Unnecessary pointer-NULL checks in pgp-pgsql.c
Hi all,
Coverity is pointing out that we are doing pointer-NULL checks on
things that cannot be NULL in decrypt_internal():
out:
- if (src)
- mbuf_free(src);
- if (ctx)
- pgp_free(ctx);
+ Assert(ctx != NULL && src != NULL && dst != NULL);
+ mbuf_free(src);
+ pgp_free(ctx); if (err)
{
px_set_debug_handler(NULL);
- if (dst)
- mbuf_free(dst);
+ mbuf_free(dst);src, dst and ctx are created respectively from mbuf_create_from_data,
mbuf_create and pgp_init which never return NULL and they are palloc'd
all the time. I think that we could simplify things with the patch
attached, note that I added an assertion for correctness but I don't
really think that it is much necessary.
Regards,
--
Michael
Attachments:
20150202_pgcrypto_fixes.patchtext/x-diff; charset=US-ASCII; name=20150202_pgcrypto_fixes.patchDownload
diff --git a/contrib/pgcrypto/pgp-pgsql.c b/contrib/pgcrypto/pgp-pgsql.c
index 1a0e710..af1a990 100644
--- a/contrib/pgcrypto/pgp-pgsql.c
+++ b/contrib/pgcrypto/pgp-pgsql.c
@@ -594,16 +594,14 @@ decrypt_internal(int is_pubenc, int need_text, text *data,
got_unicode = pgp_get_unicode_mode(ctx);
out:
- if (src)
- mbuf_free(src);
- if (ctx)
- pgp_free(ctx);
+ Assert(ctx != NULL && src != NULL && dst != NULL);
+ mbuf_free(src);
+ pgp_free(ctx);
if (err)
{
px_set_debug_handler(NULL);
- if (dst)
- mbuf_free(dst);
+ mbuf_free(dst);
ereport(ERROR,
(errcode(ERRCODE_EXTERNAL_ROUTINE_INVOCATION_EXCEPTION),
errmsg("%s", px_strerror(err))));
On Sun, Feb 1, 2015 at 9:14 PM, Michael Paquier
<michael.paquier@gmail.com> wrote:
Coverity is pointing out that we are doing pointer-NULL checks on things that cannot be NULL in decrypt_internal(): out: - if (src) - mbuf_free(src); - if (ctx) - pgp_free(ctx); + Assert(ctx != NULL && src != NULL && dst != NULL); + mbuf_free(src); + pgp_free(ctx);if (err) { px_set_debug_handler(NULL); - if (dst) - mbuf_free(dst); + mbuf_free(dst);src, dst and ctx are created respectively from mbuf_create_from_data,
mbuf_create and pgp_init which never return NULL and they are palloc'd
all the time. I think that we could simplify things with the patch
attached, note that I added an assertion for correctness but I don't
really think that it is much necessary.
Yeah, I'd drop the assertion. Also, how about changing things around
slightly so that we lose the goto-label construct? There's only one
goto, and its only about 6 lines before the label, so we could just
flip the sense of the if-test and put the code that gets skipped
inside the if-block.
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers
Robert Haas wrote:
I wrote:
src, dst and ctx are created respectively from mbuf_create_from_data,
mbuf_create and pgp_init which never return NULL and they are palloc'd
all the time. I think that we could simplify things with the patch
attached, note that I added an assertion for correctness but I don't
really think that it is much necessary.Yeah, I'd drop the assertion. Also, how about changing things around
slightly so that we lose the goto-label construct? There's only one
goto, and its only about 6 lines before the label, so we could just
flip the sense of the if-test and put the code that gets skipped
inside the if-block.
Good idea. This gives the patch attached then.
Regards,
--
Michael
Attachments:
20150204_pgcrypto_fixes_v2.patchtext/x-patch; charset=US-ASCII; name=20150204_pgcrypto_fixes_v2.patchDownload
*** a/contrib/pgcrypto/pgp-pgsql.c
--- b/contrib/pgcrypto/pgp-pgsql.c
***************
*** 575,609 **** decrypt_internal(int is_pubenc, int need_text, text *data,
err = pgp_set_symkey(ctx, (uint8 *) VARDATA(key),
VARSIZE(key) - VARHDRSZ);
! /*
! * decrypt
! */
if (err >= 0)
err = pgp_decrypt(ctx, src, dst);
! /*
! * failed?
! */
! if (err < 0)
! goto out;
!
! if (ex.expect)
! check_expect(ctx, &ex);
! /* remember the setting */
! got_unicode = pgp_get_unicode_mode(ctx);
! out:
! if (src)
! mbuf_free(src);
! if (ctx)
! pgp_free(ctx);
if (err)
{
px_set_debug_handler(NULL);
! if (dst)
! mbuf_free(dst);
ereport(ERROR,
(errcode(ERRCODE_EXTERNAL_ROUTINE_INVOCATION_EXCEPTION),
errmsg("%s", px_strerror(err))));
--- 575,599 ----
err = pgp_set_symkey(ctx, (uint8 *) VARDATA(key),
VARSIZE(key) - VARHDRSZ);
! /* decrypt */
if (err >= 0)
+ {
err = pgp_decrypt(ctx, src, dst);
! if (ex.expect)
! check_expect(ctx, &ex);
! /* remember the setting */
! got_unicode = pgp_get_unicode_mode(ctx);
! }
! mbuf_free(src);
! pgp_free(ctx);
if (err)
{
px_set_debug_handler(NULL);
! mbuf_free(dst);
ereport(ERROR,
(errcode(ERRCODE_EXTERNAL_ROUTINE_INVOCATION_EXCEPTION),
errmsg("%s", px_strerror(err))));
On Wed, Feb 4, 2015 at 7:00 AM, Michael Paquier
<michael.paquier@gmail.com> wrote:
Robert Haas wrote:
I wrote:
src, dst and ctx are created respectively from mbuf_create_from_data,
mbuf_create and pgp_init which never return NULL and they are palloc'd
all the time. I think that we could simplify things with the patch
attached, note that I added an assertion for correctness but I don't
really think that it is much necessary.Yeah, I'd drop the assertion. Also, how about changing things around
slightly so that we lose the goto-label construct? There's only one
goto, and its only about 6 lines before the label, so we could just
flip the sense of the if-test and put the code that gets skipped
inside the if-block.Good idea. This gives the patch attached then.
Committed.
(BTW, why do you not leave a blank line between quoted text and your responses?)
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers
* Robert Haas (robertmhaas@gmail.com) wrote:
(BTW, why do you not leave a blank line between quoted text and your responses?)
+1. That vertical space is really helpful, at least to me.
Thanks,
Stephen
On Thu, Feb 5, 2015 at 5:31 AM, Stephen Frost <sfrost@snowman.net> wrote:
* Robert Haas (robertmhaas@gmail.com) wrote:
(BTW, why do you not leave a blank line between quoted text and your responses?)
+1. That vertical space is really helpful, at least to me.
Will do if people here are better with that. I just did it to keep the
messages shorter.
--
Michael
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers