Github/littlefs
1
Fork 0
Code Actions Activity

Tweaked compact to use pre-commit inlined state

Browse Source 
This seems more correct and avoids an extra set of inlined state copies.

Win win.
This commit is contained in:
Christopher Haster
2023-09-27 11:55:25 -05:00
parent 488ba4b650
commit 58be838916
1 changed files with 22 additions and 31 deletions
Download Patch File Download Diff File Expand all files Collapse all files
lfs.c
+22 -31
View File
@@ -5408,6 +5408,10 @@ static int lfsr_mdir_compact__(lfs_t *lfs, lfsr_mdir_t *mdir_,
// it's really tempting to deduplicate this via recursion! but we can't // it's really tempting to deduplicate this via recursion! but we can't
// do that here // do that here
// note that any inlined data updates here check the pre-commit state
// (inlined), not the in-flight state (inlined_), this is important,
// we can't trust inlined_ after a failed commit
// copy over tags in the rbyd in order // copy over tags in the rbyd in order
lfsr_srid_t rid = start_rid; lfsr_srid_t rid = start_rid;
lfsr_tag_t tag = 0; lfsr_tag_t tag = 0;
@@ -5448,7 +5452,7 @@ static int lfsr_mdir_compact__(lfs_t *lfs, lfsr_mdir_t *mdir_,
opened; opened;
opened = opened->next) { opened = opened->next) {
lfsr_file_t *file = (lfsr_file_t*)opened; lfsr_file_t *file = (lfsr_file_t*)opened;
if (lfsr_inlined_hasshrub(&file->inlined_) if (lfsr_inlined_hasshrub(&file->inlined)
&& file->inlined.u.data.u.disk.block && file->inlined.u.data.u.disk.block
== data.u.disk.block == data.u.disk.block
&& file->inlined.u.data.u.disk.off && file->inlined.u.data.u.disk.off
@@ -5457,9 +5461,8 @@ static int lfsr_mdir_compact__(lfs_t *lfs, lfsr_mdir_t *mdir_,
// but we have just enough info // but we have just enough info
file->inlined_.u.data = LFSR_DATA_DISK( file->inlined_.u.data = LFSR_DATA_DISK(
mdir_->u.rbyd.block, mdir_->u.rbyd.block,
mdir_->u.rbyd.eoff mdir_->u.rbyd.eoff - lfsr_data_size(&data),
- lfsr_data_size(&file->inlined_.u.data), lfsr_data_size(&data));
lfsr_data_size(&file->inlined_.u.data));
} }
} }
@@ -5507,9 +5510,9 @@ static int lfsr_mdir_compact__(lfs_t *lfs, lfsr_mdir_t *mdir_,
opened; opened;
opened = opened->next) { opened = opened->next) {
lfsr_file_t *file = (lfsr_file_t*)opened; lfsr_file_t *file = (lfsr_file_t*)opened;
if (lfsr_inlined_hasshrub(&file->inlined_) if (lfsr_inlined_hasshrub(&file->inlined)
&& file->inlined_.u.rbyd.block == mdir->u.rbyd.block && file->inlined.u.rbyd.block == mdir->u.rbyd.block
&& file->inlined_.u.rbyd.trunk == shrub.trunk) { && file->inlined.u.rbyd.trunk == shrub.trunk) {
file->inlined_.u.rbyd.block = mdir_->u.rbyd.block; file->inlined_.u.rbyd.block = mdir_->u.rbyd.block;
file->inlined_.u.rbyd.trunk = mdir_->u.rbyd.trunk; file->inlined_.u.rbyd.trunk = mdir_->u.rbyd.trunk;
file->inlined_.u.rbyd.weight = mdir_->u.rbyd.weight; file->inlined_.u.rbyd.weight = mdir_->u.rbyd.weight;
@@ -5546,15 +5549,15 @@ static int lfsr_mdir_compact__(lfs_t *lfs, lfsr_mdir_t *mdir_,
opened = opened->next) { opened = opened->next) {
lfsr_file_t *file = (lfsr_file_t*)opened; lfsr_file_t *file = (lfsr_file_t*)opened;
// inlined data? // inlined data?
if (lfsr_inlined_hassprout(&file->inlined_) if (lfsr_inlined_hassprout(&file->inlined)
&& lfsr_file_isunsynced(file) && lfsr_file_isunsynced(file)
&& file->inlined_.u.data.u.disk.block == mdir->u.rbyd.block && file->inlined.u.data.u.disk.block == mdir->u.rbyd.block
&& (file->m.mdir.mid & lfsr_midrmask(lfs)) >= start_rid && (file->m.mdir.mid & lfsr_midrmask(lfs)) >= start_rid
&& (lfs_size_t)(file->m.mdir.mid & lfsr_midrmask(lfs)) && (lfs_size_t)(file->m.mdir.mid & lfsr_midrmask(lfs))
