On Sat, Aug 07, 1999 at 12:46:05PM +1000, Chris wrote:
>
> > The issue is not speed, because this is something we do in the
> > background when there's nothing else to do. The issue is to avoid
> > thrashing the cache.
[...]
> Two things,
You haven't considered SMP yet.
--
This is my .signatur
On Sat, Aug 07, 1999 at 12:46:05PM +1000, Chris wrote:
>
> > The issue is not speed, because this is something we do in the
> > background when there's nothing else to do. The issue is to avoid
> > thrashing the cache.
[...]
> Two things,
You haven't considered SMP yet.
--
This is my .signatu
Dag-Erling Smorgrav wrote:
>
> Chris writes:
> > Anyways thats all I can think of. The only way I can see that using DMA
> > to refresh pages as a faster method is if the DMA controller can do it
> > quicker than the CPU which I doubt is likely, also it will only be
> > useful if it can do 32-bi
Dag-Erling Smorgrav wrote:
>
> Chris <[EMAIL PROTECTED]> writes:
> > Anyways thats all I can think of. The only way I can see that using DMA
> > to refresh pages as a faster method is if the DMA controller can do it
> > quicker than the CPU which I doubt is likely, also it will only be
> > usefu
Chris writes:
> Anyways thats all I can think of. The only way I can see that using DMA
> to refresh pages as a faster method is if the DMA controller can do it
> quicker than the CPU which I doubt is likely, also it will only be
> useful if it can do 32-bit addresses.
Grr.. *read what I f###ing
Chris <[EMAIL PROTECTED]> writes:
> Anyways thats all I can think of. The only way I can see that using DMA
> to refresh pages as a faster method is if the DMA controller can do it
> quicker than the CPU which I doubt is likely, also it will only be
> useful if it can do 32-bit addresses.
Grr..
John-Mark Gurney wrote:
>
> Dag-Erling Smorgrav scribbled this message on Aug 4:
> > "Kelly Yancey" writes:
> > > [...]
> >
> > Which reminds me - has anyone thought of using DMA for zeroing pages,
> > to avoid cache invalidation? The idea is to keep a chunk of zeroes on
> > disk and DMA it into
Peter Jeremy writes:
> Dag-Erling Smorgrav wrote:
> > The idea is to keep a chunk of zeroes on disk and DMA it into memory
> Have you looked at disk latencies recently? A modern CPU could zero-
> fill a decent fraction of its RAM in the time taken to fetch a page of
> zeroes from the platter. A
John-Mark Gurney wrote:
>
> Dag-Erling Smorgrav scribbled this message on Aug 4:
> > "Kelly Yancey" <[EMAIL PROTECTED]> writes:
> > > [...]
> >
> > Which reminds me - has anyone thought of using DMA for zeroing pages,
> > to avoid cache invalidation? The idea is to keep a chunk of zeroes on
> > d
Peter Jeremy <[EMAIL PROTECTED]> writes:
> Dag-Erling Smorgrav <[EMAIL PROTECTED]> wrote:
> > The idea is to keep a chunk of zeroes on disk and DMA it into memory
> Have you looked at disk latencies recently? A modern CPU could zero-
> fill a decent fraction of its RAM in the time taken to fetch
Dag-Erling Smorgrav scribbled this message on Aug 4:
> "Kelly Yancey" writes:
> > [...]
>
> Which reminds me - has anyone thought of using DMA for zeroing pages,
> to avoid cache invalidation? The idea is to keep a chunk of zeroes on
> disk and DMA it into memory instead of clearing pages "manual
On Wed, Aug 04, 1999 at 01:20:59PM +0200, Dag-Erling Smorgrav wrote:
> "Kelly Yancey" writes:
> > [...]
>
> Which reminds me - has anyone thought of using DMA for zeroing pages,
> to avoid cache invalidation? The idea is to keep a chunk of zeroes on
> disk and DMA it into memory instead of cleari
Dag-Erling Smorgrav scribbled this message on Aug 4:
> "Kelly Yancey" <[EMAIL PROTECTED]> writes:
> > [...]
>
> Which reminds me - has anyone thought of using DMA for zeroing pages,
> to avoid cache invalidation? The idea is to keep a chunk of zeroes on
> disk and DMA it into memory instead of cl
On Wed, Aug 04, 1999 at 01:20:59PM +0200, Dag-Erling Smorgrav wrote:
> "Kelly Yancey" <[EMAIL PROTECTED]> writes:
> > [...]
>
> Which reminds me - has anyone thought of using DMA for zeroing pages,
> to avoid cache invalidation? The idea is to keep a chunk of zeroes on
> disk and DMA it into memo
Dag-Erling Smorgrav wrote:
>Which reminds me - has anyone thought of using DMA for zeroing pages,
This sounds reasonable. Some DMA engines support filling regions
and memory-memory copies, but I'm not sure about what can be done
with the DMA engine(s) in PCs.
> The idea is to keep a chunk of zer
Dag-Erling Smorgrav <[EMAIL PROTECTED]> wrote:
>Which reminds me - has anyone thought of using DMA for zeroing pages,
This sounds reasonable. Some DMA engines support filling regions
and memory-memory copies, but I'm not sure about what can be done
with the DMA engine(s) in PCs.
> The idea is to
As Dag-Erling Smorgrav wrote ...
> "Kelly Yancey" writes:
> > [...]
>
> Which reminds me - has anyone thought of using DMA for zeroing pages,
> to avoid cache invalidation? The idea is to keep a chunk of zeroes on
> disk and DMA it into memory instead of clearing pages "manually". This
> assumes
As Dag-Erling Smorgrav wrote ...
> "Kelly Yancey" <[EMAIL PROTECTED]> writes:
> > [...]
>
> Which reminds me - has anyone thought of using DMA for zeroing pages,
> to avoid cache invalidation? The idea is to keep a chunk of zeroes on
> disk and DMA it into memory instead of clearing pages "manual
"Kelly Yancey" writes:
> [...]
Which reminds me - has anyone thought of using DMA for zeroing pages,
to avoid cache invalidation? The idea is to keep a chunk of zeroes on
disk and DMA it into memory instead of clearing pages "manually". This
assumes your disk supports DMA, of course.
DES
--
Dag
"Kelly Yancey" <[EMAIL PROTECTED]> writes:
> [...]
Which reminds me - has anyone thought of using DMA for zeroing pages,
to avoid cache invalidation? The idea is to keep a chunk of zeroes on
disk and DMA it into memory instead of clearing pages "manually". This
assumes your disk supports DMA, of
:while the thread was still fresh, but I got distracted by work and other
:projects, so here it is a tad late.
:
: For those of you who hadn't been following the previous calloc() thread:
: I had theorized that we could see an improvement in the performance of
:...
It's always fun to run the
:while the thread was still fresh, but I got distracted by work and other
:projects, so here it is a tad late.
:
: For those of you who hadn't been following the previous calloc() thread:
: I had theorized that we could see an improvement in the performance of
:...
It's always fun to run th
Sorry for posting this out-of-the-blue, I meant to post it last week
while the thread was still fresh, but I got distracted by work and other
projects, so here it is a tad late.
For those of you who hadn't been following the previous calloc() thread:
I had theorized that we could see an imp
Sorry for posting this out-of-the-blue, I meant to post it last week
while the thread was still fresh, but I got distracted by work and other
projects, so here it is a tad late.
For those of you who hadn't been following the previous calloc() thread:
I had theorized that we could see an im
24 matches
Mail list logo