On 10/21/26 22:50, Alex Bennée wrote:
Ilya Leoshkevich <[email protected]> writes:
On 10/19/26 21:26, Alex Bennée wrote:
This resurrects the hand run test script from d84842be6c5 (tests:
remove test-gdbstub.py) and modernises it to use the shared test
infrastructure for run-test.
We then tweak the common tuxrun code to over an alternative "launch
via gdbstub" helper which reuses the existing code but:
- needs a vmlinux with symbols
- drops kaslr because it doesn't play nice with symbols
- extend qargs with stuff normally added by vm.launch
- passes kargs directly
- runs with the gdb run-test script
- log the run-test output to the console log
- makes --binary not mandatory
Message-ID: <[email protected]>
Signed-off-by: Alex Bennée <[email protected]>
---
v3
- mention the nokaslr in the commit message
- add a comment about it to the shared kernel args
- drop unused failcount
- remove stale comment about hbreak
- always delete the watch point pass or fail
- fix unbalanced tuple in sh4's direct call of fetch_tuxrun_assets
- simplify the symbol to watchpoint logic
- drop the mandatory --binary
- make the command copy and paste friendly
v2
- add self.vm._machine to qargs
- drop leading spaces in print () calls
- add register read tests
- use gdb.Breakpoint for hwbreak as for swbreak
- drop flaky symbol evaluation, verify hit_count instead
- explicit gdb.BP_BREAKPOINT for CatchBreakpoint
---
.../functional/gdb-scripts/kernel_gdb_test.py | 162 ++++++++++++++++++
tests/functional/qemu_test/tuxruntest.py | 97 ++++++++++-
tests/functional/sh4/test_tuxrun.py | 4 +-
tests/guest-debug/run-test.py | 14 +-
4 files changed, 261 insertions(+), 16 deletions(-)
create mode 100644 tests/functional/gdb-scripts/kernel_gdb_test.py
[...]
@@ -132,10 +144,77 @@ def common_tuxrun(self,
special with the command line we can process most things using
the tag metadata.
"""
- (kernel, disk, dtb) = self.fetch_tuxrun_assets(kernel_asset,
rootfs_asset,
- dtb_asset)
+ (kernel, disk, dtb, _) = self.fetch_tuxrun_assets(kernel_asset,
+ rootfs_asset,
+ dtb_asset)
self.prepare_run(accel, kernel, disk, drive, dtb,
console_index)
self.vm.launch()
self.run_tuxtest_tests(haltmsg)
os.remove(disk)
+
+ def common_tuxrun_gdb(self,
+ kernel_asset,
+ vmlinux_asset,
+ rootfs_asset,
+ accel=None,
+ dtb_asset=None,
+ drive="virtio-blk-device",
+ haltmsg="reboot: System halted",
+ console_index=0):
+ """
+ Common path for LKFT tests under gdb. Unlike the common_tuxrun
+ function we need to launch with the gdb test script and deal
+ with munging the arguments for that especially.
+ """
+ (kernel, disk, dtb, vmlinux) = self.fetch_tuxrun_assets(kernel_asset,
+ rootfs_asset,
+ dtb_asset,
+ vmlinux_asset)
+
+ gdb_path = os.getenv("QEMU_TEST_GDB")
By the way, there is a minor usability issue: if I set QEMU_TEST_GDB to
a non-existent binary, the test hangs instead of exiting with an error
message.
I can add a guard - but in normal invocation meson should take care only
to pass the env variable if configure detects gdb.
I haven't debugged this, but I wonder if it would rather be possible to
structure the code in a way that does not lead to a hang on GDB failure.
I suspect that GDB crashes or GDB script errors may also lead to hangs.
