Cleanup unneeded macros used for register space address calculation.
Now we are using the EBDA to find the space address.
Signed-off-by: Guillaume Thouvenin <[EMAIL PROTECTED]>
---
pci-calgary.c |5 -
1 file changed, 5 deletions(-)
Index: linux-2.6.20.4/arch/x86_64/kernel/pci-cal
Hello Andrew,
This patch fixes missing dependencies in drivers/connector/Kconfig
file. We have to ensure that the dependencies of the selected variable
are fulfilled otherwise it can produce some undefined references
during the kernel compilation. This problem was reported by Adrian Bunk.
Signe
Hello,
This patch removes the ugly union declaration in cn_fork.h and
cn_exit.h files. The code is cleaner without the union and the price is
only four bytes added in the structure.
Thanks to Alexander Nyberg for reporting this.
Signed-off-by: Guillaume Thouvenin <[EMAIL PROTEC
Hello,
I don't see the connector directory in the 2.6.12-rc2-mm1 tree. So it
seems that you removed the connector? Will you include it again in futur
release? At the same time, will you include the fork connector?
Thanks for your help,
Best regards,
Guillaume
-
To unsubscribe from this list:
On Tue, 2005-04-05 at 11:34 +0400, Evgeniy Polyakov wrote:
> On Tue, 2005-04-05 at 01:10 -0400, Herbert Xu wrote:
>
> >In fact to this day I still don't understand what problems this thing is
> >meant to solve.
>
> Hmm, what else can I add to my words?
> May be checking the size of the code needed
ork.c
This patch applies to 2.6.12-rc1-mm4
Many thanks to Andrew for the great help,
Best regards,
Guillaume
Signed-off-by: Guillaume Thouvenin <[EMAIL PROTECTED]>
---
drivers/connector/Kconfig | 11
drivers/connector/Makefile |1
drivers/connector/cn_fo
g of the message) is
around 2%. If we use the CBUS patch http://lkml.org/lkml/2005/3/20/14
instead of cn_netlink_send() routine we can reduce this overhead near to
0%. We're waiting for the inclusion of the CBUS in the -mm tree.
This patch applies to 2.6.12-rc1-mm4.
Signed-off-by: Guilla
On Wed, 2005-03-30 at 20:25 +1000, Herbert Xu wrote:
> Paul Jackson <[EMAIL PROTECTED]> wrote:
> >
> > So I suppose if fork_connector were not used to collect > child pid> information for accounting, then someone would have to make
> > the case that there were enough other uses, of sufficient val
On Tue, 2005-03-29 at 22:06 -0800, dean gaudet wrote:
> On Tue, 29 Mar 2005, Jay Lan wrote:
>
> > The fork_connector is not designed to solve accounting data collection
> > problem.
> >
> > The accounting data collection must be done via a hook from do_exit().
>
> by the time do_exit() occurs th
On Tue, 2005-03-29 at 07:35 -0800, Paul Jackson wrote:
> Guillaume wrote:
> > I ran some test using the CBUS instead of the cn_netlink_send() routine
> > and the overhead is nearly 0%:
>
> Overhead of what? Does this include merging the data and getting it to
> disk?
I test the overhead of sendi
On Tue, 2005-03-29 at 07:23 -0800, Paul Jackson wrote:
> Guillaume wrote:
> > The goal of the fork connector is to inform a user space application
> > that a fork occurs in the kernel. This information (cpu ID, parent PID
> > and child PID) can be used by several user space applications. It's not
On Mon, 2005-03-28 at 13:42 -0800, Paul Jackson wrote:
> Guillaume wrote:
> > The lmbench shows that the overhead (the construction and the sending
> > of the message) in the fork() routine is around 7%.
>
> Thanks for including the numbers. The 7% seems a bit costly, for a bit
> more accountin
On Tue, 2005-03-29 at 00:49 -0800, Paul Jackson wrote:
> This
> amortizes the cost of almost all the handling, and of all the disk i/o,
> over many data collection events. Correct me if I'm wrong, but
> fork_connector doesn't do this merging of events into a consolidated
> data buffer, so is at
On Mon, 2005-03-28 at 13:42 -0800, Paul Jackson wrote:
> Guillaume wrote:
> > The lmbench shows that the overhead (the construction and the sending
> > of the message) in the fork() routine is around 7%.
>
> Thanks for including the numbers. The 7% seems a bit costly, for a bit
> more accountin
d
that fix problems in the connector.c file. At least, you need to apply
the patch provided in the second email.
Signed-off-by: Guillaume Thouvenin <[EMAIL PROTECTED]>
---
drivers/connector/Kconfig | 11
drivers/connector/Makefile |1
drivers/connector/cn_f
This patch fixes a bug in the __cn_rx_skb() routine when checking the
size of the netlink message.
It applies to 2.6.12-rc1-mm3.
Signed-off-by: Evgeniy Polyakov <[EMAIL PROTECTED]>
---
connector.c |9 -
1 files changed, 4 insertions(+), 5 deletions(-)
Index: linux-2.6.12-rc1-
Hello,
The following patch set adds a fork connector in the do_fork()
routine to 2.6.12-rc1-mm3 kernel.
We provide two patches. The first one is the fork connector
implementation and the second one fix a bug in the connector.c. The
second patch has been sent by Evgeniy Polyakov for inclusion
On Tue, 2005-03-22 at 10:15 -0800, Jay Lan wrote:
> Guillaume Thouvenin wrote:
> > On Mon, 2005-03-21 at 12:52 -0800, Ram wrote:
> >
> >> If a bunch of applications are listening for fork events,
> >> your patch allows any application to turn off the
On Mon, 2005-03-21 at 12:52 -0800, Ram wrote:
> If a bunch of applications are listening for fork events,
> your patch allows any application to turn off the
> fork event notification? Is this the right behavior?
Yes it is. The main management is done by application so, if severa
rk.c2003-01-30
11:24:37.0 +0100
+++ linux-2.6.11-mm4-cnfork/drivers/connector/cn_fork.c 2005-03-16
14:34:41.0 +0100
@@ -0,0 +1,85 @@
+/*
+ * cn_fork.c
+ *
+ * 2005 Copyright (c) Guillaume Thouvenin <[EMAIL PROTECTED]>
+ * All rights reserved.
+ *
+ * This pr
On Thu, 2005-03-17 at 14:05 -0800, Jesse Barnes wrote:
> On Thursday, March 17, 2005 1:38 pm, Evgeniy Polyakov wrote:
> > The most significant part there - is requirement to store
> > u32 seq in each CPU's cache and thus flush cacheline +
> > invalidate/get from mem on each other cpus
> > each time
hat
fix problems in the connector.c file. At least, you need to apply the
patch provided in the second email.
Signed-off-by: Guillaume Thouvenin <[EMAIL PROTECTED]>
---
drivers/connector/Kconfig | 11 +
drivers/connector/Makefile |1
drivers/conn
Hello,
The following patch set adds a fork connector in the do_fork()
routine.
We provide two patches. The first one is the fork connector
implementation and the second one fix a bug in the connector.c. The
second patch has been sent by Evgeniy Polyakov for inclusion and should
be in the nex
This patch fixes a bug in the __cn_rx_skb() routine when checking the
size of the netlink message.
It applies to 2.6.11-mm4.
Signed-off-by: Evgeniy Polyakov <[EMAIL PROTECTED]>
---
connector.c |9 -
1 files changed, 4 insertions(+), 5 deletions(-)
Index: linux-2.6.11-mm4-cnfo
Hello,
In the ChangeLog-2.6.11 file I read that the enhanced I/O accounting
data patch and the enhanced memory accounting data collection patch were
added. It's cool but I don't see how this stuff is used because
information is never dump in a file or send to an accounting application
(or I miss
On Tue, 2005-03-08 at 08:29 +0100, Guillaume Thouvenin wrote:
> TODO:
> - Run the lmbench with the user space application that manages
> group of processus. if fork connector is not used the only
> overhead is a test in the do_fork() routine.
