Merge branch 'fixes' into next

Merge our fixes branch from the 4.16 cycle.

There were a number of important fixes merged, in particular some Power9
workarounds that we want in next for testing purposes. There's also been
some conflicting changes in the CPU features code which are best merged
and tested before going upstream.

4 files changed, 107 insertions, 80 deletions

diff --git a/arch/powerpc/mm/hash_native_64.c b/arch/powerpc/mm/hash_native_64.c
index 993842f1ed60..1d049c78c82a 100644
--- a/arch/powerpc/mm/hash_native_64.c
+++ b/arch/powerpc/mm/hash_native_64.c
@@ -201,6 +201,15 @@ static inline unsigned long  ___tlbie(unsigned long vpn, int psize,
 	return va;
 }
 
+static inline void fixup_tlbie(unsigned long vpn, int psize, int apsize, int ssize)
+{
+	if (cpu_has_feature(CPU_FTR_P9_TLBIE_BUG)) {
+		/* Need the extra ptesync to ensure we don't reorder tlbie*/
+		asm volatile("ptesync": : :"memory");
+		___tlbie(vpn, psize, apsize, ssize);
+	}
+}
+
 static inline void __tlbie(unsigned long vpn, int psize, int apsize, int ssize)
 {
 	unsigned long rb;
@@ -278,6 +287,7 @@ static inline void tlbie(unsigned long vpn, int psize, int apsize,
 		asm volatile("ptesync": : :"memory");
 	} else {
 		__tlbie(vpn, psize, apsize, ssize);
+		fixup_tlbie(vpn, psize, apsize, ssize);
 		asm volatile("eieio; tlbsync; ptesync": : :"memory");
 	}
 	if (lock_tlbie && !use_local)
@@ -771,7 +781,7 @@ static void native_hpte_clear(void)
  */
 static void native_flush_hash_range(unsigned long number, int local)
 {
-	unsigned long vpn;
+	unsigned long vpn = 0;
 	unsigned long hash, index, hidx, shift, slot;
 	struct hash_pte *hptep;
 	unsigned long hpte_v;
@@ -843,6 +853,10 @@ static void native_flush_hash_range(unsigned long number, int local)
 				__tlbie(vpn, psize, psize, ssize);
 			} pte_iterate_hashed_end();
 		}
+		/*
+		 * Just do one more with the last used values.
+		 */
+		fixup_tlbie(vpn, psize, psize, ssize);
 		asm volatile("eieio; tlbsync; ptesync":::"memory");
 
 		if (lock_tlbie)
diff --git a/arch/powerpc/mm/mmu_context_book3s64.c b/arch/powerpc/mm/mmu_context_book3s64.c
index 80acad52b006..422be81bf69f 100644
--- a/arch/powerpc/mm/mmu_context_book3s64.c
+++ b/arch/powerpc/mm/mmu_context_book3s64.c
@@ -166,6 +166,7 @@ int init_new_context(struct task_struct *tsk, struct mm_struct *mm)
 	mm_iommu_init(mm);
 #endif
 	atomic_set(&mm->context.active_cpus, 0);
+	atomic_set(&mm->context.copros, 0);
 
 	return 0;
 }
diff --git a/arch/powerpc/mm/pgtable_64.c b/arch/powerpc/mm/pgtable_64.c
index 28c980eb4422..adf469f312f2 100644
--- a/arch/powerpc/mm/pgtable_64.c
+++ b/arch/powerpc/mm/pgtable_64.c
@@ -481,6 +481,7 @@ void mmu_partition_table_set_entry(unsigned int lpid, unsigned long dw0,
 			     "r" (TLBIEL_INVAL_SET_LPID), "r" (lpid));
 		trace_tlbie(lpid, 0, TLBIEL_INVAL_SET_LPID, lpid, 2, 0, 0);
 	}
+	/* do we need fixup here ?*/
 	asm volatile("eieio; tlbsync; ptesync" : : : "memory");
 }
 EXPORT_SYMBOL_GPL(mmu_partition_table_set_entry);
diff --git a/arch/powerpc/mm/tlb-radix.c b/arch/powerpc/mm/tlb-radix.c
index 001c1f6458b6..291eab4ed7cd 100644
--- a/arch/powerpc/mm/tlb-radix.c
+++ b/arch/powerpc/mm/tlb-radix.c
@@ -119,6 +119,49 @@ static inline void __tlbie_pid(unsigned long pid, unsigned long ric)
 	trace_tlbie(0, 0, rb, rs, ric, prs, r);
 }
 
