Tags

gdbstub/next: re-organise and split build

I was motivated to sort this out while working on my register API
which is target agnostic but ran into the weeds when trying to link up
with the gdbstub. This was due to us building gdbstub for every single
target we support due to a few ABI sensitive bits that require CPU
specific information. This series does a bunch of surgery to break the
monolithic file apart into its constituent parts as well as simplify
the headers to users can avoid bringing in more dependencies than they
need.

I had hoped to go all the way and conditionally compile syscalls only
for the two ABIs (32 and 64 bit) unfortunately I was unable to the
appropriate meson-foo to make that happen.

This version is mostly just minor clean-ups and tag updates including
a few extra code motion and checkpatch cleanup patches. The biggest
change is replacing the probe shell script with a slightly smarter
python one and adding Mad's accelops patch.

The following patches need review:

 - gdbstub: split out softmmu/user specifics for syscall handling
 - testing: probe gdb for supported architectures ahead of time
 - gdbstub: only compile gdbstub twice for whole build
 - gdbstub: clean-up indent on gdb_exit

testing updates:

  - ensure socat available for tests
  - skip socat tests for MacOS
  - properly clean up fifos after use
  - make fp-test less chatty
  - store test artefacts on Cirrus
  - control custom runners with QEMU_CI knobs
  - disable benchmark runs under tsan build
  - update ubuntu 2004 to 2204
  - skip nios2 kernel replay test
  - add tuxrun baselines to avocado
  - binary build of tricore tools
  - export test results on cross builds
  - improve windows builds
  - ensure we properly print TAP headers
  - migrate away from docker.py for building containers
  - be more efficient in our handling of build artefacts between stages
  - enable ztsd in containers so we can run tux_baselines
  - disable heavyweight PPC64 Boot Linux test in CI

* New Sapphire Rapids model support
* x86 bugfixes
* Prepare to drop support for Python 3.6

testing updates:

  - ensure socat available for tests
  - skip socat tests for MacOS
  - properly clean up fifos after use
  - make fp-test less chatty
  - store test artefacts on Cirrus
  - control custom runners with QEMU_CI knobs
  - disable benchmark runs under tsan build
  - update ubuntu 2004 to 2204
  - skip nios2 kernel replay test
  - add tuxrun baselines to avocado
  - binary build of tricore tools
  - export test results on cross builds
  - improve windows builds

Update CentOS 8 playbook and CI configuration to Python 3.8

The playbook has already been run by Cleber on the machine.

Testing, docs, semihosting and plugin updates

  - update playbooks for custom runners
  - add section timing support to gitlab
  - upgrade fedora images to 37
  - purge perl from the build system and deps
  - disable unstable tests in CI
  - improve intro, emulation and semihosting docs
  - semihosting bug fix and O_BINARY default
  - add memory-sve test
  - fix some races in qht
  - improve plugin handling of memory helpers
  - optimise plugin hooks
  - fix some plugin deadlocks
  - reduce win64-cross build time by dropping some targets

Testing, docs, semihosting and plugin updates

  - update playbooks for custom runners
  - add section timing support to gitlab
  - upgrade fedora images to 37
  - purge perl from the build system and deps
  - disable unstable tests in CI
  - improve intro, emulation and semihosting docs
  - semihosting bug fix and O_BINARY default
  - add memory-sve test
  - fix some races in qht
  - improve plugin handling of memory helpers
  - optimise plugin hooks
  - fix some plugin deadlocks

maintainer tree pre-PR (testing/docs/semihosting/plugins)

Hi,

This is a roll-up of all my maintainer trees which contains fixes all
over QEMU. I'd hoped to minimise the disruption of multiple pull
requests but in the end spent a lot of time trying to re-create the
win64 cross failure which only occurs on CI. I'm not sure if the
failure is just down to some sort of weird registry issue so I'm kind
of hoping it disappears once merged. However if anyone is able to run
CI on their repo you can try:

  git push-ci -f gitlab

and then manually run:

  win64-fedora-cross-container (in containers)

and then:

  cross-win64-system (in builds)

and let me know if it hangs for you to. In which case I'll just have
to drop the Fedora 37 updates all together. Either way I'll base the
final PR of this series.

