Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

patch: runtime: TailscaleCurrentP #109

Open
bradfitz opened this issue Dec 18, 2024 · 0 comments
Open

patch: runtime: TailscaleCurrentP #109

bradfitz opened this issue Dec 18, 2024 · 0 comments
Labels
tsgo

Comments

@bradfitz
Copy link
Member

For tailscale/tailscale#14412 for sharded counters that have high write load, we want a ~"current CPU" hint, even if it's outdated by the time we use it.

+func TailscaleCurrentP() int {
+       return int(getg().m.p.ptr().id)
+}

/cc @raggi
@bradfitz bradfitz added the tsgo label Dec 18, 2024
bradfitz added a commit that referenced this issue Dec 18, 2024
@bradfitz
runtime: add func TailscaleCurrentP
3f910f6 
Updates #109

Signed-off-by: Brad Fitzpatrick <[email protected]>
@bradfitz bradfitz mentioned this issue Dec 18, 2024
Merged
bradfitz added a commit that referenced this issue Dec 18, 2024
@bradfitz
runtime: add func TailscaleCurrentP
161c3b7 
Updates #109

Signed-off-by: Brad Fitzpatrick <[email protected]>
raggi added a commit to tailscale/tailscale that referenced this issue Dec 19, 2024
@raggi
syncs: add ShardedInt expvar.Var type
9cfa369 
ShardedInt provides an int type expvar.Var that supports more efficient
writes at high frequencies (one order of magnigude on an M1 Max, much
more on NUMA systems).

There are two implementations of ShardValue, one that abuses sync.Pool
that will work on current public Go versions, and one that takes a
dependency on a runtime.TailscaleP function exposed in Tailscale's Go
fork. The sync.Pool variant has about 10x the throughput of a single
atomic integer on an M1 Max, and the runtime.TailscaleP variant is about
10x faster than the sync.Pool variant.

Neither variant have perfect distribution, or perfectly always avoid
cross-CPU sharing, as there is no locking or affinity to ensure that the
time of yield is on the same core as the time of core biasing, but in
the average case the distributions are enough to provide substantially
better performance.

See golang/go#18802 for a related upstream proposal.

Updates tailscale/go#109
Updates tailscale/corp#25450

Signed-off-by: James Tucker <[email protected]>
@raggi raggi mentioned this issue Dec 19, 2024
Merged
raggi added a commit to tailscale/tailscale that referenced this issue Dec 19, 2024
@raggi
syncs: add ShardedInt expvar.Var type
a3297dd 
ShardedInt provides an int type expvar.Var that supports more efficient
writes at high frequencies (one order of magnigude on an M1 Max, much
more on NUMA systems).

There are two implementations of ShardValue, one that abuses sync.Pool
that will work on current public Go versions, and one that takes a
dependency on a runtime.TailscaleP function exposed in Tailscale's Go
fork. The sync.Pool variant has about 10x the throughput of a single
atomic integer on an M1 Max, and the runtime.TailscaleP variant is about
10x faster than the sync.Pool variant.

Neither variant have perfect distribution, or perfectly always avoid
cross-CPU sharing, as there is no locking or affinity to ensure that the
time of yield is on the same core as the time of core biasing, but in
the average case the distributions are enough to provide substantially
better performance.

See golang/go#18802 for a related upstream proposal.

Updates tailscale/go#109
Updates tailscale/corp#25450

Signed-off-by: James Tucker <[email protected]>
raggi added a commit to tailscale/tailscale that referenced this issue Dec 19, 2024
@raggi
syncs: add ShardedInt expvar.Var type
7e42703 
ShardedInt provides an int type expvar.Var that supports more efficient
writes at high frequencies (one order of magnigude on an M1 Max, much
more on NUMA systems).

There are two implementations of ShardValue, one that abuses sync.Pool
that will work on current public Go versions, and one that takes a
dependency on a runtime.TailscaleP function exposed in Tailscale's Go
fork. The sync.Pool variant has about 10x the throughput of a single
atomic integer on an M1 Max, and the runtime.TailscaleP variant is about
10x faster than the sync.Pool variant.

Neither variant have perfect distribution, or perfectly always avoid
cross-CPU sharing, as there is no locking or affinity to ensure that the
time of yield is on the same core as the time of core biasing, but in
the average case the distributions are enough to provide substantially
better performance.

See golang/go#18802 for a related upstream proposal.

Updates tailscale/go#109
Updates tailscale/corp#25450

Signed-off-by: James Tucker <[email protected]>
raggi added a commit to tailscale/tailscale that referenced this issue Dec 19, 2024
@raggi
syncs: add ShardedInt expvar.Var type
e8f1721 
ShardedInt provides an int type expvar.Var that supports more efficient
writes at high frequencies (one order of magnigude on an M1 Max, much
more on NUMA systems).

There are two implementations of ShardValue, one that abuses sync.Pool
that will work on current public Go versions, and one that takes a
dependency on a runtime.TailscaleP function exposed in Tailscale's Go
fork. The sync.Pool variant has about 10x the throughput of a single
atomic integer on an M1 Max, and the runtime.TailscaleP variant is about
10x faster than the sync.Pool variant.

Neither variant have perfect distribution, or perfectly always avoid
cross-CPU sharing, as there is no locking or affinity to ensure that the
time of yield is on the same core as the time of core biasing, but in
the average case the distributions are enough to provide substantially
better performance.

See golang/go#18802 for a related upstream proposal.

Updates tailscale/go#109
Updates tailscale/corp#25450

Signed-off-by: James Tucker <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
tsgo
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@bradfitz