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]
Date:	Tue, 16 Sep 2008 14:57:16 +0000 (GMT)
From:	Michael Trimarchi <trimarchimichael@...oo.it>
To:	Nick Piggin <nickpiggin@...oo.com.au>
Cc:	Hisashi Hifumi <hifumi.hisashi@....ntt.co.jp>,
	akpm@...ux-foundation.org, linux-kernel@...r.kernel.org,
	linux-fsdevel@...r.kernel.org
Subject: Re: [PATCH] VFS: make file->f_pos access atomic on 32bit arch

Hi,

----- Messaggio originale -----
> Da: Nick Piggin <nickpiggin@...oo.com.au>
> A: Michael Trimarchi <trimarchimichael@...oo.it>
>
Cc: Hisashi Hifumi <hifumi.hisashi@....ntt.co.jp>;
akpm@...ux-foundation.org; linux-kernel@...r.kernel.org;
linux-fsdevel@...r.kernel.org
> Inviato: Martedì 16 settembre 2008, 16:11:21
> Oggetto: Re: [PATCH] VFS: make file->f_pos access atomic on 32bit arch
> 
...
> >  i was thinking about introducing a file_pos_update() not implemented using
> > file_pos_read()/file_pos_write() and taking the seq_lock or
> > disabling preemption only one time
> 
> AFAIK, the only way to do an atomic 64 bit store on 32-bit x86 is
> "cmpxchg8b" (with lock prefix on SMP). That would be far slower
> I'm sure.

Maybe it is clear with an example:
 
+/*
+ * file_pos_read/write is atomic by using sequence lock on 32bit arch.
+ */
+static inline loff_t file_pos_read(struct file *file)
+{
+#if BITS_PER_LONG == 32 && defined(CONFIG_SMP)
+	loff_t pos;
+	unsigned int seq;
+
+	do {
+		seq = read_seqbegin(&file->f_pos_seqlock);
+		pos = file->f_pos;
+	} while (read_seqretry(&file->f_pos_seqlock, seq));
+	return pos;
+#elif BITS_PER_LONG == 32 && defined(CONFIG_PREEMPT)
+	loff_t pos;
+
+	preempt_disable();
+	pos = file->f_pos;
+	preempt_enable();
+	return pos;
+#else
+	return file->f_pos;
+#endif
+}
+
+static inline void file_pos_write(struct file *file, loff_t pos)
+{
+#if BITS_PER_LONG == 32 && defined(CONFIG_SMP)
+	write_seqlock(&file->f_pos_seqlock);
+	file->f_pos = pos;
+	write_sequnlock(&file->f_pos_seqlock);
+#elif BITS_PER_LONG == 32 && defined(CONFIG_PREEMPT)
+	preempt_disable();
+	file->f_pos = pos;
+	preempt_enable();
+#else
+	file->f_pos = pos;
+#endif
+}
+
if (offset != file_pos_read(file))
                 file_pos_write(file, offset);

Compile with:
BITS_PER_LONG=32
CONFIG_PREEMPT

The code does:

file_pos_read(...)
preempt_disable
....
preempt_enable

file_pos_write(...)
preempt_disable
....
preempt_enable

with the file_pos_update() the code does:

preempt_disable
...
...
preempt_enable

Regards Michael

__________________________________________________
Do You Yahoo!?
Poco spazio e tanto spam? Yahoo! Mail ti protegge dallo spam e ti da tanto spazio gratuito per i tuoi file e i messaggi 
http://mail.yahoo.it 
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