Unnecessary subpool creation

While I was investigating issue #602, I wrote small Ruby
script to take the matching of pool creation and deletion
output from SVN_POOL_DEBUG_VERBOSE output. I attached it to
the end of this mail in case it would be of interest to
somebody.

With that script, I found two places where Subversion
creates subpool but never destroys it. One is
libsvn_fs/trail.c:52 and the other is
libsvn_ra_local/commit_editor.c:637.

For those two, the life time of subpool is exactly the same
as the parent pool. When I first saw it, I thought it was a
bug and looked abort_txn and commit_txn in trail.c. And I
found from commetn at commit_txn it is an intentional
behavior. I'd suggest removing those two subpool creation
but I don't know why it's done that way. Are there only
historical reasons?

-- 
Yoshiki Hayashi

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org