x86: introduce max_low_pfn_mapped for 64-bit

when more than 4g memory is installed, don't map the big hole below 4g.

Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

12 files changed, 74 insertions, 17 deletions

diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c
index a31a579a47ca..9c981c4a3644 100644
--- a/arch/x86/kernel/acpi/boot.c
+++ b/arch/x86/kernel/acpi/boot.c
@@ -130,7 +130,7 @@ char *__init __acpi_map_table(unsigned long phys, unsigned long size)
         if (!phys || !size)
                 return NULL;
 
-        if (phys+size <= (max_pfn_mapped << PAGE_SHIFT))
+        if (phys+size <= (max_low_pfn_mapped << PAGE_SHIFT))
                 return __va(phys);
 
         offset = phys & (PAGE_SIZE - 1);
diff --git a/arch/x86/kernel/cpu/amd_64.c b/arch/x86/kernel/cpu/amd_64.c
index 958526d6a74a..bd182b7616ee 100644
--- a/arch/x86/kernel/cpu/amd_64.c
+++ b/arch/x86/kernel/cpu/amd_64.c
@@ -199,10 +199,14 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
                  * Don't do it for gbpages because there seems very little
                  * benefit in doing so.
                  */
-                if (!rdmsrl_safe(MSR_K8_TSEG_ADDR, &tseg) &&
-                    (tseg >> PMD_SHIFT) <
-                        (max_pfn_mapped >> (PMD_SHIFT-PAGE_SHIFT)))
+                if (!rdmsrl_safe(MSR_K8_TSEG_ADDR, &tseg)) {
+                    if ((tseg>>PMD_SHIFT) <
+                                (max_low_pfn_mapped>>(PMD_SHIFT-PAGE_SHIFT)) ||
+                        ((tseg>>PMD_SHIFT) <
+                                (max_pfn_mapped>>(PMD_SHIFT-PAGE_SHIFT)) &&
+                         (tseg>>PMD_SHIFT) >= (1ULL<<(32 - PMD_SHIFT))))
                         set_memory_4k((unsigned long)__va(tseg), 1);
+                }
         }
 }
 
diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
index 3451e0b3f324..9f5002e0b35c 100644
--- a/arch/x86/kernel/e820.c
+++ b/arch/x86/kernel/e820.c
@@ -1056,7 +1056,7 @@ unsigned long __initdata end_user_pfn = MAX_ARCH_PFN;
 /*
  * Find the highest page frame number we have available
  */
-unsigned long __init e820_end(void)
+static unsigned long __init e820_end_pfn(unsigned long limit_pfn, unsigned type)
 {
         int i;
         unsigned long last_pfn = 0;
@@ -1064,12 +1064,21 @@ unsigned long __init e820_end(void)
 
         for (i = 0; i < e820.nr_map; i++) {
                 struct e820entry *ei = &e820.map[i];
+                unsigned long start_pfn;
                 unsigned long end_pfn;
 
-                if (ei->type != E820_RAM)
+                if (ei->type != type)
                         continue;
 
+                start_pfn = ei->addr >> PAGE_SHIFT;
                 end_pfn = (ei->addr + ei->size) >> PAGE_SHIFT;
+
+                if (start_pfn >= limit_pfn)
+                        continue;
+                if (end_pfn > limit_pfn) {
+                        last_pfn = limit_pfn;
+                        break;
+                }
                 if (end_pfn > last_pfn)
                         last_pfn = end_pfn;
         }
@@ -1083,7 +1092,15 @@ unsigned long __init e820_end(void)
                          last_pfn, max_arch_pfn);
         return last_pfn;
 }
+unsigned long __init e820_end_of_ram_pfn(void)
+{
+        return e820_end_pfn(MAX_ARCH_PFN, E820_RAM);
+}
 
