Revision abd96ecb298675a21c412a29f5de2f80174d5f18 authored by Oleg Nesterov on 22 August 2007, 21:01:58 UTC, committed by Linus Torvalds on 23 August 2007, 02:52:47 UTC
de_thread: if (atomic_read(&oldsighand->count) <= 1) BUG_ON(atomic_read(&sig->count) != 1); This is not safe without the rmb() in between. The results of two correctly ordered __exit_signal()->atomic_dec_and_test()'s could be seen out of order on our CPU. The same is true for the "thread_group_empty()" case, __unhash_process()'s changes could be seen before atomic_dec_and_test(&sig->count). On some platforms (including i386) atomic_read() doesn't provide even the compiler barrier, in that case these checks are simply racy. Remove these BUG_ON()'s. Alternatively, we can do something like BUG_ON( ({ smp_rmb(); atomic_read(&sig->count) != 1; }) ); Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru> Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com> Cc: Roland McGrath <roland@redhat.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
1 parent 5c076fc
File | Mode | Size |
---|---|---|
acorn | ||
acpi | ||
amba | ||
ata | ||
atm | ||
auxdisplay | ||
base | ||
block | ||
bluetooth | ||
cdrom | ||
char | ||
clocksource | ||
connector | ||
cpufreq | ||
crypto | ||
dio | ||
dma | ||
edac | ||
eisa | ||
fc4 | ||
firewire | ||
firmware | ||
hid | ||
hwmon | ||
i2c | ||
ide | ||
ieee1394 | ||
infiniband | ||
input | ||
isdn | ||
kvm | ||
leds | ||
lguest | ||
macintosh | ||
mca | ||
md | ||
media | ||
message | ||
mfd | ||
misc | ||
mmc | ||
mtd | ||
net | ||
nubus | ||
of | ||
oprofile | ||
parisc | ||
parport | ||
pci | ||
pcmcia | ||
pnp | ||
power | ||
ps3 | ||
rapidio | ||
rtc | ||
s390 | ||
sbus | ||
scsi | ||
serial | ||
sh | ||
sn | ||
spi | ||
tc | ||
telephony | ||
uio | ||
usb | ||
video | ||
w1 | ||
xen | ||
zorro | ||
Kconfig | -rw-r--r-- | 1.5 KB |
Makefile | -rw-r--r-- | 2.6 KB |
Computing file changes ...