kernel_optimize_test/arch/sparc64/lib/bitops.S
David S. Miller b445e26cbf [SPARC64]: Avoid membar instructions in delay slots.
In particular, avoid membar instructions in the delay
slot of a jmpl instruction.

UltraSPARC-I, II, IIi, and IIe have a bug, documented in
the UltraSPARC-IIi User's Manual, Appendix K, Erratum 51

The long and short of it is that if the IMU unit misses
on a branch or jmpl, and there is a store buffer synchronizing
membar in the delay slot, the chip can stop fetching instructions.

If interrupts are enabled or some other trap is enabled, the
chip will unwedge itself, but performance will suffer.

We already had a workaround for this bug in a few spots, but
it's better to have the entire tree sanitized for this rule.

Signed-off-by: David S. Miller <davem@davemloft.net>
2005-06-27 15:42:04 -07:00

155 lines
2.8 KiB
ArmAsm

/* $Id: bitops.S,v 1.3 2001/11/18 00:12:56 davem Exp $
* bitops.S: Sparc64 atomic bit operations.
*
* Copyright (C) 2000 David S. Miller (davem@redhat.com)
*/
#include <linux/config.h>
#include <asm/asi.h>
.text
/* On SMP we need to use memory barriers to ensure
* correct memory operation ordering, nop these out
* for uniprocessor.
*/
#ifdef CONFIG_SMP
#define BITOP_PRE_BARRIER membar #StoreLoad | #LoadLoad
#define BITOP_POST_BARRIER \
ba,pt %xcc, 80b; \
membar #StoreLoad | #StoreStore
80: retl
nop
#else
#define BITOP_PRE_BARRIER
#define BITOP_POST_BARRIER
#endif
.globl test_and_set_bit
.type test_and_set_bit,#function
test_and_set_bit: /* %o0=nr, %o1=addr */
BITOP_PRE_BARRIER
srlx %o0, 6, %g1
mov 1, %o2
sllx %g1, 3, %g3
and %o0, 63, %g2
sllx %o2, %g2, %o2
add %o1, %g3, %o1
1: ldx [%o1], %g7
or %g7, %o2, %g1
casx [%o1], %g7, %g1
cmp %g7, %g1
bne,pn %xcc, 1b
and %g7, %o2, %g2
clr %o0
movrne %g2, 1, %o0
BITOP_POST_BARRIER
retl
nop
.size test_and_set_bit, .-test_and_set_bit
.globl test_and_clear_bit
.type test_and_clear_bit,#function
test_and_clear_bit: /* %o0=nr, %o1=addr */
BITOP_PRE_BARRIER
srlx %o0, 6, %g1
mov 1, %o2
sllx %g1, 3, %g3
and %o0, 63, %g2
sllx %o2, %g2, %o2
add %o1, %g3, %o1
1: ldx [%o1], %g7
andn %g7, %o2, %g1
casx [%o1], %g7, %g1
cmp %g7, %g1
bne,pn %xcc, 1b
and %g7, %o2, %g2
clr %o0
movrne %g2, 1, %o0
BITOP_POST_BARRIER
retl
nop
.size test_and_clear_bit, .-test_and_clear_bit
.globl test_and_change_bit
.type test_and_change_bit,#function
test_and_change_bit: /* %o0=nr, %o1=addr */
BITOP_PRE_BARRIER
srlx %o0, 6, %g1
mov 1, %o2
sllx %g1, 3, %g3
and %o0, 63, %g2
sllx %o2, %g2, %o2
add %o1, %g3, %o1
1: ldx [%o1], %g7
xor %g7, %o2, %g1
casx [%o1], %g7, %g1
cmp %g7, %g1
bne,pn %xcc, 1b
and %g7, %o2, %g2
clr %o0
movrne %g2, 1, %o0
BITOP_POST_BARRIER
retl
nop
.size test_and_change_bit, .-test_and_change_bit
.globl set_bit
.type set_bit,#function
set_bit: /* %o0=nr, %o1=addr */
srlx %o0, 6, %g1
mov 1, %o2
sllx %g1, 3, %g3
and %o0, 63, %g2
sllx %o2, %g2, %o2
add %o1, %g3, %o1
1: ldx [%o1], %g7
or %g7, %o2, %g1
casx [%o1], %g7, %g1
cmp %g7, %g1
bne,pn %xcc, 1b
nop
retl
nop
.size set_bit, .-set_bit
.globl clear_bit
.type clear_bit,#function
clear_bit: /* %o0=nr, %o1=addr */
srlx %o0, 6, %g1
mov 1, %o2
sllx %g1, 3, %g3
and %o0, 63, %g2
sllx %o2, %g2, %o2
add %o1, %g3, %o1
1: ldx [%o1], %g7
andn %g7, %o2, %g1
casx [%o1], %g7, %g1
cmp %g7, %g1
bne,pn %xcc, 1b
nop
retl
nop
.size clear_bit, .-clear_bit
.globl change_bit
.type change_bit,#function
change_bit: /* %o0=nr, %o1=addr */
srlx %o0, 6, %g1
mov 1, %o2
sllx %g1, 3, %g3
and %o0, 63, %g2
sllx %o2, %g2, %o2
add %o1, %g3, %o1
1: ldx [%o1], %g7
xor %g7, %o2, %g1
casx [%o1], %g7, %g1
cmp %g7, %g1
bne,pn %xcc, 1b
nop
retl
nop
.size change_bit, .-change_bit