On Fri, Apr 3, 2015 at 11:42 PM, Andrew Morton
wrote:
> On Mon, 30 Mar 2015 13:26:25 -0700 Andrew Morton
> wrote:
>
>> d) fincore() is more expensive
>
> Actually, I kinda take that back. fincore() will be faster than
> preadv2() in the case of a pagecache miss, and slower in the case of a
>
On Fri, Apr 3, 2015 at 11:42 PM, Andrew Morton
a...@linux-foundation.org wrote:
On Mon, 30 Mar 2015 13:26:25 -0700 Andrew Morton a...@linux-foundation.org
wrote:
d) fincore() is more expensive
Actually, I kinda take that back. fincore() will be faster than
preadv2() in the case of a
On Mon, 30 Mar 2015 13:26:25 -0700 Andrew Morton
wrote:
> d) fincore() is more expensive
Actually, I kinda take that back. fincore() will be faster than
preadv2() in the case of a pagecache miss, and slower in the case of a
pagecache hit.
The breakpoint appears to be a hit rate of 30% - if
On Mon, 30 Mar 2015 13:26:25 -0700 Andrew Morton a...@linux-foundation.org
wrote:
d) fincore() is more expensive
Actually, I kinda take that back. fincore() will be faster than
preadv2() in the case of a pagecache miss, and slower in the case of a
pagecache hit.
The breakpoint appears to be
On Fri, Mar 27, 2015 at 12:30 PM, Andrew Morton
wrote:
> On Fri, 27 Mar 2015 08:58:54 -0700 Jeremy Allison wrote:
>
>> On Fri, Mar 27, 2015 at 02:01:59AM -0700, Andrew Morton wrote:
>> > On Fri, 27 Mar 2015 01:48:33 -0700 Christoph Hellwig
>> > wrote:
>> >
>> > > On Fri, Mar 27, 2015 at
On Mon, Mar 30, 2015 at 4:26 PM, Andrew Morton
wrote:
> On Mon, 30 Mar 2015 00:36:04 -0700 Christoph Hellwig
> wrote:
>
>> On Fri, Mar 27, 2015 at 08:58:54AM -0700, Jeremy Allison wrote:
>> > The problem with the above is that we can't tell the difference
>> > between pread2() returning a short
On Mon, Mar 30, 2015 at 3:36 AM, Christoph Hellwig wrote:
> On Fri, Mar 27, 2015 at 08:58:54AM -0700, Jeremy Allison wrote:
>> The problem with the above is that we can't tell the difference
>> between pread2() returning a short read because the pages are not
>> in cache, or because someone
On Mon, Mar 30, 2015 at 6:57 PM, Andrew Morton
wrote:
> On Mon, 30 Mar 2015 18:49:06 -0400 Milosz Tanski wrote:
>
>> > A fincore+pread solution that blocks is simply unsafe
>> > to use for us. We'll have to stay with the threadpool :-(.
>>
>> We're getting data from a network filesystem Ceph in
On Mon, 30 Mar 2015 18:49:06 -0400 Milosz Tanski wrote:
> > A fincore+pread solution that blocks is simply unsafe
> > to use for us. We'll have to stay with the threadpool :-(.
>
> We're getting data from a network filesystem Ceph in our case, but it
> could be pNFS. In many cases those
On Mon, Mar 30, 2015 at 1:19 PM, Jeremy Allison wrote:
> On Mon, Mar 30, 2015 at 12:36:04AM -0700, Christoph Hellwig wrote:
>> On Fri, Mar 27, 2015 at 08:58:54AM -0700, Jeremy Allison wrote:
>> > The problem with the above is that we can't tell the difference
>> > between pread2() returning a
On Mon, 30 Mar 2015 18:40:16 -0400 Milosz Tanski wrote:
> On Mon, Mar 30, 2015 at 2:54 PM, Andrew Morton
> wrote:
> > On Mon, 30 Mar 2015 00:40:20 -0700 Christoph Hellwig
> > wrote:
> >
> >> On Fri, Mar 27, 2015 at 10:04:11AM -0700, Andrew Morton wrote:
> >> > mm... I don't think we should
On Mon, Mar 30, 2015 at 4:32 PM, Jeremy Allison wrote:
> On Mon, Mar 30, 2015 at 01:26:25PM -0700, Andrew Morton wrote:
>>
>> cons:
>>
>> d) fincore() is more expensive
>>
>> e) fincore() will very occasionally block
>
> The above is the killer for Samba. If fincore
> returns true but when we
On Mon, Mar 30, 2015 at 2:54 PM, Andrew Morton
wrote:
> On Mon, 30 Mar 2015 00:40:20 -0700 Christoph Hellwig
> wrote:
>
>> On Fri, Mar 27, 2015 at 10:04:11AM -0700, Andrew Morton wrote:
>> > mm... I don't think we should be adding placeholders to the kernel API
>> > to support code which
On Mon, Mar 30, 2015 at 4:37 PM, Andrew Morton
wrote:
> On Mon, 30 Mar 2015 13:32:27 -0700 Jeremy Allison wrote:
>
>> On Mon, Mar 30, 2015 at 01:26:25PM -0700, Andrew Morton wrote:
>> >
>> > cons:
>> >
>> > d) fincore() is more expensive
>> >
>> > e) fincore() will very occasionally block
>>
>>
On Mon, 30 Mar 2015 13:49:37 -0700 Jeremy Allison wrote:
> > This implies that the samba main thread also has to avoid any memory
> > allocations both direct and within syscall and pagefault - those will
> > occasionally exhibit similar worse-case latency. Is this done now?
>
> We don't do
On Mon, Mar 30, 2015 at 01:37:58PM -0700, Andrew Morton wrote:
> On Mon, 30 Mar 2015 13:32:27 -0700 Jeremy Allison wrote:
>
> > On Mon, Mar 30, 2015 at 01:26:25PM -0700, Andrew Morton wrote:
> > >
> > > cons:
> > >
> > > d) fincore() is more expensive
> > >
> > > e) fincore() will very
On Mon, 30 Mar 2015 13:32:27 -0700 Jeremy Allison wrote:
> On Mon, Mar 30, 2015 at 01:26:25PM -0700, Andrew Morton wrote:
> >
> > cons:
> >
> > d) fincore() is more expensive
> >
> > e) fincore() will very occasionally block
>
> The above is the killer for Samba. If fincore
> returns true
On Mon, Mar 30, 2015 at 01:26:25PM -0700, Andrew Morton wrote:
>
> cons:
>
> d) fincore() is more expensive
>
> e) fincore() will very occasionally block
The above is the killer for Samba. If fincore
returns true but when we schedule the pread
we block, we're hosed.
Once we block, we're done
On Mon, 30 Mar 2015 00:36:04 -0700 Christoph Hellwig wrote:
> On Fri, Mar 27, 2015 at 08:58:54AM -0700, Jeremy Allison wrote:
> > The problem with the above is that we can't tell the difference
> > between pread2() returning a short read because the pages are not
> > in cache, or because someone
On Mon, 30 Mar 2015 00:40:20 -0700 Christoph Hellwig wrote:
> On Fri, Mar 27, 2015 at 10:04:11AM -0700, Andrew Morton wrote:
> > mm... I don't think we should be adding placeholders to the kernel API
> > to support code which hasn't been written, tested, reviewed, merged,
> > etc. It's
On Mon, Mar 30, 2015 at 12:36:04AM -0700, Christoph Hellwig wrote:
> On Fri, Mar 27, 2015 at 08:58:54AM -0700, Jeremy Allison wrote:
> > The problem with the above is that we can't tell the difference
> > between pread2() returning a short read because the pages are not
> > in cache, or because
On Fri, Mar 27, 2015 at 10:04:11AM -0700, Andrew Morton wrote:
> mm... I don't think we should be adding placeholders to the kernel API
> to support code which hasn't been written, tested, reviewed, merged,
> etc. It's possible none of this will ever happen and we end up with a
> syscall nobody
On Fri, Mar 27, 2015 at 08:58:54AM -0700, Jeremy Allison wrote:
> The problem with the above is that we can't tell the difference
> between pread2() returning a short read because the pages are not
> in cache, or because someone truncated the file. So we need some
> way to differentiate this.
Is
On Fri, Mar 27, 2015 at 10:04:11AM -0700, Andrew Morton wrote:
mm... I don't think we should be adding placeholders to the kernel API
to support code which hasn't been written, tested, reviewed, merged,
etc. It's possible none of this will ever happen and we end up with a
syscall nobody
On Fri, Mar 27, 2015 at 08:58:54AM -0700, Jeremy Allison wrote:
The problem with the above is that we can't tell the difference
between pread2() returning a short read because the pages are not
in cache, or because someone truncated the file. So we need some
way to differentiate this.
Is a
On Mon, Mar 30, 2015 at 12:36:04AM -0700, Christoph Hellwig wrote:
On Fri, Mar 27, 2015 at 08:58:54AM -0700, Jeremy Allison wrote:
The problem with the above is that we can't tell the difference
between pread2() returning a short read because the pages are not
in cache, or because someone
On Mon, 30 Mar 2015 13:32:27 -0700 Jeremy Allison j...@samba.org wrote:
On Mon, Mar 30, 2015 at 01:26:25PM -0700, Andrew Morton wrote:
cons:
d) fincore() is more expensive
e) fincore() will very occasionally block
The above is the killer for Samba. If fincore
returns true but
On Mon, 30 Mar 2015 00:40:20 -0700 Christoph Hellwig h...@infradead.org wrote:
On Fri, Mar 27, 2015 at 10:04:11AM -0700, Andrew Morton wrote:
mm... I don't think we should be adding placeholders to the kernel API
to support code which hasn't been written, tested, reviewed, merged,
etc.
On Mon, 30 Mar 2015 00:36:04 -0700 Christoph Hellwig h...@infradead.org wrote:
On Fri, Mar 27, 2015 at 08:58:54AM -0700, Jeremy Allison wrote:
The problem with the above is that we can't tell the difference
between pread2() returning a short read because the pages are not
in cache, or
On Mon, Mar 30, 2015 at 01:26:25PM -0700, Andrew Morton wrote:
cons:
d) fincore() is more expensive
e) fincore() will very occasionally block
The above is the killer for Samba. If fincore
returns true but when we schedule the pread
we block, we're hosed.
Once we block, we're done
On Mon, 30 Mar 2015 18:40:16 -0400 Milosz Tanski mil...@adfin.com wrote:
On Mon, Mar 30, 2015 at 2:54 PM, Andrew Morton
a...@linux-foundation.org wrote:
On Mon, 30 Mar 2015 00:40:20 -0700 Christoph Hellwig h...@infradead.org
wrote:
On Fri, Mar 27, 2015 at 10:04:11AM -0700, Andrew
On Mon, 30 Mar 2015 18:49:06 -0400 Milosz Tanski mil...@adfin.com wrote:
A fincore+pread solution that blocks is simply unsafe
to use for us. We'll have to stay with the threadpool :-(.
We're getting data from a network filesystem Ceph in our case, but it
could be pNFS. In many cases
On Mon, Mar 30, 2015 at 3:36 AM, Christoph Hellwig h...@infradead.org wrote:
On Fri, Mar 27, 2015 at 08:58:54AM -0700, Jeremy Allison wrote:
The problem with the above is that we can't tell the difference
between pread2() returning a short read because the pages are not
in cache, or because
On Mon, Mar 30, 2015 at 4:32 PM, Jeremy Allison j...@samba.org wrote:
On Mon, Mar 30, 2015 at 01:26:25PM -0700, Andrew Morton wrote:
cons:
d) fincore() is more expensive
e) fincore() will very occasionally block
The above is the killer for Samba. If fincore
returns true but when we
On Mon, 30 Mar 2015 13:49:37 -0700 Jeremy Allison j...@samba.org wrote:
This implies that the samba main thread also has to avoid any memory
allocations both direct and within syscall and pagefault - those will
occasionally exhibit similar worse-case latency. Is this done now?
