aboutsummaryrefslogtreecommitdiff
path: root/include/asm-s390/checksum.h
diff options
context:
space:
mode:
authorMartin Schwidefsky <schwidefsky@de.ibm.com>2006-09-28 16:56:43 +0200
committerMartin Schwidefsky <schwidefsky@de.ibm.com>2006-09-28 16:56:43 +0200
commit94c12cc7d196bab34aaa98d38521549fa1e5ef76 (patch)
tree8e0cec0ed44445d74a2cb5160303d6b4dfb1bc31 /include/asm-s390/checksum.h
parent25d83cbfaa44e1b9170c0941c3ef52ca39f54ccc (diff)
[S390] Inline assembly cleanup.
Major cleanup of all s390 inline assemblies. They now have a common coding style. Quite a few have been shortened, mainly by using register asm variables. Use of the EX_TABLE macro helps as well. The atomic ops, bit ops and locking inlines new use the Q-constraint if a newer gcc is used. That results in slightly better code. Thanks to Christian Borntraeger for proof reading the changes. Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Diffstat (limited to 'include/asm-s390/checksum.h')
-rw-r--r--include/asm-s390/checksum.h176
1 files changed, 55 insertions, 121 deletions
diff --git a/include/asm-s390/checksum.h b/include/asm-s390/checksum.h
index 471f2af2b16..37c362d89fa 100644
--- a/include/asm-s390/checksum.h
+++ b/include/asm-s390/checksum.h
@@ -30,57 +30,13 @@
static inline unsigned int
csum_partial(const unsigned char * buff, int len, unsigned int sum)
{
- /*
- * Experiments with ethernet and slip connections show that buf
- * is aligned on either a 2-byte or 4-byte boundary.
- */
-#ifndef __s390x__
- register_pair rp;
-
- rp.subreg.even = (unsigned long) buff;
- rp.subreg.odd = (unsigned long) len;
- __asm__ __volatile__ (
- "0: cksm %0,%1\n" /* do checksum on longs */
- " jo 0b\n"
- : "+&d" (sum), "+&a" (rp) : : "cc", "memory" );
-#else /* __s390x__ */
- __asm__ __volatile__ (
- " lgr 2,%1\n" /* address in gpr 2 */
- " lgfr 3,%2\n" /* length in gpr 3 */
- "0: cksm %0,2\n" /* do checksum on longs */
- " jo 0b\n"
- : "+&d" (sum)
- : "d" (buff), "d" (len)
- : "cc", "memory", "2", "3" );
-#endif /* __s390x__ */
- return sum;
-}
-
-/*
- * csum_partial as an inline function
- */
-static inline unsigned int
-csum_partial_inline(const unsigned char * buff, int len, unsigned int sum)
-{
-#ifndef __s390x__
- register_pair rp;
+ register unsigned long reg2 asm("2") = (unsigned long) buff;
+ register unsigned long reg3 asm("3") = (unsigned long) len;
- rp.subreg.even = (unsigned long) buff;
- rp.subreg.odd = (unsigned long) len;
- __asm__ __volatile__ (
- "0: cksm %0,%1\n" /* do checksum on longs */
- " jo 0b\n"
- : "+&d" (sum), "+&a" (rp) : : "cc", "memory" );
-#else /* __s390x__ */
- __asm__ __volatile__ (
- " lgr 2,%1\n" /* address in gpr 2 */
- " lgfr 3,%2\n" /* length in gpr 3 */
- "0: cksm %0,2\n" /* do checksum on longs */
- " jo 0b\n"
- : "+&d" (sum)
- : "d" (buff), "d" (len)
- : "cc", "memory", "2", "3" );
-#endif /* __s390x__ */
+ asm volatile(
+ "0: cksm %0,%1\n" /* do checksum on longs */
+ " jo 0b\n"
+ : "+d" (sum), "+d" (reg2), "+d" (reg3) : : "cc", "memory");
return sum;
}
@@ -114,7 +70,7 @@ static inline unsigned int
csum_partial_copy_nocheck (const char *src, char *dst, int len, unsigned int sum)
{
memcpy(dst,src,len);
- return csum_partial_inline(dst, len, sum);
+ return csum_partial(dst, len, sum);
}
/*
@@ -126,22 +82,22 @@ csum_fold(unsigned int sum)
#ifndef __s390x__
register_pair rp;
- __asm__ __volatile__ (
- " slr %N1,%N1\n" /* %0 = H L */
- " lr %1,%0\n" /* %0 = H L, %1 = H L 0 0 */
- " srdl %1,16\n" /* %0 = H L, %1 = 0 H L 0 */
- " alr %1,%N1\n" /* %0 = H L, %1 = L H L 0 */
- " alr %0,%1\n" /* %0 = H+L+C L+H */
- " srl %0,16\n" /* %0 = H+L+C */
- : "+&d" (sum), "=d" (rp) : : "cc" );
+ asm volatile(
+ " slr %N1,%N1\n" /* %0 = H L */
+ " lr %1,%0\n" /* %0 = H L, %1 = H L 0 0 */
+ " srdl %1,16\n" /* %0 = H L, %1 = 0 H L 0 */
