From: Yaakov Selkowitz <yselkowitz@cygwin.com>
To: overseers@sourceware.org
Subject: Re: sourceware and git
Date: Wed, 30 Jul 2014 23:41:00 -0000 [thread overview]
Message-ID: <1406763685.8964.38.camel@YAAKOV04> (raw)
In-Reply-To: <20140730213320.GA4760@ednor.casa.cgf.cx>
On Wed, 2014-07-30 at 17:33 -0400, Christopher Faylor wrote:
> As far as tagging is concerned, I don't see any reason why newlib
> should be tagged when cygwin is tagged. Cygwin shouldn't care about
> newlib tags.
What I don't understand yet is why our situation is different than
binutils-gdb, the two of which also have separate release schedules and
yet share enough code that they chose to use a single repository. Given
that, IIUC:
* newlib and libgloss are developed in tandem;
* newlib and libgloss are meant to be built together;
* newlib and libgloss require the top-level configury in order to build;
then wouldn't the most straightforward (and reliable) method for
newlib/libgloss be to have a single git repo matching the layout of cvs
co newlib?
As for cygwin's usage of newlib, since pulling in newlib would end up
including libgloss and the top level configury, the only difference I
see between a joint newlib/libgloss/cygwin repo and a separate cygwin
repo would be that the former would not require any changes to the
cygwin build system, where the latter would.
Either way, yes, a cygwin checkout (or source tarball) would be somewhat
larger then it is now by virtue of the addition of libgloss to the
checkout. (The top-level configury already skips building libgloss for
cygwin even when present.) If the newlib folks don't mind the same
(although winsup is smaller than libgloss, so it's not quite an equal
tradeoff :-) ), then following the lead of binutils-gdb and using a
single repository makes a lot of sense to me.
Yaakov
next prev parent reply other threads:[~2014-07-30 23:41 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <239280260.18333755.1406652387203.JavaMail.zimbra@redhat.com>
2014-07-29 16:57 ` Jeff Johnston
2014-07-29 18:14 ` Joseph S. Myers
2014-07-30 1:13 ` Christopher Faylor
2014-07-30 7:48 ` Corinna Vinschen
2014-07-30 15:18 ` Christopher Faylor
2014-07-30 15:39 ` Corinna Vinschen
2014-07-30 17:06 ` Joseph S. Myers
2014-07-30 17:27 ` Hans-Peter Nilsson
2014-07-30 17:02 ` Joseph S. Myers
2014-07-30 18:45 ` Christopher Faylor
2014-07-30 20:57 ` Joseph S. Myers
2014-07-30 21:33 ` Christopher Faylor
2014-07-30 21:47 ` Joseph S. Myers
2014-07-30 23:41 ` Yaakov Selkowitz [this message]
2014-07-31 2:22 ` Christopher Faylor
2014-08-01 11:05 ` Corinna Vinschen
2014-08-01 22:54 ` Frank Ch. Eigler
2014-08-02 0:12 ` Yaakov Selkowitz
2014-07-29 18:28 ` Tom Tromey
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1406763685.8964.38.camel@YAAKOV04 \
--to=yselkowitz@cygwin.com \
--cc=overseers@sourceware.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).