#include <linux/bootmem.h>
#include <linux/module.h>
#include <linux/sched.h>
-#include <linux/slab.h>
#include <linux/mm.h>
+#include <linux/interrupt.h>
+#include <linux/seq_file.h>
+#include <linux/debugfs.h>
+#include <linux/pfn.h>
+#include <linux/percpu.h>
+#include <linux/gfp.h>
-void clflush_cache_range(void *addr, int size)
-{
- int i;
-
- for (i = 0; i < size; i += boot_cpu_data.x86_clflush_size)
- clflush(addr+i);
-}
-
+#include <asm/e820.h>
#include <asm/processor.h>
#include <asm/tlbflush.h>
#include <asm/sections.h>
+#include <asm/setup.h>
#include <asm/uaccess.h>
#include <asm/pgalloc.h>
+#include <asm/proto.h>
+#include <asm/pat.h>
+
+/*
+ * The current flushing context - we pass it instead of 5 arguments:
+ */
+struct cpa_data {
+ unsigned long *vaddr;
+ pgprot_t mask_set;
+ pgprot_t mask_clr;
+ int numpages;
+ int flags;
+ unsigned long pfn;
+ unsigned force_split : 1;
+ int curpage;
+ struct page **pages;
+};
/*
- * We must allow the BIOS range to be executable:
+ * Serialize cpa() (for !DEBUG_PAGEALLOC which uses large identity mappings)
+ * using cpa_lock. So that we don't allow any other cpu, with stale large tlb
+ * entries change the page attribute in parallel to some other cpu
+ * splitting a large page entry along with changing the attribute.
*/
-#define BIOS_BEGIN 0x000a0000
-#define BIOS_END 0x00100000
+static DEFINE_SPINLOCK(cpa_lock);
+
+#define CPA_FLUSHTLB 1
+#define CPA_ARRAY 2
+#define CPA_PAGES_ARRAY 4
+
+#ifdef CONFIG_PROC_FS
+static unsigned long direct_pages_count[PG_LEVEL_NUM];
+
+void update_page_count(int level, unsigned long pages)
+{
+ unsigned long flags;
+
+ /* Protect against CPA */
+ spin_lock_irqsave(&pgd_lock, flags);
+ direct_pages_count[level] += pages;
+ spin_unlock_irqrestore(&pgd_lock, flags);
+}
+
+static void split_page_count(int level)
+{
+ direct_pages_count[level]--;
+ direct_pages_count[level - 1] += PTRS_PER_PTE;
+}
+
+void arch_report_meminfo(struct seq_file *m)
+{
+ seq_printf(m, "DirectMap4k: %8lu kB\n",
+ direct_pages_count[PG_LEVEL_4K] << 2);
+#if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE)
+ seq_printf(m, "DirectMap2M: %8lu kB\n",
+ direct_pages_count[PG_LEVEL_2M] << 11);
+#else
+ seq_printf(m, "DirectMap4M: %8lu kB\n",
+ direct_pages_count[PG_LEVEL_2M] << 12);
+#endif
+#ifdef CONFIG_X86_64
+ if (direct_gbpages)
+ seq_printf(m, "DirectMap1G: %8lu kB\n",
+ direct_pages_count[PG_LEVEL_1G] << 20);
+#endif
+}
+#else
+static inline void split_page_count(int level) { }
+#endif
+
+#ifdef CONFIG_X86_64
+
+static inline unsigned long highmap_start_pfn(void)
+{
+ return __pa(_text) >> PAGE_SHIFT;
+}
+
+static inline unsigned long highmap_end_pfn(void)
+{
+ return __pa(roundup(_brk_end, PMD_SIZE)) >> PAGE_SHIFT;
+}
+
+#endif
+
+#ifdef CONFIG_DEBUG_PAGEALLOC
+# define debug_pagealloc 1
+#else
+# define debug_pagealloc 0
+#endif
static inline int
within(unsigned long addr, unsigned long start, unsigned long end)
}
/*
+ * Flushing functions
+ */
+
+/**
+ * clflush_cache_range - flush a cache range with clflush
+ * @addr: virtual start address
+ * @size: number of bytes to flush
+ *
+ * clflush is an unordered instruction which needs fencing with mfence
+ * to avoid ordering issues.
+ */
+void clflush_cache_range(void *vaddr, unsigned int size)
+{
+ void *vend = vaddr + size - 1;
+
+ mb();
+
+ for (; vaddr < vend; vaddr += boot_cpu_data.x86_clflush_size)
+ clflush(vaddr);
+ /*
+ * Flush any possible final partial cacheline:
+ */
+ clflush(vend);
+
+ mb();
+}
+EXPORT_SYMBOL_GPL(clflush_cache_range);
+
+static void __cpa_flush_all(void *arg)
+{
+ unsigned long cache = (unsigned long)arg;
+
+ /*
+ * Flush all to work around Errata in early athlons regarding
+ * large page flushing.
+ */
+ __flush_tlb_all();
+
+ if (cache && boot_cpu_data.x86 >= 4)
+ wbinvd();
+}
+
+static void cpa_flush_all(unsigned long cache)
+{
+ BUG_ON(irqs_disabled());
+
+ on_each_cpu(__cpa_flush_all, (void *) cache, 1);
+}
+
+static void __cpa_flush_range(void *arg)
+{
+ /*
+ * We could optimize that further and do individual per page
+ * tlb invalidates for a low number of pages. Caveat: we must
+ * flush the high aliases on 64bit as well.
