mleku/next.orly.dev
1
0
Fork 1
Code Issues 2 Pull Requests Actions Packages Projects Releases Wiki Activity

Add NIP-11 relay synchronization and group management features
Some checks failed
Go / build (push) Has been cancelled
Details
Go / release (push) Has been cancelled
Details

Browse Source 
- Introduced a new `sync` package for managing NIP-11 relay information and relay group configurations.
- Implemented a cache for NIP-11 documents, allowing retrieval of relay public keys and authoritative configurations.
- Enhanced the sync manager to update peer lists based on authoritative configurations from relay group events.
- Updated event handling to incorporate policy checks during event imports, ensuring compliance with relay rules.
- Refactored various components to utilize the new `sha256-simd` package for improved performance.
- Added comprehensive tests to validate the new synchronization and group management functionalities.
- Bumped version to v0.24.1 to reflect these changes.
This commit is contained in:
mleku
2025-11-03 18:17:15 +00:00
parent e161d0e4be
commit e56bf76257
83 changed files with 3712 additions and 7417 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
pkg/crypto/ec/base58/base58check.go
View File

@@ -7,7 +7,7 @@ package base58
import (
"errors"
"next.orly.dev/pkg/crypto/sha256"
"github.com/minio/sha256-simd"
)
// ErrChecksum indicates that the checksum of a check-encoded string does not verify against

2
pkg/crypto/ec/chainhash/hash.go
View File

@@ -9,7 +9,7 @@ import (
"encoding/json"
"fmt"
"next.orly.dev/pkg/crypto/sha256"
"github.com/minio/sha256-simd"
"next.orly.dev/pkg/encoders/hex"
)

2
pkg/crypto/ec/chainhash/hashfuncs.go
View File

@@ -6,7 +6,7 @@
package chainhash
import (
"next.orly.dev/pkg/crypto/sha256"
"github.com/minio/sha256-simd"
)
// HashB calculates hash(b) and returns the resulting bytes.

2
pkg/crypto/ec/musig2/musig2_test.go
View File

@@ -9,7 +9,7 @@ import (
"testing"
"next.orly.dev/pkg/crypto/ec"
"next.orly.dev/pkg/crypto/sha256"
"github.com/minio/sha256-simd"
"next.orly.dev/pkg/encoders/hex"
)

2
pkg/crypto/ec/schnorr/bench_test.go
View File

@@ -11,7 +11,7 @@ import (
"next.orly.dev/pkg/crypto/ec"
"next.orly.dev/pkg/crypto/ec/secp256k1"
"next.orly.dev/pkg/crypto/sha256"
"github.com/minio/sha256-simd"
"next.orly.dev/pkg/encoders/hex"
)

2
pkg/crypto/ec/secp256k1/example_test.go
View File

@@ -12,7 +12,7 @@ import (
"fmt"
"next.orly.dev/pkg/crypto/ec/secp256k1"
"next.orly.dev/pkg/crypto/sha256"
"github.com/minio/sha256-simd"
"next.orly.dev/pkg/encoders/hex"
)

2
pkg/crypto/ec/secp256k1/nonce.go
View File

@@ -9,7 +9,7 @@ import (
"bytes"
"hash"
"next.orly.dev/pkg/crypto/sha256"
"github.com/minio/sha256-simd"
)
// References:

2
pkg/crypto/ec/secp256k1/nonce_test.go
View File

@@ -8,7 +8,7 @@ package secp256k1
import (
"testing"
"next.orly.dev/pkg/crypto/sha256"
"github.com/minio/sha256-simd"
"next.orly.dev/pkg/encoders/hex"
"next.orly.dev/pkg/utils"
)

2
pkg/crypto/encryption/nip44.go
View File

@@ -13,7 +13,7 @@ import (
"lol.mleku.dev/chk"
"lol.mleku.dev/errorf"
p256k1signer "p256k1.mleku.dev/signer"
"next.orly.dev/pkg/crypto/sha256"
"github.com/minio/sha256-simd"
"next.orly.dev/pkg/encoders/hex"
"next.orly.dev/pkg/interfaces/signer"
"next.orly.dev/pkg/utils"

