diff options
Diffstat (limited to 'lib/raid6')
| -rw-r--r-- | lib/raid6/Makefile | 37 | ||||
| -rw-r--r-- | lib/raid6/s390vx.uc | 62 |
2 files changed, 25 insertions, 74 deletions
diff --git a/lib/raid6/Makefile b/lib/raid6/Makefile index 1c5420ff254e..29127dd05d63 100644 --- a/lib/raid6/Makefile +++ b/lib/raid6/Makefile @@ -21,7 +21,7 @@ altivec_flags += -isystem $(shell $(CC) -print-file-name=include) ifdef CONFIG_CC_IS_CLANG # clang ppc port does not yet support -maltivec when -msoft-float is # enabled. A future release of clang will resolve this -# https://bugs.llvm.org/show_bug.cgi?id=31177 +# https://llvm.org/pr31177 CFLAGS_REMOVE_altivec1.o += -msoft-float CFLAGS_REMOVE_altivec2.o += -msoft-float CFLAGS_REMOVE_altivec4.o += -msoft-float @@ -33,27 +33,8 @@ CFLAGS_REMOVE_vpermxor8.o += -msoft-float endif endif -# The GCC option -ffreestanding is required in order to compile code containing -# ARM/NEON intrinsics in a non C99-compliant environment (such as the kernel) -ifeq ($(CONFIG_KERNEL_MODE_NEON),y) -NEON_FLAGS := -ffreestanding -# Enable <arm_neon.h> -NEON_FLAGS += -isystem $(shell $(CC) -print-file-name=include) -ifeq ($(ARCH),arm) -NEON_FLAGS += -march=armv7-a -mfloat-abi=softfp -mfpu=neon -endif -CFLAGS_recov_neon_inner.o += $(NEON_FLAGS) -ifeq ($(ARCH),arm64) -CFLAGS_REMOVE_recov_neon_inner.o += -mgeneral-regs-only -CFLAGS_REMOVE_neon1.o += -mgeneral-regs-only -CFLAGS_REMOVE_neon2.o += -mgeneral-regs-only -CFLAGS_REMOVE_neon4.o += -mgeneral-regs-only -CFLAGS_REMOVE_neon8.o += -mgeneral-regs-only -endif -endif - quiet_cmd_unroll = UNROLL $@ - cmd_unroll = $(AWK) -v N=$* -f $(srctree)/$(src)/unroll.awk < $< > $@ + cmd_unroll = $(AWK) -v N=$* -f $(src)/unroll.awk < $< > $@ targets += int1.c int2.c int4.c int8.c $(obj)/int%.c: $(src)/int.uc $(src)/unroll.awk FORCE @@ -75,10 +56,16 @@ targets += vpermxor1.c vpermxor2.c vpermxor4.c vpermxor8.c $(obj)/vpermxor%.c: $(src)/vpermxor.uc $(src)/unroll.awk FORCE $(call if_changed,unroll) -CFLAGS_neon1.o += $(NEON_FLAGS) -CFLAGS_neon2.o += $(NEON_FLAGS) -CFLAGS_neon4.o += $(NEON_FLAGS) -CFLAGS_neon8.o += $(NEON_FLAGS) +CFLAGS_neon1.o += $(CC_FLAGS_FPU) +CFLAGS_neon2.o += $(CC_FLAGS_FPU) +CFLAGS_neon4.o += $(CC_FLAGS_FPU) +CFLAGS_neon8.o += $(CC_FLAGS_FPU) +CFLAGS_recov_neon_inner.o += $(CC_FLAGS_FPU) +CFLAGS_REMOVE_neon1.o += $(CC_FLAGS_NO_FPU) +CFLAGS_REMOVE_neon2.o += $(CC_FLAGS_NO_FPU) +CFLAGS_REMOVE_neon4.o += $(CC_FLAGS_NO_FPU) +CFLAGS_REMOVE_neon8.o += $(CC_FLAGS_NO_FPU) +CFLAGS_REMOVE_recov_neon_inner.o += $(CC_FLAGS_NO_FPU) targets += neon1.c neon2.c neon4.c neon8.c $(obj)/neon%.c: $(src)/neon.uc $(src)/unroll.awk FORCE $(call if_changed,unroll) diff --git