]> git.kernelconcepts.de Git - karo-tx-linux.git/blobdiff - mm/filemap.c
Merge tag 'upstream-4.13-rc1' of git://git.infradead.org/linux-ubifs
[karo-tx-linux.git] / mm / filemap.c
index aea58e983a737b66257c897640b9aa62fa777bc0..a49702445ce05beeb8d80b46f0ee57c116986be2 100644 (file)
@@ -239,14 +239,16 @@ void __delete_from_page_cache(struct page *page, void *shadow)
        /* Leave page->index set: truncation lookup relies upon it */
 
        /* hugetlb pages do not participate in page cache accounting. */
-       if (!PageHuge(page))
-               __mod_node_page_state(page_pgdat(page), NR_FILE_PAGES, -nr);
+       if (PageHuge(page))
+               return;
+
+       __mod_node_page_state(page_pgdat(page), NR_FILE_PAGES, -nr);
        if (PageSwapBacked(page)) {
                __mod_node_page_state(page_pgdat(page), NR_SHMEM, -nr);
                if (PageTransHuge(page))
                        __dec_node_page_state(page, NR_SHMEM_THPS);
        } else {
-               VM_BUG_ON_PAGE(PageTransHuge(page) && !PageHuge(page), page);
+               VM_BUG_ON_PAGE(PageTransHuge(page), page);
        }
 
        /*
@@ -309,6 +311,16 @@ int filemap_check_errors(struct address_space *mapping)
 }
 EXPORT_SYMBOL(filemap_check_errors);
 
+static int filemap_check_and_keep_errors(struct address_space *mapping)
+{
+       /* Check for outstanding write errors */
+       if (test_bit(AS_EIO, &mapping->flags))
+               return -EIO;
+       if (test_bit(AS_ENOSPC, &mapping->flags))
+               return -ENOSPC;
+       return 0;
+}
+
 /**
  * __filemap_fdatawrite_range - start writeback on mapping dirty pages in range
  * @mapping:   address space structure to write
@@ -408,17 +420,16 @@ bool filemap_range_has_page(struct address_space *mapping,
 }
 EXPORT_SYMBOL(filemap_range_has_page);
 
-static int __filemap_fdatawait_range(struct address_space *mapping,
+static void __filemap_fdatawait_range(struct address_space *mapping,
                                     loff_t start_byte, loff_t end_byte)
 {
        pgoff_t index = start_byte >> PAGE_SHIFT;
        pgoff_t end = end_byte >> PAGE_SHIFT;
        struct pagevec pvec;
        int nr_pages;
-       int ret = 0;
 
        if (end_byte < start_byte)
-               goto out;
+               return;
 
        pagevec_init(&pvec, 0);
        while ((index <= end) &&
@@ -435,14 +446,11 @@ static int __filemap_fdatawait_range(struct address_space *mapping,
                                continue;
 
                        wait_on_page_writeback(page);
-                       if (TestClearPageError(page))
-                               ret = -EIO;
+                       ClearPageError(page);
                }
                pagevec_release(&pvec);
                cond_resched();
        }
-out:
-       return ret;
 }
 
 /**
@@ -462,14 +470,8 @@ out:
 int filemap_fdatawait_range(struct address_space *mapping, loff_t start_byte,
                            loff_t end_byte)
 {
-       int ret, ret2;
-
-       ret = __filemap_fdatawait_range(mapping, start_byte, end_byte);
-       ret2 = filemap_check_errors(mapping);
-       if (!ret)
-               ret = ret2;
-
-       return ret;
+       __filemap_fdatawait_range(mapping, start_byte, end_byte);
+       return filemap_check_errors(mapping);
 }
 EXPORT_SYMBOL(filemap_fdatawait_range);
 
@@ -485,15 +487,17 @@ EXPORT_SYMBOL(filemap_fdatawait_range);
  * call sites are system-wide / filesystem-wide data flushers: e.g. sync(2),
  * fsfreeze(8)
  */
-void filemap_fdatawait_keep_errors(struct address_space *mapping)
+int filemap_fdatawait_keep_errors(struct address_space *mapping)
 {
        loff_t i_size = i_size_read(mapping->host);
 
        if (i_size == 0)
-               return;
+               return 0;
 
        __filemap_fdatawait_range(mapping, 0, i_size - 1);
+       return filemap_check_and_keep_errors(mapping);
 }
+EXPORT_SYMBOL(filemap_fdatawait_keep_errors);
 
