Hi Net Gurus,
This thread came up on kgdb-bugreport mailing list. Could you please suggest
us what's the correct way of fixing this problem?
1. When running a kgdb on RTL8139 ethernet interface: 8139too driver prints
too many "Out-of-sync dirty pointer" messages on console and gdb can't
connec
From: John Heffner <[EMAIL PROTECTED]>
Date: Thu, 22 Feb 2007 16:52:03 -0500
> I think it's not unreasonable to change clamp to 32 bits now, since with
> 1500 byte packets, this corresponds to a max cwnd of ~94MB. This is
> pretty big, but we are currently right at this limit with 10 GigE.
Agr
John, what tree did you diff this against?
I can tell you didn't create this patch against anything
actually in any of my trees, because:
diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c
index 7fd2910..a0c894f 100644
--- a/net/ipv4/tcp_cong.c
+++ b/net/ipv4/tcp_cong.c
@@ -303,9 +303,9 @@ v
Hello.
I have cooked up new git tree for IPv6 development.
It is available as branch named
2.6.21-rc1-net-2.6-20070223-FOR_DAVEM-20070223
at
.
I will shift to new branch time to time (e.g. every -rc releases) in order
to chase the latest tree.
Regards,
--yoshfuji
-
To unsubscrib
On 22 Feb 2007 18:49:00 -0500, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
The rehash-every-10-minutes detail is theoretically unnecessary,
but does cover the case where a would-be attacker *does* get a chance
to look at a machine, such as by using routing delays to measure the
effectiveness of
> I think you misunderstood me. If you are trying to DoS me from
> outside with a hash collision attack, you are trying to feed me
> packets that fall into the same hash bucket. The Jenkins hash does
> not have to be artifact-free, and does not have to be
> cryptographically strong. It just has
Here is an experimental patch that changes the packet scheduler to use
ktime instead of gettimeofday. This should be faster on 64 bit and avoid some of
the math overhead issues with previous code.
Also since it uses monotonic clock, it won't cause timing glitches when NTP
adjusts the clock.
---
Steve Wise wrote:
Divy,
Do these need to be pulled into OFED 1.2 as well?
Hi Steve,
Yes, I believe so.
Cheers,
Divy
Steve.
On Thu, 2007-02-22 at 03:58 -0800, Divy Le Ray wrote:
Jeff,
I'm sending a series of incremental patches updating
the cxgb3 driver. These patches are built aga
Ilpo Järvinen wrote:
BTW, while looking this patch, I noticed that snd_cwnd_clamp is only u16
while snd_cwnd is u32, which seems rather strange since snd_cwnd is being
limited by the clamp value here and there?!?! And tcp_highspeed.c is
clearly assuming even more than this (but the problem is h
Begin forwarded message:
Date: Thu, 22 Feb 2007 07:56:27 -0800
From: [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Subject: [Bugme-new] [Bug 8057] New: slab corruption running ip6sic
http://bugzilla.kernel.org/show_bug.cgi?id=8057
Summary: slab corruption running ip6sic
Kernel Versio
On Thu, 22 Feb 2007, John Heffner wrote:
> diff --git a/net/ipv4/tcp_cong.c b/net/ipv4/tcp_cong.c
> index 7fd2910..a0c894f 100644
> --- a/net/ipv4/tcp_cong.c
> +++ b/net/ipv4/tcp_cong.c
> @@ -303,9 +303,9 @@ void tcp_slow_start(struct tcp_sock *tp)
>
> tp->snd_cwnd_cnt += cnt;
>
> On Thu, 22 Feb 2007 04:20:10 -0800 [EMAIL PROTECTED] wrote:
> http://bugzilla.kernel.org/show_bug.cgi?id=8054
>
>Summary: tipc_ref_discard tipc_deleteport locking dependency
> Kernel Version: 2.6.21-rc1
> Status: NEW
> Severity: normal
> Owner:
On Wed, Feb 21, 2007 at 04:04:41PM -0800, Stephen Hemminger wrote:
> I compared the current sky2 driver against the version in 2.6.16.40.
