linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH 1/3] arm64: Define Qualcomm Technologies Falkor v1 CPU
@ 2016-12-07 20:00 Christopher Covington
  2016-12-07 20:00 ` [PATCH 2/3] arm64: Work around Falkor erratum 1003 Christopher Covington
  2016-12-07 20:04 ` [PATCH] arm64: Work around Falkor erratum 1009 Christopher Covington
  0 siblings, 2 replies; 10+ messages in thread
From: Christopher Covington @ 2016-12-07 20:00 UTC (permalink / raw)
  To: Catalin Marinas, Will Deacon, Mark Rutland, Jayachandran C,
	Ganapatrao Kulkarni, Shanker Donthineni, James Morse,
	linux-arm-kernel, linux-kernel
  Cc: Christopher Covington

From: Shanker Donthineni <shankerd@codeaurora.org>

This patch adds the cputype info for Qualcomm Technologies ARMv8 CPU
implementer ID 0x51 and part number for Falkor v1 in cputype.h.

Signed-off-by: Shanker Donthineni <shankerd@codeaurora.org>
Signed-off-by: Christopher Covington <cov@codeaurora.org>
---
 arch/arm64/include/asm/cputype.h | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/arch/arm64/include/asm/cputype.h b/arch/arm64/include/asm/cputype.h
index 26a68dd..ee60561 100644
--- a/arch/arm64/include/asm/cputype.h
+++ b/arch/arm64/include/asm/cputype.h
@@ -71,6 +71,7 @@
 #define ARM_CPU_IMP_APM			0x50
 #define ARM_CPU_IMP_CAVIUM		0x43
 #define ARM_CPU_IMP_BRCM		0x42
+#define ARM_CPU_IMP_QCOM		0x51
 
 #define ARM_CPU_PART_AEM_V8		0xD0F
 #define ARM_CPU_PART_FOUNDATION		0xD00
@@ -84,10 +85,13 @@
 
 #define BRCM_CPU_PART_VULCAN		0x516
 
+#define QCOM_CPU_PART_FALKOR_V1		0x800
+
 #define MIDR_CORTEX_A53 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A53)
 #define MIDR_CORTEX_A57 MIDR_CPU_MODEL(ARM_CPU_IMP_ARM, ARM_CPU_PART_CORTEX_A57)
 #define MIDR_THUNDERX	MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX)
 #define MIDR_THUNDERX_81XX MIDR_CPU_MODEL(ARM_CPU_IMP_CAVIUM, CAVIUM_CPU_PART_THUNDERX_81XX)
+#define MIDR_QCOM_FALKOR_V1 MIDR_CPU_MODEL(ARM_CPU_IMP_QCOM, QCOM_CPU_PART_FALKOR_V1)
 
 #ifndef __ASSEMBLY__
 
-- 
Qualcomm Datacenter Technologies, Inc. as an affiliate of Qualcomm
Technologies, Inc. Qualcomm Technologies, Inc. is a member of the Code Aurora
Forum, a Linux Foundation Collaborative Project.

^ permalink raw reply related	[flat|nested] 10+ messages in thread

* [PATCH 2/3] arm64: Work around Falkor erratum 1003
  2016-12-07 20:00 [PATCH 1/3] arm64: Define Qualcomm Technologies Falkor v1 CPU Christopher Covington
@ 2016-12-07 20:00 ` Christopher Covington
  2016-12-08 10:31   ` Catalin Marinas
  2016-12-08 11:31   ` Mark Rutland
  2016-12-07 20:04 ` [PATCH] arm64: Work around Falkor erratum 1009 Christopher Covington
  1 sibling, 2 replies; 10+ messages in thread
From: Christopher Covington @ 2016-12-07 20:00 UTC (permalink / raw)
  To: Catalin Marinas, Will Deacon, Christopher Covington,
	Shanker Donthineni, Suzuki K Poulose, Andre Przywara,
	Ganapatrao Kulkarni, James Morse, Andrew Pinski, Mark Rutland,
	Jean-Philippe Brucker, Lorenzo Pieralisi, Geoff Levand,
	linux-arm-kernel, linux-kernel

From: Shanker Donthineni <shankerd@codeaurora.org>

On the Qualcomm Datacenter Technologies Falkor v1 CPU, memory accesses may
allocate TLB entries using an incorrect ASID when TTBRx_EL1 is being
updated. Changing the TTBRx_EL1[ASID] and TTBRx_EL1[BADDR] fields
separately using a reserved ASID will ensure that there are no TLB entries
with incorrect ASID after changing the the ASID.

Pseudo code:
  write TTBRx_EL1[ASID] to a reserved value
  ISB
  write TTBRx_EL1[BADDR] to a desired value
  ISB
  write TTBRx_EL1[ASID] to a desired value
  ISB

Signed-off-by: Shanker Donthineni <shankerd@codeaurora.org>
Signed-off-by: Christopher Covington <cov@codeaurora.org>
---
 arch/arm64/Kconfig               | 11 +++++++++++
 arch/arm64/include/asm/cpucaps.h |  3 ++-
 arch/arm64/kernel/cpu_errata.c   |  7 +++++++
 arch/arm64/mm/context.c          | 10 ++++++++++
 arch/arm64/mm/proc.S             | 21 +++++++++++++++++++++
 5 files changed, 51 insertions(+), 1 deletion(-)

diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index 969ef88..1004a3d 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -474,6 +474,17 @@ config CAVIUM_ERRATUM_27456
 
 	  If unsure, say Y.
 
+config QCOM_FALKOR_ERRATUM_E1003
+	bool "Falkor E1003: Incorrect translation due to ASID change"
+	default y
+	help
+	  An incorrect translation TLBI entry may be created while
+	  changing the ASID & translation table address together for
+	  TTBR0_EL1. The workaround for this issue is use a reserved
+	  ASID in cpu_do_switch_mm() before switching to target ASID.
+
+	  If unsure, say Y.
+
 endmenu
 
 
diff --git a/arch/arm64/include/asm/cpucaps.h b/arch/arm64/include/asm/cpucaps.h
index 87b4465..cb6a8c2 100644
--- a/arch/arm64/include/asm/cpucaps.h
+++ b/arch/arm64/include/asm/cpucaps.h
@@ -34,7 +34,8 @@
 #define ARM64_HAS_32BIT_EL0			13
 #define ARM64_HYP_OFFSET_LOW			14
 #define ARM64_MISMATCHED_CACHE_LINE_SIZE	15
+#define ARM64_WORKAROUND_QCOM_FALKOR_E1003	16
 
-#define ARM64_NCAPS				16
+#define ARM64_NCAPS				17
 
 #endif /* __ASM_CPUCAPS_H */
diff --git a/arch/arm64/kernel/cpu_errata.c b/arch/arm64/kernel/cpu_errata.c
index b75e917..3789e2f 100644
--- a/arch/arm64/kernel/cpu_errata.c
+++ b/arch/arm64/kernel/cpu_errata.c
@@ -130,6 +130,13 @@ const struct arm64_cpu_capabilities arm64_errata[] = {
 		.def_scope = SCOPE_LOCAL_CPU,
 		.enable = cpu_enable_trap_ctr_access,
 	},
+#ifdef CONFIG_QCOM_FALKOR_ERRATUM_E1003
+	{
+		.desc = "Qualcomm Falkor erratum E1003",
+		.capability = ARM64_WORKAROUND_QCOM_FALKOR_E1003,
+		MIDR_RANGE(MIDR_QCOM_FALKOR_V1, 0x00, 0x00),
+	},
+#endif
 	{
 	}
 };
diff --git a/arch/arm64/mm/context.c b/arch/arm64/mm/context.c
index efcf1f7..f8d94ff 100644
--- a/arch/arm64/mm/context.c
+++ b/arch/arm64/mm/context.c
@@ -87,6 +87,11 @@ static void flush_context(unsigned int cpu)
 	/* Update the list of reserved ASIDs and the ASID bitmap. */
 	bitmap_clear(asid_map, 0, NUM_USER_ASIDS);
 
+	/* Reserve ASID '1' for Falkor erratum E1003 */
+	if (IS_ENABLED(CONFIG_QCOM_FALKOR_ERRATUM_E1003) &&
+	    cpus_have_cap(ARM64_WORKAROUND_QCOM_FALKOR_E1003))
+		__set_bit(1, asid_map);
+
 	/*
 	 * Ensure the generation bump is observed before we xchg the
 	 * active_asids.
@@ -239,6 +244,11 @@ static int asids_init(void)
 		panic("Failed to allocate bitmap for %lu ASIDs\n",
 		      NUM_USER_ASIDS);
 
+	/* Reserve ASID '1' for Falkor erratum E1003 */
+	if (IS_ENABLED(CONFIG_QCOM_FALKOR_ERRATUM_E1003) &&
+	    cpus_have_cap(ARM64_WORKAROUND_QCOM_FALKOR_E1003))
+		__set_bit(1, asid_map);
+
 	pr_info("ASID allocator initialised with %lu entries\n", NUM_USER_ASIDS);
 	return 0;
 }
diff --git a/arch/arm64/mm/proc.S b/arch/arm64/mm/proc.S
index 352c73b..b4d6508 100644
--- a/arch/arm64/mm/proc.S
+++ b/arch/arm64/mm/proc.S
@@ -134,6 +134,27 @@ ENDPROC(cpu_do_resume)
 ENTRY(cpu_do_switch_mm)
 	mmid	x1, x1				// get mm->context.id
 	bfi	x0, x1, #48, #16		// set the ASID
+#ifdef CONFIG_QCOM_FALKOR_ERRATUM_E1003
+alternative_if_not ARM64_WORKAROUND_QCOM_FALKOR_E1003
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+	nop
+alternative_else
+	mrs     x2, ttbr0_el1                   // get cuurent TTBR0_EL1
+	mov     x3, #1                          // reserved ASID
+	bfi     x2, x3, #48, #16                // set the reserved ASID + old BADDR
+	msr     ttbr0_el1, x2                   // update TTBR0_EL1
+	isb
+	bfi     x2, x0, #0, #48                 // set the desired BADDR + reserved ASID
+	msr     ttbr0_el1, x2                   // update TTBR0_EL1
+	isb
+alternative_endif
+#endif
 	msr	ttbr0_el1, x0			// set TTBR0
 	isb
 alternative_if ARM64_WORKAROUND_CAVIUM_27456
-- 
Qualcomm Datacenter Technologies, Inc. as an affiliate of Qualcomm
Technologies, Inc. Qualcomm Technologies, Inc. is a member of the Code Aurora
Forum, a Linux Foundation Collaborative Project.

^ permalink raw reply related	[flat|nested] 10+ messages in thread

* [PATCH] arm64: Work around Falkor erratum 1009
  2016-12-07 20:00 [PATCH 1/3] arm64: Define Qualcomm Technologies Falkor v1 CPU Christopher Covington
  2016-12-07 20:00 ` [PATCH 2/3] arm64: Work around Falkor erratum 1003 Christopher Covington
@ 2016-12-07 20:04 ` Christopher Covington
  2016-12-08 11:20   ` Will Deacon
  2016-12-08 11:45   ` Mark Rutland
  1 sibling, 2 replies; 10+ messages in thread
From: Christopher Covington @ 2016-12-07 20:04 UTC (permalink / raw)
  To: Catalin Marinas, Will Deacon, Paolo Bonzini,
	Radim Krčmář,
	Christoffer Dall, Marc Zyngier, linux-arm-kernel, linux-kernel,
	kvm, kvmarm
  Cc: Shanker Donthineni, Christopher Covington

From: Shanker Donthineni <shankerd@codeaurora.org>

During a TLB invalidate sequence targeting the inner shareable
domain, Falkor may prematurely complete the DSB before all loads
and stores using the old translation are observed; instruction
fetches are not subject to the conditions of this erratum.

