x86: coding style fixes in arch/x86/lib/csum-wrappers_64.c
authorPaolo Ciarrocchi <paolo.ciarrocchi@gmail.com>
Sun, 17 Feb 2008 13:56:50 +0000 (14:56 +0100)
committerIngo Molnar <mingo@elte.hu>
Tue, 19 Feb 2008 15:18:32 +0000 (16:18 +0100)
no code changed:

arch/x86/lib/csum-wrappers_64.o:
   text    data     bss     dec     hex filename
    839       0       0     839     347 csum-wrappers_64.o.before
    839       0       0     839     347 csum-wrappers_64.o.after
md5:
b31994226c33e0b52bef5a0e110b84b0  csum-wrappers_64.o.before.asm
b31994226c33e0b52bef5a0e110b84b0  csum-wrappers_64.o.after.asm

Signed-off-by: Paolo Ciarrocchi <paolo.ciarrocchi@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
arch/x86/lib/csum-wrappers_64.c

index fd42a4a095fc1bb822f1dc033df8a35fa58070b5..95e45dcc5a297625ac6b2d8781773dc011f288db 100644 (file)
 /* Copyright 2002,2003 Andi Kleen, SuSE Labs.
  * Subject to the GNU Public License v.2
- * 
+ *
  * Wrappers of assembly checksum functions for x86-64.
  */
 
 #include <asm/checksum.h>
 #include <linux/module.h>
 
-/** 
- * csum_partial_copy_from_user - Copy and checksum from user space. 
- * @src: source address (user space) 
+/**
+ * csum_partial_copy_from_user - Copy and checksum from user space.
+ * @src: source address (user space)
  * @dst: destination address
  * @len: number of bytes to be copied.
  * @isum: initial sum that is added into the result (32bit unfolded)
  * @errp: set to -EFAULT for an bad source address.
- * 
+ *
  * Returns an 32bit unfolded checksum of the buffer.
- * src and dst are best aligned to 64bits. 
- */ 
+ * src and dst are best aligned to 64bits.
+ */
 __wsum
 csum_partial_copy_from_user(const void __user *src, void *dst,
                            int len, __wsum isum, int *errp)
-{ 
+{
        might_sleep();
        *errp = 0;
-       if (likely(access_ok(VERIFY_READ,src, len))) { 
+       if (likely(access_ok(VERIFY_READ, src, len))) {
                /* Why 6, not 7? To handle odd addresses aligned we
                   would need to do considerable complications to fix the
                   checksum which is defined as an 16bit accumulator. The
                   fix alignment code is primarily for performance
                   compatibility with 32bit and that will handle odd
                   addresses slowly too. */
-               if (unlikely((unsigned long)src & 6)) {                 
-                       while (((unsigned long)src & 6) && len >= 2) { 
-                               __u16 val16;                    
+               if (unlikely((unsigned long)src & 6)) {
+                       while (((unsigned long)src & 6) && len >= 2) {
+                               __u16 val16;
                                *errp = __get_user(val16, (const __u16 __user *)src);
                                if (*errp)
                                        return isum;
                                *(__u16 *)dst = val16;
                                isum = (__force __wsum)add32_with_carry(
                                                (__force unsigned)isum, val16);
-                               src += 2; 
-                               dst += 2; 
+                               src += 2;
+                               dst += 2;
                                len -= 2;
                        }
                }
                isum = csum_partial_copy_generic((__force const void *)src,
                                        dst, len, isum, errp, NULL);
-               if (likely(*errp == 0)) 
+               if (likely(*errp == 0))
                        return isum;
-       } 
+       }
        *errp = -EFAULT;
-       memset(dst,0,len); 
-       return isum;            
-} 
+       memset(dst, 0, len);
+       return isum;
+}
 
 EXPORT_SYMBOL(csum_partial_copy_from_user);
 
-/** 
- * csum_partial_copy_to_user - Copy and checksum to user space. 
+/**
+ * csum_partial_copy_to_user - Copy and checksum to user space.
  * @src: source address
  * @dst: destination address (user space)
  * @len: number of bytes to be copied.
  * @isum: initial sum that is added into the result (32bit unfolded)
  * @errp: set to -EFAULT for an bad destination address.
- * 
+ *
  * Returns an 32bit unfolded checksum of the buffer.
  * src and dst are best aligned to 64bits.
- */ 
+ */
 __wsum
 csum_partial_copy_to_user(const void *src, void __user *dst,
                          int len, __wsum isum, int *errp)
-{ 
+{
        might_sleep();
        if (unlikely(!access_ok(VERIFY_WRITE, dst, len))) {
                *errp = -EFAULT;
-               return 0; 
+               return 0;
        }
 
        if (unlikely((unsigned long)dst & 6)) {
-               while (((unsigned long)dst & 6) && len >= 2) { 
+               while (((unsigned long)dst & 6) && len >= 2) {
                        __u16 val16 = *(__u16 *)src;
                        isum = (__force __wsum)add32_with_carry(
                                        (__force unsigned)isum, val16);
                        *errp = __put_user(val16, (__u16 __user *)dst);
                        if (*errp)
                                return isum;
-                       src += 2; 
-                       dst += 2; 
+                       src += 2;
+                       dst += 2;
                        len -= 2;
                }
        }
 
        *errp = 0;
-       return csum_partial_copy_generic(src, (void __force *)dst,len,isum,NULL,errp); 
-} 
+       return csum_partial_copy_generic(src, (void __force *)dst, len, isum, NULL, errp);
+}
 
 EXPORT_SYMBOL(csum_partial_copy_to_user);
 
-/** 
+/**
  * csum_partial_copy_nocheck - Copy and checksum.
  * @src: source address
  * @dst: destination address
  * @len: number of bytes to be copied.
  * @isum: initial sum that is added into the result (32bit unfolded)
- * 
+ *
  * Returns an 32bit unfolded checksum of the buffer.
- */ 
+ */
 __wsum
 csum_partial_copy_nocheck(const void *src, void *dst, int len, __wsum sum)
-{ 
-       return csum_partial_copy_generic(src,dst,len,sum,NULL,NULL);
-} 
+{
+       return csum_partial_copy_generic(src, dst, len, sum, NULL, NULL);
+}
 EXPORT_SYMBOL(csum_partial_copy_nocheck);
 
 __sum16 csum_ipv6_magic(const struct in6_addr *saddr,
@@ -119,16 +119,16 @@ __sum16 csum_ipv6_magic(const struct in6_addr *saddr,
                        __u32 len, unsigned short proto, __wsum sum)
 {
        __u64 rest, sum64;
-     
+
        rest = (__force __u64)htonl(len) + (__force __u64)htons(proto) +
                (__force __u64)sum;
        asm("  addq (%[saddr]),%[sum]\n"
            "  adcq 8(%[saddr]),%[sum]\n"
-           "  adcq (%[daddr]),%[sum]\n" 
+           "  adcq (%[daddr]),%[sum]\n"
            "  adcq 8(%[daddr]),%[sum]\n"
            "  adcq $0,%[sum]\n"
-           : [sum] "=r" (sum64) 
-           : "[sum]" (rest),[saddr] "r" (saddr), [daddr] "r" (daddr));
+           : [sum] "=r" (sum64)
+           : "[sum]" (rest), [saddr] "r" (saddr), [daddr] "r" (daddr));
        return csum_fold((__force __wsum)add32_with_carry(sum64 & 0xffffffff, sum64>>32));
 }