[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20170621144725.GH1627@brightrain.aerifal.cx>
Date: Wed, 21 Jun 2017 10:47:25 -0400
From: Rich Felker <dalias@...c.org>
To: musl@...ts.openwall.com
Subject: Re: [PATCH] Handle localtime errors in ctime
On Wed, Jun 21, 2017 at 10:18:40AM +0300, Alexander Monakov wrote:
> On Tue, 20 Jun 2017, Rich Felker wrote:
> > OK, using this one with a minor change to make it more readable (at
> > least to me).
>
> Thanks. Shouldn't the ctime change be reverted?
That's a tricky question. I probably wouldn't have accepted it if I
hadn't misread the spec, but it's also not wrong (it only affects a
case with undefined behavior), and it's moderately unfriendly to
contributors to accept then revert a change that's not wrong but just
not needed. The change could mildly reduce the ability to catch/trap
code with UB, but normally the caller will crash anyway from
dereferencing the null pointer returned. So unless others have strong
feelings on this (or the patch contributor wants it reverted) my
leaning is to just leave it.
Rich
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
