Merge tag 'powerpc-3.19-1' of git://git.kernel.org/pub/scm/linux/kernel/git/mpe/linux

Pull powerpc updates from Michael Ellerman:
 "Some nice cleanups like removing bootmem, and removal of
  __get_cpu_var().

  There is one patch to mm/gup.c.  This is the generic GUP
  implementation, but is only used by us and arm(64).  We have an ack
  from Steve Capper, and although we didn't get an ack from Andrew he
  told us to take the patch through the powerpc tree.

  There's one cxl patch.  This is in drivers/misc, but Greg said he was
  happy for us to manage fixes for it.

  There is an infrastructure patch to support an IPMI driver for OPAL.

  There is also an RTC driver for OPAL.  We weren't able to get any
  response from the RTC maintainer, Alessandro Zummo, so in the end we
  just merged the driver.

  The usual batch of Freescale updates from Scott"

* tag 'powerpc-3.19-1' of git://git.kernel.org/pub/scm/linux/kernel/git/mpe/linux: (101 commits)
  powerpc/powernv: Return to cpu offline loop when finished in KVM guest
  powerpc/book3s: Fix partial invalidation of TLBs in MCE code.
  powerpc/mm: don't do tlbie for updatepp request with NO HPTE fault
  powerpc/xmon: Cleanup the breakpoint flags
  powerpc/xmon: Enable HW instruction breakpoint on POWER8
  powerpc/mm/thp: Use tlbiel if possible
  powerpc/mm/thp: Remove code duplication
  powerpc/mm/hugetlb: Sanity check gigantic hugepage count
  powerpc/oprofile: Disable pagefaults during user stack read
  powerpc/mm: Check for matching hpte without taking hpte lock
  powerpc: Drop useless warning in eeh_init()
  powerpc/powernv: Cleanup unused MCE definitions/declarations.
  powerpc/eeh: Dump PHB diag-data early
  powerpc/eeh: Recover EEH error on ownership change for BCM5719
  powerpc/eeh: Set EEH_PE_RESET on PE reset
  powerpc/eeh: Refactor eeh_reset_pe()
  powerpc: Remove more traces of bootmem
  powerpc/pseries: Initialise nvram_pstore_info's buf_lock
  cxl: Name interrupts in /proc/interrupt
  cxl: Return error to PSL if IRQ demultiplexing fails & print clearer warning
  ...

1 files changed, 36 insertions, 68 deletions

diff --git a/arch/powerpc/mm/pgtable_64.c b/arch/powerpc/mm/pgtable_64.c
index c8d709ab489d..4fe5f64cc179 100644
--- a/arch/powerpc/mm/pgtable_64.c
+++ b/arch/powerpc/mm/pgtable_64.c
@@ -33,9 +33,9 @@
 #include <linux/swap.h>
 #include <linux/stddef.h>
 #include <linux/vmalloc.h>
-#include <linux/bootmem.h>
 #include <linux/memblock.h>
 #include <linux/slab.h>
+#include <linux/hugetlb.h>
 
 #include <asm/pgalloc.h>
 #include <asm/page.h>
@@ -51,6 +51,7 @@
 #include <asm/cputable.h>
 #include <asm/sections.h>
 #include <asm/firmware.h>
+#include <asm/dma.h>
 
 #include "mmu_decl.h"
 
@@ -75,11 +76,7 @@ static __ref void *early_alloc_pgtable(unsigned long size)
 {
 	void *pt;
 
-	if (init_bootmem_done)
-		pt = __alloc_bootmem(size, size, __pa(MAX_DMA_ADDRESS));
-	else
-		pt = __va(memblock_alloc_base(size, size,
-					 __pa(MAX_DMA_ADDRESS)));
+	pt = __va(memblock_alloc_base(size, size, __pa(MAX_DMA_ADDRESS)));
 	memset(pt, 0, size);
 
 	return pt;
@@ -113,10 +110,6 @@ int map_kernel_page(unsigned long ea, unsigned long pa, int flags)
 							  __pgprot(flags)));
 	} else {
 #ifdef CONFIG_PPC_MMU_NOHASH
-		/* Warning ! This will blow up if bootmem is not initialized
-		 * which our ppc64 code is keen to do that, we'll need to
-		 * fix it and/or be more careful
-		 */
 		pgdp = pgd_offset_k(ea);
 #ifdef PUD_TABLE_SIZE
 		if (pgd_none(*pgdp)) {
@@ -352,16 +345,31 @@ EXPORT_SYMBOL(iounmap);
 EXPORT_SYMBOL(__iounmap);
 EXPORT_SYMBOL(__iounmap_at);
 
