On Mon, Mar 12, 2007 at 01:56:13PM -0700, David Miller wrote:
> From: Pekka J Enberg <[EMAIL PROTECTED]>
> Date: Mon, 12 Mar 2007 14:15:16 +0200 (EET)
>
> > On 3/9/07, David Miller <[EMAIL PROTECTED]> wrote:
> > > The whole cahce-multipath subsystem has to have it's guts revamped for
> > > proper
On Sun, Nov 05, 2006 at 12:53:23AM +0100, Christoph Hellwig wrote:
> On Sat, Nov 04, 2006 at 06:06:48PM -0500, Dave Jones wrote:
> > On Sat, Nov 04, 2006 at 11:56:29PM +0100, Christoph Hellwig wrote:
> >
> > This will break the compile for !NUMA if someone ends up doing a bisect
> > and lands here
On Sat, May 20, 2006 at 01:51:45PM -0700, Chris Wedgwood wrote:
> ...
> Anyhow, the code as-is hasn't been maintained for a long time except
> for a few minor blips (I'm using hg's annotate to find those and have
> included those people on the cc' list as presumably there are using
> these features
This patch checks device locality on every ip packet xmit.
In multipath configuration tcp connection to route association is done at
session startup time. The tcp session process is migrated to different nodes
after this association. This would mean a remote NIC is chosen for xmit,
although a lo
Following patch adds in node aware, device round robin ip multipathing.
It is based on multipath_drr.c, the multipath device round robin algorithm, and
is derived from it. This implementation maintians per node state table, and
round robins between interfaces on the same node. The implementati
On Thu, Mar 09, 2006 at 07:14:26PM +1100, Nick Piggin wrote:
> Ravikiran G Thirumalai wrote:
>
> >Here's a patch making x86_64 local_t to 64 bits like other 64 bit arches.
> >This keeps local_t unsigned long. (We can change it to signed value
> >along with other a
On Wed, Mar 08, 2006 at 04:32:58PM -0800, Andrew Morton wrote:
> Ravikiran G Thirumalai <[EMAIL PROTECTED]> wrote:
> >
> > On Wed, Mar 08, 2006 at 03:43:21PM -0800, Andrew Morton wrote:
> > > Benjamin LaHaise <[EMAIL PROTECTED]> wrote:
> > > >
On Wed, Mar 08, 2006 at 03:43:21PM -0800, Andrew Morton wrote:
> Benjamin LaHaise <[EMAIL PROTECTED]> wrote:
> >
> > I think it may make more sense to simply convert local_t into a long, given
> > that most of the users will be things like stats counters.
> >
>
> Yes, I agree that making local_t
On Wed, Mar 08, 2006 at 04:17:33PM -0500, Benjamin LaHaise wrote:
> On Wed, Mar 08, 2006 at 01:07:26PM -0800, Ravikiran G Thirumalai wrote:
>
> Last time I checked, all the major architectures had efficient local_t
> implementations. Most of the RISC CPUs are able to do a
On Wed, Mar 08, 2006 at 03:36:42PM -0500, Benjamin LaHaise wrote:
> On Wed, Mar 08, 2006 at 12:26:56PM -0800, Ravikiran G Thirumalai wrote:
> > +static inline void percpu_counter_mod_bh(struct percpu_counter *fbc, long
> > amount)
> > +{
> > + local_bh_disable();
&
On Tue, Mar 07, 2006 at 07:22:34PM -0800, Andrew Morton wrote:
> Ravikiran G Thirumalai <[EMAIL PROTECTED]> wrote:
> >
> > The problem is percpu_counter_sum has to read all the cpus cachelines. If
> > we have to use percpu_counter_sum everywhere, then might as well use
On Tue, Mar 07, 2006 at 06:16:02PM -0800, Andrew Morton wrote:
> Ravikiran G Thirumalai <[EMAIL PROTECTED]> wrote:
> >
> > +static inline int read_sockets_allocated(struct proto *prot)
> > +{
> > + int total = 0;
> > + int cpu;
> >
On Tue, Mar 07, 2006 at 06:13:01PM -0800, Andrew Morton wrote:
> Ravikiran G Thirumalai <[EMAIL PROTECTED]> wrote:
> >
> > +static inline void percpu_counter_mod_bh(struct percpu_counter *fbc, long
> > amount)
> > +{
> > + local_bh_disable();
On Tue, Mar 07, 2006 at 06:14:22PM -0800, Andrew Morton wrote:
> Ravikiran G Thirumalai <[EMAIL PROTECTED]> wrote:
> >
> > - if (atomic_read(sk->sk_prot->memory_allocated) <
> > sk->sk_prot->sysctl_mem[0]) {
> > + if (p
This patch uses alloc_percpu to allocate per-CPU memory for the proto->inuse
field. The inuse field is currently per-CPU as in NR_CPUS * cacheline size --
a big bloat on arches with large cachelines. Also marks some frequently
used protos read mostly.
Signed-off-by: Pravin B. Shelar <[EMAIL PRO
Change the atomic_t sockets_allocated member of struct proto to a
per-cpu counter.
Signed-off-by: Pravin B. Shelar <[EMAIL PROTECTED]>
Signed-off-by: Ravikiran Thirumalai <[EMAIL PROTECTED]>
Signed-off-by: Shai Fultheim <[EMAIL PROTECTED]>
Index: linux-2.6.16-rc5mm3/include/net/sock.h
==
Add percpu_counter_mod_bh for using these counters safely from
both softirq and process context.
