On 06/18/2010 06:44 AM, Daniel P. Berrange wrote:
On Wed, Jun 02, 2010 at 10:55:17AM -0400, Cole Robinson wrote:
If VM startup fails early enough (can't find a referenced USB device), libvirtd will crash trying to clear the VNC port bit, since port = 0, which overflows us out of the bitmap bounds.
Fix this by being more defensive in the bitmap operations, and only clearing a previously set VNC port.
v2: Add safety check to all relevant bitmap ops.
Signed-off-by: Cole Robinson <crobinso@redhat.com> --- src/qemu/qemu_driver.c | 2 +- src/util/bitmap.c | 6 +++--- 2 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/src/qemu/qemu_driver.c b/src/qemu/qemu_driver.c index c8cd50a..f5a1310 100644 --- a/src/qemu/qemu_driver.c +++ b/src/qemu/qemu_driver.c @@ -3741,7 +3741,7 @@ retry: if ((vm->def->ngraphics == 1) && vm->def->graphics[0]->type == VIR_DOMAIN_GRAPHICS_TYPE_VNC && vm->def->graphics[0]->data.vnc.autoport && - vm->def->graphics[0]->data.vnc.port != -1) { + vm->def->graphics[0]->data.vnc.port >= QEMU_VNC_PORT_MIN) { if (virBitmapClearBit(driver->reservedVNCPorts, vm->def->graphics[0]->data.vnc.port - \ QEMU_VNC_PORT_MIN) < 0) { diff --git a/src/util/bitmap.c b/src/util/bitmap.c index 69094a5..cef3fc4 100644 --- a/src/util/bitmap.c +++ b/src/util/bitmap.c @@ -100,7 +100,7 @@ void virBitmapFree(virBitmapPtr bitmap) */ int virBitmapSetBit(virBitmapPtr bitmap, size_t b) { - if (b > bitmap->size - 1) + if (bitmap->size <= b) return -1;
bitmap->map[VIR_BITMAP_UNIT_OFFSET(b)] |= (1 << VIR_BITMAP_BIT_OFFSET(b)); @@ -118,7 +118,7 @@ int virBitmapSetBit(virBitmapPtr bitmap, size_t b) */ int virBitmapClearBit(virBitmapPtr bitmap, size_t b) { - if (b > bitmap->size - 1) + if (bitmap->size <= b) return -1;
bitmap->map[VIR_BITMAP_UNIT_OFFSET(b)] &= ~(1 << VIR_BITMAP_BIT_OFFSET(b)); @@ -140,7 +140,7 @@ int virBitmapGetBit(virBitmapPtr bitmap, size_t b, bool *result) { uint32_t bit;
- if (b > bitmap->size - 1) + if (bitmap->size <= b) return -1;
bit = bitmap->map[VIR_BITMAP_UNIT_OFFSET(b)] &
ACK, this patch doesn't seem to have ever been pushed - current GIT still crashes on abort of VM startup
Daniel
Pushed now, accidentally in 2 commits though, since I pushed v1 first. Thanks, Cole