From: Nicolas Palix <npalix@diku.dk>
To: "Randy Dunlap" <rdunlap@xenotime.net>,
"Roland Dreier" <rdreier@cisco.com>,
"Joe Perches" <joe@perches.com>,
"Andrew Morton" <akpm@linux-foundation.org>,
"David S. Miller" <davem@davemloft.net>,
"Michal Marek" <mmarek@suse.cz>,
"Sam Ravnborg" <sam@ravnborg.org>, "Julia Lawall" <julia@diku.dk>,
"Gilles Muller" <Gilles.Muller@lip6.fr>,
linux-kernel@vger.kernel.org, linux-kbuild@vger.kernel.org,
cocci@diku.dk, "Wolfram Sang" <w.sang@pengutronix.de>,
"Kernel Janitors" <kernel-janitors@vger.kernel.org>,
"Américo Wang" <xiyou.wangcong@gmail.com>,
"Andy Isaacson" <adi@hexapodia.org>,
"Linus Torvalds" <torvalds@linux-foundation.org>,
"Pekka Enberg" <penberg@cs.helsinki.fi>,
"walter harms" <wharms@bfs.de>, "Joerg Roedel" <joro@8bytes.org>,
"Jan Engelhardt" <jengelh@medozas.de>
Cc: Nicolas Palix <npalix@diku.dk>
Subject: [PATCH 7/7] Add scripts/coccinelle/deref_null.cocci
Date: Sun, 6 Jun 2010 17:15:07 +0200 [thread overview]
Message-ID: <1275837307-4283-8-git-send-email-npalix@diku.dk> (raw)
In-Reply-To: <1275837307-4283-1-git-send-email-npalix@diku.dk>
Add a Coccinelle file to identify the dereferences of NULL variables
This semantic patch identifies when a variable is known to be NULL
after a test, but it is still dereferenced later.
Signed-off-by: Nicolas Palix <npalix@diku.dk>
Signed-off-by: Julia Lawall <julia@diku.dk>
---
scripts/coccinelle/deref_null.cocci | 293 +++++++++++++++++++++++++++++++++++
1 files changed, 293 insertions(+), 0 deletions(-)
create mode 100644 scripts/coccinelle/deref_null.cocci
diff --git a/scripts/coccinelle/deref_null.cocci b/scripts/coccinelle/deref_null.cocci
new file mode 100644
index 0000000..9969d76
--- /dev/null
+++ b/scripts/coccinelle/deref_null.cocci
@@ -0,0 +1,293 @@
+///
+/// A variable is dereference under a NULL test.
+/// Even though it is know to be NULL.
+///
+// Confidence: Moderate
+// Copyright: (C) 2010 Nicolas Palix, DIKU. GPLv2.
+// Copyright: (C) 2010 Julia Lawall, DIKU. GPLv2.
+// Copyright: (C) 2010 Gilles Muller, INRIA/LiP6. GPLv2.
+// URL: http://coccinelle.lip6.fr/
+// Comments: -I ... -all_includes can give more complete results
+// Options:
+
+virtual context
+virtual patch
+virtual org
+virtual report
+
+@initialize:python depends on !context && patch && !org && !report@
+
+import sys
+print >> sys.stderr, "This semantic patch does not support the 'patch' mode."
+
+@depends on patch@
+@@
+
+this_rule_should_never_matches();
+
+@ifm depends on !patch@
+expression *E;
+statement S1,S2;
+position p1;
+@@
+
+if@p1 ((E == NULL && ...) || ...) S1 else S2
+
+// The following two rules are separate, because both can match a single
+// expression in different ways
+@pr1 depends on !patch expression@
+expression *ifm.E;
+identifier f;
+position p1;
+@@
+
+ (E != NULL && ...) ? <+...E->f@p1...+> : ...
