| 193726354173_511860007716 | Eric Dumazet |
|
From: Jarek Poplawski <jarkao2@xxxxxxxxx
Date: Sun, 29 Aug 2010 11:59:51 +0200
Actually, when GRO compares napi- dev to skb- dev?
Hmmm, I thought the code made a skb- dev comparison with the
|
|
| |
| 122329514696_570060007514 | Stephen Hemminger |
|
On 2010-08-29 20:39, Eric Dumazet wrote:
Le dimanche 29 aoĂťt 2010 Ă 10:06 -0700, David Miller a ĂŠcrit :
From: Jarek Poplawski <jarkao2@xxxxxxxxx
Date: Sun |
|
| |
| 107724514564_588260007766 | Stephen Hemminger |
|
From: Stephen Hemminger <shemminger@xxxxxxxxxx
Date: Mon, 30 Aug 2010 08:57:21 -0700
On Mon, 30 Aug 2010 06:42:31 +0000
Jarek Poplawski <jarkao2@xxxxxxxxx wrote:
On 2010-08-29 20:3 |
|
| |
| 111120144604_565560007056 | David Miller |
|
On Mon, Aug 30, 2010 at 10:51:17AM -0700, Stephen Hemminger wrote:
Theres something very important I forgot to tell you.
What?
Dont cross the GRO streams.
Why?
It would be bad |
|
| |
| 162328674135_586760007665 | Jarek Poplawski |
|
On Wed, 01 Sep 2010 14:51:51 -0700 (PDT)
David Miller <davem@xxxxxxxxxxxxx wrote:
From: Jarek Poplawski <jarkao2@xxxxxxxxx
Date: Mon, 30 Aug 2010 21:09:00 +0200
On Mon, Aug 30, 20 |
|
| |
| 178424284300_595860007758 | David Miller |
|
On Thu, Sep 02, 2010 at 09:18:39AM +0000, Jarek Poplawski wrote:
On Wed, Sep 01, 2010 at 02:55:54PM -0700, Stephen Hemminger wrote:
On Wed, 01 Sep 2010 14:51:51 -0700 (PDT)
David Miller < |
|
| |
| 167327704709_536960007269 | Jarek Poplawski |
|
On Thu, Sep 02, 2010 at 09:30:23AM -0700, David Miller wrote:
From: Jarek Poplawski <jarkao2@xxxxxxxxx
Date: Thu, 2 Sep 2010 12:53:26 +0000
Stephen, after Erics explanation, I really t |
|
| |
| 188325484720_544760007642 | Jarek Poplawski |
|
Le jeudi 02 septembre 2010 à 08:33 +0000, Jarek Poplawski a écrit :
On Wed, Sep 01, 2010 at 02:51:51PM -0700, David Miller wrote:
The only scenerio you can guarentee that all packets |
|
| |
| 121325414657_581760007970 | Jarek Poplawski |
|
Le jeudi 02 septembre 2010 à 09:55 +0000, Jarek Poplawski a écrit :
Exactly, but there is no "a single napi -- device mapping". And sky2
uses the same model. So, there is |
|
| |
| 187428194836_500860007327 | Jarek Poplawski |
|
Le jeudi 02 septembre 2010 à 11:02 +0000, Jarek Poplawski a écrit :
On Thu, Sep 02, 2010 at 12:41:45PM +0200, Eric Dumazet wrote:
Two vlans might carry packets in different domains, |
|
| |
| 167324704755_597460007301 | Herbert Xu |
|
From: Eric Dumazet <eric.dumazet@xxxxxxxxx
Date: Thu, 02 Sep 2010 12:41:45 +0200
[PATCH net-next-2.6] gro: remove use of napi- dev
Only use of napi- dev in GRO stack is the one found in n |
|
| |
| 173221004071_544060007541 | Jarek Poplawski |
|
Le vendredi 03 septembre 2010 à 05:26 +0800, Herbert Xu a écrit :
On Thu, Sep 02, 2010 at 12:41:45PM +0200, Eric Dumazet wrote:
[PATCH net-next-2.6] gro: remove use of napi- dev
|
|
| |
| 170923524088_542860007518 | Eric Dumazet |
|
On Mon, Aug 30, 2010 at 09:50:12AM -0700, David Miller wrote:
From: Stephen Hemminger <shemminger@xxxxxxxxxx
Date: Mon, 30 Aug 2010 08:57:21 -0700
On Mon, 30 Aug 2010 06:42:31 +0000
|
|
| |
| 107020124058_572760007254 | Eric Dumazet |
|
On Mon, 30 Aug 2010 21:59:16 +0200
Eric Dumazet <eric.dumazet@xxxxxxxxx wrote:
So everybody jumped on GRO, while my concern was more the napi- dev
thing.
