summaryrefslogtreecommitdiffstats
path: root/results/classifier/gemma3:12b/device/1935
diff options
context:
space:
mode:
authorChristian Krinitsin <mail@krinitsin.com>2025-07-03 07:27:52 +0000
committerChristian Krinitsin <mail@krinitsin.com>2025-07-03 07:27:52 +0000
commitd0c85e36e4de67af628d54e9ab577cc3fad7796a (patch)
treef8f784b0f04343b90516a338d6df81df3a85dfa2 /results/classifier/gemma3:12b/device/1935
parent7f4364274750eb8cb39a3e7493132fca1c01232e (diff)
downloademulator-bug-study-d0c85e36e4de67af628d54e9ab577cc3fad7796a.tar.gz
emulator-bug-study-d0c85e36e4de67af628d54e9ab577cc3fad7796a.zip
add deepseek and gemma results
Diffstat (limited to 'results/classifier/gemma3:12b/device/1935')
-rw-r--r--results/classifier/gemma3:12b/device/19356
1 files changed, 6 insertions, 0 deletions
diff --git a/results/classifier/gemma3:12b/device/1935 b/results/classifier/gemma3:12b/device/1935
new file mode 100644
index 00000000..f04fb102
--- /dev/null
+++ b/results/classifier/gemma3:12b/device/1935
@@ -0,0 +1,6 @@
+
+migrate problem when add SCSI reservations with iSCSI backed disks
+Description of problem:
+When performing migrations with QEMU using iSCSI as the backend, it's common for the migration to start successfully. However, in scenarios where Persistent Reservations are added in the guest, the target host, under the precopy mode, preempts the Persistent Reservations right from the beginning, causing migration issues. Is there a way to control the Persistent Reservations lock within QEMU at an appropriate time, ensuring that it's only preempted during the switchover phase?
+
+Isn't libiscsi thread-safe? Can multiple threads operate on Persistent Reservations lock simultaneously?