http://bugzilla.novell.com/show_bug.cgi?id=548542 Tamás Németh changed: What |Removed |Added ---------------------------------------------------------------------------- Severity|Normal |Major -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 User nice@titanic.nyme.hu added comment http://bugzilla.novell.com/show_bug.cgi?id=548542#c2 --- Comment #2 from Tamás Németh 2009-10-26 06:12:12 MDT --- Created an attachment (id=324102) --> (http://bugzilla.novell.com/attachment.cgi?id=324102) patch for hso-v1.12 to make it able to be compiled for linux-2.6.31 OK, finally I realized that the problem is in the driver itself. The 2.6.31 kernel contains a very old, unstable version of the hso driver by default. There are newer and much better versions of the driver. The newest one is 112: http://www.pharscape.org/forum/index.php?topic=720.0 However this version cannot be compiled for 2.6.31, but a user called Knaaleman created a patch to fix this problem: http://www.pharscape.org/forum/index.php/topic,784.msg3243.html#msg3243 Now, that patch cannot be applied to the driver because he computed it by comparing his fixed version to another modified version, not the original one. I extracted the usable content from his patch, and that's what I'm attaching here. My version of the patch really makes the new hso driver to be able to compiled for linux-2.6.31, and seems to work perfectly on my 64 bit openSUSE 11.2 RC1 so far. However, this patch seems to disable some "killswitch" feature, which I don't know how relates to an USB modem. It may have some drawbacks, but I have no clue. Can you take a look at it? The driver can be installed by simply decompressing, and patching it, and the issuing 'make install' in its directory. It does not modify the kernel source (thus it's outside the kernel source), but some parts of the kernel source need to be present while compiling. Moreover it places the hso.ko module at a place which is different from the place of the kernel's builtin hso.ko module. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 User nice@titanic.nyme.hu added comment http://bugzilla.novell.com/show_bug.cgi?id=548542#c3 --- Comment #3 from Tamás Németh 2009-10-26 06:13:52 MDT --- I reassigned the bug since it seems to me that the problem is caused by the driver itself, not other parts of the kernel. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 User uwedr@novell.com added comment http://bugzilla.novell.com/show_bug.cgi?id=548542#c5 Uwe Drechsel changed: What |Removed |Added ---------------------------------------------------------------------------- Component|Mobile Devices |Network AssignedTo|mobile-bugs@forge.provo.nov |bnc-team-screening@forge.pr |ell.com |ovo.novell.com --- Comment #5 from Uwe Drechsel 2009-10-29 15:13:18 CET --- Reassgning to Network (or kernel) people. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c6 Jeff Mahoney changed: What |Removed |Added ---------------------------------------------------------------------------- Priority|P5 - None |P3 - Medium CC| |jeffm@novell.com AssignedTo|kernel-maintainers@forge.pr |jeffm@novell.com |ovo.novell.com | --- Comment #6 from Jeff Mahoney 2009-11-20 20:02:10 UTC --- I had to go hunt down where to find that driver. Interestingly at the top of the post, it says: "If your kernel is 2.6.31 or higher do not use this driver! Instead use the driver supplied with the kernel." I'm going to see what's needed to shape this up. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c8 --- Comment #8 from Jeff Mahoney 2009-11-20 22:08:24 UTC --- Created an attachment (id=328776) --> (http://bugzilla.novell.com/attachment.cgi?id=328776) Replacement hso.c Can you replace your hso.c with this one and rebuild the module? This will be the first in a few tests since I don't actually have the hardware. This is version 0.12-Option-jeffm1. The changes are: hso-remove-compat-code hso-stats hso-netdev-prints hso-max-mtu hso-proper-return hso-use-netdev_alloc_skb hso-use-drvdata-accessors hso-stupid-kfree-wrapper hso-use-netdev_ops hso-static-tiocmget_submit_urb remove-stupid-hso-mutexes hso-autopm-enabled hso-fix-newline hso-use-cpu_to_be16 hso-constify-hso_serial_ops hso-use-usb_endpoint_type hso-rfkill-api Most of the changes are just bringing unimportant things in line with mainline. There are still somewhat substantial differences between this and the one in mainline. I'll be clear: My plan here is to get the in-kernel driver working. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c10 --- Comment #10 from Jeff Mahoney 2009-11-21 07:18:03 UTC --- Great. That's really just a baseline for what I wanted to start testing with. It's based off of the hso-1.12 driver, but obviously heavily altered. As for rebuilding the module, I really meant just copying it over the hso.c in the external driver source. You don't even need to do that, though. All you need is a Makefile that has the following in it: -----8<----- obj-m := hso.o ----->8----- . and then do: make -C /lib/modules/$(uname -r)/build M=$PWD and it will build that module, completely outside of the kernel tree. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c12 --- Comment #12 from Jeff Mahoney 2009-11-21 07:22:33 UTC --- Created an attachment (id=328812) --> (http://bugzilla.novell.com/attachment.cgi?id=328812) Test hso.c #3 This version (1.12-Option-jeffm3) adds the following chunk: hso-remove-ioctl This just removes the hso-specific ioctl section. I expect it to work in this version as well. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c14 --- Comment #14 from Jeff Mahoney 2009-11-21 07:47:37 UTC --- Created an attachment (id=328814) --> (http://bugzilla.novell.com/attachment.cgi?id=328814) Test hso.c #5 This version (1.12-Option-jeffm4) adds the following chunk: hso-remove-duplicate-low_latency This removes serial->tty->low_latency = 1 from _hso_serial_set_termios. It's documented in mainline as being a duplicate. I'm not sure if this will work or not. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c16 Jeff Mahoney changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |NEEDINFO Info Provider| |nice@titanic.nyme.hu --- Comment #16 from Jeff Mahoney 2009-11-21 07:52:42 UTC --- Oops, comment #14 should have read version -jeffm5. Ok, so now we have some test cases. I'm trying to accommodate the time difference since I assume you've just woken up or are about to go to bed very late. It's 3am my time, so I'm calling it a night. Please test each of those attachments and let me know which one stops working correctly. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c18 --- Comment #18 from Tamás Németh 2009-11-21 13:43:07 UTC --- I'd like you to know on more thing: I can distinguish three states of the modem when plugged in (I will give them my own names): 1. just plugged in: LEDs are flashing relatively fast, modem-manager detected the device but nothing happened. 2. registered: LEDs are flashing slowly and gradually, the modem already registered somehow to the GSM network. The PIN code is required to enter this state. 3. connected: LEDs are lighting steadily, the modem's network interface has an IP address, etc. Communication is possible. Now, when I suspend the machine, the modem goes back from connected to registered state. As far as I know this is NOT an erroneous behaviour, so you shouldn't fix it I assume ;) The only problem is that modem-manager is unable to cope with suspend/resume. May this is not the fault of the patches applied to hso-1.12 but the driver itself. Or maybe modem-manager is faulty. I don't know. I will send you modem-manager outputs soon. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c20 Tamás Németh changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |NEW Info Provider|nice@titanic.nyme.hu | --- Comment #20 from Tamás Németh 2009-11-21 13:46:27 UTC --- Sorry, the needinfo flag. So to sum up: all versions sent by you seem to work well. Every connection attempt was successful so far, and I experienced no broken connections. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c22 --- Comment #22 from Jeff Mahoney 2009-11-21 16:41:10 UTC --- Ok, hold onto that info. That will be a separate bug report and we'll involve people who know modem-manager more intimately. I'm surprised (happily) that all those test drivers worked though. I had expected #5 or #6 to break something. Now that there is very little context left, I see something that could be as simple as an inverted if statement causing the problem. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c Jeff Mahoney changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |NEEDINFO Info Provider| |nice@titanic.nyme.hu -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c25 --- Comment #25 from Jeff Mahoney 2009-11-21 19:54:11 UTC --- Created an attachment (id=328832) --> (http://bugzilla.novell.com/attachment.cgi?id=328832) Patched in-kernel driver Ok, this is the in-kernel driver, patched to fix the reversed if statement. Can you try this one out? Obviously if your problems are made worse in areas with poor coverage, that would be good data. If it usually works either way when you have good signal, then there's not a lot to work with. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c27 --- Comment #27 from Jeff Mahoney 2009-11-21 20:48:03 UTC --- Created an attachment (id=328833) --> (http://bugzilla.novell.com/attachment.cgi?id=328833) Patched in-kernel driver #2 Ok, the reversed if statement is probably a bug - just not one that you're running into right now. Perhaps the device is getting shut off by the autosuspend code. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c29 --- Comment #29 from Tamás Németh 2009-11-21 21:50:33 UTC --- Sorry, maybe I've forgotten to mention that all the hso-1.12 versions you sent me can be compiled within the kernel tree, that is if you replace the in-kernel hso.c with any of them, you can recompile the kernel and you'll get a new working in-kernel hso.ko in the exact same place, where the original in-kernel module was. Were you aware of this? -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c31 Tamás Németh changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |NEW Info Provider|nice@titanic.nyme.hu | --- Comment #31 from Tamás Németh 2009-11-22 12:22:49 UTC --- (In reply to comment #30)

