Hi, > -----Original Message----- > From: Felipe Contreras [mailto:felipe.contreras@xxxxxxxxx] > Sent: Sunday, May 16, 2010 5:13 PM > To: Guzman Lugo, Fernando > Cc: linux-omap; Ramirez Luna, Omar > Subject: Re: [PATCH 02/14] dspbridge: deh: trivial cleanups > > On Mon, May 17, 2010 at 1:02 AM, Guzman Lugo, Fernando > <fernando.lugo@xxxxxx> wrote: > > >> + for (c = 0; c < GPTIMER_IRQ_WAIT_MAX_CNT; c++) > >> + if ((omap_dm_timer_read_status(timer) & > GPTIMER_IRQ_OVERFLOW)) > >> + return; > >> + > >> + pr_err("%s: GPTimer interrupt failed\n", __func__); > > > > In case of not timeout the timer is not disabled, that could prevent > power transition to off. It should be disable as soon as it is not needed > which is after calling dump_dsp_stack. > > Right, I originally had timer_disable() outside of wait_for_timeout(), > and only later decided to put it in. However, do we still need to let > the timer on? Can't we disable the timer always before dumping the > stack? The rest of the code seems to run fine with the timer disabled. If there is no f/i clock the DSP cannot ack the interruption, maybe reading the register without the clocks not cause a fatal in the DSP. But you will have issues if the ack is not done. Regards, Fernando. > > >> omap_dm_timer_disable(timer); > >> } > > Cheers. > > -- > Felipe Contreras -- To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
