Fix tiny memory leaks
Hi, hackers!
There is a memory leak in functions check_application_name() and
check_cluster_name().
Functions are located in src/backend/commands/variable.c
The leak can be triggered using SQL command: SET
application_name=new_name;
You can run pgbench with this command for a long time by passing this
command in file via '-f' flag.
regards,
Tofig
Attachments:
0001-Fix-tiny-memory-leaks.patchtext/x-diff; name=0001-Fix-tiny-memory-leaks.patchDownload
From a801d03c57819d3de181967d72260cb42bbcf6a3 Mon Sep 17 00:00:00 2001
From: Tofig Aliev <t.aliev@postgrespro.ru>
Date: Fri, 6 Dec 2024 14:58:38 +0700
Subject: [PATCH] Fix tiny memory leaks
Memory for (*newval) allocated earlier
in the parse_and_validate_value() method.
Since this method re-allocates memory,
it is necessary to free the previously allocated memory.
---
src/backend/commands/variable.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/src/backend/commands/variable.c b/src/backend/commands/variable.c
index 0ecff94d0e..2c8059e8d8 100644
--- a/src/backend/commands/variable.c
+++ b/src/backend/commands/variable.c
@@ -1092,6 +1092,8 @@ check_application_name(char **newval, void **extra, GucSource source)
return false;
}
+ guc_free(*newval);
+
pfree(clean);
*newval = ret;
return true;
@@ -1128,6 +1130,8 @@ check_cluster_name(char **newval, void **extra, GucSource source)
return false;
}
+ guc_free(*newval);
+
pfree(clean);
*newval = ret;
return true;
--
2.43.0
On 6 Dec 2024, at 09:38, Tofig Aliev <t.aliev@postgrespro.ru> wrote:
There is a memory leak in functions check_application_name() and check_cluster_name().
Functions are located in src/backend/commands/variable.c
LGTM.
--
Daniel Gustafsson
On Fri, Dec 6, 2024 at 12:49 AM Daniel Gustafsson <daniel@yesql.se> wrote:
On 6 Dec 2024, at 09:38, Tofig Aliev <t.aliev@postgrespro.ru> wrote:
There is a memory leak in functions check_application_name() and check_cluster_name().
Functions are located in src/backend/commands/variable.cLGTM.
LGTM. It seems commit a9d58bfe8a3a missed to fix this memory leak.
Regards,
--
Masahiko Sawada
Amazon Web Services: https://aws.amazon.com
On 9 Dec 2024, at 19:52, Masahiko Sawada <sawada.mshk@gmail.com> wrote:
On Fri, Dec 6, 2024 at 12:49 AM Daniel Gustafsson <daniel@yesql.se> wrote:
On 6 Dec 2024, at 09:38, Tofig Aliev <t.aliev@postgrespro.ru> wrote:
There is a memory leak in functions check_application_name() and check_cluster_name().
Functions are located in src/backend/commands/variable.cLGTM.
LGTM. It seems commit a9d58bfe8a3a missed to fix this memory leak.
Yeah. While fairly insignificant in nature, and there has been no complaints
until now, I'm preparing a backpatch down to REL_15_STABLE to keep the code in
sync for other backpatches to apply clean.
--
Daniel Gustafsson
On 9 Dec 2024, at 20:03, Daniel Gustafsson <daniel@yesql.se> wrote:
On 9 Dec 2024, at 19:52, Masahiko Sawada <sawada.mshk@gmail.com> wrote:
On Fri, Dec 6, 2024 at 12:49 AM Daniel Gustafsson <daniel@yesql.se> wrote:
On 6 Dec 2024, at 09:38, Tofig Aliev <t.aliev@postgrespro.ru> wrote:
There is a memory leak in functions check_application_name() and check_cluster_name().
Functions are located in src/backend/commands/variable.cLGTM.
LGTM. It seems commit a9d58bfe8a3a missed to fix this memory leak.
Yeah. While fairly insignificant in nature, and there has been no complaints
until now, I'm preparing a backpatch down to REL_15_STABLE to keep the code in
sync for other backpatches to apply clean.
Done, but to 16 and not 15 which was a typo in the above email.
--
Daniel Gustafsson
On Mon, Dec 9, 2024 at 12:04 PM Daniel Gustafsson <daniel@yesql.se> wrote:
On 9 Dec 2024, at 20:03, Daniel Gustafsson <daniel@yesql.se> wrote:
On 9 Dec 2024, at 19:52, Masahiko Sawada <sawada.mshk@gmail.com> wrote:
On Fri, Dec 6, 2024 at 12:49 AM Daniel Gustafsson <daniel@yesql.se> wrote:
On 6 Dec 2024, at 09:38, Tofig Aliev <t.aliev@postgrespro.ru> wrote:
There is a memory leak in functions check_application_name() and check_cluster_name().
Functions are located in src/backend/commands/variable.cLGTM.
LGTM. It seems commit a9d58bfe8a3a missed to fix this memory leak.
Yeah. While fairly insignificant in nature, and there has been no complaints
until now, I'm preparing a backpatch down to REL_15_STABLE to keep the code in
sync for other backpatches to apply clean.Done, but to 16 and not 15 which was a typo in the above email.
Thank you!
Regards,
--
Masahiko Sawada
Amazon Web Services: https://aws.amazon.com