As with SVE rather than impose ambitious toolchain requirements for SME we manually encode the few instructions which we require in order to perform the work the kernel needs to do. The instructions used to save and restore context are provided as assembler macros while those for entering and leaving streaming mode are done in asm volatile blocks since they are expected to be used from C.
We could do the SMSTART and SMSTOP operations with read/modify/write cycles on SVCR but using the aliases provided for individual field accesses should be slightly faster. These instructions are aliases for MSR but since our minimum toolchain requirements are old enough to mean that we can't use the sX_X_cX_cX_X form and they always use xzr rather than taking a value like write_sysreg_s() wants we just use .inst.
Signed-off-by: Mark Brown broonie@kernel.org --- arch/arm64/include/asm/fpsimd.h | 25 +++++++++++++ arch/arm64/include/asm/fpsimdmacros.h | 54 +++++++++++++++++++++++++++ 2 files changed, 79 insertions(+)
diff --git a/arch/arm64/include/asm/fpsimd.h b/arch/arm64/include/asm/fpsimd.h index cb24385e3632..6e2dc9dcbf49 100644 --- a/arch/arm64/include/asm/fpsimd.h +++ b/arch/arm64/include/asm/fpsimd.h @@ -249,6 +249,31 @@ static inline void sve_setup(void) { }
#endif /* ! CONFIG_ARM64_SVE */
+#ifdef CONFIG_ARM64_SME + +static inline void sme_smstart_sm(void) +{ + asm volatile(__msr_s(SYS_SVCR_SMSTART_SM_EL0, "xzr")); +} + +static inline void sme_smstop_sm(void) +{ + asm volatile(__msr_s(SYS_SVCR_SMSTOP_SM_EL0, "xzr")); +} + +static inline void sme_smstop(void) +{ + asm volatile(__msr_s(SYS_SVCR_SMSTOP_SMZA_EL0, "xzr")); +} + +#else + +static inline void sme_smstart_sm(void) { } +static inline void sme_smstop_sm(void) { } +static inline void sme_smstop(void) { } + +#endif /* ! CONFIG_ARM64_SME */ + /* For use by EFI runtime services calls only */ extern void __efi_fpsimd_begin(void); extern void __efi_fpsimd_end(void); diff --git a/arch/arm64/include/asm/fpsimdmacros.h b/arch/arm64/include/asm/fpsimdmacros.h index 2509d7dde55a..2e9a33155081 100644 --- a/arch/arm64/include/asm/fpsimdmacros.h +++ b/arch/arm64/include/asm/fpsimdmacros.h @@ -93,6 +93,12 @@ .endif .endm
+.macro _sme_check_wv v + .if (\v) < 12 || (\v) > 15 + .error "Bad vector select register \v." + .endif +.endm + /* SVE instruction encodings for non-SVE-capable assemblers */ /* (pre binutils 2.28, all kernel capable clang versions support SVE) */
@@ -174,6 +180,54 @@ | (\np) .endm
+/* SME instruction encodings for non-SME-capable assemblers */ +/* (pre binutils 2.38/LLVM 13) */ + +/* RDSVL X\nx, #\imm */ +.macro _sme_rdsvl nx, imm + _check_general_reg \nx + _check_num (\imm), -0x20, 0x1f + .inst 0x04bf5800 \ + | (\nx) \ + | (((\imm) & 0x3f) << 5) +.endm + +/* + * STR (vector from ZA array): + * STR ZA[\nw, #\offset], [X\nxbase, #\offset, MUL VL] + */ +.macro _sme_str_zav nw, nxbase, offset=0 + _sme_check_wv \nw + _check_general_reg \nxbase + _check_num (\offset), -0x100, 0xff + .inst 0xe1200000 \ + | (((\nw) & 3) << 13) \ + | ((\nxbase) << 5) \ + | ((\offset) & 7) +.endm + +/* + * LDR (vector to ZA array): + * LDR ZA[\nw, #\offset], [X\nxbase, #\offset, MUL VL] + */ +.macro _sme_ldr_zav nw, nxbase, offset=0 + _sme_check_wv \nw + _check_general_reg \nxbase + _check_num (\offset), -0x100, 0xff + .inst 0xe1000000 \ + | (((\nw) & 3) << 13) \ + | ((\nxbase) << 5) \ + | ((\offset) & 7) +.endm + +/* + * Zero the entire ZA array + * ZERO ZA + */ +.macro zero_za + .inst 0xc00800ff +.endm + .macro __for from:req, to:req .if (\from) == (\to) _for__body %\from