A summary of the contents:

  - some cleanups and tweaks for gitlab
  - a new playbook to install gitlab runners
  - a proposed re-org of the front about section
  - a few semihosting fixes (with docs ;-)
  - a bunch of tsan fixes touching tcg and plugins
  - fixing some plugin races
  - expunging the last vestiges of perl from the build system

The following patches need review:

  plugins: Iterate on cb_lists in qemu_plugin_user_exit
  docs: add a new section to outline emulation support (1 acks, 1 sobs)
  gitlab: add lsan suppression file to workaround tcmalloc issues
  scripts/ci: update gitlab-runner playbook to use latest runner

current maintainer trees (testing/semihosting/plugins)

Hi All,

I've wrapped up all my maintainer trees into a single branch to
minimise the disruption of multiple pull requests. Most stuff has
already been reviewed but in short we have:

  - some cleanups and tweaks for gitlab
  - a new playbook to install gitlab runners
  - a proposed re-org of the front about section
  - a few semihosting fixes (with docs ;-)
  - a bunch of tsan fixes touching tcg and plugins
  - fixing some plugin races
  - expunging the last vestiges of perl from the build system

The following patches need review:

  tests/tcg: add memory-sve test for aarch64
  semihosting: add semihosting section to the docs
  docs: add a proper feature overview in "About QEMU"
  tests/unit: drop hacky race avoidance in test-io-channel-command
  gitlab: just use plain --cc=clang for custom runner build
  gitlab: add FF_SCRIPT_SECTIONS for timings
  scripts/ci: update gitlab-runner playbook to use latest runner

testing updates:

  - fix minor shell-ism that can break check-tcg
  - turn off verbose logging on custom runners
  - make configure echo call in CI
  - fix unused variable in linux-test
  - add binary compiler docker image for hexagon
  - disable doc and gui builds for tci and disable-tcg builds

testing updates:

  - fix minor shell-ism that can break check-tcg
  - turn off verbose logging on custom runners
  - make configure echo call in CI
  - fix unused variable in linux-test
  - add binary compiler docker image for hexagon
  - disable doc and gui builds for tci and disable-tcg builds

Testing and doc updates:

  - Only probe if docker or podman binaries in path
  - reduce console noise for aspeed avocado tests
  - update documents on maintainer roles and process
  - raise timeout for ppc64 avocado tests
  - integrate coverage reports into gitlab

Testing and doc updates:

  - Only probe if docker or podman binaries in path
  - tweak avacado console to better find login prompts
  - reduce console noise for aspeed avocado tests
  - update documents on maintainer roles and process
  - raise timeout for ppc64 avocado tests
  - integrate coverage reports into gitlab

Testing and doc updates:

  - Only probe if docker or podman binaries in path
  - tweak avacado console to better find login prompts
  - reduce console noise for aspeed avocado tests
  - update documents on maintainer roles and process
  - raise timeout for ppc64 avocado tests
  - integrate coverage reports into gitlab

testing and plugin updates for 7.2:

  - cleanup win32/64 docker files
  - update test-mingw test
  - add flex/bison to debian-all-test
  - handle --enable-static/--disable-pie in config
  - extend timeouts on x86_64 avocado tests
  - add flex/bison to debian-hexagon-cross
  - use regular semihosting for nios2 check-tcg
  - fix obscure linker error to nios2 softmmu tests
  - various windows portability fixes for tests
  - clean-up of MAINTAINERS
  - use -machine none when appropriate in avocado
  - make raspi2_initrd test detect shutdown
  - disable sh4 rd2 tests on gitlab
  - re-enable threadcount/linux-test for sh4
  - clean-up s390x handling of "ex" instruction
  - better handle new CPUs in execlog plugin
  - pass CONFIG_DEBUG_TCG to plugin builds
  - try and avoid races in test-io-channel-command
  - speed up ssh key checking for tests/vm

testing and plugin updates for 7.2:

  - cleanup win32/64 docker files
  - update test-mingw test
  - add flex/bison to debian-all-test
  - handle --enable-static/--disable-pie in config
  - extend timeouts on x86_64 avocado tests
  - add flex/bison to debian-hexagon-cross
  - use regular semihosting for nios2 check-tcg
  - fix obscure linker error to nios2 softmmu tests
  - various windows portability fixes for tests
  - clean-up of MAINTAINERS
  - use -machine none when appropriate in avocado
  - make raspi2_initrd test detect shutdown
  - disable sh4 rd2 tests on gitlab
  - re-enable threadcount/linux-test for sh4
  - clean-up s390x handling of "ex" instruction
  - better handle new CPUs in execlog plugin
  - pass CONFIG_DEBUG_TCG to plugin builds
  - try and avoid races in test-io-channel-command
  - speed up ssh key checking for tests/vm