i upgraded to the latest hotplug last night, and rebooted this morning. when looking at the boot-process, i noticed hotplug showing me a bunch of new error messages regarding PCI slots and INPUT. no modules for INPUT product 0011/0001/0001/ab41 no modules for INPUT product 0003/046d/c024/9802 no modules or PCI slot 0:0.0 (various times) the screenshot i linked to shows it in its entirety. everything seems to work fine though. Reproducible: Always Steps to Reproduce: 1. emerge hotplug-20040311 and boot with it. Actual Results: i got the error messages as decribed at Details. Expected Results: it should not have produced any error messages. since i do not experience any malfunctions of devices, i am not sure about the severity. i will file it as normal.
Experiencing the same but no REAL problems appeared...
My USB mouse and keyboard failed to get recognized by udev when using this version of hotplug. I also got the same exact error messages. Downgrading to the second latest hotplug scipts fixed it. FYI, I am running a pure udev system.
Also on my system hotplug-20040311 produce few series of messages like: /etc/hotplug/ieee1394.agent: line 71: [: 0x000000: integer expression expected /etc/hotplug/ieee1394.agent: line 76: [: 0x00005e: integer expression expected <repeated 8 or 10 times, then final message> ... no drivers for IEEE1394 product 0x000000/0x00005e/0x000001 this pack of messages repeated 3 or 4 times with different product ID Only one IEEE1394 device in my system is FW controller on Audigy card, no other devices attached.
happens only with pci to me. i found no clues in logs.
Maybe I'm wrong but I think these error messages appear because of the structure of the /etc/hotplug/input.rc file looking for input devices in /proc/bus/input/* There are some input devices like Synaptics Touchpad and Sony Jog Dial that are listed here and I think hotplug does not know which kernel module should be loaded for these devices. Perhaps it's also due to changes in the /proc/ files structure because this does not appear for me with a kernel=<2.6.3. I think this bug should be reported upstreams.
I also have this bug. It breaks completly hotplug of USB Mass storage device. The first plug works, but the nexts fail. Go back to hotplug 20040105 to have this work Using kernel 2.6.4-mm2 without SCSI-emulation /dev populated with udev
modem_run (needed for my Speedtouch USB 330 adsl modem) would go "D" and gives an oops. peace
I get this error with the latest hotplug and firewire drivers but the previous version was fine with kernel 2.6.4.
Yeah, this is a known issue with the upstream hotplug scripts, people are working on it. Oh, I'd be willing to take over maintaining this package if the current maintainer doesn't mind?
As I can understand, scripts have big problem with converting dec<->hex identifiers. I make small patch for ieee1394.agent, which eliminate error messages and make script working (at least, for my Audigy FW controller and MiniDV camera)
Created attachment 27610 [details, diff] patch for ieee1394.agent hotplug script
Known issue in upstream package.
It seems that these problems have been fixed in cvs thanks to Greg Kroah-Hartman. Could it be possible to have these fixes integrated into a new hotplug ebuild revision for gentoo ?
As these were not real bugs, but rather debugging messages that got printed to the screen, I'm reluctant to create a new release just for it. There will be a new hotplug release this week to solve this issue, and other, real bugs.
*** Bug 45431 has been marked as a duplicate of this bug. ***
Fixed in 2004_03_29 release
