2013-08-29 18:11:10

by Jan-Simon Möller

Subject: [PATCH] Fix bitoperation for compilation with clang

From: Jan-Simon Möller <[email protected]>

v2: Fix bug in statement as pointed out by Herbert Xu. Kudos to pipacs.

Author: PaX Team <pageexec at freemail.hu>
ML-Post: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20120507/142707.html
URL: http://llvm.linuxfoundation.org

Merge: Jan-Simon Möller <dl9pf at gmx.de>

Description:

Fix for warning:
linux/crypto/fcrypt.c:143:47: warning: signed shift result (0x598000000) requires 36 bits to
represent, but 'int' only has 32 bits [-Wshift-overflow]
Z(0xef), Z(0x70), Z(0xcf), Z(0xc2), Z(0x2a), Z(0xb3), Z(0x61), Z(0xad),
^~~~~~~
linux/crypto/fcrypt.c:113:29: note: expanded from macro 'Z'
^ ~~
linux/include/uapi/linux/byteorder/little_endian.h:38:53: note: expanded from macro
'__cpu_to_be32'
^
linux/include/uapi/linux/swab.h:116:21: note: expanded from macro '__swab32'
___constant_swab32(x) : \
^
linux/include/uapi/linux/swab.h:18:12: note: expanded from macro '___constant_swab32'
(((__u32)(x) & (__u32)0x0000ff00UL) << 8) | \
^

Solution - make sure we don't exceed the 32 bit range.
#define Z(x) cpu_to_be32(((x & 0x1f) << 27) | (x >> 5))

Signed-off-by: Jan-Simon Möller <[email protected]>
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
CC: [email protected]
---
crypto/fcrypt.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/crypto/fcrypt.c b/crypto/fcrypt.c
index 3b2cf56..021d7fe 100644
--- a/crypto/fcrypt.c
+++ b/crypto/fcrypt.c
@@ -110,7 +110,7 @@ static const __be32 sbox0[256] = {
};

#undef Z
-#define Z(x) cpu_to_be32((x << 27) | (x >> 5))
+#define Z(x) cpu_to_be32(((x & 0x1f) << 27) | (x >> 5))
static const __be32 sbox1[256] = {
Z(0x77), Z(0x14), Z(0xa6), Z(0xfe), Z(0xb2), Z(0x5e), Z(0x8c), Z(0x3e),
Z(0x67), Z(0x6c), Z(0xa1), Z(0x0d), Z(0xc2), Z(0xa2), Z(0xc1), Z(0x85),
--
1.8.1.4

2013-09-02 10:36:18

by Herbert Xu

Subject: Re: [PATCH] Fix bitoperation for compilation with clang

On Thu, Aug 29, 2013 at 08:09:24PM +0200, [email protected] wrote:
> From: Jan-Simon M?ller <[email protected]>
>
> v2: Fix bug in statement as pointed out by Herbert Xu. Kudos to pipacs.
>
> Author: PaX Team <pageexec at freemail.hu>
> ML-Post: http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20120507/142707.html
> URL: http://llvm.linuxfoundation.org
>
> Merge: Jan-Simon M?ller <dl9pf at gmx.de>
>
> Description:
>
> Fix for warning:
> linux/crypto/fcrypt.c:143:47: warning: signed shift result (0x598000000) requires 36 bits to
> represent, but 'int' only has 32 bits [-Wshift-overflow]
> Z(0xef), Z(0x70), Z(0xcf), Z(0xc2), Z(0x2a), Z(0xb3), Z(0x61), Z(0xad),
> ^~~~~~~
> linux/crypto/fcrypt.c:113:29: note: expanded from macro 'Z'
> ^ ~~
> linux/include/uapi/linux/byteorder/little_endian.h:38:53: note: expanded from macro
> '__cpu_to_be32'
> ^
> linux/include/uapi/linux/swab.h:116:21: note: expanded from macro '__swab32'
> ___constant_swab32(x) : \
> ^
> linux/include/uapi/linux/swab.h:18:12: note: expanded from macro '___constant_swab32'
> (((__u32)(x) & (__u32)0x0000ff00UL) << 8) | \
> ^
>
> Solution - make sure we don't exceed the 32 bit range.
> #define Z(x) cpu_to_be32(((x & 0x1f) << 27) | (x >> 5))
>
> Signed-off-by: Jan-Simon M?ller <[email protected]>

Patch applied.
--
Email: Herbert Xu <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt