IF YOU WOULD LIKE TO GET AN ACCOUNT, please write an
email to Administrator. User accounts are meant only to access repo
and report issues and/or generate pull requests.
This is a purpose-specific Git hosting for
BaseALT
projects. Thank you for your understanding!
Только зарегистрированные пользователи имеют доступ к сервису!
Для получения аккаунта, обратитесь к администратору.
All code using the in-memory bucket array, excluding GC, has now been
converted to use the alloc btree directly - so we can finally delete it.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
This is one of the last steps in getting rid of the main in-memory
bucket array.
This changes bch2_dev_usage_update() to take bkey_alloc_unpacked instead
of bucket_mark, and for the places where we are in fact working with
bucket_mark and don't have bkey_alloc_unpacked, we add a wrapper that
takes bucket_mark and converts to bkey_alloc_unpacked.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
In the old allocator code, preparing an existing empty bucket was part
of the same code path that invalidated buckets containing cached data.
In the new allocator code this is no longer the case: the main allocator
path finds empty buckets (via the new freespace btree), and can't
allocate buckets that contain cached data.
We now need a separate code path to invalidate buckets containing cached
data when we're low on empty buckets, which this patch implements. When
the number of free buckets decreases that triggers the new invalidate
path to run, which uses the LRU btree to pick cached data buckets to
invalidate until we're above our watermark.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
In the old allocator code, buckets would be discarded just prior to
being used - this made sense in bcache where we were discarding buckets
just after invalidating the cached data they contain, but in a
filesystem where we typically have more free space we want to be
discarding buckets when they become empty.
This patch implements the new behaviour - it checks the need_discard
btree for buckets awaiting discards, and then clears the appropriate
bit in the alloc btree, which moves the buckets to the freespace btree.
Additionally, discards are now enabled by default.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
Now that we have new persistent data structures for the allocator, this
patch converts the allocator to use them.
Now, foreground bucket allocation uses the freespace btree to find
buckets to allocate, instead of popping buckets off the freelist.
The background allocator threads are no longer needed and are deleted,
as well as the allocator freelists. Now we only need background tasks
for invalidating buckets containing cached data (when we are low on
empty buckets), and for issuing discards.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
This adds two new btrees for the upcoming allocator rewrite: an extents
btree of free buckets, and a btree for buckets awaiting discards.
We also add a new trigger for alloc keys to keep the new btrees up to
date, and a compatibility path to initialize them on existing
filesystems.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
This introduces a new alloc key which doesn't use varints. Soon we'll be
adding backpointers and storing them in alloc keys, which means our
pack/unpack workflow for alloc keys won't really work - we'll need to be
mutating alloc keys in place.
Instead of bch2_alloc_unpack(), we now have bch2_alloc_to_v4() that
converts older types of alloc keys to v4 if needed.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
This implements new persistent LRUs, to be used for buckets containing
cached data, as well as stripes ordered by time when a block became
empty.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Add a new superblock field which represents journal buckets as ranges:
also move code for the superblock journal fields to journal_sb.c.
This also reworks the code for resizing the journal to write the new
superblock before using the new journal buckets, and thus be a bit
safer.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
In the write path, after the write to the block device(s) complete we
have to punt to process context to do the btree update.
Instead of using the work item embedded in op->cl, this patch switches
to a per write-point work item. This helps with two different issues:
- lock contention: btree updates to the same writepoint will (usually)
be updating the same alloc keys
- context switch overhead: when we're bottlenecked on btree updates,
having a thread (running out of a work item) checking the write point
for completed ops is cheaper than queueing up a new work item and
waking up a kworker.
In an arbitrary benchmark, 4k random writes with fio running inside a
VM, this patch resulted in a 10% improvement in total iops.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
This simplifies the logic in bch2_btree_update_start() a bit, handling
the unlock/block logic more locally.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
Since journal reclaim -> btree key cache flushing may require the
allocation of new btree nodes, it has an implicit dependency on copygc
in order to make forward progress - so we should avoid blocking copygc
unless the journal is really close to full.