> These were the only two patches that should be added.
> They both relate to dual port boards, and not to many people have
> them.
Thanks, I've applied both.
Minor comments:
1. Please move div64_64 out of yeah-tcp (and cubic where you copied it).
to asm-generic/div64.h
2. Don't need separate tcp_yeah.h just put it in tcp_yeah.c. Also, maybe
the vegas stuff you copied should go in one place? tcp/vegas.h?
3. whitespace:
don't do:
On Thu, 22 Feb 2007 03:07:55 -0800 (PST)
David Miller <[EMAIL PROTECTED]> wrote:
> From: "Michael K. Edwards" <[EMAIL PROTECTED]>
> Date: Thu, 22 Feb 2007 03:00:32 -0800
>
> > Besides, RCUing a hash of any sort sounds like a nightmare, whereas
> > something like a splay tree is reputed to be very
Fix arithmetic order bug in limited slow start. The subtraction needs to be
done before snd_cwnd is incremented.
Signed-off-by: John Heffner <[EMAIL PROTECTED]>
---
commit 244e7411d99443df7b7ae849ba6ebbec4c2342bc
tree e6d5985a22448f59f8bef393542e1d5497ee5684
parent 97033fa201705e6cfc68ce66f34ed
> BTW, there was a query about this back in Feb. 2006 on linux-kernel,
> nobody replied, he reposted to linux-net in September 2006 and this
> is likely where this kernel bugzilla comes from :-)
>
> This is not a kernel bug, let's close this and move on.
Agreed closed
--
Stephen Hemminger <[EM
David Miller wrote:
> From: weidong <[EMAIL PROTECTED]>
> Date: Wed, 14 Feb 2007 11:30:57 -0500
>
>> when I use linux TCP socket, and find there is a bug in function
>> sk_acceptq_is_full().
> ...
>> Consider an example:
>>
>> After listen(sockfd, 1) system call, sk->sk_max_ack_backlog is
Thanks, I'll check for formatting errors in the future.
2007/2/22, David Miller <[EMAIL PROTECTED]>:
Applied to tcp-2.6 GIT, there were a lot of broken whitespace
issues in your patch, GIT makes mention of them quite happily
even without applying the patch, which I recommend you do in the
future
Ok, I've missed this bit in the submitting patches documentation..
however sorry for that.
2007/2/22, David Miller <[EMAIL PROTECTED]>:
Please never submit patches like this, submit the infrastructure
FIRST, then submit the stuff that uses it. When a sequence of patches
is applied, in sequence,
On Wed, 2007-02-21 at 08:27 -0500, John W. Linville wrote:
> On Wed, Feb 21, 2007 at 12:46:46PM +0100, Roger While wrote:
> >
> > >
> > >Dan Williams (1):
> > > prism54: correct assignment of DOT1XENABLE in WE-19 codepaths
> >
> > Where did this spring from ?
> > I see no posting of this pat
On Thu, Feb 22, 2007 at 03:17:40PM +, Frederik Deweerdt ([EMAIL PROTECTED])
wrote:
> On Thu, Feb 22, 2007 at 04:54:52PM +0300, Evgeniy Polyakov wrote:
>
> > Changes from 'take36' patchset:
> > * ported to 2.6.21-rc1
> Hi Evgeniy,
>
> The changes below were needed to be able to compile take
On Thu, Feb 22, 2007 at 04:54:52PM +0300, Evgeniy Polyakov wrote:
> Changes from 'take36' patchset:
> * ported to 2.6.21-rc1
Hi Evgeniy,
The changes below were needed to be able to compile take37 on 2.6.21-rc1
with:
CONFIG_KEVENT=y
CONFIG_KEVENT_USER_STAT=y
CONFIG_KEVENT_TIMER=y
CONFIG_KEVENT_P
Hi David
sk_backlog is a critical field of struct sock. (known famous words)
It is (ab)used in hot paths, in particular in release_sock(), tcp_recvmsg(),
tcp_v4_rcv(), sk_receive_skb().
