| Message ID | 20090426114638.GA27129@redhat.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
Michael S. Tsirkin wrote: > By the time we've done a memset, it's too late to > check the pointer for MAP_FAILED value. Let's compare > to MAP_FAILED first. > > Applied, thanks.
diff --git a/hw/device-assignment.c b/hw/device-assignment.c index 1f0a1a7..0a5f850 100644 --- a/hw/device-assignment.c +++ b/hw/device-assignment.c @@ -1087,12 +1087,12 @@ static int assigned_dev_register_msix_mmio(AssignedDevice *dev) dev->msix_table_page = mmap(NULL, 0x1000, PROT_READ|PROT_WRITE, MAP_ANONYMOUS|MAP_PRIVATE, 0, 0); - memset(dev->msix_table_page, 0, 0x1000); if (dev->msix_table_page == MAP_FAILED) { fprintf(stderr, "fail allocate msix_table_page! %s\n", strerror(errno)); return -EFAULT; } + memset(dev->msix_table_page, 0, 0x1000); dev->mmio_index = cpu_register_io_memory(0, msix_mmio_read, msix_mmio_write, dev); return 0;
By the time we've done a memset, it's too late to check the pointer for MAP_FAILED value. Let's compare to MAP_FAILED first. Signed-off-by: Michael S. Tsikirn <mst@redhat.com> --- Noticed this when reading the code. Makes sense, does it not? hw/device-assignment.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-)