flate: improve huffman flate hcode spatial locality

[teivah]

Improving huffman hcode spatial locality and optimize accesses when iterating over len. For example:

go/src/compress/flate/huffman_bit_writer.go

Lines 211 to 213 in 4469f54

	for i := range cgnl {
	cgnl[i] = uint8(litEnc.codes[i].len)
	}

I would propose that instead of having in huffmanEncoder a slice of hcode, to have an hcode struct holding a slice of code and a slice of len. It would optimize the utilization of the CPU cache lines.

Here are the results I'm getting while comparing both benchmarks locally (x86) with benstat:

name                         old time/op    new time/op    delta
Decode/Digits/Huffman/1e4-4     112µs ± 0%     123µs ± 0%   ~     (p=1.000 n=1+1)
Decode/Digits/Huffman/1e5-4    1.07ms ± 0%    1.06ms ± 0%   ~     (p=1.000 n=1+1)
Decode/Digits/Huffman/1e6-4    10.7ms ± 0%    10.5ms ± 0%   ~     (p=1.000 n=1+1)
Decode/Newton/Huffman/1e4-4     131µs ± 0%     124µs ± 0%   ~     (p=1.000 n=1+1)
Decode/Newton/Huffman/1e5-4    1.21ms ± 0%    1.17ms ± 0%   ~     (p=1.000 n=1+1)
Decode/Newton/Huffman/1e6-4    11.5ms ± 0%    11.4ms ± 0%   ~     (p=1.000 n=1+1)
Encode/Digits/Huffman/1e4-4    45.5µs ± 0%    34.1µs ± 0%   ~     (p=1.000 n=1+1)
Encode/Digits/Huffman/1e5-4     445µs ± 0%     322µs ± 0%   ~     (p=1.000 n=1+1)
Encode/Digits/Huffman/1e6-4    4.32ms ± 0%    3.16ms ± 0%   ~     (p=1.000 n=1+1)
Encode/Newton/Huffman/1e4-4    59.6µs ± 0%    50.0µs ± 0%   ~     (p=1.000 n=1+1)
Encode/Newton/Huffman/1e5-4     507µs ± 0%     390µs ± 0%   ~     (p=1.000 n=1+1)
Encode/Newton/Huffman/1e6-4    5.00ms ± 0%    3.86ms ± 0%   ~     (p=1.000 n=1+1)

name                         old speed      new speed      delta
Decode/Digits/Huffman/1e4-4  89.6MB/s ± 0%  81.5MB/s ± 0%   ~     (p=1.000 n=1+1)
Decode/Digits/Huffman/1e5-4  93.2MB/s ± 0%  94.7MB/s ± 0%   ~     (p=1.000 n=1+1)
Decode/Digits/Huffman/1e6-4  93.4MB/s ± 0%  95.2MB/s ± 0%   ~     (p=1.000 n=1+1)
Decode/Newton/Huffman/1e4-4  76.5MB/s ± 0%  81.0MB/s ± 0%   ~     (p=1.000 n=1+1)
Decode/Newton/Huffman/1e5-4  82.9MB/s ± 0%  85.5MB/s ± 0%   ~     (p=1.000 n=1+1)
Decode/Newton/Huffman/1e6-4  86.7MB/s ± 0%  87.8MB/s ± 0%   ~     (p=1.000 n=1+1)
Encode/Digits/Huffman/1e4-4   220MB/s ± 0%   293MB/s ± 0%   ~     (p=1.000 n=1+1)
Encode/Digits/Huffman/1e5-4   225MB/s ± 0%   311MB/s ± 0%   ~     (p=1.000 n=1+1)
Encode/Digits/Huffman/1e6-4   232MB/s ± 0%   316MB/s ± 0%   ~     (p=1.000 n=1+1)
Encode/Newton/Huffman/1e4-4   168MB/s ± 0%   200MB/s ± 0%   ~     (p=1.000 n=1+1)
Encode/Newton/Huffman/1e5-4   197MB/s ± 0%   257MB/s ± 0%   ~     (p=1.000 n=1+1)
Encode/Newton/Huffman/1e6-4   200MB/s ± 0%   259MB/s ± 0%   ~     (p=1.000 n=1+1)

