== Progress ==
* More Morello (updating for new capability encoding)
* Went to the embassy to get my passport
* Read a couple ARM policy refreshers...
== Plan ==
* More Morello
== Progress ==
* BFD Linker:
- GNU-629: non-contiguous memory support: posted v4
* GCC:
* GCC-9: continued work for validation of -mpure-code on v6m:
posted updated patch for latent bug on trunk
* GCC upstream validation:
- reported a couple of failures/regressions
- fixed FDPIC build broken on trunk
* misc:
- infra fixes / troubleshooting / reviews
== Next ==
* Binutils: GNU-629: support non-contiguous memory regions in linker
* GCC-9: continue work for validation of -mpure-code on v6m
* FDPIC GDB
VirtIO Related Work
===================
This is essentially a mind-map of bullet points as we work out what
concrete work items should be implemented by the TCWG Virt hackers
(most likely just me ;-)
Xen for Automotive ([LBI-24])
- Things that need doing
- support for virtio in core (i.e. not just emulated devices)
- support for virtio-iommu and sharing memory between guests
(ivshmemv2?)
- support virtio over ivshmemv2 (would native virtio be supported?)
- support for vhost over shared memory (zero copy device emulation)
- support isolating userspace emulation from hvm (vhost-user)
- Why?
- userspace emulation has performance benefits (exit-less, polling)
- growing number of virtio devices for that guests support
[LBI-24] <https://projects.linaro.org/browse/LBI-24>
Virtualizing Android ([LBI-38])
- Builds on Google's work on Cuttlefish
- New Android Device model using VirtIO (including VirtGPU) \o/!
- Used to run on QEMU, now targeting CrosVM
- Potential interest in using as a HAL layer
- Potential work items
- Better support for ARM in CrosVM
- Is QEMU of interest to members? (emulation & tooling - unlike
CrosVM)
- Is a virtualised Android HAL possible
- Where does Hafnium fit it?
[LBI-38] <https://projects.linaro.org/browse/LBI-38>
QEMU Tooling ([VIRT-252])
=========================
Extend gdbstub for SVE ([VIRT-281])
- re-based [a v7 branch]
[VIRT-281] <https://projects.linaro.org/browse/VIRT-281>
[a v7 branch]
<https://github.com/stsquad/qemu/tree/gdbstub/sve-registers-v7>
Upstream Work ([VIRT-109])
==========================
- posted [PATCH v2 00/12] testing/next (with build fixes!) Message-Id:
<20200130113223.31046-10-alex.bennee(a)linaro.org>
- posted [PATCH v2 00/19] testing and plugin updates Message-Id:
<20200213225109.13120-1-alex.bennee(a)linaro.org>
[VIRT-109] <https://projects.linaro.org/browse/VIRT-109>
Completed Reviews [3/3]
=======================
[RFC PATCH 1/2] GitLab CI: avoid calling before_scripts on unintended jobs
Message-Id: <5d0def0e-0943-3345-784d-80f8ccc318b9(a)redhat.com>
[PATCH v1 00/14] tests/vm: Add support for aarch64 VMs
Message-Id: <20200205212920.467-1-robert.foley(a)linaro.org>
- CLOSING NOTE [2020-02-07 Fri 19:56]
Some problems with hangs, not sure about using threads/file
approach.
Added: <2020-02-06 Thu>
[PATCH 0/2] target/arm: Pass arguments by value for sve FMLA/FCMLA
Message-Id: <20200212025145.24300-1-richard.henderson(a)linaro.org>
Absences
========
- 17t-24th Feb Half Term
- 23rd-27th March BUD20
Current Review Queue
====================
* [PATCH v16 00/10] VIRTIO-IOMMU device
Message-Id: <20200214132745.23392-1-eric.auger(a)redhat.com>
Added: <2020-02-14 Fri>
* [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>
* {PATCH v2 0/2} tests/tcg/multiarch: Add tests for implemented real
Message-Id: <1581603905-21565-1-git-send-email-Filip.Bozuta(a)rt-rk.com>
Added: <2020-02-13 Thu>
* [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:
* VIRT-241 [QEMU ISA Support for A-profile]
- Looking into implementing some of the easier, smaller architecture features
- Implemented ARMv8.1-VMID16 (16-bit VMID support), which is trivial
for QEMU because it doesn't care much about VMIDs
- Working on ARMv8.1-PMU and ARMv8.4-PMU: it turns out that we already have
most of these extensions, so we just need to finish off some smaller pieces
and fix one or two bugs. Sent out a patchset implementing these.
- Found a bunch of ID-register related bugs in the process, which I have
fixes for
- Implemented the "mandatory from v8.2" ACTLR2 and HACTLR2 registers
* VIRT-65 [QEMU upstream maintainership]
- code review:
- series from Laurent fixing a long-standing issue with handling of
realtime signals in our linux-user code
- imx6 model patches to fix a watchdog device bug and add the
watchdog to our imx6 boards
- final few unreviewed patches in rth's "PAN, ATS1E1, UAO" series
- respin of the json QAPI Sphinx conversion patchset; hopefully the
first half of this is now uncontroversial cleanups that can go in
while the second half gets code reviewed
thanks
-- PMM
o LLVM:
* 10.0.0 RC:
- reported issues upstream
- upload AArch64 binaries
* IR Outliner:
- looking at the branch status
o Misc
* Resurrected an old GNU toolchain prototype w/r to Linux kernel
size reduction.
* Various meetings and discussions.
== Progress ==
* BFD Linker:
- GNU-629: non-contiguous memory support: posted v3, no feedback yet.
* GCC:
* GCC-9: continued work for validation of -mpure-code on v6m:
reproduced latent bug on trunk, fix accepted, but I'm still worried by
it not setting all insn attributes correctly.
* GCC upstream validation:
- reported a couple of failures/regressions
- FDPIC build broken on trunk
* misc:
- infra fixes / troubleshooting / reviews
== Next ==
* Binutils: GNU-629: support non-contiguous memory regions in linker
* GCC-9: continue work for validation of -mpure-code on v6m
* Fix FDPIC build
[ Linux on ARM mini-conference ]
2 days. Some interesting stuff in the future hardware directions talks.
Helped out a bit with ABI history in the toolchain breakout session. Some good
hallway discussion, especially re record-replay on aarch64.
I should make time this weekend to produce a more complete trip report.
[VIRT-263 # ARMv8.1-VHE Virtual Host Extensions ]
Posted v7 -- merged!
[VIRT-262 # ARMv8.1-PAN ]
[VIRT-273 # ARMv8.2-ATS1E1 ]
[VIRT-276 # ARMv8.2-UAO ]
Posted v3.
[VIRT-327 # Richard's upstream QEMU work ]
Random patch review.
Updated some aa32 vfp cleanup patches; still need to
re-test and post.
r~
Progress:
* VIRT-65 [QEMU upstream maintainership]
- sent pullreq with some of the pending rST doc conversion work
- finished the work on converting the json QAPI doc comments from
texinfo to rst, and sent out a patchset for this
- code review:
- Cornelia's patches rstifying some s390 docs
- rth's 'Reduce aa64_va_parameter overhead' patch
- a patchset fixing some minor memory leaks in timer devices
- rth's PAN/ATS1E1/UAO patchset
- sent out an arm pullreq
I seem to be currently keeping on top of the code-review queue
with about an afternoon a week spent on review, which is pleasing.
thanks
-- PMM