+
+@pr2 depends on !patch expression@
+expression *ifm.E;
+identifier f;
+position p2;
+@@
+
+(
+ (E != NULL) && ... && <+...E->f@p2...+>
+|
+ (E == NULL) || ... || <+...E->f@p2...+>
+|
+ sizeof(<+...E->f@p2...+>)
+)
+
+// For org and report modes
+
+@r depends on !context && !patch && (org || report) exists@
+expression subE <= ifm.E;
+expression *ifm.E;
+expression E1,E2;
+identifier f;
+statement S1,S2,S3,S4;
+iterator iter;
+position p!={pr1.p1,pr2.p2};
+position ifm.p1;
+@@
+
+if@p1 ((E == NULL && ...) || ...)
+{
+ ... when != if (...) S1 else S2
+(
+ iter(subE,...) S4 // no use
+|
+ list_remove_head(E2,subE,...)
+|
+ subE = E1
+|
+ for(subE = E1;...;...) S4
+|
+ subE++
+|
+ ++subE
+|
+ --subE
+|
+ subE--
+|
+ &subE
+|
+ E->f@p // bad use
+)
+ ... when any
+ return ...;
+}
+else S3
+
+@script:python depends on !context && !patch && !org && report@
+p << r.p;
+p1 << ifm.p1;
+x << ifm.E;
+@@
+
+msg="ERROR: %s is NULL but dereferenced." % (x)
+coccilib.report.print_report(p[0], msg)
+cocci.include_match(False)
+
+@script:python depends on !context && !patch && org && !report@
+p << r.p;
+p1 << ifm.p1;
+x << ifm.E;
+@@
+
+msg="ERROR: %s is NULL but dereferenced." % (x)
+msg_safe=msg.replace("[","@(").replace("]",")")
+cocci.print_main(msg_safe,p)
+cocci.include_match(False)
+
+@s depends on !context && !patch && (org || report) exists@
+expression subE <= ifm.E;
+expression *ifm.E;
+expression E1,E2;
+identifier f;
+statement S1,S2,S3,S4;
+iterator iter;
+position p!={pr1.p1,pr2.p2};
+position ifm.p1;
+@@
+
+if@p1 ((E == NULL && ...) || ...)
+{
+ ... when != if (...) S1 else S2
+(
+ iter(subE,...) S4 // no use
+|
+ list_remove_head(E2,subE,...)
+|
+ subE = E1
+|
+ for(subE = E1;...;...) S4
+|
+ subE++
+|
+ ++subE
+|
+ --subE
+|
+ subE--
+|
+ &subE
+|
+ E->f@p // bad use
+)
+ ... when any
+}
+else S3
+
+@script:python depends on !context && !patch && !org && report@
+p << s.p;
+p1 << ifm.p1;
+x << ifm.E;
+@@
+
+msg="ERROR: %s is NULL but dereferenced." % (x)
+coccilib.report.print_report(p[0], msg)
+
+@script:python depends on !context && !patch && org && !report@
+p << s.p;
+p1 << ifm.p1;
+x << ifm.E;
+@@
+
+msg="ERROR: %s is NULL but dereferenced." % (x)
+msg_safe=msg.replace("[","@(").replace("]",")")
+cocci.print_main(msg_safe,p)
+
+// For context mode
+
+@depends on context && !patch && !org && !report exists@
+expression subE <= ifm.E;
+expression *ifm.E;
+expression E1,E2;
+identifier f;
+statement S1,S2,S3,S4;
+iterator iter;
+position p!={pr1.p1,pr2.p2};
+position ifm.p1;
+@@
+
+if@p1 ((E == NULL && ...) || ...)
+{
+ ... when != if (...) S1 else S2
+(
+ iter(subE,...) S4 // no use
+|
+ list_remove_head(E2,subE,...)
+|
+ subE = E1
+|
+ for(subE = E1;...;...) S4
+|
+ subE++
+|
+ ++subE
+|
+ --subE
+|
+ subE--
+|
+ &subE
+|
+* E->f@p // bad use
+)
+ ... when any
+ return ...;
+}
+else S3
+
+// The following three rules are duplicates of ifm, pr1 and pr2 respectively.
+// It is need because the previous rule as already made a "change".
