From mboxrd@z Thu Jan 1 00:00:00 1970 From: jon-hunter@ti.com (Jon Hunter) Date: Thu, 9 Jun 2011 15:49:03 -0500 Subject: [PATCH v3 04/12] Serial: OMAP: Add runtime pm support for omap-serial driver In-Reply-To: References: <1307532194-13039-1-git-send-email-govindraj.raja@ti.com> <1307532194-13039-5-git-send-email-govindraj.raja@ti.com> <4DEFDDE8.3090808@ti.com> Message-ID: <4DF131BF.3050908@ti.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org Hi Govindraj On 06/08/2011 11:35 PM, Govindraj wrote: > On Thu, Jun 9, 2011 at 2:09 AM, Jon Hunter wrote: >> Hi Govindraj, >> >> On 6/8/2011 6:23 AM, Govindraj.R wrote: >> >> [snip] >> >>> + >>> +#define OMAP_UART_AUTOSUSPEND_DELAY (30 * HZ) /* Value is msecs */ >> >> [snip] >> >>> @@ -1295,18 +1381,36 @@ static int serial_omap_probe(struct >>> platform_device *pdev) >>> up->uart_dma.rx_dma_channel = OMAP_UART_DMA_CH_FREE; >>> } >>> >>> + pm_runtime_use_autosuspend(&pdev->dev); >>> + pm_runtime_set_autosuspend_delay(&pdev->dev, >>> + OMAP_UART_AUTOSUSPEND_DELAY); >> >> Something is weird here...DEFAULT_AUTOSUSPEND_DELAY is defined as (30*HZ) >> which would appear to be jiffies (ticks per second) and NOT msecs. However, >> pm_runtime_set_autosuspend is clearly expecting msecs. So this seems to >> conflict. By default 30*HZ for omap would be 30*128 = 3840ms so not quite 4 >> seconds. >> >> What were you intending here? > > Intention is to get approx 3 secs timeout for autosuspend. In that case you should just define DEFAULT_AUTOSUSPEND_DELAY as 30000. The above is just confusing as you are mixing time types and hence, it is not clear what you intend the default timeout to be. Jon