[Bug 230096] Review Request: iwlwifi-firmware - Microcode for Intel® PRO/Wireless 3945 A/B/G network adaptors

[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 report.

Summary: Review Request: iwlwifi-firmware - Microcode for Intel® PRO/Wireless 3945 A/B/G network adaptors


https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=230096





------- Additional Comments From matthias at rpmforge.net  2007-02-27 08:08 EST -------
(In reply to comment #9)

> > That's invalid for firmware. /lib is correct.
> 
> for %{_lib}, it works for me and no build error.

But %_lib = lib64 on x86_64, so your firmware file will end up in the wrong
place, since it needs to be in /lib/firmware on x86_64, not in /lib64/firmware,
pretty much like the kernel modules needs to be in /lib/modules/...

> > This driver is invalid on PPC.
> 
> So, this : https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=179260
>     should be read.

It's not that is doesn't build or work on ppc. It's just that it doesn't make
sense because it's a firmware for hardware which is only to be found in x86 and
x86_64 hardware (actually, almost certainly only x86_64 hardware, but since you
can install an x86 OS on it, it also makes sense to have available for the x86 OS).

> > For instance "you must use a %{?dist}
> > tag" is incorrect, and in this case, it's _deliberate_ to not use one, since it
> > allows hardlinking the package across multiple releases.
> 
> I understand that. also the fact it's a noarch package.
> Now i wonder if it's accepted in CVS build procedure.

Of course it is. You seem to be confusing a lot of "mandatory" vs. "suggested"
packaging points.

> > Your only valid comment is the one about the %changelog, but please realize that
> > it's pretty much useless to have multiple entries for the same day, especially
> > if they're so minor. The wrong thing would have been not to increment the
release.
> 
> I a little bit agree with you about that (i don't make multiple entries for the
> same day when i build my own packages) but, it's important for review to avoid
> confusions and to follow the work (changes, modification, ...) of the owner of
> the package.

You are right, but one should realize that this is pretty much irrelevant.

> however, I maintains that the Group tag isn't good.

It's a work in progress. Some specific packaging rules for firmwares are being
discussed right now. As of this very instant, it's the correct one. It might
change, in which case I'll change it too.


-- 
Configure bugmail: https://bugzilla.redhat.com/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.