lib/xz: use size_t instead of uint32_t in a few places

Reduce the number of uint32_t <-> size_t conversions a little. 
Eliminating such conversions entirely would require changing almost all
uint32_t to size_t, which would look confusing and increase the sizes of
the structs even more.  Going the other way, converting everything to
uint32_t, isn't possible because the input and output buffers use size_t
in struct xz_buf.

Now both arguments to min() have the same type.  This is required to for
compatibility with PowerPC boot code[1] whose min() is strict like
include/linux/minmax.h was before the commit d03eba99f5bf ("minmax: allow
min()/max()/clamp() if the arguments have the same signedness.").

Swap the order of the "state" and "len" in struct lzma_dec to avoid
padding in the middle of the struct when size_t is 64 bits.  The
reordering doesn't change the size of the struct; the padding just appears
at the end instead.

dict_flush() used to truncate size_t to uint32_t when returning.  This
wasn't a bug; the value is always small enough.

Link: https://lore.kernel.org/20260614160521.924710-1-lasse.collin@tukaani.org
Signed-off-by: Lasse Collin <lasse.collin@tukaani.org>
Reported-by: Nathan Chancellor <nathan@kernel.org>
Closes: https://lore.kernel.org/lkml/20260610232323.GA1071374@ax162/ [1]
Reviewed-by: Thorsten Blum <thorsten.blum@linux.dev>
Cc: David Laight <david.laight.linux@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

0 files changed, 0 insertions, 0 deletions