Date: Mon, 01 Oct 2001 13:24:29 +0200 From: jonas <jonas.bulow@servicefactory.se> To: Jeff Kletsky <jeff+freebsd@wagsky.com> Cc: freebsd-stable@FreeBSD.ORG, freebsd-hardware@FreeBSD.ORG Subject: Re: Promise Ultra100 TX2 and FastTrack100 TX2 (and TX4) Message-ID: <3BB8526C.11B616B9@servicefactory.se> References: <Pine.BSF.4.21.0109301646200.72313-100000@wildside.wagsky.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Hi,
I suffer the same fate! :-(
I have a Promise Fasttrak 100 TX4. It is identified by FreeBSD as two
TX2 controllers:
I experienced the same problem as you, i.e. the computer hangs during
boot. I solved that by configure an explicit IRQ on the PCI slot used by
the promise controler. But a lot of problems remains.
I can only use one disk on each TX2 controller if I want the computer to
boot at all, i.e not hang during boot. Using two disks on one controller
causes the computer to hang as you described. Using the debugger I found
out that it hangs in the function ata_command:
case ATA_WAIT_INTR:
scp->active = ATA_WAIT_INTR;
asleep((caddr_t)scp, PRIBIO, "atacmd", 10 * hz);
=> outb(scp->ioaddr + ATA_CMD, command);
Breakpoint 17, ata_command (scp=0xc17e0b00, device=0, command=236 'ì',
cylinder=0, head=0 '\000', sector=0 '\000', count=0 '\000',
feature=0 '\000', flags=2) at ../../dev/ata/ata-all.c:1518
9: scp->ioaddr = 57248
8: command = 236 'ì'
(gdb) s
(never returns)
The kernel panics if I access the ar0 device. The backtrace is:
(gdb) bt
#0 Debugger (msg=0xc027477b "panic") at
../../i386/i386/db_interface.c:321
#1 0xc015ff48 in panic (fmt=0xc0278737 "biodone: bp %p not busy %d")
at ../../kern/kern_shutdown.c:579
#2 0xc0188576 in biodone (bp=0xc19d2000) at ../../kern/vfs_bio.c:2664
#3 0xc022e536 in ad_interrupt (request=0xc19d4f80)
at ../../dev/ata/ata-disk.c:698
#4 0xc02294fb in ata_intr (data=0xc17e5d80) at
../../dev/ata/ata-all.c:1159
#5 0xc025af3f in find_idesc (maskptr=0xc104a7c0, irq=0)
at ../../i386/isa/intr_machdep.c:590
Now, I have no ideas of how to solve this. :-(
regards,
jonas
Jeff Kletsky wrote:
>
> I have tried each of these cards in two different boxes, 4.4-RELEASE
> as installed, and 4.4-STABLE:
>
> FreeBSD port6.pn.wagsky.com 4.4-STABLE
> FreeBSD 4.4-STABLE #0: Sun Sep 16 16:10:23 PDT 2001
> jeff@devel.pn.wagsky.com:/usr/obj/usr/src/sys/PIIIAV.20010916 i386
>
> and seem to be unable to get the drives to be recognized by FreeBSD.
> This seems to be in conflict with the release notes and HARDWARE.TXT.
> Reading the notes in the group, others have been experiencing similar
> issues with the TX2 series. Is the TX2 different that what is
> referred to in the notes?
>
> The symptoms are as follows:
>
> (already run sh MAKEDEV ad4 ad5 ad6 ad7 -- with no change in behavior)
>
> * One drive on card's IDE1
>
> - BIOS recognizes and loads
> - boot manager sees drive
> - atapci1 is seen on screen and in dmesg.boot
> - ad4 is not seen
>
> * One drive on card's IDE2 (or both populated)
>
> - BIOS recognizes and loads
> - boot manager sees drive
> - atapci1 is (I believe) seen on screen
> - hangs in the boot process just before identifying disks
>
> Motherboards are Intel se440-bx2, P600 and P600E, Promise card in
> first or second PCI slot (from AGP card). No issues with machines
> otherwise.
>
> Ultra100 TX2 is PDC20268 V2.10 B23
> FastTrak100 TX2 is PDC20270 V2.00 B2
>
> >From dmesg.boot:
> [...]
> chip1: <Intel 82371AB Power management controller> port 0x7000-0x700f at device
> 7.3 on pci0
> atapci1: <Promise TX2 ATA100 controller> port 0x1070-0x107f,0x1080-0x1083,0x1088
> -0x108f,0x1084-0x1087,0x1090-0x1097 mem 0xf4000000-0xf4003fff irq 10 at device 1
> 4.0 on pci0
> ata2: at 0x1090 on atapci1
> ata3: at 0x1088 on atapci1
> fxp0: <Intel Pro 10/100B/100+ Ethernet> port 0x1000-0x103f mem 0xf4100000-0xf41f
> ffff,0xf4004000-0xf4004fff irq 9 at device 16.0 on pci0
> [...]
> ppc0: parallel port not found.
> ad0: 19574MB <WDC WD205BA> [39770/16/63] at ata0-master UDMA33
> acd0: CDROM <FX4824T> at ata1-master using PIO4
> Mounting root from ufs:/dev/ad0s1a
>
> (the freezing behavious is similar, but does not proceed to ad0)
>
> Any help on this would be greatly appreciated, even if (for now) it is
> a notation in the notes so that others may not suffer the same fate.
>
> Thanks!
>
> Jeff
>
> To Unsubscribe: send mail to majordomo@FreeBSD.org
> with "unsubscribe freebsd-stable" in the body of the message
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hardware" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3BB8526C.11B616B9>