Andy Green wrote:
Somebody in the thread at some point said:
| Has anyone else used the latest andy git branch and have any experiences
| with suspend/resume? I'm currently seeing the screen coming back as
| grey-scale.
Wow. I think you're probably the only guy using that branch with your
config, and the suspend / resume stuff is mega sensitive to that. Not
to blame your config which is blameless.
Previously when we suspend we forced the LCM into hard reset for the
duration, I took that out to see what would happen and the LCM came back
in resume quicker here without other incident. You might try
uncommenting it in ./drivers/video/display/jbt6k74.c.
static int jbt_suspend(struct spi_device *spi, pm_message_t state)
{
struct jbt_info *jbt = dev_get_drvdata(&spi->dev);
struct jbt6k74_platform_data *jbt6k74_pdata = spi->dev.platform_data;
/* platform needs to register resume dependencies here */
if (jbt6k74_pdata->suspending)
(jbt6k74_pdata->suspending)(0, spi);
/* Save mode for resume */
jbt->last_state = jbt->state;
jbt6k74_enter_state(jbt, JBT_STATE_DEEP_STANDBY);
jbt->have_resumed = 0;
// (jbt6k74_pdata->reset)(0, 0); <==== uncomment
return 0;
}
But even then the problem has to be in resume ordering driven by config
you would think and I dunno how that would work.
-Andy
Thanks Andy. That works for me. The display comes out of suspend in
color again.
