Very Computer

Hi

Kernel 2.4.19-rc3 introduces a new bug regarding the Promise PDC20276
controller. One of my machines has a Gigabyte GA-8IRXP Motherboard with an
onboard Promise PDC20276. The controller can either be run in RAID or in
ATA mode. I operate it in ATA mode. Kernel 2.4.19-rc3 now incorrectly
skips IDE initialization of the PDC20276 because it thinks it's a RAID
controller which results in a kernel panic (the root filesystem is on a
harddisk connected to the Promise controller). It outputs a message like
this before it panics: PDC20276: Skipping RAID controller. This worked
correctly in 2.4.19-rc2.

Regards
Daniel

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Set CONFIG_PDC202XX_FORCE on.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

This works, thank you. This option didn't get my attention at first
because its named Special FastTrak Feature in xconfig. Perhaps this should
be changed.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Yes. The behaviour should be as previous kernels (work with the kernel
driver by default).

The following patch should fix it.

diff -Naur -X /home/marcelo/lib/dontdiff linux.orig/drivers/ide/ide-pci.c linux/drivers/ide/ide-pci.c
--- linux.orig/drivers/ide/ide-pci.c    2002-07-24 02:50:59.000000000 +0000

        {DEVID_VIA_IDE, "VIA_IDE",    NULL,           NULL,           NULL,           NULL,           {{0x00,0x00,0x00}, {0x00,0x00,0x00}},   ON_BOARD,       0 },
        {DEVID_MR_IDE,  "VP_IDE",     PCI_VIA82CXXX,  ATA66_VIA82CXXX,INIT_VIA82CXXX, DMA_VIA82CXXX,  {{0x40,0x02,0x02}, {0x40,0x01,0x01}},   ON_BOARD,       0 },
        {DEVID_VP_IDE,  "VP_IDE",     PCI_VIA82CXXX,  ATA66_VIA82CXXX,INIT_VIA82CXXX, DMA_VIA82CXXX,  {{0x40,0x02,0x02}, {0x40,0x01,0x01}},   ON_BOARD,       0 },
-#ifdef CONFIG_PDC202XX_FORCE
+#ifndef CONFIG_PDC202XX_FORCE
         {DEVID_PDC20246,"PDC20246",  PCI_PDC202XX,   NULL,           INIT_PDC202XX,  NULL,           {{0x00,0x00,0x00}, {0x00,0x00,0x00}},   OFF_BOARD,      16 },
         {DEVID_PDC20262,"PDC20262",  PCI_PDC202XX,   ATA66_PDC202XX, INIT_PDC202XX,  NULL,           {{0x00,0x00,0x00}, {0x00,0x00,0x00}},   OFF_BOARD,      48 },

         */
        pciirq = dev->irq;

-#ifndef CONFIG_PDC202XX_FORCE
+#ifdef CONFIG_PDC202XX_FORCE
        if (dev->class >> 8 == PCI_CLASS_STORAGE_RAID) {
                /*
                 * By rights we want to ignore Promise FastTrak and SuperTrak

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Hi Marcelo

Quote:> Yes. The behaviour should be as previous kernels (work with the kernel
> driver by default).

> The following patch should fix it.

Yes it does. Line 410 of ide-pci.c should be changed too, from:

#else /* !CONFIG_PDC202XX_FORCE */

to

#else /* CONFIG_PDC202XX_FORCE */

Regards
  Daniel

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/