Skip to content

Commit

Permalink
x86: Align skb w/ start of cacheline on newer core 2/Xeon Arch
Browse files Browse the repository at this point in the history
x86 architectures can handle unaligned accesses in hardware, and it has
been shown that unaligned DMA accesses can be expensive on Nehalem
architectures.  As such we should overwrite NET_IP_ALIGN to resolve
this issue.

Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: x86@kernel.org
Signed-off-by: Alexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Acked-by: H. Peter Anvin <hpa@zytor.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
  • Loading branch information
Alexander Duyck authored and David S. Miller committed Jun 30, 2010
1 parent cb836a9 commit ea812ca
Showing 1 changed file with 9 additions and 0 deletions.
9 changes: 9 additions & 0 deletions arch/x86/include/asm/system.h
Original file line number Diff line number Diff line change
Expand Up @@ -457,4 +457,13 @@ static inline void rdtsc_barrier(void)
alternative(ASM_NOP3, "lfence", X86_FEATURE_LFENCE_RDTSC);
}

#ifdef CONFIG_MCORE2
/*
* We handle most unaligned accesses in hardware. On the other hand
* unaligned DMA can be quite expensive on some Nehalem processors.
*
* Based on this we disable the IP header alignment in network drivers.
*/
#define NET_IP_ALIGN 0
#endif
#endif /* _ASM_X86_SYSTEM_H */

0 comments on commit ea812ca

Please sign in to comment.