stream-clone: disable gc for `_entries_walk` duration
The number of small container created turn Python in a gc-frenzy that seriously
impact performance.
This significantly boost performance. The following number comes from a large
private repository using perf::stream-locked-section:
base-line: 35.04 seconds
prev-change: 24.51 seconds (-30%)
this-change: 20.88 seconds (-40% from baseline; -15% from previous changes)
#require clang-format
Test that a simple "hg fix" configuration for clang-format works.
$ cat >> $HGRCPATH <<EOF
> [extensions]
> fix =
> [experimental]
> evolution.createmarkers=True
> evolution.allowunstable=True
> [fix]
> clang-format:command=clang-format --style=Google --assume-filename={rootpath}
> clang-format:linerange=--lines={first}:{last}
> clang-format:pattern=set:**.cpp or **.hpp
> EOF
$ hg init repo
$ cd repo
$ printf "void foo(){int x=2;}\n" > foo.cpp
$ printf "void\nfoo();\n" > foo.hpp
$ hg commit -Am "foo commit"
adding foo.cpp
adding foo.hpp
$ hg cat -r tip *
void foo(){int x=2;}
void
foo();
$ hg fix -r tip
$ hg cat -r tip *
void foo() { int x = 2; }
void foo();
$ cd ..