Messages in this thread |  | | | Date | Wed, 21 Sep 2005 15:08:26 +0800 | | From | jayakumar alsa <> | | Subject | Re: [PATCH 2.6.13.1 1/1] CS5535 AUDIO ALSA driver |
| |
> > +static int __devinit snd_cs5535audio_probe(struct pci_dev *pci, > > + const struct pci_device_id *pci_id) > > +{ > > + static int dev; > > + snd_card_t *card; > > + cs5535audio_t *cs5535au; > > + int err; > > + > > + if (dev >= SNDRV_CARDS) > > + return -ENODEV; > > + if (!enable[dev]) { > > + dev++; > > + return -ENOENT; > > + } > > + > > + card = snd_card_new(index[dev], id[dev], THIS_MODULE, 0); > > + if (card == NULL) > > + return -ENOMEM; > > + > > + if ((err = snd_cs5535audio_create(card, pci, &cs5535au)) < 0) { > > + snd_card_free(card); > > + return err; > > + } > > + > > + if ((err = snd_cs5535audio_mixer(cs5535au)) < 0) { > > + snd_card_free(card); > > + return err; > > + } > > + > > + if ((err = snd_cs5535audio_pcm(cs5535au)) < 0) { > > + snd_card_free(card); > > + return err; > > + } > > + > > + strcpy(card->driver, DRIVER_NAME); > > + > > + strcpy(card->shortname, "CS5535 Audio"); > > + sprintf(card->longname, "%s %s at 0x%lx, irq %i", > > + card->shortname, card->driver, > > + cs5535au->port, cs5535au->irq); > > + > > + if ((err = snd_card_register(card)) < 0) { > > + snd_card_free(card); > > + return err; > > + } > > + > > + pci_set_drvdata(pci, card); > > + dev++; > > + return 0; > > +} > > Ditto. > > The handling of `dev' is racy ;)
I took a quick look at the dev handling.
% egrep -B20 "dev\+\+" $lintree/sound/pci/*.c <snip> -- ../azt3328.c- return -ENODEV; ../azt3328.c- if (!enable[dev]) { ../azt3328.c: dev++; -- ../azt3328.c- ../azt3328.c- pci_set_drvdata(pci, card); ../azt3328.c: dev++; -- ../cmipci.c- return -ENODEV; ../cmipci.c- if (! enable[dev]) { ../cmipci.c: dev++; -- ../cmipci.c- } ../cmipci.c- pci_set_drvdata(pci, card); ../cmipci.c: dev++; -- <snip> So I think everyone is doing the same thing. I think probe is called from driver_probe_device through device_add and in this case that's from pci_bus_add_device/s from pci_do_scan_bus. I think the device probe functions are therefore called serially and there is no risk of a race condition. I hope that's right.
Thanks, jk - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
|  |