Hans de Goede wrote:
Hi Mike,
Thanks for doing this.
As discussed before I've been working on a simple userspace library for
iscsi so that various tools (such as anaconda) can use that instead of
having to invoke iscsiadm.
The latest version of the (proposed) API documentation for this library
is here:
http://people.atrpms.net/~hdegoede/html/libiscsi_8h.html
The changes to the userspace open-iscsi code are very minimal. All I did
was make 2 idbm.c functions non static, and added prototypes to idbm.h.
Note: I also put in one fix for idbm_rec_update_param, to update the
value string of the matching rec_info. Not doing this does not cause any
issues in the current use of this method, but still it seems better to
me to update this.
I did not see this fix. Could you break it out?
Talking about fixes, I also noticed various uses of strncpy without
ensuring a terminating 0, you may want to do a search replace of strncpy
with strlcpy.
Will do.
The tests under the libiscsi/tests dir are hardcoded to my setup. IOW
they need some more work.
The only currently not implemented feature of libiscsi is the
libiscsi_get_error_string() function. I'm planning on changing usr/log.c
so that dolog becomes a function pointer which can be pointer to either
a function doing the current log_daemon behavior, or to one doing the
current print to stderr behavior. Then the log_daemon variable can be
removed, and libiscsi can supply its own logging function which can
store the last error message.
But before making these changes I first wanted to discuss this with you,
so is making the proposed changes to log.c ok with you?
I am completely clueless when it comes to making libraries. I bet if I
made one it would end up being like libsysfs where everyone hates it.
Just the fwparam stuff ended up in a horrible mess :)
With that said, I think it looks ok from a iscsi point of view.
Sometimes we get a flurry of people that want to work on a lib and
sometimes we get nothing, but a interface should go to lists either way
just in case.
Thanks again for doing this work.
_______________________________________________
Anaconda-devel-list mailing list
Anaconda-devel-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/anaconda-devel-list