56
pkg/crypto/encryption/nip44_test.go
View File

@@ -10,7 +10,7 @@ import (
"github.com/stretchr/testify/assert"
"lol.mleku.dev/chk"
"next.orly.dev/pkg/crypto/keys"
"next.orly.dev/pkg/crypto/sha256"
"github.com/minio/sha256-simd"
"next.orly.dev/pkg/encoders/hex"
)
@@ -258,10 +258,10 @@ func TestCryptPriv001(t *testing.T) {
t,
"0000000000000000000000000000000000000000000000000000000000000001",
"0000000000000000000000000000000000000000000000000000000000000002",
"c41c775356fd92eadc63ff5a0dc1da211b268cbea22316767095b2871ea1412d",
"d927e07202f86f1175e9dfc90fbbcd61963c5ee2506a10654641a826dd371a1b",
"0000000000000000000000000000000000000000000000000000000000000001",
"a",
"AgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAABee0G5VSK0/9YypIObAtDKfYEAjD35uVkHyB0F4DwrcNaCXlCWZKaArsGrY6M9wnuTMxWfp1RTN9Xga8no+kF5Vsb",
"AgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAB4ZAC1J9dJuHPtWNca8rycgBrU2S0ClwfvXjrTr0BZSm54UFqMJpt2easxakffyhgWf/PrUrSLJHJg1cfJ/MAh/Wy",
)
}
@@ -643,7 +643,7 @@ func TestConversationKey001(t *testing.T) {
t,
"315e59ff51cb9209768cf7da80791ddcaae56ac9775eb25b6dee1234bc5d2268",
"c2f9d9948dc8c7c38321e4b85c8558872eafa0641cd269db76848a6073e69133",
"3dfef0ce2a4d80a25e7a328accf73448ef67096f65f79588e358d9a0eb9013f1",
"8bc1eda9f0bd37d986c4cda4872af3409d8efbf4ff93e6ab61c3cc035cc06365",
)
}
@@ -652,7 +652,7 @@ func TestConversationKey002(t *testing.T) {
t,
"a1e37752c9fdc1273be53f68c5f74be7c8905728e8de75800b94262f9497c86e",
"03bb7947065dde12ba991ea045132581d0954f042c84e06d8c00066e23c1a800",
"4d14f36e81b8452128da64fe6f1eae873baae2f444b02c950b90e43553f2178b",
"217cdcc158edaa9ebac91af882353ffc0372b450c135315c245e48ffa23efdf7",
)
}
@@ -661,7 +661,7 @@ func TestConversationKey003(t *testing.T) {
t,
"98a5902fd67518a0c900f0fb62158f278f94a21d6f9d33d30cd3091195500311",
"aae65c15f98e5e677b5050de82e3aba47a6fe49b3dab7863cf35d9478ba9f7d1",
"9c00b769d5f54d02bf175b7284a1cbd28b6911b06cda6666b2243561ac96bad7",
"17540957c96b901bd4d665ad7b33ac6144793c024f050ba460f975f1bf952b6e",
)
}
@@ -670,7 +670,7 @@ func TestConversationKey004(t *testing.T) {
t,
"86ae5ac8034eb2542ce23ec2f84375655dab7f836836bbd3c54cefe9fdc9c19f",
"59f90272378089d73f1339710c02e2be6db584e9cdbe86eed3578f0c67c23585",
"19f934aafd3324e8415299b64df42049afaa051c71c98d0aa10e1081f2e3e2ba",
"7c4af2456b151d0966b64e9e462bee907b92a3f6d253882556c254fc11c9140f",
)
}
@@ -679,7 +679,7 @@ func TestConversationKey005(t *testing.T) {
t,
"2528c287fe822421bc0dc4c3615878eb98e8a8c31657616d08b29c00ce209e34",
"f66ea16104c01a1c532e03f166c5370a22a5505753005a566366097150c6df60",
"c833bbb292956c43366145326d53b955ffb5da4e4998a2d853611841903f5442",
"652493c2472a24794907b8bdfb7dc8e56ea2022e607918ca6f9e170e9f1886bc",
)
}
@@ -688,7 +688,7 @@ func TestConversationKey006(t *testing.T) {
t,
"49808637b2d21129478041813aceb6f2c9d4929cd1303cdaf4fbdbd690905ff2",
"74d2aab13e97827ea21baf253ad7e39b974bb2498cc747cdb168582a11847b65",
"4bf304d3c8c4608864c0fe03890b90279328cd24a018ffa9eb8f8ccec06b505d",
"7f186c96ebdcb32e6ad374d33303f2d618aad43a8f965a3392ac3cb1d0e85110",
)
}
@@ -697,7 +697,7 @@ func TestConversationKey007(t *testing.T) {
t,
