[Bug 495564] Review Request: libguestfs - Access and modify virtual machine disk images

[bugzilla at redhat.com]

Please do not reply directly to this email. All additional
comments should be made in the comments box of this bug.


https://bugzilla.redhat.com/show_bug.cgi?id=495564





--- Comment #17 from Richard W.M. Jones <rjones at redhat.com>  2009-05-07 16:26:40 EDT ---
(In reply to comment #10)
> Please specify prototyping behavior for Guestfs.xs (see perlxs manual)
> 
> Is it worth addressing the "please specify..." part?  

This is fixed by this commit:
http://git.et.redhat.com/?p=libguestfs.git;a=commitdiff;h=672c4ba257924c3e64836c08fb3b55bc4a6b2aba

(In reply to comment #11)
> In src/guestfs.c, you probably want to use a read wrapper
> so you don't have to worry about EINTR and EAGAIN
> (the code should retry, not fail in those cases).
> It looks like there are a few others that would benefit.

This should fix it:
http://git.et.redhat.com/?p=libguestfs.git;a=commitdiff;h=bb349b05333aa5bf87a3882f15458d8f7341d807

(In reply to comment #12)
> Please use an unsigned type for length-only variables like "len" here:

This change isn't exhaustive by any means, but it fixes some
of these problems:

http://git.et.redhat.com/?p=libguestfs.git;a=commitdiff;h=dd8b152da0e899104fec305159640d08d7d6cdd9

(In reply to comment #13)
> This realloc (from guestfsd.c) leaks upon failure:

http://git.et.redhat.com/?p=libguestfs.git;a=commitdiff;h=427b5f079fd344919ecf568bab2084825aacf606

(In reply to comment #15)
> Since "make check" doesn't pass for me, I haven't yet tried running things
> under valgrind.  Have you?

No I haven't run it under valgrind, but obviously I should!

I've not seen the 'make check' error that we discussed on IRC, and
as you know I've tried to reproduce your setup quite closely.  I've
now built libguestfs on RHEL 5 too, which has quite a few differences
from Fedora 11.  Again, not seen that problem in 'make check' ...

-- 
Configure bugmail: https://bugzilla.redhat.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.