internal/runtime/maps: hash small keys with wyhash instead of AES

[mcy]

Replace UseAeshash with MinAeshashSize, which sets the smallest key size
at which we actually use AES hash. This tunable parameter can be
discovered using benchmarks, allowing for fine-tuning of a hybrid
wyhash/AES implementation to different microarchitectures.

It turns out that using AES for very small keys leaves a noticeable
amount of performance on the table, because SIMD instructions do not
like being used for one-offs. Unsurprisingly, the a block cipher wants
streaming inputs.

This CL includes a benchmark that can be used to calculate ideal cutoff
values for other targets. I've used the Zen 4 x86_64 implementtion for
this particular CL, but preparing similar changes keyed on stepping
numbers (e.g. to target Intel chips) should be a very easy and
worthwhile followup.

On my machine (Zen4, 7995WX), I selected the cutoff at 8 bytes, and
observed the following improvements across multiple linker-randomized
binaries:

HashInt64Speed                        4.79n  4.31n  -10.0%
MapAccessHit/int64/len=64             5.23n  4.91n   -6.0%
MapAccessHit/int64/len=65536          8.37n  7.80n   -6.8%
MapAccessMiss/int64/len=65536         7.75n  6.98n  -10.9%
MapAssignExists/int64/len=6           6.39n  5.70n  -10.7%
MapAccessHit/string/len=65536         12.3n  11.8n   -5.2%
MapAccessHit/int64/len=6              3.60n  3.60n    ~
GroupByInt64 (1M-row aggregation)     7.46m  6.38m  -14.6%

The final benchmark is new, and is intended to measure hash latency
specifically; existing benchmarks look good on AES because they focus on
throughput.

All other benchmarks do not show statistically significant changes as
reported by benchstat.

[gopherbot]

This PR (HEAD: 7a2e155) has been imported to Gerrit for code review.

Please visit Gerrit at https://go-review.googlesource.com/c/go/+/815960.

Important tips:

• Don't comment on this PR. All discussion takes place in Gerrit.
• You need a Gmail or other Google account to log in to Gerrit.
• To change your code in response to feedback:
  • Push a new commit to the branch used by your GitHub PR.
  • A new "patch set" will then appear in Gerrit.
  • Respond to each comment by marking as Done in Gerrit if implemented as suggested. You can alternatively write a reply.
  • Critical: you must click the blue Reply button near the top to publish your Gerrit responses.
  • Multiple commits in the PR will be squashed by GerritBot.
• The title and description of the GitHub PR are used to construct the final commit message.
  • Edit these as needed via the GitHub web interface (not via Gerrit or git).
  • You should word wrap the PR description at ~76 characters unless you need longer lines (e.g., for tables or URLs).
• See the Sending a change via GitHub and Reviews sections of the Contribution Guide as well as the FAQ for details.

[gopherbot]

Message from Gopher Robot:

Patch Set 1:

(1 comment)

---

Please don’t reply on this GitHub thread. Visit golang.org/cl/815960.
After addressing review feedback, remember to publish your drafts!

[gopherbot]

Message from Keith Randall:

Patch Set 2:

(1 comment)

---

Please don’t reply on this GitHub thread. Visit golang.org/cl/815960.
After addressing review feedback, remember to publish your drafts!

[gopherbot]

Message from Sunny Young de la Sota:

Patch Set 2:

(1 comment)

---

Please don’t reply on this GitHub thread. Visit golang.org/cl/815960.
After addressing review feedback, remember to publish your drafts!

[gopherbot]

Message from Keith Randall:

Patch Set 3: Auto-Submit+1 Code-Review+2 Commit-Queue+1

(1 comment)

---

Please don’t reply on this GitHub thread. Visit golang.org/cl/815960.
After addressing review feedback, remember to publish your drafts!

[gopherbot]

Message from golang-scoped@luci-project-accounts.iam.gserviceaccount.com:

Patch Set 3:

Dry run: CV is trying the patch.

Bot data: {"action":"start","triggered_at":"2026-08-17T21:17:28Z","revision":"fb52a28ee042aca8cf958e4d2404ae735ee2ab9d"}

---

Please don’t reply on this GitHub thread. Visit golang.org/cl/815960.
After addressing review feedback, remember to publish your drafts!

[gopherbot]

Message from Keith Randall:

Patch Set 3: Code-Review+1

---

Please don’t reply on this GitHub thread. Visit golang.org/cl/815960.
After addressing review feedback, remember to publish your drafts!

[gopherbot]

Message from Keith Randall:

Patch Set 3:

(1 comment)

---

Please don’t reply on this GitHub thread. Visit golang.org/cl/815960.
After addressing review feedback, remember to publish your drafts!

[gopherbot]

Message from Keith Randall:

Patch Set 3: -Commit-Queue

(Performed by <GERRIT_ACCOUNT_60063> on behalf of <GERRIT_ACCOUNT_5200>)

---

Please don’t reply on this GitHub thread. Visit golang.org/cl/815960.
After addressing review feedback, remember to publish your drafts!

[gopherbot]

Message from golang-scoped@luci-project-accounts.iam.gserviceaccount.com:

Patch Set 3:

This CL has passed the run

---

Please don’t reply on this GitHub thread. Visit golang.org/cl/815960.
After addressing review feedback, remember to publish your drafts!

[gopherbot]

Message from golang-scoped@luci-project-accounts.iam.gserviceaccount.com:

Patch Set 3: LUCI-TryBot-Result+1

---

Please don’t reply on this GitHub thread. Visit golang.org/cl/815960.
After addressing review feedback, remember to publish your drafts!

[gopherbot]

Message from Sunny Young de la Sota:

Patch Set 3:

(1 comment)

---

Please don’t reply on this GitHub thread. Visit golang.org/cl/815960.
After addressing review feedback, remember to publish your drafts!

[gopherbot]

Message from Sunny Young de la Sota:

Patch Set 3:

(1 comment)

---

Please don’t reply on this GitHub thread. Visit golang.org/cl/815960.
After addressing review feedback, remember to publish your drafts!