tests/test-mq-qqueue.t

revlog: don't flush data file after every added revision

The current behavior of revlogs is to flush the data file when writing
data to it. Tracing system calls revealed that changegroup processing
incurred numerous write(2) calls for values much smaller than the
default buffer size (Python defaults to 4096, but it can be adjusted
based on detected block size at run time by CPython).

The reason we flush revlogs is so readers have all data available.
For example, the current code in revlog.py will re-open the revlog
file (instead of seeking an existing file handle) to read the text
of a revision. This happens when starting a new delta chain when
adding several revisions from changegroups, for example. Yes, this
is likely sub-optimal (we should probably be sharing file descriptors
between readers and writers to avoid the flushing and associated
overhead of re-opening files).

While flushing revlogs is necessary, it appears all callers are
diligent about flushing files before a read is performed (see
buildtext() in _addrevision()), making the flush in
_writeentry() redundant and unncessary. So, we remove it. In practice,
this means we incur a write(2) a) when the buffer is full (typically
4096 bytes) b) when a new delta chain is created rather than after
every added revision. This applies to every revlog, but by volume
it mostly impacts filelogs.

Removing the redundant flush from _writeentry() significantly
reduces the number of write(2) calls during changegroup processing on
my Linux machine. When applying a changegroup of the hg repo based on
my local repo, the total number of write(2) calls during application
of the mercurial/localrepo.py revlogs dropped from 1,320 to 217 with
this patch applied. Total I/O related system calls dropped from 1,577
to 474.

When unbundling a mozilla-central gzipped bundle (264,403 changesets
with 1,492,215 changes to 222,507 files), total write(2) calls
dropped from 1,252,881 to 827,106 and total system calls dropped from
3,601,259 to 3,178,636 - a reduction of 425,775!

While the system call reduction is significant, it appears
to have no impact on wall time on my Linux and Windows machines. Still,
fewer syscalls is fewer syscalls. Surely this can't hurt. If nothing
else, it makes examining remaining system call usage simpler and opens
the door to experimenting with the performance impact of different
buffer sizes.

  $ echo "[extensions]" >> $HGRCPATH
  $ echo "mq=" >> $HGRCPATH

  $ hg init foo
  $ cd foo
  $ echo a > a
  $ hg ci -qAm a

Default queue:

  $ hg qqueue
  patches (active)

  $ echo b > a
  $ hg qnew -fgDU somestuff

Applied patches in default queue:

  $ hg qap
  somestuff

Try to change patch (create succeeds, switch fails):

  $ hg qqueue foo --create
  abort: new queue created, but cannot make active as patches are applied
  [255]

  $ hg qqueue
  foo
  patches (active)

Empty default queue:

  $ hg qpop
  popping somestuff
  patch queue now empty

Switch queue:

  $ hg qqueue foo
  $ hg qqueue
  foo (active)
  patches

List queues, quiet:

  $ hg qqueue --quiet
  foo
  patches

Fail creating queue with already existing name:

  $ hg qqueue --create foo
  abort: queue "foo" already exists
  [255]

  $ hg qqueue
  foo (active)
  patches

Create new queue for rename:

  $ hg qqueue --create bar

  $ hg qqueue
  bar (active)
  foo
  patches

Rename queue, same name:

  $ hg qqueue --rename bar
  abort: can't rename "bar" to its current name
  [255]

Rename queue to existing:

  $ hg qqueue --rename foo
  abort: queue "foo" already exists
  [255]

Rename queue:

  $ hg qqueue --rename buz

  $ hg qqueue
  buz (active)
  foo
  patches

Switch back to previous queue:

  $ hg qqueue foo
  $ hg qqueue --delete buz

  $ hg qqueue
  foo (active)
  patches

Create queue for purge:

  $ hg qqueue --create purge-me

  $ hg qqueue
  foo
  patches
  purge-me (active)

Create patch for purge:

  $ hg qnew patch-purge-me

  $ ls -1d .hg/patches-purge-me 2>/dev/null || true
  .hg/patches-purge-me

  $ hg qpop -a
  popping patch-purge-me
  patch queue now empty

Purge queue:

  $ hg qqueue foo
  $ hg qqueue --purge purge-me

  $ hg qqueue
  foo (active)
  patches

  $ ls -1d .hg/patches-purge-me 2>/dev/null || true

Unapplied patches:

  $ hg qun
  $ echo c > a
  $ hg qnew -fgDU otherstuff

Fail switching back:

  $ hg qqueue patches
  abort: new queue created, but cannot make active as patches are applied
  [255]

Fail deleting current:

  $ hg qqueue foo --delete
  abort: cannot delete currently active queue
  [255]

Switch back and delete foo:

  $ hg qpop -a
  popping otherstuff
  patch queue now empty

  $ hg qqueue patches
  $ hg qqueue foo --delete
  $ hg qqueue
  patches (active)

Tricky cases:

  $ hg qqueue store --create
  $ hg qnew journal

  $ hg qqueue
  patches
  store (active)

  $ hg qpop -a
  popping journal
  patch queue now empty

  $ hg qqueue patches
  $ hg qun
  somestuff

Invalid names:

  $ hg qqueue test/../../bar --create
  abort: invalid queue name, may not contain the characters ":\/."
  [255]

  $ hg qqueue . --create
  abort: invalid queue name, may not contain the characters ":\/."
  [255]

  $ cd ..