*
* You should have received a copy of the GNU General Public License
* version 2 along with this program; If not, see
- * http://www.sun.com/software/products/lustre/docs/GPLv2.pdf
- *
- * Please contact Sun Microsystems, Inc., 4150 Network Circle, Santa Clara,
- * CA 95054 USA or visit www.sun.com if you need additional information or
- * have any questions.
+ * http://www.gnu.org/licenses/gpl-2.0.html
*
* GPL HEADER END
*/
* Copyright (c) 2004, 2010, Oracle and/or its affiliates. All rights reserved.
* Use is subject to license terms.
*
- * Copyright (c) 2011, 2015, Intel Corporation.
+ * Copyright (c) 2011, 2016, Intel Corporation.
*/
/*
* This file is part of Lustre, http://www.lustre.org/
* Lustre is a trademark of Sun Microsystems, Inc.
*/
-#include <linux/kernel.h>
-#include <linux/mm.h>
-#include <linux/string.h>
-#include <linux/stat.h>
#include <linux/errno.h>
-#include <linux/unistd.h>
-#include <linux/version.h>
-#include <asm/uaccess.h>
-
-#include <linux/fs.h>
-#include <linux/stat.h>
-#include <asm/uaccess.h>
+#include <linux/delay.h>
+#include <linux/kernel.h>
#include <linux/mm.h>
-#include <linux/pagemap.h>
#define DEBUG_SUBSYSTEM S_LLITE
unsigned long addr, size_t count)
{
policy->l_extent.start = ((addr - vma->vm_start) & PAGE_MASK) +
- (vma->vm_pgoff << PAGE_CACHE_SHIFT);
+ (vma->vm_pgoff << PAGE_SHIFT);
policy->l_extent.end = (policy->l_extent.start + count - 1) |
~PAGE_MASK;
}
rc = cl_io_init(env, io, CIT_FAULT, io->ci_obj);
if (rc == 0) {
struct vvp_io *vio = vvp_env_io(env);
- struct ll_file_data *fd = LUSTRE_FPRIVATE(file);
+ struct ll_file_data *fd = file->private_data;
LASSERT(vio->vui_cl.cis_io == io);
int result;
__u16 refcheck;
sigset_t set;
- struct inode *inode;
+ struct inode *inode = NULL;
struct ll_inode_info *lli;
ENTRY;
CDEBUG(D_MMAP, "%s mkwrite with %d\n", current->comm, result);
LASSERT(ergo(result == 0, PageLocked(vmpage)));
+ /* if page has been unmapped, presumably due to lock reclaim for
+ * concurrent usage, add some delay before retrying to prevent
+ * entering live-lock situation with competitors
+ */
+ if (result == -ENODATA && inode != NULL) {
+ CDEBUG(D_MMAP, "delaying new page-fault for inode %p to "
+ "prevent live-lock\n", inode);
+ msleep(10);
+ }
+
return result;
}
case 0:
result = VM_FAULT_LOCKED;
break;
- case -EFAULT:
- result = VM_FAULT_NOPAGE;
- break;
case -ENOMEM:
result = VM_FAULT_OOM;
break;
* \retval VM_FAULT_ERROR on general error
* \retval NOPAGE_OOM not have memory for allocate new page
*/
-static int ll_fault0(struct vm_area_struct *vma, struct vm_fault *vmf)
+static vm_fault_t ll_fault0(struct vm_area_struct *vma, struct vm_fault *vmf)
{
struct lu_env *env;
struct cl_io *io;
if (ll_sbi_has_fast_read(ll_i2sbi(file_inode(vma->vm_file)))) {
/* do fast fault */
ll_cl_add(vma->vm_file, env, NULL, LCC_MMAP);
- fault_ret = filemap_fault(vma, vmf);
+ fault_ret = ll_filemap_fault(vma, vmf);
ll_cl_remove(vma->vm_file, env);
/* - If there is no error, then the page was found in cache and
vmpage = vio->u.fault.ft_vmpage;
if (result != 0 && vmpage != NULL) {