name                         old alloc/op   new alloc/op   delta
Decode/Digits/Huffman/1e4-4    40.5kB ± 0%    40.5kB ± 0%   ~     (all equal)
Decode/Digits/Huffman/1e5-4    40.5kB ± 0%    40.5kB ± 0%   ~     (all equal)
Decode/Digits/Huffman/1e6-4    40.6kB ± 0%    40.6kB ± 0%   ~     (all equal)
Decode/Newton/Huffman/1e4-4    41.2kB ± 0%    41.2kB ± 0%   ~     (all equal)
Decode/Newton/Huffman/1e5-4    45.0kB ± 0%    45.0kB ± 0%   ~     (p=1.000 n=1+1)
Decode/Newton/Huffman/1e6-4    79.9kB ± 0%    79.9kB ± 0%   ~     (all equal)
Encode/Digits/Huffman/1e4-4     0.00B          0.00B        ~     (all equal)
Encode/Digits/Huffman/1e5-4     1.00B ± 0%     1.00B ± 0%   ~     (all equal)
Encode/Digits/Huffman/1e6-4     16.0B ± 0%     12.0B ± 0%   ~     (p=1.000 n=1+1)
Encode/Newton/Huffman/1e4-4     0.00B          0.00B        ~     (all equal)
Encode/Newton/Huffman/1e5-4     1.00B ± 0%     1.00B ± 0%   ~     (all equal)
Encode/Newton/Huffman/1e6-4     19.0B ± 0%     14.0B ± 0%   ~     (p=1.000 n=1+1)

name                         old allocs/op  new allocs/op  delta
Decode/Digits/Huffman/1e4-4      5.00 ± 0%      5.00 ± 0%   ~     (all equal)
Decode/Digits/Huffman/1e5-4      5.00 ± 0%      5.00 ± 0%   ~     (all equal)
Decode/Digits/Huffman/1e6-4      5.00 ± 0%      5.00 ± 0%   ~     (all equal)
Decode/Newton/Huffman/1e4-4      14.0 ± 0%      14.0 ± 0%   ~     (all equal)
Decode/Newton/Huffman/1e5-4      23.0 ± 0%      23.0 ± 0%   ~     (all equal)
Decode/Newton/Huffman/1e6-4       161 ± 0%       161 ± 0%   ~     (all equal)
Encode/Digits/Huffman/1e4-4      0.00           0.00        ~     (all equal)
Encode/Digits/Huffman/1e5-4      0.00           0.00        ~     (all equal)
Encode/Digits/Huffman/1e6-4      0.00           0.00        ~     (all equal)
Encode/Newton/Huffman/1e4-4      0.00           0.00        ~     (all equal)
Encode/Newton/Huffman/1e5-4      0.00           0.00        ~     (all equal)
Encode/Newton/Huffman/1e6-4      0.00           0.00        ~     (all equal)

[google-cla]

Thanks for your pull request. It looks like this may be your first contribution to a Google open source project (if not, look below for help). Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).

📝 Please visit https://cla.developers.google.com/ to sign.

Once you've signed (or fixed any issues), please reply here with @googlebot I signed it! and we'll verify it.

---

What to do if you already signed the CLA

Individual signers

• It's possible we don't have your GitHub username or you're using a different email address on your commit. Check your existing CLA data and verify that your email is set on your git commits.

Corporate signers

• Your company has a Point of Contact who decides which employees are authorized to participate. Ask your POC to be added to the group of authorized contributors. If you don't know who your Point of Contact is, direct the Google project maintainer to go/cla#troubleshoot (Public version).
• The email used to register you as an authorized contributor must be the email used for the Git commit. Check your existing CLA data and verify that your email is set on your git commits.
• The email used to register you as an authorized contributor must also be attached to your GitHub account.

ℹ️ Googlers: Go here for more info.

[teivah]

@googlebot I signed it!

[gopherbot]

This PR (HEAD: 05bf781) has been imported to Gerrit for code review.

Please visit https://go-review.googlesource.com/c/go/+/317789 to see it.

Tip: You can toggle comments from me using the comments slash command (e.g. /comments off)
See the Wiki page for more info

[gopherbot]

Message from Go Bot:

Patch Set 1:

Congratulations on opening your first change. Thank you for your contribution!

Next steps:
A maintainer will review your change and provide feedback. See
https://golang.org/doc/contribute.html#review for more info and tips to get your
patch through code review.

Most changes in the Go project go through a few rounds of revision. This can be
surprising to people new to the project. The careful, iterative review process
is our way of helping mentor contributors and ensuring that their contributions
have a lasting impact.

During May-July and Nov-Jan the Go project is in a code freeze, during which
little code gets reviewed or merged. If a reviewer responds with a comment like
R=go1.11 or adds a tag like "wait-release", it means that this CL will be
reviewed as part of the next development cycle. See https://golang.org/s/release
for more details.

---

Please don’t reply on this GitHub thread. Visit golang.org/cl/317789.
After addressing review feedback, remember to publish your drafts!

[gopherbot]

Message from Joe Tsai:

Patch Set 3:

(1 comment)

---

Please don’t reply on this GitHub thread. Visit golang.org/cl/317789.
After addressing review feedback, remember to publish your drafts!

[gopherbot]

Message from Teiva Harsanyi:

Patch Set 3:

(1 comment)

---

Please don’t reply on this GitHub thread. Visit golang.org/cl/317789.
After addressing review feedback, remember to publish your drafts!