Skip to contentSkip to conversation

staging/android: use multiple futex wait queues

PATCH Superseded Hugo Lefeuvre 3 participants 6 messages Feb 14, 2019 – Feb 16, 2019 Staging

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 staging/android: use multiple futex wait queues

Use multiple per-offset wait queues instead of one big wait queue per region.

---
This patch is based on the simplify handle_vsoc_cond_wait patchset, currently under review: https://lkml.org/lkml/2019/2/7/870

---
 drivers/staging/android/TODO   |  4 ---
 drivers/staging/android/vsoc.c | 56 ++++++++++++++++++++++++++--------
 2 files changed, 44 insertions(+), 16 deletions(-)
drivers/staging/android/TODO +0 -4
index fbf015cc6d62..cd2af06341dc 100644
@@ -12,10 +12,6 @@ ion/
  - Better test framework (integration with VGEM was suggested)
 
 vsoc.c, uapi/vsoc_shm.h
- - The current driver uses the same wait queue for all of the futexes in a
-   region. This will cause false wakeups in regions with a large number of
-   waiting threads. We should eventually use multiple queues and select the
-   queue based on the region.
  - Add debugfs support for examining the permissions of regions.
  - Remove VSOC_WAIT_FOR_INCOMING_INTERRUPT ioctl. This functionality has been
    superseded by the futex and is there for legacy reasons.
drivers/staging/android/vsoc.c +44 -12

110 lines not on this page · raw message

Signature · 1 line
2.20.1
#1 lore raw
Quoted 4 lines · Hugo Lefeuvre

Use multiple per-offset wait queues instead of one big wait queue per region.

Have you tested this?

Noticed any performance speedups or slow downs?

thanks,

greg k-h

#2 lore raw
Quoted 8 lines
Quoted 4 lines

Use multiple per-offset wait queues instead of one big wait queue per region.

Have you tested this?

Noticed any performance speedups or slow downs?

Not yet. I have started to set up a cuttlefish test environment but it might take some time until I am able to fully test these changes myself (having some trouble to find documentation)... I will post the results here when ready.

thanks!

regards,
Hugo

Signature · 3 lines
                Hugo Lefeuvre (hle)    |    www.owl.eu.com
RSA4096_ 360B 03B3 BF27 4F4D 7A3F D5E8 14AA 1EB8 A247 3DFD
ed25519_ 37B2 6D38 0B25 B8A2 6B9F 3A65 A36F 5357 5F2D DC4C
#4 lore raw
In reply to Hugo Lefeuvre

drivers/staging/android/vsoc.c static int handle_vsoc_cond_wait(struct file *filp, struct vsoc_cond_wait *arg) · line 413

Quoted 5 lines · Hugo Lefeuvre
drivers/staging/android/vsoc.c +1 -0
@@ -402,6 +410,7 @@ static int handle_vsoc_cond_wait(struct file *filp, struct vsoc_cond_wait *arg)
 	struct vsoc_region_data *data = vsoc_dev.regions_data + region_number;
 	int ret = 0;
 	struct vsoc_device_region *region_p = vsoc_region_from_filep(filp);
+	struct vsoc_futex_wait_queue_t *it, *wait_queue = NULL;
                                             ^^^^^^^^^^^^^^^^^

drivers/staging/android/vsoc.c static int handle_vsoc_cond_wait(struct file *filp, struct vsoc_cond_wait *arg) · line 430

Quoted 11 lines
drivers/staging/android/vsoc.c +4 -0
 	atomic_t *address = NULL;
 	ktime_t wake_time;
 
@@ -415,10 +424,27 @@ static int handle_vsoc_cond_wait(struct file *filp, struct vsoc_cond_wait *arg)
 	address = shm_off_to_virtual_addr(region_p->region_begin_offset +
 					  arg->offset);
 
+	/* Find wait queue corresponding to offset or create it */
+	spin_lock(&data->futex_wait_queue_lock);
+	list_for_each_entry(it, &data->futex_wait_queue_list, list) {
+		if (wait_queue->offset == arg->offset) {
                    ^^^^^^^^^^^^^^^^^^

You meant "it->offset".

drivers/staging/android/vsoc.c static int handle_vsoc_cond_wait(struct file *filp, struct vsoc_cond_wait *arg) · line 442

Quoted 12 lines
drivers/staging/android/vsoc.c +12 -0
+			wait_queue = it;
+			break;
+		}
+	}
+
+	if (!wait_queue) {
+		wait_queue = kmalloc(sizeof(*wait_queue), GFP_KERNEL);
+		wait_queue->offset = arg->offset;
+		init_waitqueue_head(&wait_queue->queue);
+		list_add(&wait_queue->list, &data->futex_wait_queue_list);
+	}
+	spin_unlock(&data->futex_wait_queue_lock);

regards,
dan carpenter

#5 lore raw
Quoted 4 lines

drivers/staging/android/vsoc.c static int handle_vsoc_cond_wait(struct file *filp, struct vsoc_cond_wait *arg) · line 430

Quoted 2 lines
drivers/staging/android/vsoc.c +2 -0
+	list_for_each_entry(it, &data->futex_wait_queue_list, list) {
+		if (wait_queue->offset == arg->offset) {
                    ^^^^^^^^^^^^^^^^^^

You meant "it->offset".

Right, this is not good. Fixed in v2.

Thanks for the feedback!

regards,
Hugo

Signature · 3 lines
                Hugo Lefeuvre (hle)    |    www.owl.eu.com
RSA4096_ 360B 03B3 BF27 4F4D 7A3F D5E8 14AA 1EB8 A247 3DFD
ed25519_ 37B2 6D38 0B25 B8A2 6B9F 3A65 A36F 5357 5F2D DC4C
#3 lore raw
In reply to Greg Kroah-Hartman

Hi,

Quoted 1 line

Have you tested this?

I have finally set up a cuttlefish test env and tested both my first patch set[0] and this patch (v2).

My first patch set works fine. I have nothing to say about it.

Quoted 1 line

Noticed any performance speedups or slow downs?

This patch doesn't work.

The boot process goes well. Overall, calls to vsoc_cond_wake are executed slightly faster. However the system freezes at some point.

The graphical interface generates many calls to vsoc_cond_wake and wait, so I suspect an issue with the locks. I will investigate this and come back with an updated patch.

regards,
Hugo

[0] https://lore.kernel.org/patchwork/patch/1039712/

Signature · 3 lines
                Hugo Lefeuvre (hle)    |    www.owl.eu.com
RSA4096_ 360B 03B3 BF27 4F4D 7A3F D5E8 14AA 1EB8 A247 3DFD
ed25519_ 37B2 6D38 0B25 B8A2 6B9F 3A65 A36F 5357 5F2D DC4C