New: Hang in dc395x driver
https://bugzilla.kernel.org/show_bug.cgi?id=15950
Summary: Hang in dc395x driver
Product: SCSI Drivers
Version: 2.5
Kernel Version: 2.6.31
Platform: All
OS/Version: Linux
Tree: Mainline
Status: NEW
Severity: normal
Priority: P1
Component: DC395X
AssignedTo: [email protected]
ReportedBy: [email protected]
Regression: Yes
Kernel 2.6.31 and at least some earlier kernels.
I'm using an old Microtek ScanMaker flatbed scanner connected to a DC395x SCSI card.
dc395x 0000:01:07.0: PCI INT A -> Link[APC2] -> GSI 17 (level, low) -> IRQ 17
dc395x: Used settings: AdapterID=07, Speed=0(20.0MHz), dev_mode=0x57
dc395x: AdaptMode=0x0f, Tags=4(16), DelayReset=1s
dc395x: Connectors: ext50 Termination: Auto Low High
dc395x: Performing initial SCSI bus reset
scsi0 : Tekram DC395(U/UW/F), DC315(U) - ASIC TRM-S1040 v2.05, 2004/03/08
scsi 0:0:6:0: Scanner Scanner 1.80 PQ: 0 ANSI: 4
CCS
scsi 0:0:6:0: Attached scsi generic sg0 type 6sata_nv 0000:00:05.0: version 3.5
When scanning with xsane, after a few minutes I often get a hang in xsane and if I forcibly kill xsane, the scanner is no longer there. When the xsane hang occurs, I see this:
dc395x: reselect: w/o disconnected cmds <06-0>
dc395x: disconnect: Unexpected reselection <06-0>
I'm not sure when this problem started, but it's somewhere in the 2.6.2x region.
I'd be happy to recompile, test and debug if needed.
Thanks,
Tom
Signature · 3 lines
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are watching the assignee of the bug.
Quoted 4 lines · On Sun, 2010-05-09 at 21:19 +0000, [email protected] wrote:
I'm not sure when this problem started, but it's somewhere in the 2.6.2x region.
I'd be happy to recompile, test and debug if needed.
Retesting isn't necessary. There have only been nine changes to that driver in 2.6.2x. The versions of interest (as in works/doesn't work) should be
2.6.26 (should be identical to 2.6.31, so just a something else changed sanity check)
2.6.25 (lots of pointless changes + sense buffer replacement) 2.6.34 (pid removal and sg chaining)
2.6.23 (accessors change)
2.6.22 (spelling only fix!)
James
[Bug 15950] Hang in dc395x driver
In reply to [email protected]https://bugzilla.kernel.org/show_bug.cgi?id=15950
--- Comment #1 from Anonymous Emailer <[email protected]> 2010-05-09 22:07:58 ---
Reply-To: [email protected]
Quoted 4 lines · On Sun, 2010-05-09 at 21:19 +0000, [email protected] wrote:
I'm not sure when this problem started, but it's somewhere in the 2.6.2x region.
I'd be happy to recompile, test and debug if needed.
Retesting isn't necessary. There have only been nine changes to that driver in 2.6.2x. The versions of interest (as in works/doesn't work) should be
2.6.26 (should be identical to 2.6.31, so just a something else changed sanity check)
2.6.25 (lots of pointless changes + sense buffer replacement) 2.6.34 (pid removal and sg chaining)
2.6.23 (accessors change)
2.6.22 (spelling only fix!)
James
Signature · 3 lines
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are watching the assignee of the bug.
[Bug 15950] Hang in dc395x driver
In reply to [email protected]https://bugzilla.kernel.org/show_bug.cgi?id=15950
--- Comment #2 from [email protected] 2010-05-10 04:16:03 --- I downloaded the .1 versions of kernel's 2.6.22 through 2.6.26, and found that 2.6.25 and above had no changes in the dc395x.c file, but that a variety of changes were introduced in 2.6.24 and below. Since the older dc395x.c doesn't drop into my current 2.6.31 kernel, I tried recompiling and running 2.6.24 without success. On Gentoo, the current udev doesn't work with that older kernel, so things like networking and X fail to run properly and I was unable to test the scanner. Suggestions?
Tom
Signature · 3 lines
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are watching the assignee of the bug.
Re: [Bug 15950] Hang in dc395x driver
Quoted 8 lines · On Mon, 10 May 2010, [email protected] wrote:
I downloaded the .1 versions of kernel's 2.6.22 through 2.6.26, and found that 2.6.25 and above had no changes in the
dc395x.cfile, but that a variety of changes were introduced in 2.6.24 and below. Since the olderdc395x.cdoesn't drop into my current 2.6.31 kernel, I tried recompiling and running 2.6.24 without success. On Gentoo, the current udev doesn't work with that older kernel, so things like networking and X fail to run properly and I was unable to test the scanner. Suggestions?
You can use the command-line sane interface scanimage (from the sane-utils package).
Thanks
Guennadi
---
Guennadi Liakhovetski, Ph.D.