For information h
TODO:
- Run the lmbench with the user space application that manages
group of processus. if fork connector is not used the only
overhead is a test in the do_fork() routine.
Thank you everyone for your comments,
Best regards,
Guillaume
Signed-off by: Guillaume Thouvenin <[EM
I don't see the connector module [cn.ko] in this new release. Do you
remove it from your tree definitely?
Best regards,
Guillaume
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/
On Wed, 2005-03-02 at 01:06 -0800, Andrew Morton wrote:
> Guillaume Thouvenin <[EMAIL PROTECTED]> wrote:
> >
> > So I ran the lmbench with three different kernels with the fork
> > connector patch I just sent. Results are attached at the end of the mail
> > an
On Tue, 2005-03-01 at 22:38 +0900, Kaigai Kohei wrote:
> > I tested without user space listeners and the cost is negligible. I will
> > test with a user space listeners and see the results. I'm going to run
> > the test this week after improving the mechanism that switch on/off the
> > sending of t
This patch works with another
small patch that fix a problem in the connector. Without it, there is a
message that says "skb does not have enough length". It will be fix in
the next -mm tree I think.
Thanks everyone for the comments,
Guillaume
Signed-off-by: Guillaume Thouven
On Tue, 2005-03-01 at 10:06 -0800, Jay Lan wrote:
> Sorry I was not clear on my point.
>
> I was trying to point out that, an exit hook for BSD and CSA is
> essential to save accounting data before the data is gone. That
> can not be done with a netlink.
>
> So, my patch was to keep acct_process
On Tue, 2005-03-01 at 22:38 +0900, Kaigai Kohei wrote:
> > I tested without user space listeners and the cost is negligible. I will
> > test with a user space listeners and see the results. I'm going to run
> > the test this week after improving the mechanism that switch on/off the
> > sending of t
ere.
> Atomic allocation is fast, if it succeds, but there are no groups/socket to
> send,
> skb will be freed, if allocation fails, then group check is useless.
>
> I would prefer Guillaume Thouvenin as fork connector author to test
> his current implementation and show that conne
On Mon, 2005-02-28 at 10:56 -0800, Jay Lan wrote:
> The exit hook is essential for CSA to save off data before the data
> is gone, A netlink type of thing does not help. BSD is in the same
> situation. You can not replace the acct_process() call with a netlink.
> If ELSA is to use the enhanced acco
On Mon, 2005-02-28 at 10:59 +0900, Kaigai Kohei wrote:
> Marcelo Tosatti wrote:
> > Yep, the netlink people should be able to help - they known what would be
> > required for not sending messages in case there is no listener registered.
> >
> > Maybe its already possible? I have never used netl
On Thu, 2005-02-24 at 20:46 -0800, Andrew Morton wrote:
> Jay Lan <[EMAIL PROTECTED]> wrote:
> >
> > Since my idea of providing an accounting framework was considered
> > 'overkill', here i submit a tiny patch just to allow CSA to
> > handle end-of-process (eop) situation by saving off accounting
On Thu, 2005-02-24 at 13:45 +0300, Evgeniy Polyakov wrote:
> >
> > Todo:
> >
> > - Test the performance impact with lmbench
> > - Improve the callback that turns on/off the fork connector
> > - Create a specific module to register the callback.
>
> Besides connector.c changes I do
sage.
Todo:
- Test the performance impact with lmbench
- Improve the callback that turns on/off the fork connector
- Create a specific module to register the callback.
Thanks to Andrew, Evgeniy and everyone for the comments,
Guillaume
Signed-off-by: Guillaume Thouvenin <[EMAI
On Wed, 2005-02-23 at 11:11 -0800, Jay Lan wrote:
> Guillaume Thouvenin wrote:
> > It's what I'm proposing. The problem is to be alerted when a new process
> > is created in order to add it in the correct group of processes if the
> > parent belongs to one (or severa
On Wed, 2005-02-23 at 14:41 +0300, Evgeniy Polyakov wrote:
> > Please assume that > originally written for> will always be listening.
