Обсуждение: memory leak in ODBC driver
There is a 16-byte leak for every statement handle that has SQLFetch called on it.
I have tracked this leak down to its source:
SC_Fetch calls extend_getdata_info, which allocates memory for the stmt.getdata_info.
However, this memory is never freed when the statement is deallocated.
This patch seems to fix the problem - however I don't know enough about the internals of the driver to really
understandif this is the correct or safe fix.
Can one of the developers help me out here?
--- statement.c.old 2004-11-27 13:07:57.000000000 -0500
+++ statement.c 2004-11-27 13:09:37.000000000 -0500
@@ -354,6 +354,8 @@
if (self->callbacks)
free(self->callbacks);
+ GDATA_unbind_cols(SC_get_GDTI(self), TRUE);
+
DELETE_STMT_CS(self);
free(self);
> -----Original Message----- > From: pgsql-odbc-owner@postgresql.org > [mailto:pgsql-odbc-owner@postgresql.org] On Behalf Of Scot Loach > Sent: 27 November 2004 18:11 > To: pgsql-odbc@postgresql.org > Subject: [ODBC] memory leak in ODBC driver > > There is a 16-byte leak for every statement handle that has > SQLFetch called on it. > I have tracked this leak down to its source: > SC_Fetch calls extend_getdata_info, which allocates memory > for the stmt.getdata_info. > However, this memory is never freed when the statement is deallocated. > > This patch seems to fix the problem - however I don't know > enough about the internals of the driver to really understand > if this is the correct or safe fix. > Can one of the developers help me out here? It certainly *looks* OK to me, but I share the same concerns as you (I've never really got that far inside the driver before). Anyone else care to comment? Regards, Dave.
I've done some more code inspection and I think my fix is ok. I'll make the change to my code base and I'll let you know if I have any problems with it. I've submitted bug #1067 for this. scot. -----Original Message----- From: Dave Page [mailto:dpage@vale-housing.co.uk] Sent: Sat 11/27/2004 6:19 PM To: Scot Loach; pgsql-odbc@postgresql.org Cc: Subject: RE: [ODBC] memory leak in ODBC driver > -----Original Message----- > From: pgsql-odbc-owner@postgresql.org > [mailto:pgsql-odbc-owner@postgresql.org] On Behalf Of Scot Loach > Sent: 27 November 2004 18:11 > To: pgsql-odbc@postgresql.org > Subject: [ODBC] memory leak in ODBC driver > > There is a 16-byte leak for every statement handle that has > SQLFetch called on it. > I have tracked this leak down to its source: > SC_Fetch calls extend_getdata_info, which allocates memory > for the stmt.getdata_info. > However, this memory is never freed when the statement is deallocated. > > This patch seems to fix the problem - however I don't know > enough about the internals of the driver to really understand > if this is the correct or safe fix. > Can one of the developers help me out here? It certainly *looks* OK to me, but I share the same concerns as you (I've never really got that far inside the driver before). Anyone else care to comment? Regards, Dave.
> -----Original Message----- > From: Scot Loach [mailto:sloach@sandvine.com] > Sent: 30 November 2004 02:31 > To: Dave Page; pgsql-odbc@postgresql.org > Subject: RE: [ODBC] memory leak in ODBC driver > > I've done some more code inspection and I think my fix is ok. > > I'll make the change to my code base and I'll let you know if > I have any problems with it. > > I've submitted bug #1067 for this. OK, thanks. Let me knoiw when you're happy with your testing and I'll commit the patch given the lack of other opinions!! Regards, Dave
Dave Page wrote: > > > > -----Original Message----- > > From: Scot Loach [mailto:sloach@sandvine.com] > > Sent: 30 November 2004 02:31 > > To: Dave Page; pgsql-odbc@postgresql.org > > Subject: RE: [ODBC] memory leak in ODBC driver > > > > I've done some more code inspection and I think my fix is ok. > > > > I'll make the change to my code base and I'll let you know if > > I have any problems with it. > > > > I've submitted bug #1067 for this. > > OK, thanks. Let me knoiw when you're happy with your testing and I'll > commit the patch given the lack of other opinions!! Scot, is this ready to be applied to ODBC? -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup. | Newtown Square, Pennsylvania 19073
Sure, we can apply it, I haven't had any problems with it. -----Original Message----- From: pgsql-odbc-owner@postgresql.org [mailto:pgsql-odbc-owner@postgresql.org]On Behalf Of Bruce Momjian Sent: Monday, December 20, 2004 8:26 PM To: Dave Page Cc: Scot Loach; pgsql-odbc@postgresql.org Subject: Re: [ODBC] memory leak in ODBC driver Dave Page wrote: > > > > -----Original Message----- > > From: Scot Loach [mailto:sloach@sandvine.com] > > Sent: 30 November 2004 02:31 > > To: Dave Page; pgsql-odbc@postgresql.org > > Subject: RE: [ODBC] memory leak in ODBC driver > > > > I've done some more code inspection and I think my fix is ok. > > > > I'll make the change to my code base and I'll let you know if > > I have any problems with it. > > > > I've submitted bug #1067 for this. > > OK, thanks. Let me knoiw when you're happy with your testing and I'll > commit the patch given the lack of other opinions!! Scot, is this ready to be applied to ODBC? -- Bruce Momjian | http://candle.pha.pa.us pgman@candle.pha.pa.us | (610) 359-1001 + If your life is a hard drive, | 13 Roberts Road + Christ can be your backup. | Newtown Square, Pennsylvania 19073 ---------------------------(end of broadcast)--------------------------- TIP 1: subscribe and unsubscribe commands go to majordomo@postgresql.org
> -----Original Message----- > From: Scot Loach [mailto:sloach@sandvine.com] > Sent: 21 December 2004 15:00 > To: Bruce Momjian; Dave Page > Cc: pgsql-odbc@postgresql.org > Subject: RE: [ODBC] memory leak in ODBC driver > > Sure, we can apply it, I haven't had any problems with it. > Thanks, patch applied. Regards, Dave.