[snip]
+/**
+ * dma_async_client_register - allocate and register a &dma_client
+ * @event_callback: callback for notification of channel addition/
removal
+ */
+struct dma_client *dma_async_client_register(dma_event_callback
event_callback)
+{
+ struct dma_client *client;
+
+ c
On Thu, 16 Mar 2006 19:29:32 -0800
Skunk Worx <[EMAIL PROTECTED]> wrote:
> Rick Jones wrote:
> >>
> >> From strace :
> >>
> >> 15:27:04.568800 recv(3, " >> ) = 555 <0.000121>
> >>
> >> vs.
> >>
> >> 15:18:24.515891 recv(3, " >> ) = 566 <0.038414>
> >>
> >> Will watch replies and post more when I
Rick Jones wrote:
From strace :
15:27:04.568800 recv(3, ") = 555 <0.000121>
vs.
15:18:24.515891 recv(3, ") = 566 <0.038414>
Will watch replies and post more when I know more. Kinda new at this.
Do you have the strace from the sending side?
Yes...I hope I'm not posting too much info
On Thu, 16 Mar 2006, Jeff Garzik wrote:
>
> Please pull from 'upstream-fixes' branch of
> master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/netdev-2.6.git
The commit comments for the Chelsio driver fix are a bit unfortunate.
The array clearly _does_ have three elements, it's just that the cod
Please pull from 'upstream-fixes' branch of
master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/netdev-2.6.git
to receive the following updates:
drivers/net/chelsio/sge.c|6 +++---
drivers/net/e100.c |5 -
drivers/net/e1000/e1000_main.c |8 +++
Scott Bardone wrote:
Adrian,
This is a bug. The array should contain 2 elements.
Attached is a patch which fixes it.
Thanks.
Signed-off-by: Scott Bardone <[EMAIL PROTECTED]>
applied. please avoid attachments and use a proper patch description in
the future. I had to hand-edit and hand-app
David S. Miller wrote:
return -E_NO_BIG_ENDIAN_TESTING;
[E1000]: Fix 4 missed endianness conversions on RX descriptor fields.
Signed-off-by: David S. Miller <[EMAIL PROTECTED]>
applied
-
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to [E
John W. Linville wrote:
The following changes since commit f13b83580acef03a36c785dccc534ccdd7e43084:
Adrian Bunk:
fs/namespace.c:dup_namespace(): fix a use after free
are found in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless-2.6.git
upstrea
Jesse Brandeburg wrote:
On 3/11/06, Jeff Garzik <[EMAIL PROTECTED]> wrote:
Jeff Kirsher wrote:
Jeff - should a patch be made for 2.6.16 also?
Yes, since this is a small fix and there aren't a ton of e100 changes, I
would prefer that you create a 'git pull' against 2.6.16-rc (latest
Linus gi
On Thu, 2006-03-16 at 08:51 -0500, jamal wrote:
> > BTW, in this example, the new hash I suggested would be as good
> > as the 2.6 case.
> >
>
> Yes, if you used 256 buckets per hash table ;->
No - you haven't understood what the new algorithm does.
It will get the same performance of the 2.6 v
From: "Chris Leech" <[EMAIL PROTECTED]>
Date: Thu, 16 Mar 2006 11:24:05 -0800
> > Thanks, that clarifies things. So, if I've understood correctly, the
> > benefit kicks in when:
> >
> > 1) I/OAT is enabled :)
> > 2) The user posts a recv() (or the like) of >= 2K
> > 3) There is >= 2K of data avai
Jeff Garzik <[EMAIL PROTECTED]> writes:
> Eric W. Biederman wrote:
>> There is no maintainer contact in the driver nor in the maintainers file.
>
>> It has clearly been bug fixed but simply didn't look like there
>> was a maintainer.
>
> Then you clearly did not look at all. Simply running the co
you are correct that memory is allocated on a syn with
this patch. however, instead of using a large
open_request struct and the other resources necessary
to track the connection's state, a struct of size 20 B
is allocated and stored in a hash table using the ISN
(syn_cookie value) as the index in
On 3/16/06, Scott Feldman <[EMAIL PROTECTED]> wrote:
> Do you have any data to share on header split? Also, can other non-
> Intel nics use I/OAT copy, and if so, is header-split a requirement
> for the copy?
