-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Somebody in the thread at some point said:
| This is so much better. Screen is now resuming in color. Resume is much | more reliable as well. Yet even more reliable when I take out the | dev_info() call in the _irq routine. I still get a complete lockup | eventually, though. Seems like the longer I leave it in suspend the less | likely it will resume. Hum okay I doubted it would be the only monster. One thing about that work function, it time travels due to the deferral msleep, it can maybe try to come back for another try after, eg, the pwm timers are sleeping and normal scheduling is broken, then funny stuff could happen. I started on a new patch for that but won't be able to complete it until later thisafternoon due to wife owning me on Sundays. | The pops must be coming from the speaker on the side. I heard them even | with the top off so the earpiece isn't even connected, right? They only | happen on resume. Good, no top on my phone either. | At the time of the suspend/resume I have audio opened as a pseudo alsa | device which configures with the same info as stereoout.state. Here is | the actual states: Great I will try that tonight then. | but when it suspends I get | | timeout waiting for SYSCOMP & DBGACK, last DBG_STATUS: 0 | timeout waiting for SYSCOMP & DBGACK, last DBG_STATUS: 0 | timeout waiting for SYSCOMP & DBGACK, last DBG_STATUS: 0 | timeout waiting for SYSCOMP & DBGACK, last DBG_STATUS: 0 | | Program received signal SIGTRAP, Trace/breakpoint trap. | 0xffffffec in ?? () | | and then its pretty much dead. Yes at that time the cpu core and the jtag registers used to debug are actually unpowered, nothing good is going to happen from that direction. - -Andy -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkhU2NAACgkQOjLpvpq7dMoeCgCeLAdx2LUMj+wg2//qcYfttgXa DJgAnR7GrQOP8iPBQJzUMvdTjVw/dVfC =RI3+ -----END PGP SIGNATURE-----