This introduces watermarks to replace our single MAY_GET_UNRESERVED bit
in the journal, and adds a watermark for copygc and plumbs it through.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
We don't actually want copygc allocations to be nowait - an allocation
for copygc might fail and then later succeed due to a bucket needing to
wait on journal commit, or to be discarded.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
When bch2_journal_pin_set() is updating an existing pin, we shouldn't
call bch2_journal_reclaim_fast() after dropping the old pin and before
dropping the new pin - that could reclaim the entry we're trying to pin.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
For backpointers, we'll need to delete old backpointers before adding
new backpointers - otherwise we'll run into spurious duplicate
backpointer errors.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
For backpointers, we need to switch the order triggers are run in: we
need to run triggers for deletions/overwrites before triggers for
inserts.
To avoid breaking the reflink triggers, this patch moves deleting of
indirect extents with refcount=0 to their triggers, instead of doing it
when we update those keys.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Print bucket:offset when the filesystem is online; this makes debugging
easier when correlating with alloc updates.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Add a new helper for logging messages to the journal - a new debugging
tool, an alternative to trace_printk().
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This fixes a bug where __bch2_btree_node_update_key() wasn't clearing
should_be_locked, leading to bch2_btree_path_traverse() always failing -
all callers of btree_path_make_mut() want should_be_locked cleared, so
do it there.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
bch2_btree_iter_next_node() was mucking with other btree_path state
without setting path->update to be consistent with the fact that the
path is very much no longer uptodate - oops.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
bch2_journal_space_available -> bch2_journal_halt() self deadlocks on
journal lock; work around this by dropping/retaking journal lock before
we call bch2_fatal_error().
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
When deleting an entry from a heap that was at entry h->used - 1, we'd
end up calling heap_sift() on an entry outside the heap - the entry we
just removed - which would end up re-adding it to the heap and deleting
something we didn't want to delete. Oops...
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
We've been seeing a very strange bug where journal flush & reclaim delay
end up getting inexplicably zeroed, in the superblock. We're now
validating all the options in bch2_validate_super(), and 0 is no longer
a valid value for those options, but we need to be careful not to
prevent people's filesystems from mounting because of the new
validation.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Something funny is going on with the new code for restoring the journal
write point, and it's hard to reproduce.
We do want to debug this because resuming writing to the journal in the
wrong spot could be something serious. For now, replace the assertion
with an error message and revert to old behaviour when it happens.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
We're seeing a very strange bug where journal_flush_delay sometimes gets
set to 0 in the superblock. Together with the preceding patch, this
should help us track it down.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This moves validation of superblock options to bch2_sb_validate(), so
they'll be checked in the write path as well.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Now we've got strings for metadata versions - this changes
bch2_sb_to_text() and our mount log message to use it.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Previously, we'd go into an infinite loop when attempting to cache a
bkey in the key cache larger than 128 u64s - since we were only using a
u8 for the size field, it'd get rounded up to 256 then truncated to 0.
Oops.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
These warnings are symptomatic of something else going wrong, we don't
want them spamming up the logs as that'll make it harder to find the
real issue.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This patch tweaks the journal recovery path so that we start writing
right after where we left off, instead of the next empty bucket. This is
partly prep work for supporting zoned devices, but it's also good to do
in general to avoid the journal completely filling up and getting stuck.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
BTREE_TRIGGER_WANTS_OLD_AND_NEW didn't work correctly when the old and
new key were both alloc keys, but different versions - it required old
and new key type to be identical, and this bug is a problem for the new
allocator rewrite.
This patch fixes it by checking if the old and new key have the same
trigger functions - the different versions of alloc (and inode) keys
have the same trigger functions.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This replaces the switch statements in bch2_mark_key(),
bch2_trans_mark_key() with new bkey methods - prep work for the next
patch, which fixes BTREE_TRIGGER_WANTS_OLD_AND_NEW.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
We recently started stashing a copy of the key being overwritten in
btree_insert_entry: this is helpful for avoiding multiple calls to
bch2_btree_path_peek_slot() and bch2_journal_keys_peek() in the
transaction commit path.
But it turns out this has a problem - when we run mem/atomic triggers,
we've done a couple things that can invalidate the pointer to the old
key's value. This makes the optimization of stashing a pointer to the
old value questionable, but for now this patch revalidates that pointer
before running mem triggers.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This turns bch2_dump_trans_updates() into a to_text() method - this way
it can be used by debug tracing.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
Add a new error macro that also dumps transaction updates in addition to
doing an emergency shutdown - when a transaction update discovers or is
causing a fs inconsistency, it's helpful to see what updates it was
doing.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
As we've already reserved space in the journal this optimization doesn't
actually buy us anything, and when doing list_journal debugging it
deletes information we want.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
In BTREE_ITER_FILTER_SNAPHOTS mode, we skip over keys in unrelated
snapshots. When we hit the end of an inode, if the next inode(s) are in
a different subvolume, we could potentially have to skip past many keys
before finding a key we can return to the caller, so they can terminate
the iteration.