I don't have any header-split data. The I/OAT copy offload will work
for any TCP traffi
The ip(8) command has a bug when dealing with IPoIB link layer
addresses. Specifically it does not correctly handle the addition of
new entries in the neighbor/arp table. For example, this command will
fail:
ip neigh add 192.168.0.138 lladdr
00:00:04:04:fe:80:00:00:00:00:00:00:00:01:73:00:00:
All,
Request for comments for the upstream inclusion of the qla3xxx driver.
This is a complementary network driver for our ISP4XXX parts. There is
a concurrent effort underway to get the iSCSI driver (qla4xxx)
integrated upstream as well.
The following files are included and have been posted to
Eric W. Biederman wrote:
There is no maintainer contact in the driver nor in the maintainers file.
It has clearly been bug fixed but simply didn't look like there
was a maintainer.
Then you clearly did not look at all. Simply running the command
git-whatchanged drivers/net/forcedet
On Mar 16, 2006, at 11:20 AM, Chris Leech wrote:
On 3/16/06, Leonid Grossman <[EMAIL PROTECTED]> wrote:
Hi Chris,
Do you know what part of the performance delta is contributed by the
offload for copy operations, and what part comes from other I/OAT
features like header separation, etc. ?
Thi
jensen galan wrote:
greetings!
this is my first creation of a patch for the linux
kernel. if you have time, could you please take a look
at it and give me some feedback.
this patch creates a syn_cache for caching TCP options
when syn_cookies are in use (by default, all TCP
options are lost when
"Ayaz Abdulla" <[EMAIL PROTECTED]> writes:
> Its hard for me to believe that you would think that there is no
> indication of the driver being actively supported. Take a look at the
> revision history in the forcedeth.c file. There have been 18 revisions
> in less then a year timeframe.
There is
Its hard for me to believe that you would think that there is no
indication of the driver being actively supported. Take a look at the
revision history in the forcedeth.c file. There have been 18 revisions
in less then a year timeframe.
Please file a bug in bugzilla, thats the intended use of that
"Ayaz Abdulla" <[EMAIL PROTECTED]> writes:
> What is the bugzilla bug number for that issue?
I don't think there is a bugzilla number as there was no indication
that the driver was actively supported, in which case the practical
solution is generally to dig in and fix a problem rather than whine
Begin forwarded message:
Date: Thu, 16 Mar 2006 07:22:54 -0800
From: [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Subject: [Bugme-new] [Bug 6233] New: race condition in tcp_sendmsg when
connection became established
http://bugzilla.kernel.org/show_bug.cgi?id=6233
Summary: race conditio
What is the bugzilla bug number for that issue?
I will work on clearing up some of those names.
-Original Message-
From: Eric W. Biederman [mailto:[EMAIL PROTECTED]
Sent: Thursday, March 16, 2006 9:58 AM
To: Ayaz Abdulla
Cc: Stephen Hemminger; "Manfred Spraul <[EMAIL PROTECTED], Jeff
Gar
Just a guess, but doesn't this put a drag on apps that are legitimately using
raw sockets?
-
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
> Thanks, that clarifies things. So, if I've understood correctly, the
> benefit kicks in when:
>
> 1) I/OAT is enabled :)
> 2) The user posts a recv() (or the like) of >= 2K
> 3) There is >= 2K of data available to give them
>
> yes?
Yes
-
To unsubscribe from this list: send the line "unsubscrib
Chris Leech wrote:
I must be missing something - if the MTU was 1500 bytes, how did the
receiver's offloaded copies get to the 2k level? Were several arriving
TCP segments aggregated?
Most of the overhead (get_user_pages) is per recv, not on a per packet
basis. Regardless of packet size, we
should have kept this on list
-- Forwarded message --
From: Chris Leech <[EMAIL PROTECTED]>
Date: Mar 16, 2006 11:13 AM
Subject: Re: I/OAT performance data
To: Rick Jones <[EMAIL PROTECTED]>
> I must be missing something - if the MTU was 1500 bytes, how did the
> receiver's offlo
oops, should have kept this on list
-- Forwarded message --
From: Chris Leech <[EMAIL PROTECTED]>
Date: Mar 16, 2006 10:56 AM
Subject: Re: I/OAT performance data
To: Rick Jones <[EMAIL PROTECTED]>
> When it says "buffer size" for the Chariot stuff, is that the socket
> buffer siz
On 3/16/06, Leonid Grossman <[EMAIL PROTECTED]> wrote:
> Hi Chris,
> Do you know what part of the performance delta is contributed by the
> offload for copy operations, and what part comes from other I/OAT
> features like header separation, etc. ?
