At Wed, 24 Jun 2009 22:47:12 -0700, Joe Perches wrote:
On Thu, 2009-06-25 at 07:43 +0200, Takashi Iwai wrote:
At Wed, 24 Jun 2009 22:13:35 -0700, Joe Perches wrote:
diff --git a/sound/pci/ens1370.c b/sound/pci/ens1370.c index 18f4d1e..d589bbc 100644 --- a/sound/pci/ens1370.c +++ b/sound/pci/ens1370.c @@ -445,11 +445,11 @@ static irqreturn_t snd_audiopci_interrupt(int irq, void *dev_id);
static struct pci_device_id snd_audiopci_ids[] = { #ifdef CHIP1370
- { 0x1274, 0x5000, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, }, /* ES1370 */
- { PCI_VDEVICE(ENSONIQ, 0x5000), 0, }, /* ES1370 */
#endif #ifdef CHIP1371
- { 0x1274, 0x1371, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, }, /* ES1371 */
- { 0x1274, 0x5880, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, }, /* ES1373 - CT5880 */
- { PCI_VDEVICE(ENSONIQ, 0x1371), 0, }, /* ES1371 */
- { PCI_VDEVICE(ENSONIQ, 0x5880), 0, }, /* ES1373 - CT5880 */ { 0x1102, 0x8938, PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0, }, /* Ectiva EV1938 */
Any reason not to covert this?
Not really, 0x1102 maps to 2 PCI entries, CREATIVE and ECTIVA. The script wouldn't know which is appropriate to use, so I skipped it.
Heh, but it's obvious when you read the commend in the entry :)
(Though, this is a drawback of using PCI_* definition. The definition isn't always unique, so a simple grep isn't possible to search for a device with 1102:8938. So, I'm not 100% fan of such a conversion although I agree that this improves readability much.)
Anyway, could you change that entry and repost? Nno need to repost all in the series, of course :)
thanks,
Takashi