We don't
On Mon, Mar 30, 2015 at 2:54 PM, Andrew Morton
a...@linux-foundation.org wrote:
On Mon, 30 Mar 2015 00:40:20 -0700 Christoph Hellwig h...@infradead.org
wrote:
On Fri, Mar 27, 2015 at 10:04:11AM -0700, Andrew Morton wrote:
mm... I don't think we should be adding placeholders to the kernel
On Mon, Mar 30, 2015 at 4:26 PM, Andrew Morton
a...@linux-foundation.org wrote:
On Mon, 30 Mar 2015 00:36:04 -0700 Christoph Hellwig h...@infradead.org
wrote:
On Fri, Mar 27, 2015 at 08:58:54AM -0700, Jeremy Allison wrote:
The problem with the above is that we can't tell the difference
On Mon, Mar 30, 2015 at 1:19 PM, Jeremy Allison j...@samba.org wrote:
On Mon, Mar 30, 2015 at 12:36:04AM -0700, Christoph Hellwig wrote:
On Fri, Mar 27, 2015 at 08:58:54AM -0700, Jeremy Allison wrote:
The problem with the above is that we can't tell the difference
between pread2() returning
On Mon, Mar 30, 2015 at 01:37:58PM -0700, Andrew Morton wrote:
On Mon, 30 Mar 2015 13:32:27 -0700 Jeremy Allison j...@samba.org wrote:
On Mon, Mar 30, 2015 at 01:26:25PM -0700, Andrew Morton wrote:
cons:
d) fincore() is more expensive
e) fincore() will very occasionally
On Mon, Mar 30, 2015 at 4:37 PM, Andrew Morton
a...@linux-foundation.org wrote:
On Mon, 30 Mar 2015 13:32:27 -0700 Jeremy Allison j...@samba.org wrote:
On Mon, Mar 30, 2015 at 01:26:25PM -0700, Andrew Morton wrote:
cons:
d) fincore() is more expensive
e) fincore() will very
On Mon, Mar 30, 2015 at 6:57 PM, Andrew Morton
a...@linux-foundation.org wrote:
On Mon, 30 Mar 2015 18:49:06 -0400 Milosz Tanski mil...@adfin.com wrote:
A fincore+pread solution that blocks is simply unsafe
to use for us. We'll have to stay with the threadpool :-(.
We're getting data from
On Fri, Mar 27, 2015 at 12:30 PM, Andrew Morton
a...@linux-foundation.org wrote:
On Fri, 27 Mar 2015 08:58:54 -0700 Jeremy Allison j...@samba.org wrote:
On Fri, Mar 27, 2015 at 02:01:59AM -0700, Andrew Morton wrote:
On Fri, 27 Mar 2015 01:48:33 -0700 Christoph Hellwig h...@infradead.org
On Fri, 27 Mar 2015 11:21:26 -0400 Milosz Tanski wrote:
> On Thu, Mar 26, 2015 at 11:28 PM, Andrew Morton
> wrote:
> > On Mon, 16 Mar 2015 14:27:10 -0400 Milosz Tanski wrote:
> >
> >> This patchset introduces two new syscalls preadv2 and pwritev2. They are
> >> the
> >> same syscalls as
On Fri, Mar 27, 2015 at 12:30 PM, Andrew Morton
wrote:
> On Fri, 27 Mar 2015 08:58:54 -0700 Jeremy Allison wrote:
>
>> On Fri, Mar 27, 2015 at 02:01:59AM -0700, Andrew Morton wrote:
>> > On Fri, 27 Mar 2015 01:48:33 -0700 Christoph Hellwig
>> > wrote:
>> >
>> > > On Fri, Mar 27, 2015 at
On Fri, Mar 27, 2015 at 09:30:46AM -0700, Andrew Morton wrote:
>
> But from an interface perspective the behaviour you're asking for is
> insane, frankly - if the kernel copied out 8k of data then pread2()
> should return 8k. Otherwise there's no way for userspace to know that
> the 8k copy
On Fri, Mar 27, 2015 at 11:58 AM, Jeremy Allison wrote:
> On Fri, Mar 27, 2015 at 02:01:59AM -0700, Andrew Morton wrote:
>> On Fri, 27 Mar 2015 01:48:33 -0700 Christoph Hellwig
>> wrote:
>>
>> > On Fri, Mar 27, 2015 at 01:35:16AM -0700, Andrew Morton wrote:
>> > > fincore() doesn't have to be
On Fri, 27 Mar 2015 09:30:46 -0700 Andrew Morton
wrote:
> I expect that this situation (first part in cache, latter part not in
> cache) is rare - for reasonably small requests the common cases will be
> "all cached" and "nothing cached". So perhaps the best approach here
> is for samba to add
On Fri, 27 Mar 2015 08:58:54 -0700 Jeremy Allison wrote:
> On Fri, Mar 27, 2015 at 02:01:59AM -0700, Andrew Morton wrote:
> > On Fri, 27 Mar 2015 01:48:33 -0700 Christoph Hellwig
> > wrote:
> >
> > > On Fri, Mar 27, 2015 at 01:35:16AM -0700, Andrew Morton wrote:
> > > > fincore() doesn't have
On Fri, Mar 27, 2015 at 02:01:59AM -0700, Andrew Morton wrote:
> On Fri, 27 Mar 2015 01:48:33 -0700 Christoph Hellwig
> wrote:
>