Signed-off-by: Shanker Donthineni <shankerd@codeaurora.org>
Signed-off-by: Christopher Covington <cov@codeaurora.org>
---
 arch/arm64/Kconfig                | 10 +++++++++
 arch/arm64/include/asm/cpucaps.h  |  3 ++-
 arch/arm64/include/asm/tlbflush.h | 43 +++++++++++++++++++++++++++++++++++++++
 arch/arm64/kernel/cpu_errata.c    |  7 +++++++
 arch/arm64/kvm/hyp/tlb.c          | 39 ++++++++++++++++++++++++++++++-----
 5 files changed, 96 insertions(+), 6 deletions(-)

diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index 1004a3d..125440f 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -485,6 +485,16 @@ config QCOM_FALKOR_ERRATUM_E1003
 
 	  If unsure, say Y.
 
+config QCOM_FALKOR_ERRATUM_E1009
+	bool "Falkor E1009: Prematurely complete a DSB after a TLBI"
+	default y
+	help
+	  Falkor CPU may prematurely complete a DSB following a TLBI xxIS
+	  invalidate maintenance operations. Repeat the TLBI operation one
+	  more time to fix the issue.
+
+	  If unsure, say Y.
+
 endmenu
 
 
diff --git a/arch/arm64/include/asm/cpucaps.h b/arch/arm64/include/asm/cpucaps.h
index cb6a8c2..5357d7f 100644
--- a/arch/arm64/include/asm/cpucaps.h
+++ b/arch/arm64/include/asm/cpucaps.h
@@ -35,7 +35,8 @@
 #define ARM64_HYP_OFFSET_LOW			14
 #define ARM64_MISMATCHED_CACHE_LINE_SIZE	15
 #define ARM64_WORKAROUND_QCOM_FALKOR_E1003	16
+#define ARM64_WORKAROUND_QCOM_FALKOR_E1009	17
 
-#define ARM64_NCAPS				17
+#define ARM64_NCAPS				18
 
 #endif /* __ASM_CPUCAPS_H */
diff --git a/arch/arm64/include/asm/tlbflush.h b/arch/arm64/include/asm/tlbflush.h
index deab523..03bafc5 100644
--- a/arch/arm64/include/asm/tlbflush.h
+++ b/arch/arm64/include/asm/tlbflush.h
@@ -23,6 +23,7 @@
 
 #include <linux/sched.h>
 #include <asm/cputype.h>
+#include <asm/alternative.h>
 
 /*
  * Raw TLBI operations.
@@ -94,6 +95,13 @@ static inline void flush_tlb_all(void)
 	dsb(ishst);
 	__tlbi(vmalle1is);
 	dsb(ish);
+	asm volatile(ALTERNATIVE(
+		     "nop \n"
+		     "nop \n",
+		     "tlbi vmalle1is \n"
+		     "dsb ish \n",
+		     ARM64_WORKAROUND_QCOM_FALKOR_E1009)
+		     : :);
 	isb();
 }
 
@@ -104,6 +112,13 @@ static inline void flush_tlb_mm(struct mm_struct *mm)
 	dsb(ishst);
 	__tlbi(aside1is, asid);
 	dsb(ish);
+	asm volatile(ALTERNATIVE(
+		     "nop \n"
+		     "nop \n",
+		     "tlbi aside1is, %0 \n"
+		     "dsb ish \n",
+		     ARM64_WORKAROUND_QCOM_FALKOR_E1009)
+		     : : "r" (asid));
 }
 
 static inline void flush_tlb_page(struct vm_area_struct *vma,
@@ -114,6 +129,13 @@ static inline void flush_tlb_page(struct vm_area_struct *vma,
 	dsb(ishst);
 	__tlbi(vale1is, addr);
 	dsb(ish);
+	asm volatile(ALTERNATIVE(
+		     "nop \n"
+		     "nop \n",
+		     "tlbi vale1is, %0 \n"
+		     "dsb ish \n",
+		     ARM64_WORKAROUND_QCOM_FALKOR_E1009)
+		     : : "r" (addr));
 }
 
 /*
@@ -145,6 +167,13 @@ static inline void __flush_tlb_range(struct vm_area_struct *vma,
 			__tlbi(vae1is, addr);
 	}
 	dsb(ish);
+	asm volatile(ALTERNATIVE(
+		     "nop \n"
+		     "nop \n",
+		     "tlbi vae1is, %0 \n"
+		     "dsb ish \n",
+		     ARM64_WORKAROUND_QCOM_FALKOR_E1009)
+		     : : "r" (end));
 }
 
 static inline void flush_tlb_range(struct vm_area_struct *vma,
@@ -169,6 +198,13 @@ static inline void flush_tlb_kernel_range(unsigned long start, unsigned long end
 	for (addr = start; addr < end; addr += 1 << (PAGE_SHIFT - 12))
 		__tlbi(vaae1is, addr);
 	dsb(ish);
+	asm volatile(ALTERNATIVE(
+		     "nop \n"
+		     "nop \n",
+		     "tlbi vaae1is, %0 \n"
+		     "dsb ish \n",
+		     ARM64_WORKAROUND_QCOM_FALKOR_E1009)
+		     : : "r" (end));
 	isb();
 }
 
@@ -183,6 +219,13 @@ static inline void __flush_tlb_pgtable(struct mm_struct *mm,
 