In response to comment #29, yep. That's expected. The Makefile in drivers/net/usb/ does the same thing as the Makefile chunk that I posted in comment #10.

I'm really at a loss to explain why test7 works and the version from comment #27 doesn't. They're virtually identical. Are you sure you used the right one?

I compared the two sources with diff, and realized that there is only a little difference between them, and yet despite this, the hso.c attached to c#27 does not work! I downloaded, compiled and installed it again very carefully, checked dmesg if I'm loading the correct version, and according to my experiences it doesn't work. I had many unsuccessful connection attempts and broken connections. If I can connect finally, the connection breaks so fast that I was even unable to write any comment here. Totally unusable. On the other side 1.12-jeffm7 is so stable for me that I haven't experienced any problem with it yet. Despite this I'd like to ask you to commit a slightly modified (fit for linux-2.6.31) hso-1.12 into the kernel, rather than a slightly modified (bugfixed) hso-1.2, because hso-1.12 it told (by Option) to have many fixes, and being a much more advanced driver than 1.2. If you finally decide what version to use where will you send it? Into vanilla 2.6.31.7/2.6.32 or Novell's own repositories? -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c33 --- Comment #33 from Tamás Németh 2009-11-22 12:29:23 UTC --- Here: http://www.pharscape.org/article-1241090271.html -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c35 --- Comment #35 from Jeff Mahoney 2009-11-22 15:43:49 UTC --- Yep, and the 1.12 driver hasn't received a number of fixes that went upstream since then. That includes stability fixes and additional device IDs. I've been backporting them to the test hso.c's that I've been posting here. The ones you've said have been working fine. From http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=history;f... 2009-09-11 Marcel Holtmann net: Add DEVTYPE support for Ethernet based devices 2009-09-02 Stephen Hemminger netdev: drivers should make ethtool_ops const 2009-09-01 Stephen Hemminger usbnet: convert to netdev_tx_t 2009-07-06 Patrick McHardy net: use NETDEV_TX_OK instead of 0 in ndo_start_xmit... 2009-06-13 Patrick McHardy net: fix network drivers ndo_start_xmit() return values... 2009-06-08 Paulius Zaleckas hso: convert dev_alloc_skb() to netdev_alloc_skb() 2009-06-03 Johannes Berg rfkill: rewrite 2009-05-01 Greg Kroah-Hartman net: remove driver_data direct access of struct device 2009-04-22 Johannes Berg rfkill: remove deprecated state constants 2009-04-02 Jan Dumon hso: fix for the 'invalid frame length' messages 2009-04-02 Jan Dumon hso: fix for crash when unplugging the device 2009-04-02 Jan Dumon hso: add Vendor/Product ID's for new devices 2009-03-22 Stephen Hemminger usbnet: convert hso driver to net_device_ops 2009-02-27 Paulius Zaleckas hso: remove hso_free_device() 2009-02-27 Paulius Zaleckas hso: remove claiming and releasing of USB interface 2009-02-18 Hannes Eder drivers/net/usb: fix sparse warnings: Should it be... 2009-02-03 Filip Aben hso: add new device id's 2009-02-01 Harvey Harrison net: replace uses of __constant_{endian} 2009-01-15 Denis Joseph Barrow hso serial throttled tty kref fix. 2009-01-15 Denis Joseph Barrow tty: Fix double grabbing of a spinlock 2009-01-15 Alan Cox tty: Fix a kref leak in the HSO driver on re-open 2009-01-13 Denis Joseph Barrow hso: driver fix for big endian machines. 2009-01-02 Denis Joseph Barrow hso maintainers update patch 2009-01-02 Denis Joseph Barrow hso modem detect fix patch against Alan Cox'es tty... 2009-01-02 Denis Joseph Barrow tty: Modem functions for the HSO driver 2009-01-02 Alan Cox tty: Fix the HSO termios handling a bit 2009-01-02 Alan Cox hso: net driver using tty without locking 2008-12-30 Julia Lawall drivers/net/usb: use USB API functions rather than... 2008-12-23 Oliver Neukum net: startup race in hso driver 2008-11-25 David S. Miller Revert "hso: Fix crashes on close." 2008-11-25 David S. Miller Revert "hso: Fix free of mutexes still in use." 2008-11-25 David S. Miller Revert "hso: Add TIOCM ioctl handling." 2008-11-25 Denis Joseph Barrow hso: Hook up ->reset_resume 2008-11-25 Denis Joseph Barrow hso: Add TIOCM ioctl handling. 2008-11-25 Denis Joseph Barrow hso: Fix free of mutexes still in use. 2008-11-25 Denis Joseph Barrow hso: Fix URB submission -EINVAL. 2008-11-25 Denis Joseph Barrow hso: Fix crashes on close. 2008-11-25 Denis Joseph Barrow hso: Add new usb device id's. 2008-11-07 Jonathan McDowell [netdrvr] usb/hso: Cleanup rfkill error handling 2008-11-07 Paulius Zaleckas hso: rfkill type should be WWAN -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c37 --- Comment #37 from Tamás Németh 2009-11-22 20:14:13 UTC --- (In reply to comment #36)