> >
> > > > What happened to the idea of sending an on/off message down the netlink
> > > > socket?
> > ...
> > Arrange for the userspace daemon to send a message t
On Wed, 2005-02-23 at 00:51 -0800, Andrew Morton wrote:
> > It's what I'm proposing. The problem is to be alerted when a new process
> > is created in order to add it in the correct group of processes if the
> > parent belongs to one (or several) groups. The notification can be done
> > with the fo
cro #if defined(CONFIG_CONNECTOR)
- cn_netlink_send() sends messages to the correct group which is
CN_IDX_FORK
Thanks to Evgeniy and to everyone for the comments,
Guillaume
Signed-off-by: Guillaume Thouvenin <[EMAIL PROTECTED]>
---
drivers/connector/Kconfig |8
include/lin
On Tue, 2005-02-22 at 23:20 -0800, Andrew Morton wrote:
> Kaigai Kohei <[EMAIL PROTECTED]> wrote:
> >
> > The common agreement for the method of dealing with process aggregation
> > has not been constructed yet, I understood. And, we will not able to
> > integrate each process aggregation model
On Tue, 2005-02-22 at 12:11 -0800, Jay Lan wrote:
> How ELSA adds per process accounting data
> to your grouping (banks) when a process exit? How do you save
> accounting data you need in task_struct before it is disposed? BSD
> handles that through acct_process() hook at do_exit(). CSA also
> dep
On Mon, 2005-02-21 at 03:58 -0800, Paul Jackson wrote:
> > It's a new patch that implements a fork connector in the
> > kernel/fork.c:do_fork() routine. The connector sends information about
> > parent PID and child PID over a netlink interface. It allows to several
> > user space applications
On Mon, 2005-02-21 at 01:47 -0800, Paul Jackson wrote:
> Guillaume wrote:
> > The problem is the following: I have a user space daemon that manages
> > group of processes. The main idea is, if a parent belongs to a group
> > then its child belongs to the same group. To achieve this I need to know
On Thu, 2005-02-17 at 18:50 +0300, Evgeniy Polyakov wrote:
> >
> > +#if defined(CONFIG_CONNECTOR) && defined(CONFIG_FORK_CONNECTOR)
>
> I suspect CONFIG_FORK_CONNECTOR is enough.
The problem here is that if connector is compiled as a module and
fork_connector as builtin there will be undefined
On Thu, 2005-02-17 at 18:50 +0300, Evgeniy Polyakov wrote:
> On Thu, 2005-02-17 at 15:55 +0100, Guillaume Thouvenin wrote:
> > It's a new patch that implements a fork connector in the
> > kernel/fork.c:do_fork() routine. The connector sends information about
> > paren
On Fri, 2005-02-18 at 17:16 -0800, Andrew Morton wrote:
> Jay Lan <[EMAIL PROTECTED]> wrote:
> >
> > Since the need of Linux system accounting has gone beyond what BSD
> > accounting provides, i think it is a good idea to create a thin layer
> > of common code for various accounting packages, such
me
Signed-off-by: Guillaume Thouvenin <[EMAIL PROTECTED]>
---
drivers/connector/Kconfig | 10 ++
include/linux/connector.h |2 ++
kernel/fork.c | 41 +
3 files changed, 53 insertions(+)
diff -uprN -X dontdiff linux-2.6.11
On Mon, 2005-02-14 at 13:01 +0100, Guillaume Thouvenin wrote:
> This patch removes unreachable code in cn_netlink_send() function.
The code can be reach via
nlh = NLMSG_PUT(skb, 0, msg->seq, NLMSG_DONE, size - sizeof(*nlh));
So the patch is wrong
Sorry for that
Guillaume
-
To unsub
This patch removes unreachable code in cn_netlink_send() function.
