Skip to content

K
kernel_linux

OpenHarmony / kernel_linux
上一次同步 3 年多

通知 13
Star 8
Fork 2
K
kernel_linux
切换分支/标签
  1. 15 11月, 2012 1 次提交
  3. 07 11月, 2012 2 次提交
  5. 03 10月, 2012 1 次提交
    • M
      tg3: Fix sparse warnings. · 86449944
      由 Michael Chan 提交于 
      drivers/net/ethernet/broadcom/tg3.c:8121:8: warning: symbol 'i' shadows an earlier one
drivers/net/ethernet/broadcom/tg3.c:8003:6: originally declared here
drivers/net/ethernet/broadcom/tg3.c:785:5: warning: symbol 'tg3_ape_scratchpad_read' was not declared. Should it be static?
drivers/net/ethernet/broadcom/tg3.c:7781:19: warning: Using plain integer as NULL pointer
drivers/net/ethernet/broadcom/tg3.c:10231:31: error: bad constant expression
Reported-by: NFengguang Wu <fenguang.wu@intel.com>
Signed-off-by: NMichael Chan <mchan@broadcom.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      86449944
  7. 02 10月, 2012 1 次提交
  9. 30 9月, 2012 7 次提交
  11. 08 9月, 2012 1 次提交
  13. 24 8月, 2012 1 次提交
  15. 30 7月, 2012 5 次提交
  17. 17 7月, 2012 4 次提交
  19. 13 7月, 2012 1 次提交
  21. 05 7月, 2012 1 次提交
  23. 14 6月, 2012 1 次提交
  25. 12 6月, 2012 1 次提交
  27. 19 5月, 2012 1 次提交
  29. 01 5月, 2012 2 次提交
    • E
      tg3: provide frags as skb head · 8d4057a9
      由 Eric Dumazet 提交于 
      This patch converts tg3 driver, one of our reference drivers, to use new
build_skb() api in frag mode.

Instead of using kmalloc() to allocate the memory block that will be
used by build_skb() as skb->head, we use a page fragment.

This is a followup of patch "net: allow skb->head to be a page fragment"

This allows GRO, TCP coalescing, and splice() to be more efficient.

Incidentally, this also removes SLUB slow path contention in kfree()
Signed-off-by: NEric Dumazet <edumazet@google.com>
Cc: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: Maciej Żenczykowski <maze@google.com>
Cc: Neal Cardwell <ncardwell@google.com>
Cc: Tom Herbert <therbert@google.com>
Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Cc: Ben Hutchings <bhutchings@solarflare.com>
Cc: Matt Carlson <mcarlson@broadcom.com>
Cc: Michael Chan <mchan@broadcom.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      8d4057a9
    • E
      net: allow skb->head to be a page fragment · d3836f21
      由 Eric Dumazet 提交于 
      skb->head is currently allocated from kmalloc(). This is convenient but
has the drawback the data cannot be converted to a page fragment if
needed.

We have three spots were it hurts :

1) GRO aggregation

 When a linear skb must be appended to another skb, GRO uses the
frag_list fallback, very inefficient since we keep all struct sk_buff
around. So drivers enabling GRO but delivering linear skbs to network
stack aren't enabling full GRO power.

2) splice(socket -> pipe).

 We must copy the linear part to a page fragment.
 This kind of defeats splice() purpose (zero copy claim)

3) TCP coalescing.

 Recently introduced, this permits to group several contiguous segments
into a single skb. This shortens queue lengths and save kernel memory,
and greatly reduce probabilities of TCP collapses. This coalescing
doesnt work on linear skbs (or we would need to copy data, this would be
too slow)

Given all these issues, the following patch introduces the possibility
of having skb->head be a fragment in itself. We use a new skb flag,
skb->head_frag to carry this information.

build_skb() is changed to accept a frag_size argument. Drivers willing
to provide a page fragment instead of kmalloc() data will set a non zero
value, set to the fragment size.

Then, on situations we need to convert the skb head to a frag in itself,
we can check if skb->head_frag is set and avoid the copies or various
fallbacks we have.

This means drivers currently using frags could be updated to avoid the
current skb->head allocation and reduce their memory footprint (aka skb
truesize). (thats 512 or 1024 bytes saved per skb). This also makes
bpf/netfilter faster since the 'first frag' will be part of skb linear
part, no need to copy data.
Signed-off-by: NEric Dumazet <edumazet@google.com>
Cc: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
Cc: Maciej Żenczykowski <maze@google.com>
Cc: Neal Cardwell <ncardwell@google.com>
Cc: Tom Herbert <therbert@google.com>
Cc: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Cc: Ben Hutchings <bhutchings@solarflare.com>
Cc: Matt Carlson <mcarlson@broadcom.com>
Cc: Michael Chan <mchan@broadcom.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      d3836f21
  31. 30 4月, 2012 1 次提交
  33. 26 4月, 2012 1 次提交
  35. 04 4月, 2012 1 次提交
  37. 03 4月, 2012 1 次提交
  39. 29 3月, 2012 1 次提交
  41. 22 3月, 2012 1 次提交
    • M
      tg3: Fix RSS ring refill race condition · 7ae52890
      由 Michael Chan 提交于 
      The RSS feature in tg3 hardware has only one rx producer ring for all
RSS rings.  NAPI vector 1 is special and handles the refilling of the
rx producer ring on behalf of all RSS rings.  There is a race condition
between these RSS NAPIs and the NAPI[1].  If NAPI[1] finishes checking
for refill and then another RSS ring empties the rx producer ring
before NAPI[1] exits NAPI, the chip will be completely out of SKBs in
the rx producer ring.

We fix this by adding a flag tp->rx_refill and rely on napi_schedule()/
napi_complete() to help synchronize it to close the race condition.

Update driver version to 3.123.
Signed-off-by: NMichael Chan <mchan@broadcom.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      7ae52890
  43. 06 3月, 2012 1 次提交
  45. 05 3月, 2012 3 次提交