Opened 15 years ago
Closed 7 years ago
#4050 closed bug (not reproducible)
ATA driver cannot recognize 2nd HDD
Reported by: | mt | Owned by: | marcusoverhagen |
---|---|---|---|
Priority: | normal | Milestone: | R1 |
Component: | Drivers/Disk/ATA | Version: | R1/Development |
Keywords: | Cc: | ||
Blocked By: | Blocking: | #4535 | |
Platform: | x86 |
Description
With IDE driver, system can recognize 2nd HDD's BFS partitions.
But, with ATA driver cannot recognize 2nd HDD's BFS partitions.
(See DriveSetup screenshots)
Tested hrev31322, gcc4/gcc2 hybrid
HDDs are connected via JMB363 SATA card, not on board SATA.
Attachments (10)
Change History (32)
by , 15 years ago
Attachment: | DriveSetup_ide.png added |
---|
by , 15 years ago
Attachment: | DriveSetup_ata.png added |
---|
by , 15 years ago
Attachment: | syslog_ide.txt added |
---|
by , 15 years ago
Attachment: | syslog_ata.txt added |
---|
by , 15 years ago
Attachment: | listdev.txt added |
---|
comment:1 by , 15 years ago
Cc: | added |
---|
comment:2 by , 15 years ago
Milestone: | R1 → R1/alpha1 |
---|
follow-up: 5 comment:3 by , 15 years ago
Can you please tell me the brand and type number of the harddisks?
comment:4 by , 15 years ago
Cc: | removed |
---|
The relevant lines seem to be:
686 KERN: KDiskDeviceManager::_ScanPartition(/dev/disk/ata/4/slave/raw) 687 KERN: intel: ep_std_ops(0x1) 688 KERN: trying: partitioning_systems/intel/extended/v1 689 KERN: returned: -1 690 KERN: intel: ep_std_ops(0x2) 691 KERN: trying: partitioning_systems/intel/map/v1 692 KERN: intel: pm_identify_partition(0, 1: 0, 80026361856, 512) 694 KERN: das_read(): request.Wait() returned: Device timeout 695 KERN: returned: -1 696 KERN: trying: file_systems/bfs/v1 697 KERN: das_read(): request.Wait() returned: Device timeout
Looks like a timeout or possibly a stall. Tuning some of the timeout values could proof this.
It's mmlr btw. but I'm on the bugs-list anyway, so no CC is required.
comment:5 by , 15 years ago
Replying to marcusoverhagen:
Can you please tell me the brand and type number of the harddisks?
HDD0: Hitachi HDP725050GLA 500GB
HDD1: Hitachi HDS721680PLA 80GB
I hope this may help you.
comment:6 by , 15 years ago
Blocked By: | 4058 added |
---|---|
Priority: | normal → high |
Looks very similar to issue 4050, improper error recovery
by , 15 years ago
Attachment: | DriveSetup_ata_r32984.png added |
---|
by , 15 years ago
Attachment: | syslog_ata_r32984.txt added |
---|
comment:9 by , 15 years ago
This is a strange problem. There are no further ata related errors in the syslog, except those two in the beginning. I think that error recovery is working for those two. I analyzed them yesterday, it's a failing "check media status". This should not affect further operation.
575 KERN: ata 4 error: command failed, error bit is set. status 0x51, error 0x04 576 KERN: publish device: node 0x80fcf000, path disk/ata/4/master/raw, module drivers/disk/scsi/scsi_disk/device_v1 577 KERN: ata 4 error: command failed, error bit is set. status 0x51, error 0x04 578 KERN: publish device: node 0x80fcf230, path disk/ata/4/slave/raw, module drivers/disk/scsi/scsi_disk/device_v1
Haiku can mount disk/ata/4/master/raw and read it successfully, while disk/ata/4/slave/raw generates those errors, without generating any ata debug output.
The error message seems to be returned by the io_scheduler in scsi_disk.cpp, but there is no accompanying ata error message.
KERN: das_read(): request.Wait() returned: Device timeout
comment:11 by , 15 years ago
Owner: | changed from | to
---|
comment:12 by , 15 years ago
Milestone: | R1/alpha1 → R1/alpha2 |
---|
comment:14 by , 15 years ago
Priority: | high → normal |
---|
comment:15 by , 14 years ago
Milestone: | R1/alpha3 → R1/beta1 |
---|---|
Version: | R1/pre-alpha1 → R1/Development |
sliding this one out to R1/beta1.
comment:17 by , 10 years ago
Component: | Drivers/Disk → Drivers/Disk/ATA |
---|---|
Owner: | changed from | to
by , 10 years ago
Attachment: | Drivesetup_hrev48280.png added |
---|
by , 10 years ago
Attachment: | syslog_ata_hrev48280.txt added |
---|
by , 10 years ago
Attachment: | listdev_hrev48280.txt added |
---|
comment:19 by , 10 years ago
Milestone: | R1/beta1 → R1 |
---|
follow-up: 21 comment:20 by , 7 years ago
Is this system still available to perform another test with a recent image?
comment:21 by , 7 years ago
Replying to kallisti5:
Is this system still available to perform another test with a recent image?
Sorry, I do not have this system. So, could you close the ticket?
comment:22 by , 7 years ago
Resolution: | → not reproducible |
---|---|
Status: | new → closed |
The jmicron controller is running in IDE/ATA compatibility mode, so AHCI is not relevant here. Also partition type being BFS is not relevant.
It looks like we properly detect the harddisk, but then fail whenever trying a read access.
Relevant parts from logfile: