add deepseek and gemma results

1 files changed, 47 insertions, 0 deletions

diff --git a/results/classifier/gemma3:12b/files/1103868 b/results/classifier/gemma3:12b/files/1103868
new file mode 100644
index 000000000..39396747c
--- /dev/null
+++ b/results/classifier/gemma3:12b/files/1103868
@@ -0,0 +1,47 @@
+
+drive_mirror crashes on full disk copy of a resized disk with a backing file
+
+This bug was discovered using libvirt on ubuntu with a build of qemu 1.3 but it is trivailly reproducible with the curent git version.
+
+Repro steps:
+
+qemu-img create -f qcow2 base 32M
+qemu-img create -f qcow2 -o backing_file=base disk
+qemu-img resize /home/vishvananda/disk 64M
+qemu-system-x86_64 -drive file=disk,id=vda -vnc :1 -monitor stdio
+QEMU 1.3.0 monitor - type 'help' for more information
+(qemu) drive_mirror -f vda test
+Formatting 'test', fmt=qcow2 size=67108864 encryption=off cluster_size=65536 lazy_refcounts=off 
+qemu-system-x86_64: /build/buildd/qemu-1.3.0+dfsg/block/mirror.c:129: mirror_run: Assertion `n > 0' failed.
+Aborted
+
+Note that the command works just fine if the front image is not resized:
+
+qemu-img create -f qcow2 base 32M
+qemu-img create -f qcow2 -o backing_file=base disk
+qemu-system-x86_64 -drive file=disk,id=vda -vnc :1 -monitor stdio
+
+or if the backing file is resized as well:
+
+qemu-img create -f qcow2 base 32M
+qemu-img create -f qcow2 -o backing_file=base disk
+qemu-img resize /home/vishvananda/disk 64M
+qemu-img resize /home/vishvananda/base 64M
+qemu-system-x86_64 -drive file=disk,id=vda -vnc :1 -monitor stdio
+
+or if we don't use -f when creating the mirror:
+
+QEMU 1.3.0 monitor - type 'help' for more information
+(qemu) drive_mirror vda test
+Formatting 'test', fmt=qcow2 size=33554432 backing_file='base' backing_fmt='qcow2' encryption=off cluster_size=65536 lazy_refcounts=off 
+
+although in this final case the mirror is created the same size as the backing file which seems wrong:
+
+qemu-img info test
+image: test
+file format: qcow2
+virtual size: 32M (33554432 bytes)
+disk size: 196K
+cluster_size: 65536
+backing file: base
+backing file format: qcow2
\ No newline at end of file