It really makes sense to place it next to sk_lock, because sk_backlog is only
used after sk_lock locked (and
From: John Heffner <[EMAIL PROTECTED]>
Date: Thu, 22 Feb 2007 09:25:07 -0500
> My patch is meant as a replacement for YeAH patch 2/2, not meant to back
> it out. You do still need the second hunk below. Sorry 'bout that.
>
> If you're going to apply YeAH patch 2/2 first, you will also need to
My patch is meant as a replacement for YeAH patch 2/2, not meant to back
it out. You do still need the second hunk below. Sorry 'bout that.
If you're going to apply YeAH patch 2/2 first, you will also need to
remove the declaration of tcp_limited_slow_start() in include/net/tcp.h.
Thanks,
Divy,
Do these need to be pulled into OFED 1.2 as well?
Steve.
On Thu, 2007-02-22 at 03:58 -0800, Divy Le Ray wrote:
> Jeff,
>
> I'm sending a series of incremental patches updating
> the cxgb3 driver. These patches are built against Linus'git tree.
>
> Cheers,
> Divy
> -
> To unsubscribe fro
In article <[EMAIL PROTECTED]> (at Thu, 22 Feb 2007 04:50:05 -0800), Divy Le
Ray <[EMAIL PROTECTED]> says:
> > Even if you remove them, PLEASE DO NOT CHANGE THE VALUES,
> > unless you have very, very good reason.
:
> The cxgb3 driver has not yet appeared in a stable version of the linux
> kernel
From: Evgeniy Polyakov <[EMAIL PROTECTED]>
Date: Thu, 22 Feb 2007 16:46:21 +0300
> On Thu, Feb 22, 2007 at 05:31:38AM -0800, David Miller ([EMAIL PROTECTED])
> wrote:
> > TCP_CLOSE is where we end up on a non-error close too, this has
> > the same kind of bug as your previous attempt to set POLLE
Kevent based generic AIO.
This patch only implements network AIO, which is _COMPLETELY_
impossible and broken in _ANY_ micro-thread design. For details
and test consider following link:
http://tservice.net.ru/~s0mbre/blog/2007/02/10#2007_02_10
Designing AIO without network in mind can only be
Kevent posix timer notifications.
Simple extension to POSIX timers which allows
to deliver notification of the timer expiration
through kevent queue.
Example application posix_timer.c can be found
in archive on project homepage.
Signed-off-by: Evgeniy Polyakov <[EMAIL PROTECTED]>
diff --git a
Private userspace notifications.
Allows to register notifications of any private userspace
events over kevent. Events can be marked as ready using
kevent_ctl(KEVENT_READY) command.
Signed-off-by: Evgeniy Polyakov <[EMAIL PROTECTED]>
diff --git a/kernel/kevent/kevent_unotify.c b/kernel/kevent/k
poll/select() notifications.
This patch includes generic poll/select notifications.
kevent_poll works simialr to epoll and has the same issues (callback
is invoked not from internal state machine of the caller, but through
process awake, a lot of allocations and so on).
Signed-off-by: Evgeniy Po
Socket notifications.
This patch includes socket send/recv/accept notifications.
Signed-off-by: Evgeniy Polyakov <[EMAIL PROTECTED]>
diff --git a/fs/inode.c b/fs/inode.c
index 5abb097..b60fa63 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -21,6 +21,7 @@
#include
#include
#include
+#include
Generic event handling mechanism (socket, pipes, files, signals, AIO, timers,
posix timers...).
Kevent is a generic subsytem which allows to handle event notifications.
It supports both level and edge triggered events. It is similar to
poll/epoll in some cases, but it is more scalable, it is fas
Description.
diff --git a/Documentation/kevent.txt b/Documentation/kevent.txt
new file mode 100644
index 000..ee1c381
--- /dev/null
+++ b/Documentation/kevent.txt
@@ -0,0 +1,276 @@
+Description.