"af67c382106242c5baabf856efdc0629cc1c5b4061f85b8ceaba52aa7e4b4082",
"bdaf0001d63e7ec994fad736eab178ee3c2d7cfc925ae29f37d19224486db57b",
"a3a575dd66d45e9379904047ebfb9a7873c471687d0535db00ef2daa24b391db",
"8d4f18de53fdae5aa404547764429674f5075e589790947e248a1dcf4b867697",
)
}
@@ -706,7 +706,7 @@ func TestConversationKey008(t *testing.T) {
t,
"0e44e2d1db3c1717b05ffa0f08d102a09c554a1cbbf678ab158b259a44e682f1",
"1ffa76c5cc7a836af6914b840483726207cb750889753d7499fb8b76aa8fe0de",
"a39970a667b7f861f100e3827f4adbf6f464e2697686fe1a81aeda817d6b8bdf",
"2d90b6069def88c4fce31c28d3d9ec8328bc6893d1c5dd02235f403af7ea5540",
)
}
@@ -715,7 +715,7 @@ func TestConversationKey009(t *testing.T) {
t,
"5fc0070dbd0666dbddc21d788db04050b86ed8b456b080794c2a0c8e33287bb6",
"31990752f296dd22e146c9e6f152a269d84b241cc95bb3ff8ec341628a54caf0",
"72c21075f4b2349ce01a3e604e02a9ab9f07e35dd07eff746de348b4f3c6365e",
"8d02fe35ec3ff734de79a0da26fe38223232d2fa909e7a9438451d633f8395a1",
)
}
@@ -724,7 +724,7 @@ func TestConversationKey010(t *testing.T) {
t,
"1b7de0d64d9b12ddbb52ef217a3a7c47c4362ce7ea837d760dad58ab313cba64",
"24383541dd8083b93d144b431679d70ef4eec10c98fceef1eff08b1d81d4b065",
"dd152a76b44e63d1afd4dfff0785fa07b3e494a9e8401aba31ff925caeb8f5b1",
"e3efc88ea3b67f27602c5a0033bf57e1174eaed468d685ab6835629319a1f9f9",
)
}
@@ -733,7 +733,7 @@ func TestConversationKey011(t *testing.T) {
t,
"df2f560e213ca5fb33b9ecde771c7c0cbd30f1cf43c2c24de54480069d9ab0af",
"eeea26e552fc8b5e377acaa03e47daa2d7b0c787fac1e0774c9504d9094c430e",
"770519e803b80f411c34aef59c3ca018608842ebf53909c48d35250bd9323af6",
"77efc793bdaf6b7ea889353b68707530e615fa106d454001fd9013880576ab3f",
)
}
@@ -742,7 +742,7 @@ func TestConversationKey012(t *testing.T) {
t,
"cffff919fcc07b8003fdc63bc8a00c0f5dc81022c1c927c62c597352190d95b9",
"eb5c3cca1a968e26684e5b0eb733aecfc844f95a09ac4e126a9e58a4e4902f92",
"46a14ee7e80e439ec75c66f04ad824b53a632b8409a29bbb7c192e43c00bb795",
"248d4c8b660266a25b3e595fb51afc3f22e83db85b9ebcb8f56c4587a272701f",
)
}
@@ -751,7 +751,7 @@ func TestConversationKey013(t *testing.T) {
t,
"64ba5a685e443e881e9094647ddd32db14444bb21aa7986beeba3d1c4673ba0a",
"50e6a4339fac1f3bf86f2401dd797af43ad45bbf58e0801a7877a3984c77c3c4",
"968b9dbbfcede1664a4ca35a5d3379c064736e87aafbf0b5d114dff710b8a946",
"4fdb2226074f4cfa308fcd1a2fdf3c40e61d97b15d52d4306ae65c86cd21f25d",
)
}
@@ -760,7 +760,7 @@ func TestConversationKey014(t *testing.T) {
t,
"dd0c31ccce4ec8083f9b75dbf23cc2878e6d1b6baa17713841a2428f69dee91a",
"b483e84c1339812bed25be55cff959778dfc6edde97ccd9e3649f442472c091b",
"09024503c7bde07eb7865505891c1ea672bf2d9e25e18dd7a7cea6c69bf44b5d",
"9f865913b556656341ac1222d949d2471973f0c52af50034255489582a4421c1",
)
}
@@ -769,7 +769,7 @@ func TestConversationKey015(t *testing.T) {
t,
"af71313b0d95c41e968a172b33ba5ebd19d06cdf8a7a98df80ecf7af4f6f0358",
"2a5c25266695b461ee2af927a6c44a3c598b8095b0557e9bd7f787067435bc7c",
"fe5155b27c1c4b4e92a933edae23726a04802a7cc354a77ac273c85aa3c97a92",
"0a4be1d6c43298e93a7ca27b9f3e20b8a2a2ea9be31c8a542cf525cf85e10372",
)
}
@@ -778,7 +778,7 @@ func TestConversationKey016(t *testing.T) {
t,
