ATAPI over SATA Plextor Px-7125a

puremourning · 01-23-2005, 03:29 PM

Hey there,

I have a SATA plextor px-712sa dvd writer, which doesnt seem to be detected/working in either FC3 (2.6.9 or 2.6.10) or Gentoo (2.6.9 or 2.6.10).

Does anyone know if

a) ATAPI is supported over SATA
b) this device works under linux.

?

i've googled a fair bit, but to no avail. Its been quite frustrating as I read somewhere that atapi is in libata in 2.6.10 kernel, but theres still nothing recognising the drive. not even cdrecord --scanbus.

any help would be great.

regards

EDIT: i typed the model number wrong its px-7125SA (doh) hopefully this will make more sense now?

eeades · 01-24-2005, 10:45 AM

I have this drive working perfectly under Slackware 9.1 and 10.0, on an Intel 875 chipset (built-in SATA). What type of SATA controller is it (some do Not support atapi under any OS). If you have SATA through your chipset, I would use that.

Eric

puremourning · 01-26-2005, 10:41 AM

its an nforce 4 motherboard via the sata_nv driver. could this be an issue with the drver, as i have read that the nforce3 supported this drive (so i assume the nforce 4 does too). I wouldn't like to have to install windoze to check. maybe i could file a bug report against the driver?

this is what dmesg says on boot (it sees the device, ish, but does nothing). do i need to tell it to use a scsi emulation?

Code:

libata version 1.02 loaded.
sata_nv version 0.03
ACPI: PCI interrupt 0000:00:07.0[A] -> GSI 11 (level, low) -> IRQ 11
PCI: Setting latency timer of device 0000:00:07.0 to 64
ata1: SATA max UDMA/133 cmd 0x9F0 ctl 0xBF2 bmdma 0xCC00 irq 11
ata2: SATA max UDMA/133 cmd 0x970 ctl 0xB72 bmdma 0xCC08 irq 11
ata1: no device found (phy stat 00000000)
scsi0 : sata_nv
ata2: dev 0 cfg 49:2f00 82:7c6b 83:7f09 84:4063 85:7c69 86:3e01 87:4063 88:407f
ata2: dev 0 ATA, max UDMA/133, 586114704 sectors: lba48
ata2: dev 0 configured for UDMA/133
scsi1 : sata_nv
  Vendor: ATA       Model: Maxtor 6B300S0    Rev: BANC
  Type:   Direct-Access                      ANSI SCSI revision: 05
ACPI: PCI interrupt 0000:00:08.0[A] -> GSI 10 (level, low) -> IRQ 10
PCI: Setting latency timer of device 0000:00:08.0 to 64
ata3: SATA max UDMA/133 cmd 0x9E0 ctl 0xBE2 bmdma 0xE000 irq 10
ata4: SATA max UDMA/133 cmd 0x960 ctl 0xB62 bmdma 0xE008 irq 10
ata3: no device found (phy stat 00000000)
scsi2 : sata_nv
ata4: dev 0 cfg 49:0f00 82:0000 83:0000 84:0000 85:0000 86:0000 87:0000 88:0407
ata4: dev 0 ATAPI, max UDMA/33
ata4: dev 0 configured for UDMA/33
scsi3 : sata_nv
SCSI device sda: 586114704 512-byte hdwr sectors (300091 MB)
SCSI device sda: drive cache: write back
 /dev/scsi/host1/bus0/target0/lun0: p1 p2 < p5 p6 p7 >
Attached scsi disk sda at scsi1, channel 0, id 0, lun 0

i would have expected something like attached scsi disk sda2 at scsi2,chan... etc for the dvd drive. does anyone understand the ata4: lines ? i'm really easger to get this working.

eeades · 01-27-2005, 11:49 AM

The ata4 lines look like they are talking about the DVD drive. Have you checked for /dev/hda? as I would guess that is where it would show up.

Eric

puremourning · 01-27-2005, 12:37 PM

the only hd* i have is hdc which is my other DVD-ROM drive (PATA)

Code:

ben $ ls /dev/hd*
/dev/hdc
ben $ ls /dev/sd*
/dev/sda  /dev/sda1  /dev/sda2  /dev/sda5  /dev/sda6  /dev/sda7

/dev/sd:
c1b0t0u0  c1b0t0u0p1  c1b0t0u0p2  c1b0t0u0p5  c1b0t0u0p6  c1b0t0u0p7

*frustrating*

puremourning · 01-27-2005, 12:40 PM

ou tof interest, what do you have in dmesg for your -working- system? this may give me some clues.

GregLee · 01-28-2005, 05:27 AM

Quote:

Originally posted by puremourning
its an nforce 4 motherboard via the sata_nv driver.
...
i would have expected something like attached scsi disk sda2 at scsi2,chan... etc for the dvd drive. does anyone understand the ata4: lines ? i'm really easger to get this working.

