2.6.8-rc3-np1

2.6.8-rc3-np1

[Nick Piggin]

http://www.kerneltrap.org/~npiggin/2.6.8-rc3-np1/

Patch is against 2.6.8-rc3-mm2 only at this stage due to significant
memory management changes in there making it difficult to track Linus'
tree as well.

Feedback on the scheduler would be much appreciated, as it might get
a run in Andrew's tree soon.

Nick

Re: 2.6.8-rc3-np1

[Jason Munro]

On 4:52:14 am 08/09/04 Nick Piggin <nickpiggin@yahoo.com.au> wrote:
> http://www.kerneltrap.org/~npiggin/2.6.8-rc3-np1/
>
> Patch is against 2.6.8-rc3-mm2 only at this stage due to significant
> memory management changes in there making it difficult to track Linus'
> tree as well.
>
> Feedback on the scheduler would be much appreciated, as it might get
> a run in Andrew's tree soon.
>

Running great on my Toshiba laptop. No benchmark numbers but I played some
DVDs, ogg files, emerged some pacakges. All very smooth.


\__ Jason Munro
 \__ jason@stdbev.com
  \__ http://hastymail.sourceforge.net/

Re: 2.6.8-rc3-np1

[Martin Schlemmer]

[-- Attachment #1: Type: text/plain, Size: 818 bytes --]

On Mon, 2004-08-09 at 11:52, Nick Piggin wrote:
> http://www.kerneltrap.org/~npiggin/2.6.8-rc3-np1/
> 
> Patch is against 2.6.8-rc3-mm2 only at this stage due to significant
> memory management changes in there making it difficult to track Linus'
> tree as well.
> 
> Feedback on the scheduler would be much appreciated, as it might get
> a run in Andrew's tree soon.
> 

I am trying to get it patched against rc4-mm1, but it seems there
are some issues with the SMT bits - dependent_sleeper for example
is still around although it was removed with all previous patches
(and uses task_t.time_slice which is no longer there).

I assume you forgot to apply them?  Possible to get a complete
patch?  If not, I'll see if I can get something going after some
sleep.


Thanks,

-- 
Martin Schlemmer

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: 2.6.8-rc3-np1

[Matt Heler]

This applied fine here with some tweaking. Removing the nmi fixes from the np 
patch allowed it to apply just fine with some offsets , but not fuzz.

Matt



On Wednesday 11 August 2004 3:13 pm, Martin Schlemmer wrote:
> On Mon, 2004-08-09 at 11:52, Nick Piggin wrote:
> > http://www.kerneltrap.org/~npiggin/2.6.8-rc3-np1/
> >
> > Patch is against 2.6.8-rc3-mm2 only at this stage due to significant
> > memory management changes in there making it difficult to track Linus'
> > tree as well.
> >
> > Feedback on the scheduler would be much appreciated, as it might get
> > a run in Andrew's tree soon.
>
> I am trying to get it patched against rc4-mm1, but it seems there
> are some issues with the SMT bits - dependent_sleeper for example
> is still around although it was removed with all previous patches
> (and uses task_t.time_slice which is no longer there).
>
> I assume you forgot to apply them?  Possible to get a complete
> patch?  If not, I'll see if I can get something going after some
> sleep.
>
>
> Thanks,

Re: 2.6.8-rc3-np1

[Martin Schlemmer]

[-- Attachment #1: Type: text/plain, Size: 1303 bytes --]

On Thu, 2004-08-12 at 01:04, Matt Heler wrote:
> This applied fine here with some tweaking. Removing the nmi fixes from the np 
> patch allowed it to apply just fine with some offsets , but not fuzz.
> 


Yeah, I got it to apply with one small reject I think.  I assume you
did build it?  Do you have CONFIG_SCHED_SMT enabled?

> 
> 
> 
> On Wednesday 11 August 2004 3:13 pm, Martin Schlemmer wrote:
> > On Mon, 2004-08-09 at 11:52, Nick Piggin wrote:
> > > http://www.kerneltrap.org/~npiggin/2.6.8-rc3-np1/
> > >
> > > Patch is against 2.6.8-rc3-mm2 only at this stage due to significant
> > > memory management changes in there making it difficult to track Linus'
> > > tree as well.
> > >
> > > Feedback on the scheduler would be much appreciated, as it might get
> > > a run in Andrew's tree soon.
> >
> > I am trying to get it patched against rc4-mm1, but it seems there
> > are some issues with the SMT bits - dependent_sleeper for example
> > is still around although it was removed with all previous patches
> > (and uses task_t.time_slice which is no longer there).
> >
> > I assume you forgot to apply them?  Possible to get a complete
> > patch?  If not, I'll see if I can get something going after some
> > sleep.
> >
> >
> > Thanks,
-- 
Martin Schlemmer

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: 2.6.8-rc3-np1

[Nick Piggin]

Martin Schlemmer wrote:
> On Mon, 2004-08-09 at 11:52, Nick Piggin wrote:
> 
>>http://www.kerneltrap.org/~npiggin/2.6.8-rc3-np1/
>>
>>Patch is against 2.6.8-rc3-mm2 only at this stage due to significant
>>memory management changes in there making it difficult to track Linus'
>>tree as well.
>>
>>Feedback on the scheduler would be much appreciated, as it might get
>>a run in Andrew's tree soon.
>>
> 
> 
> I am trying to get it patched against rc4-mm1, but it seems there
> are some issues with the SMT bits - dependent_sleeper for example
> is still around although it was removed with all previous patches
> (and uses task_t.time_slice which is no longer there).
> 
> I assume you forgot to apply them?  Possible to get a complete
> patch?  If not, I'll see if I can get something going after some
> sleep.

Yeah, I've missed something. Thanks for the report. I'll fix
it shortly.