"6636e8a389f75fe068a03b3edb3ea4a785e2768e3f73f48ffb1fc5e7cb7289dc",
"514eb2064224b6a5829ea21b6e8f7d3ea15ff8e70e8555010f649eb6e09aec70",
"ff7afacd4d1a6856d37ca5b546890e46e922b508639214991cf8048ddbe9745c",
"49d2c0088e89856b56566d5a4b492ac9e7c219c1019018bca65cb465c24d3631",
)
}
@@ -787,7 +787,7 @@ func TestConversationKey017(t *testing.T) {
t,
"94b212f02a3cfb8ad147d52941d3f1dbe1753804458e6645af92c7b2ea791caa",
"f0cac333231367a04b652a77ab4f8d658b94e86b5a8a0c472c5c7b0d4c6a40cc",
"e292eaf873addfed0a457c6bd16c8effde33d6664265697f69f420ab16f6669b",
"98cd935572ff535b68990f558638ba3399c19acaea4a783a167a349bad9c4872",
)
}
@@ -796,7 +796,7 @@ func TestConversationKey018(t *testing.T) {
t,
"aa61f9734e69ae88e5d4ced5aae881c96f0d7f16cca603d3bed9eec391136da6",
"4303e5360a884c360221de8606b72dd316da49a37fe51e17ada4f35f671620a6",
"8e7d44fd4767456df1fb61f134092a52fcd6836ebab3b00766e16732683ed848",
"49d2c0088e89856b56566d5a4b492ac9e7c219c1019018bca65cb465c24d3631",
)
}
@@ -805,7 +805,7 @@ func TestConversationKey019(t *testing.T) {
t,
"5e914bdac54f3f8e2cba94ee898b33240019297b69e96e70c8a495943a72fc98",
"5bd097924f606695c59f18ff8fd53c174adbafaaa71b3c0b4144a3e0a474b198",
"f5a0aecf2984bf923c8cd5e7bb8be262d1a8353cb93959434b943a07cf5644bc",
"d9aee5a1c3491352e9cba0b8d3887c9aeb6f4a6caae19811d507bb3ef47210b2d",
)
}
@@ -814,7 +814,7 @@ func TestConversationKey020(t *testing.T) {
t,
"8b275067add6312ddee064bcdbeb9d17e88aa1df36f430b2cea5cc0413d8278a",
"65bbbfca819c90c7579f7a82b750a18c858db1afbec8f35b3c1e0e7b5588e9b8",
"2c565e7027eb46038c2263563d7af681697107e975e9914b799d425effd248d6",
"469f0da3a3b53edbb0af1db5d3d595f39e42edb3d9c916618a50927d272bff71",
)
}
@@ -886,7 +886,7 @@ func TestConversationKey028(t *testing.T) {
t,
"261a076a9702af1647fb343c55b3f9a4f1096273002287df0015ba81ce5294df",
"b2777c863878893ae100fb740c8fab4bebd2bf7be78c761a75593670380a6112",
"76f8d2853de0734e51189ced523c09427c3e46338b9522cd6f74ef5e5b475c74",
"1f70de97fd7f605973b35b5ca64b2939ce5a039e70cab88c2a088bdeccc81bf8",
)
}
@@ -913,7 +913,7 @@ func TestConversationKey031(t *testing.T) {
t,
"63bffa986e382b0ac8ccc1aa93d18a7aa445116478be6f2453bad1f2d3af2344",
"b895c70a83e782c1cf84af558d1038e6b211c6f84ede60408f519a293201031d",
"3a3b8f00d4987fc6711d9be64d9c59cf9a709c6c6481c2cde404bcc7a28f174e",
"3445872a13f45a46ecd362c0e347cd32b3532b1b4cd35ec567ad4d4afe7a1665",
)
}
@@ -922,7 +922,7 @@ func TestConversationKey032(t *testing.T) {
t,
"e4a8bcacbf445fd3721792b939ff58e691cdcba6a8ba67ac3467b45567a03e5c",
"b54053189e8c9252c6950059c783edb10675d06d20c7b342f73ec9fa6ed39c9d",
"7b3933b4ef8189d347169c7955589fc1cfc01da5239591a08a183ff6694c44ad",
"d9aee5a1c3491352e9cba0b8d3887c9aeb6f4a6caae19811d507bb3ef47210b2d",
)
}
@@ -952,7 +952,7 @@ func TestConversationKey035(t *testing.T) {
t,
"0000000000000000000000000000000000000000000000000000000000000001",
"79be667ef9dcbbac55a06295ce870b07029bfcdb2dce28d959f2815b16f81798",
"3b4610cb7189beb9cc29eb3716ecc6102f1247e8f3101a03a1787d8908aeb54e",
"7b88c5403f9b6598e1dcad39aa052aadfd50f357c7dc498b93d928e518685737",
)
}
@@ -1378,4 +1378,4 @@ func assertCryptPub(
return
}
assert.Equal(t, decrypted, plaintextBytes, "wrong decryption")
}
}