+
+Kevent documentation
+
+
+http://tservice.net.ru/~s0mbre/old/?section=proj
Pipe notifications.
diff --git a/fs/pipe.c b/fs/pipe.c
index ebafde7..f4a4c22 100644
--- a/fs/pipe.c
+++ b/fs/pipe.c
@@ -16,6 +16,7 @@
#include
#include
#include
+#include
#include
#include
@@ -314,6 +315,7 @@ redo:
break;
}
if
Signal notifications.
This type of notifications allows to deliver signals through kevent queue.
One can find example application signal.c on project homepage.
If KEVENT_SIGNAL_NOMASK bit is set in raw_u64 id then signal will be
delivered only through queue, otherwise both delivery types are use
Timer notifications.
Timer notifications can be used for fine grained per-process time
management, since interval timers are very inconvenient to use,
and they are limited.
This subsystem uses high-resolution timers.
id.raw[0] is used as number of seconds
id.raw[1] is used as number of nanose
On Thu, Feb 22, 2007 at 05:31:38AM -0800, David Miller ([EMAIL PROTECTED])
wrote:
> TCP_CLOSE is where we end up on a non-error close too, this has
> the same kind of bug as your previous attempt to set POLLERR
> here.
>
> One side gets TCP_TIMEWAIT the other goes straight to TCP_CLOSE.
>
> It r
From: "Ilpo_Järvinen" <[EMAIL PROTECTED]>
Date: Thu, 22 Feb 2007 14:25:32 +0200 (EET)
> Before preparing the submittable spurious RTO responses patch, I would
> like to know what is the preferred style for the response selector code
> (I'm controlling the response through frto_response sysctl)?
From: Evgeniy Polyakov <[EMAIL PROTECTED]>
Date: Thu, 22 Feb 2007 15:14:27 +0300
> if (sk->sk_shutdown == SHUTDOWN_MASK || sk->sk_state == TCP_CLOSE)
> mask |= POLLHUP;
>
> to
>
> if (sk->sk_shutdown == SHUTDOWN_MASK || sk->sk_state == TCP_CLOSE)
> mask |= POLLHUP | (sk->sk_state ==
YOSHIFUJI Hideaki <[EMAIL PROTECTED]> wrote:
> This sockaddr_rxrpc{} should NOT include sockaddr_in{} directly.
> Please use sockaddr_storage{} (or sockaddr{}, maybe), and make it
> sure to align on 64-bit word.
How about this then:
struct sockaddr_rxrpc {
sa_family_t srx_family;
YOSHIFUJI Hideaki / [EMAIL PROTECTED] wrote:
In article <[EMAIL PROTECTED]> (at Thu, 22 Feb 2007 04:34:34 -0800), Divy Le Ray
<[EMAIL PROTECTED]> says:
1. Please do do change the values.
2. Explain why you are going to kill these ioctls.
Jeff and Arjan have required that most of
"Hua Zhong" <[EMAIL PROTECTED]> writes:
> Correction: it's TSO, not TOE.
>
> By googling I found that specifying a big MTU (like 16500) on the "tc class"
> command avoids the "giant" packets. No idea where they come from.
They come from TCP because it will pass down a max MTU sized packet
(upto
"Hua Zhong" <[EMAIL PROTECTED]> writes:
> Just add more info: this machine has TOE. And if that's turned off,
> everything seems to work fine.
>
> Does it mean that traffic control doesn't play well with TOE? Or there are
> some tricks to make them work together?
Congratulations. You discovered
In article <[EMAIL PROTECTED]> (at Thu, 22 Feb 2007 04:34:34 -0800), Divy Le
Ray <[EMAIL PROTECTED]> says:
> > 1. Please do do change the values.
> > 2. Explain why you are going to kill these ioctls.
> >
> Jeff and Arjan have required that most of these ioctls go.
> Please read:
> http://mar
Hi Yoshfuji,
From: Divy Le Ray <[EMAIL PROTECTED]>
Clean up some private ioctls.
Signed-off-by: Divy Le Ray <[EMAIL PROTECTED]>
---
drivers/net/cxgb3/cxgb3_ioctl.h | 21 -
drivers/net/cxgb3/cxgb3_main.c | 48 +++
2 files changed, 9 inse
On Thu, 22 Feb 2007, David Miller wrote:
> If you have any further bug fixes or enhancements to FRTO we
> can put it there too.