+static inline void __tlbiel_va(unsigned long va, unsigned long pid,
+			       unsigned long ap, unsigned long ric)
+{
+	unsigned long rb,rs,prs,r;
+
+	rb = va & ~(PPC_BITMASK(52, 63));
+	rb |= ap << PPC_BITLSHIFT(58);
+	rs = pid << PPC_BITLSHIFT(31);
+	prs = 1; /* process scoped */
+	r = 1;   /* radix format */
+
+	asm volatile(PPC_TLBIEL(%0, %4, %3, %2, %1)
+		     : : "r"(rb), "i"(r), "i"(prs), "i"(ric), "r"(rs) : "memory");
+	trace_tlbie(0, 1, rb, rs, ric, prs, r);
+}
+
+static inline void __tlbie_va(unsigned long va, unsigned long pid,
+			      unsigned long ap, unsigned long ric)
+{
+	unsigned long rb,rs,prs,r;
+
+	rb = va & ~(PPC_BITMASK(52, 63));
+	rb |= ap << PPC_BITLSHIFT(58);
+	rs = pid << PPC_BITLSHIFT(31);
+	prs = 1; /* process scoped */
+	r = 1;   /* radix format */
+
+	asm volatile(PPC_TLBIE_5(%0, %4, %3, %2, %1)
+		     : : "r"(rb), "i"(r), "i"(prs), "i"(ric), "r"(rs) : "memory");
+	trace_tlbie(0, 0, rb, rs, ric, prs, r);
+}
+
+static inline void fixup_tlbie(void)
+{
+	unsigned long pid = 0;
+	unsigned long va = ((1UL << 52) - 1);
+
+	if (cpu_has_feature(CPU_FTR_P9_TLBIE_BUG)) {
+		asm volatile("ptesync": : :"memory");
+		__tlbie_va(va, pid, mmu_get_ap(MMU_PAGE_64K), RIC_FLUSH_TLB);
+	}
+}
+
 /*
  * We use 128 set in radix mode and 256 set in hpt mode.
  */
@@ -151,24 +194,25 @@ static inline void _tlbiel_pid(unsigned long pid, unsigned long ric)
 static inline void _tlbie_pid(unsigned long pid, unsigned long ric)
 {
 	asm volatile("ptesync": : :"memory");
-	__tlbie_pid(pid, ric);
-	asm volatile("eieio; tlbsync; ptesync": : :"memory");
-}
 
-static inline void __tlbiel_va(unsigned long va, unsigned long pid,
-			       unsigned long ap, unsigned long ric)
-{
-	unsigned long rb,rs,prs,r;
-
-	rb = va & ~(PPC_BITMASK(52, 63));
-	rb |= ap << PPC_BITLSHIFT(58);
-	rs = pid << PPC_BITLSHIFT(31);
-	prs = 1; /* process scoped */
-	r = 1;   /* radix format */
-
-	asm volatile(PPC_TLBIEL(%0, %4, %3, %2, %1)
-		     : : "r"(rb), "i"(r), "i"(prs), "i"(ric), "r"(rs) : "memory");
-	trace_tlbie(0, 1, rb, rs, ric, prs, r);
+	/*
+	 * Workaround the fact that the "ric" argument to __tlbie_pid
+	 * must be a compile-time contraint to match the "i" constraint
+	 * in the asm statement.
+	 */
+	switch (ric) {
+	case RIC_FLUSH_TLB:
+		__tlbie_pid(pid, RIC_FLUSH_TLB);
+		break;
+	case RIC_FLUSH_PWC:
+		__tlbie_pid(pid, RIC_FLUSH_PWC);
+		break;
+	case RIC_FLUSH_ALL:
+	default:
+		__tlbie_pid(pid, RIC_FLUSH_ALL);
+	}
+	fixup_tlbie();
+	asm volatile("eieio; tlbsync; ptesync": : :"memory");
 }
 
