localrepo: factor out base of filecache annotation class
It isn't needed that storecache is derived from repofilecache.
Changes in this patch allow repofilecache and storecache to do in own
__init__() differently from each other.
manifest: apply checkambig=True only for root 00manifest.i
This is a fix for my
14ad8e2a4abe, which used 'bool(dir)' as
checkambig value for revlog.__init__().
I can't remember why I did so in
14ad8e2a4abe, but this is obviously
wrong, because only root indexfile is cached via filecache-ed property
of localrepository.
revlog: address review feedback for deltachain C implementation
* Scope of "value" is reduced
* index_baserev() is documented
* Error is no longer redundantly set for -2 return values
* Error values are compared <= -2 instead of == -2 to protect
against odd failure scenarios
test-rebase-interruptions: stabilize for Windows
External hooks end up launching cmd.exe, which knows nothing about $VAR syntax.
For some reason, I thought that Mercurial would substitute in the value, in
order to paper over the platform difference. But I can't find that in the
documentation, and there's at least one other use of this pattern [1].
[1] https://www.mercurial-scm.org/repo/hg/file/tip/tests/test-histedit-fold.t#l477
drawdag: inline transaction() function
I suspect Jun wrote the method before he learnt that Python 2.7 allows
multiple context managers in a single with-clause.
revlog: C implementation of delta chain resolution
I've seen revlog._deltachain() appear in a number of performance
profiles. I suspect there are 2 reasons for this:
1. Delta chain resolution performs many index lookups, thus triggering
population of index tuples. Creating possibly tens of thousands of
PyObject will have overhead.
2. Delta chain resolution is a tight loop.
By moving delta chain resolution to C, we can defer instantiation
of full index entry tuples and make the loop faster courtesy of
not running in Python.
We can measure the impact to delta chain resolution via
`hg perflogrevision` using the mozilla-central repo with a recent
manifest having delta chain length of 33726:
$ hg perfrevlogrevision -m 364895
! full
! wall 0.367585 comb 0.370000 user 0.340000 sys 0.030000 (best of 27)
! wall 0.357581 comb 0.360000 user 0.350000 sys 0.010000 (best of 28)
! deltachain
! wall 0.010644 comb 0.010000 user 0.010000 sys 0.000000 (best of 270)
! wall 0.000292 comb 0.000000 user 0.000000 sys 0.000000 (best of 8729)
$ hg perfrevlogrevision --cache -m 364895
! deltachain
! wall 0.003904 comb 0.000000 user 0.000000 sys 0.000000 (best of 712)
! wall 0.000284 comb 0.000000 user 0.000000 sys 0.000000 (best of 9926)
The first test measures savings from both not instantiating index
entries and moving to C. The second test (which doesn't clear the
index caches) essentially isolates the benefits of moving from Python
to C. It still shows a 13.7x speedup (versus 36.4x). And there are
multiple milliseconds of savings within the critical path for resolving
revision data. I think that justifies the existence of C code.
A more striking example of the benefits of this change can be
demonstrated by timing `hg debugdeltachain -m` for the mozilla-central
repo:
$ time hg debugdeltachain -m > /dev/null
before: 1057.4s
after: 503.3s
PyPy2.7 5.8.0: 220.0s
It's worth noting that the C code isn't as optimal as it could be.
We're still instantiating a new PyObject for every revision. A future
optimization would be to reuse the PyObject on the cached index tuple.
We could potentially also get wins by using a memory array of raw
integers. There is also room for a delta chain cache on revlog
instances. Of course, the best optimization is to implement revlog
reading outside of Python so Python doesn't need to be concerned
about the relatively expensive index entries and operations on them.
revlog: backed out changeset
e9d325cfe071
We'll move this to the default branch.