lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <1548927785-40397-1-git-send-email-ayush.m@samsung.com>
Date:   Thu, 31 Jan 2019 15:13:05 +0530
From:   Ayush Mittal <ayush.m@...sung.com>
To:     gregkh@...uxfoundation.org, tj@...nel.org,
        linux-kernel@...r.kernel.org
Cc:     pankaj.m@...sung.com, a.sahrawat@...sung.com, v.narang@...sung.com,
        Ayush Mittal <ayush.m@...sung.com>
Subject: [PATCH 1/1] kernfs: Allocating memory for kernfs_iattrs with
 kmem_cache.

This patch creates a new cache for kernfs_iattrs.
Currently, memory is allocated with kzalloc() which always gives aligned memory.
On ARM, this is 64 aligned.
To avoid the wastage of memory in aligning the size requested,
a new cache for kernfs_iattrs is created.

Size of struct kernfs_iattrs is 80 Bytes.
On ARM, it will come in kmalloc-128 slab.
and it will come in kmalloc-192 slab if debug info is enabled.
Extra bytes taken 48 bytes.
Total number of objects created : 4096
Total saving = 48*4096 = 192 KB

After creating new slab(When debug info is enabled) :
sh-3.2# cat /proc/slabinfo
...
kernfs_iattrs_cache   4069   4096    128   32    1 : tunables    0    0    0 : slabdata    128    128      0
...

All testing has been done on ARM taregt.

Signed-off-by: Ayush Mittal <ayush.m@...sung.com>
Signed-off-by: Vaneet Narang <v.narang@...sung.com>
---
 fs/kernfs/dir.c             | 2 +-
 fs/kernfs/inode.c           | 2 +-
 fs/kernfs/kernfs-internal.h | 2 +-
 fs/kernfs/mount.c           | 7 ++++++-
 4 files changed, 9 insertions(+), 4 deletions(-)

diff --git a/fs/kernfs/dir.c b/fs/kernfs/dir.c
index 4ca0b5c..b84d635 100644
--- a/fs/kernfs/dir.c
+++ b/fs/kernfs/dir.c
@@ -536,8 +536,8 @@ void kernfs_put(struct kernfs_node *kn)
 			security_release_secctx(kn->iattr->ia_secdata,
 						kn->iattr->ia_secdata_len);
 		simple_xattrs_free(&kn->iattr->xattrs);
+		kmem_cache_free(kernfs_iattrs_cache, kn->iattr);
 	}
-	kfree(kn->iattr);
 	spin_lock(&kernfs_idr_lock);
 	idr_remove(&root->ino_idr, kn->id.ino);
 	spin_unlock(&kernfs_idr_lock);
diff --git a/fs/kernfs/inode.c b/fs/kernfs/inode.c
index 80cebcd..0c1fd94 100644
--- a/fs/kernfs/inode.c
+++ b/fs/kernfs/inode.c
@@ -42,7 +42,7 @@ static struct kernfs_iattrs *kernfs_iattrs(struct kernfs_node *kn)
 	if (kn->iattr)
 		goto out_unlock;
 
-	kn->iattr = kzalloc(sizeof(struct kernfs_iattrs), GFP_KERNEL);
+	kn->iattr = kmem_cache_zalloc(kernfs_iattrs_cache, GFP_KERNEL);
 	if (!kn->iattr)
 		goto out_unlock;
 	iattrs = &kn->iattr->ia_iattr;
diff --git a/fs/kernfs/kernfs-internal.h b/fs/kernfs/kernfs-internal.h
index 3d83b11..dba810c 100644
--- a/fs/kernfs/kernfs-internal.h
+++ b/fs/kernfs/kernfs-internal.h
@@ -78,7 +78,7 @@ static inline struct kernfs_node *kernfs_dentry_node(struct dentry *dentry)
 }
 
 extern const struct super_operations kernfs_sops;
-extern struct kmem_cache *kernfs_node_cache;
+extern struct kmem_cache *kernfs_node_cache, *kernfs_iattrs_cache;
 
 /*
  * inode.c
diff --git a/fs/kernfs/mount.c b/fs/kernfs/mount.c
index ff2716f..e2ff694 100644
--- a/fs/kernfs/mount.c
+++ b/fs/kernfs/mount.c
@@ -20,7 +20,7 @@
 
 #include "kernfs-internal.h"
 
-struct kmem_cache *kernfs_node_cache;
+struct kmem_cache *kernfs_node_cache, *kernfs_iattrs_cache;
 
 static int kernfs_sop_remount_fs(struct super_block *sb, int *flags, char *data)
 {
@@ -414,4 +414,9 @@ void __init kernfs_init(void)
 					      0,
 					      SLAB_PANIC | SLAB_TYPESAFE_BY_RCU,
 					      NULL);
+	kernfs_iattrs_cache  = kmem_cache_create("kernfs_iattrs_cache",
+					      sizeof(struct kernfs_iattrs),
+					      0,
+					      SLAB_PANIC | SLAB_TYPESAFE_BY_RCU,
+					      NULL);
 }
-- 
1.9.1

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