This adds a peek_upto() variant to solve this problem, to be used when
we know the range we're searching within.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This tweaks __bch2_set_nr_journal_buckets() so that we aren't reversing
their order in the jorunal anymore - nice for rotating disks.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This adds bch2_journal_log_msg(), which just logs a message to the
journal, and uses it to mark startup and when journal replay finishes.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
It wasn't used as iter_flags (excepting the unit tests, which this patch
fixes), and the next patch is going to need to pass in
BTREE_TRIGGER_NORUN.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
traverse_all() traverses btree paths in sorted order, so it should never
see transaction restarts due to lock ordering violations. But some code
in __bch2_btree_path_upgrade(), while necessary when not running under
traverse_all(), was causing some confusing lock ordering violations -
disabling this code under traverse_all() will let us put in some more
assertions.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
In btree_path_traverse_all() we were failing to check for -EIO in the
retry loop, and after btree node read error we'd go into an infinite
loop.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
When the iov_iter is a bvec iter, it's possible the IO was submitted
from a kthread that didn't have an mm to switch to.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This fixes a bug in the DIO read path where, when using a loopback
device in DIO mode, we'd allocate a biovec that would get overwritten
and leaked in bio_iov_iter_get_pages() -> bio_iov_bvec_set().
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
"bcachefs: Log & error message improvements" accidentally changed the
format specifier we use for converting UUIDs to strings, which broke
mounting of encrypted filesystems - this patch reverts that change.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
pr_tab_rjust() was broken and leaving a null somewhere in the output
string - this patch fixes it and simplifies it a bit.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
When bch2_trans_begin() is called and there hasn't been a transaction
restart, we presume that we're now doing something new - iterating over
different keys, and we now shouldn't keep aruond paths related to the
previous transaction, excepting the subvolumes btree.
This should fix some of our "transaction path overflow" bugs.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This fixes a shutdown race where we were rearming journal->write_work
after the journal has already shut down.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Options no longer have to be manually added to bch2_sb_to_text() - it
now uses the master list of options in opts.h. Also, improve some of the
formatting by converting it to tabstops.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
We were accidentally using default mount options and overwriting the
discard flag.
Signed-off-by: Daniel Hill <daniel@gluo.nz>
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
Six locks have a percpu mode, which we use for interior btree nodes, as
well as btree key cache keys for the subvolumes btree. We've been
switching locks back and forth between percpu and non percpu mode as
needed, but it turns out this is racy - when we're reusing an existing
node, other threads could be attempting to lock it while we're switching
it between modes.
This patch fixes this by never switching 'struct btree' between the two
modes, and instead segragating them between two different freed lists.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
This is prep work for the next patch, which is going to fix our usage of
the percpu mode of six locks by never switching struct btree between the
two modes - which means we need separate freed lists.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
We don't need to pass the number of nodes required to
bch2_btree_update_start, just whether we're doing a split at @level.
This is prep work for a fix to our usage of six lock's percpu mode,
which is going to require us to count up and allocate interior nodes and
leaf nodes seperately.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
Previously, when bch2_btree_cache_scan() attempted to reclaim a node but
failed (because trylock failed, because it was dirty, etc.), it would
count that against the number of nodes it was scanning and attempting to
free. This patch changes that behaviour, so that now we only count nodes
that we then don't free if they have the accessed bit (which we also
clear).
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
After emergency shutdown, all journal entries will be written as noflush
entries, meaning they will never be used - but they'll still exist for
debugging tools to examine.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Previous patch just moved responsibility for incrementing the journal
sequence number and initializing the new journal entry from
__journal_entry_close() to journal_entry_open(); this patch makes the
analagous change for journal reservation state, incrementing the index
into array of journal_bufs at open time.