Signed-off-by: Guillaume Thouvenin <[EMAIL PROTECTED]>
--- drivers/connector/connector.c.orig 2005-02-14 12:52:32.0 +0100
+++ drivers/connector/connector.c 2005-02-14 12:52:44.0 +0100
@@ -119,11
On Fri, 2005-02-11 at 11:11 -0800, Greg KH wrote:
> > + char *kobj_path = NULL;
> > + char *action_string = NULL;
> > + char **envp = NULL;
> > + char ppid_string[FORK_BUFFER_SIZE];
> > + char cpid_string[FORK_BUFFER_SIZE];
> > +
> > + if (!uevent_sock)
> > + return;
> > +
g | 8m08.374s 7m34.011s 0m51.282s |
| 8m09.317s 7m34.563s 0m51.259s |
| 8m08.121s 7m35.412s 0m51.069s |
---/
Regards,
Guillaume
Signed-off-by: Guillaume Thouvenin <[EMAIL PROTECTED]>
diff -uprN -X dontdiff l
On Fri, 2005-02-11 at 00:54 -0800, Andrew Morton wrote:
> > I tested this patch on a 2.6.11-rc3-mm2 kernel and there is a little
> > overhead when I compile a Linux kernel:
> >
> >#time sh -c 'make O=/home/guill/build/k2610 bzImage &&
> >make O=/home/guill/build/k2610 modules'
> >
> >
On Mon, 2005-02-07 at 15:46 -0800, Andrew Morton wrote:
> Guillaume Thouvenin <[EMAIL PROTECTED]> wrote:
> >
> > Hello,
> >
> >This module sends a signal to one or several processes (in user
> > space) when a fork occurs in the kernel. It relays
Selon Jay Lan <[EMAIL PROTECTED]>:
> CSA is currently implemented as a loadable module. I think ELSA is the
> same, right? The use of the enhanced accounting data collection
> code is not in the kernel tree. That was why Andrew did not see usage of
> the accounting patches. Should i propose to inc
On Mon, 2005-02-07 at 15:46 -0800, Andrew Morton wrote:
> Guillaume Thouvenin <[EMAIL PROTECTED]> wrote:
> >
> >This module sends a signal to one or several processes (in user
> > space) when a fork occurs in the kernel. It relays information about
> > forks
kernel/Makefile |1
kernel/fork.c | 27 ++
kernel/relay_fork.c | 477 ++++++
5 files changed, 539 insertions(+)
Signed-off-by: Guillaume Thouvenin <[EMAIL PROTECTED]>
diff -uprN -X dontdiff linux-2.6.11-rc3-mm1/include/linux/ge
On Thu, 3 Feb 2005, Andrew Morton wrote:
> Which implies that we need to see some additional accounting code, so we
> can verify that the base accumulation infrastructure is doing the expected
> thing. As well as an ack from the interested parties. Does anyone know
> what's happening with all the
On Thu, 2005-01-20 at 10:28 +0100, Guillaume Thouvenin wrote:
>The only thing I need to manage "jobs" is a hook in the
> do_fork() routine, everything else is done outside the kernel.
I have a question about this kind of hooks.
It seems that if someone needs a hook somewhe
Andrew wrote:
> Limin Gu wrote:
>>
>> Could you consider this for inclusion in the mm tree?
>> The Job patch has been posted a few times, and I've addressed
>> the issues others raised.
>
> A similar thing happened last year with the "enhanced system
> accounting" patch.
I have a different
it.
echo XX > /sys/relayfork/signal
Any comments are welcome,
Guillaume.
Reminder: This module used a hook in the kernel tree. Thus you need to
apply the corresponding patch in order to use it.
relay_fork.c -----
/*
* Relay fork interface
*
* Author: Guil
to read it.
Guillaume
Signed-Off-By: Guillaume Thouvenin <[EMAIL PROTECTED]>
--- linux-2.6.10/kernel/fork.c.orig 2005-01-04 11:28:58.0 +0100
+++ linux-2.6.10/kernel/fork.c 2005-01-19 14:23:25.0 +0100
@@ -61,6 +61,46 @@ rwlock_t tasklist_lock __cacheline_align
65 matches
Mail list logo