202
pkg/crypto/sha256/LICENSE
View File

@@ -1,202 +0,0 @@
Apache License
Version 2.0, January 2004
http://www.apache.org/licenses/
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
1. Definitions.
"License" shall mean the terms and conditions for use, reproduction,
and distribution as defined by Sections 1 through 9 of this document.
"Licensor" shall mean the copyright owner or entity authorized by
the copyright owner that is granting the License.
"Legal Entity" shall mean the union of the acting entity and all
other entities that control, are controlled by, or are under common
control with that entity. For the purposes of this definition,
"control" means (i) the power, direct or indirect, to cause the
direction or management of such entity, whether by contract or
otherwise, or (ii) ownership of fifty percent (50%) or more of the
outstanding shares, or (iii) beneficial ownership of such entity.
"You" (or "Your") shall mean an individual or Legal Entity
exercising permissions granted by this License.
"Source" form shall mean the preferred form for making modifications,
including but not limited to software source code, documentation
source, and configuration files.
"Object" form shall mean any form resulting from mechanical
transformation or translation of a Source form, including but
not limited to compiled object code, generated documentation,
and conversions to other media types.
"Work" shall mean the work of authorship, whether in Source or
Object form, made available under the License, as indicated by a
copyright notice that is included in or attached to the work
(an example is provided in the Appendix below).
"Derivative Works" shall mean any work, whether in Source or Object
form, that is based on (or derived from) the Work and for which the
editorial revisions, annotations, elaborations, or other modifications
represent, as a whole, an original work of authorship. For the purposes
of this License, Derivative Works shall not include works that remain
separable from, or merely link (or bind by name) to the interfaces of,
the Work and Derivative Works thereof.
"Contribution" shall mean any work of authorship, including
the original version of the Work and any modifications or additions
to that Work or Derivative Works thereof, that is intentionally
submitted to Licensor for inclusion in the Work by the copyright owner
or by an individual or Legal Entity authorized to submit on behalf of
the copyright owner. For the purposes of this definition, "submitted"
means any form of electronic, verbal, or written communication sent
to the Licensor or its representatives, including but not limited to
communication on electronic mailing lists, source code control systems,
and issue tracking systems that are managed by, or on behalf of, the
Licensor for the purpose of discussing and improving the Work, but
excluding communication that is conspicuously marked or otherwise
designated in writing by the copyright owner as "Not a Contribution."
"Contributor" shall mean Licensor and any individual or Legal Entity
on behalf of whom a Contribution has been received by Licensor and
subsequently incorporated within the Work.
2. Grant of Copyright License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
copyright license to reproduce, prepare Derivative Works of,
publicly display, publicly perform, sublicense, and distribute the
Work and such Derivative Works in Source or Object form.
3. Grant of Patent License. Subject to the terms and conditions of
this License, each Contributor hereby grants to You a perpetual,
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
(except as stated in this section) patent license to make, have made,
use, offer to sell, sell, import, and otherwise transfer the Work,
where such license applies only to those patent claims licensable
by such Contributor that are necessarily infringed by their
Contribution(s) alone or by combination of their Contribution(s)
with the Work to which such Contribution(s) was submitted. If You
institute patent litigation against any entity (including a
cross-claim or counterclaim in a lawsuit) alleging that the Work
or a Contribution incorporated within the Work constitutes direct
or contributory patent infringement, then any patent licenses
granted to You under this License for that Work shall terminate
as of the date such litigation is filed.
4. Redistribution. You may reproduce and distribute copies of the
Work or Derivative Works thereof in any medium, with or without
modifications, and in Source or Object form, provided that You
meet the following conditions:
(a) You must give any other recipients of the Work or
Derivative Works a copy of this License; and
(b) You must cause any modified files to carry prominent notices
stating that You changed the files; and
(c) You must retain, in the Source form of any Derivative Works
that You distribute, all copyright, patent, trademark, and
attribution notices from the Source form of the Work,
excluding those notices that do not pertain to any part of
the Derivative Works; and
(d) If the Work includes a "NOTICE" text file as part of its
distribution, then any Derivative Works that You distribute must
include a readable copy of the attribution notices contained
within such NOTICE file, excluding those notices that do not
pertain to any part of the Derivative Works, in at least one
of the following places: within a NOTICE text file distributed
as part of the Derivative Works; within the Source form or
documentation, if provided along with the Derivative Works; or,
within a display generated by the Derivative Works, if and
wherever such third-party notices normally appear. The contents
of the NOTICE file are for informational purposes only and
do not modify the License. You may add Your own attribution
notices within Derivative Works that You distribute, alongside
or as an addendum to the NOTICE text from the Work, provided
that such additional attribution notices cannot be construed
as modifying the License.
You may add Your own copyright statement to Your modifications and
may provide additional or different license terms and conditions
for use, reproduction, or distribution of Your modifications, or
for any such Derivative Works as a whole, provided Your use,
reproduction, and distribution of the Work otherwise complies with
the conditions stated in this License.
5. Submission of Contributions. Unless You explicitly state otherwise,
any Contribution intentionally submitted for inclusion in the Work
by You to the Licensor shall be under the terms and conditions of
this License, without any additional terms or conditions.
Notwithstanding the above, nothing herein shall supersede or modify
the terms of any separate license agreement you may have executed
with Licensor regarding such Contributions.
6. Trademarks. This License does not grant permission to use the trade
names, trademarks, service marks, or product names of the Licensor,
except as required for reasonable and customary use in describing the
origin of the Work and reproducing the content of the NOTICE file.
7. Disclaimer of Warranty. Unless required by applicable law or
agreed to in writing, Licensor provides the Work (and each
Contributor provides its Contributions) on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
implied, including, without limitation, any warranties or conditions
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
PARTICULAR PURPOSE. You are solely responsible for determining the
appropriateness of using or redistributing the Work and assume any
risks associated with Your exercise of permissions under this License.
8. Limitation of Liability. In no event and under no legal theory,
whether in tort (including negligence), contract, or otherwise,
unless required by applicable law (such as deliberate and grossly
negligent acts) or agreed to in writing, shall any Contributor be
liable to You for damages, including any direct, indirect, special,
incidental, or consequential damages of any character arising as a
result of this License or out of the use or inability to use the
Work (including but not limited to damages for loss of goodwill,
work stoppage, computer failure or malfunction, or any and all
other commercial damages or losses), even if such Contributor
has been advised of the possibility of such damages.
9. Accepting Warranty or Additional Liability. While redistributing
the Work or Derivative Works thereof, You may choose to offer,
and charge a fee for, acceptance of support, warranty, indemnity,
or other liability obligations and/or rights consistent with this
License. However, in accepting such obligations, You may act only
on Your own behalf and on Your sole responsibility, not on behalf
of any other Contributor, and only if You agree to indemnify,
defend, and hold each Contributor harmless for any liability
incurred by, or claims asserted against, such Contributor by reason
of your accepting any such warranty or additional liability.
END OF TERMS AND CONDITIONS
APPENDIX: How to apply the Apache License to your work.
To apply the Apache License to your work, attach the following
boilerplate notice, with the fields enclosed by brackets "[]"
replaced with your own identifying information. (Don't include
the brackets!) The text should be enclosed in the appropriate
comment syntax for the file format. We also recommend that a
file or class name and description of purpose be included on the
same "printed page" as the copyright notice for easier
identification within third-party archives.
Copyright [yyyy] [name of copyright owner]
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.

