Linux kernel stable tree https://git.rulkc.org/pub/scm/linux/kernel/git/stable/linux.git/atom?h=linux-6.7.y 2023-10-22T21:10:13+00:00 2023-10-22T21:10:13+00:00 Kent Overstreet kent.overstreet@linux.dev 2023-09-12T21:16:02+00:00 urn:sha1:6bd68ec266ad71827ef940151067b67b62fb8fed We're using more stack than we'd like in a number of functions, and btree_trans is the biggest object that we stack allocate. But we have to do a heap allocatation to initialize it anyways, so there's no real downside to heap allocating the entire thing. Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev> 2023-10-22T21:10:12+00:00 Kent Overstreet kent.overstreet@linux.dev 2023-09-10T22:05:17+00:00 urn:sha1:1809b8cba756d32bd6e976ed4ee64efdf66c6d94 More reorganization, this splits up io.c into - io_read.c - io_misc.c - fallocate, fpunch, truncate - io_write.c Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev> 2023-10-22T21:10:12+00:00 Kent Overstreet kent.overstreet@linux.dev 2023-09-11T05:37:34+00:00 urn:sha1:e46c181af9e230c4c5dbc701fdadc295d6191eec Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev> 2023-10-22T21:10:05+00:00 Kent Overstreet kent.overstreet@linux.dev 2023-06-27T21:32:48+00:00 urn:sha1:f33c58fc46a9c5bd6cbf90edb6ce17fa3fd912d5 Now that we have journal watermarks and alloc watermarks unified, BTREE_INSERT_USE_RESERVE is redundant and can be deleted. Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev> 2023-10-22T21:10:05+00:00 Kent Overstreet kent.overstreet@linux.dev 2023-06-26T22:36:24+00:00 urn:sha1:0fb3355d0a3b055af8735fa25b5af63f4dd9a034 bch2_bkey_make_mut() now takes the bkey_s_c by reference and points it at the new, mutable key. This helps in some fsck paths that may have multiple repair operations on the same key. Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev> 2023-10-22T21:10:01+00:00 Kent Overstreet kent.overstreet@linux.dev 2023-04-30T23:21:06+00:00 urn:sha1:dbda63bbb0dbce070f22132339a07146bf1af850 It's safe to call bch2_trans_update with a k/v pair where the value hasn't been filled out, as long as the key part has been and the value is filled out by transaction commit time. This patch folds the bch2_trans_update() call into bch2_bkey_make_mut(), eliminating a bit of boilerplate. Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev> 2023-10-22T21:09:56+00:00 Kent Overstreet kent.overstreet@linux.dev 2023-03-10T21:28:37+00:00 urn:sha1:702ffea204840455e4f2d918538c39cc5c59666b - __bch2_bkey_drop_ptr() -> bch2_bkey_drop_ptr_noerror(), now available outside extents. - Split bch2_bkey_has_device() and bch2_bkey_has_device_c(), const and non const versions - bch2_extent_has_ptr() now returns the pointer it found Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev> 2023-10-22T21:09:48+00:00 Kent Overstreet kent.overstreet@linux.dev 2022-11-24T03:13:19+00:00 urn:sha1:994ba475433a9395cb71e2ffb1928ce9fdb98e80 This introduces some new conveniences, to help cut down on boilerplate: - bch2_trans_kmalloc_nomemzero() - performance optimiation - bch2_bkey_make_mut() - bch2_bkey_get_mut() - bch2_bkey_get_mut_typed() - bch2_bkey_alloc() Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev> 2023-10-22T21:09:37+00:00 Kent Overstreet kent.overstreet@gmail.com 2022-07-18T03:06:38+00:00 urn:sha1:549d173c1bd9b58c2ad41217522462e012a6545f Now that we have error codes, with subtypes, we can switch to our own error code for transaction restarts - and even better, a distinct error code for each transaction restart reason: clearer code and better debugging. Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com> 2023-10-22T21:09:36+00:00 Kent Overstreet kent.overstreet@gmail.com 2022-07-18T23:42:58+00:00 urn:sha1:d4bf5eecd78a90d019b933929a14c91d6d41af62 Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>