I pointed to gro because the co |
|
| |
| 169727024072_553360007758 | David Miller |
|
The string clone is only used as a temporary copy of the argument val
within the while loop, and so it should be freed before leaving the
function. The call to strsep, however, modifies clone, so a |
|
| |
| 186023844228_561560007655 | Kyle Moffett |
|
Hello,
Im in the process of dusting off some old unidirectional fiber link
patches as part of support for a new custom hardware chassis.
Specifically, a critical piece of the hardware is unidirecti |
|
| |
| 191720884756_598860007225 | Kyle Moffett |
|
From: Kyle Moffett <Kyle.D.Moffett@xxxxxxxxxx
Date: Fri, 27 Aug 2010 15:42:17 -0400
A large variety of fiber ethernet PHYs and some copper ethernet PHYs
support forced "unidirectional l |
|
| |
| 152922374577_561760007377 | Stephen Hemminger |
|
Some sky2 fiber hardware seems to support configuring unidirectional
fiber links (using the phy bit PHY_M_FIB_FORCE_LNK). There are three
parts to enabling this hardware support:
(1) Allow DUPLEX |
|
| |
| 113527374389_519360007218 | Stephen Hemminger |
|
On Aug 27, 2010, at 16:38, Stephen Hemminger wrote:
On Fri, 27 Aug 2010 15:42:18 -0400 Kyle Moffett <Kyle.D.Moffett@xxxxxxxxxx wrote:
+ /*
+ * Once interrupts are reenabled, reset the P |
|
| |
| 144721604920_552060007631 | Kyle Moffett |
|
On Fri, Aug 27, 2010 at 17:22, Stephen Hemminger
<shemminger@xxxxxxxxxxxxxxxxxxxx wrote:
On Fri, 27 Aug 2010 15:51:58 -0500 "Moffett, Kyle D" <Kyle.D.Moffett@xxxxxxxxxx wrote:
|
|
| |
| 127825534014_550660007371 | Jesse Brandeburg |
|
On Fri, Aug 27, 2010 at 12:10, Kyle Moffett <Kyle.D.Moffett@xxxxxxxxxx wrote:
The Intel 82571EB chipset can be used in an unmanaged configuration as a
fast dual-port Gig-E controller. U |
|
| |
| 116120844459_591760007988 | David Miller |
|
|
|
| |
| 114126004225_571960007891 | Glen Turner |
|
From: Nicolas Dichtel <nicolas.dichtel@xxxxxxxxx
Date: Fri, 27 Aug 2010 17:47:40 +0200
From 1683b838a42429af30d6ab76d2d15d267c93c455 Mon Sep 17 00:00:00 2001
From: Nicolas Dichtel <nicola |
|
| |
| 109523534628_548460007091 | Eilon Greenstein |
|
bnx2x_trylock_hw_lock() returns a bool :
true if succeeded to acquire the lock.
false in case of error.
-EINVAL is not an acceptable value, since its promoted to true.