I tried the rate-halving approach in spurious RTO response and it looks
really nice indeed. I'll try to also run couple of tests with the
undo_cwr(,1) response too in
Hi Dave,
@@ -919,10 +920,7 @@ int dev_close(struct net_device *dev)
* engine, but this requires more changes in devices. */
smp_mb__after_clear_bit(); /* Commit netif_running(). */
- while (test_bit(__LINK_STATE_RX_SCHED, &dev->state)) {
- /* No hurry. */
-
Hello,
This is a rather strange e-mail for these mailing
lists, I know. I am a third year Social Anthropology
student in the University of Durham doing my
dissertation (thesis) on the Anthropology of
GNU/Linux. I would really appreciate if you could help
me out and offer some of your time to fill
On Wed, Feb 21, 2007 at 10:45:43PM -0800, David Miller ([EMAIL PROTECTED])
wrote:
> POLLHUP is a complete mess, because it means different things on
> different kinds of fds. And exactly, because it is not maskable,
> this makes it totally useless for sockets.
POLLHUP does show that socket can n
In article <[EMAIL PROTECTED]> (at Thu, 22 Feb 2007 03:59:25 -0800), [EMAIL
PROTECTED] says:
> From: Divy Le Ray <[EMAIL PROTECTED]>
>
> Clean up some private ioctls.
>
> Signed-off-by: Divy Le Ray <[EMAIL PROTECTED]>
> ---
>
> drivers/net/cxgb3/cxgb3_ioctl.h | 21 -
> drive
Hello,
I am working on optimizing the TCP-code for a certain type of TCP-stream,
and to make one of my optimizations work I need to copy data from one SKB
(the data field of the skb) to another SKB (data field).
Currently I am using memcpy, and it does what it is supposed to, but I am
curious as
From: Divy Le Ray <[EMAIL PROTECTED]>
Offload packets may be DMAed long after their SGE Tx descriptors are done
so they must remain mapped until they are freed rather than until their
descriptors are freed. Unmap such packets through an skb destructor.
Signed-off-by: Divy Le Ray <[EMAIL PROTECTE
From: Divy Le Ray <[EMAIL PROTECTED]>
Add all-in-sw lro support.
Signed-off-by: Divy Le Ray <[EMAIL PROTECTED]>
---
drivers/net/cxgb3/adapter.h | 21 ++
drivers/net/cxgb3/common.h |1
drivers/net/cxgb3/cxgb3_ioctl.h |1
drivers/net/cxgb3/cxgb3_main.c | 16 ++
drivers/net
From: Divy Le Ray <[EMAIL PROTECTED]>
Update FW version to 3.2
Signed-off-by: Divy Le Ray <[EMAIL PROTECTED]>
---
drivers/net/cxgb3/t3_hw.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/net/cxgb3/t3_hw.c b/drivers/net/cxgb3/t3_hw.c
index 365a7f5..ec06ad6 1006
From: Divy Le Ray <[EMAIL PROTECTED]>
Populate Rx free list with pages.
Signed-off-by: Divy Le Ray <[EMAIL PROTECTED]>
---
drivers/net/cxgb3/adapter.h |9 +
drivers/net/cxgb3/sge.c | 322 +++
2 files changed, 237 insertions(+), 94 deletions(-)
d
From: Divy Le Ray <[EMAIL PROTECTED]>
Improve the traffic recovery after the HW ran out of response queue entries.
Signed-off-by: Divy Le Ray <[EMAIL PROTECTED]>
---
drivers/net/cxgb3/adapter.h |2 ++
drivers/net/cxgb3/sge.c | 15 ++-
2 files changed, 16 insertions(+), 1 d
From: Divy Le Ray <[EMAIL PROTECTED]>
Clean up some private ioctls.
