[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20181113105530.GE21289@port70.net>
Date: Tue, 13 Nov 2018 11:55:30 +0100
From: Szabolcs Nagy <nsz@...t70.net>
To: musl@...ts.openwall.com
Cc: Sedrubal <dev@...rubal.de>
Subject: Re: Re: [PATCH] Fix many compiler warnings
* Sedrubal <dev@...rubal.de> [2018-11-13 01:15:11 +0100]:
> > Well, hope belongs in the church. Whether or not you introduced bugs is
> > a thing you can test for. There is libc-test.
>
> You're right, I'm sorry. Unfortunately for me libc-test contains 5 errors for
> master, v1.1.20 and for master+my changes. So I'm not sure, if I broke
> something...
>
you don't need to run libc-test.
you have to ensure that before and after the patch the
exact same stripped libc binaries are generated and point
this out in the commit message.
(in principle register allocation, instruction scheduling
etc can change, in practice that should not happen if you
get all parenthesis right and use the same toolchain.
not all header files are used in libc code so you could
write further tests for them or compile code using them
if you want to make sure everything works, but that's
not critical.)
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.
