powerpc/mm: Drop WIMG in favour of new constants

PowerISA 3.0 introduces two pte bits with the below meaning for radix:
  00 -> Normal Memory
  01 -> Strong Access Order (SAO)
  10 -> Non idempotent I/O (Cache inhibited and guarded)
  11 -> Tolerant I/O (Cache inhibited)

We drop the existing WIMG bits in the Linux page table in favour of the
above constants. We loose _PAGE_WRITETHRU with this conversion. We only
use writethru via pgprot_cached_wthru() which is used by
fbdev/controlfb.c which is Apple control display and also PPC32.

With respect to _PAGE_COHERENCE, we have been marking hpte always
coherent for some time now. htab_convert_pte_flags() always added
HPTE_R_M.

NOTE: KVM changes need closer review.

Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>

arch/powerpc/include/asm/book3s/64/hash.h

@@ -21,11 +21,9 @@
 #define _PAGE_RW		(_PAGE_READ | _PAGE_WRITE)
 #define _PAGE_RWX		(_PAGE_READ | _PAGE_WRITE | _PAGE_EXEC)
 #define _PAGE_PRIVILEGED	0x00008 /* kernel access only */
-#define _PAGE_GUARDED		0x00010 /* G: guarded (side-effect) page */
-/* M (memory coherence) is always set in the HPTE, so we don't need it here */
-#define _PAGE_COHERENT		0x0
-#define _PAGE_NO_CACHE		0x00020 /* I: cache inhibit */
-#define _PAGE_WRITETHRU		0x00040 /* W: cache write-through */
+#define _PAGE_SAO		0x00010 /* Strong access order */
+#define _PAGE_NON_IDEMPOTENT	0x00020 /* non idempotent memory */
+#define _PAGE_TOLERANT		0x00030 /* tolerant memory, cache inhibited */
 #define _PAGE_DIRTY		0x00080 /* C: page changed */
 #define _PAGE_ACCESSED		0x00100 /* R: page referenced */
 #define _PAGE_SPECIAL		0x00400 /* software: special page */
@@ -43,7 +41,12 @@
 #define _PAGE_HASHPTE		(1ul << 61)	/* PTE has associated HPTE */
 #define _PAGE_PTE		(1ul << 62)	/* distinguishes PTEs from pointers */
 #define _PAGE_PRESENT		(1ul << 63)	/* pte contains a translation */
-
+/*
+ * Drivers request for cache inhibited pte mapping using _PAGE_NO_CACHE
+ * Instead of fixing all of them, add an alternate define which
+ * maps CI pte mapping.
+ */
+#define _PAGE_NO_CACHE		_PAGE_TOLERANT
 /*
  * We need to differentiate between explicit huge page and THP huge
  * page, since THP huge page also need to track real subpage details
@@ -126,9 +129,6 @@
 #define _PAGE_KERNEL_RWX	(_PAGE_PRIVILEGED | _PAGE_DIRTY | \
 				 _PAGE_RW | _PAGE_EXEC)
 
-/* Strong Access Ordering */
-#define _PAGE_SAO		(_PAGE_WRITETHRU | _PAGE_NO_CACHE | _PAGE_COHERENT)
-
 /* No page size encoding in the linux PTE */
 #define _PAGE_PSIZE		0
 
@@ -147,10 +147,9 @@
 /*
  * Mask of bits returned by pte_pgprot()
  */
-#define PAGE_PROT_BITS	(_PAGE_GUARDED | _PAGE_COHERENT | _PAGE_NO_CACHE | \
-			 _PAGE_WRITETHRU | _PAGE_4K_PFN | \
-			 _PAGE_PRIVILEGED | _PAGE_ACCESSED |  _PAGE_READ |\
-			 _PAGE_WRITE |  _PAGE_DIRTY | _PAGE_EXEC | \
+#define PAGE_PROT_BITS  (_PAGE_SAO | _PAGE_NON_IDEMPOTENT | _PAGE_TOLERANT | \
+			 _PAGE_4K_PFN | _PAGE_PRIVILEGED | _PAGE_ACCESSED | \
+			 _PAGE_READ | _PAGE_WRITE |  _PAGE_DIRTY | _PAGE_EXEC | \
 			 _PAGE_SOFT_DIRTY)
 /*
  * We define 2 sets of base prot bits, one for basic pages (ie,
@@ -159,7 +158,7 @@
  * the processor might need it for DMA coherency.
  */
 #define _PAGE_BASE_NC	(_PAGE_PRESENT | _PAGE_ACCESSED | _PAGE_PSIZE)
-#define _PAGE_BASE	(_PAGE_BASE_NC | _PAGE_COHERENT)
+#define _PAGE_BASE	(_PAGE_BASE_NC)
 