This means that __journal_entry_close() never fails to close an open
journal entry, which is important for the next patch that will change
our emergency shutdown behaviour.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
It makes the code more readable if we work off of sequence numbers,
instead of direct indexes into the array of journal buffers.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This patch changes journal_entry_open() to initialize the new journal
entry, not __journal_entry_close().
This also means that journal_cur_seq() refers to the sequence number of
the last journal entry when we don't have an open journal entry, not the
next one.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
When we do an interior btree update, we create new btree nodes and link
them into the btree in memory, but they don't become reachable on disk
until later, when btree_update_nodes_written_trans() runs.
Updates to the new nodes can thus happen before they're reachable on
disk, and if the updates to those new nodes are written before the nodes
become reachable, we would then drop the journal pin for those updates
before the btree has them.
This is what the journal pin in bch2_btree_update_start() was protecting
against. However, it's not actually needed because we don't allow
subsequent append writes to btree nodes until the node is reachable on
disk.
Dropping this unneeded pin also fixes a bug introduced by "bcachefs:
Journal seq now incremented at entry open, not close" - in the new code,
if the journal is completely empty a journal pin list for
journal_cur_seq() won't exist.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This change is prep work for moving some work from
__journal_entry_close() to journal_entry_open(): without this change,
journal_entry_open() doesn't know if it's going to be able to open a new
journal entry until the cmpxchg loop, meaning it can't create the new
journal pin entry and update other global state because those have to be
done prior to the cmpxchg opening the new journal entry.
Fortunately, we don't call bch2_journal_halt() from interrupt context.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This replaces the journal flag JOURNAL_NEED_WRITE with per-journal buf
state - more explicit, and solving a race in the old code that would
lead to entries being opened and written unnecessarily.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This fixes a regression from "bcachefs: Stash a copy of key being
overwritten in btree_insert_entry". In btree_key_can_insert_cached(), we
may reallocate the key cache key, invalidating pointers previously
returned by peek() - fix it by issuing a transaction restart.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This fixes a regression from "bcachefs: Heap allocate printbufs" -
bch2_sb_field_validate() was leaking an error string.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Checking btree_node_may_write() isn't atomic with the other btree flags,
dirty and need_write in particular. There was a rare race where we'd
unblock a node from writing while __btree_node_flush() was setting
need_write, and no thread would notice that the node was now both able
to write and needed to be written.
Fix this by adding btree node flags for will_make_reachable and
write_blocked that can be checked in the cmpxchg loop in
__bch2_btree_node_write.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
bch2_btree_node_write_cond() was only used in one place - this inlines
it into __btree_node_flush() and makes the cmpxchg loop actually
correct.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
btree_node_write_if_need() kicks off a btree node write only if
need_write is set; this makes the locking easier to reason about by
moving the check into the cmpxchg loop in __bch2_btree_node_write().
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
There was a rare recursive locking bug, in __bch2_btree_node_write()
nowrite path -> btree_node_write_done(), in the path that kicks off
another write.
This splits out an inner __btree_node_write_done() that expects to be
run with the btree node lock held.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
In sysfs, files can only output at most PAGE_SIZE. This is a problem for
debug info that needs to list an arbitrary number of times, and because
of this limit some of our debug info has been terser and harder to read
than we'd like.
This patch moves info about journal pins and cached btree nodes to
debugfs, and greatly expands and improves the output we return.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This cacheline aligns struct journal, and puts j->reservations and
j->prereserved on their own cacheline - we may want to split them up in
a separate patch.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
In btree_update_interior.c, we were changing a path's level directly -
which affects path sort order - without re-sorting paths, leading to
assertions when bch2_path_get() verified paths were sorted correctly.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
journal_flush_done() was overwriting did_work, thus occasionally
returning false when it did do work and occasional assertions in the
shutdown sequence because we didn't completely flush the key cache.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This patch changes printbufs dynamically allocate and reallocate a
buffer as needed. Stack usage has become a bit of a problem, and a major
cause of that has been static size string buffers on the stack.
The most involved part of this refactoring is that printbufs must now be
exited with printbuf_exit().
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
We're hitting a strange bug with transaction paths not being sorted
correctly - this dumps transaction paths in the order we thought was
sorted, which will hopefully shed some light as to what's going on.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
When key cache pins were put onto their own list, we neglected to update
bch2_journal_pins_to_text() to print them.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
bch2_trans_begin() invalidates all iterators, until they're revalidated
by calling peek() or traverse().