 /**
  * filemap_fdatawait - wait for all under-writeback pages to complete
@@ -535,6 +539,9 @@ int filemap_write_and_wait(struct address_space *mapping)
                        int err2 = filemap_fdatawait(mapping);
                        if (!err)
                                err = err2;
+               } else {
+                       /* Clear any previously stored errors */
+                       filemap_check_errors(mapping);
                }
        } else {
                err = filemap_check_errors(mapping);
@@ -569,6 +576,9 @@ int filemap_write_and_wait_range(struct address_space *mapping,
                                                lstart, lend);
                        if (!err)
                                err = err2;
+               } else {
+                       /* Clear any previously stored errors */
+                       filemap_check_errors(mapping);
                }
        } else {
                err = filemap_check_errors(mapping);
@@ -577,6 +587,90 @@ int filemap_write_and_wait_range(struct address_space *mapping,
 }
 EXPORT_SYMBOL(filemap_write_and_wait_range);
 
+void __filemap_set_wb_err(struct address_space *mapping, int err)
+{
+       errseq_t eseq = __errseq_set(&mapping->wb_err, err);
+
+       trace_filemap_set_wb_err(mapping, eseq);
+}
+EXPORT_SYMBOL(__filemap_set_wb_err);
+
+/**
+ * file_check_and_advance_wb_err - report wb error (if any) that was previously
+ *                                and advance wb_err to current one
+ * @file: struct file on which the error is being reported
+ *
+ * When userland calls fsync (or something like nfsd does the equivalent), we
+ * want to report any writeback errors that occurred since the last fsync (or
+ * since the file was opened if there haven't been any).
+ *
+ * Grab the wb_err from the mapping. If it matches what we have in the file,
+ * then just quickly return 0. The file is all caught up.
+ *
+ * If it doesn't match, then take the mapping value, set the "seen" flag in
+ * it and try to swap it into place. If it works, or another task beat us
+ * to it with the new value, then update the f_wb_err and return the error
+ * portion. The error at this point must be reported via proper channels
+ * (a'la fsync, or NFS COMMIT operation, etc.).
+ *
+ * While we handle mapping->wb_err with atomic operations, the f_wb_err
+ * value is protected by the f_lock since we must ensure that it reflects
+ * the latest value swapped in for this file descriptor.
+ */
+int file_check_and_advance_wb_err(struct file *file)
+{
+       int err = 0;
+       errseq_t old = READ_ONCE(file->f_wb_err);
+       struct address_space *mapping = file->f_mapping;
+
+       /* Locklessly handle the common case where nothing has changed */
+       if (errseq_check(&mapping->wb_err, old)) {
+               /* Something changed, must use slow path */
+               spin_lock(&file->f_lock);
+               old = file->f_wb_err;
+               err = errseq_check_and_advance(&mapping->wb_err,
+                                               &file->f_wb_err);
+               trace_file_check_and_advance_wb_err(file, old);
+               spin_unlock(&file->f_lock);
+       }
+       return err;
+}
+EXPORT_SYMBOL(file_check_and_advance_wb_err);
+
+/**
+ * file_write_and_wait_range - write out & wait on a file range
+ * @file:      file pointing to address_space with pages
+ * @lstart:    offset in bytes where the range starts
+ * @lend:      offset in bytes where the range ends (inclusive)
+ *
+ * Write out and wait upon file offsets lstart->lend, inclusive.
+ *
+ * Note that @lend is inclusive (describes the last byte to be written) so
+ * that this function can be used to write to the very end-of-file (end = -1).
+ *
+ * After writing out and waiting on the data, we check and advance the
+ * f_wb_err cursor to the latest value, and return any errors detected there.
+ */
+int file_write_and_wait_range(struct file *file, loff_t lstart, loff_t lend)
+{
+       int err = 0, err2;
+       struct address_space *mapping = file->f_mapping;
+
+       if ((!dax_mapping(mapping) && mapping->nrpages) ||
+           (dax_mapping(mapping) && mapping->nrexceptional)) {
+               err = __filemap_fdatawrite_range(mapping, lstart, lend,
+                                                WB_SYNC_ALL);
+               /* See comment of filemap_write_and_wait() */
+               if (err != -EIO)
+                       __filemap_fdatawait_range(mapping, lstart, lend);
+       }
+       err2 = file_check_and_advance_wb_err(file);
+       if (!err)
+               err = err2;
+       return err;
+}
+EXPORT_SYMBOL(file_write_and_wait_range);
+
 /**
  * replace_page_cache_page - replace a pagecache page with a new one
  * @old:       page to be replaced
@@ -2265,7 +2359,7 @@ int filemap_fault(struct vm_fault *vmf)
                /* No page in the page cache at all */
                do_sync_mmap_readahead(vmf->vma, ra, file, offset);
                count_vm_event(PGMAJFAULT);
-               mem_cgroup_count_vm_event(vmf->vma->vm_mm, PGMAJFAULT);
+               count_memcg_event_mm(vmf->vma->vm_mm, PGMAJFAULT);
                ret = VM_FAULT_MAJOR;
 retry_find:
                page = find_get_page(mapping, offset);