 /* Permission masks used to generate the __P and __S table,
  *
@@ -200,9 +199,9 @@
 /* Permission masks used for kernel mappings */
 #define PAGE_KERNEL	__pgprot(_PAGE_BASE | _PAGE_KERNEL_RW)
 #define PAGE_KERNEL_NC	__pgprot(_PAGE_BASE_NC | _PAGE_KERNEL_RW | \
-				 _PAGE_NO_CACHE)
+				 _PAGE_TOLERANT)
 #define PAGE_KERNEL_NCG	__pgprot(_PAGE_BASE_NC | _PAGE_KERNEL_RW | \
-				 _PAGE_NO_CACHE | _PAGE_GUARDED)
+				 _PAGE_NON_IDEMPOTENT)
 #define PAGE_KERNEL_X	__pgprot(_PAGE_BASE | _PAGE_KERNEL_RWX)
 #define PAGE_KERNEL_RO	__pgprot(_PAGE_BASE | _PAGE_KERNEL_RO)
 #define PAGE_KERNEL_ROX	__pgprot(_PAGE_BASE | _PAGE_KERNEL_ROX)
@@ -509,45 +508,26 @@ static inline void __set_pte_at(struct mm_struct *mm, unsigned long addr,
 	*ptep = pte;
 }
 
-/*
- * Macro to mark a page protection value as "uncacheable".
- */
-
-#define _PAGE_CACHE_CTL	(_PAGE_COHERENT | _PAGE_GUARDED | _PAGE_NO_CACHE | \
-			 _PAGE_WRITETHRU)
+#define _PAGE_CACHE_CTL	(_PAGE_NON_IDEMPOTENT | _PAGE_TOLERANT)
 
 #define pgprot_noncached pgprot_noncached
 static inline pgprot_t pgprot_noncached(pgprot_t prot)
 {
 	return __pgprot((pgprot_val(prot) & ~_PAGE_CACHE_CTL) |
-			_PAGE_NO_CACHE | _PAGE_GUARDED);
+			_PAGE_NON_IDEMPOTENT);
 }
 
 #define pgprot_noncached_wc pgprot_noncached_wc
 static inline pgprot_t pgprot_noncached_wc(pgprot_t prot)
 {
 	return __pgprot((pgprot_val(prot) & ~_PAGE_CACHE_CTL) |
-			_PAGE_NO_CACHE);
+			_PAGE_TOLERANT);
 }
 
 #define pgprot_cached pgprot_cached
 static inline pgprot_t pgprot_cached(pgprot_t prot)
 {
-	return __pgprot((pgprot_val(prot) & ~_PAGE_CACHE_CTL) |
-			_PAGE_COHERENT);
-}
-
-#define pgprot_cached_wthru pgprot_cached_wthru
-static inline pgprot_t pgprot_cached_wthru(pgprot_t prot)
-{
-	return __pgprot((pgprot_val(prot) & ~_PAGE_CACHE_CTL) |
-			_PAGE_COHERENT | _PAGE_WRITETHRU);
-}
-
-#define pgprot_cached_noncoherent pgprot_cached_noncoherent
-static inline pgprot_t pgprot_cached_noncoherent(pgprot_t prot)
-{
-	return __pgprot(pgprot_val(prot) & ~_PAGE_CACHE_CTL);
+	return __pgprot((pgprot_val(prot) & ~_PAGE_CACHE_CTL));
 }
 
