Re: [Adeos-main] [PATCH] I-pipe tracer for ARM

[Jan Kiszka <jan.kiszka@domain.hid>]

[-- Attachment #1: Type: text/plain, Size: 1340 bytes --]

Sebastian Smolorz wrote:
> Jan Kiszka wrote:
>> Sebastian Smolorz wrote:
>>> Jan Kiszka wrote:
>>>> Try if smaller CONFIG_IPIPE_TRACE_SHIFT makes it boot without
>>>> CONFIG_IPIPE_TRACE_VMALLOC.
>>> I already tried that but it didn't help. First with IPIPE_TRACE_SHIFT set
>>> to 10 and after modifying Kconfig.debug with a value of 5. Booting
>>> impossible.
>> What about starting disabled and arming the tracer later via
>> /proc/ipipe/trace/enable? This sounds like it's not a size issue
>> (TRACE_SHIFT=5 is ridiculous small), rather some memory section is
>> accessed too early.
> 
> It seems to be a combined problem. When the tracer is disabled at boot time, 
> the system starts if IPIPE_TRACE_SHIFT < 14, but not if >= 14. With the 
> tracer enabled at boot time the IPIPE_TRACE_SHIFT value is not relevant, no 
> boot at all.

Then would you mind moving the trace enabling point as I said and post a
patch? And if you have any idea for a characteristic CONFIG-switch that
should lower the default TRACE_SHIFT (CONFIG_EMBEDDED? Or per arch?),
you are welcome as well. TIA.

> 
> But the freeze lasts again a long time (between 90-100). I wonder if this is 
> related to the ARM cache nature.

Hmm, maybe each piece of memory is affected by this lacy inclusion in
some process's page list...

Jan


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 250 bytes --]