From 1da177e4c3f41524e886b7f1b8a0c1fc7321cac2 Mon Sep 17 00:00:00 2001 From: Linus Torvalds Date: Sat, 16 Apr 2005 15:20:36 -0700 Subject: Linux-2.6.12-rc2 Initial git repository build. I'm not bothering with the full history, even though we have it. We can create a separate "historical" git archive of that later if we want to, and in the meantime it's about 3.2GB when imported into git - space that would just make the early git days unnecessarily complicated, when we don't have a lot of good infrastructure for it. Let it rip! --- include/asm-sparc/checksum.h | 253 +++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 253 insertions(+) create mode 100644 include/asm-sparc/checksum.h (limited to 'include/asm-sparc/checksum.h') diff --git a/include/asm-sparc/checksum.h b/include/asm-sparc/checksum.h new file mode 100644 index 000000000000..286158108974 --- /dev/null +++ b/include/asm-sparc/checksum.h @@ -0,0 +1,253 @@ +/* $Id: checksum.h,v 1.33 2002/02/01 22:01:05 davem Exp $ */ +#ifndef __SPARC_CHECKSUM_H +#define __SPARC_CHECKSUM_H + +/* checksum.h: IP/UDP/TCP checksum routines on the Sparc. + * + * Copyright(C) 1995 Linus Torvalds + * Copyright(C) 1995 Miguel de Icaza + * Copyright(C) 1996 David S. Miller + * Copyright(C) 1996 Eddie C. Dost + * Copyright(C) 1997 Jakub Jelinek + * + * derived from: + * Alpha checksum c-code + * ix86 inline assembly + * RFC1071 Computing the Internet Checksum + */ + +#include +#include + +/* computes the checksum of a memory block at buff, length len, + * and adds in "sum" (32-bit) + * + * returns a 32-bit number suitable for feeding into itself + * or csum_tcpudp_magic + * + * this function must be called with even lengths, except + * for the last fragment, which may be odd + * + * it's best to have buff aligned on a 32-bit boundary + */ +extern unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum); + +/* the same as csum_partial, but copies from fs:src while it + * checksums + * + * here even more important to align src and dst on a 32-bit (or even + * better 64-bit) boundary + */ + +extern unsigned int __csum_partial_copy_sparc_generic (const unsigned char *, unsigned char *); + +static inline unsigned int +csum_partial_copy_nocheck (const unsigned char *src, unsigned char *dst, int len, + unsigned int sum) +{ + register unsigned int ret asm("o0") = (unsigned int)src; + register char *d asm("o1") = dst; + register int l asm("g1") = len; + + __asm__ __volatile__ ( + "call __csum_partial_copy_sparc_generic\n\t" + " mov %6, %%g7\n" + : "=&r" (ret), "=&r" (d), "=&r" (l) + : "0" (ret), "1" (d), "2" (l), "r" (sum) + : "o2", "o3", "o4", "o5", "o7", + "g2", "g3", "g4", "g5", "g7", + "memory", "cc"); + return ret; +} + +static inline unsigned int +csum_partial_copy_from_user(const unsigned char __user *src, unsigned char *dst, int len, + unsigned int sum, int *err) + { + if (!access_ok (VERIFY_READ, src, len)) { + *err = -EFAULT; + memset (dst, 0, len); + return sum; + } else { + register unsigned long ret asm("o0") = (unsigned long)src; + register char *d asm("o1") = dst; + register int l asm("g1") = len; + register unsigned int s asm("g7") = sum; + + __asm__ __volatile__ ( + ".section __ex_table,#alloc\n\t" + ".align 4\n\t" + ".word 1f,2\n\t" + ".previous\n" + "1:\n\t" + "call __csum_partial_copy_sparc_generic\n\t" + " st %8, [%%sp + 64]\n" + : "=&r" (ret), "=&r" (d), "=&r" (l), "=&r" (s) + : "0" (ret), "1" (d), "2" (l), "3" (s), "r" (err) + : "o2", "o3", "o4", "o5", "o7", "g2", "g3", "g4", "g5", + "cc", "memory"); + return ret; + } + } + +static inline unsigned int +csum_partial_copy_to_user(const unsigned char *src, unsigned char __user *dst, int len, + unsigned int sum, int *err) +{ + if (!access_ok (VERIFY_WRITE, dst, len)) { + *err = -EFAULT; + return sum; + } else { + register unsigned long ret asm("o0") = (unsigned long)src; + register char __user *d asm("o1") = dst; + register int l asm("g1") = len; + register