From owner-cvs-sys  Sun Jun 21 14:08:05 1998
Return-Path: <owner-cvs-sys>
Received: (from daemon@localhost)
          by hub.freebsd.org (8.8.8/8.8.8) id OAA29690
          for cvs-sys-outgoing; Sun, 21 Jun 1998 14:08:05 -0700 (PDT)
          (envelope-from owner-cvs-sys)
Received: from freefall.freebsd.org (freefall.FreeBSD.ORG [204.216.27.21])
          by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id OAA29594;
          Sun, 21 Jun 1998 14:07:39 -0700 (PDT)
          (envelope-from bde@FreeBSD.org)
From: Bruce Evans <bde@FreeBSD.ORG>
Received: (from bde@localhost)
	by freefall.freebsd.org (8.8.8/8.8.5) id OAA26914;
	Sun, 21 Jun 1998 14:06:04 -0700 (PDT)
Date: Sun, 21 Jun 1998 14:06:04 -0700 (PDT)
Message-Id: <199806212106.OAA26914@freefall.freebsd.org>
To: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-sys@FreeBSD.ORG
Subject: cvs commit: src/sys/gnu/ext2fs ext2_linux_ialloc.c
Sender: owner-cvs-sys@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

bde         1998/06/21 14:06:04 PDT

  Modified files:
    sys/gnu/ext2fs       ext2_linux_ialloc.c 
  Log:
  Fixed (?) races in mark_buffer_dirty().  We abuse the buffer cache
  by hacking on locked buffers without getblk()ing them, and we didn't
  even use splbio() to prevent biodone() changing the buffer underneath
  use when a write completes.  I think there was no problem in practice
  on i386's because the operations on b_flags and numdirtybufs happen to
  be atomic.  We still depend on biodone()'s operations on b_flags not
  interfering with ours.  I think there is only interference for B_ERROR,
  and this is harmless because errors for async writes are ignored anyway.
  
  Don't use mark_buffer_dirty() except for superblock-related metadata.
  It was used in just one case where ordinary BSD buffering is more
  natural.
  
  Revision  Changes    Path
  1.10      +6 -3      src/sys/gnu/ext2fs/ext2_linux_ialloc.c