Freelance Open-Source Software Developer
http://www.open-technology.de/
[Bug 15950] Hang in dc395x driver
In reply to [email protected]https://bugzilla.kernel.org/show_bug.cgi?id=15950
--- Comment #3 from [email protected] 2010-05-10 05:14:36 ---
Quoted 27 lines · On Sun, May 9, 2010 at 6:07 PM, <[email protected]> wrote:
https://bugzilla.kernel.org/show_bug.cgi?id=15950
--- Comment #1 from Anonymous Emailer <[email protected]> 2010-05-09 22:07:58 ---
Reply-To: [email protected]Quoted 4 lines · On Sun, 2010-05-09 at 21:19 +0000, [email protected] wrote:
I'm not sure when this problem started, but it's somewhere in the 2.6.2x region.
I'd be happy to recompile, test and debug if needed.
Retesting isn't necessary. There have only been nine changes to that driver in 2.6.2x. The versions of interest (as in works/doesn't work) should be
2.6.26 (should be identical to 2.6.31, so just a something else changed sanity check)
2.6.25 (lots of pointless changes + sense buffer replacement) 2.6.34 (pid removal and sg chaining)
2.6.23 (accessors change)
2.6.22 (spelling only fix!)
Quoted 7 lines
James
--
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You reported the bug.
I downloaded the .1 versions of kernel's 2.6.22 through 2.6.26, and found that 2.6.25 and above had no changes in the dc395x.c file, but that a variety of changes were introduced in 2.6.24 and below. Since the older dc395x.c doesn't drop into my current 2.6.31 kernel, I tried recompiling and running 2.6.24 without success. On Gentoo, the current udev doesn't work with that older kernel, so things like networking and X fail to run properly and I was unable to test the scanner. Suggestions?
Tom
Signature · 3 lines
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are watching the assignee of the bug.
[Bug 15950] Hang in dc395x driver
In reply to [email protected]https://bugzilla.kernel.org/show_bug.cgi?id=15950
--- Comment #4 from Guennadi Liakhovetski <[email protected]> 2010-05-10 19:59:25 ---
Quoted 8 lines · On Mon, 10 May 2010, [email protected] wrote:
I downloaded the .1 versions of kernel's 2.6.22 through 2.6.26, and found that 2.6.25 and above had no changes in the
dc395x.cfile, but that a variety of changes were introduced in 2.6.24 and below. Since the olderdc395x.cdoesn't drop into my current 2.6.31 kernel, I tried recompiling and running 2.6.24 without success. On Gentoo, the current udev doesn't work with that older kernel, so things like networking and X fail to run properly and I was unable to test the scanner. Suggestions?
You can use the command-line sane interface scanimage (from the sane-utils package).
Thanks
Guennadi
---
Guennadi Liakhovetski, Ph.D.
Freelance Open-Source Software Developer
http://www.open-technology.de/
Signature · 3 lines
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are watching the assignee of the bug.
[Bug 15950] Hang in dc395x driver
In reply to [email protected]https://bugzilla.kernel.org/show_bug.cgi?id=15950
[email protected] changed:
What |Removed |Added
----------------------------------------------------------------------------
Regression|Yes |No
--- Comment #5 from [email protected] 2010-05-16 03:49:14 --- Guennadi - good call.
I was able to go back to kernel 2.6.22.1 (through 2.6.25/24/23) and all versions exhibit the same behavior. With the command-line scanimage software, I eventually see an error returned and the driver resets after a two minute delay.
I've removed the regression flag from this bug, but still think there's a problem.
Tom
Signature · 3 lines
Configure bugmail: https://bugzilla.kernel.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are watching the assignee of the bug.
[Bug 15950] Hang in dc395x driver
In reply to [email protected]https://bugzilla.kernel.org/show_bug.cgi?id=15950
Alan <[email protected]> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |RESOLVED
CC| |[email protected]
Resolution|--- |OBSOLETESignature · 2 lines
You are receiving this mail because: You are watching the assignee of the bug.
[Bug 15950] Hang in dc395x driver
In reply to [email protected]https://bugzilla.kernel.org/show_bug.cgi?id=15950
[email protected] changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|RESOLVED |REOPENED
Resolution|OBSOLETE |------ Comment #6 from [email protected] ---
Reopening. Not obsolete. Reproduceable in 3.5.3, which is what I'm using at present. (I can upgrade to a newer version if need-be.)
Signature · 2 lines
You are receiving this mail because: You are watching the assignee of the bug.
[Bug 15950] Hang in dc395x driver
In reply to [email protected]https://bugzilla.kernel.org/show_bug.cgi?id=15950
Alan <[email protected]> changed:
What |Removed |Added
----------------------------------------------------------------------------
Kernel Version|2.6.31 |3.5.3
Signature · 2 lines
You are receiving this mail because: You are watching the assignee of the bug.
[Bug 15950] Hang in dc395x driver
In reply to [email protected]https://bugzilla.kernel.org/show_bug.cgi?id=15950
--- Comment #7 from Alan <[email protected]> ---
No need - there is one obscure bug fix in early 2013, but nothing else since 2010. You may be the only person on the planet who still uses one!
Signature · 2 lines
You are receiving this mail because: You are watching the assignee of the bug.
[Bug 15950] Hang in dc395x driver
In reply to [email protected]https://bugzilla.kernel.org/show_bug.cgi?id=15950
--- Comment #8 from [email protected] ---
Could be. Haven't found another inexpensive SCSI adapter.
I say the driver probably needs to get fixed or removed, though. No point in having it in the kernel giving folks false hope that it'll work!
Signature · 2 lines
You are receiving this mail because: You are watching the assignee of the bug.