[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAJgzZoqZij4QgbmOUwz1vvHwaC-C_fRPy9MoTHR-fftEgTERBw@mail.gmail.com>
Date: Wed, 26 Jun 2024 06:38:13 -0400
From: enh <enh@...gle.com>
To: musl@...ts.openwall.com
Subject: Re: roundf() (and round(), and ...)
On Tue, Jun 25, 2024 at 9:49 PM Thorsten Glaser <tg@...bsd.de> wrote:
>
> enh dixit:
>
> >bionic actually uses __builtin_fabs() [and friends] to _implement_
> >these functions, should someone be foolish enough to be using a
>
> That’ll be fun should the compiler decide to insert a call to
> fabs() at the call site instead, which it’d be allowed to do ☺
yeah, even for our limited set of supported architectures we can't do
all the functions for all the architectures. arm32 in particular is
missing a lot of stuff.
> Stories of compilers doing that abound… IIRC it was the compiler
> recognising the implementation pattern of memcpy and replacing
> that by a call to memcpy… in some libc’s that was not careful to
> use -ffreestanding memcpy implementation.
well, unlike musl we don't implement those functions in C. (but, yes,
bionic's had similar fun. stack protector trying to protect stuff
before we've set up the stack protector infrastructure is our usual
one.)
> bye,
> //mirabilos
> --
> 15:41⎜<Lo-lan-do:#fusionforge> Somebody write a testsuite for helloworld :-)
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.