 	__tlbi(vae1is, addr);
 	dsb(ish);
+	asm volatile(ALTERNATIVE(
+		     "nop \n"
+		     "nop \n",
+		     "tlbi vae1is, %0 \n"
+		     "dsb ish \n",
+		     ARM64_WORKAROUND_QCOM_FALKOR_E1009)
+		     : : "r" (addr));
 }
 
 #endif
diff --git a/arch/arm64/kernel/cpu_errata.c b/arch/arm64/kernel/cpu_errata.c
index 3789e2f..8013579 100644
--- a/arch/arm64/kernel/cpu_errata.c
+++ b/arch/arm64/kernel/cpu_errata.c
@@ -137,6 +137,13 @@ const struct arm64_cpu_capabilities arm64_errata[] = {
 		MIDR_RANGE(MIDR_QCOM_FALKOR_V1, 0x00, 0x00),
 	},
 #endif
+#ifdef CONFIG_QCOM_FALKOR_ERRATUM_E1009
+	{
+		.desc = "Qualcomm Falkor erratum E1009",
+		.capability = ARM64_WORKAROUND_QCOM_FALKOR_E1009,
+		MIDR_RANGE(MIDR_QCOM_FALKOR_V1, 0x00, 0x00),
+	},
+#endif
 	{
 	}
 };
diff --git a/arch/arm64/kvm/hyp/tlb.c b/arch/arm64/kvm/hyp/tlb.c
index 88e2f2b..dfd3a77 100644
--- a/arch/arm64/kvm/hyp/tlb.c
+++ b/arch/arm64/kvm/hyp/tlb.c
@@ -16,6 +16,7 @@
  */
 
 #include <asm/kvm_hyp.h>
+#include <asm/alternative.h>
 
 void __hyp_text __kvm_tlb_flush_vmid_ipa(struct kvm *kvm, phys_addr_t ipa)
 {
@@ -32,7 +33,14 @@ void __hyp_text __kvm_tlb_flush_vmid_ipa(struct kvm *kvm, phys_addr_t ipa)
 	 * whole of Stage-1. Weep...
 	 */
 	ipa >>= 12;
-	asm volatile("tlbi ipas2e1is, %0" : : "r" (ipa));
+	asm volatile("tlbi ipas2e1is, %0 \n"
+		     ALTERNATIVE(
+		     "nop \n"
+		     "nop \n",
+		     "dsb ish \n"
+		     "tlbi ipas2e1is, %0 \n",
+		     ARM64_WORKAROUND_QCOM_FALKOR_E1009)
+		     : : "r" (ipa));
 
 	/*
 	 * We have to ensure completion of the invalidation at Stage-2,
@@ -41,7 +49,14 @@ void __hyp_text __kvm_tlb_flush_vmid_ipa(struct kvm *kvm, phys_addr_t ipa)
 	 * the Stage-1 invalidation happened first.
 	 */
 	dsb(ish);
-	asm volatile("tlbi vmalle1is" : : );
+	asm volatile("tlbi vmalle1is \n"
+		     ALTERNATIVE(
+		     "nop \n"
+		     "nop \n",
+		     "dsb ish \n"
+		     "tlbi vmalle1is \n",
+		     ARM64_WORKAROUND_QCOM_FALKOR_E1009)
+		     : : );
 	dsb(ish);
 	isb();
 
@@ -57,7 +72,14 @@ void __hyp_text __kvm_tlb_flush_vmid(struct kvm *kvm)
 	write_sysreg(kvm->arch.vttbr, vttbr_el2);
 	isb();
 
-	asm volatile("tlbi vmalls12e1is" : : );
+	asm volatile("tlbi vmalls12e1is \n"
+		     ALTERNATIVE(
+		     "nop \n"
+		     "nop \n",
+		     "dsb ish \n"
+		     "tlbi vmalls12e1is \n",
+		     ARM64_WORKAROUND_QCOM_FALKOR_E1009)
+		     : : );
 	dsb(ish);
 	isb();
 
@@ -82,7 +104,14 @@ void __hyp_text __kvm_tlb_flush_local_vmid(struct kvm_vcpu *vcpu)
 void __hyp_text __kvm_flush_vm_context(void)
 {
 	dsb(ishst);
-	asm volatile("tlbi alle1is	\n"
-		     "ic ialluis	  ": : );
+	asm volatile("tlbi alle1is \n"
+		     ALTERNATIVE(
+		     "nop \n"
+		     "nop \n",
+		     "dsb ish \n"
+		     "tlbi alle1is \n",
+		     ARM64_WORKAROUND_QCOM_FALKOR_E1009)
+		     "ic ialluis \n"
+		     : : );
 	dsb(ish);
 }
-- 
Qualcomm Datacenter Technologies, Inc. as an affiliate of Qualcomm
Technologies, Inc. Qualcomm Technologies, Inc. is a member of the Code Aurora
Forum, a Linux Foundation Collaborative Project.

