Fix cpumask manipulation on (64-bit) Windows systems.
cpus_allowed=nn values greater than 32 does not work, due
to the compiler not promoting expressions like "1 << cpu" to
64 bits, the clear function clearing the wrong bit (using "cpu - 1"),
and the clear function using XOR to clear (which only works if the
bit was previously set).
Example problems (from extra debug prints):
* setting CPU 32 really sets bit 1 (CPU 0)
Set mask of 0000000000000000 to add 0000000000000001 (32)
* setting CPU 63 really sets bit 31 (CPU 32)
Set mask of 0000000000000000 to add 0000000080000000 (63)
* clearing CPU 0 really clears bit 63 (CPU 64)
Clear mask of 0000000055555555 to remove 8000000000000000 (0)
* clearing CPU 2 really clears bit 2 (CPU 1)
Clear mask of 0000000055555555 to remove 0000000000000002 (2)
Tested with x86_64-w64-mingw32-gcc 6.4.0 from cygwin on a system
with 64 CPU cores (all fitting in one Windows processor group).
---
os/os-windows.h | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/os/os-windows.h b/os/os-windows.h
index 520da19..c7941f4 100644
--- a/os/os-windows.h
+++ b/os/os-windows.h
@@ -209,17 +209,17 @@ static inline int fio_getaffinity(int pid, os_cpu_mask_t *mask)
static inline void fio_cpu_clear(os_cpu_mask_t *mask, int cpu)
{
- *mask ^= 1 << (cpu-1);
+ *mask &= ~(1ULL << cpu);
}
static inline void fio_cpu_set(os_cpu_mask_t *mask, int cpu)
{
- *mask |= 1 << cpu;
+ *mask |= 1ULL << cpu;
}
static inline int fio_cpu_isset(os_cpu_mask_t *mask, int cpu)
{
- return (*mask & (1U << cpu));
+ return (*mask & (1ULL << cpu));
}
static inline int fio_cpu_count(os_cpu_mask_t *mask)
--
2.9.3
--
To unsubscribe from this list: send the line "unsubscribe fio" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