- page_cache_release(vmpage);
+ put_page(vmpage);
vmf->page = NULL;
}
}
RETURN(fault_ret);
}
-static int ll_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
+#ifdef HAVE_VM_OPS_USE_VM_FAULT_ONLY
+static vm_fault_t ll_fault(struct vm_fault *vmf)
{
+ struct vm_area_struct *vma = vmf->vma;
+#else
+static vm_fault_t ll_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
+{
+#endif
int count = 0;
bool printed = false;
- int result;
+ bool cached;
+ vm_fault_t result;
+ ktime_t kstart = ktime_get();
sigset_t set;
+ result = pcc_fault(vma, vmf, &cached);
+ if (cached)
+ goto out;
+
/* Only SIGKILL and SIGTERM is allowed for fault/nopage/mkwrite
* so that it can be killed by admin but not cause segfault by
* other signals. */
set = cfs_block_sigsinv(sigmask(SIGKILL) | sigmask(SIGTERM));
- ll_stats_ops_tally(ll_i2sbi(file_inode(vma->vm_file)),
- LPROC_LL_FAULT, 1);
-
+ /* make sure offset is not a negative number */
+ if (vmf->pgoff > (MAX_LFS_FILESIZE >> PAGE_SHIFT))
+ return VM_FAULT_SIGBUS;
restart:
result = ll_fault0(vma, vmf);
- if (!(result & (VM_FAULT_RETRY | VM_FAULT_ERROR | VM_FAULT_LOCKED))) {
- struct page *vmpage = vmf->page;
-
- /* check if this page has been truncated */
- lock_page(vmpage);
- if (unlikely(vmpage->mapping == NULL)) { /* unlucky */
- unlock_page(vmpage);
- page_cache_release(vmpage);
- vmf->page = NULL;
+ if (vmf->page &&
+ !(result & (VM_FAULT_RETRY | VM_FAULT_ERROR | VM_FAULT_LOCKED))) {
+ struct page *vmpage = vmf->page;
+
+ /* check if this page has been truncated */
+ lock_page(vmpage);
+ if (unlikely(vmpage->mapping == NULL)) { /* unlucky */
+ unlock_page(vmpage);
+ put_page(vmpage);
+ vmf->page = NULL;
- if (!printed && ++count > 16) {
- CWARN("the page is under heavy contention,"
- "maybe your app(%s) needs revising :-)\n",
- current->comm);
- printed = true;
- }
+ if (!printed && ++count > 16) {
+ CWARN("the page is under heavy contention, maybe your app(%s) needs revising :-)\n",
+ current->comm);
+ printed = true;
+ }
- goto restart;
- }
+ goto restart;
+ }
- result |= VM_FAULT_LOCKED;
- }
+ result |= VM_FAULT_LOCKED;
+ }
cfs_restore_sigs(set);
- return result;
+
+out:
+ if (vmf->page && result == VM_FAULT_LOCKED) {
+ ll_rw_stats_tally(ll_i2sbi(file_inode(vma->vm_file)),
+ current->pid, vma->vm_file->private_data,
+ cl_offset(NULL, vmf->page->index), PAGE_SIZE,
+ READ);
+ ll_stats_ops_tally(ll_i2sbi(file_inode(vma->vm_file)),
+ LPROC_LL_FAULT,
+ ktime_us_delta(ktime_get(), kstart));
+ }
+
+ return result;
}
-static int ll_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
+#ifdef HAVE_VM_OPS_USE_VM_FAULT_ONLY
+static vm_fault_t ll_page_mkwrite(struct vm_fault *vmf)
+{
+ struct vm_area_struct *vma = vmf->vma;
+#else
+static vm_fault_t ll_page_mkwrite(struct vm_area_struct *vma,
+ struct vm_fault *vmf)
{
+#endif
int count = 0;
bool printed = false;
bool retry;
- int result;
+ bool cached;
+ ktime_t kstart = ktime_get();
+ vm_fault_t result;
- ll_stats_ops_tally(ll_i2sbi(file_inode(vma->vm_file)),
- LPROC_LL_MKWRITE, 1);
+ result = pcc_page_mkwrite(vma, vmf, &cached);
+ if (cached)
+ goto out;
