Skip to contentSkip to conversation

wait/9p: Account 9P RPC waiting time as I/O wait

PATCH Superseded Remi Pommarel 3 participants 5 messages Jan 21

Subsystems (3)

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

cover wait/9p: Account 9P RPC waiting time as I/O wait

This patch serie helps to attribute the time spent waiting for server responses during 9P RPC calls to I/O wait time in system metrics. As a result, I/O-intensive operations on a 9pfs mount will now be reflected in the "wa" column instead of the "id" one of tools like top.

Thanks,

Remi Pommarel (2):
wait: Introduce io_wait_event_killable()
9p: Track 9P RPC waiting time as IO

 include/linux/wait.h | 15 +++++++++++++++
 net/9p/client.c      |  4 ++--
 2 files changed, 17 insertions(+), 2 deletions(-)
Signature · 1 line
2.50.1
#1 lore raw

PATCH 1/2 wait: Introduce io_wait_event_killable()

Add io_wait_event_killable(), a variant of wait_event_killable() that uses io_schedule() instead of schedule(). This is to be used in situation where waiting time is to be accounted as IO wait time.

---
 include/linux/wait.h | 15 +++++++++++++++
 1 file changed, 15 insertions(+)
include/linux/wait.h +15 -0
index f648044466d5..dce055e6add3 100644
@@ -937,6 +937,21 @@ extern int do_wait_intr_irq(wait_queue_head_t *, wait_queue_entry_t *);
 	__ret;									\
 })
 
+#define __io_wait_event_killable(wq, condition)					\
+	___wait_event(wq, condition, TASK_KILLABLE, 0, 0, io_schedule())
+
+/*
+ * wait_event_killable() - link wait_event_killable but with io_schedule()
+ */
+#define io_wait_event_killable(wq_head, condition)				\
+({										\
+	int __ret = 0;								\
+	might_sleep();								\
+	if (!(condition))							\
+		__ret = __io_wait_event_killable(wq_head, condition);		\
+	__ret;									\
+})
+
 #define __wait_event_state(wq, condition, state)				\
 	___wait_event(wq, condition, state, 0, 0, schedule())
 
Signature · 1 line
2.50.1
maintainer Acked-by on v1 1/2
#2 lore raw

Re: [PATCH 1/2] wait: Introduce io_wait_event_killable()

Quoted 5 lines · Remi Pommarel

Add io_wait_event_killable(), a variant of wait_event_killable() that uses io_schedule() instead of schedule(). This is to be used in situation where waiting time is to be accounted as IO wait time.

Quoted 33 lines
---
 include/linux/wait.h | 15 +++++++++++++++
 1 file changed, 15 insertions(+)
include/linux/wait.h +15 -0
index f648044466d5..dce055e6add3 100644
@@ -937,6 +937,21 @@ extern int do_wait_intr_irq(wait_queue_head_t *, wait_queue_entry_t *);
 	__ret;									\
 })
 
+#define __io_wait_event_killable(wq, condition)					\
+	___wait_event(wq, condition, TASK_KILLABLE, 0, 0, io_schedule())
+
+/*
+ * wait_event_killable() - link wait_event_killable but with io_schedule()
+ */
+#define io_wait_event_killable(wq_head, condition)				\
+({										\
+	int __ret = 0;								\
+	might_sleep();								\
+	if (!(condition))							\
+		__ret = __io_wait_event_killable(wq_head, condition);		\
+	__ret;									\
+})
+
 #define __wait_event_state(wq, condition, state)				\
 	___wait_event(wq, condition, state, 0, 0, schedule())
 
Signature · 1 line
2.50.1
#3 lore raw

PATCH 2/2 9p: Track 9P RPC waiting time as IO

In reply to Remi Pommarel

Use io_wait_event_killable() to ensure that time spent waiting for 9P RPC transactions is accounted as IO wait time.

---
 net/9p/client.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
net/9p/client.c +2 -2
index f60d1d041adb..1b475525ac5b 100644
@@ -590,8 +590,8 @@ p9_client_rpc(struct p9_client *c, int8_t type, const char *fmt, ...)
 	}
 again:
 	/* Wait for the response */
-	err = wait_event_killable(req->wq,
-				  READ_ONCE(req->status) >= REQ_STATUS_RCVD);
+	err = io_wait_event_killable(req->wq,
+				     READ_ONCE(req->status) >= REQ_STATUS_RCVD);
 
 	/* Make sure our req is coherent with regard to updates in other
 	 * threads - echoes to wmb() in the callback
Signature · 1 line
2.50.1
#4 lore raw

Re: [PATCH 2/2] 9p: Track 9P RPC waiting time as IO

Remi Pommarel wrote on Wed, Jan 21, 2026 at 08:21:59PM +0100:

Quoted 2 lines

Use io_wait_event_killable() to ensure that time spent waiting for 9P RPC transactions is accounted as IO wait time.

Thanks for splitting this out of your other 9p improvements!

I was about to ask Peter/Ingo which tree this should go through, but could you also convert the other few wait_event_killable() calls in net/9p/trans_*.c ?
They're either waiting for other IO to complete (virtio x2) or for the current IO to complete (virtio/xen), so I think they qualify just as much.

(the virtio ones will likely conflict with some other rework that's been dragging on last month, but given the patch is trivial it won't matter much, you can send as of master)

Thanks,

Signature · 1 line
Dominique Martinet | Asmadeus