Re: Disabling IRQ #20 problem with JMicron SATA-PATA Controller

From: Robert Hancock
Date: Wed Dec 30 2009 - 14:09:28 EST


On 12/30/2009 03:03 AM, Ozan ÃaÄlayan wrote:
Hi,


I have a new box with Asus P6T-SE board and Intel i7-920 nehalem cpu. The board has a JMicron JMB363 PATA and SATA controller. When the controller is enabled and set to AHCI mode in BIOS, the system slows down dramatically after disabling IRQ #20.

I'm using 2.6.30.9.

I'm posting full dmesg, thanks:

Dec 28 11:32:07 redd kernel: [ 56.833649] ahci 0000:00:1f.2: version 3.0
Dec 28 11:32:07 redd kernel: [ 56.833658] ahci 0000:00:1f.2: PCI INT B -> GSI 20 (level, low) -> IRQ 20
Dec 28 11:32:07 redd kernel: [ 56.833691] ahci: SSS flag set, parallel bus scan disabled
Dec 28 11:32:07 redd kernel: [ 56.833728] ahci 0000:00:1f.2: AHCI 0001.0200 32 slots 6 ports 3 Gbps 0x3f impl SATA mode
Dec 28 11:32:07 redd kernel: [ 56.833730] ahci 0000:00:1f.2: flags: 64bit ncq sntf stag pm led clo pmp pio slum part ems
Dec 28 11:32:07 redd kernel: [ 56.833734] ahci 0000:00:1f.2: setting latency timer to 64
Dec 28 11:32:07 redd kernel: [ 56.839192] scsi0 : ahci
Dec 28 11:32:07 redd kernel: [ 56.839263] scsi1 : ahci
Dec 28 11:32:07 redd kernel: [ 56.839310] scsi2 : ahci
Dec 28 11:32:07 redd kernel: [ 56.839353] scsi3 : ahci
Dec 28 11:32:07 redd kernel: [ 56.839396] scsi4 : ahci
Dec 28 11:32:07 redd kernel: [ 56.839447] scsi5 : ahci
Dec 28 11:32:07 redd kernel: [ 56.839583] ata1: SATA max UDMA/133 abar m2048@0xf7ffc000 port 0xf7ffc100 irq 20
Dec 28 11:32:07 redd kernel: [ 56.839586] ata2: SATA max UDMA/133 abar m2048@0xf7ffc000 port 0xf7ffc180 irq 20
Dec 28 11:32:07 redd kernel: [ 56.839590] ata3: SATA max UDMA/133 abar m2048@0xf7ffc000 port 0xf7ffc200 irq 20
Dec 28 11:32:07 redd kernel: [ 56.839593] ata4: SATA max UDMA/133 abar m2048@0xf7ffc000 port 0xf7ffc280 irq 20
Dec 28 11:32:07 redd kernel: [ 56.839596] ata5: SATA max UDMA/133 irq_stat 0x00400040, connection status changed irq 20
Dec 28 11:32:07 redd kernel: [ 56.839600] ata6: SATA max UDMA/133 irq_stat 0x00400040, connection status changed irq 20
Dec 28 11:32:07 redd kernel: [ 56.839623] ahci 0000:04:00.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16
Dec 28 11:32:07 redd kernel: [ 56.839717] ahci 0000:04:00.0: AHCI 0001.0000 32 slots 2 ports 3 Gbps 0x3 impl SATA mode
Dec 28 11:32:07 redd kernel: [ 56.839721] ahci 0000:04:00.0: flags: 64bit ncq pm led clo pmp pio slum part
Dec 28 11:32:07 redd kernel: [ 56.839728] ahci 0000:04:00.0: setting latency timer to 64
Dec 28 11:32:07 redd kernel: [ 56.839861] scsi6 : ahci
Dec 28 11:32:07 redd kernel: [ 56.839912] scsi7 : ahci
Dec 28 11:32:07 redd kernel: [ 56.839978] ata7: SATA max UDMA/133 abar m8192@0xfbcfe000 port 0xfbcfe100 irq 16
Dec 28 11:32:07 redd kernel: [ 56.839983] ata8: SATA max UDMA/133 abar m8192@0xfbcfe000 port 0xfbcfe180 irq 16
Dec 28 11:32:07 redd kernel: [ 57.147188] ata1: SATA link down (SStatus 0 SControl 300)
Dec 28 11:32:07 redd kernel: [ 57.147218] ata7: SATA link down (SStatus 0 SControl 300)
Dec 28 11:32:07 redd kernel: [ 57.147249] ata8: SATA link down (SStatus 0 SControl 300)
Dec 28 11:32:07 redd kernel: [ 57.201093] irq 20: nobody cared (try booting with the "irqpoll" option)
Dec 28 11:32:07 redd kernel: [ 57.201098] Pid: 0, comm: swapper Not tainted 2.6.30.9-128 #1


