| To: | colin <colin@realtek.com.tw> |
|---|---|
| Subject: | Re: Can I use this kind of performance counters to implement oProfile? |
| From: | Fuxin Zhang <fxzhang@ict.ac.cn> |
| Date: | Mon, 16 Jan 2006 23:18:27 +0800 |
| Cc: | linux-mips@linux-mips.org |
| In-reply-to: | <005101c61a75$43edc6b0$106215ac@realtek.com.tw> |
| Original-recipient: | rfc822;linux-mips@linux-mips.org |
| References: | <005101c61a75$43edc6b0$106215ac@realtek.com.tw> |
| Sender: | linux-mips-bounce@linux-mips.org |
| User-agent: | Thunderbird 1.5 (Windows/20051201) |
You should be able to accumulate counters during process switching or a timer routine, I managed to implement perfctr(another profiling software) for godson-2 cpu, for both cases with/without interrupt support. You can look at generic perfctr code. colin 写道: > Hi all, > Our SOC has performance counters, and we would like to use oProfile on it. > After surveying the oProfile doc, I found that the model of our performance > counters donot seem to fit oProfile. > This is because oProfile uses the interrupts caused by overflow of, say, > cache miss count to estimate the probability of this event in every portion. > Our SOC doesn't emit interrupt when event count overflow. Therefore, > oProfile cannot be used to estimate cache miss event on our chip. Is that > right? > > Regards, > Colin > > > > > > > > > > > > |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | RE: how to emdedded ramdisk.gz in vmlinux for linux-2.6.14?, Marc Karasek |
|---|---|
| Next by Date: | Timer Interrupt, Bharathi Subramanian |
| Previous by Thread: | Can I use this kind of performance counters to implement oProfile?, colin |
| Next by Thread: | Setting gp on pic code, Alex Gonzalez |
| Indexes: | [Date] [Thread] [Top] [All Lists] |