((weak)) in the declaration. linker
should do the job.
Signed-off-by: Eric Dumazet [EMAIL PROTECTED]
--- linux-2.6.19/kernel/timer.c 2006-12-11 11:25:50.0 +0100
+++ linux-2.6.19-ed/kernel/timer.c 2006-12-11 11:31:30.0 +0100
@@ -1020,11 +1020,9 @@ static inline void calc_load
and nr_uninterruptible of all online CPUS, bringing foreign dirty
cache lines.
This patch is an optimization of calc_load() so that nr_active() is called
only if we need it.
The use of unlikely() is welcome since the condition is true only once every
5*HZ time.
Signed-off-by: Eric Dumazet [EMAIL
pipe/splice should use const pipe_buf_operations and file_operations
struct pipe_inode_info has an unused field start : get rid of it.
Signed-off-by: Eric Dumazet [EMAIL PROTECTED]
--- linux-2.6.19/include/linux/pipe_fs_i.h 2006-12-11 17:00:21.0
+0100
+++ linux-2.6.19-ed/include
Andrew Morton a écrit :
hm, the patch seems to transform a mess into a mess. I guess it's a messy
problem.
I agree that aggregating all the time-related things into a struct like
this makes some sense. As does aggregating them all into a similar-looking
namespace, but that'd probably be too
vmlinux.pre vmlinux
textdata bss dec hex filename
3268989 664356 492196 4425541 438745 vmlinux.pre
3268765 664356 492196 4425317 438665 vmlinux
So this patch reduces text size by 224 bytes on my x86_64 machine. Similar
results on ia32.
Signed-off-by: Eric Dumazet [EMAIL
).
Signed-off-by: Eric Dumazet [EMAIL PROTECTED]
--- linux-2.6.19/include/linux/jiffies.h2006-12-12 00:32:00.0
+0100
+++ linux-2.6.19-ed/include/linux/jiffies.h 2006-12-12 00:41:40.0
+0100
@@ -80,6 +80,11 @@
*/
extern u64 __jiffy_data jiffies_64;
extern unsigned
David Miller a écrit :
From: Eric Dumazet [EMAIL PROTECTED]
Date: Mon, 11 Dec 2006 23:58:06 +0100
Some subsystems dont need more than 32bits timestamps.
See for example net/ipv4/inetpeer.c and include/net/tcp.h :
#define tcp_time_stamp((__u32)(jiffies))
Because most timeouts
David Miller a écrit :
From: Eric Dumazet [EMAIL PROTECTED]
Date: Tue, 12 Dec 2006 04:47:14 +0100
I doubt being able to extend the expiration of a dst above 2^32
ticks (49 days if HZ=1000, 198 days if HZ=250) is worth the ram
wastage.
And this doesn't apply for all jiffies uses because
David Miller a écrit :
From: Eric Dumazet [EMAIL PROTECTED]
Date: Tue, 12 Dec 2006 05:09:23 +0100
We definitly *like* being able to use bigger timeouts on 64bits platforms.
Not that they are mandatory since the same application should run fine on
32bits kernel. But as the standard type
/jiffies in time_data. This patch does the thing for i386 and
x86_64.
avenrun, xtime, xtime_lock, wall_to_monotonic, are now temporary defined as
macros to make this patch not too invasive, but we can in future patches
gradually deletes these macros.
Signed-off-by: Eric Dumazet [EMAIL PROTECTED
Andrew Morton a écrit :
On Wed, 13 Dec 2006 16:12:46 -0800
Greg KH [EMAIL PROTECTED] wrote:
Original comment seemed to indicate that this conditional thing was
performance related. Is it really? If not, we should consider the below patch.
Yes, it's a performance gain and I don't see how this
applies to 2.6.19-rc4 and has been tested.
This patch needs testing on a K8 NUMA platform.
Thanks to Eric Dumazet and Andi Kleen for their improvement suggestions.
I had the patch in, but had to drop it again because it makes one of my
test system triple fault. Haven't done much investigation
Andrew Morton a écrit :
On Thu, 23 Nov 2006 11:57:29 +0100
Eric Dumazet [EMAIL PROTECTED] wrote:
On 32bits SMP platforms, 64bits i_size is protected by a seqcount
(i_size_seqcount).
When i_size is read or written, i_size_seqcount is read/written as well, so it
make sense to group these two
Andrew Morton a écrit :
This all depends on the offset of the inode, and you don't know what that is.
offsetof(ext3_inode_info, vfs_inode) != offsetof(nfs_inode, vfs_inode), etc.
Doh... yes you are absolutly right :) I feel dumb now :(
-
To unsubscribe from this list: send the line
On Wednesday 29 November 2006 00:12, Jeremy Fitzhardinge wrote:
Hi Eric,
Could you try this patch out and see if it makes much performance
difference for you. You should apply this on top of the %fs patch I
posted earlier (and use the %fs patch as the baseline for your
comparisons).
Hi
On Thursday 30 November 2006 02:25, Paul E. McKenney wrote:
On Wed, Nov 22, 2006 at 04:02:29PM +0100, Eric Dumazet wrote:
On some workloads, (for example when lot of close() syscalls are done),
RCU qlen can be quite large, and RCU heads are no longer in cpu cache
when rcu_do_batch
Oleg Nesterov a écrit :
On top of rcu-add-a-prefetch-in-rcu_do_batch.patch
rcu_do_batch:
struct rcu_head *next, *list;
while (list) {
next = list-next; -- [1]
list-func(list);
list = next;
}
We can't trust *list
Oleg Nesterov a écrit :
On 12/03, Eric Dumazet wrote:
Oleg Nesterov a ?crit :
On top of rcu-add-a-prefetch-in-rcu_do_batch.patch
rcu_do_batch:
struct rcu_head *next, *list;
while (list) {
next = list-next; -- [1]
list-func(list
Oleg Nesterov a écrit :
On 12/03, Eric Dumazet wrote:
Oleg Nesterov a ?crit :
Yes, but how is it related to RCU ?
I mean, rcu_do_batch() is just a loop like others in kernel.
The loop itself is not buggy, but can call a buggy function, you are right.
int start_me_again
),%ebx
Signed-off-by: Eric Dumazet [EMAIL PROTECTED]
--- linux-2.6.19/mm/slab.c 2006-12-04 11:50:19.0 +0100
+++ linux-2.6.19-ed/mm/slab.c 2006-12-04 17:25:02.0 +0100
@@ -371,6 +371,19 @@ static void kmem_list3_init(struct kmem_
} while (0)
/*
+ * Define the reciprocal
(%esp) // useless
mov (%esp),%ebx
Signed-off-by: Eric Dumazet [EMAIL PROTECTED]
--- linux-2.6.19/include/linux/reciprocal_div.h 1970-01-01 01:00:00.0
+0100
+++ linux-2.6.19-ed/include/linux/reciprocal_div.h 2006-12-04
19:01:44.0 +0100
@@ -0,0 +1,30 @@
+#ifndef
) // useless
mov(%esp),%ebx
Signed-off-by: Eric Dumazet [EMAIL PROTECTED]
--- linux-2.6.19/include/linux/reciprocal_div.h 1970-01-01 01:00:00.0
+0100
+++ linux-2.6.19-ed/include/linux/reciprocal_div.h 2006-12-04
23:12:34.0 +0100
@@ -0,0 +1,30 @@
+#ifndef
David Miller a écrit :
From: Eric Dumazet [EMAIL PROTECTED]
Date: Mon, 04 Dec 2006 22:34:29 +0100
On a 200 MHz sparcv9 machine, the division takes 64 cycles instead of 1 cycle
for a multiply.
For UltraSPARC I and II (which is what this 200mhz guy probably is),
it's 4 cycle latency
Philippe Elie a écrit :
On Fri, 09 Sep 2005 at 11:23 +, Andi Kleen wrote:
Indeed. Someone must have fixed it. But why would anyone want frame pointers
on x86-64?
Oprofile can use it, I though it was already used but apparently only
to backtrace userspace actually.
Hi Pilippe
Last
On Sun, 2012-09-09 at 17:03 +0200, Davidlohr Bueso wrote:
Account for properties when a and/or b are 0:
gcd(0, 0) = 0
gcd(a, 0) = a
gcd(0, b) = b
Cc: sta...@vger.kernel.org
Signed-off-by: Davidlohr Bueso d...@gnu.org
---
lib/gcd.c |3 +++
1 file changed, 3 insertions(+)
diff
On Mon, 2012-09-10 at 09:44 +0100, ch...@chris-wilson.co.uk wrote:
I've not seen this reported yet, so here's a warning that happens
occasionally:
[192979.475833]
[192979.475840] ===
[192979.475841] [ INFO: suspicious RCU usage. ]
[192979.475844] 3.6.0-rc2+ #33
On Mon, 2012-09-10 at 16:13 +0300, Kirill A. Shutemov wrote:
From: Kirill A. Shutemov kirill.shute...@linux.intel.com
H. Peter Anvin doesn't like huge zero page which sticks in memory forever
after the first allocation. Here's implementation of lockless refcounting
for huge zero page.
...
On Mon, 2012-09-10 at 17:44 +0300, Kirill A. Shutemov wrote:
On Mon, Sep 10, 2012 at 04:02:39PM +0200, Eric Dumazet wrote:
On Mon, 2012-09-10 at 16:13 +0300, Kirill A. Shutemov wrote:
From: Kirill A. Shutemov kirill.shute...@linux.intel.com
H. Peter Anvin doesn't like huge zero page
On Mon, 2012-09-10 at 17:44 +0300, Kirill A. Shutemov wrote:
Yes, disabling preemption before alloc_pages() and enabling after
atomic_set() looks reasonable. Thanks.
In fact, as alloc_pages(GFP_TRANSHUGE | __GFP_ZERO, HPAGE_PMD_ORDER);
might sleep, it would be better to disable preemption
From: Eric Dumazet eduma...@google.com
We currently use a per socket page reserve for tcp_sendmsg() operations.
This page is used to build fragments for skbs.
Its done to increase probability of coalescing small write() into
single segments in skbs still in write queue (not yet sent
, bnx2x, tg3, mellanox mlx4)
Signed-off-by: Eric Dumazet eduma...@google.com
Cc: Ben Hutchings bhutchi...@solarflare.com
Cc: Vijay Subramanian subramanian.vi...@gmail.com
Cc: Alexander Duyck alexander.h.du...@intel.com
---
v2: uses existing page_frag structure to hold page/offset/size
convert
On Fri, 2012-09-21 at 13:27 -0700, Vijay Subramanian wrote:
I get the following compile error with the newer version of the patch
net/sched/em_meta.c: In function ‘meta_int_sk_sendmsg_off’:
net/sched/em_meta.c:464: error: ‘struct sock’ has no member named
‘sk_sndmsg_off’
make[1]: ***
On Tue, 2012-09-11 at 16:25 -0700, Andrew Morton wrote:
Asking for a 256k allocation is pretty crazy - this is an operating
system kernel, not a userspace application.
I'm wondering if this is due to a recent change, but I'm having trouble
working out where the allocation call site is.
--
On Wed, 2012-09-12 at 00:46 -0700, Joe Perches wrote:
auto75914...@hushmail.com reports that iptables does not correctly
output the KERN_level.
$IPTABLES -A RULE_0_in -j LOG --log-level notice --log-prefix DENY in:
result with linux 3.6-rc5
Sep 12 06:37:29 x kernel: 5DENY in:
On Wed, 2012-09-12 at 16:13 +0200, Stephane Eranian wrote:
void hrtimer_init_on_stack(struct hrtimer *timer, clockid_t clock_id,
enum hrtimer_mode mode)
{
debug_object_init_on_stack(timer, hrtimer_debug_descr);
- __hrtimer_init(timer, clock_id, mode);
+
On Thu, 2012-09-13 at 17:16 +0800, Fengguang Wu wrote:
Hi Patrick,
This happens in today's linux-next tree and is pretty reproducible.
Bisection has been started.
[1.834544] nf_conntrack version 0.5.0 (1786 buckets, 7144 max)
[1.835406] ctnetlink v0.93: registering with nfnetlink.
On Thu, 2012-09-13 at 09:43 -0400, Cyril Chemparathy wrote:
This patch fixes an apparent bug in the running weighted average calculation
used in the RED algorithm.