+ " alr %1,%N1\n" /* %0 = H L, %1 = L H L 0 */
+ " alr %0,%1\n" /* %0 = H+L+C L+H */
+ " srl %0,16\n" /* %0 = H+L+C */
+ : "+&d" (sum), "=d" (rp) : : "cc");
#else /* __s390x__ */
- __asm__ __volatile__ (
- " sr 3,3\n" /* %0 = H*65536 + L */
- " lr 2,%0\n" /* %0 = H L, R2/R3 = H L / 0 0 */
- " srdl 2,16\n" /* %0 = H L, R2/R3 = 0 H / L 0 */
- " alr 2,3\n" /* %0 = H L, R2/R3 = L H / L 0 */
- " alr %0,2\n" /* %0 = H+L+C L+H */
- " srl %0,16\n" /* %0 = H+L+C */
+ asm volatile(
+ " sr 3,3\n" /* %0 = H*65536 + L */
+ " lr 2,%0\n" /* %0 = H L, 2/3 = H L / 0 0 */
+ " srdl 2,16\n" /* %0 = H L, 2/3 = 0 H / L 0 */
+ " alr 2,3\n" /* %0 = H L, 2/3 = L H / L 0 */
+ " alr %0,2\n" /* %0 = H+L+C L+H */
+ " srl %0,16\n" /* %0 = H+L+C */
: "+&d" (sum) : : "cc", "2", "3");
#endif /* __s390x__ */
return ((unsigned short) ~sum);
@@ -155,29 +111,7 @@ csum_fold(unsigned int sum)
static inline unsigned short
ip_fast_csum(unsigned char *iph, unsigned int ihl)
{
- unsigned long sum;
-#ifndef __s390x__
- register_pair rp;
-
- rp.subreg.even = (unsigned long) iph;
- rp.subreg.odd = (unsigned long) ihl*4;
- __asm__ __volatile__ (
- " sr %0,%0\n" /* set sum to zero */
- "0: cksm %0,%1\n" /* do checksum on longs */
- " jo 0b\n"
- : "=&d" (sum), "+&a" (rp) : : "cc", "memory" );
-#else /* __s390x__ */
- __asm__ __volatile__ (
- " slgr %0,%0\n" /* set sum to zero */
- " lgr 2,%1\n" /* address in gpr 2 */
- " lgfr 3,%2\n" /* length in gpr 3 */
- "0: cksm %0,2\n" /* do checksum on ints */
- " jo 0b\n"
- : "=&d" (sum)
- : "d" (iph), "d" (ihl*4)
- : "cc", "memory", "2", "3" );
-#endif /* __s390x__ */
- return csum_fold(sum);
+ return csum_fold(csum_partial(iph, ihl*4, 0));
}
/*
@@ -190,47 +124,47 @@ csum_tcpudp_nofold(unsigned long saddr, unsigned long daddr,
unsigned int sum)
{
#ifndef __s390x__
- __asm__ __volatile__ (
- " alr %0,%1\n" /* sum += saddr */
- " brc 12,0f\n"
- " ahi %0,1\n" /* add carry */
+ asm volatile(
+ " alr %0,%1\n" /* sum += saddr */
+ " brc 12,0f\n"
+ " ahi %0,1\n" /* add carry */
"0:"
- : "+&d" (sum) : "d" (saddr) : "cc" );
- __asm__ __volatile__ (
- " alr %0,%1\n" /* sum += daddr */
- " brc 12,1f\n"
- " ahi %0,1\n" /* add carry */
+ : "+&d" (sum) : "d" (saddr) : "cc");
+ asm volatile(
+ " alr %0,%1\n" /* sum += daddr */
+ " brc 12,1f\n"
+ " ahi %0,1\n" /* add carry */
"1:"
- : "+&d" (sum) : "d" (daddr) : "cc" );
- __asm__ __volatile__ (
- " alr %0,%1\n" /* sum += (len<<16) + (proto<<8) */
- " brc 12,2f\n"
- " ahi %0,1\n" /* add carry */
+ : "+&d" (sum) : "d" (daddr) : "cc");
+ asm volatile(
+ " alr %0,%1\n" /* sum += (len<<16) + (proto<<8) */
+ " brc 12,2f\n"
+ " ahi %0,1\n" /* add carry */
"2:"
: "+&d" (sum)
: "d" (((unsigned int) len<<16) + (unsigned int) proto)
- : "cc" );
+ : "cc");
#else /* __s390x__ */
- __asm__ __volatile__ (
- " lgfr %0,%0\n"
- " algr %0,%1\n" /* sum += saddr */
- " brc 12,0f\n"
- " aghi %0,1\n" /* add carry */
- "0: algr %0,%2\n" /* sum += daddr */
- " brc 12,1f\n"
- " aghi %0,1\n" /* add carry */
- "1: algfr %0,%3\n" /* sum += (len<<16) + proto */
- " brc 12,2f\n"
- " aghi %0,1\n" /* add carry */
- "2: srlg 0,%0,32\n"
- " alr %0,0\n" /* fold to 32 bits */
- " brc 12,3f\n"
- " ahi %0,1\n" /* add carry */
- "3: llgfr %0,%0"
+ asm volatile(
+ " lgfr %0,%0\n"
+ " algr %0,%1\n" /* sum += saddr */
+ " brc 12,0f\n"
+ " aghi %0,1\n" /* add carry */
+ "0: algr %0,%2\n" /* sum += daddr */
+ " brc 12,1f\n"
+ " aghi %0,1\n" /* add carry */
+ "1: algfr %0,%3\n" /* sum += (len<<16) + proto */
+ " brc 12,2f\n"
+ " aghi %0,1\n" /* add carry */
+ "2: srlg 0,%0,32\n"
+ " alr %0,0\n" /* fold to 32 bits */
+ " brc 12,3f\n"
+ " ahi %0,1\n" /* add carry */
+ "3: llgfr %0,%0"
: "+&d" (sum)
: "d" (saddr), "d" (daddr),
"d" (((unsigned int) len<<16) + (unsigned int) proto)
- : "cc", "0" );
+ : "cc", "0");
#endif /* __s390x__ */
return sum;
}