Signed-off by: Pravin B. Shelar <[EMAIL PROTECTED]>
Signed-off by: Ravikiran G Thirumalai <[EMAIL PROTECTED]>
Signed-off by: Shai Fultheim <[EMAIL PROTECTED]>
Index: linux-2.6.16-rc
Change struct proto->memory_allocated to a batching per-CPU counter
(percpu_counter) from an atomic_t. A batching counter is better than a
plain per-CPU counter as this field is read often.
Signed-off-by: Pravin B. Shelar <[EMAIL PROTECTED]>
Signed-off-by: Ravikiran Thirumalai <[EMAIL PROTECTED
Following patchset converts struct proto.memory_allocated to use batching
per-cpu counters, struct proto.sockets_allocated to use per-cpu counters and
changes the proto.inuse per-cpu variable to use alloc_percpu instead of the
NR_CPUS x cacheline size padding.
We observed 5% improvement in apache
On Thu, Feb 02, 2006 at 07:16:00PM -0800, Andrew Morton wrote:
> Ravikiran G Thirumalai <[EMAIL PROTECTED]> wrote:
> >
> > On Fri, Jan 27, 2006 at 03:01:06PM -0800, Andrew Morton wrote:
> > Here's an implementation which delegates tuning of batching to the user.
On Fri, Jan 27, 2006 at 03:01:06PM -0800, Andrew Morton wrote:
> Ravikiran G Thirumalai <[EMAIL PROTECTED]> wrote:
>
>
> > >
> > > If the benchmarks say that we need to. If we cannot observe any problems
> > > in testing of existing code and if we ca
On Sat, Jan 28, 2006 at 01:35:03AM +0100, Eric Dumazet wrote:
> Eric Dumazet a écrit :
> >Andrew Morton a écrit :
> >>Eric Dumazet <[EMAIL PROTECTED]> wrote:
> >
> >#ifdef CONFIG_SMP
> >void percpu_counter_mod(struct percpu_counter *fbc, long amount)
> >{
> > long old, new;
> > atomic_long_
On Sat, Jan 28, 2006 at 12:21:07AM +0100, Eric Dumazet wrote:
> Ravikiran G Thirumalai a écrit :
> >On Fri, Jan 27, 2006 at 11:30:23PM +0100, Eric Dumazet wrote:
>
> Why not use a boot time allocated percpu area (as done today in
> setup_per_cpu_areas()), but instead of reservi
On Fri, Jan 27, 2006 at 03:08:47PM -0800, Andrew Morton wrote:
> Andrew Morton <[EMAIL PROTECTED]> wrote:
> >
> > Oh, and because vm_acct_memory() is counting a singleton object, it can use
> > DEFINE_PER_CPU rather than alloc_percpu(), so it saves on a bit of kmalloc
> > overhead.
>
> Actually, I
On Fri, Jan 27, 2006 at 11:30:23PM +0100, Eric Dumazet wrote:
>
> There are several issues here :
>
> alloc_percpu() current implementation is a a waste of ram. (because it uses
> slab allocations that have a minimum size of 32 bytes)
Oh there was a solution for that :).
http://lwn.net/Artic
On Fri, Jan 27, 2006 at 12:16:02PM -0800, Andrew Morton wrote:
> Ravikiran G Thirumalai <[EMAIL PROTECTED]> wrote:
> >
> > which can be assumed as not frequent.
> > At sk_stream_mem_schedule(), read_sockets_allocated() is invoked only
> > certain conditions, un
On Fri, Jan 27, 2006 at 09:53:53AM +0100, Eric Dumazet wrote:
> Ravikiran G Thirumalai a écrit :
> >Change the atomic_t sockets_allocated member of struct proto to a
> >per-cpu counter.
> >
> >Signed-off-by: Pravin B. Shelar <[EMAIL PROTECTED]>
> >Sign
This patch uses alloc_percpu to allocate per-CPU memory for the proto->inuse
field. The inuse field is currently per-CPU as in NR_CPUS * cacheline size --
a big bloat on arches with large cachelines. Also marks some frequently
used protos read mostly.
Signed-off-by: Pravin B. Shelar <[EMAIL PRO
Add percpu_counter_mod_bh for using these counters safely from
both softirq and process context.
Signed-off by: Pravin B. Shelar <[EMAIL PROTECTED]>
Signed-off by: Ravikiran G Thirumalai <[EMAIL PROTECTED]>
Signed-off by: Shai Fultheim <[EMAIL PROTECTED]>
Index: linux-2.6.16
Change struct proto->memory_allocated to a batching per-CPU counter
(percpu_counter) from an atomic_t. A batching counter is better than a
plain per-CPU counter as this field is read often.
Signed-off-by: Pravin B. Shelar <[EMAIL PROTECTED]>
Signed-off-by: Ravikiran Thirumalai <[EMAIL PROTECTED
Change the atomic_t sockets_allocated member of struct proto to a
per-cpu counter.
Signed-off-by: Pravin B. Shelar <[EMAIL PROTECTED]>
Signed-off-by: Ravikiran Thirumalai <[EMAIL PROTECTED]>
Signed-off-by: Shai Fultheim <[EMAIL PROTECTED]>
Index: linux-2.6.16-rc1/include/net/sock.h
=
The following patches change struct proto.memory_allocated,
proto.sockets_allocated to use per-cpu counters. This patchset also switches
the proto.inuse percpu varible to use alloc_percpu, instead of NR_CPUS *
cacheline size padding.
We saw 5 % improvement in apache bench requests per second with
32 matches
Mail list logo
