view tests/test-push-warn @ 5339:058e93c3d07d

I have spotted the biggest bottleneck in "bdiff.c". Actually it was
pretty easy to find after I recompiled the python interpreter and
mercurial for profiling.

In "bdiff.c" function "equatelines" allocates the minimum hash table
size, which can lead to tons of collisions. I introduced an
"overcommit" factor of 16, this is, I allocate 16 times more memory
than the minimum value. Overcommiting 128 times does not improve the
performance over the 16-times case.

#!/bin/sh

mkdir a
cd a
hg init
echo foo > t1
hg add t1
hg commit -m "1" -d "1000000 0"

cd ..
hg clone a b

cd a
echo foo > t2
hg add t2
hg commit -m "2" -d "1000000 0"

cd ../b
echo foo > t3
hg add t3
hg commit -m "3" -d "1000000 0"

hg push ../a
hg pull ../a
hg push ../a
hg merge
hg commit -m "4" -d "1000000 0"
hg push ../a
cd ..

hg init c
cd c
for i in 0 1 2; do
    echo $i >> foo
    hg ci -Am $i -d "1000000 0"
done
cd ..

hg clone c d
cd d
for i in 0 1; do
    hg co -C $i
    echo d-$i >> foo
    hg ci -m d-$i -d "1000000 0"
done

HGMERGE=true hg merge 3
hg ci -m c-d -d "1000000 0"

hg push ../c; echo $?
hg push -r 2 ../c; echo $?
hg push -r 3 ../c; echo $?
hg push -r 3 -r 4 ../c; echo $?
hg push -f -r 3 -r 4 ../c; echo $?
hg push -r 5 ../c; echo $?

# issue 450
hg init ../e
hg push -r 0 ../e ; echo $?
hg push -r 1 ../e ; echo $?

exit 0