2023-02-02 01:24:12

by Nathan Huckleberry

Subject: [PATCH] dm-verity: Remove WQ_UNBOUND.

Setting WQ_UNBOUND increases scheduler latency on ARM64. This is likely
due to the asymmetric architecture of ARM64 processors.

I've been unable to reproduce the results that claim WQ_UNBOUND gives a
performance boost on x86-64.

This flag is causing performance issues for multiple subsystems within
Android. Notably, the same slowdown exists for decompression with
EROFS.

| open-prebuilt-camera | WQ_UNBOUND | ~WQ_UNBOUND |
|-----------------------|------------|---------------|
| verity wait time (us) | 11746 | 119 (-98%) |
| erofs wait time (us) | 357805 | 174205 (-51%) |

| sha256 ramdisk random read | WQ_UNBOUND | ~WQ_UNBOUND |
|----------------------------|-----------=---|-------------|
| arm64 (accelerated) | bw=42.4MiB/s | bw=212MiB/s |
| arm64 (generic) | bw=16.5MiB/s | bw=48MiB/s |
| x86_64 (generic) | bw=233MiB/s | bw=230MiB/s |

Cc: Sami Tolvanen <[email protected]>
Cc: Eric Biggers <[email protected]>
Signed-off-by: Nathan Huckleberry <[email protected]>
---
drivers/md/dm-verity-target.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/md/dm-verity-target.c b/drivers/md/dm-verity-target.c
index ccf5b852fbf7..020fd2341025 100644
--- a/drivers/md/dm-verity-target.c
+++ b/drivers/md/dm-verity-target.c
@@ -1399,8 +1399,8 @@ static int verity_ctr(struct dm_target *ti, unsigned argc, char **argv)
goto bad;
}

- /* WQ_UNBOUND greatly improves performance when running on ramdisk */
- wq_flags = WQ_MEM_RECLAIM | WQ_UNBOUND;
+ wq_flags = WQ_MEM_RECLAIM;
+
/*
* Using WQ_HIGHPRI improves throughput and completion latency by
* reducing wait times when reading from a dm-verity device.
--
2.39.1.456.gfc5497dd1b-goog

2023-02-02 16:42:24

by Mike Snitzer

Subject: Re: dm-verity: Remove WQ_UNBOUND.

On Wed, Feb 01 2023 at 8:23P -0500,
Nathan Huckleberry <[email protected]> wrote:

> Setting WQ_UNBOUND increases scheduler latency on ARM64. This is likely
> due to the asymmetric architecture of ARM64 processors.
>
> I've been unable to reproduce the results that claim WQ_UNBOUND gives a
> performance boost on x86-64.
>
> This flag is causing performance issues for multiple subsystems within
> Android. Notably, the same slowdown exists for decompression with
> EROFS.
>
> | open-prebuilt-camera | WQ_UNBOUND | ~WQ_UNBOUND |
> |-----------------------|------------|---------------|
> | verity wait time (us) | 11746 | 119 (-98%) |
> | erofs wait time (us) | 357805 | 174205 (-51%) |
>
> | sha256 ramdisk random read | WQ_UNBOUND | ~WQ_UNBOUND |
> |----------------------------|-----------=---|-------------|
> | arm64 (accelerated) | bw=42.4MiB/s | bw=212MiB/s |
> | arm64 (generic) | bw=16.5MiB/s | bw=48MiB/s |
> | x86_64 (generic) | bw=233MiB/s | bw=230MiB/s |
>
> Cc: Sami Tolvanen <[email protected]>
> Cc: Eric Biggers <[email protected]>
> Signed-off-by: Nathan Huckleberry <[email protected]>
> ---
> drivers/md/dm-verity-target.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/md/dm-verity-target.c b/drivers/md/dm-verity-target.c
> index ccf5b852fbf7..020fd2341025 100644
> --- a/drivers/md/dm-verity-target.c
> +++ b/drivers/md/dm-verity-target.c
> @@ -1399,8 +1399,8 @@ static int verity_ctr(struct dm_target *ti, unsigned argc, char **argv)
> goto bad;
> }
>
> - /* WQ_UNBOUND greatly improves performance when running on ramdisk */
> - wq_flags = WQ_MEM_RECLAIM | WQ_UNBOUND;
> + wq_flags = WQ_MEM_RECLAIM;
> +
> /*
> * Using WQ_HIGHPRI improves throughput and completion latency by
> * reducing wait times when reading from a dm-verity device.
> --
> 2.39.1.456.gfc5497dd1b-goog

Hi,

I've discussed with Mikulas, tweaked your patch slightly but accepted
your change, please see:

https://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm.git/commit/?h=dm-6.3&id=6f30cc248507ee96c22ff4c3cbc86099ff12b7a9

Thanks,
Mike

2023-02-02 19:11:34

by Nathan Huckleberry

Subject: Re: dm-verity: Remove WQ_UNBOUND.