Signed-off-by: Divy Le Ray <[EMAIL PROTECTED]>
---
drivers/net/cxgb3/cxgb3_ioctl.h | 21 -
drivers/net/cxgb3/cxgb3_main.c | 48 +++
2 files changed, 9 insertions(+), 60 d
From: Divy Le Ray <[EMAIL PROTECTED]>
sysfs attributes are now managed per port, no longer per adapter.
Signed-off-by: Divy Le Ray <[EMAIL PROTECTED]>
---
drivers/net/cxgb3/cxgb3_main.c | 21 -
1 files changed, 12 insertions(+), 9 deletions(-)
diff --git a/drivers/net/cxg
Jeff,
I'm sending a series of incremental patches updating
the cxgb3 driver. These patches are built against Linus'git tree.
Cheers,
Divy
-
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.o
From: Andi Kleen <[EMAIL PROTECTED]>
Date: Thu, 22 Feb 2007 12:41:05 +0100
> > With 100k flows we typically see an aver depth 1.3 and a max depth 4.
> > Right
>
> This includes full srcport:dstport?
Yes.
-
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a mess
> With 100k flows we typically see an aver depth 1.3 and a max depth 4. Right
This includes full srcport:dstport?
> now there is only a dst entry in leaf nodes. So yes anything else we can put
> in leaf is "free".
How much would fit there?
-Andi
-
To unsubscribe from this list: send the l
From: Eric Dumazet <[EMAIL PROTECTED]>
Date: Thu, 22 Feb 2007 12:11:50 +0100
> I noticed in oprofile study a cache miss in tcp_rcv_established() to read
> copied_seq.
...
> As you suggested, I changed tcp_create_openreq_child() so that these fields
> are changed together, to avoid adding a new
From: Benjamin Herrenschmidt <[EMAIL PROTECTED]>
Date: Thu, 22 Feb 2007 10:24:31 +1100
> The device has a single interrupt though that interrupt at least can
> tell you which queues need servicing. It can't mask the interrupt per
> queue though, which is the main issue.
Right, which effectively t
From: weidong <[EMAIL PROTECTED]>
Date: Wed, 14 Feb 2007 11:30:57 -0500
> when I use linux TCP socket, and find there is a bug in function
> sk_acceptq_is_full().
...
> Consider an example:
>
> After listen(sockfd, 1) system call, sk->sk_max_ack_backlog is set to
> 1. As we know, sk->sk_
Hi David
I noticed in oprofile study a cache miss in tcp_rcv_established() to read
copied_seq.
80400a80 : /* tcp_rcv_established total: 4034293
2.0400 */
55493 0.0281 :80400bc9: mov 0x4c8(%r12),%eax copied_seq
543103 0.2746 :80400bd1: cmp 0x3e0(%r12),%eax
From: "Michael K. Edwards" <[EMAIL PROTECTED]>
Date: Thu, 22 Feb 2007 03:00:32 -0800
> Besides, RCUing a hash of any sort sounds like a nightmare, whereas
> something like a splay tree is reputed to be very straightforward to
> make "persistent" in the functional-programming sense. I haven't
> co
From: Eric Dumazet <[EMAIL PROTECTED]>
Date: Thu, 22 Feb 2007 11:43:32 +0100
> I wonder if :
>
> newtp->snd_nxt = newtp->snd_una = newtp->snd_sml = treq->snt_isn + 1;
>
> is correct, since compiler will write snd_sml, then snd_una, and snd_nxt
>
> Maybe we should also reorder this too ?
Yes, t
On 2/22/07, David Miller <[EMAIL PROTECTED]> wrote:
From: "Michael K. Edwards" <[EMAIL PROTECTED]>
Date: Wed, 21 Feb 2007 13:15:51 -0800
> it had better be a 2-left hash with a compact
> overflow pool for the rare case of second collision.
Just like Evgeniy, you should do your research too :-))
On Thursday 22 February 2007 11:32, David Miller wrote:
> From: Eric Dumazet <[EMAIL PROTECTED]>
> Date: Thu, 22 Feb 2007 11:22:02 +0100
>
> > rcv_wup and copied_seq should be next to rcv_nxt field, to lower number
> > of active cache lines in hot paths. (tcp_rcv_established(), tcp_poll(),
> > ...)
