On Dec 5, 2012, at 4:48 PM, Edgar Fuß wrote:
>> The attached diff tries to coalesce writes to the journal in MAXPHYS
>> sized and aligned blocks.
> First, many thanks to hannken@ for ending my monthlong performance nightmare!
>
> Btw., this also explains my tstile lockups: writing the journal co
> The attached diff tries to coalesce writes to the journal in MAXPHYS
> sized and aligned blocks.
First, many thanks to hannken@ for ending my monthlong performance nightmare!
Btw., this also explains my tstile lockups: writing the journal could take
tens of seconds; because the journal was locke
On Dec 4, 2012, at 10:11 PM, David Laight wrote:
> On Tue, Dec 04, 2012 at 09:53:11PM +0100, J. Hannken-Illjes wrote:
>>
>> On Dec 4, 2012, at 8:10 PM, Michael van Elst wrote:
>>
>>> hann...@eis.cs.tu-bs.de ("J. Hannken-Illjes") writes:
>>>
The attached diff tries to coalesce writes to t
On Tue, Dec 04, 2012 at 09:53:11PM +0100, J. Hannken-Illjes wrote:
>
> On Dec 4, 2012, at 8:10 PM, Michael van Elst wrote:
>
> > hann...@eis.cs.tu-bs.de ("J. Hannken-Illjes") writes:
> >
> >> The attached diff tries to coalesce writes to the journal in MAXPHYS
> >> sized and aligned blocks.
> >
On Dec 4, 2012, at 8:10 PM, Michael van Elst wrote:
> hann...@eis.cs.tu-bs.de ("J. Hannken-Illjes") writes:
>
>> The attached diff tries to coalesce writes to the journal in MAXPHYS
>> sized and aligned blocks.
> [...]
>> Comments or objections anyone?
>
>> + * Write data to the log.
>> + * Tr
On Tue, Dec 04, 2012 at 07:10:47PM +, Michael van Elst wrote:
> hann...@eis.cs.tu-bs.de ("J. Hannken-Illjes") writes:
>
> >The attached diff tries to coalesce writes to the journal in MAXPHYS
> >sized and aligned blocks.
> [...]
> >Comments or objections anyone?
>
> >+ * Write data to the log
hann...@eis.cs.tu-bs.de ("J. Hannken-Illjes") writes:
>The attached diff tries to coalesce writes to the journal in MAXPHYS
>sized and aligned blocks.
[...]
>Comments or objections anyone?
>+ * Write data to the log.
>+ * Try to coalesce writes and emit MAXPHYS aligned blocks.
Looks fine, but I
The attached diff tries to coalesce writes to the journal in MAXPHYS
sized and aligned blocks.
Background is Edgar Fuß (see some long threads on tech-kern) seeing
very bad I/O speed for a wapbl-enabled file system on a raid5
when wapbl_flush() has to emit a long journal.
On a little test raid5 I