kernel/linux.git/arch/s390/lib/string.c, branch v7.2-rc2

s390/memmove: Optimize backward copy case

2026-06-10T14:55:21+00:00

memmove() copies byte wise for the backward copy case, when the mvc instruction cannot be used. This is quite slow, but can be optimized with the mvcrl instruction, which is available since z15. Some numbers (measured on a shared z16 LPAR) show that the new implementation is nearly always faster, except for the non realistic one and two byte cases: size old new 1 2ns 3ns 2 4ns 5ns 4 5ns 5ns 8 8ns 5ns 16 12ns 6ns 32 8ns 7ns 64 15ns 7ns 128 31ns 9ns 256 64ns 10ns 512 129ns 18ns 1024 250ns 19ns 2048 498ns 38ns Reviewed-by: Juergen Christ Signed-off-by: Heiko Carstens Signed-off-by: Alexander Gordeev

s390/string: Convert memset(16|32|64)() to C

2026-06-10T14:55:21+00:00

Convert memset(16|32|64)() from assembler to C, which should make it easier to read and change, if required. And it allows the compiler to optimize the code, and use different instructions, except for the used inline assemblies. Reviewed-by: Juergen Christ Signed-off-by: Heiko Carstens Signed-off-by: Alexander Gordeev

s390/string: Convert memcpy() to C

2026-06-10T14:55:21+00:00

Convert memcpy() from assembler to C, which should make it easier to read and change, if required. And it allows the compiler to optimize the code, and use different instructions, except for the used inline assemblies. Reviewed-by: Juergen Christ Signed-off-by: Heiko Carstens Signed-off-by: Alexander Gordeev

s390/string: Convert memset() to C

2026-06-10T14:55:21+00:00

Convert memset() from assembler to C, which should make it easier to read and change, if required. And it allows the compiler to optimize the code, and use different instructions, except for the used inline assemblies. Reviewed-by: Juergen Christ Signed-off-by: Heiko Carstens Signed-off-by: Alexander Gordeev

s390/string: Convert memmove() to C

2026-06-10T14:55:20+00:00

Convert memmove() from assembler to C, which should make it easier to read and change, if required. And it allows the compiler to optimize the code, and use different instructions, except for the used inline assemblies. Reviewed-by: Juergen Christ Signed-off-by: Heiko Carstens Signed-off-by: Alexander Gordeev

s390/string: Remove strlcat() implementation

2026-05-30T09:37:25+00:00

strlcat() shouldn't be used anymore (see fortify-string.h), and will be deprecated / removed sooner or later [1]. Therefore remove the s390 implementation of strlcat() in favor of the generic variant. [1] https://lore.kernel.org/all/20260514160719.105084-3-manuelebner@mailbox.org/ Reviewed-by: Alexander Gordeev Signed-off-by: Heiko Carstens Signed-off-by: Alexander Gordeev

s390: Remove superfluous newlines from inline assemblies

2025-09-29T11:52:08+00:00

Remove superfluous newlines from inline assemblies. Compilers use the number of lines of inline assemblies as heuristic for the complexity and inline decisions. Therefore inline assemblies should only contain as many lines as required. A lot of inline assemblies contain a superfluous newline for the last line. Remove such newlines to improve compiler inlining decisions. Suggested-by: Juergen Christ Signed-off-by: Heiko Carstens Reviewed-by: Alexander Gordeev Reviewed-by: Juergen Christ Signed-off-by: Alexander Gordeev

s390/string: Remove strcpy() implementation

2025-04-30T09:41:28+00:00

Remove the optimized strcpy() library implementation. This doesn't make any difference since gcc recognizes all strcpy() usages anyway and uses the builtin variant. There is not a single branch to strcpy() within the generated kernel image, which also seems to be the reason why most other architectures don't have a strcpy() implementation anymore. Reviewed-by: Mikhail Zaslonko Signed-off-by: Heiko Carstens

s390/string: Remove optimized strncpy()

2025-04-17T13:16:10+00:00

There are hardly any strncpy() users left, therefore drop the optimized s390 variant. Acked-by: Alexander Gordeev Signed-off-by: Heiko Carstens

s390/string: Convert to use flag output macros

2024-11-13T13:31:33+00:00

Use flag output macros in inline asm to allow for better code generation if the compiler has support for the flag output constraint. Reviewed-by: Juergen Christ Signed-off-by: Heiko Carstens