^ permalink raw reply related	[flat|nested] 10+ messages in thread

* Re: [PATCH 2/3] arm64: Work around Falkor erratum 1003
  2016-12-07 20:00 ` [PATCH 2/3] arm64: Work around Falkor erratum 1003 Christopher Covington
@ 2016-12-08 10:31   ` Catalin Marinas
  2016-12-19 21:27     ` Christopher Covington
  2016-12-08 11:31   ` Mark Rutland
  1 sibling, 1 reply; 10+ messages in thread
From: Catalin Marinas @ 2016-12-08 10:31 UTC (permalink / raw)
  To: Christopher Covington
  Cc: Will Deacon, Shanker Donthineni, Suzuki K Poulose,
	Andre Przywara, Ganapatrao Kulkarni, James Morse, Andrew Pinski,
	Mark Rutland, Jean-Philippe Brucker, Lorenzo Pieralisi,
	Geoff Levand, linux-arm-kernel, linux-kernel

On Wed, Dec 07, 2016 at 03:00:26PM -0500, Christopher Covington wrote:
> From: Shanker Donthineni <shankerd@codeaurora.org>
> 
> On the Qualcomm Datacenter Technologies Falkor v1 CPU, memory accesses may
> allocate TLB entries using an incorrect ASID when TTBRx_EL1 is being
> updated. Changing the TTBRx_EL1[ASID] and TTBRx_EL1[BADDR] fields
> separately using a reserved ASID will ensure that there are no TLB entries
> with incorrect ASID after changing the the ASID.
> 
> Pseudo code:
>   write TTBRx_EL1[ASID] to a reserved value
>   ISB
>   write TTBRx_EL1[BADDR] to a desired value
>   ISB
>   write TTBRx_EL1[ASID] to a desired value
>   ISB

While the new ASID probably won't have incorrect TLB entries, the
reserved ASID will have random entries from all over the place. That's
because in step 1 you change the ASID to the reserved one while leaving
the old BADDR in place. There is a brief time before changing the ASID
when speculative page table walks will populate the TLB with entries
tagged with the reserved ASID. Such entries are never removed during TLB
shoot-down for the real ASID, so, depending on how this CPU implements
the walk cache, you could end up with intermediate level entries still
active and pointing to freed/reused pages. It will eventually hit an
entry that looks global with weird consequences.

We've been bitten by this in the past on arm32: 52af9c6cd863 ("ARM:
6943/1: mm: use TTBR1 instead of reserved context ID").

-- 
Catalin

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH] arm64: Work around Falkor erratum 1009
  2016-12-07 20:04 ` [PATCH] arm64: Work around Falkor erratum 1009 Christopher Covington
@ 2016-12-08 11:20   ` Will Deacon
  2016-12-08 11:35     ` Marc Zyngier
  2016-12-08 11:45   ` Mark Rutland
  1 sibling, 1 reply; 10+ messages in thread
From: Will Deacon @ 2016-12-08 11:20 UTC (permalink / raw)
  To: Christopher Covington
  Cc: Catalin Marinas, Paolo Bonzini, Radim Krčmář,
	Christoffer Dall, Marc Zyngier, linux-arm-kernel, linux-kernel,
	kvm, kvmarm, Shanker Donthineni

On Wed, Dec 07, 2016 at 03:04:31PM -0500, Christopher Covington wrote:
> From: Shanker Donthineni <shankerd@codeaurora.org>
> 
> During a TLB invalidate sequence targeting the inner shareable
> domain, Falkor may prematurely complete the DSB before all loads
> and stores using the old translation are observed; instruction
> fetches are not subject to the conditions of this erratum.
> 
> Signed-off-by: Shanker Donthineni <shankerd@codeaurora.org>
> Signed-off-by: Christopher Covington <cov@codeaurora.org>
> ---
>  arch/arm64/Kconfig                | 10 +++++++++
>  arch/arm64/include/asm/cpucaps.h  |  3 ++-
>  arch/arm64/include/asm/tlbflush.h | 43 +++++++++++++++++++++++++++++++++++++++
>  arch/arm64/kernel/cpu_errata.c    |  7 +++++++
>  arch/arm64/kvm/hyp/tlb.c          | 39 ++++++++++++++++++++++++++++++-----
>  5 files changed, 96 insertions(+), 6 deletions(-)
> 
> diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
> index 1004a3d..125440f 100644
> --- a/arch/arm64/Kconfig
> +++ b/arch/arm64/Kconfig
> @@ -485,6 +485,16 @@ config QCOM_FALKOR_ERRATUM_E1003
>  
>  	  If unsure, say Y.
>  
> +config QCOM_FALKOR_ERRATUM_E1009
> +	bool "Falkor E1009: Prematurely complete a DSB after a TLBI"
> +	default y
> +	help
> +	  Falkor CPU may prematurely complete a DSB following a TLBI xxIS
> +	  invalidate maintenance operations. Repeat the TLBI operation one
> +	  more time to fix the issue.
> +
> +	  If unsure, say Y.

Call me perverse, but I like this workaround. People often tend to screw
up TLBI and DVM sync, but the IPI-based workaround is horribly invasive
and fragile. Simply repeating the operation tends to be enough to make
the chance of failure small enough to be acceptable.

> diff --git a/arch/arm64/include/asm/cpucaps.h b/arch/arm64/include/asm/cpucaps.h
> index cb6a8c2..5357d7f 100644
> --- a/arch/arm64/include/asm/cpucaps.h
> +++ b/arch/arm64/include/asm/cpucaps.h
> @@ -35,7 +35,8 @@
>  #define ARM64_HYP_OFFSET_LOW			14
>  #define ARM64_MISMATCHED_CACHE_LINE_SIZE	15
>  #define ARM64_WORKAROUND_QCOM_FALKOR_E1003	16
> +#define ARM64_WORKAROUND_QCOM_FALKOR_E1009	17

Could you rename this to something like ARM64_WORKAROUND_REPEAT_TLBI, so
that it could potentially be used by others?

