Re: [PATCH] Backport recent IDE updates, take 2

[Erik Andersen <andersen@codepoet.org>]

On Wed Aug 27, 2003 at 04:35:07PM +0100, Alan Cox wrote:
> On Gwe, 2003-08-22 at 23:11, Erik Andersen wrote:
> > CHS), and sanely clamps LBA48 when not doing DMA, preventing data
> > corruption, and avoids idedisk_supports_host_protected_area()
> 
> Doesnt seem to do that right. Marcelo, please don't apply any of this
> stuff until someone actually fixes the lba48 clipping properly

        if (idedisk_supports_lba48(id)) {
                /* drive speaks 48-bit LBA */
                drive->select.b.lba = 1;
                drive->capacity64 = id->lba_capacity_2;
                if (hpa)
                        idedisk_check_hpa(drive);
                if (drive->addressing == 0 && drive->capacity64 > (1ULL)<<28) {
                        /* FIXME:  most controllers that won't do LBA48 with
                         * DMA will do it via PIO so we ought to implement a
                         * PIO fallback...   For now, punt and limit the drive
                         * to 128 GiB to prevent bad things from happening... */
                        drive->capacity64 = (1ULL)<<28;
                }
	}

This is not sufficient?

 -Erik

--
Erik B. Andersen             http://codepoet-consulting.com/
--This message was written using 73% post-consumer electrons--