Changed all dir tests to be reentrant

To help with this, added TEST_PL, which is set to true when powerloss
testing. This way tests can check for stronger conditions (no EEXIST)
when not powerloss testing.

With TEST_PL, there's really no reason every test in t5_dirs shouldn't
be reentrant, and this gives us a huge improvement of test coverage very
cheaply.

---

The increased test coverage caught a bug, which is that gstate wasn't
being consumed properly when mtree uninlining. Humorously, this went
unnoticed because the most common form of mtree uninlining, mdir splitting,
ended up incorrectly consuming the gstate twice, which canceled itself
out since the consume operation is basically just xor.

Also added support for printing dstarts to dbglfs.py, to help debugging.
This commit is contained in:
Christopher Haster
2023-07-18 16:32:49 -05:00
parent 97f867b28d
commit c928ed131f
7 changed files with 270 additions and 175 deletions
+19 -10
View File
@@ -5386,18 +5386,9 @@ static int lfsr_mdir_compact_(lfs_t *lfs, lfsr_mdir_t *mdir,
// - mid = wl => only alloc if mdir is tired (wear-leveling)
// - otherwise => always alloc, use this mid (new mdir)
// consume gstate on original rbyd, we need this even if we drop
// our mdir to avoid losing info
//
// if succesful, this should get immediately appended to our new commit
int err = lfsr_fs_consumegdelta(lfs, msource);
if (err) {
return err;
}
// first thing we need to do is read our current revision count
uint32_t rev;
err = lfsr_bd_read(lfs, msource->rbyd.block, 0, sizeof(uint32_t),
int err = lfsr_bd_read(lfs, msource->rbyd.block, 0, sizeof(uint32_t),
&rev, sizeof(uint32_t));
if (err && err != LFS_ERR_CORRUPT) {
return err;
@@ -5430,6 +5421,19 @@ static int lfsr_mdir_compact_(lfs_t *lfs, lfsr_mdir_t *mdir,
rev = (err != LFS_ERR_CORRUPT ? lfs_fromle32_(&rev) : 0);
}
// only consume gstate here during normal compacts
// TODO avoid duplicate conditions somehow?
if (mid == LFSR_MID_WL) {
// consume gstate on original rbyd, we need this even if we drop
// our mdir to avoid losing info
//
// if succesful, this should get immediately appended to our new commit
int err = lfsr_fs_consumegdelta(lfs, msource);
if (err) {
return err;
}
}
// swap our rbyds
lfs_swap32(&mdir->rbyd.block, &mdir->redund_block);
// update our revision count
@@ -7509,6 +7513,11 @@ static int lfsr_fs_preparemutation(lfs_t *lfs) {
LFS_DEBUG("Fixing grm (%"PRId32".%"PRId32")",
lfs->grm_.mid,
lfs->grm_.rid);
// checkpoint the allocator in case fixing the grm causes mtree
// manipulation
lfs_alloc_ack(lfs);
int err = lfsr_fs_fixgrm(lfs);
if (err) {
return err;