I have essentially the same problem, with a Plextor PX-716SA, VIA controller. The BIOS recognizes the Plextor, and Linux 2.6.10 sort of sees it:

Code:

/var/log/debug:
Jan 27 17:19:03 hinalea kernel: sata_via version 1.0
Jan 27 17:19:03 hinalea kernel: ata3: dev 0 cfg 49:0f00 82:0000 83:0000 84:0000 85:0000 86:0000 87:0000 88:00
1f
/var/log/messages:
Jan 27 17:19:03 hinalea kernel: sata_via(0000:00:0f.0): routed to hard irq line 10
Jan 27 17:19:03 hinalea kernel: ata3: SATA max UDMA/133 cmd 0xD400 ctl 0xD002 bmdma 0xC000 irq 20
Jan 27 17:19:03 hinalea kernel: ata4: SATA max UDMA/133 cmd 0xC800 ctl 0xC402 bmdma 0xC008 irq 20
Jan 27 17:19:03 hinalea kernel: ata3: dev 0 ATAPI, max UDMA/66
Jan 27 17:19:04 hinalea kernel: ata3: dev 0 configured for UDMA/66
Jan 27 17:19:04 hinalea kernel: scsi2 : sata_via

but looking in /proc and /dev, I don't see any sign of a device to mount.

puremourning · 01-30-2005, 12:45 PM

what is the kernel module called for the Intel 875 chipset? i will look at the source and see if it is indeed the drivers that are causing the problem. it seems sata_nv and sata_via just ignore the device. perhaps the itel driver does something interesting with it.

eeades · 01-31-2005, 07:47 PM

The kernel module for the 875 chipset is sata_PIIX (or really close, I cannot look it up right now). That is really strange that it does not show up. I would think that it would show up as /dev/sdb from the looks of things (and I think it has when I have installed SUSE, but I currently am running Slack with the chipset in compatibility mode, not native, so everything shows up as /dev/hd?). If only the Intel chipsets currently work with SATA optical drives then no wonder they haven't really caught on yet.

Eric

puremourning · 02-01-2005, 12:00 AM

SOLVED AT LAST! solution is to edit include/linux/libata.h and change #undef ATA_ENABLE_ATAPI to #define ATA_ENABLE_ATAPI and recompile kernel. caveat is that it only works in kernel 2.6.10 which is broken on my system. boo. looks like i will be stuck with only 2 usb ports until its fixed.

GregLee · 02-01-2005, 08:55 AM

Quote:

Originally posted by puremourning
SOLVED AT LAST! solution is to edit include/linux/libata.h and change #undef ATA_ENABLE_ATAPI to #define ATA_ENABLE_ATAPI and recompile kernel. caveat is that it only works in kernel 2.6.10 which is broken on my system. boo. looks like i will be stuck with only 2 usb ports until its fixed.

I tried this (Linux 2.6.10, sata_via), and it sort of partially worked. During bootup, Linux reported the name of the drive, but then hung. I had to unplug the power. I had had a cdrom in the drive, so I took this out and tried again. This time I got the message "scsi scan 96 byte inquiry failed. Consider BLIST_INQUIRY_96 for this device." It hung at this point.

Looking at the source in scsi_devinfo.c (where I guess the BLIST flag could be added), I see comments that some drives hang when luns are probed, and I recalled in configuring the kernel I had checked the options to probe all luns and enable vendor-specific options. I unchecked these options, recompiled and tried again. Now it's back to ground zero. The drive isn't seen.

I don't know exactly how to add the BLIST flag, and I haven't tried that yet.

puremourning · 02-01-2005, 09:44 AM

i'm sorry, you've stumped me. for me, it-just-works(tm). Try the kernel dev mailing list for info on how to set that flag. its a brand new drive, so tetsing will have been minimal. looks like we're the guinaepigs!

jourbans · 02-17-2005, 01:46 PM

You might want to try the latest libata dev kernel patches for 2.6.11-rc4:

http://www.kernel.org/pub/linux/kern...garzik/libata/

I'm fighting with the same problem. Try to get this model DVDRW to work in a "pure" sata configuration. I haven't gotten it to work so far, but I haven't finished playing with it yet. I does appear that ATA_ENABLE_ATAPI is definite in libata.h. g/l!

puremourning · 05-22-2005, 11:29 AM

did you manage to get it working?

i use gentoo-sources (gentoo pathched kernel) and i'm unsure how to apply those patches. as of now the 2.6.12 kernels aren't available in gentoo portage so i can't try them - i believe libata is updated in 2.6.12?

jourbans · 05-23-2005, 02:32 PM

As of right now, using 2.6.12-rc3, and enabling ATA_ENABLE_ATAPI in libata.h, I can do the following:
- Mount cd's and dvd's to my filesystem
- Watch DVD's (reading the drive as a raw device)
- Rip CD's with cdparanoia (more raw device reading again I believe)

I can't burn with it though. Looks like the packet writing code isn't quite there yet.