| To: | fche@xxxxxxxxxx (Frank Ch. Eigler) |
|---|---|
| Subject: | Re: pcp updates: build, drop pcp-compat, pmatop |
| From: | Lukas Berk <lberk@xxxxxxxxxx> |
| Date: | Mon, 11 Jan 2016 10:59:55 -0500 |
| Cc: | pcp@xxxxxxxxxxx |
| Delivered-to: | pcp@xxxxxxxxxxx |
| In-reply-to: | <y0my4bzh8ee.fsf@xxxxxxxx> (Frank Ch. Eigler's message of "Fri, 08 Jan 2016 20:02:49 -0500") |
| References: | <87io33pxe6.fsf@xxxxxxxxxx> <y0my4bzh8ee.fsf@xxxxxxxx> |
| User-agent: | Gnus/5.13 (Gnus v5.13) Emacs/24.5 (gnu/linux) |
Hi, fche@xxxxxxxxxx (Frank Ch. Eigler) writes: [...] >> Lukas Berk (3): >> [...] >> Make more Requires: versions explicit > > By the way, this ... > > -Requires: pcp-libs >= @package_version@ > +Requires: pcp-libs = @package_version@ > > ... seems unnecessary. The >= would allow one to retain an older pcp > subrpm while upgrading pcp-libs. Because we don't change the libpcp* > SONAMEs nor ABIs (except by extending, marked by symbol-versioning @@ > suffixes), this configuration would work fine, so ISTM we don't need > to go out of our way to prevent it. It's already this way in the fedora/centOS spec files. This was simply mirroring the change to pcp.spec.in for consistency. The smaller the difference between the rpm's we create with ./Makepkgs and those we actually ship in distro's the better, imo. Cheers, Lukas |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | pcp updates: qa, Nathan Scott |
|---|---|
| Next by Date: | pcp updates: early registration of anon metrics for event records, Ken McDonell |
| Previous by Thread: | Re: [pcp] pcp updates: build, drop pcp-compat, pmatop, Nathan Scott |
| Next by Thread: | pcp updates: merges (lberk,fche,mgoodwin), Nathan Scott |
| Indexes: | [Date] [Thread] [Top] [All Lists] |