Skip to contentSkip to conversation

Fix potential issue on memmove

PATCH Namhyung Kim 4 participants 4 messages Aug 12, 2010

lore
Share & tools
lore
#0 lore raw
Quoted 22 lines
arch/x86/lib/memcpy_32.c +12 -3
@@ -25,19 +25,35 @@ void *memmove(void *dest, const void *src, size_t n)
 	int d0, d1, d2;
 
 	if (dest < src) {
-		memcpy(dest, src, n);
+		if ((dest + n) < src)
+			 return memcpy(dest, src, n);
+		else
+			__asm__ __volatile__(
+				"rep\n\t"
+				"movsb\n\t"
+				: "=&c" (d0), "=&S" (d1), "=&D" (d2)
+				:"0" (n),
+				 "1" (src),
+				 "2" (dest)
+				:"memory");
+
 	} else {
-		__asm__ __volatile__(
-			"std\n\t"

arch/x86/lib/memcpy_32.c void *memmove(void *dest, const void *src, size_t n) · line 42

Quoted 10 lines
arch/x86/lib/memcpy_32.c +2 -8
-			"rep\n\t"
-			"movsb\n\t"
-			"cld"
-			: "=&c" (d0), "=&S" (d1), "=&D" (d2)
-			:"0" (n),
-			 "1" (n-1+src),
-			 "2" (n-1+dest)
-			:"memory");
+		if((src + count) < dest)
+			return memcpy(dest, src, count);

'count' should be 'n'.

Signature · 2 lines
Regards,
Namhyung Kim
#1 lore raw

Thanks, I will fix it in next version.

Quoted 44 lines

-----Original Message-----
From: Namhyung Kim [mailto:[email protected]]
Sent: Thursday, August 12, 2010 2:49 PM
To: Ma, Ling
Cc: [email protected]; [email protected]; [email protected]; [email protected]
Subject: Re: [PATCH V2] [X86] Fix potential issue on memmove

arch/x86/lib/memcpy_32.c void *memmove(void *dest, const void *src, size_t

Quoted 3 lines
arch/x86/lib/memcpy_32.c +0 -0
@@ -25,19 +25,35 @@ void *memmove(void *dest, const void *src, size_t

n)

Quoted 19 lines
arch/x86/lib/memcpy_32.c +12 -3
 	int d0, d1, d2;
 
 	if (dest < src) {
-		memcpy(dest, src, n);
+		if ((dest + n) < src)
+			 return memcpy(dest, src, n);
+		else
+			__asm__ __volatile__(
+				"rep\n\t"
+				"movsb\n\t"
+				: "=&c" (d0), "=&S" (d1), "=&D" (d2)
+				:"0" (n),
+				 "1" (src),
+				 "2" (dest)
+				:"memory");
+
 	} else {
-		__asm__ __volatile__(
-			"std\n\t"

arch/x86/lib/memcpy_32.c void *memmove(void *dest, const void *src, size_t · line 42

Quoted 10 lines
arch/x86/lib/memcpy_32.c +2 -8
-			"rep\n\t"
-			"movsb\n\t"
-			"cld"
-			: "=&c" (d0), "=&S" (d1), "=&D" (d2)
-			:"0" (n),
-			 "1" (n-1+src),
-			 "2" (n-1+dest)
-			:"memory");
+		if((src + count) < dest)
+			return memcpy(dest, src, count);

'count' should be 'n'.

Quoted 4 lines

--
Regards,
Namhyung Kim

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

#2 lore raw
Quoted 1 line · Ling

Thanks, I will fix it in next version.

Also, please split this into two patches -- one for fixing memmove, and the other for re-enabling the alias avoidance in memcpy.

-hpa

Signature · 2 lines
H. Peter Anvin, Intel Open Source Technology Center
I work for Intel.  I don't speak on their behalf.
#3 lore raw
Quoted 8 lines · H. Peter Anvin
Quoted 1 line · Ling

Thanks, I will fix it in next version.

Also, please split this into two patches -- one for fixing memmove, and the other for re-enabling the alias avoidance in memcpy.

-hpa

Nevermind, the patch doesn't actually contain the x86/mem patch.

It's rather confusing to put "re-active patch id a1e5278e40f16a4611264f8da9e557c16cb6f6ed" for a patch that was caught in testing rather than ever included; it doesn't even exist under that patch ID anymore.

-hpa