaging check-lldb failures on AArch64 [LLVM-512]
- Committed a trivial fix for one of the tests
- Still looking into the other failures
* Received and setup laptop for Morello work
- Awaiting further instructions
== Plan ==
* Start working on Morello
* Keep triaging check-lldb failures
* If time permits, look into SVE asserts some more
# Progress #
o Annual Review
* Concluded
o Upstream GDB
* Patch reviews on gerrit and answering questions.
* Fixed PR25124 - [ARM] regression: thumb-bx-pc.exp
(https://sourceware.org/bugzilla/show_bug.cgi?id=25124)
* Investigated an odd failure in gdb.base/step-over-syscall.exp and
ended up finding some odd stepping behavior that may or may not be
kernel related. Still investigating under GNU-644
(https://projects.linaro.org/browse/GNU-644)
o GDB:
* GNU-170 - GDB BZ #21221 - gdb hangs while stepping an empty loop
- On hold for now. It would be nice to be able to use a NOP instead
of a dummy label + jump.
* Setup QEMU for running SVE bits and reproduced some SVE hiccups in
gdbserver that Alex mentioned.
* Misc discussions about SVE and the GDB implementation in general.
# Plan #
o Upstream GDB
* More patch reviewing and answering questions.
o GDB:
* GNU-644 - [GDB, AArch64] gdb.base/step-over-syscalls.exp failures
- Continue investigating
* Tweak QEMU setup a little for SVE testing
