memory leaks in ACEDB

Richard Durbin rd at mrc-lmb.cam.ac.uk
Mon Apr 4 13:05:23 EST 1994

Henry Cobb at Berkeley used "purify" to look for memory leaks in ACEDB.  I
hope it's OK to reproduce his results message here; they seemed to be
of general interest.


Date: Thu, 31 Mar 1994 14:11:39 -0800
From: "Henry J. Cobb" <hcobb at fly2.berkeley.edu>
Message-Id: <199403312211.OAA03820 at fly2.berkeley.edu>
To: acedev at peru.lbl.gov
Subject: Some questions about purity of essence.
Status: R

        Just a few things pure spoted.

        wgraph/graphsub.c, line 131, what is the reference to gBox going

Purify: Searching for all memory leaks...
There are 16926 leaked bytes (0.27% of the 6263736 allocated bytes in the heap)

        There seems to be a small leak in ksetdisp that I'll look into, but
the rest seems to be flydb porblems.


The gBox reference picks up a global pointer to the current open box
(if it exists, which it will if k is non-zero).  We set the default
foreground colour for the new box to the current foreground colour for
gBox.  (I am presuming you are referring to graphBoxStart() -- my
coordinates are slightly off from yours.)

Overall it looks like there is no major memory leak, but we would
certainly like to hear of any you do find.  In particular it would be
nice to try after "Add Update".  That seems to be very hard on the
heap for some reason.

Thanks Henry,  Richard

More information about the Acedb mailing list

Send comments to us at biosci-help [At] net.bio.net