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Triggers can generate additional btree updates - we need to run alloc
triggers after all other triggers have run, because they generate
updates for the alloc btree.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Upcoming patches are doing more work on the triggers code, this patch
just moves code around.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
We're now coming up with triggers that modify the update being done. A
bkey_s_c is const - bkey_i is the correct type to be using here.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
vstruct_bytes() was returning a u64 - it should be a size_t, the corect
type for the size of anything that fits in memory.
Also replace a 64 bit divide with div_u64().
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
On btree node read error, it's helpful to see what we were trying to
read - was it all zeroes?
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
When a reflink pointer points to a missing indirect extent, we replace
it with an error key. Instead of replacing the entire reflink pointer
with an error key, this patch replaces only the missing range with an
error key.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This improves __bch2_trans_commit - early in the recovery process, when
we're running btree_gc and before we want to go RW, it now uses
bch2_journal_key_insert() to add the update to the list of updates for
journal replay to do, instead of btree_gc having to use separate
interfaces depending on whether we're running at bringup or, later,
runtime.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This patch was originally to work around the journal geting stuck in
nochanges mode - but that was just a hack, we needed to fix the actual
bug. It should be fixed now, so revert it.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
The journal can get stuck if we need to get a journal reservation for
something we have a pre-reservation for, but aren't able to reclaim
space, or if the pin fifo is full - it's impractical to resize the pin
fifo at runtime.
Previously, we reserved 8 entries in the pin fifo for pre-reservations,
but that seems small - we're seeing the journal occasionally get stuck.
Let's reserve a quarter of it.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
BTREE_NODE_SEQ() is supposed to give us a time ordering of btree nodes
on disk, so that we can tell which btree node is newer if we ever have
to scan the entire device to find btree nodes.
The btree node merge path wasn't setting it correctly on the new node -
oops.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Long ago it was possible to get a journal reservation and not use it,
but that's no longer allowed, which means journal_write_compact() has
very little work to do, and isn't really worth the code anymore.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This helps with lock contention in the journalling code: instead of
updating our journal pin on every write, only get a journal pin if we
don't have one.
This means we can avoid hammering on journal locks nearly so much, at
the cost of carrying around a journal pin for an older entry than the
one we actually need. To handle that, if needed we update our journal
pin to the correct one when flushed by journal reclaim.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
Now, when outputting to printbufs, we can set tabstops and left or right
justify text to them - this is to be used by the userspace 'bcachefs fs
usage' command.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
In move_read_endio, we were checking if the next pending write has its
read completed - but this can turn after a use after free (and we were
accessing the list without a lock), so instead just better to just
unconditionally do the wakeup.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This patch improves the superblock .to_text() methods and adds methods
for all types that were missing them. It also improves printbufs by
allowing them to specfiy what units we want to be printing in, and adds
new wrapper methods for unifying our kernel and userspace environments.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
bch_scnmemcpy was for printing length-limited strings that might not
have a terminating null - turns out sprintf & pr_buf can do this with
%.*s.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
When the nochanges option is selected, we're supposed to never issue
writes. Unfortunately, it seems discards were missed when implemnting
this, leading to some painful filesystem corruption.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Add an option that tells recovery to only read the journal, to be used
by the list_journal command.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This is prep work for the next patch, which is going to change
__bch2_trans_commit() to use bch2_journal_key_insert() when very early
in the recovery process, so that we have a unified interface for doing
btree updates.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
When viewing what's in the journal, it's more useful to have the logical
location - journal bucket and offset within that bucket - than just the
offset on that device.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Apparently it actually is possible for crypto_skcipher_encrypt() to
return an error - not sure why that would be - but we need to replace
our assertion with actual error handling.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
The error code when we fail to allocate a node in the btree node cache
doesn't make it to bch2_btree_path_traverse_all(). Instead, we need to
stash a flag in btree_trans so we know we have to take the cannibalize
lock.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
bch2_dev_lookup() is used from the extended attribute set methods, for
setting the target options, where we're already holding an inode lock -
it turns out pathname lookups also take inode locks, so that was
susceptible to deadlocks.
Fortunately we already stash the device name in ca->name. This does
change user-visible behaviour though: instead of specifying e.g.