Going by the described formula:
qavg = qavg*(1-W) + backlog*W
= qavg = qavg + (backlog - qavg) * W
... with W
On Fri, 2012-09-14 at 13:01 +, Dowdal, John wrote:
Eric, thank you for reviewing the code. I now see the problem with
the patch since backlog is an integer and qavg is a fixed point number
at logW.
We are considering another patch to update the comments to this code
(with the actual
On Fri, 2012-07-27 at 22:09 -0500, sheng qiu wrote:
Hi all,
i am comparing network throughput performance under bare-metal case
with that running VM with assigned-device (assigned NIC). i have two
physical machines (each has a 10Gbit NIC), one is used as remote
server (run netserver) and
On Sat, 2012-07-28 at 12:41 -0400, Mikulas Patocka wrote:
Introduce percpu rw semaphores
When many CPUs are locking a rw semaphore for read concurrently, cache
line bouncing occurs. When a CPU acquires rw semaphore for read, the
CPU writes to the cache line holding the semaphore.
On Sun, 2012-07-29 at 01:13 -0400, Mikulas Patocka wrote:
Each cpu should have its own rw semaphore in its cache, so I don't see a
problem there.
When you change block size, all 4096 rw semaphores are locked for write,
but changing block size is not a performance sensitive operation.
On Sun, 2012-07-29 at 12:10 +0200, Eric Dumazet wrote:
You can probably design something needing no more than 4 bytes per cpu,
and this thing could use non locked operations as bonus.
like the following ...
Coming back from my bike ride, here is a more polished version with
proper
On Tue, 2012-08-28 at 23:38 +0300, Alexey Dobriyan wrote:
Nothing can stop RCU!
After running modprobe;rmmod in a loop and cat in another loop for a while
rmmod got stuck in D-state inside remove_proc_entry() with trace amounts of
CPU time
being consumed.
It didn't oopsed, though.
On Tue, 2012-08-28 at 21:34 -0700, H.K. Jerry Chu wrote:
IMHO 31secs seem a little short. Why not change it to 6 as well because 63
secs still beats 93secs with 3sec initRTO and 5 retries.
Jerry
My rationale was that such increase were going to amplify SYN attacks
impact by 20% (if we
On Wed, 2012-08-29 at 16:50 +0300, Alexey Dobriyan wrote:
On Wed, Aug 29, 2012 at 7:11 AM, Eric Dumazet eric.duma...@gmail.com wrote:
I'll polish this patch once LKS/LPC is over...
It should oops in the following way (excuse Gmail please):
PDEO is removed from lists
-pde_users is 0
PDE
On Wed, 2012-08-29 at 10:25 -0700, H.K. Jerry Chu wrote:
But it probably matter slightly more for TCP Fast Open (the server
side patch has
been completed and will be posted soon, after I finish breaking it up
into smaller
pieces for ease of review purpose), when a full socket will be created
On Wed, 2012-08-29 at 22:39 -0400, Alan Ott wrote:
Since lowpan_process_data() modifies the skb (by calling skb_pull()), we
need our own copy so that it doesn't affect the data received by other
protcols (in this case, af_ieee802154).
Signed-off-by: Alan Ott a...@signal11.us
---
updated to
describe the current settings. The same goes for the documentation file
Documentation/networking/ip-sysctl.txt.
Signed-off-by: Alexander Bergmann a...@linlab.net
---
Thanks for your patience and followup, this seems good to me !
