[Qemu-devel] virtio-blk wierd memcpy

[Qemu-devel] virtio-blk wierd memcpy

[Dave Airlie]

In my failing attempts to valgrind qemu (how to people live like
this?), I spotted this in the logs

==19521== Source and destination overlap in memcpy(0x31d38938, 0x31d38938, 64)
==19521==    at 0x4A0A343: memcpy@@GLIBC_2.14 (in
/usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so)
==19521==    by 0x42774E: virtio_blk_device_init (virtio-blk.c:686)
==19521==    by 0x46EE9E: virtio_device_init (virtio.c:1158)
==19521==    by 0x25405E: device_realize (qdev.c:178)
==19521==    by 0x2559B5: device_set_realized (qdev.c:699)
==19521==    by 0x3A819B: property_set_bool (object.c:1315)
==19521==    by 0x3A6CE0: object_property_set (object.c:803)

The memcpy looks completely superfluous.

Dave.

Re: [Qemu-devel] virtio-blk wierd memcpy

[Stefan Weil]

Am 17.10.2013 20:49, schrieb Dave Airlie:
> In my failing attempts to valgrind qemu (how to people live like
> this?), I spotted this in the logs
>
> ==19521== Source and destination overlap in memcpy(0x31d38938, 0x31d38938, 64)
> ==19521==    at 0x4A0A343: memcpy@@GLIBC_2.14 (in
> /usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so)
> ==19521==    by 0x42774E: virtio_blk_device_init (virtio-blk.c:686)
> ==19521==    by 0x46EE9E: virtio_device_init (virtio.c:1158)
> ==19521==    by 0x25405E: device_realize (qdev.c:178)
> ==19521==    by 0x2559B5: device_set_realized (qdev.c:699)
> ==19521==    by 0x3A819B: property_set_bool (object.c:1315)
> ==19521==    by 0x3A6CE0: object_property_set (object.c:803)
>
> The memcpy looks completely superfluous.
>
> Dave.

That's correct - thank you for your report.

I have prepared a patch which removes that memcpy:
http://patchwork.ozlabs.org/patch/284355/

Regards,
Stefan

Re: [Qemu-devel] virtio-blk wierd memcpy

[Christian Borntraeger]

On 17/10/13 20:49, Dave Airlie wrote:
> In my failing attempts to valgrind qemu (how to people live like
> this?), I spotted this in the logs

What are the failures? Do you use emulation or KVM? I know that there seems to be some issues regarding
async I/O but otherwise that that it seems to run. (Some KVm ioctls miss special handling, but in
general it should work).
> 
> ==19521== Source and destination overlap in memcpy(0x31d38938, 0x31d38938, 64)
> ==19521==    at 0x4A0A343: memcpy@@GLIBC_2.14 (in
> /usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so)
> ==19521==    by 0x42774E: virtio_blk_device_init (virtio-blk.c:686)
> ==19521==    by 0x46EE9E: virtio_device_init (virtio.c:1158)
> ==19521==    by 0x25405E: device_realize (qdev.c:178)
> ==19521==    by 0x2559B5: device_set_realized (qdev.c:699)
> ==19521==    by 0x3A819B: property_set_bool (object.c:1315)
> ==19521==    by 0x3A6CE0: object_property_set (object.c:803)
> 
> The memcpy looks completely superfluous.
> 
> Dave.
>