197
pkg/crypto/sha256/README.md
View File

@@ -1,197 +0,0 @@
# sha256-simd
Accelerate SHA256 computations in pure Go using AVX512, SHA Extensions for x86
and ARM64 for ARM.
On AVX512 it provides an up to 8x improvement (over 3 GB/s per core).
SHA Extensions give a performance boost of close to 4x over native.
## Introduction
This package is designed as a replacement for `crypto/sha256`.
For ARM CPUs with the Cryptography Extensions, advantage is taken of the SHA2
instructions resulting in a massive performance improvement.
This package uses Golang assembly.
The AVX512 version is based on the Intel's "multi-buffer crypto library for
IPSec" whereas the other Intel implementations are described in "Fast SHA-256
Implementations on Intel Architecture Processors" by J. Guilford et al.
## Support for Intel SHA Extensions
Support for the Intel SHA Extensions has been added by Kristofer Peterson (
@svenski123), originally developed for
spacemeshos [here](https://github.com/spacemeshos/POET/issues/23). On CPUs that
support it (known thus far Intel Celeron J3455 and AMD Ryzen) it gives a
significant boost in performance (with thanks to @AudriusButkevicius for
reporting the results; full
results [here](https://github.com/minio/sha256-simd/pull/37#issuecomment-451607827)).
```
$ benchcmp avx2.txt sha-ext.txt
benchmark AVX2 MB/s SHA Ext MB/s speedup
BenchmarkHash5M 514.40 1975.17 3.84x
```
Thanks to Kristofer Peterson, we also added additional performance changes such
as optimized padding,
endian conversions which sped up all implementations i.e. Intel SHA alone while
doubled performance for small sizes,
the other changes increased everything roughly 50%.
## Support for AVX512
We have added support for AVX512 which results in an up to 8x performance
improvement over AVX2 (3.0 GHz Xeon Platinum 8124M CPU):
```
$ benchcmp avx2.txt avx512.txt
benchmark AVX2 MB/s AVX512 MB/s speedup
BenchmarkHash5M 448.62 3498.20 7.80x
```
The original code was developed by Intel as part of
the [multi-buffer crypto library](https://github.com/intel/intel-ipsec-mb) for
IPSec or more specifically
this [AVX512](https://github.com/intel/intel-ipsec-mb/blob/master/avx512/sha256_x16_avx512.asm)
implementation. The key idea behind it is to process a total of 16 checksums in
parallel by “transposing” 16 (independent) messages of 64 bytes between a total
of 16 ZMM registers (each 64 bytes wide).
Transposing the input messages means that in order to take full advantage of the
speedup you need to have a (server) workload where multiple threads are doing
SHA256 calculations in parallel. Unfortunately for this algorithm it is not
possible for two message blocks processed in parallel to be dependent on one
another — because then the (interim) result of the first part of the message has
to be an input into the processing of the second part of the message.
Whereas the original Intel C implementation requires some sort of explicit
scheduling of messages to be processed in parallel, for Golang it makes sense to
take advantage of channels in order to group messages together and use channels
as well for sending back the results (thereby effectively decoupling the
calculations). We have implemented a fairly simple scheduling mechanism that
seems to work well in practice.
Due to this different way of scheduling, we decided to use an explicit method to
instantiate the AVX512 version. Essentially one or more AVX512 processing
servers ([
`Avx512Server`](https://github.com/minio/sha256-simd/blob/master/sha256blockAvx512_amd64.go#L294))
have to be created whereby each server can hash over 3 GB/s on a single core. An
`hash.Hash` object ([
`Avx512Digest`](https://github.com/minio/sha256-simd/blob/master/sha256blockAvx512_amd64.go#L45))
is then instantiated using one of these servers and used in the regular fashion:
```go
import "mleku.dev/pkg/sha256"
func main() {
server := sha256.NewAvx512Server()
h512 := sha256.NewAvx512(server)
h512.Write(fileBlock)
digest := h512.Sum([]byte{})
}
```
Note that, because of the scheduling overhead, for small messages (< 1 MB) you
will be better off using the regular SHA256 hashing (but those are typically not
performance critical anyway). Some other tips to get the best performance:
- Have many go routines doing SHA256 calculations in parallel.
- Try to Write() messages in multiples of 64 bytes.
- Try to keep the overall length of messages to a roughly similar size ie. 5
MB (this way all 16 ‘lanes’ in the AVX512 computations are contributing as
much as possible).
More detailed information can be found in
this [blog](https://blog.minio.io/accelerate-sha256-up-to-8x-over-3-gb-s-per-core-with-avx512-a0b1d64f78f)
post including scaling across cores.
## Drop-In Replacement
The following code snippet shows how you can use `github.com/minio/sha256-simd`.
This will automatically select the fastest method for the architecture on which
it will be executed.
```go
import "crypto.orly/sha256"
func main() {
...
shaWriter := sha256.New()
io.Copy(shaWriter, file)
...
}
```
## Performance
Below is the speed in MB/s for a single core (ranked fast to slow) for blocks
larger than 1 MB.
| Processor | SIMD | Speed (MB/s) |
| --------------------------------- | ------- | -----------: |
| 3.0 GHz Intel Xeon Platinum 8124M | AVX512 | 3498 |
| 3.7 GHz AMD Ryzen 7 2700X | SHA Ext | 1979 |
| 1.2 GHz ARM Cortex-A53 | ARM64 | 638 |
## asm2plan9s
In order to be able to work more easily with AVX512/AVX2 instructions, a
separate tool was developed to convert SIMD instructions into the corresponding
BYTE sequence as accepted by Go assembly.
See [asm2plan9s](https://github.com/minio/asm2plan9s) for more information.
## Why and benefits
One of the most performance sensitive parts of
the [Minio](https://github.com/minio/minio) object storage server is related to
SHA256 hash sums calculations. For instance during multi part uploads each part
that is uploaded needs to be verified for data integrity by the server.
Other applications that can benefit from enhanced SHA256 performance are
deduplication in storage systems, intrusion detection, version control systems,
integrity checking, etc.
## ARM SHA Extensions
The 64-bit ARMv8 core has introduced new instructions for SHA1 and SHA2
acceleration as part of
the [Cryptography Extensions](http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.ddi0501f/CHDFJBCJ.html).
Below you can see a small excerpt highlighting one of the rounds as is done for
the SHA256 calculation process (for full code
see [sha256block_arm64.s](https://github.com/minio/sha256-simd/blob/master/sha256block_arm64.s)).
```
sha256h q2, q3, v9.4s
sha256h2 q3, q4, v9.4s
sha256su0 v5.4s, v6.4s
rev32 v8.16b, v8.16b
add v9.4s, v7.4s, v18.4s
mov v4.16b, v2.16b
sha256h q2, q3, v10.4s
sha256h2 q3, q4, v10.4s
sha256su0 v6.4s, v7.4s
sha256su1 v5.4s, v7.4s, v8.4s
```
### Detailed benchmarks
Benchmarks generated on a 1.2 Ghz Quad-Core ARM Cortex A53
equipped [Pine64](https://www.pine64.com/).
```
minio@minio-arm:$ benchcmp golang.txt arm64.txt
benchmark golang arm64 speedup
BenchmarkHash8Bytes-4 0.68 MB/s 5.70 MB/s 8.38x
BenchmarkHash1K-4 5.65 MB/s 326.30 MB/s 57.75x
BenchmarkHash8K-4 6.00 MB/s 570.63 MB/s 95.11x
BenchmarkHash1M-4 6.05 MB/s 638.23 MB/s 105.49x
```
## License
Released under the Apache License v2.0. You can find the complete text in the
file LICENSE.
## Contributing
Contributions are welcome, please send PRs for any enhancements.

