aboutsummaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorThanos Makatos <thanos.makatos@nutanix.com>2019-11-05 11:29:24 -0500
committerThanos Makatos <thanos.makatos@nutanix.com>2019-11-05 11:29:24 -0500
commit15e6ede93fbf7195c3eb425b3900065d59a19e35 (patch)
tree2f1e8a97a1f31de85ccb9f87d4911ca5c790135a
parentb24e948208e03e5e00e4654d64c4190c1889a90a (diff)
downloadlibvfio-user-15e6ede93fbf7195c3eb425b3900065d59a19e35.zip
libvfio-user-15e6ede93fbf7195c3eb425b3900065d59a19e35.tar.gz
libvfio-user-15e6ede93fbf7195c3eb425b3900065d59a19e35.tar.bz2
remove obsolete comment, this is fixed by the IOMMU patch
Signed-off-by: Thanos Makatos <thanos.makatos@nutanix.com>
-rw-r--r--kmod/muser.c3
1 files changed, 1 insertions, 2 deletions
diff --git a/kmod/muser.c b/kmod/muser.c
index 9cb9256..cd60021 100644
--- a/kmod/muser.c
+++ b/kmod/muser.c
@@ -817,8 +817,7 @@ out:
* are serialised (this can be enforced by muser), then we tell libmuser to
* mmap the control device. Do we need to distinguish between the different
* DMA map calls at this stage if we can enforce only one outstanding DMA map
- * call? What happens when the DMA map happens too early, before GET_DEVICE_FD
- * is called?
+ * call?
*/
static int muser_iommu_notifier(struct notifier_block *nb, unsigned long action,
void *data)