[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20121220111313.GB1241@krava.brq.redhat.com>
Date: Thu, 20 Dec 2012 12:13:13 +0100
From: Jiri Olsa <jolsa@...hat.com>
To: Andi Kleen <andi@...stfloor.org>
Cc: linux-kernel@...r.kernel.org,
Arnaldo Carvalho de Melo <acme@...hat.com>,
Namhyung Kim <namhyung@...nel.org>,
Corey Ashford <cjashfor@...ux.vnet.ibm.com>,
Frederic Weisbecker <fweisbec@...il.com>,
Ingo Molnar <mingo@...e.hu>, Paul Mackerras <paulus@...ba.org>,
Peter Zijlstra <a.p.zijlstra@...llo.nl>
Subject: Re: [PATCH 3/3] perf tool: Add non arch events for SandyBridge
microarchitecture
On Wed, Dec 19, 2012 at 01:55:43PM -0800, Andi Kleen wrote:
> Jiri Olsa <jolsa@...hat.com> writes:
>
> > Adding non architectural event aliases for Sandy Bridge
> > microarchitecture.
>
> "SandyBridge" is not necessarily unique. There are flavours
> of sandy bridge with slightly different events (e.g. in offcore)
> Also there's uncores too, which are very different between
> different SandyBridges.
>
> I would merge all these little files into a single file.
> I don't think anyone wants hundreds of new files for this,
> if you do it for multiple CPUs.
right, that's better
>
> Besides it would waste insane amounts of space on most file systems
> without tail packing -- each of your little files takes 4K plus an inode
> and directory entry.
>
> Intel normally uses a spreadsheet CSV format.
Is there anything in public I could download and check?
>
> Best would be probably to just teach perf about reading one of those
> standard formats, then they could be just downloaded somewhere.
so, when you say standard format.. CSV? like:
BR_MISP_EXEC.ALL_BRANCHES,event=0x89,umask=0xff
BR_MISP_EXEC.COND,event=0x89,umask=0x1
BR_MISP_EXEC.DIRECT_NEAR_CALL,event=0x89,umask=0x10
BR_MISP_EXEC.INDIRECT_JMP_NON_CALL_RET,event=0x89,umask=0x4
BR_MISP_EXEC.INDIRECT_NEAR_CALL,event=0x89,umask=0x20
BR_MISP_EXEC.NONTAKEN,event=0x89,umask=0x40
BR_MISP_EXEC.RETURN_NEAR,event=0x89,umask=0x8
BR_MISP_EXEC.TAKEN,event=0x89,umask=0x80
would be no big change for alias loader
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