Linux patches for 1.0.9 release
Hans de Goede
hdegoede at redhat.com
Tue Mar 27 04:01:32 EDT 2012
Hi,
On 03/26/2012 11:28 PM, Pete Batard wrote:
> I have now applied the first 2 to libusbx master [1], which therefore marks the actual beginning of our fork.
>
Thanks, and thanks in general for all your hard work on making libusbx a success.
> You may notice that I mistakenly added a "Linux:" prefix to the second one, which I only realized after pushing, but, while I'll try to avoid that in the future, I'm not exactly planning to go out of my way fix innocuous mistakes such as these.
Agreed, mistakes happen, and it is better to just live with them (in case
of a commit message) or do correcting patches later then have a weird
testing branch with forced pushes to allow to correct them.
I'm all for RERO including not only at the tarbal level but also at
the level of pushing patches to master.
> I decided to apply these directly to master, rather than go through a branch, because:
>
> 1. At least the first 2 have been previously acked by Peter, and they looked fine to me too.
>
> 2. Those have been posted those for some time, and nobody objected.
>
> 3. Come what may, I don't really see a problem applying things that at least a couple of people have deemed go for mainline, even if someone later makes the case to have them reverted/amended. Better 2 steps forwards and one step back than not making a move and we really shouldn't be afraid to try to push master forward.
Agreed, I'm in favor of pushing simple bugfixes / enhancements
directly to master in general. Only big feature work / re-organization
(or ABI breaking changes) need a branch IMHO.
> If I'm happy with them, I'm planning to push the other 4 in a couple hours as well, so if you have something to say about Hans' patches, now's the time.
>
Great! No objections from me :)
> After that I'll produce an xusb addon patch for review, with the aim of pushing it tomorrow.
>
> If there's anything anybody would like to have before we go 1.0.9-rc4, please try to send it to the list before Thursday.
>
> Finally, am I the only one unhappy with lsusb having (unilaterally?) been renamed to listdevs? If we are worried about name collision, which I don't exactly remember people complaining about, and since a few of us use that sample regularly, I think maybe a shorter name such as lsdevs or even lsu may be better. Anybody would also like to see the litsdevs name either reverted or changed to something else?
I agree that listdevs is not a good name, but lsusb indeed is a name conflict with a
very well known tool on Linux. So lets try to come up with something new,
how about for example: libusb_listdevs ? Or libusb_lsusb The reason I'm putting libusb
in the name is to:
1) avoid further namespace conflicts
2) because I think it makes sense to make clear in the name that this is a libusb
utility program
Regards,
Hans
More information about the libusbx
mailing list