Merge branch kvm-arm64/misc-6.16 into kvmarm-master/next

Marc Zyngier · Marc Zyngier · commit 1b85d923ba8c · 2025-05-23T10:59:43.000+01:00
* kvm-arm64/misc-6.16:
  : .
  : Misc changes and improvements for 6.16:
  :
  : - Add a new selftest for the SVE host state being corrupted by a guest
  :
  : - Keep HCR_EL2.xMO set at all times for systems running with the kernel at EL2,
  :   ensuring that the window for interrupts is slightly bigger, and avoiding
  :   a pretty bad erratum on the AmpereOne HW
  :
  : - Replace a couple of open-coded on/off strings with str_on_off()
  :
  : - Get rid of the pKVM memblock sorting, which now appears to be superflous
  :
  : - Drop superflous clearing of ICH_LR_EOI in the LR when nesting
  :
  : - Add workaround for AmpereOne's erratum AC04_CPU_23, which suffers from
  :   a pretty bad case of TLB corruption unless accesses to HCR_EL2 are
  :   heavily synchronised
  :
  : - Add a per-VM, per-ITS debugfs entry to dump the state of the ITS tables
  :   in a human-friendly fashion
  : .
  KVM: arm64: Fix documentation for vgic_its_iter_next()
  KVM: arm64: vgic-its: Add debugfs interface to expose ITS tables
  arm64: errata: Work around AmpereOne's erratum AC04_CPU_23
  KVM: arm64: nv: Remove clearing of ICH_LR&lt;n&gt;.EOI if ICH_LR&lt;n&gt;.HW == 1
  KVM: arm64: Drop sort_memblock_regions()
  KVM: arm64: selftests: Add test for SVE host corruption
  KVM: arm64: Force HCR_EL2.xMO to 1 at all times in VHE mode
  KVM: arm64: Replace ternary flags with str_on_off() helper

Signed-off-by: Marc Zyngier &lt;maz@kernel.org&gt;
diff --git a/Documentation/arch/arm64/silicon-errata.rst b/Documentation/arch/arm64/silicon-errata.rst
@@ -57,6 +57,8 @@ stable kernels.
 +----------------+-----------------+-----------------+-----------------------------+
 | Ampere         | AmpereOne AC04  | AC04_CPU_10     | AMPERE_ERRATUM_AC03_CPU_38  |
 +----------------+-----------------+-----------------+-----------------------------+
+| Ampere         | AmpereOne AC04  | AC04_CPU_23     | AMPERE_ERRATUM_AC04_CPU_23  |
++----------------+-----------------+-----------------+-----------------------------+
 +----------------+-----------------+-----------------+-----------------------------+
 | ARM            | Cortex-A510     | #2457168        | ARM64_ERRATUM_2457168       |
 +----------------+-----------------+-----------------+-----------------------------+
diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
@@ -464,6 +464,23 @@ config AMPERE_ERRATUM_AC03_CPU_38
 
 	  If unsure, say Y.
 
+config AMPERE_ERRATUM_AC04_CPU_23
+        bool "AmpereOne: AC04_CPU_23:  Failure to synchronize writes to HCR_EL2 may corrupt address translations."
+	default y
+	help
+	  This option adds an alternative code sequence to work around Ampere
+	  errata AC04_CPU_23 on AmpereOne.
+
+	  Updates to HCR_EL2 can rarely corrupt simultaneous translations for
+	  data addresses initiated by load/store instructions. Only
+	  instruction initiated translations are vulnerable, not translations
+	  from prefetches for example. A DSB before the store to HCR_EL2 is
+	  sufficient to prevent older instructions from hitting the window
+	  for corruption, and an ISB after is sufficient to prevent younger
+	  instructions from hitting the window for corruption.
+
+	  If unsure, say Y.
+
 config ARM64_WORKAROUND_CLEAN_CACHE
 	bool
 
diff --git a/arch/arm64/include/asm/el2_setup.h b/arch/arm64/include/asm/el2_setup.h
@@ -38,7 +38,7 @@
 
 	orr	x0, x0, #HCR_E2H
 .LnVHE_\@:
-	msr	hcr_el2, x0
+	msr_hcr_el2 x0
 	isb
 .endm
 
diff --git a/arch/arm64/include/asm/hardirq.h b/arch/arm64/include/asm/hardirq.h
@@ -41,7 +41,7 @@ do {									\
 									\
 	___hcr = read_sysreg(hcr_el2);					\
 	if (!(___hcr & HCR_TGE)) {					\
-		write_sysreg(___hcr | HCR_TGE, hcr_el2);		\
+		write_sysreg_hcr(___hcr | HCR_TGE);			\
 		isb();							\
 	}								\
 	/*								\
@@ -82,7 +82,7 @@ do {									\
 	 */								\
 	barrier();							\
 	if (!___ctx->cnt && !(___hcr & HCR_TGE))			\
-		write_sysreg(___hcr, hcr_el2);				\
+		write_sysreg_hcr(___hcr);				\
 } while (0)
 
 static inline void ack_bad_irq(unsigned int irq)
diff --git a/arch/arm64/include/asm/kvm_arm.h b/arch/arm64/include/asm/kvm_arm.h
@@ -103,7 +103,7 @@
 			 HCR_FMO | HCR_IMO | HCR_PTW | HCR_TID3 | HCR_TID1)
 #define HCR_HOST_NVHE_FLAGS (HCR_RW | HCR_API | HCR_APK | HCR_ATA)
 #define HCR_HOST_NVHE_PROTECTED_FLAGS (HCR_HOST_NVHE_FLAGS | HCR_TSC)
-#define HCR_HOST_VHE_FLAGS (HCR_RW | HCR_TGE | HCR_E2H)
+#define HCR_HOST_VHE_FLAGS (HCR_RW | HCR_TGE | HCR_E2H | HCR_AMO | HCR_IMO | HCR_FMO)
 
 #define HCRX_HOST_FLAGS (HCRX_EL2_MSCEn | HCRX_EL2_TCR2En | HCRX_EL2_EnFPM)
 #define MPAMHCR_HOST_FLAGS	0
diff --git a/arch/arm64/include/asm/sysreg.h b/arch/arm64/include/asm/sysreg.h
@@ -1093,6 +1093,15 @@
 	__emit_inst(0xd5000000|(\sreg)|(.L__gpr_num_\rt))
 	.endm
 
+	.macro	msr_hcr_el2, reg
+#if IS_ENABLED(CONFIG_AMPERE_ERRATUM_AC04_CPU_23)
+	dsb	nsh
+	msr	hcr_el2, \reg
+	isb
+#else
+	msr	hcr_el2, \reg
+#endif
+	.endm
 #else
 
 #include <linux/bitfield.h>
@@ -1180,13 +1189,31 @@
 		write_sysreg(__scs_new, sysreg);			\
 } while (0)
 
+#define sysreg_clear_set_hcr(clear, set) do {				\
+	u64 __scs_val = read_sysreg(hcr_el2);				\
+	u64 __scs_new = (__scs_val & ~(u64)(clear)) | (set);		\
+	if (__scs_new != __scs_val)					\
+		write_sysreg_hcr(__scs_new);			\
+} while (0)
+
 #define sysreg_clear_set_s(sysreg, clear, set) do {			\
 	u64 __scs_val = read_sysreg_s(sysreg);				\
 	u64 __scs_new = (__scs_val & ~(u64)(clear)) | (set);		\
 	if (__scs_new != __scs_val)					\
 		write_sysreg_s(__scs_new, sysreg);			\
 } while (0)
 
+#define write_sysreg_hcr(__val) do {					\
+	if (IS_ENABLED(CONFIG_AMPERE_ERRATUM_AC04_CPU_23) &&		\
+	   (!system_capabilities_finalized() ||				\
+	    alternative_has_cap_unlikely(ARM64_WORKAROUND_AMPERE_AC04_CPU_23))) \
+		asm volatile("dsb nsh; msr hcr_el2, %x0; isb"		\
+			     : : "rZ" (__val));				\
+	else								\
+		asm volatile("msr hcr_el2, %x0"				\
+			     : : "rZ" (__val));				\
+} while (0)
+
 #define read_sysreg_par() ({						\
 	u64 par;							\
 	asm(ALTERNATIVE("nop", "dmb sy", ARM64_WORKAROUND_1508412));	\
diff --git a/arch/arm64/kernel/cpu_errata.c b/arch/arm64/kernel/cpu_errata.c
@@ -557,6 +557,13 @@ static const struct midr_range erratum_ac03_cpu_38_list[] = {
 };
 #endif
 
+#ifdef CONFIG_AMPERE_ERRATUM_AC04_CPU_23
+static const struct midr_range erratum_ac04_cpu_23_list[] = {
+	MIDR_ALL_VERSIONS(MIDR_AMPERE1A),
+	{},
+};
+#endif
+
 const struct arm64_cpu_capabilities arm64_errata[] = {
 #ifdef CONFIG_ARM64_WORKAROUND_CLEAN_CACHE
 	{
@@ -875,6 +882,13 @@ const struct arm64_cpu_capabilities arm64_errata[] = {
 		.capability = ARM64_WORKAROUND_AMPERE_AC03_CPU_38,
 		ERRATA_MIDR_RANGE_LIST(erratum_ac03_cpu_38_list),
 	},
+#endif
+#ifdef CONFIG_AMPERE_ERRATUM_AC04_CPU_23
+	{
+		.desc = "AmpereOne erratum AC04_CPU_23",
+		.capability = ARM64_WORKAROUND_AMPERE_AC04_CPU_23,
+		ERRATA_MIDR_RANGE_LIST(erratum_ac04_cpu_23_list),
+	},
 #endif
 	{
 		.desc = "Broken CNTVOFF_EL2",
diff --git a/arch/arm64/kernel/hyp-stub.S b/arch/arm64/kernel/hyp-stub.S
@@ -97,7 +97,7 @@ SYM_CODE_START_LOCAL(__finalise_el2)
 2:
 	// Engage the VHE magic!
 	mov_q	x0, HCR_HOST_VHE_FLAGS
-	msr	hcr_el2, x0
+	msr_hcr_el2 x0
 	isb
 
 	// Use the EL1 allocated stack, per-cpu offset
diff --git a/arch/arm64/kvm/at.c b/arch/arm64/kvm/at.c
@@ -1245,7 +1245,7 @@ static u64 __kvm_at_s1e01_fast(struct kvm_vcpu *vcpu, u32 op, u64 vaddr)
 
 skip_mmu_switch:
 	/* Temporarily switch back to guest context */
-	write_sysreg(vcpu->arch.hcr_el2, hcr_el2);
+	write_sysreg_hcr(vcpu->arch.hcr_el2);
 	isb();
 
 	switch (op) {
@@ -1277,7 +1277,7 @@ static u64 __kvm_at_s1e01_fast(struct kvm_vcpu *vcpu, u32 op, u64 vaddr)
 	if (!fail)
 		par = read_sysreg_par();
 
-	write_sysreg(HCR_HOST_VHE_FLAGS, hcr_el2);
+	write_sysreg_hcr(HCR_HOST_VHE_FLAGS);
 
 	if (!(vcpu_el2_e2h_is_set(vcpu) && vcpu_el2_tge_is_set(vcpu)))
 		__mmu_config_restore(&config);
@@ -1340,7 +1340,7 @@ void __kvm_at_s1e2(struct kvm_vcpu *vcpu, u32 op, u64 vaddr)
 		if (!vcpu_el2_e2h_is_set(vcpu))
 			val |= HCR_NV | HCR_NV1;
 
-		write_sysreg(val, hcr_el2);
+		write_sysreg_hcr(val);
 		isb();
 
 		par = SYS_PAR_EL1_F;
@@ -1365,7 +1365,7 @@ void __kvm_at_s1e2(struct kvm_vcpu *vcpu, u32 op, u64 vaddr)
 		if (!fail)
 			par = read_sysreg_par();
 
-		write_sysreg(hcr, hcr_el2);
+		write_sysreg_hcr(hcr);
 		isb();
 	}
 
diff --git a/arch/arm64/kvm/hyp/include/hyp/switch.h b/arch/arm64/kvm/hyp/include/hyp/switch.h
@@ -345,7 +345,7 @@ static inline void ___activate_traps(struct kvm_vcpu *vcpu, u64 hcr)
 	if (cpus_have_final_cap(ARM64_WORKAROUND_CAVIUM_TX2_219_TVM))
 		hcr |= HCR_TVM;
 
-	write_sysreg(hcr, hcr_el2);
+	write_sysreg_hcr(hcr);
 
 	if (cpus_have_final_cap(ARM64_HAS_RAS_EXTN) && (hcr & HCR_VSE))
 		write_sysreg_s(vcpu->arch.vsesr_el2, SYS_VSESR_EL2);
diff --git a/arch/arm64/kvm/hyp/nvhe/host.S b/arch/arm64/kvm/hyp/nvhe/host.S
@@ -124,7 +124,7 @@ SYM_FUNC_START(__hyp_do_panic)
 	/* Ensure host stage-2 is disabled */
 	mrs	x0, hcr_el2
 	bic	x0, x0, #HCR_VM
-	msr	hcr_el2, x0
+	msr_hcr_el2 x0
 	isb
 	tlbi	vmalls12e1
 	dsb	nsh
diff --git a/arch/arm64/kvm/hyp/nvhe/hyp-init.S b/arch/arm64/kvm/hyp/nvhe/hyp-init.S
@@ -100,7 +100,7 @@ SYM_CODE_START_LOCAL(___kvm_hyp_init)
 	msr	mair_el2, x1
 
 	ldr	x1, [x0, #NVHE_INIT_HCR_EL2]
-	msr	hcr_el2, x1
+	msr_hcr_el2 x1
 
 	mov	x2, #HCR_E2H
 	and	x2, x1, x2
@@ -262,7 +262,7 @@ reset:
 
 alternative_if ARM64_KVM_PROTECTED_MODE
 	mov_q	x5, HCR_HOST_NVHE_FLAGS
-	msr	hcr_el2, x5
+	msr_hcr_el2 x5
 alternative_else_nop_endif
 
 	/* Install stub vectors */
diff --git a/arch/arm64/kvm/hyp/nvhe/mem_protect.c b/arch/arm64/kvm/hyp/nvhe/mem_protect.c
@@ -333,7 +333,7 @@ int __pkvm_prot_finalize(void)
 	 */
 	kvm_flush_dcache_to_poc(params, sizeof(*params));
 
-	write_sysreg(params->hcr_el2, hcr_el2);
+	write_sysreg_hcr(params->hcr_el2);
 	__load_stage2(&host_mmu.arch.mmu, &host_mmu.arch);
 
 	/*
diff --git a/arch/arm64/kvm/hyp/nvhe/switch.c b/arch/arm64/kvm/hyp/nvhe/switch.c
@@ -154,7 +154,7 @@ static void __deactivate_traps(struct kvm_vcpu *vcpu)
 
 	__deactivate_traps_common(vcpu);
 
-	write_sysreg(this_cpu_ptr(&kvm_init_params)->hcr_el2, hcr_el2);
+	write_sysreg_hcr(this_cpu_ptr(&kvm_init_params)->hcr_el2);
 
 	__deactivate_cptr_traps(vcpu);
 	write_sysreg(__kvm_hyp_host_vector, vbar_el2);
diff --git a/arch/arm64/kvm/hyp/vgic-v3-sr.c b/arch/arm64/kvm/hyp/vgic-v3-sr.c
@@ -429,35 +429,41 @@ u64 __vgic_v3_get_gic_config(void)
 	/*
 	 * To check whether we have a MMIO-based (GICv2 compatible)
 	 * CPU interface, we need to disable the system register
-	 * view. To do that safely, we have to prevent any interrupt
-	 * from firing (which would be deadly).
+	 * view.
 	 *
-	 * Note that this only makes sense on VHE, as interrupts are
-	 * already masked for nVHE as part of the exception entry to
-	 * EL2.
-	 */
-	if (has_vhe())
-		flags = local_daif_save();
-
-	/*
 	 * Table 11-2 "Permitted ICC_SRE_ELx.SRE settings" indicates
 	 * that to be able to set ICC_SRE_EL1.SRE to 0, all the
 	 * interrupt overrides must be set. You've got to love this.
+	 *
+	 * As we always run VHE with HCR_xMO set, no extra xMO
+	 * manipulation is required in that case.
+	 *
+	 * To safely disable SRE, we have to prevent any interrupt
+	 * from firing (which would be deadly). This only makes sense
+	 * on VHE, as interrupts are already masked for nVHE as part
+	 * of the exception entry to EL2.
 	 */
-	sysreg_clear_set(hcr_el2, 0, HCR_AMO | HCR_FMO | HCR_IMO);
-	isb();
+	if (has_vhe()) {
+		flags = local_daif_save();
+	} else {
+		sysreg_clear_set_hcr(0, HCR_AMO | HCR_FMO | HCR_IMO);
+		isb();
+	}
+
 	write_gicreg(0, ICC_SRE_EL1);
 	isb();
 
 	val = read_gicreg(ICC_SRE_EL1);
 
 	write_gicreg(sre, ICC_SRE_EL1);
 	isb();
-	sysreg_clear_set(hcr_el2, HCR_AMO | HCR_FMO | HCR_IMO, 0);
-	isb();
 
-	if (has_vhe())
+	if (has_vhe()) {
 		local_daif_restore(flags);
+	} else {
+		sysreg_clear_set_hcr(HCR_AMO | HCR_FMO | HCR_IMO, 0);
+		isb();
+	}
 
 	val  = (val & ICC_SRE_EL1_SRE) ? 0 : (1ULL << 63);
 	val |= read_gicreg(ICH_VTR_EL2);
diff --git a/arch/arm64/kvm/hyp/vhe/switch.c b/arch/arm64/kvm/hyp/vhe/switch.c
@@ -209,7 +209,7 @@ static void __deactivate_traps(struct kvm_vcpu *vcpu)
 
 	___deactivate_traps(vcpu);
 
-	write_sysreg(HCR_HOST_VHE_FLAGS, hcr_el2);
+	write_sysreg_hcr(HCR_HOST_VHE_FLAGS);
 
 	if (has_cntpoff()) {
 		struct timer_map map;
diff --git a/arch/arm64/kvm/hyp/vhe/tlb.c b/arch/arm64/kvm/hyp/vhe/tlb.c
@@ -63,7 +63,7 @@ static void enter_vmid_context(struct kvm_s2_mmu *mmu,
 	__load_stage2(mmu, mmu->arch);
 	val = read_sysreg(hcr_el2);
 	val &= ~HCR_TGE;
-	write_sysreg(val, hcr_el2);
+	write_sysreg_hcr(val);
 	isb();
 }
 
@@ -73,7 +73,7 @@ static void exit_vmid_context(struct tlb_inv_context *cxt)
 	 * We're done with the TLB operation, let's restore the host's
 	 * view of HCR_EL2.
 	 */
-	write_sysreg(HCR_HOST_VHE_FLAGS, hcr_el2);
+	write_sysreg_hcr(HCR_HOST_VHE_FLAGS);
 	isb();
 
 	/* ... and the stage-2 MMU context that we switched away from */
diff --git a/arch/arm64/kvm/pkvm.c b/arch/arm64/kvm/pkvm.c
@@ -11,7 +11,6 @@
 #include <asm/kvm_mmu.h>
 #include <linux/memblock.h>
 #include <linux/mutex.h>
-#include <linux/sort.h>
 
 #include <asm/kvm_pkvm.h>
 
@@ -25,23 +24,6 @@ static unsigned int *hyp_memblock_nr_ptr = &kvm_nvhe_sym(hyp_memblock_nr);
 phys_addr_t hyp_mem_base;
 phys_addr_t hyp_mem_size;
 
-static int cmp_hyp_memblock(const void *p1, const void *p2)
-{
-	const struct memblock_region *r1 = p1;
-	const struct memblock_region *r2 = p2;
-
-	return r1->base < r2->base ? -1 : (r1->base > r2->base);
-}
-
-static void __init sort_memblock_regions(void)
-{
-	sort(hyp_memory,
-	     *hyp_memblock_nr_ptr,
-	     sizeof(struct memblock_region),
-	     cmp_hyp_memblock,
-	     NULL);
-}
-
 static int __init register_memblock_regions(void)
 {
 	struct memblock_region *reg;
@@ -53,7 +35,6 @@ static int __init register_memblock_regions(void)
 		hyp_memory[*hyp_memblock_nr_ptr] = *reg;
 		(*hyp_memblock_nr_ptr)++;
 	}
-	sort_memblock_regions();
 
 	return 0;
 }
diff --git a/arch/arm64/kvm/trace_arm.h b/arch/arm64/kvm/trace_arm.h
@@ -176,7 +176,7 @@ TRACE_EVENT(kvm_set_way_flush,
 	    ),
 
 	    TP_printk("S/W flush at 0x%016lx (cache %s)",
-		      __entry->vcpu_pc, __entry->cache ? "on" : "off")
+		      __entry->vcpu_pc, str_on_off(__entry->cache))
 );
 
 TRACE_EVENT(kvm_toggle_cache,
@@ -196,8 +196,8 @@ TRACE_EVENT(kvm_toggle_cache,
 	    ),
 
 	    TP_printk("VM op at 0x%016lx (cache was %s, now %s)",
-		      __entry->vcpu_pc, __entry->was ? "on" : "off",
-		      __entry->now ? "on" : "off")
+		      __entry->vcpu_pc, str_on_off(__entry->was),
+		      str_on_off(__entry->now))
 );
 
 /*
diff --git a/arch/arm64/kvm/vgic/vgic-debug.c b/arch/arm64/kvm/vgic/vgic-debug.c
diff --git a/arch/arm64/kvm/vgic/vgic-its.c b/arch/arm64/kvm/vgic/vgic-its.c
diff --git a/arch/arm64/kvm/vgic/vgic-v3-nested.c b/arch/arm64/kvm/vgic/vgic-v3-nested.c
diff --git a/arch/arm64/kvm/vgic/vgic.h b/arch/arm64/kvm/vgic/vgic.h
diff --git a/arch/arm64/tools/cpucaps b/arch/arm64/tools/cpucaps
diff --git a/tools/testing/selftests/kvm/Makefile.kvm b/tools/testing/selftests/kvm/Makefile.kvm
diff --git a/tools/testing/selftests/kvm/arm64/host_sve.c b/tools/testing/selftests/kvm/arm64/host_sve.c