+ */
+ __flush_tlb_all();
+}
+
+static void cpa_flush_range(unsigned long start, int numpages, int cache)
+{
+ unsigned int i, level;
+ unsigned long addr;
+
+ BUG_ON(irqs_disabled());
+ WARN_ON(PAGE_ALIGN(start) != start);
+
+ on_each_cpu(__cpa_flush_range, NULL, 1);
+
+ if (!cache)
+ return;
+
+ /*
+ * We only need to flush on one CPU,
+ * clflush is a MESI-coherent instruction that
+ * will cause all other CPUs to flush the same
+ * cachelines:
+ */
+ for (i = 0, addr = start; i < numpages; i++, addr += PAGE_SIZE) {
+ pte_t *pte = lookup_address(addr, &level);
+
+ /*
+ * Only flush present addresses:
+ */
+ if (pte && (pte_val(*pte) & _PAGE_PRESENT))
+ clflush_cache_range((void *) addr, PAGE_SIZE);
+ }
+}
+
+static void cpa_flush_array(unsigned long *start, int numpages, int cache,
+ int in_flags, struct page **pages)
+{
+ unsigned int i, level;
+ unsigned long do_wbinvd = cache && numpages >= 1024; /* 4M threshold */
+
+ BUG_ON(irqs_disabled());
+
+ on_each_cpu(__cpa_flush_all, (void *) do_wbinvd, 1);
+
+ if (!cache || do_wbinvd)
+ return;
+
+ /*
+ * We only need to flush on one CPU,
+ * clflush is a MESI-coherent instruction that
+ * will cause all other CPUs to flush the same
+ * cachelines:
+ */
+ for (i = 0; i < numpages; i++) {
+ unsigned long addr;
+ pte_t *pte;
+
+ if (in_flags & CPA_PAGES_ARRAY)
+ addr = (unsigned long)page_address(pages[i]);
+ else
+ addr = start[i];
+
+ pte = lookup_address(addr, &level);
+
+ /*
+ * Only flush present addresses:
+ */
+ if (pte && (pte_val(*pte) & _PAGE_PRESENT))
+ clflush_cache_range((void *)addr, PAGE_SIZE);
+ }
+}
+
+/*
* Certain areas of memory on x86 require very specific protection flags,
* for example the BIOS area or kernel text. Callers don't always get this
* right (again, ioremap() on BIOS memory is not uncommon) so this function
* checks and fixes these known static required protection bits.
*/
-static inline pgprot_t static_protections(pgprot_t prot, unsigned long address)
+static inline pgprot_t static_protections(pgprot_t prot, unsigned long address,
+ unsigned long pfn)
{
pgprot_t forbidden = __pgprot(0);
* The BIOS area between 640k and 1Mb needs to be executable for
* PCI BIOS based config access (CONFIG_PCI_GOBIOS) support.
*/
- if (within(__pa(address), BIOS_BEGIN, BIOS_END))
+ if (within(pfn, BIOS_BEGIN >> PAGE_SHIFT, BIOS_END >> PAGE_SHIFT))
pgprot_val(forbidden) |= _PAGE_NX;
/*
* The kernel text needs to be executable for obvious reasons
- * Does not cover __inittext since that is gone later on
+ * Does not cover __inittext since that is gone later on. On
+ * 64bit we do not enforce !NX on the low mapping
*/
if (within(address, (unsigned long)_text, (unsigned long)_etext))
pgprot_val(forbidden) |= _PAGE_NX;
-#ifdef CONFIG_DEBUG_RODATA
- /* The .rodata section needs to be read-only */
- if (within(address, (unsigned long)__start_rodata,
- (unsigned long)__end_rodata))
+ /*
+ * The .rodata section needs to be read-only. Using the pfn
+ * catches all aliases.
+ */
+ if (within(pfn, __pa((unsigned long)__start_rodata) >> PAGE_SHIFT,
+ __pa((unsigned long)__end_rodata) >> PAGE_SHIFT))
pgprot_val(forbidden) |= _PAGE_RW;
+
+#if defined(CONFIG_X86_64) && defined(CONFIG_DEBUG_RODATA)
+ /*
+ * Once the kernel maps the text as RO (kernel_set_to_readonly is set),
+ * kernel text mappings for the large page aligned text, rodata sections
+ * will be always read-only. For the kernel identity mappings covering
+ * the holes caused by this alignment can be anything that user asks.
+ *
+ * This will preserve the large page mappings for kernel text/data
+ * at no extra cost.
+ */
+ if (kernel_set_to_readonly &&
+ within(address, (unsigned long)_text,
+ (unsigned long)__end_rodata_hpage_align)) {
+ unsigned int level;
+
+ /*
+ * Don't enforce the !RW mapping for the kernel text mapping,
+ * if the current mapping is already using small page mapping.
+ * No need to work hard to preserve large page mappings in this
+ * case.
+ *
+ * This also fixes the Linux Xen paravirt guest boot failure
+ * (because of unexpected read-only mappings for kernel identity
+ * mappings). In this paravirt guest case, the kernel text
+ * mapping and the kernel identity mapping share the same
+ * page-table pages. Thus we can't really use different
+ * protections for the kernel text and identity mappings. Also,
+ * these shared mappings are made of small page mappings.
+ * Thus this don't enforce !RW mapping for small page kernel
+ * text mapping logic will help Linux Xen parvirt guest boot
+ * aswell.
+ */
+ if (lookup_address(address, &level) && (level != PG_LEVEL_4K))
+ pgprot_val(forbidden) |= _PAGE_RW;
+ }
#endif
prot = __pgprot(pgprot_val(prot) & ~pgprot_val(forbidden));
return prot;
}
-pte_t *lookup_address(unsigned long address, int *level)
+/*
+ * Lookup the page table entry for a virtual address. Return a pointer
+ * to the entry and the level of the mapping.
+ *
+ * Note: We return pud and pmd either when the entry is marked large
+ * or when the present bit is not set. Otherwise we would return a
+ * pointer to a nonexisting mapping.