+unsigned long __init e820_end_of_low_ram_pfn(void)
+{
+        return e820_end_pfn(1UL<<(32 - PAGE_SHIFT), E820_RAM);
+}
 /*
  * Finds an active region in the address range from start_pfn to last_pfn and
  * returns its range in ei_startpfn and ei_endpfn for the e820 entry.
@@ -1206,7 +1223,7 @@ static int __init parse_memmap_opt(char *p)
                  * the real mem size before original memory map is
                  * reset.
                  */
-                saved_max_pfn = e820_end();
+                saved_max_pfn = e820_end_of_ram_pfn();
 #endif
                 e820.nr_map = 0;
                 userdef = 1;
diff --git a/arch/x86/kernel/efi.c b/arch/x86/kernel/efi.c
index 94382faeadb6..06cc8d4254b1 100644
--- a/arch/x86/kernel/efi.c
+++ b/arch/x86/kernel/efi.c
@@ -473,7 +473,7 @@ void __init efi_enter_virtual_mode(void)
                 size = md->num_pages << EFI_PAGE_SHIFT;
                 end = md->phys_addr + size;
 
-                if (PFN_UP(end) <= max_pfn_mapped)
+                if (PFN_UP(end) <= max_low_pfn_mapped)
                         va = __va(md->phys_addr);
                 else
                         va = efi_ioremap(md->phys_addr, size);
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index a7c3471ea17c..86fc2d624270 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -713,14 +713,14 @@ void __init setup_arch(char **cmdline_p)
          * partially used pages are not usable - thus
          * we are rounding upwards:
          */
-        max_pfn = e820_end();
+        max_pfn = e820_end_of_ram_pfn();
 
         /* preallocate 4k for mptable mpc */
         early_reserve_e820_mpc_new();
         /* update e820 for memory not covered by WB MTRRs */
         mtrr_bp_init();
         if (mtrr_trim_uncached_memory(max_pfn))
-                max_pfn = e820_end();
+                max_pfn = e820_end_of_ram_pfn();
 
 #ifdef CONFIG_X86_32
         /* max_low_pfn get updated here */
@@ -732,12 +732,26 @@ void __init setup_arch(char **cmdline_p)
 
         /* How many end-of-memory variables you have, grandma! */
         /* need this before calling reserve_initrd */
-        max_low_pfn = max_pfn;
+        if (max_pfn > (1UL<<(32 - PAGE_SHIFT)))
+                max_low_pfn = e820_end_of_low_ram_pfn();
+        else
+                max_low_pfn = max_pfn;
+
         high_memory = (void *)__va(max_pfn * PAGE_SIZE - 1) + 1;
 #endif
 
         /* max_pfn_mapped is updated here */
-        max_pfn_mapped = init_memory_mapping(0, (max_low_pfn << PAGE_SHIFT));
+        max_low_pfn_mapped = init_memory_mapping(0, max_low_pfn<<PAGE_SHIFT);
+        max_pfn_mapped = max_low_pfn_mapped;
+
+#ifdef CONFIG_X86_64
+        if (max_pfn > max_low_pfn) {
+                max_pfn_mapped = init_memory_mapping(1UL<<32,
+                                                     max_pfn<<PAGE_SHIFT);
+                /* can we preseve max_low_pfn ?*/
+                max_low_pfn = max_pfn;
+        }
+#endif
 
         /*
          * NOTE: On x86-32, only from this point on, fixmaps are ready for use.
diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c
index b5a0fd5f4c5f..029e8cffca9e 100644
--- a/arch/x86/mm/init_32.c
+++ b/arch/x86/mm/init_32.c
@@ -50,6 +50,7 @@
 
 unsigned int __VMALLOC_RESERVE = 128 << 20;
 
+unsigned long max_low_pfn_mapped;
 unsigned long max_pfn_mapped;
 
 DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c
index 48548ef7ddf8..122bcef222fc 100644
--- a/arch/x86/mm/init_64.c
+++ b/arch/x86/mm/init_64.c
@@ -53,6 +53,7 @@
  * The direct mapping extends to max_pfn_mapped, so that we can directly access
  * apertures, ACPI and other tables without having to play with fixmaps.
  */
+unsigned long max_low_pfn_mapped;
 unsigned long max_pfn_mapped;
 
 static unsigned long dma_reserve __initdata;
diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index afd40054d157..0389cb8f6b1a 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -536,8 +536,14 @@ static int split_large_page(pte_t *kpte, unsigned long address)
                 set_pte(&pbase[i], pfn_pte(pfn, ref_prot));
 
         if (address >= (unsigned long)__va(0) &&
+                address < (unsigned long)__va(max_low_pfn_mapped << PAGE_SHIFT))
+                split_page_count(level);
+
+#ifdef CONFIG_X86_64
+        if (address >= (unsigned long)__va(1UL<<32) &&
                 address < (unsigned long)__va(max_pfn_mapped << PAGE_SHIFT))
                 split_page_count(level);
+#endif
 