/dev/sda1, user must now specify sda1.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Before we had dedicated gc code for bucket->oldest_gen this was
btree_gc's responsibility, but now that we have that we can rip it out,
simplifying the already overcomplicated btree_gc.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This improves the formatting of journal_entry_btree_keys_to_text() by
putting each key on its own line.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
The loop that traverses paths in traverse_all() needs to be a little bit
tricky, because traversing a path can cause other paths to be added (or
perhaps removed) at about the same position.
The old logic was buggy, replace it with simpler logic.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Some of our tracepoints were calling snprintf("pS") - which does symbol
table lookups - in TP_fast_assign(), which turns out to be a really bad
idea.
This was done because perf trace wasn't correctly printing tracepoints
that use %pS anymore - but it turns out trace-cmd does handle it
correctly.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Since we retry reads when we discover we read from a pointer that went
stale, if a dirty pointer is erroniously stale it would cause us to loop
retrying that read forever - unless we check before issuing the read,
while the btree is still locked, when we know that a dirty pointer
should never be stale.
This patch adds that check, along with printing some helpful debug info.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
__bch2_btree_node_lock() was implementing the wrong lock ordering for
cached vs. non cached paths - this fixes it to match the btree path sort
order as defined by __btree_path_cmp(), and also simplifies the code
some.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This consolidates some of the btree node lock path, so that when we're
blocked taking a write lock on a node it shows up in
bch2_btree_trans_to_text(), along with intent and read locks.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
We were emitting two trace events on transaction restart in this code
path - delete the redundant one.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
We need to ensure we don't have any btree locks held when calling
do_pending_writes() - besides issuing IOs, upcoming allocator changes
will have allocations doing btree lookups directly.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
The old .debugcheck methods are no more and this just calls the .invalid
method, which doesn't add much since we already check that when doing
btree updates and when reading metadata in.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
The check_dirents pass handles transaction restarts at the toplevel -
check_subdir_count() was incorrectly handling transaction restarts
without returning -EINTR, meaning that the iterator pointing to the
dirent being checked was left invalid.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
The reflink repair code was incorrectly inserting a nonzero deleted key
via journal replay - this is due to bch2_journal_key_insert() being
somewhat hacky, and so this fix is also hacky for now.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Like the previous patches, this converts bch2_gc_gens() to use the alloc
btree directly, and private arrays of generation numbers for its own
recalculation of oldest_gen.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This converts the copygc code to use the alloc btree directly to find
buckets that need to be evacuated instead of the in-memory bucket array,
which is finally going away soon.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This changes the btree_gc code to only use the second bucket array, the
one dedicated to GC. On completion, it compares what's in its in memory
bucket array to the allocation information in the btree and writes it
directly, instead of updating the main in-memory bucket array and
writing that.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
- Updates to non key cache iterators will now be transparently
redirected to the key cache for cached btrees.
- Except when creating new keys: then the update goes to underlying
btree
For for iterating over a cached btree to work, we need to ensure that if
a key exists in the key cache, it also exists in the btree - otherwise
the iterator code will skip past it and not check the key cache.
Otherwise, for consistency, all updates should go to the same place -
the key cache.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
This is the start of cache coherency with the btree key cache - this
adds a btree iterator flag that causes lookups to also check the key
cache when we're iterating over the btree (not iterating over the key
cache).
Note that we could still race with another thread creating at item in
the key cache and updating it, since we aren't holding the key cache
locked if it wasn't found. The next patch for the update path will
address this by causing the transaction to restart if the key cache is
found to be dirty.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
Previously, when doing updates and running triggers before journal
replay completes, triggers would see the incorrect key for the old key
being overwritten - this patch updates the trigger code to check the
journal keys when necessary, needed for the upcoming allocator rewrite.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
We currently need to call bch2_btree_path_peek_slot() multiple times in
the transaction commit path - and some of those need to be updated to
also check the keys from journal replay, too. Let's consolidate this and
stash the key being overwritten in btree_insert_entry.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Add a new helper that returns true if the given btree ID uses the btree
key cache. This enables some new cleanups, since the helper can check
the options for whether caching is enabled on a given btree.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
btree_key_cache_flush_pos() uses BTREE_ITER_CACHED_NOFILL - but it
wasn't checking for !ck->valid. It does check for the entry being dirty,
so it shouldn't matter, but this refactor it a bit and adds and
assertion.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
We were double-freeing old_buckets and not freeing old_buckets_gens:
also, the code was supposed to free buckets, not old_buckets;
old_buckets is only needed because we have to use rcu_assign_pointer()
instead of swap(), and won't be set if we hit the error path.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
These nodes aren't reachable by other threads, so there's no need to
keep it locked - and this fixes a bug with the assertion in
bch2_trans_unlock() firing on transaction restart.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Change the error messages in bch2_inconsistent_error() and
bch2_fatal_error() so we can distinguish them.
