powerpc/mm/radix: Change pte relax sequence to handle nest MMU hang

When relaxing access (read -> read_write update), pte needs to be marked invalid
to handle a nest MMU bug. We also need to do a tlb flush after the pte is
marked invalid before updating the pte with new access bits.

We also move tlb flush to platform specific __ptep_set_access_flags. This will
help us to gerid of unnecessary tlb flush on BOOK3S 64 later. We don't do that
in this patch. This also helps in avoiding multiple tlbies with coprocessor
attached.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

arch/powerpc/include/asm/book3s/32/pgtable.h

@@ -245,6 +245,8 @@ static inline void __ptep_set_access_flags(struct vm_area_struct *vma,
 	unsigned long clr = ~pte_val(entry) & _PAGE_RO;
 
 	pte_update(ptep, clr, set);
+
+	flush_tlb_page(vma, address);
 }
 
 #define __HAVE_ARCH_PTE_SAME

arch/powerpc/include/asm/nohash/32/pgtable.h

@@ -266,6 +266,8 @@ static inline void __ptep_set_access_flags(struct vm_area_struct *vma,
 	unsigned long clr = ~pte_val(entry) & (_PAGE_RO | _PAGE_NA);
 
 	pte_update(ptep, clr, set);
+
+	flush_tlb_page(vma, address);
 }
 
 static inline int pte_young(pte_t pte)

arch/powerpc/include/asm/nohash/64/pgtable.h

@@ -304,6 +304,8 @@ static inline void __ptep_set_access_flags(struct vm_area_struct *vma,
 	unsigned long old = pte_val(*ptep);
 	*ptep = __pte(old | bits);
 #endif
+
+	flush_tlb_page(vma, address);
 }
 
 #define __HAVE_ARCH_PTE_SAME

arch/powerpc/include/asm/pgtable.h

@@ -8,6 +8,7 @@
 #include <asm/processor.h>		/* For TASK_SIZE */
 #include <asm/mmu.h>
 #include <asm/page.h>
+#include <asm/tlbflush.h>
 
 struct mm_struct;
 

arch/powerpc/mm/pgtable-book3s64.c

@@ -52,7 +52,6 @@ int pmdp_set_access_flags(struct vm_area_struct *vma, unsigned long address,
 		 */
 		__ptep_set_access_flags(vma, pmdp_ptep(pmdp),
 					pmd_pte(entry), address, MMU_PAGE_2M);
-		flush_pmd_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
 	}
 	return changed;
 }

arch/powerpc/mm/pgtable-radix.c

@@ -1091,8 +1091,12 @@ void radix__ptep_set_access_flags(struct vm_area_struct *vma, pte_t *ptep,
 	struct mm_struct *mm = vma->vm_mm;
 	unsigned long set = pte_val(entry) & (_PAGE_DIRTY | _PAGE_ACCESSED |
 					      _PAGE_RW | _PAGE_EXEC);
-
-	if (cpu_has_feature(CPU_FTR_POWER9_DD1)) {
+	/*
+	 * To avoid NMMU hang while relaxing access, we need mark
+	 * the pte invalid in between.
+	 */
+	if (cpu_has_feature(CPU_FTR_POWER9_DD1) ||
+	    atomic_read(&mm->context.copros) > 0) {
 		unsigned long old_pte, new_pte;
 
 		old_pte = __radix_pte_update(ptep, ~0, 0);
@@ -1100,9 +1104,11 @@ void radix__ptep_set_access_flags(struct vm_area_struct *vma, pte_t *ptep,
 		 * new value of pte
 		 */
 		new_pte = old_pte | set;
-		radix__flush_tlb_pte_p9_dd1(old_pte, mm, address);
+		radix__flush_tlb_page_psize(mm, address, psize);
 		__radix_pte_update(ptep, 0, new_pte);
-	} else
+	} else {
 		__radix_pte_update(ptep, 0, set);
+		radix__flush_tlb_page_psize(mm, address, psize);
+	}
 	asm volatile("ptesync" : : : "memory");
 }

arch/powerpc/mm/pgtable.c

@@ -224,7 +224,6 @@ int ptep_set_access_flags(struct vm_area_struct *vma, unsigned long address,
 		assert_pte_locked(vma->vm_mm, address);
 		__ptep_set_access_flags(vma, ptep, entry,
 					address, mmu_virtual_psize);
-		flush_tlb_page(vma, address);
 	}
 	return changed;
 }
@@ -263,7 +262,6 @@ extern int huge_ptep_set_access_flags(struct vm_area_struct *vma,
 		assert_spin_locked(&vma->vm_mm->page_table_lock);
 #endif
 		__ptep_set_access_flags(vma, ptep, pte, addr, psize);
-		flush_hugetlb_page(vma, addr);
 	}
 	return changed;
 #endif