[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CA+PODjrSGXXZTqSsZ2oVbZ5G=aYu6Bb_6OZEmiWuLn0H-5zvjQ@mail.gmail.com>
Date: Thu, 28 Jan 2021 15:45:08 +0300
From: Andrey Melnikov <temnota.am@...il.com>
To: Rich Felker <dalias@...ifal.cx>
Cc: musl@...ts.openwall.com
Subject: Re: Re: move __BYTE_ORDER definition to alltypes.h
ср, 27 янв. 2021 г. в 18:45, Rich Felker <dalias@...ifal.cx>:
>
> On Wed, Jan 27, 2021 at 11:12:41AM +0300, Andrey Melnikov wrote:
> > ср, 27 янв. 2021 г. в 00:40, Rich Felker <dalias@...ifal.cx>:
> > >
> > > On Wed, Jan 27, 2021 at 12:29:05AM +0300, Andrey Melnikov wrote:
> > > > вт, 26 янв. 2021 г. в 20:23, Rich Felker <dalias@...ifal.cx>:
> > > > >
> > > > > On Tue, Jan 26, 2021 at 02:55:09PM +0300, Andrey Melnikov wrote:
> > > > > > Hi.
[...]
>
> Again, you cannot use libc headers for building the kernel or kernel
> modules. musl is not glibc. It uses a BSD-like include order and
> defines all of the headers itself rather than using a mix of gcc
> headers and libc-provided ones.
I'm already able to build kernel, modules and other stuff after
removing the installed by musl stdarg.h.
> Add -nostdinc and -I the gcc header path (like the kernel's own build system does) or you will hit obscure
> problems like this again even if changes are made to the endian macros that make your immediate problem go away.
stdarg.h is a compiler-specific thing, and -nostdinc disables that
patch too. Search, where located RIGHT stdarg.h - not trivial.
For me - patching ONE library with broken design is preferred, rather
than search in all other programs where authors define by itself
"reserved identifier for ANY USE" and patch it.
Thnx. But, if you think what linux kernel have broken design and
misuse defines - you may tell kernel developers about that.
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.
