The attachment contains an UNTESTED patch (though a similar patch was tested
with a
3.10 kernel).
We've been chasing a bug where packet corruption is seen on a tap device. We
have a
PACKET_MMAP socket which is bound to a tap interface. When throughput goes
above a
threshold, we begin to se
On Thu, 16 Feb 2006, Jesse Brandeburg wrote:
>
> should i test this patch to see if it fixes the pktgen panic when
> removing a network driver module (or renegotiating link, i don't
> remember where i saw the failure) in the middle of a pktgen run?
>
Yes, please.
Removing the network dri
On Thu, 16 Feb 2006, Robert Olsson wrote:
> .
> >if(remove) {
>
> > + t->control |= T_REMDEV;
> > + pkt_dev->removal_mark = 1;
>
> >}
>
> Guess you should mark b
an be freed once again in pktgen_stop_device(). This
patch gives all responsibility for cleaning up a
pktgen_dev's skb to the associated worker thread.
Signed-off-by: Arthur Kepner <[EMAIL PROTECTED]>
---
diff --git a/net/core/pktgen.c b/net/core/pktgen.c
--- a/net/core/pktgen.c
+++
On Wed, 15 Feb 2006, David S. Miller wrote:
>
> I don't see how this fixes the race
>
Well, of course. I was just checking that you were
paying attention.
It looks like the right thing to do is to let the
worker thread free up the skb in all cases. New
patch soon.
--
Arthur
-
To un
removes the race by setting the pktgen_dev's
skb to NULL before freeing it.
Signed-off-by: Arthur Kepner <[EMAIL PROTECTED]>
---
diff --git a/net/core/pktgen.c b/net/core/pktgen.c
--- a/net/core/pktgen.c
+++ b/net/core/pktgen.c
@@ -2431,6 +2431,8 @@ static void show_result
The mac_addr variable doesn't get reset between
(re)additions of multicast addresses. One byte
of all multicast addresses (except the first)
can be incorrect.
Signed-off-by: Arthur Kepner <[EMAIL PROTECTED]>
--
Arthur
--- linux-2.6.orig/drivers/net/s2io.c 2006-01-24 04:26:09.4250
ufficient.
>
How does the following look?
---
Add NETIF_F_TSO (NETIF_F_UFO) to BOND_INTERSECT_FEATURES so that it can
be used by a bonding device iff all its slave devices support TSO (UFO).
Signed-off-by: Arthur Kepner <[EMAIL PROTECTED]>
--
Arthurdiff --git a/drivers/net/bonding/bond_main.
On Mon, 30 Jan 2006, Jay Vosburgh wrote:
>
> I believe that bond_compute_features() also needs logic to
> disable NETIF_F_TSO if NETIF_F_SG is not set.
>
But if each slave device supports TSO, then each slave device
also must support SG, and therefore the bond device will support
SG
Add NETIF_F_TSO to BOND_INTERSECT_FEATURES so that it can be
used by a bonding device iff all its slave devices support TSO.
Signed-off-by: Arthur Kepner <[EMAIL PROTECTED]>
--
Arthurdiff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
--- a/drivers/net/b
The mac_addr variable doesn't get reset between
(re)additions of multicast addresses. One byte
of all multicast addresses (except the first)
can be incorrect.
Signed-off-by: Arthur Kepner <[EMAIL PROTECTED]>
--
Arthur
--- linux-2.6.orig/drivers/net/s2io.c 2006-01-24 04:26:09.4250
On Wed, 18 Jan 2006, David S. Miller wrote:
>
> I thought we had fixed this already? Guess now :)
>
There was a patch some months ago to allow a bonding
device to use zero-copy if all its slaves did - same
idea (and implemented with the same code.)
--
Arthur
-
To unsubscribe from thi
Add NETIF_F_TSO to BOND_INTERSECT_FEATURES so that it can be
used by a bonding device iff all its slave devices support TSO.