Created an attachment (id=328856) --> (http://bugzilla.novell.com/attachment.cgi?id=328856) [details] Test hso.c #8

This version moves the wakeup and killing of the urb up in hso_stop_serial_device

Comparing this by diff to the one from c#27 doesn't show any semanical difference I can notice. And in fact this version also produces the usual in-kernel driver error. Just to remind you these errors are denoted by messages like "usb 3-2: reset full speed USB device using uhci_hcd and address 13" in dmesg. The bug may be hidden somewhere in the difference between this (or c#27's) version and c#23's version (which works). -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c40 --- Comment #40 from Tamás Németh 2009-11-30 09:41:30 UTC --- Jeff, are you still alive? -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c42 --- Comment #42 from Tamás Németh 2009-12-05 07:58:32 UTC --- (In reply to comment #41)

I'm really having a hard time believing that the c#23 version worked as expected and c#27/33 version doesn't work. I don't doubt that you've been testing. I just wonder if you've loaded the right version for some of your tests.

OK, I tested it again carefully for you. I downloaded hso.c from c#36 again, then compiled and loaded it. I double-checked everything during this process, and the results are the same. I'm getting loads of "reset full speed USB device" messages in dmesg, when using this driver. At my home, where the 3D signal is exceptionally good, the modem is sometimes able to recover from this problem, but now I'm at my working place, and I was even unable to do a single successful connection despite trying it many times. See the dmesg output showing the driver version: milleniumfalcon:~ # dmesg | egrep '(hso)|(reset full)' [ 6194.653345] hso: unloaded [ 6194.653371] usbcore: deregistering interface driver hso [ 6236.878437] hso: drivers/net/usb/hso.c: 1.2-test8 Option Wireless [ 6236.882215] hso0: Disabled Privacy Extensions [ 6236.884788] usbcore: registered new interface driver hso [ 6268.809107] usb 3-2: reset full speed USB device using uhci_hcd and address 15 [ 6311.059107] usb 3-2: reset full speed USB device using uhci_hcd and address 15 [ 6351.059084] usb 3-2: reset full speed USB device using uhci_hcd and address 15 [ 6390.809119] usb 3-2: reset full speed USB device using uhci_hcd and address 15 [ 6428.059108] usb 3-2: reset full speed USB device using uhci_hcd and address 15 [ 6449.809105] usb 3-2: reset full speed USB device using uhci_hcd and address 15 [ 6492.058110] usb 3-2: reset full speed USB device using uhci_hcd and address 15 [ 6549.059062] usb 3-2: reset full speed USB device using uhci_hcd and address 15 [ 6568.059112] usb 3-2: reset full speed USB device using uhci_hcd and address 15 [ 6612.058063] usb 3-2: reset full speed USB device using uhci_hcd and address 15 [ 6648.309116] usb 3-2: reset full speed USB device using uhci_hcd and address 15 -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c44 Jeff Mahoney changed: What |Removed |Added ---------------------------------------------------------------------------- AssignedTo|jeffm@novell.com |gregkh@novell.com --- Comment #44 from Jeff Mahoney 2009-12-14 14:54:09 UTC --- Greg, can you take a look at this one? I'm out of ideas. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c46 --- Comment #46 from Tamás Németh 2010-01-12 10:12:44 UTC --- Anybody there? I'm constantly testing the hso.c from c#23 on both 32 and 64 bit openSUSE 11.2 systems very intensively. I's been working without any problem for months. Please commit that version to the vannilla kernel (or at least to the openSUSE kernel), otherwise I'm the only one who can use that excellent driver version and everyone else has to suffer with digging for a patched driver, since drivers included in officially released kernels don't work at all (AFAIK). -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c48 --- Comment #48 from Tamás Németh 2010-01-12 15:04:30 UTC --- (In reply to comment #47)

I can not do a wholesale replacement of the driver, especially when the code is so different from the upstream kernel version.

And what about including it in the openSUSE kernels only?

If you could contact the upstream driver author, and ask them why their changes are not in the mainline kernel.org tree, that might be interesting to find out, especially as they should have pushed their changes a long time ago.

OK, I just did so. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c50 Greg Kroah-Hartman changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |NEEDINFO Info Provider| |nice@titanic.nyme.hu --- Comment #50 from Greg Kroah-Hartman 2010-01-22 18:15:46 UTC --- any response from upstream? -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c52 --- Comment #52 from Tamás Németh 2010-02-16 14:21:09 UTC --- Today I asked upstream again. BTW, sorry for asking this, but why can't you replace (e.g. with https://bugzilla.novell.com/show_bug.cgi?id=548542#c23 ) (or patch?) the in-kernel hso.c just for openSUSE 11.2. like you did to solve https://bugzilla.novell.com/show_bug.cgi?id=540589 ? -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c54 --- Comment #54 from Tamás Németh 2010-03-23 20:49:59 UTC --- (In reply to comment #50)

any response from upstream?

I saw this today: http://www.pharscape.org/article-1267446554.html Have you already received some update for 2.6.33.x? Can it be backported (or simply placed into) to 2.6.31? May I test it before an eventual backport? -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c56 --- Comment #56 from Tamás Németh 2010-04-29 20:53:58 UTC --- (In reply to comment #55)

Can you try the latest FACTORY kernel to see if this resolves the issue for you now? That is based on 2.6.34-rc.

What is the exact URL of the repository holding the mentioned kernel? Is it http://download.opensuse.org/repositories/Kernel:/openSUSE-11.2/openSUSE_112... as said at https://bugzilla.novell.com/show_bug.cgi?id=590756#c1 ? I'd like to carry out he necessary test for both bugreports. Thank you in advance, and thank you for the kernel in general! (Its quality seems to be improving for me with each version nowadays!) -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c58 --- Comment #58 from Tamás Németh 2010-05-12 07:53:50 UTC --- I only tried 2.6.31.13-14-desktop from http://download.opensuse.org/repositories/Kernel:/openSUSE-11.2/openSUSE_112 so far but that version seems to solve the problem!!! Is it possible? -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.

http://bugzilla.novell.com/show_bug.cgi?id=548542 http://bugzilla.novell.com/show_bug.cgi?id=548542#c60 Greg Kroah-Hartman changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED --- Comment #60 from Greg Kroah-Hartman 2010-05-19 22:14:55 UTC --- Glad that .34 works for you, will close this out now. -- Configure bugmail: http://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.