summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
9 daysMerge pull request #44 from samhug/snapshotsHEADmasterkoverstreet
tests/fs/bcachfs/subvol: add snapshots_while_stressed test
9 daystests/fs/bcachfs/subvol: add snapshots_while_stressed testSam Hug
2025-07-18tests/bcachefs/replication/largebuckets: up timeoutKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-07-16tests/fs/bcachefs: image_createKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-07-16tests/fs/bcachefs/replication: add --runtime to certain fio testsKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-07-16tests/bcachefs/single_device/reflink: bump timeoutKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-07-16tests/bcachefs/tier: Bump timeout on recover_superKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-07-16tests/bcachefs/single_device recovery: up bucket size to 64kKent Overstreet
We can't run with 8k buckets anymore - we deadlock on the journal. Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-07-13Merge pull request #43 from samhug/rereplicate2koverstreet
tests/fs/bcachefs: add rereplicate2 test
2025-07-13tests/fs/bcachefs: add rereplicate2 testSam Hug
wipe the superblock on a device before doing degraded mount
2025-07-13bcachefs: blacklist btree path events from slowpath countersKent Overstreet
these are low level events, and don't necessarily cause transaction restarts Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-07-06root_image: extra packagesKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-07-05tests/bcachefs: up online_fsck timeoutKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-07-05bcachefs: tweak failure rate for test_btree_read_corruptKent Overstreet
We can't always recover from a btree read error during recovery on the first mount - it might take a second mount, if it happens in the interior update path. Adjust the failure rate so we see fewer spurious test failures. Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-07-04tests/bcachefs/replication: New error injection testsKent Overstreet
Add tests that use our new knobs for fine grained IO corruption injection. Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-06-30bcachefs: online fsck testKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-06-30Merge pull request #42 from tim-day-387/pci-passthroughkoverstreet
PCI passthrough tests for Lustre
2025-06-30tests/fs/lustre: mount server via PCI passthroughTimothy Day
Signed-off-by: Timothy Day <timday@amazon.com>
2025-06-27tests/fs/lustre: mount client via PCI passthroughTimothy Day
Signed-off-by: Timothy Day <timday@amazon.com>
2025-06-26misc: QEMU prepend for NUMA pinningTimothy Day
It can be useful to use numactl to pin a VM to a NUMA node. Add ktest_qemu_prepend to enable this in ktests. Signed-off-by: Timothy Day <timday@amazon.com>
2025-06-26misc: add IdentitiesOnly to ssh commandTimothy Day
Specify IdentitiesOnly=yes during ssh to prevent ssh from trying keys that won't work. Without this, we can sometimes fail to ssh if we have too many keys. Signed-off-by: Timothy Day <timday@amazon.com>
2025-06-25Merge remote-tracking branch 'timday/lustre-updates' into HEADKent Overstreet
2025-06-25tests/fs/bcachefs: kill_btree_root_autofixKent Overstreet
Test killing a btree root that we should be able to automatically repair node (extents, when tree depth > 0). Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-06-22Update memory OSD mount codeTimothy Day
Signed-off-by: Timothy Day <timday@amazon.com>
2025-06-22Improve LLVM supportTimothy Day
Enable $CC to be passed on the build-test-kernel CLI. Properly set the $LLVM environment variable when using Clang. This ensures that the full LLVM toolchain is used (rather than just the compiler). Signed-off-by: Timothy Day <timday@amazon.com>
2025-06-17tests/bcachefs: fix casefold + overlayfs testKent Overstreet
We're now going with the simpler fix from Amir, update test for different behaviour. Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-06-15test-git-branh.sh: fewer tmpdirsKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-06-15tests/bcachefs: fix single device testKent Overstreet
--single_device was a development thing, the option no longer exists Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-06-15tests/bcachefs: don't check write_buffer_flush eventsKent Overstreet
There are tests where many of these are expected. Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-06-11tests/bcachefs: fix merge_torture_flakeyKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-06-07Merge pull request #40 from christoph-heinrich/add_unmount_mount_testkoverstreet
bcachefs: test_device_add_remount
2025-06-07bcachefs: test_device_add_remountChristoph Heinrich
2025-05-25bcachefs: test_casefold_overlayfsKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-05-25bcachefs: test_single_deviceKent Overstreet
Add a test for verifying that single device mode works, i.e. we can mount multiple single device filesystems with the same UUID. Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-05-25bcachefs: test_recover_superKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-05-25bcachefs: require BCACHEFS_TRANS_KMALLOC_TRACEKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-05-15tests/bcachefs: enable async obj listsKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-05-10tests/bcachefs: hook up new debug paramsKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-05-04tests/bcachefs: replication recovery testKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-04-30tests/bcachefs: new test for koverstreet/bcachefs#834Stijn Tintel
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be> Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-04-29gen-job-list: Don't hide output for fetch remotesKent Overstreet
this is running slowly, we need to know which remote Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-04-22bcachefs: fix units.ktestKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-04-18bcachefs: test_tiering_misalignedKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-04-17bcachefs: test_subvol_mv_snapshot2()Kent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-04-16bcachefs: improve test_subvol_mv_snapshot()Kent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-04-14tests/fs/bcachefs/snapshot_file_delete_2Kent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-04-14tests/fs/bcachefs/subvol.ktest snapshot_nestedKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-04-14root_image: include squashfs, erofs toolsKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-04-03fixup! root_image: provide a persistent bash historyKent Overstreet
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
2025-04-02root_image: provide a persistent bash historyKent Overstreet
also, some cleanups Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
generated by cgit v1.2.3 (git 2.46.0) at 2025-08-03 02:46:07 +0000