[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Weird cpio bug

> root@orson:/linux# find test1 d e -print | cpio -pdum test2
> cpio: write error: (system kern) error with unknown subsystem
> Does this mean the Kernel is raising the exception?  This doesn't seem
> as whimsical as the usual Hurd error message...

No, all this means is that this is an unrecognized error number, almost
surely a bogus value that is not an error number at all.  The thing to do
is run gdb on cpio and find what failure causes this message.  If it's a
function like read or write and errno winds up with the bogus value, then
knowing those exact details is helpful.

> Do you think this is a 'cpio' bug in the hurd build, or a
> Hurd bug?  It works fine under Linux (even with the 0-length
> files).

In general it's impossible to be sure about this question without looking
into the particular problem in detail (i.e. with gdb).  And, chances are
the answer might be "both".  The most frequent kind of bug we find in other
programs is bad handling of obscure error conditions that arise due to some
actual problem with the hurd, but that are in a class of failures that the
program in question ought to handle more gracefully than it does.

Reply to: