Re: [PATCH] scsi: smartpqi: mark PM functions as __maybe_unused

2017-05-23 Thread Martin K. Petersen
Arnd, > The newly added suspend/resume support causes harmless warnings when > CONFIG_PM is disabled: > We can avoid the warnings by removing the #ifdef around the handlers > and instead marking them as __maybe_unused, which will let gcc drop > the unused code silently. Applied to 4.13/scsi-que

Re: [PATCH] scsi: smartpqi: mark PM functions as __maybe_unused

2017-05-22 Thread Arnd Bergmann
On Fri, May 19, 2017 at 3:53 AM, Martin K. Petersen wrote: > > Arnd, > >> I notice that today's linux-next no longer contains the patch that >> introduced the warning. > > I had tagged my 4.12 fixes branch with for-next. It should be back to > 4.13 material shortly. Ok, the patch is back now, so

Re: [PATCH] scsi: smartpqi: mark PM functions as __maybe_unused

2017-05-18 Thread Martin K. Petersen
Arnd, > I notice that today's linux-next no longer contains the patch that > introduced the warning. I had tagged my 4.12 fixes branch with for-next. It should be back to 4.13 material shortly. -- Martin K. Petersen Oracle Linux Engineering

Re: [PATCH] scsi: smartpqi: mark PM functions as __maybe_unused

2017-05-18 Thread Arnd Bergmann
On Thu, May 18, 2017 at 10:32 AM, Arnd Bergmann wrote: > The newly added suspend/resume support causes harmless warnings > when CONFIG_PM is disabled: > > smartpqi/smartpqi_init.c:5147:12: error: 'pqi_ctrl_wait_for_pending_io' > defined but not used [-Werror=unused-function] > smartpqi/smartpqi_i

[PATCH] scsi: smartpqi: mark PM functions as __maybe_unused

2017-05-18 Thread Arnd Bergmann
The newly added suspend/resume support causes harmless warnings when CONFIG_PM is disabled: smartpqi/smartpqi_init.c:5147:12: error: 'pqi_ctrl_wait_for_pending_io' defined but not used [-Werror=unused-function] smartpqi/smartpqi_init.c:2019:13: error: 'pqi_wait_until_lun_reset_finished' defined