a/lib/raid6/s390vx.uc b/lib/raid6/s390vx.uc index cd0b9e95f499..863e2d320938 100644 --- a/lib/raid6/s390vx.uc +++ b/lib/raid6/s390vx.uc @@ -12,15 +12,14 @@ */ #include <linux/raid/pq.h> -#include <asm/fpu/api.h> -#include <asm/vx-insn.h> +#include <asm/fpu.h> #define NSIZE 16 -static inline void LOAD_CONST(void) +static __always_inline void LOAD_CONST(void) { - asm volatile("VREPIB %v24,7"); - asm volatile("VREPIB %v25,0x1d"); + fpu_vrepib(24, 0x07); + fpu_vrepib(25, 0x1d); } /* @@ -28,10 +27,7 @@ static inline void LOAD_CONST(void) * vector register y left by 1 bit and stores the result in * vector register x. */ -static inline void SHLBYTE(int x, int y) -{ - asm volatile ("VAB %0,%1,%1" : : "i" (x), "i" (y)); -} +#define SHLBYTE(x, y) fpu_vab(x, y, y) /* * For each of the 16 bytes in the vector register y the MASK() @@ -39,49 +35,17 @@ static inline void SHLBYTE(int x, int y) * or 0x00 if the high bit is 0. The result is stored in vector * register x. */ -static inline void MASK(int x, int y) -{ - asm volatile ("VESRAVB %0,%1,24" : : "i" (x), "i" (y)); -} - -static inline void AND(int x, int y, int z) -{ - asm volatile ("VN %0,%1,%2" : : "i" (x), "i" (y), "i" (z)); -} - -static inline void XOR(int x, int y, int z) -{ - asm volatile ("VX %0,%1,%2" : : "i" (x), "i" (y), "i" (z)); -} +#define MASK(x, y) fpu_vesravb(x, y, 24) -static inline void LOAD_DATA(int x, u8 *ptr) -{ - typedef struct { u8 _[16 * $#]; } addrtype; - register addrtype *__ptr asm("1") = (addrtype *) ptr; - - asm volatile ("VLM %2,%3,0,%1" - : : "m" (*__ptr), "a" (__ptr), "i" (x), - "i" (x + $# - 1)); -} - -static inline void STORE_DATA(int x, u8 *ptr) -{ - typedef struct { u8 _[16 * $#]; } addrtype; - register addrtype *__ptr asm("1") = (addrtype *) ptr; - - asm volatile ("VSTM %2,%3,0,1" - : "=m" (*__ptr) : "a" (__ptr), "i" (x), - "i" (x + $# - 1)); -} - -static inline void COPY_VEC(int x, int y) -{ - asm volatile ("VLR %0,%1" : : "i" (x), "i" (y)); -} +#define AND(x, y, z) fpu_vn(x, y, z) +#define XOR(x, y, z) fpu_vx(x, y, z) +#define LOAD_DATA(x, ptr) fpu_vlm(x, x + $# - 1, ptr) +#define STORE_DATA(x, ptr) fpu_vstm(x, x + $# - 1, ptr) +#define COPY_VEC(x, y) fpu_vlr(x, y) static void raid6_s390vx$#_gen_syndrome(int disks, size_t bytes, void **ptrs) { - struct kernel_fpu vxstate; + DECLARE_KERNEL_FPU_ONSTACK32(vxstate); u8 **dptr, *p, *q; int d, z, z0; @@ -114,7 +78,7 @@ static void raid6_s390vx$#_gen_syndrome(int disks, size_t bytes, void **ptrs) static void raid6_s390vx$#_xor_syndrome(int disks, int start, int stop, size_t bytes, void **ptrs) { - struct kernel_fpu vxstate; + DECLARE_KERNEL_FPU_ONSTACK32(vxstate); u8 **dptr, *p, *q; int d, z, z0; |