linux

mirror of https://github.com/armbian/linux.git synced 2026-01-06 10:13:00 -08:00

Files

Dean Nelson 52f81dc563 e1000: save skb counts in TX to avoid cache misses

commit 31c15a2f24 upstream.

Virtual Machines with emulated e1000 network adapter running on Parallels'
server were seeing kernel panics due to the e1000 driver dereferencing an
unexpected NULL pointer retrieved from buffer_info->skb.

The problem has been addressed for the e1000e driver, but not for the e1000.
Since the two drivers share similar code in the affected area, a port of the
following e1000e driver commit solves the issue for the e1000 driver:

commit 9ed318d546
Author: Tom Herbert <therbert@google.com>
Date:   Wed May 5 14:02:27 2010 +0000

    e1000e: save skb counts in TX to avoid cache misses

    In e1000_tx_map, precompute number of segements and bytecounts which
    are derived from fields in skb; these are stored in buffer_info.  When
    cleaning tx in e1000_clean_tx_irq use the values in the associated
    buffer_info for statistics counting, this eliminates cache misses
    on skb fields.

Signed-off-by: Dean Nelson <dnelson@redhat.com>
Acked-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Roman Kagan <rkagan@parallels.com>

2012-06-17 11:23:12 -07:00

e1000_ethtool.c

e1000: convert to set_phys_id

2011-05-07 03:10:13 -07:00

e1000_hw.c

e1000: Fix driver to be used on PA RISC C8000 workstations

2011-10-03 11:40:25 -07:00

e1000_hw.h

Fix common misspellings

2011-03-31 11:26:23 -03:00

e1000_main.c

e1000: save skb counts in TX to avoid cache misses