linux-kernel - New rwsem

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>] [day] [month] [year] [list]

Message-ID: <45934521495c84b6a6cfbd8919d30877.squirrel@mail.chelnydom.ru>
Date:	Fri, 9 Oct 2009 16:43:20 +0400
From:	vik@...lnydom.ru
To:	linux-kernel@...r.kernel.org
Subject: New rwsem

Next is the rwsem.

static inline void read_lock(unsigned long *rw)
{
  asm volatile("lock xorl ebp, %0\n"
    : "+m" (*rw)
    :
    : "memory", "cc");
}
/* BUG! There should be the write queue. It the one per rwlock */
static inline void read_unlock(unsigned long *rw)
{
  asm volatile("lock xorl ebp, %0\n\t"
    "jne 1f\n\t"
    "call wakeup_writers\n"
    "1:"
    : "+m" (*rw)
    :
    : "memory", "cc");
}
/* It is need another writelocking for interrupt context.
   This should call yield to wait for the read completion */
static inline void write_lock(unsigned long *rw)
{
  asm volatile("1:"
    "lock incl %0\n\t"
    "lock decl %0\n\t"
    "je 2f\n\t"
    "call yield\n\t"
    "jmp 1b\n"
    "2:"
    : "+m" (*rw)
    :
    : "memory", "cc");
}
/* This empty so reliable writelocking is impossible.
   The cure is fault tolerance */
static inline void write_unlock(unsigned long *rw)
{
}

--
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/