> Linux detects the two HPT374 controllers, but not the attached drives.
>
> The Mainboard is an EPOX EP-8K5A3+ with 2 VIA vt8235 ide controllers
> and 4 Highpoint 374.
>
> I increased the channels: limit for the HPT374 in drivers/ide/pci/hpt366.h
> to 4.
> --- drivers/ide/pci/hpt366.h.old 2002-11-19 17:41:33.000000000 +0100
> +++ drivers/ide/pci/hpt366.h 2002-11-19 17:41:45.000000000 +0100
> @@ -508,5 +508,5 @@
> init_hwif: init_hwif_hpt366,
> init_dma: init_dma_hpt366,
> - channels: 2, /* 4 */
> + channels: 4, /* 2 */
> autodma: AUTODMA,
> enablebits: {{0x00,0x00,0x00}, {0x00,0x00,0x00}},
>
I have the same board, and the controller works fine for me in 2.5.4*, as
2.4-ac doesn't contain xfs suport. I only have one drive attached, but as I
remember I first had to configure the (raid) controller' BIOS (Ctrl-H at boot
time) (even for just a bunch of disks) before using the drives. But I'm not
100%ly sure.
I don't think there's any need for your patch, cause the hpt374 only has two
channels, but there are two controllers on that board.
ANother QUestion: Did you ever get the onboard via-rhine NIC working with
IO-APIC (both BIOS and kernel) enabled?
Christian
> I have the same board, and the controller works fine for me in 2.5.4*, as
> 2.4-ac doesn't contain xfs suport. I only have one drive attached, but as I
> remember I first had to configure the (raid) controller' BIOS (Ctrl-H at boot
> time) (even for just a bunch of disks) before using the drives. But I'm not
> 100%ly sure.
I've "been in" the controller BIOS a few times, but never configured
anything because I'm using the linux md driver.
> I don't think there's any need for your patch, cause the hpt374 only has two
> channels, but there are two controllers on that board.
This patch was just a desperate attempt since it wasn't working before.
I reverted the patch right after I sent the mail to the lkml.
> ANother QUestion: Did you ever get the onboard via-rhine NIC working with
> IO-APIC (both BIOS and kernel) enabled?
No I didn't.
via-rhine.c:v1.10-LK1.1.14 May-3-2002 Written by Donald Becker
http://www.scyld.com/network/via-rhine.html
eth1: VIA VT6102 Rhine-II at 0xcc00, 00:04:61:43:88:d9, IRQ 16.
eth1: MII PHY found at address 1, status 0x7849 advertising 05e1 Link 0000.
eth1: Setting full-duplex based on MII #1 link partner capability of 45e1.
...
eth1: Transmit timed out, status 0003, PHY status 786d, resetting...
eth1: reset did not complete in 10 ms.
NETDEV WATCHDOG: eth1: transmit timed out
On 19.11.2002 11:23 Karsten 'soohrt' Desler wrote:
> > I have the same board, and the controller works fine for me in 2.5.4*, as
> > 2.4-ac doesn't contain xfs suport. I only have one drive attached, but as I
>
> > remember I first had to configure the (raid) controller' BIOS (Ctrl-H at
> boot
> > time) (even for just a bunch of disks) before using the drives. But I'm not
>
> > 100%ly sure.
>
> I've "been in" the controller BIOS a few times, but never configured
> anything because I'm using the linux md driver.
Maybe U would try 2.5.48, just to see if it works then.
When I'm back home in about 7 hours, I'll check my bios settings, maybe this
could help you.
> > ANother QUestion: Did you ever get the onboard via-rhine NIC working with
> > IO-APIC (both BIOS and kernel) enabled?
>
> No I didn't.
> via-rhine.c:v1.10-LK1.1.14 May-3-2002 Written by Donald Becker
> http://www.scyld.com/network/via-rhine.html
> eth1: VIA VT6102 Rhine-II at 0xcc00, 00:04:61:43:88:d9, IRQ 16.
> eth1: MII PHY found at address 1, status 0x7849 advertising 05e1 Link 0000.
> eth1: Setting full-duplex based on MII #1 link partner capability of 45e1.
> ...
> eth1: Transmit timed out, status 0003, PHY status 786d, resetting...
> eth1: reset did not complete in 10 ms.
> NETDEV WATCHDOG: eth1: transmit timed out
>
thats just too bad, last I managed to get it working with IO-APICs for the
first time, but i simply was too tired to remember what things I did. I can't
reproduce it anymore:( this NIC's driving me crazy.
Christian
On Tue, 2002-11-19 at 10:33, Christian Guggenberger wrote:
> On 19.11.2002 11:23 Karsten 'soohrt' Desler wrote:
> > > I have the same board, and the controller works fine for me in 2.5.4*, as
> > > 2.4-ac doesn't contain xfs suport. I only have one drive attached, but as I
> >
> > > remember I first had to configure the (raid) controller' BIOS (Ctrl-H at
> > boot
> > > time) (even for just a bunch of disks) before using the drives. But I'm not
> >
> > > 100%ly sure.