+ */
+pte_t *lookup_address(unsigned long address, unsigned int *level)
{
pgd_t *pgd = pgd_offset_k(address);
pud_t *pud;
if (pgd_none(*pgd))
return NULL;
+
pud = pud_offset(pgd, address);
if (pud_none(*pud))
return NULL;
+
+ *level = PG_LEVEL_1G;
+ if (pud_large(*pud) || !pud_present(*pud))
+ return (pte_t *)pud;
+
pmd = pmd_offset(pud, address);
if (pmd_none(*pmd))
return NULL;
*level = PG_LEVEL_2M;
- if (pmd_large(*pmd))
+ if (pmd_large(*pmd) || !pmd_present(*pmd))
return (pte_t *)pmd;
*level = PG_LEVEL_4K;
+
return pte_offset_kernel(pmd, address);
}
+EXPORT_SYMBOL_GPL(lookup_address);
+/*
+ * Set the new pmd in all the pgds we know about:
+ */
static void __set_pmd_pte(pte_t *kpte, unsigned long address, pte_t pte)
{
/* change init_mm */
if (!SHARED_KERNEL_PMD) {
struct page *page;
- for (page = pgd_list; page; page = (struct page *)page->index) {
+ list_for_each_entry(page, &pgd_list, lru) {
pgd_t *pgd;
pud_t *pud;
pmd_t *pmd;
#endif
}
+static int
+try_preserve_large_page(pte_t *kpte, unsigned long address,
+ struct cpa_data *cpa)
+{
+ unsigned long nextpage_addr, numpages, pmask, psize, flags, addr, pfn;
+ pte_t new_pte, old_pte, *tmp;
+ pgprot_t old_prot, new_prot;
+ int i, do_split = 1;
+ unsigned int level;
+
+ if (cpa->force_split)
+ return 1;
+
+ spin_lock_irqsave(&pgd_lock, flags);
+ /*
+ * Check for races, another CPU might have split this page
+ * up already:
+ */
+ tmp = lookup_address(address, &level);
+ if (tmp != kpte)
+ goto out_unlock;
+
+ switch (level) {
+ case PG_LEVEL_2M:
+ psize = PMD_PAGE_SIZE;
+ pmask = PMD_PAGE_MASK;
+ break;
+#ifdef CONFIG_X86_64
+ case PG_LEVEL_1G:
+ psize = PUD_PAGE_SIZE;
+ pmask = PUD_PAGE_MASK;
+ break;
+#endif
+ default:
+ do_split = -EINVAL;
+ goto out_unlock;
+ }
+
+ /*
+ * Calculate the number of pages, which fit into this large
+ * page starting at address:
+ */
+ nextpage_addr = (address + psize) & pmask;
+ numpages = (nextpage_addr - address) >> PAGE_SHIFT;
+ if (numpages < cpa->numpages)
+ cpa->numpages = numpages;
+
+ /*
+ * We are safe now. Check whether the new pgprot is the same:
+ */
+ old_pte = *kpte;
+ old_prot = new_prot = pte_pgprot(old_pte);
+
+ pgprot_val(new_prot) &= ~pgprot_val(cpa->mask_clr);
+ pgprot_val(new_prot) |= pgprot_val(cpa->mask_set);
+
+ /*
+ * old_pte points to the large page base address. So we need
+ * to add the offset of the virtual address:
+ */
+ pfn = pte_pfn(old_pte) + ((address & (psize - 1)) >> PAGE_SHIFT);
+ cpa->pfn = pfn;
+
+ new_prot = static_protections(new_prot, address, pfn);
+
+ /*
+ * We need to check the full range, whether
+ * static_protection() requires a different pgprot for one of
+ * the pages in the range we try to preserve:
+ */
+ addr = address + PAGE_SIZE;
+ pfn++;
+ for (i = 1; i < cpa->numpages; i++, addr += PAGE_SIZE, pfn++) {
+ pgprot_t chk_prot = static_protections(new_prot, addr, pfn);
+
+ if (pgprot_val(chk_prot) != pgprot_val(new_prot))
+ goto out_unlock;
+ }
+
+ /*
+ * If there are no changes, return. maxpages has been updated
+ * above:
+ */
+ if (pgprot_val(new_prot) == pgprot_val(old_prot)) {
+ do_split = 0;
+ goto out_unlock;
+ }
+
+ /*
+ * We need to change the attributes. Check, whether we can
+ * change the large page in one go. We request a split, when
+ * the address is not aligned and the number of pages is
+ * smaller than the number of pages in the large page. Note
+ * that we limited the number of possible pages already to
+ * the number of pages in the large page.
+ */
+ if (address == (nextpage_addr - psize) && cpa->numpages == numpages) {
+ /*
+ * The address is aligned and the number of pages
+ * covers the full page.
+ */
+ new_pte = pfn_pte(pte_pfn(old_pte), canon_pgprot(new_prot));
+ __set_pmd_pte(kpte, address, new_pte);
+ cpa->flags |= CPA_FLUSHTLB;
+ do_split = 0;
+ }
+
+out_unlock:
+ spin_unlock_irqrestore(&pgd_lock, flags);
+
+ return do_split;
+}
+
static int split_large_page(pte_t *kpte, unsigned long address)
{
- pgprot_t ref_prot = pte_pgprot(pte_clrhuge(*kpte));
- gfp_t gfp_flags = GFP_KERNEL;
- unsigned long flags;
- unsigned long addr;
+ unsigned long flags, pfn, pfninc = 1;
+ unsigned int i, level;
pte_t *pbase, *tmp;
+ pgprot_t ref_prot;
struct page *base;
- int i, level;
-#ifdef CONFIG_DEBUG_PAGEALLOC
- gfp_flags = GFP_ATOMIC;
-#endif
- base = alloc_pages(gfp_flags, 0);
+ if (!debug_pagealloc)
+ spin_unlock(&cpa_lock);
+ base = alloc_pages(GFP_KERNEL | __GFP_NOTRACK, 0);
+ if (!debug_pagealloc)
+ spin_lock(&cpa_lock);
if (!base)
return -ENOMEM;
* up for us already:
*/
tmp = lookup_address(address, &level);
- if (tmp != kpte) {
- WARN_ON_ONCE(1);
+ if (tmp != kpte)
goto out_unlock;
- }
- address = __pa(address);
- addr = address & LARGE_PAGE_MASK;
pbase = (pte_t *)page_address(base);
-#ifdef CONFIG_X86_32
- paravirt_alloc_pt(&init_mm, page_to_pfn(base));
+ paravirt_alloc_pte(&init_mm, page_to_pfn(base));
+ ref_prot = pte_pgprot(pte_clrhuge(*kpte));
+ /*
+ * If we ever want to utilize the PAT bit, we need to
+ * update this function to make sure it's converted from
+ * bit 12 to bit 7 when we cross from the 2MB level to
+ * the 4K level:
+ */
+ WARN_ON_ONCE(pgprot_val(ref_prot) & _PAGE_PAT_LARGE);
+
+#ifdef CONFIG_X86_64
+ if (level == PG_LEVEL_1G) {
+ pfninc = PMD_PAGE_SIZE >> PAGE_SHIFT;
+ pgprot_val(ref_prot) |= _PAGE_PSE;
+ }
#endif
- for (i = 0; i < PTRS_PER_PTE; i++, addr += PAGE_SIZE)
- set_pte(&pbase[i], pfn_pte(addr >> PAGE_SHIFT, ref_prot));
+ /*
+ * Get the target pfn from the original entry:
+ */
+ pfn = pte_pfn(*kpte);
+ for (i = 0; i < PTRS_PER_PTE; i++, pfn += pfninc)
+ 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.
