On Thu, Jun 18, 2015 at 05:27:31PM +0800, Gao Pan wrote: > In our former i2c driver, i2c clk is enabled and disabled in > xfer function, which contributes to power saving. However, > the clk enable process brings a busy wait delay until the core > is stable. As a result, the performance is sacrificed. > > To weigh the power consumption and i2c bus performance, runtime > pm is the good solution for it. The clk is enabled when a i2c > transfer starts, and disabled after a specifically defined delay. > > Without the patch the test case (many eeprom reads) executes with approx: > real 1m7.735s > user 0m0.488s > sys 0m20.040s > > With the patch the same test case (many eeprom reads) executes with approx: > real 0m54.241s > user 0m0.440s > sys 0m5.920s > > From the test result, the patch get better performance. Looks OK to me. coccicheck says something, though: drivers/i2c/busses/i2c-imx.c:1098:7-26: pm_runtime_get_sync returns < 0 as error. Unecessary IS_ERR_VALUE at line 1099 drivers/i2c/busses/i2c-imx.c:896:10-29: pm_runtime_get_sync returns < 0 as error. Unecessary IS_ERR_VALUE at line 897 Please fix that and add "kernel@xxxxxxxxxxxxxx" to CC when resending. I know they use the driver a lot. > V2: > As Uwe Kleine-König's suggestion, the version do below changes: > - call clk_prepare_enable in probe to avoid never enabling clock > if CONFIG_PM is disabled > - enable clock before request IRQ in probe > - remove the pm staff in i2c_imx_isr Thanks for this info. If you could put it below "---" that would be preferred by me.
Attachment:
signature.asc
Description: Digital signature
