Bug 12576 - Motherboard fails to detect more than 4 SATA
Summary: Motherboard fails to detect more than 4 SATA
Status: RESOLVED PATCH_ALREADY_AVAILABLE
Alias: None
Product: IO/Storage
Classification: Unclassified
Component: Serial ATA (show other bugs)
Hardware: All Linux
: P1 normal
Assignee: Jeff Garzik
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2009-01-29 13:39 UTC by Rich Ercolani
Modified: 2009-01-31 04:56 UTC (History)
1 user (show)

See Also:
Kernel Version: 2.6.28.1
Subsystem:
Regression: ---
Bisected commit-id:


Attachments

Description Rich Ercolani 2009-01-29 13:39:08 UTC
Latest working kernel version: None
Earliest failing kernel version: 2.6.28.1
Distribution: Fedora 10
Hardware Environment: Intel BX97520J, two SATA chips onboard, Marvell 88SE6145 and Intel 82801GR/GH AHCI.
Software Environment: Fedora 10
Problem Description:
I have a motherboard with the two above SATA chipsets, each with four ports.

It seems like the motherboard is failing to detect more than 4 SATA disks at a time. I've tried to plug in a fifth disk, and I receive a timeout on the SATA port.

It does not matter if it's on one chipset or the other, and I've swapped it across multiple disks, SATA cables, and SATA ports. I still can't see more than 4 devices at a time.

Steps to reproduce:
1) Plug 5 SATA devices in, any 5.
2) Watch as you find 4 devices.
3) Be sad.

Only interesting thing I spotted in dmesg:
ata8: link is slow to respond, please be patient (ready=0)
ata8: SRST failed (errno=-16)
ata8: link is slow to respond, please be patient (ready=0)
ata8: SRST failed (errno=-16)
ata8: link is slow to respond, please be patient (ready=0)
irq 16: nobody cared (try booting with the "irqpoll" option)
Pid: 0, comm: swapper Not tainted 2.6.28.1-19.fc10.x86_64 #1
Call Trace:
 <IRQ>  [<ffffffff810896f3>] __report_bad_irq+0x38/0x87
 [<ffffffff81089850>] note_interrupt+0x10e/0x172
 [<ffffffff81089f0e>] handle_fasteoi_irq+0xac/0xdb
 [<ffffffff81013dcf>] do_IRQ+0xfc/0x175
 [<ffffffff81011758>] ret_from_intr+0x0/0x2e
 <EOI>  [<ffffffff810184a0>] ? mwait_idle+0x3e/0x4f
 [<ffffffff81018497>] ? mwait_idle+0x35/0x4f
 [<ffffffff81010242>] ? cpu_idle+0x5b/0xb4
 [<ffffffff81350f34>] ? start_secondary+0x197/0x19c
handlers:
[<ffffffff8125512a>] (usb_hcd_irq+0x0/0xb3)
[<ffffffff8123e5bd>] (ata_sff_interrupt+0x0/0xc2)
Disabling IRQ #16
Comment 1 Rich Ercolani 2009-01-29 14:01:18 UTC
Oh, I told a lie.

Apparently I fail at counting cables, and the machine is just failing on the 6145 part.

Isn't this a pata_marvell bug?
Comment 2 Rich Ercolani 2009-01-29 19:34:10 UTC
Resolved by ahci.marvell_enable=1
Comment 3 Tejun Heo 2009-01-30 21:54:45 UTC
cc'ing Mark.  Any ideas?
Comment 4 Mark Lord 2009-01-31 04:56:07 UTC
No ideas here.  Marvell won't let me even peek at the 6145 datasheets (AHCI, not sata_mv).  I do have a motherboard here with one of those chips on it, though -- never tried using the ports for anything yet.

-ml

Note You need to log in before you can comment on or make changes to this bug.