This is showing the offloaded copy as the only dif
Hi Chris,
Do you know what part of the performance delta is contributed by the
offload for copy operations, and what part comes from other I/OAT
features like header separation, etc. ?
Thanks, Leonid
> -Original Message-
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED] On Behalf Of
Chris Leech wrote:
When it says "buffer size" for the Chariot stuff, is that the socket
buffer size, or the size of the buffer(s) being passed to the transport?
That's the I/O size for the application, being passed to the transport.
Was the MTU 1500 or 9000 bytes?
1500 byte MTU
Can the Ch
Observed problems when multiple processes request scans and subsequently
scan results. This causes a scan result request to hit card registers
before the scan is complete, returning an incomplete scan list and
possibly making the card very angry. Instead, cache the results of a
wireless scan and
Show the specific device that driver messages are about.
Signed-off-by: Dan Williams <[EMAIL PROTECTED]>
--- a/drivers/net/wireless/airo.c 2006-03-16 11:19:21.0 -0500
+++ b/drivers/net/wireless/airo.c 2006-03-16 11:19:27.0 -0500
@@ -1216,6 +1216,22 @@
static int flash
The number 2312 was used all over the place to refer to the card's
default MTU. Make it a #define and use that everywhere rather than the
number.
Signed-off-by: Dan Williams <[EMAIL PROTECTED]>
--- a/drivers/net/wireless/airo.c 2006-03-16 11:21:10.0 -0500
+++ b/drivers/net/wireless
Chris Leech wrote:
Sorry this took so long. The attached PDF show the benefit of I/OAT
for bulk data receives on 1, 2, 4, 6, and 8 gigabit Ethernet ports.
The baseline is a 2.6.15 kernel with an updated e1000 driver.
When it says "buffer size" for the Chariot stuff, is that the socket
buffer
I wonder if they would be more open to accepting that patch now?
- Greg Scott
-Original Message-
From: Stephen Hemminger [mailto:[EMAIL PROTECTED]
Sent: Thursday, March 16, 2006 11:55 AM
To: Chris Wedgwood
Cc: Greg Scott; Chuck Ebbert; linux-kernel; David S. Miller;
netdev@vger.kernel.
I thought that in the java case the problem only occurred because
TCP_NODELAY was set?
Perhaps/probably, since they were sending something in five spearate
peices and so were "semistuck" until cwnd grew to five packets from the
initial three.
However, had it not set TCP_NODELAY, the first of
Stephen Hemminger wrote:
Also, even if the TCP connection stalls for the ACK, why does it take multiple
miliseconds for the ACK to come back over the localhost?
Thanks,
Ben
NODELAY effects whether we send the second message right away (without waiting
for
the first ACK).
And it also disab
From strace :
15:27:04.568800 recv(3, "= 555 <0.000121>
vs.
15:18:24.515891 recv(3, "= 566 <0.038414>
Will watch replies and post more when I know more. Kinda new at this.
Do you have the strace from the sending side?
-
To unsubscribe from this list: send the line "unsubscribe netdev" i
>
> Also, even if the TCP connection stalls for the ACK, why does it take multiple
> miliseconds for the ACK to come back over the localhost?
>
> Thanks,
> Ben
>
NODELAY effects whether we send the second message right away (without waiting
for
the first ACK).
Normal (small writer):
"Ayaz Abdulla" <[EMAIL PROTECTED]> writes:
> Can we remove "Reverse Engineered" and "(EXPERIMENTAL)" now?
Can we get better bit names in the driver?
Having unknown xyz in half of the names is scare?
Can we have some help fixing the problem where the on some boots
the driver cannot receive packets
Stephen Hemminger wrote:
On Wed, 15 Mar 2006 20:13:01 -0800
Skunk Worx <[EMAIL PROTECTED]> wrote:
Hello,
I've taken a performance hit over localhost between kernels 2.6.14 and
2.6.15 in my client/server application.
I'm trying to gut things down to a simple test case, in the meantime,
thi
On Thu, 16 Mar 2006 08:07:43 -0800
Chris Wedgwood <[EMAIL PROTECTED]> wrote:
> On Mon, Mar 13, 2006 at 10:00:41AM -0800, Stephen Hemminger wrote:
>
> > There still is a bug in the 3c59x driver. It doesn't include any
> > code to handle changing the mac address. It will work if you take
> > the
Patches against jwl's wireless-2.6 tree.