Acked-by: Eric Dumazet eduma...@google.com
On Wed, 2012-08-08 at 08:17 +0200, Willy Tarreau wrote:
Well, if the drivers provided with the kernel don't work out of the box
anymore, maybe we should also move them to a separate repository. All it
is going to do otherwise is to cause invalid bug reports because users
don't understand
On Wed, 2012-08-08 at 08:49 +0200, Willy Tarreau wrote:
Hi Eric,
On Wed, Aug 08, 2012 at 08:27:52AM +0200, Eric Dumazet wrote:
On Wed, 2012-08-08 at 08:17 +0200, Willy Tarreau wrote:
Well, if the drivers provided with the kernel don't work out of the box
anymore, maybe we should
this down and it seems to be the following commit:
commit be9f4a44e7d41cee50ddb5f038fc2391cbbb4046
Author: Eric Dumazet eduma...@google.com
Date: Thu Jul 19 07:34:03 2012 +
ipv4: tcp: remove per net tcp_sock
It doesn't revert totally cleanly, but after fixing up the rejections
On Wed, 2012-08-08 at 15:26 -0400, Paul Moore wrote:
On Wednesday, August 08, 2012 12:14:42 PM John Stultz wrote:
So I bisected this down and it seems to be the following commit:
commit be9f4a44e7d41cee50ddb5f038fc2391cbbb4046
Author: Eric Dumazet eduma...@google.com
Date: Thu Jul 19
On Wed, 2012-08-08 at 15:50 -0400, Paul Moore wrote:
Yep. I was just trying to see if there was a way we could avoid having to
make it conditional on CONFIG_SECURITY, but I think this is a better approach
than the alternatives.
I'm also looking into making sure we get a sane LSM label on
On Wed, 2012-08-08 at 12:49 -0700, John Stultz wrote:
I can't comment on the patch itself, but I tested it against Linus' HEAD
and it seems to resolve the oops on shutdown for me.
OK, thanks !
--
To unsubscribe from this list: send the line unsubscribe linux-kernel in
the body of a message
On Wed, 2012-08-08 at 15:59 -0400, Eric Paris wrote:
Seems wrong. We shouldn't ever need ifdef CONFIG_SECURITY in core
code.
Sure but it seems include file misses an accessor for this.
We could add it on a future cleanup patch, as Paul mentioned.
Ifndef CONF_SECURITY then
On Wed, 2012-08-08 at 22:09 +0200, Eric Dumazet wrote:
On Wed, 2012-08-08 at 15:59 -0400, Eric Paris wrote:
Seems wrong. We shouldn't ever need ifdef CONFIG_SECURITY in core
code.
Sure but it seems include file misses an accessor for this.
We could add it on a future cleanup patch
On Wed, 2012-08-08 at 16:46 -0400, Paul Moore wrote:
On Wednesday, August 08, 2012 10:32:52 PM Eric Dumazet wrote:
On Wed, 2012-08-08 at 22:09 +0200, Eric Dumazet wrote:
On Wed, 2012-08-08 at 15:59 -0400, Eric Paris wrote:
Seems wrong. We shouldn't ever need ifdef CONFIG_SECURITY
From: Eric Dumazet eduma...@google.com
Various /proc/net files sometimes report crazy timer values, expressed
in clock_t units.
This happens when an expired timer delta (expires - jiffies) is passed
to jiffies_to_clock_t().
This function has an overflow in :
return div_u64((u64)x * TICK_NSEC
On Thu, 2012-08-09 at 09:30 -0400, Paul Moore wrote:
In the case of a TCP syn-recv and timewait ACK things are a little less
clear.
Eric (Dumazet), it looks like we have a socket in tcp_v4_reqsk_send_ack() and
tcp_v4_timewait_ack(), any reason why we can't propagate the socket down
From: Eric Dumazet eduma...@google.com
commit be9f4a44e7d41cee (ipv4: tcp: remove per net tcp_sock) added a
selinux regression, reported and bisected by John Stultz
selinux_ip_postroute_compat() expect to find a valid sk-sk_security
pointer, but this field is NULL for unicast_sock
Fix
On Thu, 2012-08-09 at 11:07 -0400, Paul Moore wrote:
Is is possible to do the call to security_sk_alloc() in the ip_init()
function
or does the per-cpu nature of the socket make this a pain?
Its a pain, if we want NUMA affinity.
Here, each cpu should get memory from its closest node.
On Thu, 2012-08-09 at 12:05 -0400, Eric Paris wrote:
On Thu, Aug 9, 2012 at 11:36 AM, Eric Dumazet eric.duma...@gmail.com wrote:
On Thu, 2012-08-09 at 11:07 -0400, Paul Moore wrote:
Is is possible to do the call to security_sk_alloc() in the ip_init()
function
or does the per-cpu
On Thu, 2012-08-09 at 16:06 -0400, Eric Paris wrote:
NAK.
