[lkundrak]

See https://bugzilla.redhat.com/show_bug.cgi?id=1358472

Patch: https://bugzilla.redhat.com/attachment.cgi?id=1182236

Thank you
Lubo

[Josh]

Hmm, that bug report seems not to be public yet ... Can't see it.

Edit: to be more precise - it says I can't see it without logging in.

[LOM]

Logging in doesn't get you any further, the bug seems to be RedHat specific and not something we should care about.

You are not authorized to access bug #1358472.

Most likely the bug has been restricted for internal development processes and we cannot grant access.

[lkundrak]

Sorry, my bad. The bug was automatically filed and I didn't realize it defaults to private. There's nothing secret there; nor too much interesting stuff -- just a traceback and the patch.

[Josh]

I'll try to wrap the whole thing up because it was not immediately clear to me what this is all about:

There was some automatic Fedora package auditing, and it discovered a crash if the command line parameter "-c" was given a file name of a non-existent file.

Is that correct?

Anyway, I will test this and apply the patch accordingly.
However, the priority is not very high as the "-c" parameter is never used in the standard 'automatic' setup. The bug will affect only people who run usb_modeswitch manually on the command line.

[lkundrak]

No, the testing was not automatic really; I was hunting down another bug (see http://www.draisberghof.de/usb_modeswit ... f=2&t=2557) and happened to call the tool with a wrong parameter. The crash was then caught by ABRT which makes it convenient to track the crashes and also saves the core.

Agreed about the severity -- not too important at all.

The another bug hits undefined behavior and indeed makes usb_modeswitch act very unpredictably though; that's a bit more severe. Patched it for Fedora already, but other distros may need a fix too.