Github/littlefs
1
Fork 0
Code Actions Activity

Added big vestigial-name-split comment

Browse Source 
This is the _nth_ time I've tried to force arbitrary btree name inserts
to work, so _clearly_ I need a bigger comment.

Hopefully this will prevent me from trying to delete the LFSR_RATTR_NOOP
in test_btree_find_general_fuzz _again_.

---

The gist is that insert-before-bid+1 is fundamentally different from
insert-after-bid when named btrees are involved:

    .-----f-----.    insert-after-d     .-------f-----.
  .-b--.     .--j-.        =>         .-b---.      .--j-.
  |   .-.   .-.   |                   |   .---.   .-.   |
  a   c d   h i   k                   a   c d e   h i   k
                                              ^
                     insert-before-h
                           =>           .-----f-------.
                                      .-b--.      .---j-.
                                      |   .-.   .---.   |
                                      a   c d   g h i   k
                                                ^

The problem is that lfsr_btree_commit_ needs to find the same leaf
rbyd as lfsr_btree_namelookup, and potentially insert-before the
first rid or insert-after the last rid.

Instead of separate insert-before/after flags, we make the first tag
in a commit insert-before, and all following non-grow tags
insert-after (splits).

This info is now captured in the above mentioned comment.
This commit is contained in:
Christopher Haster
2025-04-28 17:30:05 -05:00
parent 677c078b50
commit 27dd339a6a
2 changed files with 77 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tests/test_btree.toml
+47 -6
View File
@@ -3866,6 +3866,28 @@ code = '''
}
// split btree
//
// note! all name updates _must_ be via splits (except for
// the first one)
//
// This is because our btrees contain vestigial names, i.e.
// our inner nodes may contain names no longer in the tree.
// This simplifies lfsr_btree_commit_, but means
// insert-before-bid+1 is _not_ the same as insert-after-bid
// when named btrees are involved. If you try this it _will
// not_ work and if try to make it work you _will_ cry:
//
// .-----f-----. insert-after-d .-------f-----.
// .-b--. .--j-. => .-b---. .--j-.
// | .-. .-. | | .---. .-. |
// a c d h i k a c d e h i k
// ^
// insert-before-h
// => .-----f-------.
// .-b--. .---j-.
// | .-. .---. |
// a c d g h i k
// ^
lfsr_bid_t split_bid;
lfs_scmp_t cmp = lfsr_btree_namelookupleaf(&lfs, &btree,
0, name, 3,
@@ -3882,9 +3904,7 @@ code = '''
&(uint8_t){'0'+(i % 10)}, 1))) => 0;
} else {
lfsr_btree_commit(&lfs, &btree, split_bid, LFSR_RATTRS(
// TODO can we avoid this noop? the problem is we need
// some way to differentiate inserting before vs after
// the requested bid
// yes, we need this noop, see above
LFSR_RATTR_NOOP(),
LFSR_RATTR_NAME(
LFSR_TAG_REG, +1,
@@ -4038,6 +4058,29 @@ code = '''
}
// split btree
//
// note! all name updates _must_ be via splits (except for
// the first one)
//
// This is because our btrees contain vestigial names, i.e.
// our inner nodes may contain names no longer in the tree.
// This simplifies lfsr_btree_commit_, but means
// insert-before-bid+1 is _not_ the same as insert-after-bid
// when named btrees are involved. If you try this it _will
// not_ work and if try to make it work you _will_ cry:
//
//
// .-----f-----. insert-after-d .-------f-----.
// .-b--. .--j-. => .-b---. .--j-.
// | .-. .-. | | .---. .-. |
// a c d h i k a c d e h i k
// ^
// insert-before-h
// => .-----f-------.
// .-b--. .---j-.
// | .-. .---. |
// a c d g h i k
// ^
lfsr_bid_t split_bid;
lfsr_bid_t split_weight;
lfs_scmp_t cmp = lfsr_btree_namelookupleaf(&lfs, &btree,
@@ -4056,9 +4099,7 @@ code = '''
&(uint8_t){'0'+(i % 10)}, 1))) => 0;
} else {
lfsr_btree_commit(&lfs, &btree, split_bid, LFSR_RATTRS(
// TODO can we avoid this noop? the problem is we need
// some way to differentiate inserting before vs after
// the requested bid
// yes, we need this noop, see above
LFSR_RATTR_NOOP(),
LFSR_RATTR_NAME(
LFSR_TAG_REG, +weight,