+ *
+ * We use the standard kernel pagetable protections for the new
+ * pagetable protections, the actual ptes set above control the
+ * primary protection behavior:
+ */
+ __set_pmd_pte(kpte, address, mk_pte(base, __pgprot(_KERNPG_TABLE)));
/*
- * Install the new, split up pagetable. Important detail here:
+ * Intel Atom errata AAH41 workaround.
*
- * On Intel the NX bit of all levels must be cleared to make a
- * page executable. See section 4.13.2 of Intel 64 and IA-32
- * Architectures Software Developer's Manual).
+ * The real fix should be in hw or in a microcode update, but
+ * we also probabilistically try to reduce the window of having
+ * a large TLB mixed with 4K TLBs while instruction fetches are
+ * going on.
*/
- ref_prot = pte_pgprot(pte_mkexec(pte_clrhuge(*kpte)));
- __set_pmd_pte(kpte, address, mk_pte(base, ref_prot));
+ __flush_tlb_all();
+
base = NULL;
out_unlock:
- spin_unlock_irqrestore(&pgd_lock, flags);
-
+ /*
+ * If we dropped out via the lookup_address check under
+ * pgd_lock then stick the page back into the pool:
+ */
if (base)
- __free_pages(base, 0);
+ __free_page(base);
+ spin_unlock_irqrestore(&pgd_lock, flags);
return 0;
}
-static int
-__change_page_attr(unsigned long address, unsigned long pfn, pgprot_t prot)
+static int __cpa_process_fault(struct cpa_data *cpa, unsigned long vaddr,
+ int primary)
{
- struct page *kpte_page;
- int level, err = 0;
- pte_t *kpte;
+ /*
+ * Ignore all non primary paths.
+ */
+ if (!primary)
+ return 0;
-#ifdef CONFIG_X86_32
- BUG_ON(pfn > max_low_pfn);
-#endif
+ /*
+ * Ignore the NULL PTE for kernel identity mapping, as it is expected
+ * to have holes.
+ * Also set numpages to '1' indicating that we processed cpa req for
+ * one virtual address page and its pfn. TBD: numpages can be set based
+ * on the initial value and the level returned by lookup_address().
+ */
+ if (within(vaddr, PAGE_OFFSET,
+ PAGE_OFFSET + (max_pfn_mapped << PAGE_SHIFT))) {
+ cpa->numpages = 1;
+ cpa->pfn = __pa(vaddr) >> PAGE_SHIFT;
+ return 0;
+ } else {
+ WARN(1, KERN_WARNING "CPA: called for zero pte. "
+ "vaddr = %lx cpa->vaddr = %lx\n", vaddr,
+ *cpa->vaddr);
+
+ return -EFAULT;
+ }
+}
+static int __change_page_attr(struct cpa_data *cpa, int primary)
+{
+ unsigned long address;
+ int do_split, err;
+ unsigned int level;
+ pte_t *kpte, old_pte;
+
+ if (cpa->flags & CPA_PAGES_ARRAY) {
+ struct page *page = cpa->pages[cpa->curpage];
+ if (unlikely(PageHighMem(page)))
+ return 0;
+ address = (unsigned long)page_address(page);
+ } else if (cpa->flags & CPA_ARRAY)
+ address = cpa->vaddr[cpa->curpage];
+ else
+ address = *cpa->vaddr;
repeat:
kpte = lookup_address(address, &level);
if (!kpte)
- return -EINVAL;
-
- kpte_page = virt_to_page(kpte);
- BUG_ON(PageLRU(kpte_page));
- BUG_ON(PageCompound(kpte_page));
+ return __cpa_process_fault(cpa, address, primary);
- prot = static_protections(prot, address);
+ old_pte = *kpte;
+ if (!pte_val(old_pte))
+ return __cpa_process_fault(cpa, address, primary);
if (level == PG_LEVEL_4K) {
- set_pte_atomic(kpte, pfn_pte(pfn, canon_pgprot(prot)));
- } else {
- err = split_large_page(kpte, address);
- if (!err)
- goto repeat;
+ pte_t new_pte;
+ pgprot_t new_prot = pte_pgprot(old_pte);
+ unsigned long pfn = pte_pfn(old_pte);
+
+ pgprot_val(new_prot) &= ~pgprot_val(cpa->mask_clr);
+ pgprot_val(new_prot) |= pgprot_val(cpa->mask_set);
+
+ new_prot = static_protections(new_prot, address, pfn);
+
+ /*
+ * We need to keep the pfn from the existing PTE,
+ * after all we're only going to change it's attributes
+ * not the memory it points to
+ */
+ new_pte = pfn_pte(pfn, canon_pgprot(new_prot));
+ cpa->pfn = pfn;
+ /*
+ * Do we really change anything ?
+ */
+ if (pte_val(old_pte) != pte_val(new_pte)) {
+ set_pte_atomic(kpte, new_pte);
+ cpa->flags |= CPA_FLUSHTLB;
+ }
+ cpa->numpages = 1;
+ return 0;
}
+
+ /*
+ * Check, whether we can keep the large page intact
+ * and just change the pte:
+ */
+ do_split = try_preserve_large_page(kpte, address, cpa);
+ /*
+ * When the range fits into the existing large page,
+ * return. cp->numpages and cpa->tlbflush have been updated in
+ * try_large_page:
+ */
+ if (do_split <= 0)
+ return do_split;
+
+ /*
+ * We have to split the large page:
+ */
+ err = split_large_page(kpte, address);
+ if (!err) {
+ /*
+ * Do a global flush tlb after splitting the large page
+ * and before we do the actual change page attribute in the PTE.