Dec 28 11:32:07 redd kernel: [ 57.201100] Call Trace:
Dec 28 11:32:07 redd kernel: [ 57.201108] [<c01759cf>] __report_bad_irq+0x2e/0x6f
Dec 28 11:32:07 redd kernel: [ 57.201113] [<c0175b00>] note_interrupt+0xf0/0x148
Dec 28 11:32:07 redd kernel: [ 57.201116] [<c017601e>] handle_fasteoi_irq+0x7d/0x9b
Dec 28 11:32:07 redd kernel: [ 57.201122] [<c0104cdb>] handle_irq+0x3b/0x48
Dec 28 11:32:07 redd kernel: [ 57.201126] [<c0104648>] do_IRQ+0x40/0x83
Dec 28 11:32:07 redd kernel: [ 57.201130] [<c0103889>] common_interrupt+0x29/0x30
Dec 28 11:32:07 redd kernel: [ 57.201135] [<c0108662>] ? mwait_idle+0x6b/0x89
Dec 28 11:32:07 redd kernel: [ 57.201139] [<c01024b1>] cpu_idle+0x49/0x64
Dec 28 11:32:07 redd kernel: [ 57.201145] [<c03fde07>] start_secondary+0xc6/0xc8
Dec 28 11:32:07 redd kernel: [ 57.201148] handlers:
Dec 28 11:32:07 redd kernel: [ 57.201149] [<f8076414>] (ahci_interrupt+0x0/0xa8 [ahci])
Dec 28 11:32:07 redd kernel: [ 57.201158] Disabling IRQ #20


Dec 28 11:32:07 redd kernel: [ 57.463017] ata2: SATA link down (SStatus 0 SControl 300)
Dec 28 11:32:07 redd kernel: [ 57.778848] ata3: SATA link down (SStatus 0 SControl 300)
Dec 28 11:32:07 redd kernel: [ 58.094678] ata4: SATA link down (SStatus 0 SControl 300)
Dec 28 11:32:07 redd kernel: [ 58.981252] ata5: SATA link up 3.0 Gbps (SStatus 123 SControl 300)
Dec 28 11:32:07 redd kernel: [ 59.099172] ata5.00: ATA-8: ST3750528AS, CC37, max UDMA/133
Dec 28 11:32:07 redd kernel: [ 59.099175] ata5.00: 1465149168 sectors, multi 16: LBA48 NCQ (depth 31/32)
Dec 28 11:32:07 redd kernel: [ 59.399031] ata5.00: configured for UDMA/133
Dec 28 11:32:07 redd kernel: [ 59.410064] scsi 4:0:0:0: Direct-Access ATA ST3750528AS CC37 PQ: 0 ANSI: 5
Dec 28 11:32:07 redd kernel: [ 59.410185] sd 4:0:0:0: Attached scsi generic sg0 type 0
Dec 28 11:32:07 redd kernel: [ 59.410206] sd 4:0:0:0: [sda] 1465149168 512-byte hardware sectors: (750 GB/698 GiB)
Dec 28 11:32:07 redd kernel: [ 59.410222] sd 4:0:0:0: [sda] Write Protect is off
Dec 28 11:32:07 redd kernel: [ 59.410226] sd 4:0:0:0: [sda] Mode Sense: 00 3a 00 00
Dec 28 11:32:07 redd kernel: [ 59.410246] sd 4:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA
Dec 28 11:32:07 redd kernel: [ 59.410331] sda: sda1 sda2< sda5 sda6>
Dec 28 11:32:07 redd kernel: [ 59.699037] sd 4:0:0:0: [sda] Attached SCSI disk
Dec 28 11:32:07 redd kernel: [ 60.285554] ata6: SATA link up 1.5 Gbps (SStatus 113 SControl 300)
Dec 28 11:32:07 redd kernel: [ 60.398475] ata6.00: ATAPI: HL-DT-ST DVDRAM GH22NS50, TN01, max UDMA/100, ATAPI AN
Dec 28 11:32:07 redd kernel: [ 60.698335] ata6.00: configured for UDMA/100

You say this only happens if the JMicron controller is enabled and set to AHCI mode? That's odd because it's the IRQ for the main Intel controller that it's complaining about. If that's really the IRQ it's using then I wonder how the system continues booting at all after it's disabled. Can you post /proc/interrupts output?

My other suggestions would be make sure you have the latest BIOS installed, and try a newer kernel if possible..
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/