On Thu, 2014-05-01 at 13:41 +0000, Mark Hambleton wrote:
Wouldn't it have been better to just revert the patch that added the platform specific pre-processing?
I confused. The platform specific pre-processing was part of the
original commit d8063e7015 (HMP: Implement task packing for small tasks
in HMP systems)
This was then modified by commit ba8ed8301f (sched: hmp: Change TC2
packing config to disabled default if present)
What you are doing now is introducing a second config change into LSK...
Only if you compile with TC2.
The revert would have looked like this:
Which just reverts the second commit I mentioned and still leaves TC2
specific pre-processing.
I guess what you're saying is that you also want the default for
hmp_packing_enabled to change to 0, so your multi-platform kernels keep
the behaviour they had before this proposed patch? But to do that would
mean that everyone else who wasn't compiling with TC2 enabled in their
kernel now getting a configuration change instead.
Guess that means then that we should just stick with the status quo
then?
I personally don't care one way or the other, I'm just a
piggy-in-the-middle patch monkey chimera who'd wish people would make up
their minds what they want :-)
So if Mark H, Mark B, and Robin can come to a consensus on what the task
packing config should be, then tell me...
--
Tixy
> #ifdef CONFIG_SCHED_HMP_LITTLE_PACKING
> -#ifndef CONFIG_ARCH_VEXPRESS_TC2
> unsigned int hmp_packing_enabled = 1;
> +#ifndef CONFIG_ARCH_VEXPRESS_TC2
> unsigned int hmp_full_threshold = (NICE_0_LOAD * 9) / 8;
> #else
> /* TC2 has a sharp consumption curve @ around 800Mhz, so
> we aim to spread the load around that frequency. */
> -unsigned int hmp_packing_enabled;
> unsigned int hmp_full_threshold = 650; /* 80% of the 800Mhz freq * NICE_0_LOAD */
> #endif
> #endif
--
Tixy
>
> Mark
>
> > -----Original Message-----
> > From: Jon Medhurst (Tixy) [mailto:tixy@linaro.org]
> > Sent: 01 May 2014 14:37
> > To: Mark Brown; Alex Shi
> > Cc: Linaro Kernel; Chris Redpath; Robin Randhawa; Mark Hambleton
> > Subject: [PATCH] sched: hmp: Remove compile-time configuration of task
> > packing
> >
> > Compile-time configuration is messy and precludes multi-platform kernels
> > so remove it. Task packing should instead be tuned at run-time, e.g. in
> > user-side boot scripts, with something like:
> >
> > echo 1 > /sys/kernel/hmp/packing_enable
> > echo 650 > /sys/kernel/hmp/packing_limit
> >
> > Signed-off-by: Jon Medhurst
tixy@linaro.org
> > ---
> >
> > Mark and Alex, please don't apply this directly. Assuming that this
> > change doesn't provoke any dissent, it can be pulled from the usual
> > place... git://git.linaro.org/arm/big.LITTLE/mp.git for-lsk
> >
> > kernel/sched/fair.c | 7 -------
> > 1 file changed, 7 deletions(-)
> >
> > diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> > index 43857fe..6610622 100644
> > --- a/kernel/sched/fair.c
> > +++ b/kernel/sched/fair.c
> > @@ -3676,15 +3676,8 @@ unsigned int hmp_next_up_threshold = 4096;
> > unsigned int hmp_next_down_threshold = 4096;
> >
> > #ifdef CONFIG_SCHED_HMP_LITTLE_PACKING
> > -#ifndef CONFIG_ARCH_VEXPRESS_TC2
> > unsigned int hmp_packing_enabled = 1;
> > unsigned int hmp_full_threshold = (NICE_0_LOAD * 9) / 8;
> > -#else
> > -/* TC2 has a sharp consumption curve @ around 800Mhz, so
> > - we aim to spread the load around that frequency. */
> > -unsigned int hmp_packing_enabled;
> > -unsigned int hmp_full_threshold = 650; /* 80% of the 800Mhz freq *
> > NICE_0_LOAD */
> > -#endif
> > #endif
> >
> > static unsigned int hmp_up_migration(int cpu, int *target_cpu, struct
> > sched_entity *se);
> > --
> > 1.7.10.4
> >
> >
>