I personally think commit be9f4a44e7d41cee should be reverted until it
is fixed. Let me explain what all I believe it broke and how.
Suggesting to revert this commit while we have known working fixes is a
bit of strange reaction.
On Thu, 2012-08-09 at 14:53 -0700, Casey Schaufler wrote:
On 8/9/2012 2:29 PM, Eric Dumazet wrote:
smack_sk_alloc_security() uses smk_of_current() : What can be the
meaning of smk_of_current() in the context of 'kernel' sockets...
Yes, and all of it's callers - to date - have had
From: Eric Dumazet eduma...@google.com
commit be9f4a44e7d41cee (ipv4: tcp: remove per net tcp_sock) added a
selinux regression, reported and bisected by John Stultz
selinux_ip_postroute_compat() expect to find a valid sk-sk_security
pointer, but this field is NULL for unicast_sock
It turns out
On Tue, 2012-08-21 at 16:37 -0400, Mimi Zohar wrote:
We're here, because fput() called schedule_work() to delay the last
fput(). The execution needs to take place before the syscall returns to
userspace. Need to read __schedule()... Do you know if cond_resched()
can guarantee that it will
On Tue, 2012-08-21 at 23:07 -0500, Larry Finger wrote:
Hi,
The commit entitled tcp: reduce out_of_order memory use turns out to cause
problems with a number of USB drivers.
The first one called to my attention was for staging/r8712u. For this driver,
there are problems with SSL
On Wed, 2012-08-22 at 01:29 +0200, Alex Bergmann wrote:
Hi David,
I'm not 100% sure, but it looks like I found an RFC mismatch with the
current default values of the TCP implementation.
Alex
From 8b854a525eb45f64ad29dfab16f9d9f681e84495 Mon Sep 17 00:00:00 2001
From: Alexander
On Wed, 2012-08-22 at 10:48 +0200, Alex Bergmann wrote:
On 08/22/2012 10:06 AM, Eric Dumazet wrote:
Prior to 9ad7c049 the timeout was defined with 189secs. Now we have only
a timeout of 63secs.
((2 5) - 1) * 3 secs = 189 secs
((2 5) - 1) * 1 secs = 63 secs
On Wed, 2012-08-22 at 11:29 +0200, Alex Bergmann wrote:
Actual 6 SYN frames are sent. The initial one and 5 retries.
first one had a t0 + 0 delay. How can it count ???
The kernel is waiting another 32 seconds for a SYN+ACK and then gives
the ETIMEDOUT back to userspace.
Do you mean
On Wed, 2012-08-22 at 12:00 +0200, Eric Dumazet wrote:
On Wed, 2012-08-22 at 11:29 +0200, Alex Bergmann wrote:
Actual 6 SYN frames are sent. The initial one and 5 retries.
first one had a t0 + 0 delay. How can it count ???
The kernel is waiting another 32 seconds for a SYN+ACK
On Wed, 2012-08-22 at 11:38 -0500, Nathan Zimmer wrote:
This moves a kfree outside a spinlock to help scaling on larger (512 core)
systems.
I ran a simple test which just reads from /proc/cpuinfo.
Lower is better, as you can see the worst case scenario is improved.
baseline
On Wed, 2012-08-22 at 20:28 +0200, Eric Dumazet wrote:
Thats interesting, but if you really want this to fly, one RCU
conversion would be much better ;)
pde_users would be an atomic_t and you would avoid the spinlock
contention.
Here is what I had in mind, I would be interested to know
On Wed, 2012-08-22 at 16:33 -0500, Larry Finger wrote:
On 08/22/2012 12:15 AM, Eric Dumazet wrote:
This particular commit is the start of a patches batch that ended in the
generic TCP coalescing mechanism.
It is known to have problem on drivers doing skb_clone() in their rx
path
On Thu, 2012-08-23 at 15:19 +0800, Wei Yongjun wrote:
From: Wei Yongjun yongjun_...@trendmicro.com.cn
Using is_zero_ether_addr() instead of directly use
memcmp() to determine if the ethernet address is all
zeros.
spatch with a semantic match is used to found this problem.