>  
> -#define ARM64_NCAPS				17
> +#define ARM64_NCAPS				18
>  
>  #endif /* __ASM_CPUCAPS_H */
> diff --git a/arch/arm64/include/asm/tlbflush.h b/arch/arm64/include/asm/tlbflush.h
> index deab523..03bafc5 100644
> --- a/arch/arm64/include/asm/tlbflush.h
> +++ b/arch/arm64/include/asm/tlbflush.h
> @@ -23,6 +23,7 @@
>  
>  #include <linux/sched.h>
>  #include <asm/cputype.h>
> +#include <asm/alternative.h>
>  
>  /*
>   * Raw TLBI operations.
> @@ -94,6 +95,13 @@ static inline void flush_tlb_all(void)
>  	dsb(ishst);
>  	__tlbi(vmalle1is);
>  	dsb(ish);
> +	asm volatile(ALTERNATIVE(
> +		     "nop \n"
> +		     "nop \n",
> +		     "tlbi vmalle1is \n"
> +		     "dsb ish \n",
> +		     ARM64_WORKAROUND_QCOM_FALKOR_E1009)
> +		     : :);

I'd much rather this was part of the __tlbi macro, which would hopefully
restrict this to one place in the code.

Will

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH 2/3] arm64: Work around Falkor erratum 1003
  2016-12-07 20:00 ` [PATCH 2/3] arm64: Work around Falkor erratum 1003 Christopher Covington
  2016-12-08 10:31   ` Catalin Marinas
@ 2016-12-08 11:31   ` Mark Rutland
  1 sibling, 0 replies; 10+ messages in thread
From: Mark Rutland @ 2016-12-08 11:31 UTC (permalink / raw)
  To: Christopher Covington
  Cc: Catalin Marinas, Will Deacon, Shanker Donthineni,
	Suzuki K Poulose, Andre Przywara, Ganapatrao Kulkarni,
	James Morse, Andrew Pinski, Jean-Philippe Brucker,
	Lorenzo Pieralisi, Geoff Levand, linux-arm-kernel, linux-kernel

On Wed, Dec 07, 2016 at 03:00:26PM -0500, Christopher Covington wrote:
> From: Shanker Donthineni <shankerd@codeaurora.org>
> 
> On the Qualcomm Datacenter Technologies Falkor v1 CPU, memory accesses may
> allocate TLB entries using an incorrect ASID when TTBRx_EL1 is being
> updated. Changing the TTBRx_EL1[ASID] and TTBRx_EL1[BADDR] fields
> separately using a reserved ASID will ensure that there are no TLB entries
> with incorrect ASID after changing the the ASID.
> 
> Pseudo code:
>   write TTBRx_EL1[ASID] to a reserved value
>   ISB
>   write TTBRx_EL1[BADDR] to a desired value
>   ISB
>   write TTBRx_EL1[ASID] to a desired value
>   ISB
> 
> Signed-off-by: Shanker Donthineni <shankerd@codeaurora.org>
> Signed-off-by: Christopher Covington <cov@codeaurora.org>
> ---
>  arch/arm64/Kconfig               | 11 +++++++++++
>  arch/arm64/include/asm/cpucaps.h |  3 ++-
>  arch/arm64/kernel/cpu_errata.c   |  7 +++++++
>  arch/arm64/mm/context.c          | 10 ++++++++++
>  arch/arm64/mm/proc.S             | 21 +++++++++++++++++++++
>  5 files changed, 51 insertions(+), 1 deletion(-)

This needs an update to Documentation/arm64/silicon-errata.txt.

> diff --git a/arch/arm64/mm/context.c b/arch/arm64/mm/context.c
> index efcf1f7..f8d94ff 100644
> --- a/arch/arm64/mm/context.c
> +++ b/arch/arm64/mm/context.c
> @@ -87,6 +87,11 @@ static void flush_context(unsigned int cpu)
>  	/* Update the list of reserved ASIDs and the ASID bitmap. */
>  	bitmap_clear(asid_map, 0, NUM_USER_ASIDS);
>  
> +	/* Reserve ASID '1' for Falkor erratum E1003 */
> +	if (IS_ENABLED(CONFIG_QCOM_FALKOR_ERRATUM_E1003) &&
> +	    cpus_have_cap(ARM64_WORKAROUND_QCOM_FALKOR_E1003))
> +		__set_bit(1, asid_map);
> +
>  	/*
>  	 * Ensure the generation bump is observed before we xchg the
>  	 * active_asids.
> @@ -239,6 +244,11 @@ static int asids_init(void)
>  		panic("Failed to allocate bitmap for %lu ASIDs\n",
>  		      NUM_USER_ASIDS);
>  
> +	/* Reserve ASID '1' for Falkor erratum E1003 */
> +	if (IS_ENABLED(CONFIG_QCOM_FALKOR_ERRATUM_E1003) &&
> +	    cpus_have_cap(ARM64_WORKAROUND_QCOM_FALKOR_E1003))
> +		__set_bit(1, asid_map);
> +
>  	pr_info("ASID allocator initialised with %lu entries\n", NUM_USER_ASIDS);
>  	return 0;
>  }
> diff --git a/arch/arm64/mm/proc.S b/arch/arm64/mm/proc.S
> index 352c73b..b4d6508 100644
> --- a/arch/arm64/mm/proc.S
> +++ b/arch/arm64/mm/proc.S
> @@ -134,6 +134,27 @@ ENDPROC(cpu_do_resume)
>  ENTRY(cpu_do_switch_mm)
>  	mmid	x1, x1				// get mm->context.id
>  	bfi	x0, x1, #48, #16		// set the ASID
> +#ifdef CONFIG_QCOM_FALKOR_ERRATUM_E1003
> +alternative_if_not ARM64_WORKAROUND_QCOM_FALKOR_E1003
> +	nop
> +	nop
> +	nop
> +	nop
> +	nop
> +	nop
> +	nop
> +	nop
> +alternative_else
> +	mrs     x2, ttbr0_el1                   // get cuurent TTBR0_EL1
> +	mov     x3, #1                          // reserved ASID

It might be best to define a FALCOR_E1003_RESERVED_ASID constant
somewhere, rather than using 1 directly here and in the ASID allocator.