 static inline void __tlbiel_va_range(unsigned long start, unsigned long end,
@@ -203,22 +247,6 @@ static inline void _tlbiel_va_range(unsigned long start, unsigned long end,
 	asm volatile("ptesync": : :"memory");
 }
 
-static inline void __tlbie_va(unsigned long va, unsigned long pid,
-			     unsigned long ap, unsigned long ric)
-{
-	unsigned long rb,rs,prs,r;
-
-	rb = va & ~(PPC_BITMASK(52, 63));
-	rb |= ap << PPC_BITLSHIFT(58);
-	rs = pid << PPC_BITLSHIFT(31);
-	prs = 1; /* process scoped */
-	r = 1;   /* radix format */
-
-	asm volatile(PPC_TLBIE_5(%0, %4, %3, %2, %1)
-		     : : "r"(rb), "i"(r), "i"(prs), "i"(ric), "r"(rs) : "memory");
-	trace_tlbie(0, 0, rb, rs, ric, prs, r);
-}
-
 static inline void __tlbie_va_range(unsigned long start, unsigned long end,
 				    unsigned long pid, unsigned long page_size,
 				    unsigned long psize)
@@ -237,6 +265,7 @@ static inline void _tlbie_va(unsigned long va, unsigned long pid,
 
 	asm volatile("ptesync": : :"memory");
 	__tlbie_va(va, pid, ap, ric);
+	fixup_tlbie();
 	asm volatile("eieio; tlbsync; ptesync": : :"memory");
 }
 
@@ -248,6 +277,7 @@ static inline void _tlbie_va_range(unsigned long start, unsigned long end,
 	if (also_pwc)
 		__tlbie_pid(pid, RIC_FLUSH_PWC);
 	__tlbie_va_range(start, end, pid, page_size, psize);
+	fixup_tlbie();
 	asm volatile("eieio; tlbsync; ptesync": : :"memory");
 }
 
@@ -311,6 +341,16 @@ void radix__local_flush_tlb_page(struct vm_area_struct *vma, unsigned long vmadd
 }
 EXPORT_SYMBOL(radix__local_flush_tlb_page);
 
+static bool mm_needs_flush_escalation(struct mm_struct *mm)
+{
+	/*
+	 * P9 nest MMU has issues with the page walk cache
+	 * caching PTEs and not flushing them properly when
+	 * RIC = 0 for a PID/LPID invalidate
+	 */
+	return atomic_read(&mm->context.copros) != 0;
+}
+
 #ifdef CONFIG_SMP
 void radix__flush_tlb_mm(struct mm_struct *mm)
 {
@@ -321,9 +361,12 @@ void radix__flush_tlb_mm(struct mm_struct *mm)
 		return;
 
 	preempt_disable();
-	if (!mm_is_thread_local(mm))
-		_tlbie_pid(pid, RIC_FLUSH_TLB);
-	else
+	if (!mm_is_thread_local(mm)) {
+		if (mm_needs_flush_escalation(mm))
+			_tlbie_pid(pid, RIC_FLUSH_ALL);
+		else
+			_tlbie_pid(pid, RIC_FLUSH_TLB);
+	} else
 		_tlbiel_pid(pid, RIC_FLUSH_TLB);
 	preempt_enable();
 }
@@ -435,10 +478,14 @@ void radix__flush_tlb_range(struct vm_area_struct *vma, unsigned long start,
 	}
 
