From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Rafael J. Wysocki" Date: Mon, 09 Jul 2012 16:53:19 +0000 Subject: Re: ARM: mach-shmobile: recent next build failure Message-Id: <201207091853.19815.rjw@sisk.pl> List-Id: References: In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: linux-sh@vger.kernel.org On Monday, July 09, 2012, Magnus Damm wrote: > On Mon, Jul 9, 2012 at 6:37 AM, Rafael J. Wysocki wrote: > > On Sunday, July 08, 2012, Rafael J. Wysocki wrote: > >> On Sunday, July 08, 2012, Rafael J. Wysocki wrote: > >> > On Sunday, July 08, 2012, Guennadi Liakhovetski wrote: > >> > > Hi > >> > > > >> > > Got this build failure with Friday's next: > >> > > > >> > > linux/arch/arm/mach-shmobile/pm-sh7372.c: In function 'sh7372_pm_init': > >> > > linux/arch/arm/mach-shmobile/pm-sh7372.c:598: error: implicit declaration of function 'sh7372_suspend_init' > >> > > linux/arch/arm/mach-shmobile/pm-sh7372.c:599: error: implicit declaration of function 'sh7372_cpuidle_init' > >> > > make[2]: *** [arch/arm/mach-shmobile/pm-sh7372.o] Error 1 > >> > > > >> > > .config available on request. > >> > > >> > Please attach. > > Guennadi, can you please share? Giving build error reports without > kernel configuration or information about the problematic kconfig > parameters is not very helpful at all. > > If you are worried about attaching large files, why don't you use the > "savedefconfig" make traget? > > >> > Magnus, that appears to be a result of your recent wakeup changes. > >> > >> Guennadi, does the appended patch fix the problem for you? > > > > Well, scratch that, it's not so simple. > > > > The patch below should help instead. > > It looks like you are trying to fix the cpu_resume() and cpu_suspend() > mess. Thanks! > > I somehow thought that the cpu_resume()/cpu_suspend() issue was solved > already, but obviously that's not the case. My feeling is that I think > we should try to fix it on a higher level. This because more or less > all our SoCs that use low level sleep code will have to share it > between Suspend-to-RAM and CPUIdle. > > Also, I wonder if what you are seeing is the same issue as Guennadi? > Let's wait for Guennadi to share his kernel configuration... I'm quite sure it is, perhaps with a slight modification that Guennadi had both CONFIG_PM and CONFIG_CPU_IDLE unset. However, fixing that particular case alone leads to a build problem when CONFIG_PM is unset and CONFIG_CPU_IDLE is set. Thanks, Rafael