> +	bfi     x2, x3, #48, #16                // set the reserved ASID + old BADDR
> +	msr     ttbr0_el1, x2                   // update TTBR0_EL1
> +	isb
> +	bfi     x2, x0, #0, #48                 // set the desired BADDR + reserved ASID
> +	msr     ttbr0_el1, x2                   // update TTBR0_EL1
> +	isb
> +alternative_endif

Please use alternative_if and alternative_else_nop_endif.

As Catalin noted, there are issues with stale and/or conflicting TLB
entries allocated with the reserved ASID, so we likely have to
invalidate that after the final switch.

Thanks,
Mark.

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH] arm64: Work around Falkor erratum 1009
  2016-12-08 11:20   ` Will Deacon
@ 2016-12-08 11:35     ` Marc Zyngier
  0 siblings, 0 replies; 10+ messages in thread
From: Marc Zyngier @ 2016-12-08 11:35 UTC (permalink / raw)
  To: Will Deacon, Christopher Covington
  Cc: Catalin Marinas, Paolo Bonzini, Radim Krčmář,
	Christoffer Dall, linux-arm-kernel, linux-kernel, kvm, kvmarm,
	Shanker Donthineni

On 08/12/16 11:20, Will Deacon wrote:
> On Wed, Dec 07, 2016 at 03:04:31PM -0500, Christopher Covington wrote:
>> From: Shanker Donthineni <shankerd@codeaurora.org>
>>
>> During a TLB invalidate sequence targeting the inner shareable
>> domain, Falkor may prematurely complete the DSB before all loads
>> and stores using the old translation are observed; instruction
>> fetches are not subject to the conditions of this erratum.
>>
>> Signed-off-by: Shanker Donthineni <shankerd@codeaurora.org>
>> Signed-off-by: Christopher Covington <cov@codeaurora.org>
>> ---
>>  arch/arm64/Kconfig                | 10 +++++++++
>>  arch/arm64/include/asm/cpucaps.h  |  3 ++-
>>  arch/arm64/include/asm/tlbflush.h | 43 +++++++++++++++++++++++++++++++++++++++
>>  arch/arm64/kernel/cpu_errata.c    |  7 +++++++
>>  arch/arm64/kvm/hyp/tlb.c          | 39 ++++++++++++++++++++++++++++++-----
>>  5 files changed, 96 insertions(+), 6 deletions(-)
>>
>> diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
>> index 1004a3d..125440f 100644
>> --- a/arch/arm64/Kconfig
>> +++ b/arch/arm64/Kconfig
>> @@ -485,6 +485,16 @@ config QCOM_FALKOR_ERRATUM_E1003
>>  
>>  	  If unsure, say Y.
>>  
>> +config QCOM_FALKOR_ERRATUM_E1009
>> +	bool "Falkor E1009: Prematurely complete a DSB after a TLBI"
>> +	default y
>> +	help
>> +	  Falkor CPU may prematurely complete a DSB following a TLBI xxIS
>> +	  invalidate maintenance operations. Repeat the TLBI operation one
>> +	  more time to fix the issue.
>> +
>> +	  If unsure, say Y.
> 
> Call me perverse, but I like this workaround. People often tend to screw
> up TLBI and DVM sync, but the IPI-based workaround is horribly invasive
> and fragile. Simply repeating the operation tends to be enough to make
> the chance of failure small enough to be acceptable.
> 
>> diff --git a/arch/arm64/include/asm/cpucaps.h b/arch/arm64/include/asm/cpucaps.h
>> index cb6a8c2..5357d7f 100644
>> --- a/arch/arm64/include/asm/cpucaps.h
>> +++ b/arch/arm64/include/asm/cpucaps.h
>> @@ -35,7 +35,8 @@
>>  #define ARM64_HYP_OFFSET_LOW			14
>>  #define ARM64_MISMATCHED_CACHE_LINE_SIZE	15
>>  #define ARM64_WORKAROUND_QCOM_FALKOR_E1003	16
>> +#define ARM64_WORKAROUND_QCOM_FALKOR_E1009	17
> 
> Could you rename this to something like ARM64_WORKAROUND_REPEAT_TLBI, so
> that it could potentially be used by others?

And add a parameter to it so that we can generate multiple TLBIs,
depending on the level of brokenness? ;-)

	M.
-- 
Jazz is not dead. It just smells funny...

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH] arm64: Work around Falkor erratum 1009
  2016-12-07 20:04 ` [PATCH] arm64: Work around Falkor erratum 1009 Christopher Covington
  2016-12-08 11:20   ` Will Deacon
@ 2016-12-08 11:45   ` Mark Rutland
  2016-12-08 13:27     ` Catalin Marinas
  1 sibling, 1 reply; 10+ messages in thread
From: Mark Rutland @ 2016-12-08 11:45 UTC (permalink / raw)
  To: Christopher Covington
  Cc: Catalin Marinas, Will Deacon, Paolo Bonzini,
	Radim Krčmář,
	Christoffer Dall, Marc Zyngier, linux-arm-kernel, linux-kernel,
	kvm, kvmarm

On Wed, Dec 07, 2016 at 03:04:31PM -0500, Christopher Covington wrote:
> From: Shanker Donthineni <shankerd@codeaurora.org>
> 
> During a TLB invalidate sequence targeting the inner shareable
> domain, Falkor may prematurely complete the DSB before all loads
> and stores using the old translation are observed; instruction
> fetches are not subject to the conditions of this erratum.
> 
> Signed-off-by: Shanker Donthineni <shankerd@codeaurora.org>
> Signed-off-by: Christopher Covington <cov@codeaurora.org>
> ---
>  arch/arm64/Kconfig                | 10 +++++++++
>  arch/arm64/include/asm/cpucaps.h  |  3 ++-
>  arch/arm64/include/asm/tlbflush.h | 43 +++++++++++++++++++++++++++++++++++++++
>  arch/arm64/kernel/cpu_errata.c    |  7 +++++++
>  arch/arm64/kvm/hyp/tlb.c          | 39 ++++++++++++++++++++++++++++++-----
>  5 files changed, 96 insertions(+), 6 deletions(-)

Please update Documentation/arm64/silicon-errata.txt respectively.