+ *
+ * With out this, we violate the TLB application note, that says
+ * "The TLBs may contain both ordinary and large-page
+ * translations for a 4-KByte range of linear addresses. This
+ * may occur if software modifies the paging structures so that
+ * the page size used for the address range changes. If the two
+ * translations differ with respect to page frame or attributes
+ * (e.g., permissions), processor behavior is undefined and may
+ * be implementation-specific."
+ *
+ * We do this global tlb flush inside the cpa_lock, so that we
+ * don't allow any other cpu, with stale tlb entries change the
+ * page attribute in parallel, that also falls into the
+ * just split large page entry.
+ */
+ flush_tlb_all();
+ goto repeat;
+ }
+
return err;
}
-/**
- * change_page_attr_addr - Change page table attributes in linear mapping
- * @address: Virtual address in linear mapping.
- * @numpages: Number of pages to change
- * @prot: New page table attribute (PAGE_*)
- *
- * Change page attributes of a page in the direct mapping. This is a variant
- * of change_page_attr() that also works on memory holes that do not have
- * mem_map entry (pfn_valid() is false).
- *
- * See change_page_attr() documentation for more details.
- */
+static int __change_page_attr_set_clr(struct cpa_data *cpa, int checkalias);
-int change_page_attr_addr(unsigned long address, int numpages, pgprot_t prot)
+static int cpa_process_alias(struct cpa_data *cpa)
{
- int err = 0, kernel_map = 0, i;
+ struct cpa_data alias_cpa;
+ unsigned long laddr = (unsigned long)__va(cpa->pfn << PAGE_SHIFT);
+ unsigned long vaddr;
+ int ret;
+
+ if (cpa->pfn >= max_pfn_mapped)
+ return 0;
#ifdef CONFIG_X86_64
- if (address >= __START_KERNEL_map &&
- address < __START_KERNEL_map + KERNEL_TEXT_SIZE) {
+ 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 (cpa->flags & CPA_PAGES_ARRAY) {
+ struct page *page = cpa->pages[cpa->curpage];
+ if (unlikely(PageHighMem(page)))
+ return 0;
+ vaddr = (unsigned long)page_address(page);
+ } else if (cpa->flags & CPA_ARRAY)
+ vaddr = cpa->vaddr[cpa->curpage];
+ else
+ vaddr = *cpa->vaddr;
+
+ if (!(within(vaddr, PAGE_OFFSET,
+ PAGE_OFFSET + (max_pfn_mapped << PAGE_SHIFT)))) {
- address = (unsigned long)__va(__pa(address));
- kernel_map = 1;
+ alias_cpa = *cpa;
+ alias_cpa.vaddr = &laddr;
+ alias_cpa.flags &= ~(CPA_PAGES_ARRAY | CPA_ARRAY);
+
+ ret = __change_page_attr_set_clr(&alias_cpa, 0);
+ if (ret)
+ return ret;
+ }
+
+#ifdef CONFIG_X86_64
+ /*
+ * If the primary call didn't touch the high mapping already
+ * and the physical address is inside the kernel map, we need
+ * to touch the high mapped kernel as well:
+ */
+ if (!within(vaddr, (unsigned long)_text, _brk_end) &&
+ within(cpa->pfn, highmap_start_pfn(), highmap_end_pfn())) {
+ unsigned long temp_cpa_vaddr = (cpa->pfn << PAGE_SHIFT) +
+ __START_KERNEL_map - phys_base;
+ alias_cpa = *cpa;
+ alias_cpa.vaddr = &temp_cpa_vaddr;
+ alias_cpa.flags &= ~(CPA_PAGES_ARRAY | CPA_ARRAY);
+
+ /*
+ * The high mapping range is imprecise, so ignore the
+ * return value.
+ */
+ __change_page_attr_set_clr(&alias_cpa, 0);
}
#endif
- for (i = 0; i < numpages; i++, address += PAGE_SIZE) {
- unsigned long pfn = __pa(address) >> PAGE_SHIFT;
+ return 0;
+}
+
+static int __change_page_attr_set_clr(struct cpa_data *cpa, int checkalias)
+{
+ int ret, numpages = cpa->numpages;
+
+ while (numpages) {
+ /*
+ * Store the remaining nr of pages for the large page
+ * preservation check.
+ */
+ cpa->numpages = numpages;
+ /* for array changes, we can't use large page */
+ if (cpa->flags & (CPA_ARRAY | CPA_PAGES_ARRAY))
+ cpa->numpages = 1;
- if (!kernel_map || pte_present(pfn_pte(0, prot))) {
- err = __change_page_attr(address, pfn, prot);
- if (err)
- break;
+ if (!debug_pagealloc)
+ spin_lock(&cpa_lock);
+ ret = __change_page_attr(cpa, checkalias);
+ if (!debug_pagealloc)
+ spin_unlock(&cpa_lock);
+ if (ret)
+ return ret;
+
+ if (checkalias) {
+ ret = cpa_process_alias(cpa);
+ if (ret)
+ return ret;
}
-#ifdef CONFIG_X86_64
+
/*
- * Handle kernel mapping too which aliases part of
- * lowmem:
+ * Adjust the number of pages with the result of the
+ * CPA operation. Either a large page has been
+ * preserved or a single page update happened.
