> On Wed, Oct 19, 2005 at 09:05:41AM -0600, Galen Mark Shipman wrote:
>> We changed things a bit in the mpool, the deregister will remove the
>> registration from the cache and then call release, if the reference
>> count
>> is <=0, the memory is then deregistered, otherwise it will be
>> deregiste
On Wed, Oct 19, 2005 at 09:05:41AM -0600, Galen Mark Shipman wrote:
> We changed things a bit in the mpool, the deregister will remove the
> registration from the cache and then call release, if the reference count
> is <=0, the memory is then deregistered, otherwise it will be deregistered
> later
We changed things a bit in the mpool, the deregister will remove the
registration from the cache and then call release, if the reference count
is <=0, the memory is then deregistered, otherwise it will be deregistered
later via another release call. The BTL module increments the reference
count on
Hello Galen,
It seams this issue is still present and can be easily triggered.
(see attached program). Do you have plans to fix it?
On Wed, Sep 21, 2005 at 12:06:18PM -0600, Galen M. Shipman wrote:
> Gleb,
>
> >
> > Gleb Natapov wrote:
> >
> >> Hello Galen,
> >>
> >> Finally I've got some time t
Gleb,
Gleb Natapov wrote:
Hello Galen,
Finally I've got some time to look through the new code.
I have couple of notes. In pml_ob1_rdma.c you try to merge
registrations in the number of places. The code looks like this:
btl_mpool->mpool_deregister(btl_mpool, reg);
btl_mpool->mpool_regis
Gleb Natapov wrote:
Hello Galen,
Finally I've got some time to look through the new code.
I have couple of notes. In pml_ob1_rdma.c you try to merge
registrations in the number of places. The code looks like this:
btl_mpool->mpool_deregister(btl_mpool, reg);
btl_mpool->mpool_register(bt
Hello Galen,
Finally I've got some time to look through the new code.
I have couple of notes. In pml_ob1_rdma.c you try to merge
registrations in the number of places. The code looks like this:
btl_mpool->mpool_deregister(btl_mpool, reg);
btl_mpool->mpool_register(btl_mpool,
On Wed, Sep 14, 2005 at 08:40:49AM -0600, Galen M. Shipman wrote:
> Hi Gleb,
Hello Galen,
I checked out the new code and am looking at it now. Thanks!
>
> Tim and I have incorporated some of the changes you mentioned into a
> new rcache framework. Currently there is a single component in this
Hi Gleb,
Tim and I have incorporated some of the changes you mentioned into a
new rcache framework. Currently there is a single component in this
framework, rcache_rb which is a registration cache based on a red-
black tree and uses an MRU list if the registration is not
persistent. Note
