Kevin Korb (kmk@crc.com)
Wed, 3 Feb 1999 17:50:08 -0500 (EST)
It did not work:
paride: version 1.02 installed
paride: epat registered as protocol 0
pd: pd version 1.04s, major 45, cluster 64, nice 0
pda: epat 1.01, Shuttle EPAT chip c6 at 0x278, mode 3 (EPP-8), delay 1
pda: Maxtor 91360D8, master, 63504 blocks [31M], (63/16/63), fixed media
pda:pda: do_pd_read: status = 0x10000 = TMO
pda1
Now it sees my nice big 13g as a 31m ;-(
~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~
Kevin Korb Phone: (407) 354-0047 x2122
Systems Administrator Email:
Coleman Research Corp. kmk@crc.com (work)
Launch Systems Group kmk@sanitarium.net (personal)
Orlando, Florida http://www.sanitarium.net/
PGP public key available: finger:kmk@magicnet.net
~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~
On Wed, 3 Feb 1999 grant@torque.net wrote:
> > I have a 13.6gig Maxtor IDE hard drive in an H45 (Shuttle EPAT)
> > parrallel enclosure.
>
> Boy, that's sick. ;-)
>
> > Linux can only "see" the first 8gigs of the drive. I
> > read in the large-disk-HOWTO that this was a problem with the IDE driver but
> > it has been fixed in THAT driver but there is no mention of the parport
> > driver.
>
> There's no common code between the "real" IDE driver and the pd driver.
>
> I don't have a disk that reports 16383 cylinders, so you're going to have
> to test the fix for me. If you also happen to have a smaller disk
> that works with the current driver, I'd appreciate it if you could test
> that too after applying this patch.
>
> Now, I'm coding on the fly here, so please be _very_ careful.
>
> Please locate linux/drivers/block/paride/pd.c and go to the function
> pd_identify(). Modify it to look like like the version below, and remake
> the pd module.
>
>
>
> static int pd_identify( int unit )
>
> { int j, cyls;
> char id[PD_ID_LEN+1];
>
> /* WARNING: here there may be dragons. reset() applies to both drives,
> but we call it only on probing the MASTER. This should allow most
> common configurations to work, but be warned that a reset can clear
> settings on the SLAVE drive.
> */
>
> /* Patched to detect and adjust for disks with more than 16383 cylinders
> in the default translation.
> */
>
> if (PD.drive == 0) pd_reset(unit);
>
> pi_connect(PI);
> WR(0,6,DRIVE);
> pd_wait_for(unit,0,DBMSG("before IDENT"));
> pd_send_command(unit,1,0,0,0,0,IDE_IDENTIFY);
>
> if (pd_wait_for(unit,STAT_DRQ,DBMSG("IDENT DRQ")) & STAT_ERR) {
> pi_disconnect(PI);
> return 0;
> }
> pi_read_block(PI,pd_scratch,512);
> pi_disconnect(PI);
>
> cyls = word_val(1);
> if (cyls) pd_init_dev_parms(unit);
>
> PD.sectors = word_val(6);
> PD.heads = word_val(3);
> PD.cylinders = cyls;
>
> if (cyls == 16383) PD.cylinders = word_val(56);
>
> PD.capacity = PD.sectors*PD.heads*PD.cylinders;
>
> for(j=0;j<PD_ID_LEN;j++) id[j^1] = pd_scratch[j+PD_ID_OFF];
> j = PD_ID_LEN-1;
> while ((j >= 0) && (id[j] <= 0x20)) j--;
> j++; id[j] = 0;
>
> PD.removable = (word_val(0) & 0x80);
>
> printk("%s: %s, %s, %d blocks [%dM], (%d/%d/%d), %s media\n",
> PD.name,id,
> PD.drive?"slave":"master",
> PD.capacity,PD.capacity/2048,
> PD.cylinders,PD.heads,PD.sectors,
> PD.removable?"removable":"fixed");
>
> if (!PD.standby) pd_standby_off(unit);
>
> pd_hd[unit<<PD_BITS].nr_sects = PD.capacity;
> pd_hd[unit<<PD_BITS].start_sect = 0;
>
> return 1;
> }
>
> --------------------------------------------------------------------------
> Grant R. Guenther grant@torque.net
> --------------------------------------------------------------------------
>
-- To unsubscribe, send mail to: linux-parport-request@torque.net --
-- with the single word "unsubscribe" in the body of the message. --
This archive was generated by hypermail 2.0b3 on Wed 03 Feb 1999 - 17:51:44 EST