*/
- if (__pa(address) < KERNEL_TEXT_SIZE) {
- unsigned long addr2;
- pgprot_t prot2;
-
- addr2 = __START_KERNEL_map + __pa(address);
- /* Make sure the kernel mappings stay executable */
- prot2 = pte_pgprot(pte_mkexec(pfn_pte(0, prot)));
- err = __change_page_attr(addr2, pfn, prot2);
+ BUG_ON(cpa->numpages > numpages);
+ numpages -= cpa->numpages;
+ if (cpa->flags & (CPA_PAGES_ARRAY | CPA_ARRAY))
+ cpa->curpage++;
+ else
+ *cpa->vaddr += cpa->numpages * PAGE_SIZE;
+
+ }
+ return 0;
+}
+
+static inline int cache_attr(pgprot_t attr)
+{
+ return pgprot_val(attr) &
+ (_PAGE_PAT | _PAGE_PAT_LARGE | _PAGE_PWT | _PAGE_PCD);
+}
+
+static int change_page_attr_set_clr(unsigned long *addr, int numpages,
+ pgprot_t mask_set, pgprot_t mask_clr,
+ int force_split, int in_flag,
+ struct page **pages)
+{
+ struct cpa_data cpa;
+ int ret, cache, checkalias;
+ unsigned long baddr = 0;
+
+ /*
+ * Check, if we are requested to change a not supported
+ * feature:
+ */
+ mask_set = canon_pgprot(mask_set);
+ mask_clr = canon_pgprot(mask_clr);
+ if (!pgprot_val(mask_set) && !pgprot_val(mask_clr) && !force_split)
+ return 0;
+
+ /* Ensure we are PAGE_SIZE aligned */
+ if (in_flag & CPA_ARRAY) {
+ int i;
+ for (i = 0; i < numpages; i++) {
+ if (addr[i] & ~PAGE_MASK) {
+ addr[i] &= PAGE_MASK;
+ WARN_ON_ONCE(1);
+ }
}
-#endif
+ } else if (!(in_flag & CPA_PAGES_ARRAY)) {
+ /*
+ * in_flag of CPA_PAGES_ARRAY implies it is aligned.
+ * No need to cehck in that case
+ */
+ if (*addr & ~PAGE_MASK) {
+ *addr &= PAGE_MASK;
+ /*
+ * People should not be passing in unaligned addresses:
+ */
+ WARN_ON_ONCE(1);
+ }
+ /*
+ * Save address for cache flush. *addr is modified in the call
+ * to __change_page_attr_set_clr() below.
+ */
+ baddr = *addr;
}
- return err;
+ /* Must avoid aliasing mappings in the highmem code */
+ kmap_flush_unused();
+
+ vm_unmap_aliases();
+
+ cpa.vaddr = addr;
+ cpa.pages = pages;
+ cpa.numpages = numpages;
+ cpa.mask_set = mask_set;
+ cpa.mask_clr = mask_clr;
+ cpa.flags = 0;
+ cpa.curpage = 0;
+ cpa.force_split = force_split;
+
+ if (in_flag & (CPA_ARRAY | CPA_PAGES_ARRAY))
+ cpa.flags |= in_flag;
+
+ /* No alias checking for _NX bit modifications */
+ checkalias = (pgprot_val(mask_set) | pgprot_val(mask_clr)) != _PAGE_NX;
+
+ ret = __change_page_attr_set_clr(&cpa, checkalias);
+
+ /*
+ * Check whether we really changed something:
+ */
+ if (!(cpa.flags & CPA_FLUSHTLB))
+ goto out;
+
+ /*
+ * No need to flush, when we did not set any of the caching
+ * attributes:
+ */
+ cache = cache_attr(mask_set);
+
+ /*
+ * On success we use clflush, when the CPU supports it to
+ * avoid the wbindv. If the CPU does not support it and in the
+ * error case we fall back to cpa_flush_all (which uses
+ * wbindv):
+ */
+ if (!ret && cpu_has_clflush) {
+ if (cpa.flags & (CPA_PAGES_ARRAY | CPA_ARRAY)) {
+ cpa_flush_array(addr, numpages, cache,
+ cpa.flags, pages);
+ } else
+ cpa_flush_range(baddr, numpages, cache);
+ } else
+ cpa_flush_all(cache);
+
+out:
+ return ret;
}
-/**
- * change_page_attr - Change page table attributes in the linear mapping.
- * @page: First page to change
- * @numpages: Number of pages to change
- * @prot: New protection/caching type (PAGE_*)
- *
- * Returns 0 on success, otherwise a negated errno.
- *
- * This should be used when a page is mapped with a different caching policy
- * than write-back somewhere - some CPUs do not like it when mappings with
- * different caching policies exist. This changes the page attributes of the
- * in kernel linear mapping too.
- *
- * Caller must call global_flush_tlb() later to make the changes active.
- *
- * The caller needs to ensure that there are no conflicting mappings elsewhere
- * (e.g. in user space) * This function only deals with the kernel linear map.
- *
- * For MMIO areas without mem_map use change_page_attr_addr() instead.
- */
-int change_page_attr(struct page *page, int numpages, pgprot_t prot)
+static inline int change_page_attr_set(unsigned long *addr, int numpages,
+ pgprot_t mask, int array)
{
- unsigned long addr = (unsigned long)page_address(page);
+ return change_page_attr_set_clr(addr, numpages, mask, __pgprot(0), 0,
+ (array ? CPA_ARRAY : 0), NULL);
+}
+
+static inline int change_page_attr_clear(unsigned long *addr, int numpages,
+ pgprot_t mask, int array)
+{
+ return change_page_attr_set_clr(addr, numpages, __pgprot(0), mask, 0,
+ (array ? CPA_ARRAY : 0), NULL);
+}
- return change_page_attr_addr(addr, numpages, prot);
+static inline int cpa_set_pages_array(struct page **pages, int numpages,
+ pgprot_t mask)
+{
+ return change_page_attr_set_clr(NULL, numpages, mask, __pgprot(0), 0,
+ CPA_PAGES_ARRAY, pages);
}
-EXPORT_SYMBOL(change_page_attr);
-static void flush_kernel_map(void *arg)
+static inline int cpa_clear_pages_array(struct page **pages, int numpages,
+ pgprot_t mask)
+{
+ return change_page_attr_set_clr(NULL, numpages, __pgprot(0), mask, 0,
+ CPA_PAGES_ARRAY, pages);
+}
+
+int _set_memory_uc(unsigned long addr, int numpages)
{
/*
- * Flush all to work around Errata in early athlons regarding
- * large page flushing.