[PATCH 1/3] clean up printk usage to print device name
[PATCH 2/3] define default MTU
[PATCH 3/3] cache wireless scans
Dan
-
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to [EMAIL PROTECTED]
More majordomo inf
This came up with java debugging already.
Interesting how this has come-up a second time isn't it.
The problem is when the sender writes a message in separate write()
system calls, each one becomes a separate packet.
And indeed, that is doubleplusungood and should be fixed in the applicati
On Thu, Mar 16, 2006 at 05:45:48PM +0100, Jiri Benc wrote:
> On Wed, 15 Mar 2006 16:36:16 -0800, Jouni Malinen wrote:
> See the patch below. Is it viable?
I'll test this with our low-level driver.
> > This and similar change for ieee80211_get_buffered_bc() add more
> > requirements for the low-l
Daniele,
Which version of linux are you testing with? We have test results of
linux high speed TCP options in
http://www.csc.ncsu.edu/faculty/rhee/export/bitcp/asteppaper.htm
> -Original Message-
> From: [EMAIL PROTECTED] [mailto:netdev-
> [EMAIL PROTECTED] On Behalf Of Daniele Lacamera
On Wed, 15 Mar 2006 16:36:16 -0800, Jouni Malinen wrote:
> In theory, the low-level driver can determine the needed mask itself.
> However, it would need to be somehow notified of allowed BSSID values.
> By removing this entry, this information would need to fetched from
> somewhere else before int
On Wed, 15 Mar 2006 15:58:45 -0800, Jouni Malinen wrote:
> Neither did I at first, but after some debugging, I would assume that
> you actually meant to return from ieee80211_ioctl_siwap() if the
> previous address is identical to the new one, not if it has changed.. In
> other words:
>
> [...]
>
Hi,
I am making some tests to compare some linux congestion control
algorithms performance with ns-2 modules ones.
Everything looks coherent, except for tcp_highspeed, which seems to
suffer of very bad performance in linux implementation in every
scenario, if compared with ns-2 results (and oft
On Mon, Mar 13, 2006 at 10:00:41AM -0800, Stephen Hemminger wrote:
> There still is a bug in the 3c59x driver. It doesn't include any
> code to handle changing the mac address. It will work if you take
> the device down, change address, then bring it up. But you shouldn't
> have to do that.
I s
On Thu, 2006-16-03 at 15:47 +1000, Russell Stuart wrote:
> On Wed, 2006-03-15 at 22:07 -0500, jamal wrote:
[..]
> With a divisor of 64 bytes, the 2.6 hash produces 6 bit
> quantity which enumerates to 64 unique values, and the 2.4
> hash produces 4 bits which enumerates to 16 unique values.
> E
Sorry, but I don't think this patch is needed. It is OK to add a route
with the RTF_EXPIRES flag set and rtmsg_info == 0. It's simply a route
that expires straight away. So there is no inconsistency in allowing this.
I agree
In fact if anything we should find a way to export the RTF_EXPIR
On Thu, Mar 16, 2006 at 12:22:17AM -0800, Jean-Mickael Guerin wrote:
>
> I'm resending the patch to fix the lifetime of IPv6 routes, you may want
> to include it in your tree or let me know if there is something wrong or
> incomplete.
Sorry, but I don't think this patch is needed. It is OK to a
On Thu, Mar 16, 2006 at 02:06:26AM -0800, David S. Miller ([EMAIL PROTECTED])
wrote:
> From: Evgeniy Polyakov <[EMAIL PROTECTED]>
> Date: Thu, 16 Mar 2006 13:02:41 +0300
>
> > Actually I added sock_async() checks to detect cases, when ucopy is set
> > somewhere, but socket was marked as async lat
From: Evgeniy Polyakov <[EMAIL PROTECTED]>
Date: Thu, 16 Mar 2006 13:02:41 +0300
> Actually I added sock_async() checks to detect cases, when ucopy is set
> somewhere, but socket was marked as async later, so I just use
> sock_async() where other cheks could be done.
