[PATCH] proc: Make inline name size calculation automatic

[PATCH] proc: Make inline name size calculation automatic

[David Howells]

Make calculation of the size of the inline name in struct proc_dir_entry
automatic, rather than having to manually encode the numbers and failing to
allow for lockdep.

Require a minimum inline name size of 33+1 to allow for names that look
like two hex numbers with a dash between.

Reported-by: Al Viro <viro@ZenIV.linux.org.uk>
Signed-off-by: David Howells <dhowells@redhat.com>
---

 fs/proc/generic.c  |    2 +-
 fs/proc/inode.c    |    5 ++---
 fs/proc/internal.h |   18 ++++++++++++------
 fs/proc/root.c     |    3 +--
 4 files changed, 16 insertions(+), 12 deletions(-)

diff --git a/fs/proc/generic.c b/fs/proc/generic.c
index fa5b28f0a3a3..6ac1c92997ea 100644
--- a/fs/proc/generic.c
+++ b/fs/proc/generic.c
@@ -409,7 +409,7 @@ static struct proc_dir_entry *__proc_create(struct proc_dir_entry **parent,
 	if (!ent)
 		goto out;
 
-	if (qstr.len + 1 <= sizeof(ent->inline_name)) {
+	if (qstr.len + 1 <= SIZEOF_PDE_INLINE_NAME) {
 		ent->name = ent->inline_name;
 	} else {
 		ent->name = kmalloc(qstr.len + 1, GFP_KERNEL);
diff --git a/fs/proc/inode.c b/fs/proc/inode.c
index 7aa86dd65ba8..38155bec4a54 100644
--- a/fs/proc/inode.c
+++ b/fs/proc/inode.c
@@ -104,9 +104,8 @@ void __init proc_init_kmemcache(void)
 		kmem_cache_create("pde_opener", sizeof(struct pde_opener), 0,
 				  SLAB_ACCOUNT|SLAB_PANIC, NULL);
 	proc_dir_entry_cache = kmem_cache_create_usercopy(
-		"proc_dir_entry", sizeof(struct proc_dir_entry), 0, SLAB_PANIC,
-		offsetof(struct proc_dir_entry, inline_name),
-		sizeof_field(struct proc_dir_entry, inline_name), NULL);
+		"proc_dir_entry", SIZEOF_PDE_SLOT, 0, SLAB_PANIC,
+		OFFSETOF_PDE_NAME, SIZEOF_PDE_INLINE_NAME, NULL);
 }
 
 static int proc_show_options(struct seq_file *seq, struct dentry *root)
diff --git a/fs/proc/internal.h b/fs/proc/internal.h
index 0644b29e53b7..539c6566752a 100644
--- a/fs/proc/internal.h
+++ b/fs/proc/internal.h
@@ -62,14 +62,20 @@ struct proc_dir_entry {
 	char *name;
 	umode_t mode;
 	u8 namelen;
-#ifdef CONFIG_64BIT
-#define SIZEOF_PDE_INLINE_NAME	(192-155)
-#else
-#define SIZEOF_PDE_INLINE_NAME	(128-95)
-#endif
-	char inline_name[SIZEOF_PDE_INLINE_NAME];
+	char inline_name[];
 } __randomize_layout;
 
+#define OFFSETOF_PDE_NAME offsetof(struct proc_dir_entry, inline_name)
+#define SIZEOF_PDE_SLOT					\
+	(OFFSETOF_PDE_NAME + 34 <= 64 ? 64 :		\
+	 OFFSETOF_PDE_NAME + 34 <= 128 ? 128 :		\
+	 OFFSETOF_PDE_NAME + 34 <= 192 ? 192 :		\
+	 OFFSETOF_PDE_NAME + 34 <= 256 ? 256 :		\
+	 OFFSETOF_PDE_NAME + 34 <= 512 ? 512 :		\
+	 0)
+
+#define SIZEOF_PDE_INLINE_NAME (SIZEOF_PDE_SLOT - OFFSETOF_PDE_NAME)
+
 extern struct kmem_cache *proc_dir_entry_cache;
 void pde_free(struct proc_dir_entry *pde);
 
diff --git a/fs/proc/root.c b/fs/proc/root.c
index a379edccd880..efbdc08a3c86 100644
--- a/fs/proc/root.c
+++ b/fs/proc/root.c
@@ -288,8 +288,7 @@ struct proc_dir_entry proc_root = {
 	.proc_fops	= &proc_root_operations,
 	.parent		= &proc_root,
 	.subdir		= RB_ROOT,
-	.name		= proc_root.inline_name,
-	.inline_name	= "/proc",
+	.name		= "/proc",
 };
 
 int pid_ns_prepare_proc(struct pid_namespace *ns)

Re: [PATCH] proc: Make inline name size calculation automatic

[Andrew Morton]

On Wed, 13 Jun 2018 19:43:19 +0100 David Howells <dhowells@redhat.com> wrote:

> Make calculation of the size of the inline name in struct proc_dir_entry
> automatic, rather than having to manually encode the numbers and failing to
> allow for lockdep.
> 
> Require a minimum inline name size of 33+1 to allow for names that look
> like two hex numbers with a dash between.

Please cc Alexey on /proc patches.

What are the user-visible runtime effects of this change?

Re: [PATCH] proc: Make inline name size calculation automatic

[David Howells]

Andrew Morton <akpm@linux-foundation.org> wrote:

> Please cc Alexey on /proc patches.

If Alexey is responsible for procfs he should declare this in MAINTAINERS.

> What are the user-visible runtime effects of this change?

Memory usage only.

David

Re: [PATCH] proc: Make inline name size calculation automatic

[Andrew Morton]

On Tue, 19 Jun 2018 09:01:01 +0100 David Howells <dhowells@redhat.com> wrote:

> Andrew Morton <akpm@linux-foundation.org> wrote:
> 
> > Please cc Alexey on /proc patches.
> 
> If Alexey is responsible for procfs he should declare this in MAINTAINERS.

oop.  I thought we did that but it isn't there.

> > What are the user-visible runtime effects of this change?
> 
> Memory usage only.

Please always include such info in changelogs.

Re: [PATCH] proc: Make inline name size calculation automatic

[Alexey Dobriyan]

> Require a minimum inline name size of 33+1 to allow for names that look
> like two hex numbers with a dash between.

Hi, David.

Why 34? /proc will fallback to separate allocation for name anyway.

I sent nearly identical patch earlier.
https://marc.info/?l=linux-kernel&m=152667374404900&w=4

If you compare, the differences are:

* no BUILD_BUG_ON,
* 64 bytes is too litle even on 32-bit,
* 512 bytes is probably too much even on 64-bit.

> -	.name		= proc_root.inline_name,
> -	.inline_name	= "/proc",
> +	.name		= "/proc",

This will "uninline" 5 bytes wasting space.

Re: [PATCH] proc: Make inline name size calculation automatic

[David Howells]

Alexey Dobriyan <adobriyan@gmail.com> wrote:

> > Require a minimum inline name size of 33+1 to allow for names that look
> > like two hex numbers with a dash between.
> 
> Why 34? /proc will fallback to separate allocation for name anyway.

See above comment.  I ran find on /proc and there were a bunch of files whose
names were "<hex>-<hex>".  Allow for 16-char hex addresses and add a NUL char
to that and you get 34.

David

Re: [PATCH] proc: Make inline name size calculation automatic

[Alexey Dobriyan]

On Thu, Jun 14, 2018 at 09:30:42PM +0100, David Howells wrote:
> Alexey Dobriyan <adobriyan@gmail.com> wrote:
> 
> > > Require a minimum inline name size of 33+1 to allow for names that look
> > > like two hex numbers with a dash between.
> > 
> > Why 34? /proc will fallback to separate allocation for name anyway.
> 
> See above comment.  I ran find on /proc and there were a bunch of files whose
> names were "<hex>-<hex>".  Allow for 16-char hex addresses and add a NUL char
> to that and you get 34.

Those must be /proc/*/map_files symlinks. If that's the case,
they don't have PDEs allocated.