+ * for now UC MINUS. see comments in ioremap_nocache()
*/
- __flush_tlb_all();
+ return change_page_attr_set(&addr, numpages,
+ __pgprot(_PAGE_CACHE_UC_MINUS), 0);
+}
- if (boot_cpu_data.x86_model >= 4)
- wbinvd();
+int set_memory_uc(unsigned long addr, int numpages)
+{
+ int ret;
+
+ /*
+ * for now UC MINUS. see comments in ioremap_nocache()
+ */
+ ret = reserve_memtype(__pa(addr), __pa(addr) + numpages * PAGE_SIZE,
+ _PAGE_CACHE_UC_MINUS, NULL);
+ if (ret)
+ goto out_err;
+
+ ret = _set_memory_uc(addr, numpages);
+ if (ret)
+ goto out_free;
+
+ return 0;
+
+out_free:
+ free_memtype(__pa(addr), __pa(addr) + numpages * PAGE_SIZE);
+out_err:
+ return ret;
}
+EXPORT_SYMBOL(set_memory_uc);
-void global_flush_tlb(void)
+int set_memory_array_uc(unsigned long *addr, int addrinarray)
{
- BUG_ON(irqs_disabled());
+ int i, j;
+ int ret;
- on_each_cpu(flush_kernel_map, NULL, 1, 1);
+ /*
+ * for now UC MINUS. see comments in ioremap_nocache()
+ */
+ for (i = 0; i < addrinarray; i++) {
+ ret = reserve_memtype(__pa(addr[i]), __pa(addr[i]) + PAGE_SIZE,
+ _PAGE_CACHE_UC_MINUS, NULL);
+ if (ret)
+ goto out_free;
+ }
+
+ ret = change_page_attr_set(addr, addrinarray,
+ __pgprot(_PAGE_CACHE_UC_MINUS), 1);
+ if (ret)
+ goto out_free;
+
+ return 0;
+
+out_free:
+ for (j = 0; j < i; j++)
+ free_memtype(__pa(addr[j]), __pa(addr[j]) + PAGE_SIZE);
+
+ return ret;
+}
+EXPORT_SYMBOL(set_memory_array_uc);
+
+int _set_memory_wc(unsigned long addr, int numpages)
+{
+ int ret;
+ unsigned long addr_copy = addr;
+
+ ret = change_page_attr_set(&addr, numpages,
+ __pgprot(_PAGE_CACHE_UC_MINUS), 0);
+ if (!ret) {
+ ret = change_page_attr_set_clr(&addr_copy, numpages,
+ __pgprot(_PAGE_CACHE_WC),
+ __pgprot(_PAGE_CACHE_MASK),
+ 0, 0, NULL);
+ }
+ return ret;
+}
+
+int set_memory_wc(unsigned long addr, int numpages)
+{
+ int ret;
+
+ if (!pat_enabled)
+ return set_memory_uc(addr, numpages);
+
+ ret = reserve_memtype(__pa(addr), __pa(addr) + numpages * PAGE_SIZE,
+ _PAGE_CACHE_WC, NULL);
+ if (ret)
+ goto out_err;
+
+ ret = _set_memory_wc(addr, numpages);
+ if (ret)
+ goto out_free;
+
+ return 0;
+
+out_free:
+ free_memtype(__pa(addr), __pa(addr) + numpages * PAGE_SIZE);
+out_err:
+ return ret;
+}
+EXPORT_SYMBOL(set_memory_wc);
+
+int _set_memory_wb(unsigned long addr, int numpages)
+{
+ return change_page_attr_clear(&addr, numpages,
+ __pgprot(_PAGE_CACHE_MASK), 0);
+}
+
+int set_memory_wb(unsigned long addr, int numpages)
+{
+ int ret;
+
+ ret = _set_memory_wb(addr, numpages);
+ if (ret)
+ return ret;
+
+ free_memtype(__pa(addr), __pa(addr) + numpages * PAGE_SIZE);
+ return 0;
+}
+EXPORT_SYMBOL(set_memory_wb);
+
+int set_memory_array_wb(unsigned long *addr, int addrinarray)
+{
+ int i;
+ int ret;
+
+ ret = change_page_attr_clear(addr, addrinarray,
+ __pgprot(_PAGE_CACHE_MASK), 1);
+ if (ret)
+ return ret;
+
+ for (i = 0; i < addrinarray; i++)
+ free_memtype(__pa(addr[i]), __pa(addr[i]) + PAGE_SIZE);
+
+ return 0;
+}
+EXPORT_SYMBOL(set_memory_array_wb);
+
+int set_memory_x(unsigned long addr, int numpages)
+{
+ if (!(__supported_pte_mask & _PAGE_NX))
+ return 0;
+
+ return change_page_attr_clear(&addr, numpages, __pgprot(_PAGE_NX), 0);
+}
+EXPORT_SYMBOL(set_memory_x);
+
+int set_memory_nx(unsigned long addr, int numpages)
+{
+ if (!(__supported_pte_mask & _PAGE_NX))
+ return 0;
+
+ return change_page_attr_set(&addr, numpages, __pgprot(_PAGE_NX), 0);
+}
+EXPORT_SYMBOL(set_memory_nx);
+
+int set_memory_ro(unsigned long addr, int numpages)
+{
+ return change_page_attr_clear(&addr, numpages, __pgprot(_PAGE_RW), 0);
+}
+EXPORT_SYMBOL_GPL(set_memory_ro);
+
+int set_memory_rw(unsigned long addr, int numpages)
+{
+ return change_page_attr_set(&addr, numpages, __pgprot(_PAGE_RW), 0);
+}
+EXPORT_SYMBOL_GPL(set_memory_rw);
+
+int set_memory_np(unsigned long addr, int numpages)
+{
+ return change_page_attr_clear(&addr, numpages, __pgprot(_PAGE_PRESENT), 0);
+}
+
+int set_memory_4k(unsigned long addr, int numpages)
+{
+ return change_page_attr_set_clr(&addr, numpages, __pgprot(0),
+ __pgprot(0), 1, 0, NULL);
+}
+
+int set_pages_uc(struct page *page, int numpages)
+{
+ unsigned long addr = (unsigned long)page_address(page);
+
+ return set_memory_uc(addr, numpages);
+}
+EXPORT_SYMBOL(set_pages_uc);
+
+int set_pages_array_uc(struct page **pages, int addrinarray)
+{
+ unsigned long start;
+ unsigned long end;
+ int i;
+ int free_idx;
+
+ for (i = 0; i < addrinarray; i++) {
+ if (PageHighMem(pages[i]))