Also, prefer bch2_fs_fatal_error() (which also logs an error message) to
bch2_fatal_error(), and change a call to bch2_inconsistent_error() to
bch2_fatal_error() when we can't continue.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Previously, bucket fragmentation was considered to be bucket size -
total amount of live data, both dirty and cached.
This meant that if a bucket was full but only a small amount of data in
it was dirty - the rest cached, we'd get stuck: copygc wouldn't move the
dirty data out of the bucket and the allocator wouldn't be able to
invalidate and drop the cached data.
This changes fragmentation to exclude cached data, so that copygc will
evacuate these buckets and copygc/the allocator will always be able to
make forward progress.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
More prep work for getting rid of the in-memory bucket array: now that
we have BTREE_ITER_WITH_JOURNAL, the allocator code can do ntree lookups
before journal replay is finished, and there's no longer any need for it
to get allocation information from the in-memory bucket array.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
The allocator thread invalidates buckets (increments their generation
number) prior to discarding them and putting them on freelists. We've
had a short circuit path for some time to only update the in-memory
bucket mark when doing the invalidate if we're not invalidating cached
data, but that short-circuit path hasn't really been needed for quite
some time (likely since the btree key cache code was added).
We're deleting it now as part of deleting/converting code that uses the
in memory bucket array.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
BTREE_INSERT_LAZY_RW shouldn't do anything after the filesystem has
finished starting up - otherwise, it might interfere with going
read-only as part of shutting down.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
We weren't checking for -EINTR in the main loop in __bch2_move_data -
this code predates modern transaction restarts.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Had a bug report that implies bch2_inode_delete_keys() returned -EINTR
before it completed, so this patch simplifies it and makes the flow
control a little more conventional.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
With BTREE_ITER_FILTER_SNAPSHOTS, we have to distinguish between the
path where the key was found, and the path for inserting into the
current snapshot. This adds a new field to struct btree_iter for saving
a path for the current snapshot, and plumbs it through
bch2_trans_update().
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
This splits bch2_btree_iter() up into two functions: an inner function
that handles BTREE_ITER_WITH_JOURNAL, BTREE_ITER_WITH_UPDATES, and
iterating acrcoss leaf nodes, and an outer one that implements
BTREE_ITER_FILTER_SNAPHSOTS.
This is prep work for remember a btree_path at our update position in
BTREE_ITER_FILTER_SNAPSHOTS mode.
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
This improves the transaction restart tracepoints - adding distinct
tracepoints for all the locations and reasons a transaction might have
been restarted, and ensures that there's a tracepoint for every
transaction restart.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This still needs to be expanded more, but this adds a basic test for
BTREE_ITER_FILTER_SNAPSHOTS.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
bch2_btree_update_key() is used in the btree node write path - before
delivering the completion we have to update the parent pointer with the
number of sectors written.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This breaks bch2_trans_commit_run_triggers() up into multiple functions,
and deletes a bit of duplication - prep work for triggers on alloc keys,
which will need to run last.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Implement a hash table, using cuckoo hashing, for empty buckets that are
waiting on a journal commit before they can be reused.
This replaces the journal_seq field of bucket_mark, and is part of
eventually getting rid of the in memory bucket array.
We may need to make bch2_bucket_needs_journal_commit() lockless, pending
profiling and testing.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
We're trying to track down a bug that shows itself as newly-created
extents having stale dirty pointers - possibly due to the in memory gen
and the btree gen being inconsistent. This patch changes the error
message to also print out the in memory bucket gen when this happens.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
btree_gc sometimes needs another pass when it corrects bucket generation
numbers or data types - when it finds multiple pointers of different
data types to the same bucket, it may want to keep the second one it
found.
When this happens, we now clear out bucket sector counts _without_
resetting the bucket generation/data types that we already found,
instead of resetting them to what we have in the alloc btree.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
The repair for for btree_ptrs was saying one thing and doing another -
fortunately, that code can just be deleted.