> > On Fri, Mar 27, 2015 at 01:35:16AM -0700, Andrew Morton wrote:
> > > fincore() doesn't have to be ugly. Please address the design issues I
> > > raised. How is
On Thu, Mar 26, 2015 at 11:28 PM, Andrew Morton
wrote:
> On Mon, 16 Mar 2015 14:27:10 -0400 Milosz Tanski wrote:
>
>> This patchset introduces two new syscalls preadv2 and pwritev2. They are the
>> same syscalls as preadv and pwrite but with a flag argument. Additionally,
>> preadv2 implements
On Fri, Mar 27, 2015 at 02:01:59AM -0700, Andrew Morton wrote:
> On Fri, 27 Mar 2015 01:48:33 -0700 Christoph Hellwig
> wrote:
>
> > On Fri, Mar 27, 2015 at 01:35:16AM -0700, Andrew Morton wrote:
> > > fincore() doesn't have to be ugly. Please address the design issues I
> > > raised. How is
On Fri, 27 Mar 2015 01:48:33 -0700 Christoph Hellwig wrote:
> On Fri, Mar 27, 2015 at 01:35:16AM -0700, Andrew Morton wrote:
> > fincore() doesn't have to be ugly. Please address the design issues I
> > raised. How is pread2() useful to the class of applications which
> > cannot proceed until
On Fri, Mar 27, 2015 at 01:35:16AM -0700, Andrew Morton wrote:
> fincore() doesn't have to be ugly. Please address the design issues I
> raised. How is pread2() useful to the class of applications which
> cannot proceed until all data is available?
It actually makes them work correctly?
On Fri, 27 Mar 2015 01:18:22 -0700 Christoph Hellwig wrote:
> On Thu, Mar 26, 2015 at 08:28:24PM -0700, Andrew Morton wrote:
> > I still don't understand why pwritev() exists. We discussed this last
> > time but it seems nothing has changed. I'm not seeing here an adequate
> > description of
On Thu, Mar 26, 2015 at 08:28:24PM -0700, Andrew Morton wrote:
> I still don't understand why pwritev() exists. We discussed this last
> time but it seems nothing has changed. I'm not seeing here an adequate
> description of why it exists nor a justification for its addition.
pwritev2? I have
On Fri, Mar 27, 2015 at 09:02:51AM +0100, Volker Lendecke wrote:
> No, this is not the case. Maybe my whole understanding of
> pread is wrong: I always thought that it won't return short
> if the file spans the pread range. EINTR nonwithstanding.
Per Posix it could, however if we do it for
On Thu, Mar 26, 2015 at 11:08:33PM -0700, Andrew Morton wrote:
> On Fri, 27 Mar 2015 06:41:25 +0100 Volker Lendecke
> wrote:
>
> > On Thu, Mar 26, 2015 at 08:28:24PM -0700, Andrew Morton wrote:
> > > A thing which bugs me about pread2() is that it is specifically
> > > tailored to applications
On Fri, 27 Mar 2015 06:41:25 +0100 Volker Lendecke
wrote:
> On Thu, Mar 26, 2015 at 08:28:24PM -0700, Andrew Morton wrote:
> > A thing which bugs me about pread2() is that it is specifically
> > tailored to applications which are able to use a partial read result.
> > ie, by sending it over
On Thu, Mar 26, 2015 at 08:28:24PM -0700, Andrew Morton wrote:
> A thing which bugs me about pread2() is that it is specifically
> tailored to applications which are able to use a partial read result.
> ie, by sending it over the network.
Can you explain what you mean by this? Samba gets a pread
On Thu, Mar 26, 2015 at 08:28:24PM -0700, Andrew Morton wrote:
A thing which bugs me about pread2() is that it is specifically
tailored to applications which are able to use a partial read result.
ie, by sending it over the network.
Can you explain what you mean by this? Samba gets a pread
On Fri, 27 Mar 2015 06:41:25 +0100 Volker Lendecke volker.lende...@sernet.de
wrote:
On Thu, Mar 26, 2015 at 08:28:24PM -0700, Andrew Morton wrote:
A thing which bugs me about pread2() is that it is specifically
tailored to applications which are able to use a partial read result.
ie, by
On Fri, Mar 27, 2015 at 01:35:16AM -0700, Andrew Morton wrote:
fincore() doesn't have to be ugly. Please address the design issues I
raised. How is pread2() useful to the class of applications which
cannot proceed until all data is available?
