I fixed my other suspend/resume problem - it was unrelated to spi. So
restoring the SSP registers is the fix. It's all working correctly for
me now.
I've never submitted a patch before so I'd appreciate a review of this
before I submit it properly:
Index: pxa2xx_spi.c
===================================================================
--- pxa2xx_spi.c (revision 562)
+++ pxa2xx_spi.c (working copy)
@@ -102,6 +102,12 @@
u32 int_cr1;
u32 clear_sr;
u32 mask_sr;
+
+ /* saved state for suspend/resume */
+ u32 saved_sscr0;
+ u32 saved_sscr1;
+ u32 saved_sst0;
+ u32 saved_sspsp;
/* Driver message queue */
struct workqueue_struct *workqueue;
@@ -1570,11 +1576,16 @@
static int pxa2xx_spi_suspend(struct platform_device *pdev, pm_message_t state)
{
struct driver_data *drv_data = platform_get_drvdata(pdev);
+ void *reg = drv_data->ioaddr;
int status = 0;
- /* Check all childern for current power state */
+ /* Save register state */
+ drv_data->saved_sscr0 = read_SSCR0(reg);
+ drv_data->saved_sscr1 = read_SSCR1(reg);
+ if (drv_data->ssp_type != PXA25x_SSP)
+ drv_data->saved_sst0 = read_SSTO(reg);
+ drv_data->saved_sspsp = read_SSPSP(reg);
+
+ /* Check all children for current power state */
if (device_for_each_child(&pdev->dev, &state, suspend_devices) != 0) {
dev_warn(&pdev->dev, "suspend aborted\n");
return -1;
@@ -1592,13 +1604,21 @@
static int pxa2xx_spi_resume(struct platform_device *pdev)
{
struct driver_data *drv_data = platform_get_drvdata(pdev);
+ void *reg = drv_data->ioaddr;
int status = 0;
/* Enable the SSP clock */
pxa_set_cken(drv_data->master_info->clock_enable, 1);
+ /* Load saved SSP configuration */
+ write_SSCR0(0, reg);
+ write_SSCR1(drv_data->saved_sscr1, reg);
+ write_SSCR0(drv_data->saved_sscr0, reg);
+ write_SSITR(0, reg);
+ if (drv_data->ssp_type != PXA25x_SSP)
+ write_SSTO(drv_data->saved_sst0, reg);
+ write_SSPSP(drv_data->saved_sspsp, reg);
+
/* Start the queue running */
status = start_queue(drv_data);
if (status != 0) {
On Fri, Feb 29, 2008 at 10:02 AM, Zik Saleeba <[EMAIL PROTECTED]> wrote:
> On Thu, Feb 21, 2008 at 2:32 PM, Zik Saleeba <[EMAIL PROTECTED]> wrote:
> > I've been having some more problems with my SPI application on the
> > PXA270. When I suspend the system it all goes into suspend mode
> > correctly but when it comes back on resume any call to the pxa2xx_spi
> > code just hangs.
>
> I've looked into this problem a bit more. It looks like suspend/resume
> can never have worked on the pxa2xx spi subsystem.
>
> One problem I've found is that after suspend the SSP registers don't
> come back in their previous state. They need to be saved and restored.
> I've added in some code to do that and it means that at least the
> system doesn't hang as soon as spi is accessed during a resume. It
> does however still hang at some later point. I'm still tracking that
> one down.
>
> I found that saving and restoring SSCR0, SSCR1, SSITR, SST0 (for
> pxa270) and SSPSP was enough to bring the system back to life again.
>
> Would more experienced people like to comment on whether I'm on the
> right track here?
>
> Cheers,
> Zik
>
-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
spi-devel-general mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/spi-devel-general
