bdiff: don't check border condition in loop
This is pretty much a copy of d500ddae7494, just to a different loop.
The condition `p == plast` (`plast == a + len - 1`) was only true on
the final iteration of the loop. So it was wasteful to check for it
on every iteration. We decrease the iteration count by 1 and add an
explicit check for `p == plast` after the loop.
Again, we see modest wins.
From the mozilla-unified repository:
$ perfbdiff -m 3041e4d59df2
! wall 0.035502 comb 0.040000 user 0.040000 sys 0.000000 (best of 100)
! wall 0.030480 comb 0.030000 user 0.030000 sys 0.000000 (best of 100)
$ perfbdiff 0e9928989e9c --alldata --count 100
! wall 4.097394 comb 4.100000 user 4.100000 sys 0.000000 (best of 3)
! wall 3.597798 comb 3.600000 user 3.600000 sys 0.000000 (best of 3)
The 2nd example throws a total of ~3.3GB of data at bdiff. This
change increases the throughput from ~811 MB/s to ~924 MB/s.
#!/bin/rc
# 9diff - Mercurial extdiff wrapper for diff(1)
rfork e
fn getfiles {
cd $1 &&
for(f in `{du -as | awk '{print $2}'})
test -f $f && echo `{cleanname $f}
}
fn usage {
echo >[1=2] usage: 9diff [diff options] parent child root
exit usage
}
opts=()
while(~ $1 -*){
opts=($opts $1)
shift
}
if(! ~ $#* 3)
usage
# extdiff will set the parent and child to a single file if there is
# only one change. If there are multiple changes, directories will be
# set. diff(1) does not cope particularly with directories; instead we
# do the recursion ourselves and diff each file individually.
if(test -f $1)
diff $opts $1 $2
if not{
# extdiff will create a snapshot of the working copy to prevent
# conflicts during the diff. We circumvent this behavior by
# diffing against the repository root to produce plumbable
# output. This is antisocial.
for(f in `{sort -u <{getfiles $1} <{getfiles $2}}){
file1=$1/$f; test -f $file1 || file1=/dev/null
file2=$3/$f; test -f $file2 || file2=/dev/null
diff $opts $file1 $file2
}
}
exit ''