It actually makes them work correctly? preadv2(
On Thu, Mar 26, 2015 at 08:28:24PM -0700, Andrew Morton wrote:
I still don't understand why pwritev() exists. We discussed this last
time but it seems nothing has changed. I'm not seeing here an adequate
description of why it exists nor a justification for its addition.
pwritev2? I have
On Fri, 27 Mar 2015 01:18:22 -0700 Christoph Hellwig h...@infradead.org wrote:
On Thu, Mar 26, 2015 at 08:28:24PM -0700, Andrew Morton wrote:
I still don't understand why pwritev() exists. We discussed this last
time but it seems nothing has changed. I'm not seeing here an adequate
On Fri, 27 Mar 2015 01:48:33 -0700 Christoph Hellwig h...@infradead.org wrote:
On Fri, Mar 27, 2015 at 01:35:16AM -0700, Andrew Morton wrote:
fincore() doesn't have to be ugly. Please address the design issues I
raised. How is pread2() useful to the class of applications which
cannot
On Thu, Mar 26, 2015 at 11:08:33PM -0700, Andrew Morton wrote:
On Fri, 27 Mar 2015 06:41:25 +0100 Volker Lendecke
volker.lende...@sernet.de wrote:
On Thu, Mar 26, 2015 at 08:28:24PM -0700, Andrew Morton wrote:
A thing which bugs me about pread2() is that it is specifically
tailored to
On Fri, Mar 27, 2015 at 02:01:59AM -0700, Andrew Morton wrote:
On Fri, 27 Mar 2015 01:48:33 -0700 Christoph Hellwig h...@infradead.org
wrote:
On Fri, Mar 27, 2015 at 01:35:16AM -0700, Andrew Morton wrote:
fincore() doesn't have to be ugly. Please address the design issues I
raised.
On Fri, Mar 27, 2015 at 09:02:51AM +0100, Volker Lendecke wrote:
No, this is not the case. Maybe my whole understanding of
pread is wrong: I always thought that it won't return short
if the file spans the pread range. EINTR nonwithstanding.
Per Posix it could, however if we do it for regular
On Fri, 27 Mar 2015 11:21:26 -0400 Milosz Tanski mil...@adfin.com wrote:
On Thu, Mar 26, 2015 at 11:28 PM, Andrew Morton
a...@linux-foundation.org wrote:
On Mon, 16 Mar 2015 14:27:10 -0400 Milosz Tanski mil...@adfin.com wrote:
This patchset introduces two new syscalls preadv2 and
On Fri, Mar 27, 2015 at 11:58 AM, Jeremy Allison j...@samba.org wrote:
On Fri, Mar 27, 2015 at 02:01:59AM -0700, Andrew Morton wrote:
On Fri, 27 Mar 2015 01:48:33 -0700 Christoph Hellwig h...@infradead.org
wrote:
On Fri, Mar 27, 2015 at 01:35:16AM -0700, Andrew Morton wrote:
fincore()
On Fri, Mar 27, 2015 at 12:30 PM, Andrew Morton
a...@linux-foundation.org wrote:
On Fri, 27 Mar 2015 08:58:54 -0700 Jeremy Allison j...@samba.org wrote:
On Fri, Mar 27, 2015 at 02:01:59AM -0700, Andrew Morton wrote:
On Fri, 27 Mar 2015 01:48:33 -0700 Christoph Hellwig h...@infradead.org
On Fri, Mar 27, 2015 at 09:30:46AM -0700, Andrew Morton wrote:
But from an interface perspective the behaviour you're asking for is
insane, frankly - if the kernel copied out 8k of data then pread2()
should return 8k. Otherwise there's no way for userspace to know that
the 8k copy actually
On Fri, 27 Mar 2015 08:58:54 -0700 Jeremy Allison j...@samba.org wrote:
On Fri, Mar 27, 2015 at 02:01:59AM -0700, Andrew Morton wrote:
On Fri, 27 Mar 2015 01:48:33 -0700 Christoph Hellwig h...@infradead.org
wrote:
On Fri, Mar 27, 2015 at 01:35:16AM -0700, Andrew Morton wrote:
On Fri, 27 Mar 2015 09:30:46 -0700 Andrew Morton a...@linux-foundation.org
wrote:
I expect that this situation (first part in cache, latter part not in
cache) is rare - for reasonably small requests the common cases will be
all cached and nothing cached. So perhaps the best approach here
is
On Thu, Mar 26, 2015 at 11:28 PM, Andrew Morton
a...@linux-foundation.org wrote:
On Mon, 16 Mar 2015 14:27:10 -0400 Milosz Tanski mil...@adfin.com wrote:
This patchset introduces two new syscalls preadv2 and pwritev2. They are the
same syscalls as preadv and pwrite but with a flag argument.