Also, when we update a btree node pointer, we also have to update node
in memery, if it exists in the btree node cache - this fixes
bch2_check_fix_ptrs() to do that.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This reverts commit f95b61228efd04c9c158123da5827c96e9773b29.
It turns out, we're seeing filesystems in the wild end up with
blacklisted btree node bsets - this should not be happening, and until
we understand why and fix it we need to keep this code around.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
- Add a shim uuid_unparse_lower() in the kernel, since %pU doesn't work
in userspace
- We don't need to print the bcachefs: or the filesystem name prefix in
userspace
- Improve a few error messages
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Symbol decoding, via %ps, isn't supported in userspace - this will also
be faster when we're using trans->fn in the fast path, as with the new
BCH_JSET_ENTRY_log journal messages.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
The old code correctly handled the case where we were blacklisting a
range that exactly matched an existing entry, but not the case where the
new range partially overlaps an existing entry.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This patch converts bch2_sb_validate() and the .validate methods for the
various superblock sections to take printbuf, to which they can print
detailed error messages, including printing the entire section that was
invalid.
This is a great improvement over the previous situation, where we could
only return static strings that didn't have precise information about
what was wrong.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
bch2_trans_commit() can legitimately return -ENOSPC with
BTREE_INSERT_NOFAIL set if BTREE_INSERT_NOWAIT was also set.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Add a field to bch_dev for the dev_t of the underlying block device -
this fixes a null ptr deref in tracepoints.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
With BTREE_ITER_WITH_JOURNAL, there's no longer any restrictions on the
order we have to replay keys from the journal in, and we can also start
up journal reclaim right away - and delete a bunch of code.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This adds a new btree iterator flag, BTREE_ITER_WITH_JOURNAL, that is
automatically enabled when initializing a btree iterator before journal
replay has completed - it overlays the contents of the journal with the
btree.
This lets us delete bch2_btree_and_journal_walk() and just use the
normal btree iterator interface instead - which also lets us delete a
significant amount of duplicated code.
Note that BTREE_ITER_WITH_JOURNAL is still unoptimized in this patch -
we're redoing the binary search over keys in the journal every time we
call bch2_btree_iter_peek().
Signed-off-by: Kent Overstreet <kent.overstreet@linux.dev>
If we're not running fsck we still want to set BCH_FS_FSCK_DONE, so that
bch2_fsck_err() calls are interpreted as bch2_inconsistent_error()
calls().
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Add a flag to indicate whether a journal replay key has been
overwritten, and set/test it with appropriate btree locks held.
This fixes a race between the allocator - invalidating buckets, and
doing btree updates - and journal replay, which before this patch could
clobber the allocator thread's update with an older version of the key
from the journal.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This adds a _to_text() pretty printer for journal entries - including
every subtype - which will shortly be used by the 'bcachefs
list_journal' subcommand.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Add a journal entry type for logging messages, and add an option to use
it to log the transaction name - this makes for a very handy debugging
tool, as with it we can use the 'bcachefs list_journal' command to see
not only what updates were done, but what was doing them.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This adds some missing diagnostics from rare but annoying to debug
runtime allocation failure paths.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
The upcoming BTREE_ITER_WITH_JOURNAL patch will require journal keys to
stay in sorted order, so the btree iterator code can overlay them over
btree keys.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
This converts the error messages in the device add to a better style,
and adds some missing ones.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
There's places where we parse these numbers, and our parsing doesn't
cope with decimals currently - this is a hack to get the device_add path
working again where for the device blocksize there doesn't ever need to
be a decimal.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
If the btree updates pointing to a bucket were never flushed by the
journal before the bucket became empty again, we can reuse the bucket
without a journal flush.
This tweaks the tracking of journal sequence numbers in alloc keys to
implement this optimization: now, we only update the journal sequence
number in alloc keys on transitions to and from empty. When a bucket
becomes empty, we check if we can tell the journal not to flush entries
starting from when the bucket was used.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
Since dirty extents can be moved or overwritten, it's not just cached
data that we need the ptr_stale() check in bc2h_read_endio for - this
fixes data checksum errors seen in the tiering ktest tests.
Signed-off-by: Kent Overstreet <kent.overstreet@gmail.com>
