[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20111219143706.GC1606@m.brq.redhat.com>
Date: Mon, 19 Dec 2011 15:37:06 +0100
From: Jiri Olsa <jolsa@...hat.com>
To: Peter Zijlstra <a.p.zijlstra@...llo.nl>
Cc: acme@...hat.com, mingo@...e.hu, paulus@...ba.org,
cjashfor@...ux.vnet.ibm.com, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 1/3] perf, tool: Add parser generator for events parsing
On Fri, Dec 16, 2011 at 03:02:39PM +0100, Peter Zijlstra wrote:
> On Thu, 2011-12-15 at 16:30 +0100, Jiri Olsa wrote:
SNIP
> > event_breakpoint: PE_SEP_BP ':' PE_VALUE event_breakpoint_type modifier
> > event_breakpoint_type: PE_MODIFIER_BPTYPE | empty
> > modifier: PE_MODIFIER_EVENT | empty
>
> This isn't complete, we need means of specifying
> perf_event_attr::config[12] when specifying a raw event.
ok
>
>
SNIP
> > cpu-migrations|migrations |
> > alignment-faults |
> > emulation-faults
>
> It might make sense to allow means of custom symbol -> raw mappings.
>
right, will save some code..
> > PE_NAME_CACHE_TYPE: L1-dcache|l1-d|l1d|L1-data |
> > L1-icache|l1-i|l1i|L1-instruction |
> > LLC|L2 |
> > dTLB|d-tlb|Data-TLB |
> > iTLB|i-tlb|Instruction-TLB |
> > branch|branches|bpu|btb|bpc |
> > node
> >
> > PE_NAME_CACHE_OP_RESULT: load|loads|read |
> > store|stores|write |
> > prefetch|prefetches |
> > speculative-read|speculative-load |
> > refs|Reference|ops|access |
> > misses|miss
> >
> > PE_SEP_RAW: 'r'
> > PE_SEP_BP: 'mem'
> > PE_MODIFIER_EVENT: :[ukhp]{1,2}
>
> You can actually have more than 2 modifiers, eg. "upp".
>
> [u] | [k] | [h] | [p]{,3} would be more correct I think.
yay, the boundaries are indeed [0,3], I thought it was bool..
* precise_ip:
*
* 0 - SAMPLE_IP can have arbitrary
* skid
* 1 - SAMPLE_IP must have constant
* skid
* 2 - SAMPLE_IP requested to have 0
* skid
* 3 - SAMPLE_IP must have 0 skid
*
>
> > PE_MODIFIER_BPTYPE: :[rwx]{1,3}
> > PE_NAME_TP: [a-zA-Z_*?]+
> > PE_VALUE: number
> >
> > Added flex/bison files for event grammar parsing. Added
> > flex/bison Makefile rules plus few special dependencies.
>
> It might be good to include the .c output of flex/bison so that we can
> build perf without a hard dependency on them -- only people changing the
> parser will need them.
ok
thanks,
jirka
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists