kvm.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] vfio/pci: Separate INTx-enabled vfio_pci_device from unenabled to make the code logic clearer.
@ 2023-11-20  3:17 JianChunfu
  2023-11-20 17:38 ` Alex Williamson
  0 siblings, 1 reply; 2+ messages in thread
From: JianChunfu @ 2023-11-20  3:17 UTC (permalink / raw)
  To: alex.williamson, cohuck; +Cc: kvm, linux-kernel, shenghui.qu, JianChunfu

It seems a little unclear when dealing with vfio_intx_set_signal()
because of vfio_pci_device which is irq_none,
so separate the two situations.

Signed-off-by: JianChunfu <chunfu.jian@shingroup.cn>
---
 drivers/vfio/pci/vfio_pci_intrs.c | 31 +++++++++++++++----------------
 1 file changed, 15 insertions(+), 16 deletions(-)

diff --git a/drivers/vfio/pci/vfio_pci_intrs.c b/drivers/vfio/pci/vfio_pci_intrs.c
index 6069a11fb51a..b6d126c48393 100644
--- a/drivers/vfio/pci/vfio_pci_intrs.c
+++ b/drivers/vfio/pci/vfio_pci_intrs.c
@@ -468,6 +468,8 @@ static int vfio_pci_set_intx_trigger(struct vfio_pci_core_device *vdev,
 				     unsigned index, unsigned start,
 				     unsigned count, uint32_t flags, void *data)
 {
+	int32_t fd = *(int32_t *)data;
+
 	if (is_intx(vdev) && !count && (flags & VFIO_IRQ_SET_DATA_NONE)) {
 		vfio_intx_disable(vdev);
 		return 0;
@@ -476,28 +478,25 @@ static int vfio_pci_set_intx_trigger(struct vfio_pci_core_device *vdev,
 	if (!(is_intx(vdev) || is_irq_none(vdev)) || start != 0 || count != 1)
 		return -EINVAL;
 
-	if (flags & VFIO_IRQ_SET_DATA_EVENTFD) {
-		int32_t fd = *(int32_t *)data;
+	if (!is_intx(vdev)) {
 		int ret;
+		if (flags & VFIO_IRQ_SET_DATA_EVENTFD) {
+			ret = vfio_intx_enable(vdev);
+			if (ret)
+				return ret;
 
-		if (is_intx(vdev))
-			return vfio_intx_set_signal(vdev, fd);
+			ret = vfio_intx_set_signal(vdev, fd);
+			if (ret)
+				vfio_intx_disable(vdev);
 
-		ret = vfio_intx_enable(vdev);
-		if (ret)
 			return ret;
-
-		ret = vfio_intx_set_signal(vdev, fd);
-		if (ret)
-			vfio_intx_disable(vdev);
-
-		return ret;
+		} else
+			return -EINVAL;
 	}
 
-	if (!is_intx(vdev))
-		return -EINVAL;
-
-	if (flags & VFIO_IRQ_SET_DATA_NONE) {
+	if (flags & VFIO_IRQ_SET_DATA_EVENTFD) {
+		return vfio_intx_set_signal(vdev, fd);
+	} else if (flags & VFIO_IRQ_SET_DATA_NONE) {
 		vfio_send_intx_eventfd(vdev, NULL);
 	} else if (flags & VFIO_IRQ_SET_DATA_BOOL) {
 		uint8_t trigger = *(uint8_t *)data;
-- 
2.27.0


^ permalink raw reply related	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2023-11-20 17:38 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2023-11-20  3:17 [PATCH] vfio/pci: Separate INTx-enabled vfio_pci_device from unenabled to make the code logic clearer JianChunfu
2023-11-20 17:38 ` Alex Williamson

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).