[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH 7/7] migration: Finer grained tracepoints for POSTCOPY_LISTEN
From: |
Dr. David Alan Gilbert |
Subject: |
Re: [PATCH 7/7] migration: Finer grained tracepoints for POSTCOPY_LISTEN |
Date: |
Wed, 8 Dec 2021 19:46:20 +0000 |
User-agent: |
Mutt/2.1.3 (2021-09-10) |
* Peter Xu (peterx@redhat.com) wrote:
> The enablement of postcopy listening has a few steps, add a few tracepoints to
> be there ready for some basic measurements for them.
>
> Signed-off-by: Peter Xu <peterx@redhat.com>
> ---
> migration/savevm.c | 5 ++++-
> migration/trace-events | 2 +-
> 2 files changed, 5 insertions(+), 2 deletions(-)
>
> diff --git a/migration/savevm.c b/migration/savevm.c
> index 17b8e25e00..5b3f31eab2 100644
> --- a/migration/savevm.c
> +++ b/migration/savevm.c
> @@ -1946,7 +1946,7 @@ static void *postcopy_ram_listen_thread(void *opaque)
> static int loadvm_postcopy_handle_listen(MigrationIncomingState *mis)
> {
> PostcopyState ps = postcopy_state_set(POSTCOPY_INCOMING_LISTENING);
> - trace_loadvm_postcopy_handle_listen();
> + trace_loadvm_postcopy_handle_listen(1);
I think we tend just to split this into separate traces in many places;
or if we're using the same one then we should use a string
I'd make this:
trace_loadvm_postcopy_handle_listen_entry();
for example.
> Error *local_err = NULL;
>
> if (ps != POSTCOPY_INCOMING_ADVISE && ps != POSTCOPY_INCOMING_DISCARD) {
> @@ -1962,6 +1962,7 @@ static int
> loadvm_postcopy_handle_listen(MigrationIncomingState *mis)
> postcopy_ram_prepare_discard(mis);
> }
> }
> + trace_loadvm_postcopy_handle_listen(2);
>
> /*
> * Sensitise RAM - can now generate requests for blocks that don't exist
> @@ -1974,6 +1975,7 @@ static int
> loadvm_postcopy_handle_listen(MigrationIncomingState *mis)
> return -1;
> }
> }
> + trace_loadvm_postcopy_handle_listen(3);
>
> if (postcopy_notify(POSTCOPY_NOTIFY_INBOUND_LISTEN, &local_err)) {
> error_report_err(local_err);
> @@ -1988,6 +1990,7 @@ static int
> loadvm_postcopy_handle_listen(MigrationIncomingState *mis)
> QEMU_THREAD_DETACHED);
> qemu_sem_wait(&mis->listen_thread_sem);
> qemu_sem_destroy(&mis->listen_thread_sem);
> + trace_loadvm_postcopy_handle_listen(4);
trace_loadvm_postcopy_handle_listen_entry_end();
>
> return 0;
> }
> diff --git a/migration/trace-events b/migration/trace-events
> index d63a5915f5..1aa6937dc1 100644
> --- a/migration/trace-events
> +++ b/migration/trace-events
> @@ -14,7 +14,7 @@ loadvm_handle_cmd_packaged_main(int ret) "%d"
> loadvm_handle_cmd_packaged_received(int ret) "%d"
> loadvm_handle_recv_bitmap(char *s) "%s"
> loadvm_postcopy_handle_advise(void) ""
> -loadvm_postcopy_handle_listen(void) ""
> +loadvm_postcopy_handle_listen(int i) "%d"
> loadvm_postcopy_handle_run(void) ""
> loadvm_postcopy_handle_run_cpu_sync(void) ""
> loadvm_postcopy_handle_run_vmstart(void) ""
> --
> 2.32.0
>
--
Dr. David Alan Gilbert / dgilbert@redhat.com / Manchester, UK
- [PATCH 3/7] migration: Drop postcopy_chunk_hostpages(), (continued)
- [PATCH 3/7] migration: Drop postcopy_chunk_hostpages(), Peter Xu, 2021/12/07
- [PATCH 4/7] migration: Do chunk page in postcopy_each_ram_send_discard(), Peter Xu, 2021/12/07
- [PATCH 5/7] migration: Drop return code for disgard ram process, Peter Xu, 2021/12/07
- [PATCH 6/7] migration: Dump sub-cmd name in loadvm_process_command tp, Peter Xu, 2021/12/07
- [PATCH 7/7] migration: Finer grained tracepoints for POSTCOPY_LISTEN, Peter Xu, 2021/12/07
- Re: [PATCH 7/7] migration: Finer grained tracepoints for POSTCOPY_LISTEN,
Dr. David Alan Gilbert <=