Can you see that these two st
On Thu, Mar 16, 2006 at 01:45:20AM -0800, David S. Miller ([EMAIL PROTECTED])
wrote:
> From: Evgeniy Polyakov <[EMAIL PROTECTED]>
> Date: Thu, 16 Mar 2006 12:42:40 +0300
>
> > Why process should be there? It can request for data and sleep
> > away. ucopy is never used for async sockets - all ne
From: Evgeniy Polyakov <[EMAIL PROTECTED]>
Date: Thu, 16 Mar 2006 12:42:40 +0300
> Why process should be there? It can request for data and sleep
> away. ucopy is never used for async sockets - all needed for
> transfer data is stored inside kevents.
The point is that if tp->ucopy.task is non-N
On Thu, Mar 16, 2006 at 01:22:59AM -0800, David S. Miller ([EMAIL PROTECTED])
wrote:
> From: Evgeniy Polyakov <[EMAIL PROTECTED]>
> Date: Thu, 16 Mar 2006 12:07:38 +0300
>
> > Prequeue is turned off for async sockets, since process can be
> > not scheduled when new data arrives or is being sent,
On Thu, Mar 16, 2006 at 01:09:02AM -0800, Andrew Morton wrote:
> Mark Brown <[EMAIL PROTECTED]> wrote:
> > + if (np->ignore_phy && (ecmd->autoneg == AUTONEG_ENABLE ||
> > + ecmd->port == PORT_INTERNAL)) {
> What's PORT_INTERNAL? ethtool doesn't appear to define that.
From: Evgeniy Polyakov <[EMAIL PROTECTED]>
Date: Thu, 16 Mar 2006 12:07:38 +0300
> Prequeue is turned off for async sockets, since process can be
> not scheduled when new data arrives or is being sent, and data is not
> copied through usual datagram methods, but directly to/from pinned
> userspac
Mark Brown <[EMAIL PROTECTED]> wrote:
>
> +if (np->ignore_phy && (ecmd->autoneg == AUTONEG_ENABLE ||
> + ecmd->port == PORT_INTERNAL)) {
What's PORT_INTERNAL? ethtool doesn't appear to define that.
drivers/net/natsemi.c: In function `netdev_set_ecmd':
drivers/net/
> > 4) I have 2 major problems with tcp_async_{send,recv}().
> >
> >a) Much duplication of existing tcp sendmsg/recvmsg code.
> > Consolidation and code sharing is needed.
>
> Yes, especially receiving code - it is 99% what sendpage() does.
recvpage() of course.
--
Evgeniy Po
On Thu, Mar 16, 2006 at 12:36:22AM -0800, David S. Miller ([EMAIL PROTECTED])
wrote:
> From: Evgeniy Polyakov <[EMAIL PROTECTED]>
> Date: Sat, 18 Feb 2006 19:11:40 +0300
>
> > Hello developers.
> >
> > I'm pleased to announce combined patch of kevent and network AIO subsytems,
> > which are impl
From: Evgeniy Polyakov <[EMAIL PROTECTED]>
Date: Sat, 18 Feb 2006 19:11:40 +0300
> Hello developers.
>
> I'm pleased to announce combined patch of kevent and network AIO subsytems,
> which are implemented using three system calls:
> * kevent_ctl(), which fully controls kevent subsystem.
>It
From: Ingo Oeser <[EMAIL PROTECTED]>
Date: Fri, 10 Mar 2006 23:34:26 +0100
> Here are some possible (and trivial) cleanups.
> - use kzalloc() where possible
> - invert allocation failure test like
> if (object) {
> /* Rest of function here */
> }
> to
>
> if (object == NULL)
>
From: Ingo Oeser <[EMAIL PROTECTED]>
Date: Sat, 11 Mar 2006 21:36:42 +0100
> Stupidly use kzalloc() instead of kmalloc()/memset()
> everywhere where this is possible in net/ipv6/*.c .
>
> Signed-off-by: Ingo Oeser <[EMAIL PROTECTED]>
This patch also looks fine.
Applied, thanks Ingo.
-
To unsu
From: Ingo Oeser <[EMAIL PROTECTED]>
Date: Sun, 12 Mar 2006 00:49:48 +0100
> Two minor cleanups:
>
> 1. Using kzalloc() in fraq_alloc_queue()
>saves the memset() in ipv6_frag_create().
>
> 2. Invert sense of if-statements to streamline code.
>Inverts the comment, too.
>
> Signed-off-by
69 matches
Mail list logo
