fix/tests/gc-concurrent.builder.sh

echo "Build started" > "$lockFifo"

mkdir $out
echo $(cat $input1/foo)$(cat $input2/bar) > $out/foobar

# Wait for someone to write on the fifo
cat "$lockFifo"

# $out should not have been GC'ed while we were sleeping, but just in
# case...
mkdir -p $out

# Check that the GC hasn't deleted the lock on our output.
test -e "$out.lock"

ln -s $input2 $out/input-2
Make the gc-concurrent test more reliable Use a fifo pipe to handle the synchronisation between the different threads rather than relying on delays 2020-07-02 15:43:35 +02:00			`echo "Build started" > "$lockFifo"`

* Add a test to check whether concurrent garbage collection (i.e., running the collector while builds are in progress) works correctly. The test currently fails. 2005-01-28 21:36:46 +01:00			`mkdir $out`
			`echo $(cat $input1/foo)$(cat $input2/bar) > $out/foobar`

Make the gc-concurrent test more reliable Use a fifo pipe to handle the synchronisation between the different threads rather than relying on delays 2020-07-02 15:43:35 +02:00			`# Wait for someone to write on the fifo`
			`cat "$lockFifo"`
* Increase the sleep periods a bit to make the test less likely to fail on slow machines. Of course it would be better if this test wasn't timing dependent... 2008-08-14 11:26:30 +02:00
			`# $out should not have been GC'ed while we were sleeping, but just in`
			`# case...`
			`mkdir -p $out`
* Add a test to check whether concurrent garbage collection (i.e., running the collector while builds are in progress) works correctly. The test currently fails. 2005-01-28 21:36:46 +01:00
* Don't delete active lock files. 2005-01-31 13:19:53 +01:00			`# Check that the GC hasn't deleted the lock on our output.`
			`test -e "$out.lock"`

			`ln -s $input2 $out/input-2`