From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: <0x0000@163.com> Received: from m12-11.163.com (m12-11.163.com [220.181.12.11]) by ozlabs.org (Postfix) with SMTP id AF66EDDE0A for ; Mon, 11 May 2009 22:52:48 +1000 (EST) Date: Mon, 11 May 2009 20:36:58 +0800 From: "0x0000" <0x0000@163.com> To: "linuxppc-dev" Subject: some questions about the implementation of oprofile of mpc8560 in linux kernel source code Message-ID: <200905112036570173816@163.com> Mime-Version: 1.0 Content-Type: multipart/alternative; boundary="=====003_Dragon502454831168_=====" List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , This is a multi-part message in MIME format. --=====003_Dragon502454831168_===== Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Dear All, I am a user of oprofile. When I use oprofile, I find some problems. Could you please help me? I have some questions about the open source code of the implementation of oprofile in linux kernel 2.6.26, which is presented op_model_fsl_emb.c, and the path in linux kernel source tree is arch/powerpc/oprofile/op_model_fsl_emb.c. To present this question more clearly, It needs the following steps: 1> I give some cpu event and its counter values, which finally will call the function fsl_emb_reg_setup 2> I execute setup, which finally will call the function fsl_emb_cpu_setup 3> I execute start, which finally will call the function fsl_emb_start 4> After calling start, the MPC8560 CPU will continuing increasing the PMCn value until its value reach 0x80000000. Basing on what metioned in E500CORERM.pdf, this value should cause a exception, and should goto the interrupt routine:fsl_emb_handle_interrupt. But in fact, It never goto this interrupt routine! It always cause a reboot without any information! But while we using MPC7450 CPU, this situation never happened, it works well. Basing on the information metioned above, my questions are: 1> Is there any bug in op_model_fsl_emb.c and its related source code? 2> If possible, could you please tell me what the reason it is? And How to use op_model_fsl_emb.c correctly? Any feedback is appreciated! Thanks & regards Jike 2009-05-11 0x0000 --=====003_Dragon502454831168_===== Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: 7bit

Dear All,

I am a user of oprofile. When I use oprofile, I find some problems. Could you please help me?

I have some questions about the open source code of the implementation of oprofile in linux

kernel 2.6.26, which is presented op_model_fsl_emb.c, and the path in linux kernel source

tree is arch/powerpc/oprofile/op_model_fsl_emb.c.

To present this question more clearly, It needs the following steps:

1> I give some cpu event and its counter values, which finally will call the function fsl_emb_reg_setup

2> I execute setup, which finally will call the function fsl_emb_cpu_setup

3> I execute start, which finally will call the function fsl_emb_start

4> After calling start, the MPC8560 CPU will continuing increasing the PMCn value until its value reach
0x80000000. Basing on what metioned in E500CORERM.pdf, this value should cause a exception, and
should goto the interrupt routine:fsl_emb_handle_interrupt.
But in fact, It never goto this interrupt routine! It always cause a reboot without any information!

But while we using MPC7450 CPU, this situation never happened, it works well.

Basing on the information metioned above, my questions are:
1> Is there any bug in op_model_fsl_emb.c and its related source code?
2> If possible, could you please tell me what the reason it is? And How to use op_model_fsl_emb.c correctly?

Any feedback is appreciated!

Thanks & regards
Jike

2009-05-11

0x0000

--=====003_Dragon502454831168_=====--