 #define pgprot_writecombine pgprot_writecombine
@@ -555,6 +535,18 @@ static inline pgprot_t pgprot_writecombine(pgprot_t prot)
 {
 	return pgprot_noncached_wc(prot);
 }
+/*
+ * check a pte mapping have cache inhibited property
+ */
+static inline bool pte_ci(pte_t pte)
+{
+	unsigned long pte_v = pte_val(pte);
+
+	if (((pte_v & _PAGE_CACHE_CTL) == _PAGE_TOLERANT) ||
+	    ((pte_v & _PAGE_CACHE_CTL) == _PAGE_NON_IDEMPOTENT))
+		return true;
+	return false;
+}
 
 #ifdef CONFIG_TRANSPARENT_HUGEPAGE
 extern void hpte_do_hugepage_flush(struct mm_struct *mm, unsigned long addr,

arch/powerpc/include/asm/kvm_book3s_64.h

@@ -276,19 +276,24 @@ static inline unsigned long hpte_make_readonly(unsigned long ptel)
 	return ptel;
 }
 
-static inline int hpte_cache_flags_ok(unsigned long ptel, unsigned long io_type)
+static inline bool hpte_cache_flags_ok(unsigned long hptel, bool is_ci)
 {
-	unsigned int wimg = ptel & HPTE_R_WIMG;
+	unsigned int wimg = hptel & HPTE_R_WIMG;
 
 	/* Handle SAO */
 	if (wimg == (HPTE_R_W | HPTE_R_I | HPTE_R_M) &&
 	    cpu_has_feature(CPU_FTR_ARCH_206))
 		wimg = HPTE_R_M;
 
-	if (!io_type)
+	if (!is_ci)
 		return wimg == HPTE_R_M;
-
-	return (wimg & (HPTE_R_W | HPTE_R_I)) == io_type;
+	/*
+	 * if host is mapped cache inhibited, make sure hptel also have
+	 * cache inhibited.
+	 */
+	if (wimg & HPTE_R_W) /* FIXME!! is this ok for all guest. ? */
+		return false;
+	return !!(wimg & HPTE_R_I);
 }
 
 /*
@@ -325,18 +330,6 @@ static inline pte_t kvmppc_read_update_linux_pte(pte_t *ptep, int writing)
 	return new_pte;
 }
 
-
-/* Return HPTE cache control bits corresponding to Linux pte bits */
-static inline unsigned long hpte_cache_bits(unsigned long pte_val)
-{
-#if _PAGE_NO_CACHE == HPTE_R_I && _PAGE_WRITETHRU == HPTE_R_W
-	return pte_val & (HPTE_R_W | HPTE_R_I);
-#else
-	return ((pte_val & _PAGE_NO_CACHE) ? HPTE_R_I : 0) +
-		((pte_val & _PAGE_WRITETHRU) ? HPTE_R_W : 0);
-#endif
-}
-
 static inline bool hpte_read_permission(unsigned long pp, unsigned long key)
 {
 	if (key)

arch/powerpc/kvm/book3s_64_mmu_hv.c

@@ -447,7 +447,7 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu,
 	struct revmap_entry *rev;
 	struct page *page, *pages[1];
 	long index, ret, npages;
-	unsigned long is_io;
+	bool is_ci;
 	unsigned int writing, write_ok;
 	struct vm_area_struct *vma;
 	unsigned long rcbits;
@@ -503,7 +503,7 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu,
 	smp_rmb();
 
 	ret = -EFAULT;
-	is_io = 0;
+	is_ci = false;
 	pfn = 0;
 	page = NULL;
 	pte_size = PAGE_SIZE;
@@ -521,7 +521,7 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu,
 			pfn = vma->vm_pgoff +
 				((hva - vma->vm_start) >> PAGE_SHIFT);
 			pte_size = psize;
-			is_io = hpte_cache_bits(pgprot_val(vma->vm_page_prot));
+			is_ci = pte_ci(__pte((pgprot_val(vma->vm_page_prot))));
 			write_ok = vma->vm_flags & VM_WRITE;
 		}
 		up_read(&current->mm->mmap_sem);
@@ -558,10 +558,9 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu,
 		goto out_put;
 