< (lfs_size_t)end_rid) { < (lfs_size_t)end_rid) {
// write the data as a shrub tag // write the data as a shrub tag
err = lfsr_rbyd_appendcompactattr(lfs, &mdir_->u.rbyd, err = lfsr_rbyd_appendcompactattr(lfs, &mdir_->u.rbyd,
LFSR_TAG_SHRUB(INLINED), 0, file->inlined_.u.data); LFSR_TAG_SHRUB(INLINED), 0, file->inlined.u.data);
if (err) { if (err) {
LFS_ASSERT(err != LFS_ERR_RANGE); LFS_ASSERT(err != LFS_ERR_RANGE);
return err; return err;
@@ -5565,13 +5568,13 @@ static int lfsr_mdir_compact__(lfs_t *lfs, lfsr_mdir_t *mdir_,
file->inlined_.u.data = LFSR_DATA_DISK( file->inlined_.u.data = LFSR_DATA_DISK(
mdir_->u.rbyd.block, mdir_->u.rbyd.block,
mdir_->u.rbyd.eoff mdir_->u.rbyd.eoff
- lfsr_data_size(&file->inlined_.u.data), - lfsr_data_size(&file->inlined.u.data),
lfsr_data_size(&file->inlined_.u.data)); lfsr_data_size(&file->inlined.u.data));
// inlined tree? // inlined tree?
} else if (lfsr_inlined_hasshrub(&file->inlined_) } else if (lfsr_inlined_hasshrub(&file->inlined)
&& lfsr_file_isunsynced(file) && lfsr_file_isunsynced(file)
&& file->inlined_.u.rbyd.block == mdir->u.rbyd.block && file->inlined.u.rbyd.block == mdir->u.rbyd.block
&& (file->m.mdir.mid & lfsr_midrmask(lfs)) >= start_rid && (file->m.mdir.mid & lfsr_midrmask(lfs)) >= start_rid
&& (lfs_size_t)(file->m.mdir.mid & lfsr_midrmask(lfs)) && (lfs_size_t)(file->m.mdir.mid & lfsr_midrmask(lfs))
< (lfs_size_t)end_rid) { < (lfs_size_t)end_rid) {
@@ -5670,16 +5673,16 @@ static lfs_ssize_t lfsr_mdir_estimate_(lfs_t *lfs, const lfsr_mdir_t *mdir,
opened = opened->next) { opened = opened->next) {
lfsr_file_t *file = (lfsr_file_t*)opened; lfsr_file_t *file = (lfsr_file_t*)opened;
// inlined data? // inlined data?
if (lfsr_inlined_hassprout(&file->inlined_) if (lfsr_inlined_hassprout(&file->inlined)
&& lfsr_file_isunsynced(file) && lfsr_file_isunsynced(file)
&& file->inlined_.u.data.u.disk.block == mdir->u.rbyd.block && file->inlined.u.data.u.disk.block == mdir->u.rbyd.block
&& (file->m.mdir.mid & lfsr_midrmask(lfs)) == rid) { && (file->m.mdir.mid & lfsr_midrmask(lfs)) == rid) {
dsize += lfsr_data_estimate(&file->inlined_.u.data); dsize += lfsr_data_estimate(&file->inlined.u.data);
// inlined tree? // inlined tree?
} else if (lfsr_inlined_hasshrub(&file->inlined_) } else if (lfsr_inlined_hasshrub(&file->inlined)
&& lfsr_file_isunsynced(file) && lfsr_file_isunsynced(file)
&& file->inlined_.u.rbyd.block == mdir->u.rbyd.block && file->inlined.u.rbyd.block == mdir->u.rbyd.block
&& (file->m.mdir.mid & lfsr_midrmask(lfs)) == rid) { && (file->m.mdir.mid & lfsr_midrmask(lfs)) == rid) {
lfs_ssize_t dsize_ = lfsr_rbyd_estimate(lfs, lfs_ssize_t dsize_ = lfsr_rbyd_estimate(lfs,
&file->inlined.u.rbyd, -1, -1, NULL); &file->inlined.u.rbyd, -1, -1, NULL);
@@ -5776,18 +5779,6 @@ static int lfsr_mdir_commit_(lfs_t *lfs, lfsr_mdir_t *mdir,
compact:; compact:;
// can't commit, try to compact // can't commit, try to compact
// before we do anything else we need to revert any changes to inlined
// files in this mdir
for (lfsr_openedmdir_t *opened = lfs->opened[
LFS_TYPE_REG-LFS_TYPE_REG];
opened;
opened = opened->next) {
lfsr_file_t *file = (lfsr_file_t*)opened;
if (lfsr_inlined_block(&file->inlined_) == mdir->u.rbyd.block) {
file->inlined_ = file->inlined;
}
}
// check if we're within our compaction threshold // check if we're within our compaction threshold
lfs_ssize_t estimate = lfsr_mdir_estimate(lfs, mdir, start_rid, end_rid, lfs_ssize_t estimate = lfsr_mdir_estimate(lfs, mdir, start_rid, end_rid,
split_rid_); split_rid_);