Signed-off-by: Eric Dumazet |
|
| |
| 170229944251_562260007944 | Eric Dumazet |
|
From: "Eilon Greenstein" <eilong@xxxxxxxxxxxx
Date: Fri, 27 Aug 2010 07:54:44 +0300
On Thu, 2010-08-26 at 20:49 -0700, Eric Dumazet wrote:
bnx2x_trylock_hw_lock() returns a bool : |
|
| |
| 125227924497_526760007665 | Anirban Chakraborty |
|
From: Sritej Velaga <sritej.velaga@xxxxxxxxxx
Added PCI ID to board info.
Signed-off-by: Sritej Velaga <sritej.velaga@xxxxxxxxxx
Signed-off-by: Anirban Chakraborty <anirban.chakraborty@x |
|
| |
| 163824584749_509760007821 | Anirban Chakraborty |
|
From: Sony Chacko <sony.chacko@xxxxxxxxxx
In the original code, the error codes returned from some of the functions
are not caught and sent up the caller chain. Fixed it here.
Signed-off-by: So |
|
| |
| 108525164421_531060007639 | Bernard F6BVP |
|
From: Anirban Chakraborty <anirban.chakraborty@xxxxxxxxxx
Date: Thu, 26 Aug 2010 17:02:31 -0700
Please apply the following series to net-next.
All applied, thank you.
--
To unsubscribe from t |
|
| |
| 120123294992_582960007537 | Christoph Lameter |
|
From: Bernard F6BVP <f6bvp@xxxxxxx
Date: Fri, 27 Aug 2010 00:15:01 +0200
When a ROSE node was deleted, node number was not
decremented and addr kept going up when a new node
was added.
|
|
| |
| 192025164530_517160007043 | Jason Gunthorpe |
|
From: Christoph Lameter <cl@xxxxxxxxx
Date: Thu, 26 Aug 2010 16:31:14 -0500 (CDT)
@@ -271,6 +271,13 @@ static void ipoib_ib_handle_rx_wc(struct
ipoib_ud_dma_unmap_rx(priv, mapping);
ip |
|
| |
| 111523914835_561260007132 | Jason Gunthorpe |
|
On Thu, 26 Aug 2010, Jason Gunthorpe wrote:
The 40 bytes at this location are defined by the HW specification to
be an IB GRH which has an identical layout to an IPv6 header. Roland
is right, |
|
| |
| 127622964591_502560007966 | David Miller |
|
On Thu, 26 Aug 2010, Jason Gunthorpe wrote:
I think doing the memcmp only in the multicast path should be
reasonable overhead wise.
Thats is not always possible. Here the multicast path is the
|
|
| |
| 184624784482_526660007716 | Yossi Etigin |
|
On Thu, 26 Aug 2010, David Miller wrote:
The highest cost is bringing in that packet headers cache line, which
youve already done by reading the byte and checking for 0xff.
And then you need t |
|
| |
| 190629914045_589760007859 | Christoph Lameter |
|
On Fri, Aug 27, 2010 at 03:02:54AM +0300, Yossi Etigin wrote:
Shouldnt struct ib_grh be packed to make this really work?
No idea what the kernel convention for this is. It looks OK to me, in
that |
|
| |
| 143327394375_596460007513 | David Miller |
|
|
|
| |
| 100720254904_584260007562 | David Miller |
|
|
|
| |
| 197529144858_556560007250 | Ron Mercer |
|
Qlge is freeing the buffers before stopping the card DMA, and
this can cause some severe error, as a EEH event on PPC.
This patch just stop the card and then free the resources.
Signed-off-by: Bren |
|
| |
| 153027854729_501160007518 | Ville Sundell |
|
From: Ron Mercer <ron.mercer@xxxxxxxxxx
Date: Thu, 26 Aug 2010 13:19:06 -0700
Breno,
looks good.
Thanks
Please do not top-post.