 	/* Check WIMG vs. the actual page we're accessing */
-	if (!hpte_cache_flags_ok(r, is_io)) {
-		if (is_io)
+	if (!hpte_cache_flags_ok(r, is_ci)) {
+		if (is_ci)
 			goto out_put;
-
 		/*
 		 * Allow guest to map emulated device memory as
 		 * uncacheable, but actually make it cacheable.

arch/powerpc/kvm/book3s_hv_rm_mmu.c

@@ -175,7 +175,7 @@ long kvmppc_do_h_enter(struct kvm *kvm, unsigned long flags,
 	unsigned long g_ptel;
 	struct kvm_memory_slot *memslot;
 	unsigned hpage_shift;
-	unsigned long is_io;
+	bool is_ci;
 	unsigned long *rmap;
 	pte_t *ptep;
 	unsigned int writing;
@@ -199,7 +199,7 @@ long kvmppc_do_h_enter(struct kvm *kvm, unsigned long flags,
 	gfn = gpa >> PAGE_SHIFT;
 	memslot = __gfn_to_memslot(kvm_memslots_raw(kvm), gfn);
 	pa = 0;
-	is_io = ~0ul;
+	is_ci = false;
 	rmap = NULL;
 	if (!(memslot && !(memslot->flags & KVM_MEMSLOT_INVALID))) {
 		/* Emulated MMIO - mark this with key=31 */
@@ -250,7 +250,7 @@ long kvmppc_do_h_enter(struct kvm *kvm, unsigned long flags,
 			if (writing && !pte_write(pte))
 				/* make the actual HPTE be read-only */
 				ptel = hpte_make_readonly(ptel);
-			is_io = hpte_cache_bits(pte_val(pte));
+			is_ci = pte_ci(pte);
 			pa = pte_pfn(pte) << PAGE_SHIFT;
 			pa |= hva & (host_pte_size - 1);
 			pa |= gpa & ~PAGE_MASK;
@@ -267,9 +267,9 @@ long kvmppc_do_h_enter(struct kvm *kvm, unsigned long flags,
 	else
 		pteh |= HPTE_V_ABSENT;
 
-	/* Check WIMG */
-	if (is_io != ~0ul && !hpte_cache_flags_ok(ptel, is_io)) {
-		if (is_io)
+	/*If we had host pte mapping then  Check WIMG */
+	if (ptep && !hpte_cache_flags_ok(ptel, is_ci)) {
+		if (is_ci)
 			return H_PARAMETER;
 		/*
 		 * Allow guest to map emulated device memory as

arch/powerpc/mm/hash64_64k.c

@@ -244,7 +244,7 @@ int __hash_page_64K(unsigned long ea, unsigned long access,
 		 * If so, bail out and refault as a 4k page
 		 */
 		if (!mmu_has_feature(MMU_FTR_CI_LARGE_PAGE) &&
-		    unlikely(old_pte & _PAGE_NO_CACHE))
+		    unlikely(pte_ci(pte)))
 			return 0;
 		/*
 		 * Try to lock the PTE, add ACCESSED and DIRTY if it was

arch/powerpc/mm/hash_utils_64.c

@@ -192,12 +192,13 @@ unsigned long htab_convert_pte_flags(unsigned long pteflags)
 	/*
 	 * Add in WIG bits
 	 */
-	if (pteflags & _PAGE_WRITETHRU)
-		rflags |= HPTE_R_W;
-	if (pteflags & _PAGE_NO_CACHE)
+
+	if ((pteflags & _PAGE_CACHE_CTL) == _PAGE_TOLERANT)
 		rflags |= HPTE_R_I;
-	if (pteflags & _PAGE_GUARDED)
-		rflags |= HPTE_R_G;
+	if ((pteflags & _PAGE_CACHE_CTL ) == _PAGE_NON_IDEMPOTENT)
+		rflags |= (HPTE_R_I | HPTE_R_G);
+	if ((pteflags & _PAGE_CACHE_CTL) == _PAGE_SAO)
+		rflags |= (HPTE_R_I | HPTE_R_W);
 
