diff options
author | Jeff Dike <jdike@addtoit.com> | 2006-11-03 06:07:22 (GMT) |
---|---|---|
committer | Linus Torvalds <torvalds@g5.osdl.org> | 2006-11-03 20:27:58 (GMT) |
commit | 53b173327d283b9bdbfb0c3b6de6f0eb197819d6 (patch) | |
tree | 0b41508295ec6a7a826ec36f72f77433da460e57 /arch/um/include/sysdep-i386/barrier.h | |
parent | d2c89a4284ea4ecfba77c6f2d7d6f96d52e801e5 (diff) | |
download | linux-53b173327d283b9bdbfb0c3b6de6f0eb197819d6.tar.xz |
[PATCH] uml: fix I/O hang
Fix a UML hang in which everything would just stop until some I/O happened
- a ping, someone whacking the keyboard - at which point everything would
start up again as though nothing had happened.
The cause was gcc reordering some code which absolutely needed to be
executed in the order in the source. When unblock_signals switches signals
from off to on, it needs to see if any interrupts had happened in the
critical section. The interrupt handlers check signals_enabled - if it is
zero, then the handler adds a bit to the "pending" bitmask and returns.
unblock_signals checks this mask to see if any signals need to be
delivered.
The crucial part is this:
signals_enabled = 1;
save_pending = pending;
if(save_pending == 0)
return;
pending = 0;
In order to avoid an interrupt arriving between reading pending and setting
it to zero, in which case, the record of the interrupt would be erased,
signals are enabled.
What happened was that gcc reordered this so that 'save_pending = pending'
came before 'signals_enabled = 1', creating a one-instruction window within
which an interrupt could arrive, set its bit in pending, and have it be
immediately erased.
When the I/O workload is purely disk-based, the loss of a block device
interrupt stops the entire I/O system because the next block request will
wait for the current one to finish. Thus the system hangs until something
else causes some I/O to arrive, such as a network packet or console input.
The fix to this particular problem is a memory barrier between enabling
signals and reading the pending signal mask. An xchg would also probably
work.
Looking over this code for similar problems led me to do a few more
things:
- make signals_enabled and pending volatile so that they don't get cached
in registers
- add an mb() to the return paths of block_signals and unblock_signals so
that the modification of signals_enabled doesn't get shuffled into the
caller in the event that these are inlined in the future.
Signed-off-by: Jeff Dike <jdike@addtoit.com>
Cc: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Diffstat (limited to 'arch/um/include/sysdep-i386/barrier.h')
-rw-r--r-- | arch/um/include/sysdep-i386/barrier.h | 9 |
1 files changed, 9 insertions, 0 deletions
diff --git a/arch/um/include/sysdep-i386/barrier.h b/arch/um/include/sysdep-i386/barrier.h new file mode 100644 index 0000000..b58d52c --- /dev/null +++ b/arch/um/include/sysdep-i386/barrier.h @@ -0,0 +1,9 @@ +#ifndef __SYSDEP_I386_BARRIER_H +#define __SYSDEP_I386_BARRIER_H + +/* Copied from include/asm-i386 for use by userspace. i386 has the option + * of using mfence, but I'm just using this, which works everywhere, for now. + */ +#define mb() asm volatile("lock; addl $0,0(%esp)") + +#endif |