Looks good. Thanks.

On Thu, Feb 2, 2023 at 8:41 AM Mike Snitzer <[email protected]> wrote:
>
> On Wed, Feb 01 2023 at 8:23P -0500,
> Nathan Huckleberry <[email protected]> wrote:
>
> > Setting WQ_UNBOUND increases scheduler latency on ARM64. This is likely
> > due to the asymmetric architecture of ARM64 processors.
> >
> > I've been unable to reproduce the results that claim WQ_UNBOUND gives a
> > performance boost on x86-64.
> >
> > This flag is causing performance issues for multiple subsystems within
> > Android. Notably, the same slowdown exists for decompression with
> > EROFS.
> >
> > | open-prebuilt-camera | WQ_UNBOUND | ~WQ_UNBOUND |
> > |-----------------------|------------|---------------|
> > | verity wait time (us) | 11746 | 119 (-98%) |
> > | erofs wait time (us) | 357805 | 174205 (-51%) |
> >
> > | sha256 ramdisk random read | WQ_UNBOUND | ~WQ_UNBOUND |
> > |----------------------------|-----------=---|-------------|
> > | arm64 (accelerated) | bw=42.4MiB/s | bw=212MiB/s |
> > | arm64 (generic) | bw=16.5MiB/s | bw=48MiB/s |
> > | x86_64 (generic) | bw=233MiB/s | bw=230MiB/s |
> >
> > Cc: Sami Tolvanen <[email protected]>
> > Cc: Eric Biggers <[email protected]>
> > Signed-off-by: Nathan Huckleberry <[email protected]>
> > ---
> > drivers/md/dm-verity-target.c | 4 ++--
> > 1 file changed, 2 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/md/dm-verity-target.c b/drivers/md/dm-verity-target.c
> > index ccf5b852fbf7..020fd2341025 100644
> > --- a/drivers/md/dm-verity-target.c
> > +++ b/drivers/md/dm-verity-target.c
> > @@ -1399,8 +1399,8 @@ static int verity_ctr(struct dm_target *ti, unsigned argc, char **argv)
> > goto bad;
> > }
> >
> > - /* WQ_UNBOUND greatly improves performance when running on ramdisk */
> > - wq_flags = WQ_MEM_RECLAIM | WQ_UNBOUND;
> > + wq_flags = WQ_MEM_RECLAIM;
> > +
> > /*
> > * Using WQ_HIGHPRI improves throughput and completion latency by
> > * reducing wait times when reading from a dm-verity device.
> > --
> > 2.39.1.456.gfc5497dd1b-goog
>
> Hi,
>
> I've discussed with Mikulas, tweaked your patch slightly but accepted
> your change, please see:
>
> https://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm.git/commit/?h=dm-6.3&id=6f30cc248507ee96c22ff4c3cbc86099ff12b7a9
>
> Thanks,
> Mike

2023-02-07 08:17:47

by Mikulas Patocka

Subject: Re: [dm-devel] [PATCH] dm-verity: Remove WQ_UNBOUND.

On Wed, 1 Feb 2023, Nathan Huckleberry wrote:

