We never supported the ATA VIRTIO_BLK_F_IDENTIFY anyway, but simply
including it in the struct definition broke us (see commit 8ef562d112
"lguest: fix descriptor corruption in example launcher").
Now we can acknowledge that it's deprecated, and get the struct without it.
Signed-off-by: Rusty Russell <rusty@xxxxxxxxxxxxxxx>
---
Documentation/lguest/lguest.c | 18 ++++++++++++++++--
1 file changed, 16 insertions(+), 2 deletions(-)
diff --git a/Documentation/lguest/lguest.c b/Documentation/lguest/lguest.c
--- a/Documentation/lguest/lguest.c
+++ b/Documentation/lguest/lguest.c
@@ -6,6 +6,7 @@
:*/
#define _LARGEFILE64_SOURCE
#define _GNU_SOURCE
+#define VIRTIO_BLK_IDENTIFY_DEPRECATED
#include <stdio.h>
#include <string.h>
#include <unistd.h>
@@ -1750,8 +1751,21 @@ static void setup_block_file(const char
add_feature(dev, VIRTIO_BLK_F_SEG_MAX);
conf.seg_max = cpu_to_le32(VIRTQUEUE_NUM - 2);
- /* Don't try to put whole struct: we have 8 bit limit. */
- set_config(dev, offsetof(struct virtio_blk_config, geometry), &conf);
+ /*
+ * We only have 8 bits of configuration space for each device. But a
+ * new feature for virtio_blk added a 1k 'identify' field to struct
+ * virtio_blk_config. We figured that this was a bad idea, but too
+ * late: the code was already in the wild.
+ *
+ * Normally, new features get added to the end of the config struct,
+ * but that won't do in this case. So now, if you define
+ * VIRTIO_BLK_IDENTIFY_DEPRECATED (as we do at the top of this file),
+ * it pretends that field doesn't exist at all, and we can add new ones
+ * where it once was.
+ *
+ * This way, old code still compiles. Bit of a mess, though.
+ */
+ set_config(dev, sizeof(struct virtio_blk_config), &conf);
verbose("device %u: virtblock %llu sectors\n",
++devices.device_num, le64_to_cpu(conf.capacity));
_______________________________________________
Virtualization mailing list
Virtualization@xxxxxxxxxxxxxxxxxxxxxxxxxx
https://lists.linux-foundation.org/mailman/listinfo/virtualization