On Fri, Mar 27, 2015 at 02:01:59AM -0700, Andrew Morton wrote:
On Fri, 27 Mar 2015 01:48:33 -0700 Christoph Hellwig h...@infradead.org
wrote:
On Fri, Mar 27, 2015 at 01:35:16AM -0700, Andrew Morton wrote:
fincore() doesn't have to be ugly. Please address the design issues I
raised.
On Mon, 16 Mar 2015 14:27:10 -0400 Milosz Tanski wrote:
> This patchset introduces two new syscalls preadv2 and pwritev2. They are the
> same syscalls as preadv and pwrite but with a flag argument. Additionally,
> preadv2 implements an extra RWF_NONBLOCK flag.
I still don't understand why
On Thu, Mar 26, 2015 at 3:12 PM, Milosz Tanski wrote:
> On Thu, Mar 26, 2015 at 7:55 AM, Christoph Hellwig wrote:
>>
>> On Mon, Mar 16, 2015 at 02:27:10PM -0400, Milosz Tanski wrote:
>> > This patchset introduces two new syscalls preadv2 and pwritev2. They are
>> > the
>> > same syscalls as
On Thu, Mar 26, 2015 at 7:55 AM, Christoph Hellwig wrote:
>
> On Mon, Mar 16, 2015 at 02:27:10PM -0400, Milosz Tanski wrote:
> > This patchset introduces two new syscalls preadv2 and pwritev2. They are the
> > same syscalls as preadv and pwrite but with a flag argument. Additionally,
> > preadv2
On Mon, Mar 16, 2015 at 02:27:10PM -0400, Milosz Tanski wrote:
> This patchset introduces two new syscalls preadv2 and pwritev2. They are the
> same syscalls as preadv and pwrite but with a flag argument. Additionally,
> preadv2 implements an extra RWF_NONBLOCK flag.
There was some arugment that
On Mon, Mar 16, 2015 at 02:27:10PM -0400, Milosz Tanski wrote:
This patchset introduces two new syscalls preadv2 and pwritev2. They are the
same syscalls as preadv and pwrite but with a flag argument. Additionally,
preadv2 implements an extra RWF_NONBLOCK flag.
There was some arugment that we
On Thu, Mar 26, 2015 at 3:12 PM, Milosz Tanski mil...@adfin.com wrote:
On Thu, Mar 26, 2015 at 7:55 AM, Christoph Hellwig h...@infradead.org wrote:
On Mon, Mar 16, 2015 at 02:27:10PM -0400, Milosz Tanski wrote:
This patchset introduces two new syscalls preadv2 and pwritev2. They are
the
On Mon, 16 Mar 2015 14:27:10 -0400 Milosz Tanski mil...@adfin.com wrote:
This patchset introduces two new syscalls preadv2 and pwritev2. They are the
same syscalls as preadv and pwrite but with a flag argument. Additionally,
preadv2 implements an extra RWF_NONBLOCK flag.
I still don't
On Thu, Mar 26, 2015 at 7:55 AM, Christoph Hellwig h...@infradead.org wrote:
On Mon, Mar 16, 2015 at 02:27:10PM -0400, Milosz Tanski wrote:
This patchset introduces two new syscalls preadv2 and pwritev2. They are the
same syscalls as preadv and pwrite but with a flag argument. Additionally,
This patchset introduces two new syscalls preadv2 and pwritev2. They are the
same syscalls as preadv and pwrite but with a flag argument. Additionally,
preadv2 implements an extra RWF_NONBLOCK flag.
The RWF_NONBLOCK flag in preadv2 introduces an ability to perform a
non-blocking read from
This patchset introduces two new syscalls preadv2 and pwritev2. They are the
same syscalls as preadv and pwrite but with a flag argument. Additionally,
preadv2 implements an extra RWF_NONBLOCK flag.
The RWF_NONBLOCK flag in preadv2 introduces an ability to perform a
non-blocking read from
86 matches
Mail list logo