staging/lustre/llite: move /proc/fs/lustre/llite/max_read_ahead_mb to sysfs
authorOleg Drokin <green@linuxhacker.ru>
Thu, 21 May 2015 19:31:44 +0000 (15:31 -0400)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Sun, 31 May 2015 01:48:03 +0000 (10:48 +0900)
Move max_read_ahead_mb file from /proc/fs/lustre/llite/*
to /sys/fs/lustre/llite/*/

Signed-off-by: Oleg Drokin <green@linuxhacker.ru>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drivers/staging/lustre/lustre/llite/lproc_llite.c
drivers/staging/lustre/sysfs-fs-lustre

index deee702fa33c4521bfebdfe67b0110b8a9edb128..5baffc1c31af7631032525a25eb2de0a17085b06 100644 (file)
@@ -220,10 +220,11 @@ static int ll_site_stats_seq_show(struct seq_file *m, void *v)
 }
 LPROC_SEQ_FOPS_RO(ll_site_stats);
 
-static int ll_max_readahead_mb_seq_show(struct seq_file *m, void *v)
+static ssize_t max_read_ahead_mb_show(struct kobject *kobj,
+                                     struct attribute *attr, char *buf)
 {
-       struct super_block *sb = m->private;
-       struct ll_sb_info *sbi = ll_s2sbi(sb);
+       struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info,
+                                             ll_kobj);
        long pages_number;
        int mult;
 
@@ -232,23 +233,27 @@ static int ll_max_readahead_mb_seq_show(struct seq_file *m, void *v)
        spin_unlock(&sbi->ll_lock);
 
        mult = 1 << (20 - PAGE_CACHE_SHIFT);
-       return lprocfs_seq_read_frac_helper(m, pages_number, mult);
+       return lprocfs_read_frac_helper(buf, PAGE_SIZE, pages_number, mult);
 }
 
-static ssize_t ll_max_readahead_mb_seq_write(struct file *file,
-                                            const char __user *buffer,
-                                            size_t count, loff_t *off)
+static ssize_t max_read_ahead_mb_store(struct kobject *kobj,
+                                      struct attribute *attr,
+                                      const char *buffer,
+                                      size_t count)
 {
-       struct super_block *sb = ((struct seq_file *)file->private_data)->private;
-       struct ll_sb_info *sbi = ll_s2sbi(sb);
-       int mult, rc, pages_number;
+       struct ll_sb_info *sbi = container_of(kobj, struct ll_sb_info,
+                                             ll_kobj);
+       int rc;
+       unsigned long pages_number;
 
-       mult = 1 << (20 - PAGE_CACHE_SHIFT);
-       rc = lprocfs_write_frac_helper(buffer, count, &pages_number, mult);
+       rc = kstrtoul(buffer, 10, &pages_number);
        if (rc)
                return rc;
 
-       if (pages_number < 0 || pages_number > totalram_pages / 2) {
+       pages_number *= 1 << (20 - PAGE_CACHE_SHIFT); /* MB -> pages */
+
+       if (pages_number > totalram_pages / 2) {
+
                CERROR("can't set file readahead more than %lu MB\n",
                       totalram_pages >> (20 - PAGE_CACHE_SHIFT + 1)); /*1/2 of RAM*/
                return -ERANGE;
@@ -260,7 +265,7 @@ static ssize_t ll_max_readahead_mb_seq_write(struct file *file,
 
        return count;
 }
-LPROC_SEQ_FOPS(ll_max_readahead_mb);
+LUSTRE_RW_ATTR(max_read_ahead_mb);
 
 static int ll_max_readahead_per_file_mb_seq_show(struct seq_file *m, void *v)
 {
@@ -840,7 +845,6 @@ static struct lprocfs_vars lprocfs_llite_obd_vars[] = {
        /* { "mntpt_path",   ll_rd_path,             0, 0 }, */
        { "site",         &ll_site_stats_fops,    NULL, 0 },
        /* { "filegroups",   lprocfs_rd_filegroups,  0, 0 }, */
-       { "max_read_ahead_mb", &ll_max_readahead_mb_fops, NULL },
        { "max_read_ahead_per_file_mb", &ll_max_readahead_per_file_mb_fops,
                NULL },
        { "max_read_ahead_whole_mb", &ll_max_read_ahead_whole_mb_fops, NULL },
@@ -875,6 +879,7 @@ static struct attribute *llite_attrs[] = {
        &lustre_attr_client_type.attr,
        &lustre_attr_fstype.attr,
        &lustre_attr_uuid.attr,
+       &lustre_attr_max_read_ahead_mb.attr,
        NULL,
 };
 
index f353d129ea81784b259c6cbc8b2d02a6aeb8a472..fa32197f656be3e3d72ab5eac2a0d043afbb7f38 100644 (file)
@@ -97,3 +97,10 @@ Date:                May 2015
 Contact:       "Oleg Drokin" <oleg.drokin@intel.com>
 Description:
                Shows this filesystem superblock uuid
+
+What:          /sys/fs/lustre/llite/<fsname>-<uuid>/max_read_ahead_mb
+Date:          May 2015
+Contact:       "Oleg Drokin" <oleg.drokin@intel.com>
+Description:
+               Sets maximum number of megabytes in system memory to be
+               given to read-ahead cache.