 	return rflags;
 }
@@ -1142,8 +1143,7 @@ int hash_page_mm(struct mm_struct *mm, unsigned long ea,
 	/* If this PTE is non-cacheable and we have restrictions on
 	 * using non cacheable large pages, then we switch to 4k
 	 */
-	if (mmu_ci_restrictions && psize == MMU_PAGE_64K &&
-	    (pte_val(*ptep) & _PAGE_NO_CACHE)) {
+	if (mmu_ci_restrictions && psize == MMU_PAGE_64K && pte_ci(*ptep)) {
 		if (user_region) {
 			demote_segment_4k(mm, ea);
 			psize = MMU_PAGE_4K;
@@ -1297,13 +1297,13 @@ void hash_preload(struct mm_struct *mm, unsigned long ea,
 
 	WARN_ON(hugepage_shift);
 #ifdef CONFIG_PPC_64K_PAGES
-	/* If either _PAGE_4K_PFN or _PAGE_NO_CACHE is set (and we are on
+	/* If either _PAGE_4K_PFN or cache inhibited is set (and we are on
 	 * a 64K kernel), then we don't preload, hash_page() will take
 	 * care of it once we actually try to access the page.
 	 * That way we don't have to duplicate all of the logic for segment
 	 * page size demotion here
 	 */
-	if (pte_val(*ptep) & (_PAGE_4K_PFN | _PAGE_NO_CACHE))
+	if ((pte_val(*ptep) & _PAGE_4K_PFN) || pte_ci(*ptep))
 		goto out_exit;
 #endif /* CONFIG_PPC_64K_PAGES */
 

arch/powerpc/mm/pgtable.c

@@ -38,16 +38,16 @@ static inline int is_exec_fault(void)
 
 /* We only try to do i/d cache coherency on stuff that looks like
  * reasonably "normal" PTEs. We currently require a PTE to be present
- * and we avoid _PAGE_SPECIAL and _PAGE_NO_CACHE. We also only do that
+ * and we avoid _PAGE_SPECIAL and cache inhibited pte. We also only do that
  * on userspace PTEs
  */
 static inline int pte_looks_normal(pte_t pte)
 {
 
 #if defined(CONFIG_PPC_BOOK3S_64)
-	if ((pte_val(pte) &
-	     (_PAGE_PRESENT | _PAGE_SPECIAL | _PAGE_NO_CACHE)) ==
-	    _PAGE_PRESENT) {
+	if ((pte_val(pte) & (_PAGE_PRESENT | _PAGE_SPECIAL)) == _PAGE_PRESENT) {
+		if (pte_ci(pte))
+			return 0;
 		if (pte_user(pte))
 			return 1;
 	}

arch/powerpc/mm/pgtable_64.c

@@ -167,10 +167,6 @@ void __iomem * __ioremap_at(phys_addr_t pa, void *ea, unsigned long size,
 	if ((flags & _PAGE_PRESENT) == 0)
 		flags |= pgprot_val(PAGE_KERNEL);
 
-	/* Non-cacheable page cannot be coherent */
-	if (flags & _PAGE_NO_CACHE)
-		flags &= ~_PAGE_COHERENT;
-
 	/* We don't support the 4K PFN hack with ioremap */
 	if (flags & _PAGE_4K_PFN)
 		return NULL;

arch/powerpc/platforms/pseries/lpar.c

@@ -152,10 +152,6 @@ static long pSeries_lpar_hpte_insert(unsigned long hpte_group,
 	/* Exact = 0                   */
 	flags = 0;
 
-	/* Make pHyp happy */
-	if ((rflags & _PAGE_NO_CACHE) && !(rflags & _PAGE_WRITETHRU))
-		hpte_r &= ~HPTE_R_M;
-
 	if (firmware_has_feature(FW_FEATURE_XCMO) && !(hpte_r & HPTE_R_N))
 		flags |= H_COALESCE_CAND;