Signed-off-by: Arthur Kepner <[EMAIL PROTECTED]>
--
Arthurdiff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
--- a/drivers/net/b
accepting IP fragments, which can greatly reduce the
possibility of reassembling fragments which originated
from different IP datagrams.
Signed off by: Arthur Kepner <[EMAIL PROTECTED]>
diffstat:
Documentation/networking/ip-sysctl.txt | 23 +++
include/linux/sysctl.h |
On Tue, 16 Aug 2005, Herbert Xu wrote:
> .
> So my preference for this isn't really based on particularly strong
> reasons. However, as I couldn't see any strong reason for it to be
> more strict either I'd rather stay with the one since it's slightly
> simpler to code.
>
Fair enough.
--
On Tue, 16 Aug 2005, Herbert Xu wrote:
> On Thu, Aug 04, 2005 at 05:48:17PM -0700, Arthur Kepner wrote:
> >
> > @@ -366,6 +376,9 @@ static struct ipq *ip_frag_create(unsign
> > .
> > + qp->peer = sysctl_ipfrag_max_dist ? inet_getpeer(iph->saddr
On Sun, 31 Jul 2005, Herbert Xu wrote:
>
> Anyway, here is a patch to use inetpeer instead of that icky ipc
> structure.? It sure cuts down the size of the patch :)
>
Herbert;
I've been using the updated patch, and I like it. struct inet_peer
is the right place to do this accounting.
I m
?l=bonding-devel&m=111925242706297&w=2
and it's largely a combination of the patches that appear in those
emails.
Patch is against 2.6.13-rc5.
Signed-off-by: Arthur Kepner <[EMAIL PROTECTED]>
--
Arthurdiff -pur linux.orig/drivers/net/bonding/bonding.h
linux.new/drivers/net/bonding/bonding
On Sun, 31 Jul 2005, Herbert Xu wrote:
> .
> Anyway, here is a patch to use inetpeer instead of that icky ipc
> structure. It sure cuts down the size of the patch :)
Icky?? (That's my baby you're talking about! :)
The patch size reduction is impressive.
>
> I've only compile-tested this
On Sat, 30 Jul 2005, Herbert Xu wrote:
>
> I've finished my patch based on your work. Unfortunately the result
> isn't as clean as I would've liked. However, it does point out a couple
> of problems with your patch more clearly. Perhaps you could work on
> those and come up with a better solu
On Sun, 31 Jul 2005, David S. Miller wrote:
> .
> I would also suggest that some reasonable non-zero default max_win
> be choosen. This stuff will not get tested otherwise.
>
Agree with this.
(And I'll comment on the previous several mails in this thread once
I've had time to review mor
net/ipv4/ip_fragment.c | 195 +
net/ipv4/sysctl_net_ipv4.c | 11 ++
Signed-off-by: Arthur Kepner <[EMAIL PROTECTED]>
On Tue, 28 Jun 2005, Arthur Kepner wrote:
>
> On Sun, 26 Jun 2005, Herbert Xu wrote:
>
&g
On Thu, 7 Jul 2005, Raghavendra Koushik wrote:
>
> On an Altix machine I believe the readq was necessary to flush
> the PIO writes. How long did you run the tests? I had seen
> in long duration tests that an occasional write
> (TXDL control word and the address) would be missed and the xmit
On Thu, 7 Jul 2005 [EMAIL PROTECTED] wrote:
> ...
> 2. Removed unnecessary PIOs(read/write of tx_traffic_int and
>rx_traffic_int) from interrupt handler and removed read of
>general_int_status register from xmit routine.
> ..
> @@ -2891,6 +2869,8 @@ int s
Version 2 of the rfc/patch is attached. It has been changed
as indicated in the commentary below.
Diffstat:
include/linux/sysctl.h |1
net/ipv4/ip_fragment.c | 195 +
net/ipv4/sysctl_net_ipv4.c | 11 ++
Signed-off-by: Arthur Kepner
25 matches
Mail list logo
