Re: how to know when much page reclamation

Eugene Teo <eugene.teo <at> eugeneteo.net> writes:

> <quote sender="Ed L Cashin">
>> Hi.  I can choose one of two different related things in userland, and
>> one takes more physical memory than the other but runs faster.
>
> you mean one memory intensive, one io intensive?

Nope.  :)  

Strategy "A" uses more page tables (pinned in RAM) and runs fast,
while strategy "B" uses fewer page tables and runs somewhat slower.

>> Looking at the number of free pages in the system doesn't really tell
>> me whether memory is "tight" or not.  There could be plenty of pages
>> being used for caches, and at the first sign of memory pressure, those
>> caches may shrink.
>> 
>> The best way I can think of to know whether memory is really tight is
>> to measure how often shrink_zone is getting called.  (I'm looking at a
>> 2.6.0-test11 kernel.)
>> 
>> I can add some light instrumentation to track the rate at which
>> shrink_zone gets called and a system call to make that rate visible to
>> userland.  Is there an existing way already in place or a better way
>> to know when memory is tight?
>
> cat /proc/vmstat, see pgscan. but the stat is commulative. 

At a glance, that looks like just the thing, with stuff like "pgsteal"

Re: eth2 error : (3 cards on one linux machine)

pankaj chauhan <chauhan_ait <at> yahoo.co.in> writes:

> hi all ,
>
>
> i have a small problem , i need to install more than
> three network cards (ethernet )on a linux machine (
> makin it gateway).
>
> two cards are working fine ,but when i ifconfig for
> third card ( eth2 )it  gives an error 
>
> eth2 : unknown interface eth2.
the only thing i can think of is whether the appropriate module is
inserted into the kernel or not ? what does ifconfig tell you ?

anupam

>
>
> do i need to give some extra kernel config option to
> have more than 2 cards.
>
> i am sorry for asking a non kernel question on list ..
>
>
> rgds,
> pankaj
>
>

Re: __get_fre_pages problem

On Sat, 28 Feb 2004 17:26:44 +0000 (GMT)
pankaj chauhan <chauhan_ait <at> yahoo.co.in> wrote:

>
>hi all , 
>
>i have a machine with 256Mb of RAM.
>
>i am using __get_free_pages{order 9} to get 2Mb memory
>( want it to be continous) in a module. i am accessing
>this memory in interrupt context..
>
>my system crashes down when i access this memory  in
>interrupt context. during allocation i check that if
>__get_free_pages fails ( pointer returned is NULL) , i
>return an error.so i hope i am using memory in the
>interrupt context only if __get_free_pages succeeded.
>
>plz tell me what may be possible problem 

Please take a look at this (grepping kernel source code is
always a good thing to do in such cases).

 * The  <at> flags argument may be one of:
 *
 * %GFP_USER - Allocate memory on behalf of user.  May sleep.
 *
 * %GFP_KERNEL - Allocate normal kernel ram.  May sleep.
 *

Blocked delivery of your email to .vinodk <at> persistent.co.in

<<< No Message Collected >>>

--
Kernelnewbies: Help each other learn about the Linux kernel.
Archive:       http://mail.nl.linux.org/kernelnewbies/
FAQ:           http://kernelnewbies.org/faq/

remap_page_range & mm semaphore

Hello,
above the implementation of the function remap_page_range() is this comment:
	"Note: this is only safe if the mm semaphore is held when called."
Is this obsolete, because I can't find such mm semaphore?

/FAU

--
Kernelnewbies: Help each other learn about the Linux kernel.
Archive:       http://mail.nl.linux.org/kernelnewbies/
FAQ:           http://kernelnewbies.org/faq/

swap page issue

 Hi,

  Please help me to understand the issue with a swap page.

  this is in 2.4 kernel. An anonymous page of a process is added to the 
swap cache using add_to_swap_cache() and the page table entry is set to 
the swap entry. ( try_to_swap_out()). And at sometime later in 
shrink_cache(), that page is written to the swap disk using 
swap_writepage(). But in swap_writepage(), the page is first tested to see 
if it is an exclusive swap page ie the page is owned by only one process 
and it is removed from the swap cache without writing to the swap disk. 
Please correct if I have misunderstood over here.  Now the page is no more 
in the swap cache and if the process faults on this page and it tries to 
get this page from the swap cache which is no more in the cache now. I 
definitely am losing a link over 
here. 

 Can you please explain me what correctly happens during swapping out of a 
page?

 Thanks,
 Raghu.

--
Kernelnewbies: Help each other learn about the Linux kernel.
Archive:       http://mail.nl.linux.org/kernelnewbies/
FAQ:           http://kernelnewbies.org/faq/

http://helpline.wipro.com

Dear Friend,

Dear Friend,

Tks for your mail.  Webeditor, Dinamalar.

+++++++

VISIT www.dinamalar.com for hourly updation

++++++++

--
Kernelnewbies: Help each other learn about the Linux kernel.
Archive:       http://mail.nl.linux.org/kernelnewbies/
FAQ:           http://kernelnewbies.org/faq/

Re: remap_page_range & mm semaphore

On Mon, Mar 01, 2004 at 08:51:25PM +0100, Frank A. Uepping wrote:
> Hello,
> above the implementation of the function remap_page_range() is this comment:
> 	"Note: this is only safe if the mm semaphore is held when called."
> Is this obsolete, because I can't find such mm semaphore?
>

If you look at mm_struct you find that there's no only page table spinlock but 
also a R/W sem.
This function like others of the same family requires that this sem was already hold.

struct mm_struct {
	.......
	.......
	struct rw_semaphore mmap_sem;
	spinlock_t page_table_lock;
	.......
};

--
Kernelnewbies: Help each other learn about the Linux kernel.
Archive:       http://mail.nl.linux.org/kernelnewbies/
FAQ:           http://kernelnewbies.org/faq/

test

test

--

-- 

--
Kernelnewbies: Help each other learn about the Linux kernel.
Archive:       http://mail.nl.linux.org/kernelnewbies/
FAQ:           http://kernelnewbies.org/faq/