Signed-off-by: Ron Mercer <ron.mercer@xxxxxxxxxx
... |
|
| |
| 172829574508_577360007327 | Arthur Kepner |
|
On Thu, Aug 26, 2010 at 12:13 PM, Ville Sundell
<ville.sundell@xxxxxxxxxxx wrote:
This patch adds an interface to the ethtool subsystem and
implements read/write functionality for handling se |
|
| |
| 163222864066_500760007977 | Arthur Kepner |
|
From: Arthur Kepner <akepner@xxxxxxx
Date: Thu, 26 Aug 2010 10:06:46 -0700
Use a module parameter to limit the number of MSI-X vectors that
the mlx4_core driver will attempt to allocate.
|
|
| |
| 152022574001_542460007639 | Arthur Kepner |
|
From: Arthur Kepner <akepner@xxxxxxx
Date: Fri, 3 Sep 2010 13:30:44 -0700
I have that running on a (very small) system and it seems to
work OK. Will send out the patch in just a minute.
Fa |
|
| |
| 198620734552_584460007490 | Arthur Kepner |
|
From: Arthur Kepner <akepner@xxxxxxx
Date: Fri, 3 Sep 2010 15:13:03 -0700
On Fri, Sep 03, 2010 at 02:46:58PM -0700, David Miller wrote:
No, you really cant do it this way, IRQ allocation ma |
|
| |
| 109321684275_582360007326 | Simon Horman |
|
From: Arthur Kepner <akepner@xxxxxxx
Date: Mon, 6 Sep 2010 10:42:45 -0700
On Fri, Sep 03, 2010 at 07:30:46PM -0700, David Miller wrote:
This also means there would have to be notifications |
|
| |
| 134029084282_584960007134 | Julian Anastasov |
|
Cc: Xiaoyu Du <tingsrain@xxxxxxxxx
Signed-off-by: Julian Anastasov <ja@xxxxxx
Signed-off-by: Simon Horman <horms@xxxxxxxxxxxx
Index: nf-next-2.6/net/netfilter/ipvs/ip_vs_core.c
========= |
|
| |
| 117623134420_524260007670 | Simon Horman |
|
From: Julian Anastasov <ja@xxxxxx
Date: Thu, 26 Aug 2010 22:36:24 +0300 (EEST)
+ skb- csum_offset = offsetof(struct icmp6hdr, icmp6_cksum);
skb- ip_summed = CHECKSUM_UNNECESSARY;
O |
|
| |
| 121928414928_522060007026 | Simon Horman |
|
Also rename __ip_vs_securetcp_lock to ip_vs_securetcp_lock.
Spinlock conversion was suggested by Eric Dumazet.
Acked-by: Eric Dumazet <eric.dumazet@xxxxxxxxx
Signed-off-by: Simon Horman <hor |
|
| |
| 185922504099_541160007737 | Simon Horman |
|
From: Simon Horman <horms@xxxxxxxxxxxx
Date: Thu, 26 Aug 2010 21:54:27 +0900
Hi Dave,
as mentioned earlier today I have a handful of IPVS patches queued up.
Please consider them for net |
|
| |
| 190221874906_517660007085 | Sam Ravnborg |
|
---
Gigabit Ethernet driver of Topcliff PCH
Topcliff PCH is the platform controller hub that is going to be used in
Intels upcoming general embedded platform. All IO peripherals in
Topcliff PCH are |
|
| |
| 150223504993_578460007673 | Joe Perches |
|
Hi Sam
Thank you for your comment.
The replys comment was buried in following.
I will modify this patch.
Thanks, Ohtake(OKISEMI)
----- Original Message -----
From: "Sam Ravnborg" <sa |
|
| |
| 191923754986_533960007306 | Stephen Hemminger |
|
On Thu, 26 Aug 2010 18:56:55 +0900
Masayuki Ohtake <masa-korg@xxxxxxxxxxxxxxx wrote:
---
Gigabit Ethernet driver of Topcliff PCH
Topcliff PCH is the platform controller hub that is goin |
|
| |