UPSTREAM: arm64/sme: Implement ZA context switching

Allocate space for storing ZA on first access to SME and use that to save
and restore ZA state when context switching. We do this by using the vector
form of the LDR and STR ZA instructions, these do not require streaming
mode and have implementation recommendations that they avoid contention
issues in shared SMCU implementations.

Since ZA is architecturally guaranteed to be zeroed when enabled we do not
need to explicitly zero ZA, either we will be restoring from a saved copy
or trapping on first use of SME so we know that ZA must be disabled.

Signed-off-by: Mark Brown <broonie@kernel.org>
Reviewed-by: Catalin Marinas <catalin.marinas@arm.com>
Link: https://lore.kernel.org/r/20220419112247.711548-16-broonie@kernel.org
Signed-off-by: Catalin Marinas <catalin.marinas@arm.com>
(cherry picked from commit 0033cd9339642f9b7bef23f96aa2e7277ab51cce)
Signed-off-by: Will Deacon <willdeacon@google.com>
Bug: 233587962
Bug: 233588291
Change-Id: I95f189d064fb54f6d7d632e3b6c07d11aef9c6f0

arch/arm64/include/asm/fpsimd.h

@@ -47,7 +47,8 @@ extern void fpsimd_update_current_state(struct user_fpsimd_state const *state);
 
 extern void fpsimd_bind_state_to_cpu(struct user_fpsimd_state *state,
 				     void *sve_state, unsigned int sve_vl,
-				     unsigned int sme_vl, u64 *svcr);
+				     void *za_state, unsigned int sme_vl,
+				     u64 *svcr);
 
 extern void fpsimd_flush_task_state(struct task_struct *target);
 extern void fpsimd_save_and_flush_cpu_state(void);
@@ -90,6 +91,8 @@ extern void sve_flush_live(bool flush_ffr, unsigned long vq_minus_1);
 extern unsigned int sve_get_vl(void);
 extern void sve_set_vq(unsigned long vq_minus_1);
 extern void sme_set_vq(unsigned long vq_minus_1);
+extern void za_save_state(void *state);
+extern void za_load_state(void const *state);
 
 struct arm64_cpu_capabilities;
 extern void sve_kernel_enable(const struct arm64_cpu_capabilities *__unused);

arch/arm64/include/asm/fpsimdmacros.h

@@ -319,3 +319,25 @@
 		ldr		w\nxtmp, [\xpfpsr, #4]
 		msr		fpcr, x\nxtmp
 .endm
+
+.macro sme_save_za nxbase, xvl, nw
+	mov	w\nw, #0
+
+423:
+	_sme_str_zav \nw, \nxbase
+	add	x\nxbase, x\nxbase, \xvl
+	add	x\nw, x\nw, #1
+	cmp	\xvl, x\nw
+	bne	423b
+.endm
+
+.macro sme_load_za nxbase, xvl, nw
+	mov	w\nw, #0
+
+423:
+	_sme_ldr_zav \nw, \nxbase
+	add	x\nxbase, x\nxbase, \xvl
+	add	x\nw, x\nw, #1
+	cmp	\xvl, x\nw
+	bne	423b
+.endm

arch/arm64/include/asm/kvm_host.h

@@ -295,8 +295,11 @@ struct vcpu_reset_state {
 
 struct kvm_vcpu_arch {
 	struct kvm_cpu_context ctxt;
+
+	/* Guest floating point state */
 	void *sve_state;
 	unsigned int sve_max_vl;
+	u64 svcr;
 
 	/* Stage 2 paging state used by the hardware on next switch */
 	struct kvm_s2_mmu *hw_mmu;

arch/arm64/include/asm/processor.h

@@ -157,6 +157,7 @@ struct thread_struct {
 
 	unsigned int		fpsimd_cpu;
 	void			*sve_state;	/* SVE registers, if any */
+	void			*za_state;	/* ZA register, if any */
 	unsigned int		vl[ARM64_VEC_MAX];	/* vector length */
 	unsigned int		vl_onexec[ARM64_VEC_MAX]; /* vl after next exec */
 	unsigned long		fault_address;	/* fault info */

arch/arm64/kernel/entry-fpsimd.S

@@ -99,4 +99,26 @@ SYM_FUNC_START(sme_set_vq)
 	ret
 SYM_FUNC_END(sme_set_vq)
 
+/*
+ * Save the SME state
+ *
+ * x0 - pointer to buffer for state
+ */
+SYM_FUNC_START(za_save_state)
+	_sme_rdsvl	1, 1		// x1 = VL/8
+	sme_save_za 0, x1, 12
+	ret
+SYM_FUNC_END(za_save_state)
+
+/*
+ * Load the SME state
+ *
+ * x0 - pointer to buffer for state
+ */
+SYM_FUNC_START(za_load_state)
+	_sme_rdsvl	1, 1		// x1 = VL/8
+	sme_load_za 0, x1, 12
+	ret
+SYM_FUNC_END(za_load_state)
+
 #endif /* CONFIG_ARM64_SME */

arch/arm64/kernel/fpsimd.c

@@ -121,6 +121,7 @@
 struct fpsimd_last_state_struct {
 	struct user_fpsimd_state *st;
 	void *sve_state;
+	void *za_state;
 	u64 *svcr;
 	unsigned int sve_vl;
 	unsigned int sme_vl;
@@ -387,11 +388,15 @@ static void task_fpsimd_load(void)
 	if (system_supports_sme()) {
 		unsigned long sme_vl = task_get_sme_vl(current);
 
+		/* Ensure VL is set up for restoring data */
 		if (test_thread_flag(TIF_SME))
 			sme_set_vq(sve_vq_from_vl(sme_vl) - 1);
 
 		write_sysreg_s(current->thread.svcr, SYS_SVCR_EL0);
 
+		if (thread_za_enabled(&current->thread))
+			za_load_state(current->thread.za_state);
+
 		if (thread_sm_enabled(&current->thread)) {
 			restore_sve_regs = true;
 			restore_ffr = system_supports_fa64();
@@ -441,11 +446,10 @@ static void fpsimd_save(void)
 		u64 *svcr = last->svcr;
 		*svcr = read_sysreg_s(SYS_SVCR_EL0);
 
-		if (thread_za_enabled(&current->thread)) {
-			/* ZA state managment is not implemented yet */
-			force_signal_inject(SIGKILL, SI_KERNEL, 0, 0);
-			return;
-		}
+		*svcr = read_sysreg_s(SYS_SVCR_EL0);
+
+		if (*svcr & SYS_SVCR_EL0_ZA_MASK)
+			za_save_state(last->za_state);
 
 		/* If we are in streaming mode override regular SVE. */
 		if (*svcr & SYS_SVCR_EL0_SM_MASK) {
@@ -1483,6 +1487,7 @@ static void fpsimd_bind_task_to_cpu(void)
 	WARN_ON(!system_supports_fpsimd());
 	last->st = &current->thread.uw.fpsimd_state;
 	last->sve_state = current->thread.sve_state;
+	last->za_state = current->thread.za_state;
 	last->sve_vl = task_get_sve_vl(current);
 	last->sme_vl = task_get_sme_vl(current);
 	last->svcr = &current->thread.svcr;
@@ -1500,8 +1505,8 @@ static void fpsimd_bind_task_to_cpu(void)
 }
 
 void fpsimd_bind_state_to_cpu(struct user_fpsimd_state *st, void *sve_state,
-			      unsigned int sve_vl, unsigned int sme_vl,
-			      u64 *svcr)
+			      unsigned int sve_vl, void *za_state,
+			      unsigned int sme_vl, u64 *svcr)
 {
 	struct fpsimd_last_state_struct *last =
 		this_cpu_ptr(&fpsimd_last_state);
@@ -1512,6 +1517,7 @@ void fpsimd_bind_state_to_cpu(struct user_fpsimd_state *st, void *sve_state,
 	last->st = st;
 	last->svcr = svcr;
 	last->sve_state = sve_state;
+	last->za_state = za_state;
 	last->sve_vl = sve_vl;
 	last->sme_vl = sme_vl;
 }

arch/arm64/kvm/fpsimd.c

@@ -116,7 +116,7 @@ void kvm_arch_vcpu_ctxsync_fp(struct kvm_vcpu *vcpu)
 		fpsimd_bind_state_to_cpu(&vcpu->arch.ctxt.fp_regs,
 					 vcpu->arch.sve_state,
 					 vcpu->arch.sve_max_vl,
-					 0, NULL);
+					 NULL, 0, &vcpu->arch.svcr);
 
 		clear_thread_flag(TIF_FOREIGN_FPSTATE);
 		update_thread_flag(TIF_SVE, vcpu_has_sve(vcpu));