linux.git/arch/m68k/lib, branch v6.16

m68k: Use kernel's generic muldi3 libgcc function

2024-12-09T12:29:17+00:00

Use the kernels own generic lib/muldi3.c implementation of muldi3 for
68K machines. Some 68K CPUs support 64bit multiplies so move the arch
specific umul_ppmm() macro into a header file that is included by
lib/muldi3.c. That way it can take advantage of the single instruction
when available.

There does not appear to be any existing mechanism for the generic
lib/muldi3.c code to pick up an external arch definition of umul_ppmm().
Create an arch specific libgcc.h that can optionally be included by
the system include/linux/libgcc.h to allow for this.

Somewhat oddly there is also a similar definition of umul_ppmm() in
the non-architecture code in lib/crypto/mpi/longlong.h for a wide range
or machines. Its presence ends up complicating the include setup and
means not being able to use something like compiler.h instead. Actually
there is a few other defines of umul_ppmm() macros spread around in
various architectures, but not directly usable for the m68k case.

Signed-off-by: Greg Ungerer 
Link: https://lore.kernel.org/20231113133209.1367286-1-gerg@linux-m68k.org
Reviewed-by: Geert Uytterhoeven 
Reviewed-by: Arnd Bergmann 
Signed-off-by: Geert Uytterhoeven

m68k: lib: Include for __muldi3()

2023-10-16T12:30:40+00:00

When building with W=1:

    arch/m68k/lib/muldi3.c:82:1: warning: no previous prototype for ‘__muldi3’ [-Wmissing-prototypes]
       82 | __muldi3 (DItype u, DItype v)
	  | ^~~~~~~~

Fix this by including .

Signed-off-by: Geert Uytterhoeven 
Link: https://lore.kernel.org/r/160c1fd14b4798f576d9649334b1d2c77db5cb07.1697008341.git.geert@linux-m68k.org

m68k: Use kernel's generic libgcc functions

2023-10-06T08:03:01+00:00

The arch/m68k/lib versions of the libgcc functions: ashldi3, ashrdi3
and lshrdi3 were taken directly from an older version of gcc.
We can use the kernel's own generic lib versions of these - they are
virtually identical. Switch to those and remove the m68k local ones.

Signed-off-by: Greg Ungerer 
Reviewed-by: Geert Uytterhoeven 
Link: https://lore.kernel.org/r/20230913071350.1939818-1-gerg@kernel.org
Signed-off-by: Geert Uytterhoeven

m68k: Replace #include with #include

2023-08-21T11:27:43+00:00

Commit ddb5cdbafaaa ("kbuild: generate KSYMTAB entries by modpost")
deprecated , which is now a wrapper of .

Replace #include  with #include .

After all the  lines are converted,  and
 will be removed.

Signed-off-by: Masahiro Yamada 
Reviewed-by: Geert Uytterhoeven 
Link: https://lore.kernel.org/r/20230807153654.997091-1-masahiroy@kernel.org
Signed-off-by: Geert Uytterhoeven

net: unexport csum_and_copy_{from,to}_user

2022-04-29T21:37:59+00:00

csum_and_copy_from_user and csum_and_copy_to_user are exported by a few
architectures, but not actually used in modular code.  Drop the exports.

Link: https://lkml.kernel.org/r/20220421070440.1282704-1-hch@lst.de
Signed-off-by: Christoph Hellwig 
Acked-by: Jakub Kicinski 
Acked-by: Geert Uytterhoeven 
Acked-by: Arnd Bergmann 
Acked-by: Michael Ellerman  (powerpc)
Cc: David Miller 
Signed-off-by: Andrew Morton

m68k: muldi3: Use semicolon instead of comma

2021-09-24T11:35:23+00:00

This code works, but it is cleaner to use a semicolon at the end of a
statement instead of a comma.

Extracted from a big anonymous patch by Julia Lawall
.

Signed-off-by: Geert Uytterhoeven 
Link: https://lore.kernel.org/r/20210914073155.3883532-1-geert@linux-m68k.org

m68k: get rid of zeroing destination on error in csum_and_copy_from_user()

2020-08-20T19:45:17+00:00

Signed-off-by: Al Viro

saner calling conventions for csum_and_copy_..._user()

2020-08-20T19:45:15+00:00

All callers of these primitives will
	* discard anything we might've copied in case of error
	* ignore the csum value in case of error
	* always pass 0xffffffff as the initial sum, so the
resulting csum value (in case of success, that is) will never be 0.

That suggest the following calling conventions:
	* don't pass err_ptr - just return 0 on error.
	* don't bother with zeroing destination, etc. in case of error
	* don't pass the initial sum - just use 0xffffffff.

This commit does the minimal conversion in the instances of csum_and_copy_...();
the changes of actual asm code behind them are done later in the series.
Note that this asm code is often shared with csum_partial_copy_nocheck();
the difference is that csum_partial_copy_nocheck() passes 0 for initial
sum while csum_and_copy_..._user() pass 0xffffffff.  Fortunately, we are
free to pass 0xffffffff in all cases and subsequent patches will use that
freedom without any special comments.

A part that could be split off: parisc and uml/i386 claimed to have
csum_and_copy_to_user() instances of their own, but those were identical
to the generic one, so we simply drop them.  Not sure if it's worth
a separate commit...

Signed-off-by: Al Viro

csum_partial_copy_nocheck(): drop the last argument

2020-08-20T19:45:14+00:00

It's always 0.  Note that we theoretically could use ~0U as well -
result will be the same modulo 0xffff, _if_ the damn thing did the
right thing for any value of initial sum; later we'll make use of
that when convenient.

However, unlike csum_and_copy_..._user(), there are instances that
did not work for arbitrary initial sums; c6x is one such.

Signed-off-by: Al Viro

m68k: convert to csum_and_copy_from_user()

2020-05-29T20:11:49+00:00

trivial access_ok() there...

Signed-off-by: Al Viro