Skip to contentSkip to conversation

fs, binfmt_aout: Fix pointer warnings

PATCH In review Borislav Petkov 2 participants 2 messages Mar 8, 2010 – Mar 9, 2010

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 fs, binfmt_aout: Fix pointer warnings

Hi,

I'm getting the warnings in the commit message below with current git. Maybe fix them like this:

--
From: Borislav Petkov <[email protected]>
Date: Mon, 8 Mar 2010 16:37:42 +0100
Subject: [PATCH] fs, binfmt_aout: Fix pointer warnings
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

I get

fs/binfmt_aout.c: In function ‘aout_core_dump’:
fs/binfmt_aout.c:125: warning: passing argument 2 of ‘dump_write’ makes pointer from integer without a cast include/linux/coredump.h:12: note: expected ‘const void *’ but argument is of type ‘long unsigned int’ fs/binfmt_aout.c:132: warning: passing argument 2 of ‘dump_write’ makes pointer from integer without a cast include/linux/coredump.h:12: note: expected ‘const void *’ but argument is of type ‘long unsigned int’

due to dump_write() expecting a user void *. Fold casts into the START_DATA/START_STACK macros and shut up the warnings.

---
 fs/binfmt_aout.c |   14 ++++++++------
 1 files changed, 8 insertions(+), 6 deletions(-)
fs/binfmt_aout.c +8 -6
index 15d80bb..9b6aef0 100644
@@ -75,14 +75,16 @@ static int aout_core_dump(struct coredump_params *cprm)
 	struct file *file = cprm->file;
 	mm_segment_t fs;
 	int has_dumped = 0;
-	unsigned long dump_start, dump_size;
+	void __user *dump_start;
+	int dump_size;
 	struct user dump;
 #ifdef __alpha__
-#       define START_DATA(u)	(u.start_data)
+#       define START_DATA(u)	((void __user *)u.start_data)
 #else
-#	define START_DATA(u)	((u.u_tsize << PAGE_SHIFT) + u.start_code)
+#	define START_DATA(u)	((void __user *)((u.u_tsize << PAGE_SHIFT) + \
+				 u.start_code))
 #endif
-#       define START_STACK(u)   (u.start_stack)
+#       define START_STACK(u)   ((void __user *)u.start_stack)
 
 	fs = get_fs();
 	set_fs(KERNEL_DS);
@@ -104,9 +106,9 @@ static int aout_core_dump(struct coredump_params *cprm)
 
 /* make sure we actually have a data and stack area to dump */
 	set_fs(USER_DS);
-	if (!access_ok(VERIFY_READ, (void __user *)START_DATA(dump), dump.u_dsize << PAGE_SHIFT))
+	if (!access_ok(VERIFY_READ, START_DATA(dump), dump.u_dsize << PAGE_SHIFT))
 		dump.u_dsize = 0;
-	if (!access_ok(VERIFY_READ, (void __user *)START_STACK(dump), dump.u_ssize << PAGE_SHIFT))
+	if (!access_ok(VERIFY_READ, START_STACK(dump), dump.u_ssize << PAGE_SHIFT))
 		dump.u_ssize = 0;
 
 	set_fs(KERNEL_DS);
Signature · 1 line
1.6.6.1
#1 lore raw

Thanks for your report.

From: Borislav Petkov <[email protected]>
Subject: [PATCH] fs, binfmt_aout: Fix pointer warnings
Date: Mon, 8 Mar 2010 16:46:01 +0100

Quoted 55 lines

Hi,

I'm getting the warnings in the commit message below with current git. Maybe fix them like this:

--
From: Borislav Petkov <[email protected]>
Date: Mon, 8 Mar 2010 16:37:42 +0100
Subject: [PATCH] fs, binfmt_aout: Fix pointer warnings
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

I get

fs/binfmt_aout.c: In function ‘aout_core_dump’:
fs/binfmt_aout.c:125: warning: passing argument 2 of ‘dump_write’ makes pointer from integer without a cast include/linux/coredump.h:12: note: expected ‘const void *’ but argument is of type ‘long unsigned int’ fs/binfmt_aout.c:132: warning: passing argument 2 of ‘dump_write’ makes pointer from integer without a cast include/linux/coredump.h:12: note: expected ‘const void *’ but argument is of type ‘long unsigned int’

due to dump_write() expecting a user void *. Fold casts into the START_DATA/START_STACK macros and shut up the warnings.

---
 fs/binfmt_aout.c |   14 ++++++++------
 1 files changed, 8 insertions(+), 6 deletions(-)
fs/binfmt_aout.c +6 -4
index 15d80bb..9b6aef0 100644
@@ -75,14 +75,16 @@ static int aout_core_dump(struct coredump_params *cprm)
 	struct file *file = cprm->file;
 	mm_segment_t fs;
 	int has_dumped = 0;
-	unsigned long dump_start, dump_size;
+	void __user *dump_start;
+	int dump_size;
 	struct user dump;
 #ifdef __alpha__
-#       define START_DATA(u)	(u.start_data)
+#       define START_DATA(u)	((void __user *)u.start_data)
 #else
-#	define START_DATA(u)	((u.u_tsize << PAGE_SHIFT) + u.start_code)
+#	define START_DATA(u)	((void __user *)((u.u_tsize << PAGE_SHIFT) + \
+				 u.start_code))
 #endif
-#       define START_STACK(u)   (u.start_stack)
+#       define START_STACK(u)   ((void __user *)u.start_stack)
 
 	fs = get_fs();
 	set_fs(KERNEL_DS);

fs/binfmt_aout.c static int aout_core_dump(struct coredump_params *cprm) · line 112

Quoted 9 lines
fs/binfmt_aout.c +2 -2
@@ -104,9 +106,9 @@ static int aout_core_dump(struct coredump_params *cprm)
 
 /* make sure we actually have a data and stack area to dump */
 	set_fs(USER_DS);
-	if (!access_ok(VERIFY_READ, (void __user *)START_DATA(dump), dump.u_dsize << PAGE_SHIFT))
+	if (!access_ok(VERIFY_READ, START_DATA(dump), dump.u_dsize << PAGE_SHIFT))
 		dump.u_dsize = 0;
-	if (!access_ok(VERIFY_READ, (void __user *)START_STACK(dump), dump.u_ssize << PAGE_SHIFT))
+	if (!access_ok(VERIFY_READ, START_STACK(dump), dump.u_ssize << PAGE_SHIFT))
 		dump.u_ssize = 0;

Your fixing looks reasonable to me. I understand your patch set as follows.

- The START_DATA() and START_STACK() macro calls are all passed to the arguments of type (void *), meaning that they need casts in the appropreate positions.

- Moreover, they are more natural to be used directly as (void *) than to be casted outside the macro.

Thanks.

Quoted 5 lines
fs/binfmt_aout.c +0 -0
 	set_fs(KERNEL_DS);
Signature · 1 line
1.6.6.1

���{.n�+�������+%�����ݶ��w��{.n�+����{��G�����{ay�ʇڙ�,j��f���h��������z_�(�階�ݢj"���m������G������?����&���~��iO��z��v�^�m���� ������?�I