revlog: increase I/O bound to 4x the amount of data consumed
This doesn't affect normal clones since they'd be bound by the CPU bound below
anyway -- it does, however, improve generaldelta clones significantly.
This also results in better deltaing for generaldelta clones -- in generaldelta
clones, we calculate deltas with respect to the closest base if it has a higher
revision number than either parent. If the base is on a significantly different
branch, this can result in pointlessly massive deltas. This reduces the number
of bases and hence the number of bad deltas.
Empirically, for a highly branchy repository, this resulted in an improvement
of around 15% to manifest size.
#!/usr/bin/env python
import sys
import os
os.chdir(os.getenv('TESTTMP'))
if sys.argv[1] != "user@dummy":
sys.exit(-1)
os.environ["SSH_CLIENT"] = "127.0.0.1 1 2"
log = open("dummylog", "ab")
log.write("Got arguments")
for i, arg in enumerate(sys.argv[1:]):
log.write(" %d:%s" % (i + 1, arg))
log.write("\n")
log.close()
hgcmd = sys.argv[2]
if os.name == 'nt':
# hack to make simple unix single quote quoting work on windows
hgcmd = hgcmd.replace("'", '"')
r = os.system(hgcmd)
sys.exit(bool(r))