> >
> > I've "been in" the controller BIOS a few times, but never configured
> > anything because I'm using the linux md driver.
>
> Maybe U would try 2.5.48, just to see if it works then.
> When I'm back home in about 7 hours, I'll check my bios settings, maybe this
> could help you.
For IDE 2.5.47-ac is current head. 2.5.48 was just too weird and broken
for me to move to a 2.5.48-ac just yet. I hope to push Linus the next
IDE updates pretty soon
> Maybe U would try 2.5.48, just to see if it works then.
I tried 2.5.48 but Linux didn't detect ide3,4,5,6.
---
DEV: registering device: ID = '00:0e.0', name = Triones Technologies HPT374
kobject 00:0e.0: registering
parent is pci0
bus pci: add device 00:0e.0
DEV: registering device: ID = '00:0e.1', name = Triones Technologies HPT374 (#2)
kobject 00:0e.1: registering
parent is pci0
bus pci: add device 00:0e.1
[..]
bus type 'ide' registered
HPT374: IDE controller at PCI slot 00:0e.0
PCI: Found IRQ 10 for device 00:0e.0
PCI: Sharing IRQ 10 with 00:0e.1
HPT374: chipset revision 7
HPT374: not 100% native mode: will probe irqs later
PCI: Found IRQ 10 for device 00:0e.1
PCI: Sharing IRQ 10 with 00:0e.0
---
2.5.47 won't boot.
2.5.46 is working:
---
Linux version 2.5.46 (root@pikelot) (gcc version 2.95.4 20011002 (Debian prerelease)) #2 Tue Nov 19 15:19:44 CET 2002
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
HPT374: IDE controller at PCI slot 00:0e.0
PCI: Found IRQ 10 for device 00:0e.0
PCI: Sharing IRQ 10 with 00:0e.1
HPT374: chipset revision 7
HPT374: not 100%% native mode: will probe irqs later
HPT37X: using 33MHz PCI clock
ide2: BM-DMA at 0xb400-0xb407, BIOS settings: hde:pio, hdf:pio
ide3: BM-DMA at 0xb408-0xb40f, BIOS settings: hdg:DMA, hdh:DMA
PCI: Found IRQ 10 for device 00:0e.1
PCI: Sharing IRQ 10 with 00:0e.0
HPT37X: using 33MHz PCI clock
ide4: BM-DMA at 0xc800-0xc807, BIOS settings: hdi:DMA, hdj:DMA
ide5: BM-DMA at 0xc808-0xc80f, BIOS settings: hdk:DMA, hdl:DMA
hdg: MAXTOR 4K080H4, ATA DISK drive
hdh: WDC WD800AB-00CBA0, ATA DISK drive
ide3 at 0xac00-0xac07,0xb002 on irq 10
hdi: MAXTOR 4K080H4, ATA DISK drive
hdj: WDC WD800AB-00CBA0, ATA DISK drive
ide4 at 0xb800-0xb807,0xbc02 on irq 10
hdk: MAXTOR 4K080H4, ATA DISK drive
hdl: MAXTOR 4K080H4, ATA DISK drive
ide5 at 0xc000-0xc007,0xc402 on irq 10
---
> When I'm back home in about 7 hours, I'll check my bios settings, maybe this
> could help you.
That would be great, thanks.
Karsten
On 19.11.2002 16:22 Karsten Desler wrote:
> ---
>
> > When I'm back home in about 7 hours, I'll check my bios settings, maybe
> this
> > could help you.
>
> That would be great, thanks.
>
okay, my brain really shrinks; with only one hdd attached, you can't create an
array. So here it seems to work out of the box. I just tried 2.4.20-rc-ac1,
which detects my drive connected to that hpt374 (hde).
Maybe you can give this one a try?
Are using the latest BIOS for your mobo? About a week ago there was a new one
posted, somewhere at epox.com...
Christian
> okay, my brain really shrinks; with only one hdd attached, you can't create an
> array. So here it seems to work out of the box. I just tried 2.4.20-rc-ac1,
> which detects my drive connected to that hpt374 (hde).
>
> Maybe you can give this one a try?
It's partially working with 2.4.20-rc2-ac1. Linux detects the HPT 372
controllers /from time to time/.
First (re)boot:
---
Linux version 2.4.20-rc2-ac1 (root@pikelot) (gcc version 2.95.4 20011002 (Debian prerelease)) #2 Tue Nov 19 20:19:24 CET 2002
[..]
PCI: PCI BIOS revision 2.10 entry at 0xfb3d0, last bus=1
PCI: Using configuration type 1
PCI: Probing PCI hardware
Scanning bus 00
Found 00:00 [1106/3099] 000600 00
Found 00:08 [1106/b099] 000604 01
Found 00:40 [5333/8811] 000300 00
Found 00:60 [10ec/8139] 000200 00
Found 00:70 [1103/0008] 000104 00
Found 00:71 [1103/0008] 000104 00
Found 00:88 [1106/3177] 000601 00
Found 00:89 [1106/0571] 000101 00
Found 00:8d [1106/3059] 000401 00
Found 00:90 [1106/3065] 000200 00
Fixups for bus 00
Scanning behind PCI bridge 00:01.0, config 010100, pass 0
Scanning bus 01
Fixups for bus 01
Bus scan for 01 returning with max=01
Scanning behind PCI bridge 00:01.0, config 010100, pass 1
Bus scan for 00 returning with max=01
[..]
