This is the mail archive of the
libc-ports@sources.redhat.com
mailing list for the libc-ports project.
Re: [PATCH 0/2] AArch64 glibc port
- From: "Joseph S. Myers" <joseph at codesourcery dot com>
- To: Carlos O'Donell <carlos_odonell at mentor dot com>
- Cc: Marcus Shawcroft <marcus dot shawcroft at linaro dot org>, <libc-alpha at sourceware dot org>, libc-ports <libc-ports at sourceware dot org>
- Date: Wed, 3 Oct 2012 12:04:43 +0000
- Subject: Re: [PATCH 0/2] AArch64 glibc port
- References: <CABXK9ncT6kQQ_YR1kyBMRfg6uAu5LryX33soiAtURN6x-mnkWQ@mail.gmail.com> <5069B013.1090906@mentor.com>
On Mon, 1 Oct 2012, Carlos O'Donell wrote:
> (2) Split up your patches.
>
> One huge patch is impossible to review.
>
> I can see about 25-30 groupings of files that relate and your patch should
> be split up along those lines.
No, absolutely do not split it up artificially. Patch submissions should
be self-contained. The port (the new sysdeps files) should be submitted
as a single patch, to libc-ports only (not libc-alpha), with appropriate
rationale for any design choices it seems appropriate to draw attention
to. One patch is much easier to review, and spot whether anything is
missing in it, than 25-30.
The only things to split up are any libc changes required (generally,
anything outside of ports/sysdeps/.../aarch64/), where each logical change
should be sent in its own self-contained patch submission with its own
self-contained rationale.
(If any bits of the port are *not needed at all* for a functional port -
if they are purely optimized functions - they can be omitted from the
submission and dealt with later afer the port is in. But still don't send
multiple ports patches in the initial submission.)
--
Joseph S. Myers
joseph@codesourcery.com