qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
* [Qemu-devel] [PATCH for-2.12] memfd: fix vhost-user-test on non-memfd capable host
@ 2018-03-28 12:18 Marc-André Lureau
  2018-03-31  6:57 ` no-reply
                   ` (3 more replies)
  0 siblings, 4 replies; 5+ messages in thread
From: Marc-André Lureau @ 2018-03-28 12:18 UTC (permalink / raw)
  To: qemu-devel; +Cc: pbonzini, Marc-André Lureau

On RHEL7, memfd is not supported, and vhost-user-test fails:
TEST: tests/vhost-user-test... (pid=10248)
  /x86_64/vhost-user/migrate:
  qemu-system-x86_64: -object memory-backend-memfd,id=mem,size=2M,: failed to create memfd
FAIL

There is a qemu_memfd_check() to prevent running memfd path, but it
also checks for fallback implementation. Let's specialize
qemu_memfd_check() to check memfd only, while qemu_memfd_alloc_check()
checks for the qemu_memfd_alloc() API.

Reported-by: Miroslav Rezanina <mrezanin@redhat.com>
Tested-by: Miroslav Rezanina <mrezanin@redhat.com>
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
---
 include/qemu/memfd.h |  1 +
 hw/virtio/vhost.c    |  2 +-
 util/memfd.c         | 30 +++++++++++++++++++++++++++++-
 3 files changed, 31 insertions(+), 2 deletions(-)

diff --git a/include/qemu/memfd.h b/include/qemu/memfd.h
index de10198ed6..49e79634da 100644
--- a/include/qemu/memfd.h
+++ b/include/qemu/memfd.h
@@ -18,6 +18,7 @@
 
 int qemu_memfd_create(const char *name, size_t size, bool hugetlb,
                       uint64_t hugetlbsize, unsigned int seals, Error **errp);
+bool qemu_memfd_alloc_check(void);
 void *qemu_memfd_alloc(const char *name, size_t size, unsigned int seals,
                        int *fd, Error **errp);
 void qemu_memfd_free(void *ptr, size_t size, int fd);
diff --git a/hw/virtio/vhost.c b/hw/virtio/vhost.c
index 250f886acb..27c1ec5fe8 100644
--- a/hw/virtio/vhost.c
+++ b/hw/virtio/vhost.c
@@ -1223,7 +1223,7 @@ int vhost_dev_init(struct vhost_dev *hdev, void *opaque,
         if (!(hdev->features & (0x1ULL << VHOST_F_LOG_ALL))) {
             error_setg(&hdev->migration_blocker,
                        "Migration disabled: vhost lacks VHOST_F_LOG_ALL feature.");
-        } else if (vhost_dev_log_is_shared(hdev) && !qemu_memfd_check()) {
+        } else if (vhost_dev_log_is_shared(hdev) && !qemu_memfd_alloc_check()) {
             error_setg(&hdev->migration_blocker,
                        "Migration disabled: failed to allocate shared memory");
         }
diff --git a/util/memfd.c b/util/memfd.c
index 07d579ea7d..277f7211af 100644
--- a/util/memfd.c
+++ b/util/memfd.c
@@ -173,7 +173,13 @@ enum {
     MEMFD_TODO
 };
 
-bool qemu_memfd_check(void)
+/**
+ * qemu_memfd_alloc_check():
+ *
+ * Check if qemu_memfd_alloc() can allocate, including using a
+ * fallback implementation when host doesn't support memfd.
+ */
+bool qemu_memfd_alloc_check(void)
 {
     static int memfd_check = MEMFD_TODO;
 
@@ -188,3 +194,25 @@ bool qemu_memfd_check(void)
 
     return memfd_check == MEMFD_OK;
 }
+
+/**
+ * qemu_memfd_check():
+ *
+ * Check if host supports memfd.
+ */
+bool qemu_memfd_check(void)
+{
+    static int memfd_check = MEMFD_TODO;
+
+    if (memfd_check == MEMFD_TODO) {
+        int mfd = memfd_create("test", 0);
+        if (mfd >= 0) {
+            memfd_check = MEMFD_OK;
+            close(mfd);
+        } else {
+            memfd_check = MEMFD_KO;
+        }
+    }
+
+    return memfd_check == MEMFD_OK;
+}
-- 
2.17.0.rc1.1.g4c4f2b46a3

^ permalink raw reply related	[flat|nested] 5+ messages in thread
end of thread, other threads:[~2018-04-05 15:26 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2018-03-28 12:18 [Qemu-devel] [PATCH for-2.12] memfd: fix vhost-user-test on non-memfd capable host Marc-André Lureau
2018-03-31  6:57 ` no-reply
2018-04-04 13:51 ` Marc-André Lureau
2018-04-05  3:56 ` Miroslav Rezanina
2018-04-05 15:26 ` Paolo Bonzini

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).