set pthread stack size to at least PTHREAD_STACK_MIN
Attempting to set the stacksize of a pthread to below
PTHREAD_STACK_MIN causes pthread_attr_setstacksize() to fail, which
means that the thread will use the default stack size. This fix
makes sure that multipathd never requests a stack size less than
PTHREAD_STACK_MIN.
--
dm-devel mailing list
dm-devel@redhat.com
https://www.redhat.com/mailman/listinfo/dm-devel
03-13-2009, 07:55 PM
Benjamin Marzinski
set pthread stack size to at least PTHREAD_STACK_MIN
> Hmm. I don't quite agree. I run into the same problem, but having
> discovered that we're not checking any return values at all here
> we should rather do the prudent thing and check them once and for all.
>
> I've chosen this approach:
>
> diff --git a/libmultipath/log_pthread.c b/libmultipath/log_pthread.c
> index 9e9aebe..c33480e 100644
> --- a/libmultipath/log_pthread.c
> +++ b/libmultipath/log_pthread.c
> @@ -53,9 +53,30 @@ static void * log_thread (void * et)
> void log_thread_start (void)
> {
> pthread_attr_t attr;
> + size_t stacksize;
>
> logdbg(stderr,"enter log_thread_start
");
>
> + if (pthread_attr_init(&attr)) {
> + fprintf(stderr,"can't initialize log thread
");
> + exit(1);
> + }
> +
> + if (pthread_attr_getstacksize(&attr, &stacksize) != 0)
> + stacksize = PTHREAD_STACK_MIN:
> +
> + /* Check if the stacksize is large enough */
> + if (stacksize < (64 * 1024))
> + stacksize = 64 * 1024;
> +
> + /* Set stacksize and try to reinitialize attr if failed */
> + if (stacksize > PTHREAD_STACK_MIN &&
> + pthread_attr_setstacksize(&attr, stacksize) != 0 &&
> + pthread_attr_init(&attr)) {
> + fprintf(stderr,"can't set log thread stack size
");
> + exit(1);
> + }
> +
> logq_lock = (pthread_mutex_t *) malloc(sizeof(pthread_mutex_t));
> logev_lock = (pthread_mutex_t *) malloc(sizeof(pthread_mutex_t));
> logev_cond = (pthread_cond_t *) malloc(sizeof(pthread_cond_t));
> @@ -64,9 +85,6 @@ void log_thread_start (void)
> pthread_mutex_init(logev_lock, NULL);
> pthread_cond_init(logev_cond, NULL);
>
> - pthread_attr_init(&attr);
> - pthread_attr_setstacksize(&attr, 64 * 1024);
> -
> if (log_init("multipathd", 0)) {
> fprintf(stderr,"can't initialize log buffer
");
> exit(1);
>
>
> This way we'll at least be notified if something goes wrong in
> the future. We shouldn't make the same mistake again and
> ignore error codes which don't happen to trigger now.
>
> If agreed I'll post the full patch here.
This approach doesn't doesn't actually fix the bug that I see. The
problem I was seeing is that setting the stacksize too small just causes
pthread_attr_setstacksize() to fail, leaving you with the default stack
size. On some architectures, the default stacksize is large, like 10Mb.
Since you start one waiter thread per multipath device, every 100
devices eats up 1Gb of memory. Your approach always uses the default
stack size, unless it's too small. I've never seen problems with the
stack being too small. Only too large. Maybe your experience has been
different.
The other problem is that when I actually read the pthread_attr_init man
page (it can fail. who knew?), I saw that it can fail with ENOMEM. Also,
that it had a function to free it, and that the result of reinitializing
an attr that hadn't been freed was undefined. Clearly, this function
wasn't intended to be called over and over without ever freeing the
attr, which is how we've been using it in multipathd. So, in the spirit
of writing code to the interface, instead of to how it appears to be
currently implemented, how about this.
---
libmultipath/log_pthread.c | 9 +-------
libmultipath/log_pthread.h | 4 ++-
libmultipath/waiter.c | 11 ++--------
libmultipath/waiter.h | 2 +
multipathd/main.c | 48 +++++++++++++++++++++++++++++++++++----------
5 files changed, 48 insertions(+), 26 deletions(-)
--
dm-devel mailing list
dm-devel@redhat.com
https://www.redhat.com/mailman/listinfo/dm-devel
03-16-2009, 10:08 PM
Benjamin Marzinski
set pthread stack size to at least PTHREAD_STACK_MIN
> On Mon, Mar 16, 2009 at 05:12:57PM +0100, Hannes Reinecke wrote:
>> This approach doesn't doesn't actually fix the bug that I see. The
>> problem I was seeing is that setting the stacksize too small just causes
>> pthread_attr_setstacksize() to fail, leaving you with the default stack
>> size. On some architectures, the default stacksize is large, like 10Mb.
>> Since you start one waiter thread per multipath device, every 100
>> devices eats up 1Gb of memory. Your approach always uses the default
>> stack size, unless it's too small. I've never seen problems with the
>> stack being too small. Only too large. Maybe your experience has been
>> different.
>>
> Me neither. Makes me wonder if we _really_ need to set the stacksize.
> After all, I'm not aware that we're having any excessive stack usage
> somewhere. Maybe we can simplify it by removing the stack attribute
> setting altogether?
>
> I'll see if I can get the different stacksizes and just compare them
> to the 'updated' setting. Maybe there's no big difference after all...
I definitely see a problem if we use the default stacksize on ia64
machines. In RHEL5 at least, it's 10Mb per thread. With one waiter
thread per multipath device, you get a gigabyte of memory wasted on
machines with over a hundred multipath devices.
>> The other problem is that when I actually read the pthread_attr_init man
>> page (it can fail. who knew?), I saw that it can fail with ENOMEM. Also,
>> that it had a function to free it, and that the result of reinitializing
>> an attr that hadn't been freed was undefined. Clearly, this function
>> wasn't intended to be called over and over without ever freeing the
>> attr, which is how we've been using it in multipathd. So, in the spirit
>> of writing code to the interface, instead of to how it appears to be
>> currently implemented, how about this.
> Hmm. You're not freeing the attribute for all non-logging threads neither.
But I only allocate it once.
> Oversight?
Any time a new device is added, we need the waiter thread attribute. I
suppose I could free it after each waiter thread is started, and then
reallocate it again, but it seems like sort of a waste since we want the
same values every time.
I don't explicitly deallocate it on shutdown, but no matter what the
implementation is, I expect it will be cleaned up when multipathd
ends.
Or am I missing something?
-Ben
--
dm-devel mailing list
dm-devel@redhat.com
https://www.redhat.com/mailman/listinfo/dm-devel