From: Eric Dumazet <[EMAIL PROTECTED]>
Date: Thu, 22 Feb 2007 11:22:02 +0100
> rcv_wup and copied_seq should be next to rcv_nxt field, to lower number of
> active cache lines in hot paths. (tcp_rcv_established(), tcp_poll(), ...)
Please fixup the assignment order in tcp_create_openreq_child() el
Hi David
I noticed in oprofile study a cache miss in tcp_rcv_established() to read
copied_seq.
80400a80 : /* tcp_rcv_established total: 4034293
2.0400 */
55493 0.0281 :80400bc9: mov0x4c8(%r12),%eax copied_seq
543103 0.2746 :80400bd1: cmp0x3e0(%r12),%eax
On 2/22/07, Pekka Enberg <[EMAIL PROTECTED]> wrote:
So you are only interested in rough estimation of how much many pages
you need for a given amount of objects? Why not use ksize() for that
then?
Uhm, I obviously meant, why not expose obj_size() instead.
-
To unsubscribe from this list: send t
On Thu, 2007-02-22 at 10:16 +0100, Peter Zijlstra wrote:
> On Wed, 2007-02-21 at 16:53 +0100, Arjan van de Ven wrote:
> > > Index: linux-2.6-git/kernel/softirq.c
> > > ===
> > > --- linux-2.6-git.orig/kernel/softirq.c 2006-12-14 10:0
Hi Peter,
On Wed, 2007-02-21 at 17:47 +0200, Pekka Enberg wrote:
> So how does this work? You ask the slab allocator how many pages you
> need for a given number of objects and then those pages are available
> to it via the page allocator? Can other users also dip into those
> reserves?
On 2/2
On Wed, 2007-02-21 at 17:47 +0200, Pekka Enberg wrote:
> Hi Peter,
>
> On 2/21/07, Peter Zijlstra <[EMAIL PROTECTED]> wrote:
> > Provide a method to calculate the number of pages needed to store a given
> > number of slab objects (upper bound when considering possible partial and
> > free slabs).
From: Joe Perches <[EMAIL PROTECTED]>
Date: Thu, 15 Feb 2007 15:46:44 -0800
> Einar Lueck's email addresses bounce
> <[EMAIL PROTECTED]><[EMAIL PROTECTED]>
>
> Removed local random number generator function
>
> Signed-off-by: Joe Perches <[EMAIL PROTECTED]>
Applied, thanks Joe.
-
To unsubscribe
On Wed, 2007-02-21 at 16:53 +0100, Arjan van de Ven wrote:
> > Index: linux-2.6-git/kernel/softirq.c
> > ===
> > --- linux-2.6-git.orig/kernel/softirq.c 2006-12-14 10:02:18.0
> > +0100
> > +++ linux-2.6-git/kernel/softirq.
From: "Ilpo_Järvinen" <[EMAIL PROTECTED]>
Date: Tue, 20 Feb 2007 17:17:02 +0200 (EET)
> In general, TCP code uses "sk" for struct sock pointer.
>
> Signed-off-by: Ilpo Järvinen <[EMAIL PROTECTED]>
Applied to tcp-2.6, thank you.
-
To unsubscribe from this list: send the line "unsubscribe netdev"
From: Stephen Hemminger <[EMAIL PROTECTED]>
Date: Wed, 21 Feb 2007 14:41:07 -0800
> Several patches to bridge. #1 is a nit, #2 is a bug fix,
> #3 and #4 are manageability enhancements (they can wait for
> 2.6.22 if needed).
I've added #1 and #2, #3 and #4 are for 2.6.22
-
To unsubscribe from this
From: Stephen Hemminger <[EMAIL PROTECTED]>
Date: Wed, 21 Feb 2007 14:41:09 -0800
> Having a work queue for checking carrier leads to lots of race issues.
> Simpler to just get the cost when data structure is created and
> update on change.