+
+@ifm1 depends on !patch@
+expression *E;
+statement S1,S2;
+position p1;
+@@
+
+if@p1 ((E == NULL && ...) || ...) S1 else S2
+
+@pr11 depends on !patch expression@
+expression *ifm1.E;
+identifier f;
+position p1;
+@@
+
+ (E != NULL && ...) ? <+...E->f@p1...+> : ...
+
+@pr12 depends on !patch expression@
+expression *ifm1.E;
+identifier f;
+position p2;
+@@
+
+(
+ (E != NULL) && ... && <+...E->f@p2...+>
+|
+ (E == NULL) || ... || <+...E->f@p2...+>
+|
+ sizeof(<+...E->f@p2...+>)
+)
+
+@depends on context && !patch && !org && !report exists@
+expression subE <= ifm1.E;
+expression *ifm1.E;
+expression E1,E2;
+identifier f;
+statement S1,S2,S3,S4;
+iterator iter;
+position p!={pr11.p1,pr12.p2};
+position ifm1.p1;
+@@
+
+if@p1 ((E == NULL && ...) || ...)
+{
+ ... when != if (...) S1 else S2
+(
+ iter(subE,...) S4 // no use
+|
+ list_remove_head(E2,subE,...)
+|
+ subE = E1
+|
+ for(subE = E1;...;...) S4
+|
+ subE++
+|
+ ++subE
+|
+ --subE
+|
+ subE--
+|
+ &subE
+|
+* E->f@p // bad use
+)
+ ... when any
+}
+else S3
--
1.7.0.4
next prev parent reply other threads:[~2010-06-06 15:15 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-06-06 15:15 [PATCH 0/7] Add some checks to be used with Coccinelle Nicolas Palix
2010-06-06 15:15 ` [PATCH 1/7] Add a target to use the Coccinelle checker Nicolas Palix
2010-06-06 18:16 ` Sam Ravnborg
2010-06-07 12:35 ` Nicolas Palix
2010-06-30 18:58 ` Kulikov Vasiliy
2010-06-30 21:52 ` Nicolas Palix
2010-07-02 13:52 ` Kulikov Vasiliy
2010-07-02 14:30 ` Nicolas Palix
2010-07-02 20:11 ` Michal Marek
2010-06-06 15:15 ` [PATCH 2/7] Add Documentation/coccinelle.txt Nicolas Palix
2010-06-06 15:15 ` [PATCH 3/7] Add scripts/coccinelle/alloc/drop_kmalloc_cast.cocci Nicolas Palix
2010-06-06 15:15 ` [PATCH 4/7] Add scripts/coccinelle/alloc/kzalloc-simple.cocci Nicolas Palix
2010-06-06 15:15 ` [PATCH 5/7] Add scripts/coccinelle/resource_size.cocci Nicolas Palix
2010-06-06 15:15 ` [PATCH 6/7] Add scripts/coccinelle/err_cast.cocci Nicolas Palix
2010-06-06 15:15 ` Nicolas Palix [this message]
2010-06-10 11:52 ` [PATCH 0/7] Add some checks to be used with Coccinelle Joerg Roedel
2010-06-11 22:02 ` Michal Marek
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1275837307-4283-8-git-send-email-npalix@diku.dk \
--to=npalix@diku.dk \
--cc=Gilles.Muller@lip6.fr \
--cc=adi@hexapodia.org \
--cc=akpm@linux-foundation.org \
--cc=cocci@diku.dk \
--cc=davem@davemloft.net \
--cc=jengelh@medozas.de \
--cc=joe@perches.com \
--cc=joro@8bytes.org \
--cc=julia@diku.dk \
--cc=kernel-janitors@vger.kernel.org \
--cc=linux-kbuild@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mmarek@suse.cz \
--cc=penberg@cs.helsinki.fi \
--cc=rdreier@cisco.com \
--cc=rdunlap@xenotime.net \
--cc=sam@ravnborg.org \
--cc=torvalds@linux-foundation.org \
--cc=w.sang@pengutronix.de \
--cc=wharms@bfs.de \
--cc=xiyou.wangcong@gmail.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®