[...]

>  #include <linux/sched.h>
>  #include <asm/cputype.h>
> +#include <asm/alternative.h>

Nit: please keep includes (alphabetically) ordered (at least below the
linux/ or asm/ level).

[...]

> +	asm volatile(ALTERNATIVE(
> +		     "nop \n"
> +		     "nop \n",
> +		     "tlbi vmalle1is \n"
> +		     "dsb ish \n",

As a general note, perhaps we want a C compatible NOP_ALTERNATIVE() so
that the nop case can be implicitly generated for sequences like this.

Thanks,
Mark.

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH] arm64: Work around Falkor erratum 1009
  2016-12-08 11:45   ` Mark Rutland
@ 2016-12-08 13:27     ` Catalin Marinas
  0 siblings, 0 replies; 10+ messages in thread
From: Catalin Marinas @ 2016-12-08 13:27 UTC (permalink / raw)
  To: Mark Rutland
  Cc: Christopher Covington, kvm, Radim Krčmář,
	Marc Zyngier, Will Deacon, linux-kernel, linux-arm-kernel,
	Paolo Bonzini, kvmarm, Christoffer Dall

On Thu, Dec 08, 2016 at 11:45:12AM +0000, Mark Rutland wrote:
> On Wed, Dec 07, 2016 at 03:04:31PM -0500, Christopher Covington wrote:
> > +	asm volatile(ALTERNATIVE(
> > +		     "nop \n"
> > +		     "nop \n",
> > +		     "tlbi vmalle1is \n"
> > +		     "dsb ish \n",
> 
> As a general note, perhaps we want a C compatible NOP_ALTERNATIVE() so
> that the nop case can be implicitly generated for sequences like this.

It's also worth checking what cpus_have_const_cap() would generate for
the default (no workaround required) case.

-- 
Catalin

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH 2/3] arm64: Work around Falkor erratum 1003
  2016-12-08 10:31   ` Catalin Marinas
@ 2016-12-19 21:27     ` Christopher Covington
  0 siblings, 0 replies; 10+ messages in thread
From: Christopher Covington @ 2016-12-19 21:27 UTC (permalink / raw)
  To: Catalin Marinas
  Cc: Will Deacon, Shanker Donthineni, Suzuki K Poulose,
	Andre Przywara, Ganapatrao Kulkarni, James Morse, Andrew Pinski,
	Mark Rutland, Jean-Philippe Brucker, Lorenzo Pieralisi,
	Geoff Levand, linux-arm-kernel, linux-kernel

Hi Catalin,

On 12/08/2016 05:31 AM, Catalin Marinas wrote:
> On Wed, Dec 07, 2016 at 03:00:26PM -0500, Christopher Covington wrote:
>> From: Shanker Donthineni <shankerd@codeaurora.org>
>>
>> On the Qualcomm Datacenter Technologies Falkor v1 CPU, memory accesses may
>> allocate TLB entries using an incorrect ASID when TTBRx_EL1 is being
>> updated. Changing the TTBRx_EL1[ASID] and TTBRx_EL1[BADDR] fields
>> separately using a reserved ASID will ensure that there are no TLB entries
>> with incorrect ASID after changing the the ASID.
>>
>> Pseudo code:
>>   write TTBRx_EL1[ASID] to a reserved value
>>   ISB
>>   write TTBRx_EL1[BADDR] to a desired value
>>   ISB
>>   write TTBRx_EL1[ASID] to a desired value
>>   ISB
> 
> While the new ASID probably won't have incorrect TLB entries, the
> reserved ASID will have random entries from all over the place. That's
> because in step 1 you change the ASID to the reserved one while leaving
> the old BADDR in place. There is a brief time before changing the ASID
> when speculative page table walks will populate the TLB with entries
> tagged with the reserved ASID. Such entries are never removed during TLB
> shoot-down for the real ASID, so, depending on how this CPU implements
> the walk cache, you could end up with intermediate level entries still
> active and pointing to freed/reused pages. It will eventually hit an
> entry that looks global with weird consequences.
> 
> We've been bitten by this in the past on arm32: 52af9c6cd863 ("ARM:
> 6943/1: mm: use TTBR1 instead of reserved context ID").

Thanks for bringing this up, but I'm told the scenario you describe won't
happen on the Falkor 1.0 CPU.

Thanks,
Cov

-- 
Qualcomm Datacenter Technologies, Inc. as an affiliate of Qualcomm
Technologies, Inc. Qualcomm Technologies, Inc. is a member of the Code
Aurora Forum, a Linux Foundation Collaborative Project.

^ permalink raw reply	[flat|nested] 10+ messages in thread

end of thread, other threads:[~2016-12-19 21:27 UTC | newest]

Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2016-12-07 20:00 [PATCH 1/3] arm64: Define Qualcomm Technologies Falkor v1 CPU Christopher Covington
2016-12-07 20:00 ` [PATCH 2/3] arm64: Work around Falkor erratum 1003 Christopher Covington
2016-12-08 10:31   ` Catalin Marinas
2016-12-19 21:27     ` Christopher Covington
2016-12-08 11:31   ` Mark Rutland
2016-12-07 20:04 ` [PATCH] arm64: Work around Falkor erratum 1009 Christopher Covington
2016-12-08 11:20   ` Will Deacon
2016-12-08 11:35     ` Marc Zyngier
2016-12-08 11:45   ` Mark Rutland
2016-12-08 13:27     ` Catalin Marinas

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).