On Thu, 2012-08-23 at 13:58 +0200, Alex Bergmann wrote:
On 08/22/2012 06:41 PM, H.K. Jerry Chu wrote:
This issue occurred to me right after I submitted the patch for RFC6298.
I did not commit any more change because RFC compliance aside, 180secs
just seem like eternity in the Internet age.
*/
#define TCP_SYNACK_RETRIES 5 /* number of times to retry passive opening a
Acked-by: Eric Dumazet eduma...@google.com
A change of the comment might be good, to help future readers.
--
To unsubscribe from this list: send the line unsubscribe linux-kernel in
the body of a message to majord
On Thu, 2012-08-23 at 13:35 +0100, David Laight wrote:
I would suggest to increase TCP_SYN_RETRIES from 5 to 6.
180 secs is eternity, but 31 secs is too small.
Wasn't the intention of the long delay to allow a system
acting as a router to reboot?
I suspect that is why it (and some
On Thu, 2012-08-23 at 22:35 +0200, Martin Steigerwald wrote:
Hi!
Its a bit difficult to describe. With 3.6-rc2 and 3.6-rc3 on an Lenovo
ThinkPad T520 from Linus git, I get occasional network hangs:
On for example sending a small mail via SMTP to my Debian Squeeze
based server via a ASUS
On Thu, 2012-08-23 at 15:57 -0500, Larry Finger wrote:
On 08/22/2012 11:03 PM, Eric Dumazet wrote:
Changing the allocation size removes the problem ? thats really strange.
If you try different sizes in the 9100-30720 range, can you pinpoint the
failure threshold ?
The allocation size
Le vendredi 24 août 2012 à 09:09 -0500, Larry Finger a écrit :
With kernel 3.6-rc2, the error changes to the following:
--2012-08-24 08:26:42-- https://bugzilla.redhat.com/show_bug.cgi?id=847525
Resolving
Le vendredi 24 août 2012 à 09:48 -0500, Nathan Zimmer a écrit :
On Wed, Aug 22, 2012 at 11:42:58PM +0200, Eric Dumazet wrote:
On Wed, 2012-08-22 at 20:28 +0200, Eric Dumazet wrote:
Thats interesting, but if you really want this to fly, one RCU
conversion would be much better
On Fri, 2012-08-24 at 10:49 -0500, Larry Finger wrote:
There is nothing in kernel log when it happens. The file STRACE is attached.
So there is indeed a corruption. What was the driver you used in this
case ?
--
To unsubscribe from this list: send the line unsubscribe linux-kernel in
the
On Fri, 2012-08-24 at 10:58 -0500, Larry Finger wrote:
On 08/24/2012 10:19 AM, David Miller wrote:
This looks like full-on data corruption to me.
I agree. The question is why does it happen with r8712u, and only after the
commit in the subject. Drivers for other devices that I have are
On Fri, 2012-08-24 at 18:18 +0200, Eric Dumazet wrote:
On Fri, 2012-08-24 at 10:58 -0500, Larry Finger wrote:
On 08/24/2012 10:19 AM, David Miller wrote:
This looks like full-on data corruption to me.
I agree. The question is why does it happen with r8712u, and only after
On Fri, 2012-08-24 at 11:58 -0500, Larry Finger wrote:
On 08/24/2012 11:23 AM, Eric Dumazet wrote:
On Fri, 2012-08-24 at 18:18 +0200, Eric Dumazet wrote:
On Fri, 2012-08-24 at 10:58 -0500, Larry Finger wrote:
On 08/24/2012 10:19 AM, David Miller wrote:
This looks like full-on data
of 63secs.
The comments for SYN and SYNACK retries have also been updated to
describe the current settings.
Signed-off-by: Alexander Bergmann a...@linlab.net
---
Acked-by: Eric Dumazet eduma...@google.com
--
To unsubscribe from this list: send the line unsubscribe linux-kernel in
the body
On Mon, 2012-08-27 at 12:55 -0500, Larry Finger wrote:
I have prepared a patch to fix all the unchecked allocations.
Over the weekend I made some progress. To test the latest vendor driver, I
installed a 32-bit system. Their driver is not compatible with a 64-bit
system.