+#ifndef __PAGETABLE_PUD_FOLDED
+/* 4 level page table */
+struct page *pgd_page(pgd_t pgd)
+{
+	if (pgd_huge(pgd))
+		return pte_page(pgd_pte(pgd));
+	return virt_to_page(pgd_page_vaddr(pgd));
+}
+#endif
+
+struct page *pud_page(pud_t pud)
+{
+	if (pud_huge(pud))
+		return pte_page(pud_pte(pud));
+	return virt_to_page(pud_page_vaddr(pud));
+}
+
 /*
  * For hugepage we have pfn in the pmd, we use PTE_RPN_SHIFT bits for flags
  * For PTE page, we have a PTE_FRAG_SIZE (4K) aligned virtual address.
  */
 struct page *pmd_page(pmd_t pmd)
 {
-#ifdef CONFIG_TRANSPARENT_HUGEPAGE
-	if (pmd_trans_huge(pmd))
+	if (pmd_trans_huge(pmd) || pmd_huge(pmd))
 		return pfn_to_page(pmd_pfn(pmd));
-#endif
 	return virt_to_page(pmd_page_vaddr(pmd));
 }
 
@@ -731,29 +739,15 @@ void pmdp_invalidate(struct vm_area_struct *vma, unsigned long address,
 void hpte_do_hugepage_flush(struct mm_struct *mm, unsigned long addr,
 			    pmd_t *pmdp, unsigned long old_pmd)
 {
-	int ssize, i;
-	unsigned long s_addr;
-	int max_hpte_count;
-	unsigned int psize, valid;
-	unsigned char *hpte_slot_array;
-	unsigned long hidx, vpn, vsid, hash, shift, slot;
-
-	/*
-	 * Flush all the hptes mapping this hugepage
-	 */
-	s_addr = addr & HPAGE_PMD_MASK;
-	hpte_slot_array = get_hpte_slot_array(pmdp);
-	/*
-	 * IF we try to do a HUGE PTE update after a withdraw is done.
-	 * we will find the below NULL. This happens when we do
-	 * split_huge_page_pmd
-	 */
-	if (!hpte_slot_array)
-		return;
+	int ssize;
+	unsigned int psize;
+	unsigned long vsid;
+	unsigned long flags = 0;
+	const struct cpumask *tmp;
 
 	/* get the base page size,vsid and segment size */
 #ifdef CONFIG_DEBUG_VM
-	psize = get_slice_psize(mm, s_addr);
+	psize = get_slice_psize(mm, addr);
 	BUG_ON(psize == MMU_PAGE_16M);
 #endif
 	if (old_pmd & _PAGE_COMBO)
@@ -761,46 +755,20 @@ void hpte_do_hugepage_flush(struct mm_struct *mm, unsigned long addr,
 	else
 		psize = MMU_PAGE_64K;
 
-	if (!is_kernel_addr(s_addr)) {
-		ssize = user_segment_size(s_addr);
-		vsid = get_vsid(mm->context.id, s_addr, ssize);
+	if (!is_kernel_addr(addr)) {
+		ssize = user_segment_size(addr);
+		vsid = get_vsid(mm->context.id, addr, ssize);
 		WARN_ON(vsid == 0);
 	} else {
-		vsid = get_kernel_vsid(s_addr, mmu_kernel_ssize);
+		vsid = get_kernel_vsid(addr, mmu_kernel_ssize);
 		ssize = mmu_kernel_ssize;
 	}
 
-	if (ppc_md.hugepage_invalidate)
-		return ppc_md.hugepage_invalidate(vsid, s_addr,
-						  hpte_slot_array,
-						  psize, ssize);
-	/*
-	 * No bluk hpte removal support, invalidate each entry
-	 */
-	shift = mmu_psize_defs[psize].shift;
-	max_hpte_count = HPAGE_PMD_SIZE >> shift;
-	for (i = 0; i < max_hpte_count; i++) {
-		/*
-		 * 8 bits per each hpte entries
-		 * 000| [ secondary group (one bit) | hidx (3 bits) | valid bit]
-		 */
-		valid = hpte_valid(hpte_slot_array, i);
-		if (!valid)
-			continue;
-		hidx =  hpte_hash_index(hpte_slot_array, i);
-
-		/* get the vpn */
-		addr = s_addr + (i * (1ul << shift));
-		vpn = hpt_vpn(addr, vsid, ssize);
-		hash = hpt_hash(vpn, shift, ssize);
-		if (hidx & _PTEIDX_SECONDARY)
-			hash = ~hash;
-
-		slot = (hash & htab_hash_mask) * HPTES_PER_GROUP;
-		slot += hidx & _PTEIDX_GROUP_IX;
-		ppc_md.hpte_invalidate(slot, vpn, psize,
-				       MMU_PAGE_16M, ssize, 0);
-	}
+	tmp = cpumask_of(smp_processor_id());
+	if (cpumask_equal(mm_cpumask(mm), tmp))
+		flags |= HPTE_LOCAL_UPDATE;
+
+	return flush_hash_hugepage(vsid, addr, pmdp, psize, ssize, flags);
 }
 
 static pmd_t pmd_set_protbits(pmd_t pmd, pgprot_t pgprot)