>
> Signed-off-by: StephenHemminger <[EMAIL PROTECTED]>
From: Stephen Hemminger <[EMAIL PROTECTED]>
Date: Wed, 21 Feb 2007 14:41:08 -0800
> The bridge hasn't used miscdevice for a long long time.
Applied, thanks Stephen.
> Signed-off-by: StephenHemminger <[EMAIL PROTECTED]>
You don't type
From: "Michael K. Edwards" <[EMAIL PROTECTED]>
Date: Wed, 21 Feb 2007 13:15:51 -0800
> it had better be a 2-left hash with a compact
> overflow pool for the rare case of second collision.
Just like Evgeniy, you should do your research too :-
The gain for multiple-hash schemes, such as 2-left
On Wed, Feb 21, 2007 at 10:55:55AM -0800, Stephen Hemminger wrote:
> This is what I was suggesting by getting rid of the work queue completely.
...
> --- bridge.orig/net/bridge/br_if.c2007-02-21 10:22:46.0 -0800
> +++ bridge/net/bridge/br_if.c 2007-02-21 10:53:25.0 -0800
> @@ -7
From: David Miller <[EMAIL PROTECTED]>
Date: Thu, 22 Feb 2007 00:27:04 -0800 (PST)
> I'll apply this, but could you please also when making suggestions
> like this provide the patch necessary to kill the function added for
> YeaH and the call site in the YeaH algorithm?
Here is how I'm resolving
From: John Heffner <[EMAIL PROTECTED]>
Date: Mon, 19 Feb 2007 16:00:06 -0500
> I'd prefer to make it apply automatically across all congestion controls
> that do slow-start, and also make the max_ssthresh parameter
> controllable via sysctl. This patch (attached) should implement this.
> Note
From: "Angelo P. Castellani" <[EMAIL PROTECTED]>
Date: Mon, 19 Feb 2007 11:35:09 +0100
> From: Angelo P. Castellani <[EMAIL PROTECTED]>
>
> RFC3742: limited slow start
>
> See http://www.ietf.org/rfc/rfc3742.txt
>
> Signed-off-by: Angelo P. Castellani <[EMAIL PROTECTED]>
> ---
>
> To allow cod
From: "Angelo P. Castellani" <[EMAIL PROTECTED]>
Date: Mon, 19 Feb 2007 11:30:58 +0100
> From: Angelo P. Castellani <[EMAIL PROTECTED]>
>
> YeAH-TCP is a sender-side high-speed enabled TCP congestion control
> algorithm, which uses a mixed loss/delay approach to compute the
> congestion window.
In article <[EMAIL PROTECTED]> (at Thu, 22 Feb 2007 00:12:45 -0800 (PST)),
David Miller <[EMAIL PROTECTED]> says:
> From: YOSHIFUJI Hideaki / 吉藤英明 <[EMAIL PROTECTED]>
> Date: Thu, 22 Feb 2007 16:48:52 +0900 (JST)
>
> > I think it's time to distribute all EXPORT_SYMBOL
> > things in ipv6_syms.c
From: Ilpo Järvinen <[EMAIL PROTECTED]>
Date: Sat, 17 Feb 2007 22:25:48 +0200
> Here is a set of patches that fixes most of the flaws the current FRTO
> implementation (specified in RFC4138) has, besides that, the last two
> patches add SACK-enhanced FRTO (not enabled unless frto sysctl is set
> t
From: YOSHIFUJI Hideaki / 吉藤英明 <[EMAIL PROTECTED]>
Date: Thu, 22 Feb 2007 16:48:52 +0900 (JST)
> I think it's time to distribute all EXPORT_SYMBOL
> things in ipv6_syms.c to each file which each symbol is
> actually defined. Don't you think so?
I very much agree. Let's do this in 2.6.22
-
To
Edward added to Cc:
Dirk <[EMAIL PROTECTED]> :
[...]
Can you try the patch below against 2.6.21-rc1:
http://www.fr.zoreil.com/people/francois/misc/20070221-2.6.21-rc1-r8169-test.patch
A serie of incremental patches for the same changes is available at:
http://www.fr.zoreil.com/linux/kernel/2.6.
94 matches
Mail list logo