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
HPT374: IDE controller at PCI slot 00:0e.0
PCI: Found IRQ 10 for device 00:0e.0
PCI: Sharing IRQ 10 with 00:0e.1
HPT374: chipset revision 7
HPT374: not 100%% native mode: will probe irqs later
HPT37X: using 33MHz PCI clock
ide2: BM-DMA at 0xb400-0xb407, BIOS settings: hde:pio, hdf:pio
ide3: BM-DMA at 0xb408-0xb40f, BIOS settings: hdg:DMA, hdh:DMA
PCI: Found IRQ 10 for device 00:0e.1
PCI: Sharing IRQ 10 with 00:0e.0
[..]
---
So he did find the first two controllers.
I rebooted the machine after about 3 minutes to attach the drives to
ide2, but Linux didn't find the controllers.
---
Linux version 2.4.20-rc2-ac1 (root@pikelot) (gcc version 2.95.4 20011002 (Debian prerelease)) #2 Tue Nov 19 20:19:24 CET 2002
PCI: PCI BIOS revision 2.10 entry at 0xfb3d0, last bus=1
PCI: Using configuration type 1
PCI: Probing PCI hardware
Scanning bus 00
Found 00:00 [1106/3099] 000600 00
Found 00:08 [1106/b099] 000604 01
Found 00:40 [5333/8811] 000300 00
Found 00:60 [10ec/8139] 000200 00
Found 00:70 [1103/0008] 000104 00
Found 00:71 [1103/0008] 000104 00
Found 00:88 [1106/3177] 000601 00
Found 00:89 [1106/0571] 000101 00
Found 00:8d [1106/3059] 000401 00
Found 00:90 [1106/3065] 000200 00
Fixups for bus 00
Scanning behind PCI bridge 00:01.0, config 010100, pass 0
Scanning bus 01
Fixups for bus 01
Bus scan for 01 returning with max=01
Scanning behind PCI bridge 00:01.0, config 010100, pass 1
Bus scan for 00 returning with max=01
PCI: Using IRQ router VIA [1106/3177] at 00:11.0
PCI: Found IRQ 11 for device 00:08.0
PCI: Sharing IRQ 11 with 00:12.0
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
HPT374: IDE controller at PCI slot 00:0e.0
PCI: Found IRQ 10 for device 00:0e.0
PCI: Sharing IRQ 10 with 00:0e.1
HPT374: chipset revision 7
HPT374: not 100%% native mode: will probe irqs later
PCI: Found IRQ 10 for device 00:0e.1
PCI: Sharing IRQ 10 with 00:0e.0
---
So I removed the drives again, but the controllers are still undetected.
(Same dmesg excerpt as above)
I'm confused :-/
Karsten
Hello,
I had nearly the same problem in april. The linux-kernel found the
raid-controller and detected the devices at startup but I couldn't fdisk
these devices or just find the controller after startup.
It's a HPT372 here.
I couldn't solve the problem but will try it soon again. If you find the
solution, let me know.
Gru?
Uli
--
'The box said, 'Requires Windows 95 or better', so I installed Linux - TKK 5
On 19.11.2002 20:35 Karsten Desler wrote:
> > okay, my brain really shrinks; with only one hdd attached, you can't create
> an
> > array. So here it seems to work out of the box. I just tried 2.4.20-rc-ac1,
>
> > which detects my drive connected to that hpt374 (hde).
> >
> > Maybe you can give this one a try?
>
Karsten,
so today I added 4 more drives to my hpt374, tried 2.5.47-ac6 and works
flawlessy...
Sorry, but I can't try 2.4-ac-latest, cause of missing xfs support.
There are some things you could be bitten with:
My Distribution (Debian) only ships with device names up to /dev/hdh ...
So, by default, you'll see the drives attached to ide4(hdi,j) and to
ide5(hdk,l) in the kernel messages, but you can't use 'em...
You have to make device nodes yourself:
mknod /dev/hd* b MAJOR MINOR
where:
hdi : 56 0
hdi1: 56 1
...
hdj : 56 64
hdj1: 56 65
...
hdk : 57 0
hdk1: 57 1
...
hdl : 57 64
hdl1: 57 65
...
see Documentation/devices.txt in the linux kernel sources.
hope this helps for ya!
Good Luck
CHristian
> so today I added 4 more drives to my hpt374, tried 2.5.47-ac6 and works
> flawlessy...
Yep, 2.5.46 and 2.5.47-ac6 work for me too.
> [/dev/hd* device creation]
You're right, thanks - but that's not the problem since the md raid
autodetection happenes before / is mounted.
Anyways, I wondered why I couldn't use hdparm on all the raid devices
but haven't tried to solve the problem :-), thanks.
Bye
Karsten