diff options
| author | Bart Oldeman <bartoldeman@users.sourceforge.net> | 2005-11-06 12:54:07 +1300 |
|---|---|---|
| committer | Linus Torvalds <torvalds@g5.osdl.org> | 2005-11-05 16:31:36 -0800 |
| commit | f912696ab330bf539231d1f8032320f2a08b850f (patch) | |
| tree | d3e4cf70323348b341b894af57a69896c755cac1 | |
| parent | 88dcb6c4113afe93a6c4891ec43bef6a17c83155 (diff) | |
| download | linux-f912696ab330bf539231d1f8032320f2a08b850f.tar.gz linux-f912696ab330bf539231d1f8032320f2a08b850f.tar.bz2 linux-f912696ab330bf539231d1f8032320f2a08b850f.zip | |
[PATCH] reset tss->io_bitmap_owner in sys_ioperm()
my patch "x86: initialise tss->io_bitmap_owner to something" (commit ID
d5cd4aadd3d220afac8e3e6d922e333592551f7d) introduced a problem with a
program (DOSEMU) that called ioperm after already doing some port i/o.
The problem is that a process switch return causes tss->io_bitmap_base
to be set to IO_BITMAP_OFFSET so that the fault (that *really* sets the
io bitmap) never triggers.
This fixes that regression.
Signed-off-by: Bart Oldeman <bartoldeman@users.sourceforge.net>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
| -rw-r--r-- | arch/i386/kernel/ioport.c | 3 |
1 files changed, 3 insertions, 0 deletions
diff --git a/arch/i386/kernel/ioport.c b/arch/i386/kernel/ioport.c index f2b37654777f..b59a34dbe262 100644 --- a/arch/i386/kernel/ioport.c +++ b/arch/i386/kernel/ioport.c @@ -108,8 +108,11 @@ asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int turn_on) /* * Sets the lazy trigger so that the next I/O operation will * reload the correct bitmap. + * Reset the owner so that a process switch will not set + * tss->io_bitmap_base to IO_BITMAP_OFFSET. */ tss->io_bitmap_base = INVALID_IO_BITMAP_OFFSET_LAZY; + tss->io_bitmap_owner = NULL; put_cpu(); |