From mboxrd@z Thu Jan 1 00:00:00 1970 From: will.deacon@arm.com (Will Deacon) Date: Fri, 17 May 2013 16:59:36 +0100 Subject: [Bug] ARM 'perf' regression by commit a43cb95d5 In-Reply-To: References: <20130517085530.GB22992@mudshark.cambridge.arm.com> <20130517093616.GC23112@mudshark.cambridge.arm.com> <20130517095423.GD23112@mudshark.cambridge.arm.com> Message-ID: <20130517155936.GQ23112@mudshark.cambridge.arm.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Fri, May 17, 2013 at 11:07:37AM +0100, Ming Lei wrote: > On Fri, May 17, 2013 at 5:54 PM, Will Deacon wrote: > > On Fri, May 17, 2013 at 10:48:23AM +0100, Ming Lei wrote: > >> On Fri, May 17, 2013 at 5:36 PM, Will Deacon wrote: > >> > > >> > It's probably easier if you choose a workload, otherwise it's difficult to > >> > see what is `correct' and what is broken. For example, your broken output > >> > seems to be in the smsc95xx driver, so assumedly there's a bunch of > >> > networking going on whereas your other output is in cpuidle_enter_state. > >> > >> If all modules are removed, the result will become OK. And if I only insert one > >> module, the top sample will fall inside the only module, but I am sure > >> nothing is working on the symbols of the module. Ok, I tried to reproduce this on my TC2 with 3.10-rc1 and perf top looks fine. I also tried loading your spi-altera module and it made no difference. I'm using a fairly old perf tool (reports its version as 3.5.5). Will