+ continue;
+ start = page_to_pfn(pages[i]) << PAGE_SHIFT;
+ end = start + PAGE_SIZE;
+ if (reserve_memtype(start, end, _PAGE_CACHE_UC_MINUS, NULL))
+ goto err_out;
+ }
+
+ if (cpa_set_pages_array(pages, addrinarray,
+ __pgprot(_PAGE_CACHE_UC_MINUS)) == 0) {
+ return 0; /* Success */
+ }
+err_out:
+ free_idx = i;
+ for (i = 0; i < free_idx; i++) {
+ if (PageHighMem(pages[i]))
+ continue;
+ start = page_to_pfn(pages[i]) << PAGE_SHIFT;
+ end = start + PAGE_SIZE;
+ free_memtype(start, end);
+ }
+ return -EINVAL;
+}
+EXPORT_SYMBOL(set_pages_array_uc);
+
+int set_pages_wb(struct page *page, int numpages)
+{
+ unsigned long addr = (unsigned long)page_address(page);
+
+ return set_memory_wb(addr, numpages);
+}
+EXPORT_SYMBOL(set_pages_wb);
+
+int set_pages_array_wb(struct page **pages, int addrinarray)
+{
+ int retval;
+ unsigned long start;
+ unsigned long end;
+ int i;
+
+ retval = cpa_clear_pages_array(pages, addrinarray,
+ __pgprot(_PAGE_CACHE_MASK));
+ if (retval)
+ return retval;
+
+ for (i = 0; i < addrinarray; i++) {
+ if (PageHighMem(pages[i]))
+ continue;
+ start = page_to_pfn(pages[i]) << PAGE_SHIFT;
+ end = start + PAGE_SIZE;
+ free_memtype(start, end);
+ }
+
+ return 0;
+}
+EXPORT_SYMBOL(set_pages_array_wb);
+
+int set_pages_x(struct page *page, int numpages)
+{
+ unsigned long addr = (unsigned long)page_address(page);
+
+ return set_memory_x(addr, numpages);
+}
+EXPORT_SYMBOL(set_pages_x);
+
+int set_pages_nx(struct page *page, int numpages)
+{
+ unsigned long addr = (unsigned long)page_address(page);
+
+ return set_memory_nx(addr, numpages);
+}
+EXPORT_SYMBOL(set_pages_nx);
+
+int set_pages_ro(struct page *page, int numpages)
+{
+ unsigned long addr = (unsigned long)page_address(page);
+
+ return set_memory_ro(addr, numpages);
+}
+
+int set_pages_rw(struct page *page, int numpages)
+{
+ unsigned long addr = (unsigned long)page_address(page);
+
+ return set_memory_rw(addr, numpages);
}
-EXPORT_SYMBOL(global_flush_tlb);
#ifdef CONFIG_DEBUG_PAGEALLOC
+
+static int __set_pages_p(struct page *page, int numpages)
+{
+ unsigned long tempaddr = (unsigned long) page_address(page);
+ struct cpa_data cpa = { .vaddr = &tempaddr,
+ .numpages = numpages,
+ .mask_set = __pgprot(_PAGE_PRESENT | _PAGE_RW),
+ .mask_clr = __pgprot(0),
+ .flags = 0};
+
+ /*
+ * No alias checking needed for setting present flag. otherwise,
+ * we may need to break large pages for 64-bit kernel text
+ * mappings (this adds to complexity if we want to do this from
+ * atomic context especially). Let's keep it simple!
+ */
+ return __change_page_attr_set_clr(&cpa, 0);
+}
+
+static int __set_pages_np(struct page *page, int numpages)
+{
+ unsigned long tempaddr = (unsigned long) page_address(page);
+ struct cpa_data cpa = { .vaddr = &tempaddr,
+ .numpages = numpages,
+ .mask_set = __pgprot(0),
+ .mask_clr = __pgprot(_PAGE_PRESENT | _PAGE_RW),
+ .flags = 0};
+
+ /*
+ * No alias checking needed for setting not present flag. otherwise,
+ * we may need to break large pages for 64-bit kernel text
+ * mappings (this adds to complexity if we want to do this from
+ * atomic context especially). Let's keep it simple!
+ */
+ return __change_page_attr_set_clr(&cpa, 0);
+}
+
void kernel_map_pages(struct page *page, int numpages, int enable)
{
if (PageHighMem(page))
return;
/*
- * The return value is ignored - the calls cannot fail,
- * large pages are disabled at boot time:
+ * The return value is ignored as the calls cannot fail.
+ * Large pages for identity mappings are not used at boot time
+ * and hence no memory allocations during large page split.
*/
- change_page_attr(page, numpages, enable ? PAGE_KERNEL : __pgprot(0));
+ if (enable)
+ __set_pages_p(page, numpages);
+ else
+ __set_pages_np(page, numpages);
/*
* We should perform an IPI and flush all tlbs,
*/
__flush_tlb_all();
}
+
+#ifdef CONFIG_HIBERNATION
+
+bool kernel_page_present(struct page *page)
+{
+ unsigned int level;
+ pte_t *pte;
+
+ if (PageHighMem(page))
+ return false;
+
+ pte = lookup_address((unsigned long)page_address(page), &level);
+ return (pte_val(*pte) & _PAGE_PRESENT);
+}
+
+#endif /* CONFIG_HIBERNATION */
+
+#endif /* CONFIG_DEBUG_PAGEALLOC */
+
+/*
+ * The testcases use internal knowledge of the implementation that shouldn't
+ * be exposed to the rest of the kernel. Include these directly here.
+ */
+#ifdef CONFIG_CPA_DEBUG
+#include "pageattr-test.c"
#endif
git://ftp.safe.ca / safe / jmp / linux-2.6 / blobdiff