file_update_time(vma->vm_file);
- do {
- retry = false;
- result = ll_page_mkwrite0(vma, vmf->page, &retry);
+ do {
+ retry = false;
+ result = ll_page_mkwrite0(vma, vmf->page, &retry);
- if (!printed && ++count > 16) {
+ if (!printed && ++count > 16) {
const struct dentry *de = file_dentry(vma->vm_file);
- CWARN("app(%s): the page %lu of file "DFID" is under"
- " heavy contention\n",
+ CWARN("app(%s): the page %lu of file "DFID" is under heavy contention\n",
current->comm, vmf->pgoff,
PFID(ll_inode2fid(de->d_inode)));
- printed = true;
- }
- } while (retry);
-
- switch(result) {
- case 0:
- LASSERT(PageLocked(vmf->page));
- result = VM_FAULT_LOCKED;
- break;
- case -ENODATA:
- case -EFAULT:
- result = VM_FAULT_NOPAGE;
- break;
- case -ENOMEM:
- result = VM_FAULT_OOM;
- break;
- case -EAGAIN:
- result = VM_FAULT_RETRY;
- break;
- default:
- result = VM_FAULT_SIGBUS;
- break;
- }
+ printed = true;
+ }
+ } while (retry);
- return result;
+ switch (result) {
+ case 0:
+ LASSERT(PageLocked(vmf->page));
+ result = VM_FAULT_LOCKED;
+ break;
+ case -ENODATA:
+ case -EFAULT:
+ result = VM_FAULT_NOPAGE;
+ break;
+ case -ENOMEM:
+ result = VM_FAULT_OOM;
+ break;
+ case -EAGAIN:
+ result = VM_FAULT_RETRY;
+ break;
+ default:
+ result = VM_FAULT_SIGBUS;
+ break;
+ }
+
+out:
+ if (result == VM_FAULT_LOCKED) {
+ ll_rw_stats_tally(ll_i2sbi(file_inode(vma->vm_file)),
+ current->pid, vma->vm_file->private_data,
+ cl_offset(NULL, vmf->page->index), PAGE_SIZE,
+ WRITE);
+ ll_stats_ops_tally(ll_i2sbi(file_inode(vma->vm_file)),
+ LPROC_LL_MKWRITE,
+ ktime_us_delta(ktime_get(), kstart));
+ }
+
+ return result;
}
/**
ENTRY;
LASSERT(atomic_read(&vob->vob_mmap_cnt) >= 0);
atomic_inc(&vob->vob_mmap_cnt);
+ pcc_vm_open(vma);
EXIT;
}
ENTRY;
atomic_dec(&vob->vob_mmap_cnt);
LASSERT(atomic_read(&vob->vob_mmap_cnt) >= 0);
+ pcc_vm_close(vma);
EXIT;
}
LASSERTF(last > first, "last %llu first %llu\n", last, first);
if (mapping_mapped(mapping)) {
rc = 0;
- unmap_mapping_range(mapping, first + PAGE_CACHE_SIZE - 1,
- last - first + 1, 0);
+ unmap_mapping_range(mapping, first + PAGE_SIZE - 1,
+ last - first + 1, 1);
}
RETURN(rc);
int ll_file_mmap(struct file *file, struct vm_area_struct * vma)
{
struct inode *inode = file_inode(file);
- int rc;
- ENTRY;
+ ktime_t kstart = ktime_get();
+ bool cached;
+ int rc;
- if (ll_file_nolock(file))
- RETURN(-EOPNOTSUPP);
-
- ll_stats_ops_tally(ll_i2sbi(inode), LPROC_LL_MAP, 1);
- rc = generic_file_mmap(file, vma);
- if (rc == 0) {
- vma->vm_ops = &ll_file_vm_ops;
- vma->vm_ops->open(vma);
- /* update the inode's size and mtime */
- rc = ll_glimpse_size(inode);
- }
+ ENTRY;
- RETURN(rc);
+ if (ll_file_nolock(file))
+ RETURN(-EOPNOTSUPP);
+
+ rc = pcc_file_mmap(file, vma, &cached);
+ if (cached && rc != 0)
+ RETURN(rc);
+
+ rc = generic_file_mmap(file, vma);
+ if (rc == 0) {
+ vma->vm_ops = &ll_file_vm_ops;
+ vma->vm_ops->open(vma);
+ /* update the inode's size and mtime */
+ if (!cached)
+ rc = ll_glimpse_size(inode);
+ }
+
+ if (!rc)
+ ll_stats_ops_tally(ll_i2sbi(inode), LPROC_LL_MMAP,
+ ktime_us_delta(ktime_get(), kstart));
+
+ RETURN(rc);
}