 	if (full) {
-		if (local)
+		if (local) {
 			_tlbiel_pid(pid, RIC_FLUSH_TLB);
-		else
-			_tlbie_pid(pid, RIC_FLUSH_TLB);
+		} else {
+			if (mm_needs_flush_escalation(mm))
+				_tlbie_pid(pid, RIC_FLUSH_ALL);
+			else
+				_tlbie_pid(pid, RIC_FLUSH_TLB);
+		}
 	} else {
 		bool hflush = false;
 		unsigned long hstart, hend;
@@ -465,6 +512,7 @@ void radix__flush_tlb_range(struct vm_area_struct *vma, unsigned long start,
 			if (hflush)
 				__tlbie_va_range(hstart, hend, pid,
 						HPAGE_PMD_SIZE, MMU_PAGE_2M);
+			fixup_tlbie();
 			asm volatile("eieio; tlbsync; ptesync": : :"memory");
 		}
 	}
@@ -548,6 +596,9 @@ static inline void __radix__flush_tlb_range_psize(struct mm_struct *mm,
 	}
 
 	if (full) {
+		if (!local && mm_needs_flush_escalation(mm))
+			also_pwc = true;
+
 		if (local)
 			_tlbiel_pid(pid, also_pwc ? RIC_FLUSH_ALL : RIC_FLUSH_TLB);
 		else
@@ -603,46 +654,6 @@ void radix__flush_tlb_collapsed_pmd(struct mm_struct *mm, unsigned long addr)
 }
 #endif /* CONFIG_TRANSPARENT_HUGEPAGE */
 
-void radix__flush_tlb_lpid_va(unsigned long lpid, unsigned long gpa,
-			      unsigned long page_size)
-{
-	unsigned long rb,rs,prs,r;
-	unsigned long ap;
-	unsigned long ric = RIC_FLUSH_TLB;
-
-	ap = mmu_get_ap(radix_get_mmu_psize(page_size));
-	rb = gpa & ~(PPC_BITMASK(52, 63));
-	rb |= ap << PPC_BITLSHIFT(58);
-	rs = lpid & ((1UL << 32) - 1);
-	prs = 0; /* process scoped */
-	r = 1;   /* radix format */
-
-	asm volatile("ptesync": : :"memory");
-	asm volatile(PPC_TLBIE_5(%0, %4, %3, %2, %1)
-		     : : "r"(rb), "i"(r), "i"(prs), "i"(ric), "r"(rs) : "memory");
-	asm volatile("eieio; tlbsync; ptesync": : :"memory");
-	trace_tlbie(lpid, 0, rb, rs, ric, prs, r);
-}
-EXPORT_SYMBOL(radix__flush_tlb_lpid_va);
-
-void radix__flush_tlb_lpid(unsigned long lpid)
-{
-	unsigned long rb,rs,prs,r;
-	unsigned long ric = RIC_FLUSH_ALL;
-
-	rb = 0x2 << PPC_BITLSHIFT(53); /* IS = 2 */
-	rs = lpid & ((1UL << 32) - 1);
-	prs = 0; /* partition scoped */
-	r = 1;   /* radix format */
-
-	asm volatile("ptesync": : :"memory");
-	asm volatile(PPC_TLBIE_5(%0, %4, %3, %2, %1)
-		     : : "r"(rb), "i"(r), "i"(prs), "i"(ric), "r"(rs) : "memory");
-	asm volatile("eieio; tlbsync; ptesync": : :"memory");
-	trace_tlbie(lpid, 0, rb, rs, ric, prs, r);
-}
-EXPORT_SYMBOL(radix__flush_tlb_lpid);
-
 void radix__flush_pmd_tlb_range(struct vm_area_struct *vma,
 				unsigned long start, unsigned long end)
 {