Tags give the ability to mark specific points in history as being important
-
block-pull-request
6910f04a · ·Pull request Fiona's virtio-scsi TMF deadlock fix. Paolo: I merged the scsi fix in my block tree, but realize now that it belongs to the scsi subsystem. Sorry about that, I'll be more careful next time. Please reply if you want to handle this patch yourself.
-
tracing-pull-request
1461752f · ·Pull request Tanish Desai and Paolo Bonzini's tracing Rust support.
-
aio_add_sqe-v1
75aa2902 · ·block: unify block and fdmon io_uring Every AioContext has an io_uring context for file descriptor monitoring. In addition, block/io_uring.c also has an io_uring context for disk I/O. This patch series eliminates the extra io_uring context so that AioContext has a single io_uring context. Patch 2 introduces the aio_add_sqe() API for submitting custom io_uring SQEs. Patch 3 then replaces its io_uring context with the aio_add_sqe() API so that disk I/O requests share the AioContext's io_uring context. I will post performance results when they become available. Based-on: 20250326183340.1010531-1-stefanha@redhat.com
-
fdmon-glib-integration-v1
9aa014eb · ·aio-posix: integrate fdmon into glib event loop This series hooks the epoll and io_uring file descriptor monitoring implementations up to the glib event loop so that even the main loop thread can use them. Until now the main loop thread was stuck using ppoll(2) because the glib event loop semantics weren't supported by fdmon-epoll.c and fdmon-io_uring.c. The motivation for doing this is to make io_uring available in every QEMU thread on Linux hosts. This will make it easier to add features that use io_uring in the future. I am preparing a follow-up series that unifies block/io_uring.c and util/fdmon-io_uring.c so that AioContext has only one io_uring context instead of two. Expect performance results in the coming days comparing as ppoll(2) vs epoll(7) vs io_uring(7).
-