Hi,
On Sun, 2005-01-30 at 10:55, Alex Tomas wrote:
> yup, you're right. so, here is an implementation of this.
> tested on UP/SMP by dbench/fsx. Stephen, Andrew, could you
> review the patch and give it a run?
Just to say I haven't forgotten, just been battling this week against
all sorts of ap
> Stephen C Tweedie (SCT) writes:
SCT> Hi,
SCT> On Tue, 2005-01-25 at 19:30, Alex Tomas wrote:
>> >> journal_dirty_metadata(handle, bh)
>> >> {
>> >> transaction->t_reserved--;
>> >> handle->h_buffer_credits--;
>> >> if (jh->b_tcount > 0) {
>> >> /* modifed, no nee
Hi,
On Tue, 2005-01-25 at 19:30, Alex Tomas wrote:
> >> journal_dirty_metadata(handle, bh)
> >> {
> >> transaction->t_reserved--;
> >> handle->h_buffer_credits--;
> >> if (jh->b_tcount > 0) {
> >> /* modifed, no need to track it any more */
> >> transaction-> t
> Stephen C Tweedie (SCT) writes:
>> journal_dirty_metadata(handle, bh)
>> {
>> transaction->t_reserved--;
>> handle->h_buffer_credits--;
>> if (jh->b_tcount > 0) {
>> /* modifed, no need to track it any more */
>> transaction-> t_outstanding_credits++;
>>
Hi,
On Tue, 2005-01-25 at 14:36, Alex Tomas wrote:
> Hi, could you review the following solution?
>
>
> t_outstanding_credits - number of _modified_ blocks in the transaction
> t_reserved - number of blocks all running handle reserved
> transaction size = t_outstanding_credits + t_reserved;
Hi, could you review the following solution?
t_outstanding_credits - number of _modified_ blocks in the transaction
t_reserved - number of blocks all running handle reserved
transaction size = t_outstanding_credits + t_reserved;
#define TSIZE(t)((t)->t_outstanding_credits + (t)
> Stephen C Tweedie (SCT) writes:
>> + /* return credit back to the handle if it was really spent */
>> + if (credits) {
>> + handle->h_buffer_credits++;
>> + spin_lock(&handle->h_transaction->t_handle_lock);
>> + handle->h_transaction->t_outstandi
Hi,
On Mon, 2005-01-24 at 22:24, Alex Tomas wrote:
> hmmm. that's a good catch. so, with this patch A increments h_buffer_credits
> and this one will go to the t_outstanding_credits while the buffer is still
> part of the transaction. indeed, an imbalance.
>
> probably something like the followin
> Stephen C Tweedie (SCT) writes:
>> + /* return credit back to the handle if it was really spent */
>> + if (credits)
>> + handle->h_buffer_credits++;
>> + jh->b_tcount--;
>> + if (jh->b_tcount == 0) {
>> + /*
>> +* this was last reference to
Hi,
On Wed, 2005-01-19 at 15:32, Alex Tomas wrote:
> @@ -1178,8 +1199,40 @@
> void
> journal_release_buffer(handle_t *handle, struct buffer_head *bh, int credits)
> {
> + /* return credit back to the handle if it was really spent */
> + if (credits)
> + handle->h_buffer_cre
10 matches
Mail list logo