unsigned int s asm("g7") = sum; + + __asm__ __volatile__ ( + ".section __ex_table,#alloc\n\t" + ".align 4\n\t" + ".word 1f,1\n\t" + ".previous\n" + "1:\n\t" + "call __csum_partial_copy_sparc_generic\n\t" + " st %8, [%%sp + 64]\n" + : "=&r" (ret), "=&r" (d), "=&r" (l), "=&r" (s) + : "0" (ret), "1" (d), "2" (l), "3" (s), "r" (err) + : "o2", "o3", "o4", "o5", "o7", + "g2", "g3", "g4", "g5", + "cc", "memory"); + return ret; + } +} + +#define HAVE_CSUM_COPY_USER +#define csum_and_copy_to_user csum_partial_copy_to_user + +/* ihl is always 5 or greater, almost always is 5, and iph is word aligned + * the majority of the time. + */ +static inline unsigned short ip_fast_csum(const unsigned char *iph, + unsigned int ihl) +{ + unsigned short sum; + + /* Note: We must read %2 before we touch %0 for the first time, + * because GCC can legitimately use the same register for + * both operands. + */ + __asm__ __volatile__("sub\t%2, 4, %%g4\n\t" + "ld\t[%1 + 0x00], %0\n\t" + "ld\t[%1 + 0x04], %%g2\n\t" + "ld\t[%1 + 0x08], %%g3\n\t" + "addcc\t%%g2, %0, %0\n\t" + "addxcc\t%%g3, %0, %0\n\t" + "ld\t[%1 + 0x0c], %%g2\n\t" + "ld\t[%1 + 0x10], %%g3\n\t" + "addxcc\t%%g2, %0, %0\n\t" + "addx\t%0, %%g0, %0\n" + "1:\taddcc\t%%g3, %0, %0\n\t" + "add\t%1, 4, %1\n\t" + "addxcc\t%0, %%g0, %0\n\t" + "subcc\t%%g4, 1, %%g4\n\t" + "be,a\t2f\n\t" + "sll\t%0, 16, %%g2\n\t" + "b\t1b\n\t" + "ld\t[%1 + 0x10], %%g3\n" + "2:\taddcc\t%0, %%g2, %%g2\n\t" + "srl\t%%g2, 16, %0\n\t" + "addx\t%0, %%g0, %0\n\t" + "xnor\t%%g0, %0, %0" + : "=r" (sum), "=&r" (iph) + : "r" (ihl), "1" (iph) + : "g2", "g3", "g4", "cc"); + return sum; +} + +/* Fold a partial checksum without adding pseudo headers. */ +static inline unsigned int csum_fold(unsigned int sum) +{ + unsigned int tmp; + + __asm__ __volatile__("addcc\t%0, %1, %1\n\t" + "srl\t%1, 16, %1\n\t" + "addx\t%1, %%g0, %1\n\t" + "xnor\t%%g0, %1, %0" + : "=&r" (sum), "=r" (tmp) + : "0" (sum), "1" (sum<<16) + : "cc"); + return sum; +} + +static inline unsigned long csum_tcpudp_nofold(unsigned long saddr, + unsigned long daddr, + unsigned int len, + unsigned short proto, + unsigned int sum) +{ + __asm__ __volatile__("addcc\t%1, %0, %0\n\t" + "addxcc\t%2, %0, %0\n\t" + "addxcc\t%3, %0, %0\n\t" + "addx\t%0, %%g0, %0\n\t" + : "=r" (sum), "=r" (saddr) + : "r" (daddr), "r" ((proto<<16)+len), "0" (sum), + "1" (saddr) + : "cc"); + return sum; +} + +/* + * computes the checksum of the TCP/UDP pseudo-header + * returns a 16-bit checksum, already complemented + */ +static inline unsigned short int csum_tcpudp_magic(unsigned long saddr, + unsigned long daddr, + unsigned short len, + unsigned short proto, + unsigned int sum) +{ + return csum_fold(csum_tcpudp_nofold(saddr,daddr,len,proto,sum)); +} + +#define _HAVE_ARCH_IPV6_CSUM + +static inline unsigned short int csum_ipv6_magic(struct in6_addr *saddr, + struct in6_addr *daddr, + __u32 len, + unsigned short proto, + unsigned int sum) +{ + __asm__ __volatile__ ( + "addcc %3, %4, %%g4\n\t" + "addxcc %5, %%g4, %%g4\n\t" + "ld [%2 + 0x0c], %%g2\n\t" + "ld [%2 + 0x08], %%g3\n\t" + "addxcc %%g2, %%g4, %%g4\n\t" + "ld [%2 + 0x04], %%g2\n\t" + "addxcc %%g3, %%g4, %%g4\n\t" + "ld [%2 + 0x00], %%g3\n\t" + "addxcc %%g2, %%g4, %%g4\n\t" + "ld [%1 + 0x0c], %%g2\n\t" + "addxcc %%g3, %%g4, %%g4\n\t" + "ld [%1 + 0x08], %%g3\n\t" + "addxcc %%g2, %%g4, %%g4\n\t" + "ld [%1 + 0x04], %%g2\n\t" + "addxcc %%g3, %%g4, %%g4\n\t" + "ld [%1 + 0x00], %%g3\n\t" + "addxcc %%g2, %%g4, %%g4\n\t" + "addxcc %%g3, %%g4, %0\n\t" + "addx 0, %0, %0\n" + : "=&r" (sum) + : "r" (saddr), "r" (daddr), + "r"(htonl(len)), "r"(htonl(proto)), "r"(sum) + : "g2", "g3", "g4", "cc"); + + return csum_fold(sum); +} + +/* this routine is used for miscellaneous IP-like checksums, mainly in icmp.c */ +static inline unsigned short ip_compute_csum(unsigned char * buff, int len) +{ + return csum_fold(csum_partial(buff, len, 0)); +} + +#endif /* !(__SPARC_CHECKSUM_H) */ -- cgit v1.2.2