> Setting WQ_UNBOUND increases scheduler latency on ARM64. This is likely
> due to the asymmetric architecture of ARM64 processors.
>
> I've been unable to reproduce the results that claim WQ_UNBOUND gives a
> performance boost on x86-64.
>
> This flag is causing performance issues for multiple subsystems within
> Android. Notably, the same slowdown exists for decompression with
> EROFS.
>
> | open-prebuilt-camera | WQ_UNBOUND | ~WQ_UNBOUND |
> |-----------------------|------------|---------------|
> | verity wait time (us) | 11746 | 119 (-98%) |
> | erofs wait time (us) | 357805 | 174205 (-51%) |
>
> | sha256 ramdisk random read | WQ_UNBOUND | ~WQ_UNBOUND |
> |----------------------------|-----------=---|-------------|
> | arm64 (accelerated) | bw=42.4MiB/s | bw=212MiB/s |
> | arm64 (generic) | bw=16.5MiB/s | bw=48MiB/s |
> | x86_64 (generic) | bw=233MiB/s | bw=230MiB/s |
>
> Cc: Sami Tolvanen <[email protected]>
> Cc: Eric Biggers <[email protected]>
> Signed-off-by: Nathan Huckleberry <[email protected]>
> ---
> drivers/md/dm-verity-target.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/md/dm-verity-target.c b/drivers/md/dm-verity-target.c
> index ccf5b852fbf7..020fd2341025 100644
> --- a/drivers/md/dm-verity-target.c
> +++ b/drivers/md/dm-verity-target.c
> @@ -1399,8 +1399,8 @@ static int verity_ctr(struct dm_target *ti, unsigned argc, char **argv)
> goto bad;
> }
>
> - /* WQ_UNBOUND greatly improves performance when running on ramdisk */
> - wq_flags = WQ_MEM_RECLAIM | WQ_UNBOUND;
> + wq_flags = WQ_MEM_RECLAIM;
> +
> /*
> * Using WQ_HIGHPRI improves throughput and completion latency by
> * reducing wait times when reading from a dm-verity device.

Hi

If you remove WQ_UNBOUND, you should also change the last argument of
alloc_workqueue from num_online_cpus() to either 0 or 1. Try both 0 and 1
and tell us which performs better.

Mikulas

2023-02-07 22:54:39

by Nathan Huckleberry

Subject: Re: [dm-devel] [PATCH] dm-verity: Remove WQ_UNBOUND.

On Tue, Feb 7, 2023 at 12:16 AM Mikulas Patocka <[email protected]> wrote:
>
>
>
> On Wed, 1 Feb 2023, Nathan Huckleberry wrote:
>
> > Setting WQ_UNBOUND increases scheduler latency on ARM64. This is likely
> > due to the asymmetric architecture of ARM64 processors.
> >
> > I've been unable to reproduce the results that claim WQ_UNBOUND gives a
> > performance boost on x86-64.
> >
> > This flag is causing performance issues for multiple subsystems within
> > Android. Notably, the same slowdown exists for decompression with
> > EROFS.
> >
> > | open-prebuilt-camera | WQ_UNBOUND | ~WQ_UNBOUND |
> > |-----------------------|------------|---------------|
> > | verity wait time (us) | 11746 | 119 (-98%) |
> > | erofs wait time (us) | 357805 | 174205 (-51%) |
> >
> > | sha256 ramdisk random read | WQ_UNBOUND | ~WQ_UNBOUND |
> > |----------------------------|-----------=---|-------------|
> > | arm64 (accelerated) | bw=42.4MiB/s | bw=212MiB/s |
> > | arm64 (generic) | bw=16.5MiB/s | bw=48MiB/s |
> > | x86_64 (generic) | bw=233MiB/s | bw=230MiB/s |
> >
> > Cc: Sami Tolvanen <[email protected]>
> > Cc: Eric Biggers <[email protected]>
> > Signed-off-by: Nathan Huckleberry <[email protected]>
> > ---
> > drivers/md/dm-verity-target.c | 4 ++--
> > 1 file changed, 2 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/md/dm-verity-target.c b/drivers/md/dm-verity-target.c
> > index ccf5b852fbf7..020fd2341025 100644
> > --- a/drivers/md/dm-verity-target.c
> > +++ b/drivers/md/dm-verity-target.c
> > @@ -1399,8 +1399,8 @@ static int verity_ctr(struct dm_target *ti, unsigned argc, char **argv)
> > goto bad;
> > }
> >
> > - /* WQ_UNBOUND greatly improves performance when running on ramdisk */
> > - wq_flags = WQ_MEM_RECLAIM | WQ_UNBOUND;
> > + wq_flags = WQ_MEM_RECLAIM;
> > +
> > /*
> > * Using WQ_HIGHPRI improves throughput and completion latency by
> > * reducing wait times when reading from a dm-verity device.
>
> Hi
>
> If you remove WQ_UNBOUND, you should also change the last argument of
> alloc_workqueue from num_online_cpus() to either 0 or 1. Try both 0 and 1
> and tell us which performs better.

They look roughly the same on ARM64. There's a slight advantage for
using 1, but they're too close to conclusively say one is better.

Thanks,
Huck

>
> Mikulas
>