I found that
On Thu, 2012-08-02 at 14:27 +0200, leroy christophe wrote:
Hi
I'm having a big issue with UDP. Using a powerpc board (MPC860).
With our board running kernel 2.4.17, I'm able to send 16 voice
packets (UDP, 96 bytes per packet) in 11 seconds.
With the same board running either Kernel
On Thu, 2012-08-02 at 10:32 -0700, Linus Torvalds wrote:
On Thu, Aug 2, 2012 at 9:40 AM, Eric W. Biederman ebied...@xmission.com
wrote:
For a trivial hash table I don't know if the abstraction is worth it.
For a hash table that starts off small and grows as big as you need it
the incent
On Fri, 2012-07-27 at 23:37 +0800, Cong Wang wrote:
slave_enable_netpoll() and __netpoll_setup() may be called
with read_lock() held, so should use GFP_ATOMIC to allocate
memory.
Cc: David S. Miller da...@davemloft.net
Reported-by: Dan Carpenter dan.carpen...@oracle.com
Signed-off-by: Cong
On Fri, 2012-08-03 at 17:34 +0800, Cong Wang wrote:
On Fri, 2012-08-03 at 11:17 +0200, Eric Dumazet wrote:
On Fri, 2012-07-27 at 23:37 +0800, Cong Wang wrote:
slave_enable_netpoll() and __netpoll_setup() may be called
with read_lock() held, so should use GFP_ATOMIC to allocate
memory
On Fri, 2012-08-03 at 16:23 +0200, Sasha Levin wrote:
This hashtable implementation is using hlist buckets to provide a simple
hashtable to prevent it from getting reimplemented all over the kernel.
+static void hash_add(struct hash_table *ht, struct hlist_node *node, long
key)
+{
+
On Fri, 2012-08-03 at 16:23 +0200, Sasha Levin wrote:
Switch 9p error table to use the new hashtable implementation. This reduces
the amount of
generic unrelated code in 9p.
Signed-off-by: Sasha Levin levinsasha...@gmail.com
---
net/9p/error.c | 17 -
1 files changed,
On Sat, 2012-08-04 at 16:51 +0200, richard -rw- weinberger wrote:
On Sat, Aug 4, 2012 at 4:45 PM, Sławek Janecki jane...@gmail.com wrote:
I have a node.js client (10.177.62.7) requesting some data from http
rest service from server (10.177.0.1).
Client is simply using nodejs http.request()
On Sun, 2012-08-05 at 10:16 +0200, LEROY christophe wrote:
Le 02/08/2012 16:13, Eric Dumazet a écrit :
On Thu, 2012-08-02 at 14:27 +0200, leroy christophe wrote:
Hi
I'm having a big issue with UDP. Using a powerpc board (MPC860).
With our board running kernel 2.4.17, I'm able to send
On Mon, 2012-08-06 at 17:08 +0800, Cong Wang wrote:
On Fri, 2012-08-03 at 12:10 +0200, Eric Dumazet wrote:
I did this , just take it ;)
Do we have to pass gfp to -ndo_netpoll_setup() too? It seems no, so far
I don't think we have to do that.
Thanks.
It is needed
On Sat, 2012-08-18 at 10:19 +0800, Fengguang Wu wrote:
Hi David,
The bug should be introduced somewhere between 3.5 and 3.6-rc1.
[ 2866.131281] IPv4: Attempt to release TCP socket in state 1 880019ec
[ 2866.131726]
[ 2866.132188] =
[ 2866.132281] [ BUG:
On Sun, 2012-08-19 at 22:15 +0800, Lin Ming wrote:
Will it still has problem if code goes here without sock_hold(sk)?
Not sure of what you mean.
At the time tcp_write_timer() runs, we own one reference on the socket.
(this reference was taken in sk_reset_timer())
On old kernels, if we found
On Sun, 2012-08-19 at 23:05 +0800, Lin Ming wrote:
On Sun, Aug 19, 2012 at 10:45 PM, Eric Dumazet eric.duma...@gmail.com wrote:
On Sun, 2012-08-19 at 22:15 +0800, Lin Ming wrote:
Will it still has problem if code goes here without sock_hold(sk)?
Not sure of what you mean.
See my
501 - 600 of 5670 matches
Mail list logo