         /*
          * Install the new, split up pagetable. Important details here:
@@ -655,12 +661,21 @@ static int cpa_process_alias(struct cpa_data *cpa)
         if (cpa->pfn > max_pfn_mapped)
                 return 0;
 
+#ifdef CONFIG_X86_64
+        if (cpa->pfn > max_low_pfn_mapped && cpa->pfn < (1UL<<(32-PAGE_SHIFT)))
+                return 0;
+#endif
         /*
          * No need to redo, when the primary call touched the direct
          * mapping already:
          */
-        if (!within(cpa->vaddr, PAGE_OFFSET,
-                    PAGE_OFFSET + (max_pfn_mapped << PAGE_SHIFT))) {
+        if (!(within(cpa->vaddr, PAGE_OFFSET,
+                    PAGE_OFFSET + (max_low_pfn_mapped << PAGE_SHIFT))
+#ifdef CONFIG_X86_64
+                || within(cpa->vaddr, PAGE_OFFSET + (1UL<<32),
+                    PAGE_OFFSET + (max_pfn_mapped << PAGE_SHIFT))
+#endif
+        )) {
 
                 alias_cpa = *cpa;
                 alias_cpa.vaddr = (unsigned long) __va(cpa->pfn << PAGE_SHIFT);
diff --git a/arch/x86/mm/pat.c b/arch/x86/mm/pat.c
index a885a1019b8a..749766c3c5cd 100644
--- a/arch/x86/mm/pat.c
+++ b/arch/x86/mm/pat.c
@@ -449,7 +449,8 @@ int phys_mem_access_prot_allowed(struct file *file, unsigned long pfn,
         if (retval < 0)
                 return 0;
 
-        if (pfn <= max_pfn_mapped &&
+        if (((pfn <= max_low_pfn_mapped) ||
+             (pfn >= (1UL<<(32 - PAGE_SHIFT)) && pfn <= max_pfn_mapped)) &&
             ioremap_change_attr((unsigned long)__va(offset), size, flags) < 0) {
                 free_memtype(offset, offset + size);
                 printk(KERN_INFO
diff --git a/arch/x86/pci/i386.c b/arch/x86/pci/i386.c
index 6ccd7a108cd4..5281e343dd9f 100644
--- a/arch/x86/pci/i386.c
+++ b/arch/x86/pci/i386.c
@@ -334,7 +334,9 @@ int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma,
                 flags = new_flags;
         }
 
-        if (vma->vm_pgoff <= max_pfn_mapped &&
+        if (((vma->vm_pgoff <= max_low_pfn_mapped) ||
+             (vma->vm_pgoff >= (1UL<<(32 - PAGE_SHIFT)) &&
+              vma->vm_pgoff <= max_pfn_mapped)) &&
             ioremap_change_attr((unsigned long)__va(addr), len, flags)) {
                 free_memtype(addr, addr + len);
                 return -EINVAL;
diff --git a/include/asm-x86/e820.h b/include/asm-x86/e820.h
index 78c03d7bf441..33e793e991d0 100644
--- a/include/asm-x86/e820.h
+++ b/include/asm-x86/e820.h
@@ -99,7 +99,8 @@ extern void free_early(u64 start, u64 end);
 extern void early_res_to_bootmem(u64 start, u64 end);
 extern u64 early_reserve_e820(u64 startt, u64 sizet, u64 align);
 
-extern unsigned long e820_end(void);
+extern unsigned long e820_end_of_ram_pfn(void);
+extern unsigned long e820_end_of_low_ram_pfn(void);
 extern int e820_find_active_region(const struct e820entry *ei,
                                   unsigned long start_pfn,
                                   unsigned long last_pfn,
diff --git a/include/asm-x86/page.h b/include/asm-x86/page.h
index b52ed85f32f5..28d7b4533b1a 100644
--- a/include/asm-x86/page.h
+++ b/include/asm-x86/page.h
@@ -61,6 +61,7 @@ extern void map_devmem(unsigned long pfn, unsigned long size,
 extern void unmap_devmem(unsigned long pfn, unsigned long size,
                          pgprot_t vma_prot);
 
+extern unsigned long max_low_pfn_mapped;
 extern unsigned long max_pfn_mapped;
 
 struct page;