Hi, Jeff:

 Please see my reply below.
I'll resend the patch according to your review/comments later.
Thanks for your patience for newbie like me. :-)

Thanks,
Albert

----- Original Message ----- 
From: "Jeff Garzik" <redacted>
To: "Albert Lee" <redacted>
Cc: "Bartlomiej Zolnierkiewicz" <bzolnier@gmail.com>; "IDE Linux"
[off-list ref]
Sent: Tuesday, October 19, 2004 1:00 AM
Subject: Re: [PATCH/RFC] Port of pdc202xx_new driver to libata

I'm applying to libata-dev, but won't push this upstream for a while.

My comments:

1) why is ROM enabled?  Normally you don't want to do this.

+       /* Enable the ROM */
+       /*
+       if (pdev->resource[PCI_ROM_RESOURCE].start) {
+               pci_write_config_dword(pdev, PCI_ROM_ADDRESS,
+                       pdev->resource[PCI_ROM_RESOURCE].start |
PCI_ROM_ADDRESS
_ENABLE);

Ported from the original driver blindly. I'll remove this stuff.

2) why does this quirk exist?  if pata_pdc2027x (libata) will work with
this hardware, then remove this code (and quirk) completely.  It is up
to the user to choose the correct driver, if multiple drivers will
function properly on the hardware.

+#ifdef PDC_20276_QUIRK
+       /*
+        * Skip Promise PDC20276 attached to I2O RAID controller
+        */
+       if ((pdev->device == PCI_DEVICE_ID_PROMISE_20276) &&
+           (pdev->bus->self) &&
+           (pdev->bus->self->vendor == PCI_VENDOR_ID_INTEL) &&
+           ((pdev->bus->self->device == PCI_DEVICE_ID_INTEL_I960) ||
+            (pdev->bus->self->device == PCI_DEVICE_ID_INTEL_I960RM))) {
+               printk(KERN_INFO DRV_NAME ": Skipping Promise PDC20276 "
+                      "attached to I2O RAID controller.\n");
+               return -ENODEV;
+       }
+#endif

  Ported from the original driver blindly, too.  :-(
I don't know why the original driver doesn't support it.
I'll remove it, too.

3) the quirk PDC_20270_QUIRK needs work to conform to current PCI API
standards.

  Could this also be removed?
This is also ported blindly from the original driver. :-(
It seems to be a bug of DEC21150 PCI south bridge chip,
but there is no detail about the actual problem.
(I've searched the archive, but no clue found.)

-
To unsubscribe from this list: send the line "unsubscribe linux-ide" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html