-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Somebody in the thread at some point said: | Andy Green wrote: |> + } while (kbd->hp_irq_count != kbd->hp_irq_count_in_work); | [...] |> + kbd->work_in_progress = 0; | | There's a bit of a race condition here: if an interrupt happens | just betwen checking and clearing work_in_progress, this | interrupt is lost. | | Do we care ? Chances for this to happen look extremely small.
Nicely spotted... if it did happen that the interrupt came again after 100ms of silence at exactly the wrong moment in a window of a couple of us, he would have to replug his jack to get correct operation. But I guess it happens once per million times or less. Anyway I made a patch on top to try to deal with it I will test and send shortly. - -Andy -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkf/c0cACgkQOjLpvpq7dMqi8ACeIIBCp1KEcl4F/szMET6LXa5C YJsAnR9n6Yx9ZkjaSy6MNR/ZmivgDn0w =bkJP -----END PGP SIGNATURE-----