55
pkg/crypto/sha256/cpuid_other.go
View File

@@ -1,55 +0,0 @@
// Minio Cloud Storage, (C) 2021 Minio, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
//
package sha256
import (
"bytes"
"github.com/klauspost/cpuid/v2"
"io/ioutil"
"runtime"
)
var (
hasIntelSha = runtime.GOARCH == "amd64" && cpuid.CPU.Supports(
cpuid.SHA, cpuid.SSSE3,
cpuid.SSE4,
)
hasAvx512 = cpuid.CPU.Supports(
cpuid.AVX512F, cpuid.AVX512DQ, cpuid.AVX512BW,
cpuid.AVX512VL,
)
)
func hasArmSha2() bool {
if cpuid.CPU.Has(cpuid.SHA2) {
return true
}
if runtime.GOARCH != "arm64" || runtime.GOOS != "linux" {
return false
}
// Fall back to hacky cpuinfo parsing...
const procCPUInfo = "/proc/cpuinfo"
// Feature to check for.
const sha256Feature = "sha2"
cpuInfo, err := ioutil.ReadFile(procCPUInfo)
if err != nil {
return false
}
return bytes.Contains(cpuInfo, []byte(sha256Feature))
}

6
pkg/crypto/sha256/doc.go
View File

@@ -1,6 +0,0 @@
// Package sha256 is taken from github.com/minio/sha256-simd, implementing,
// where available, an accelerated SIMD implementation of sha256.
//
// This package should be updated against the upstream version from time to
// time.
package sha256

470
pkg/crypto/sha256/sha256.go
View File

@@ -1,470 +0,0 @@
/*
* Minio Cloud Storage, (C) 2016 Minio, Inc.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package sha256
import (
"crypto/sha256"
"encoding/binary"
"errors"
"hash"
)
// Size - The size of a SHA256 checksum in bytes.
const Size = 32
// BlockSize - The blocksize of SHA256 in bytes.
const BlockSize = 64
const (
chunk = BlockSize
init0 = 0x6A09E667
init1 = 0xBB67AE85
init2 = 0x3C6EF372
init3 = 0xA54FF53A
init4 = 0x510E527F
init5 = 0x9B05688C
init6 = 0x1F83D9AB
init7 = 0x5BE0CD19
)
// digest represents the partial evaluation of a checksum.
type digest struct {
h [8]uint32
x [chunk]byte
nx int
len uint64
}
// Reset digest back to default
func (d *digest) Reset() {
d.h[0] = init0
d.h[1] = init1
d.h[2] = init2
d.h[3] = init3
d.h[4] = init4
d.h[5] = init5
d.h[6] = init6
d.h[7] = init7
d.nx = 0
d.len = 0
}
type blockfuncType int
const (
blockfuncStdlib blockfuncType = iota
blockfuncIntelSha
blockfuncArmSha2
blockfuncForceGeneric = -1
)
var blockfunc blockfuncType
func init() {
switch {
case hasIntelSha:
blockfunc = blockfuncIntelSha
case hasArmSha2():
blockfunc = blockfuncArmSha2
}
}
// New returns a new hash.Hash computing the SHA256 checksum.
func New() hash.Hash {
if blockfunc == blockfuncStdlib {
// Fallback to the standard golang implementation
// if no features were found.
return sha256.New()
}
d := new(digest)
d.Reset()
return d
}
// Sum256 - single caller sha256 helper
func Sum256(data []byte) (result [Size]byte) {
var d digest
d.Reset()
d.Write(data)
result = d.checkSum()
return
}
// Return size of checksum
func (d *digest) Size() int { return Size }
// Return blocksize of checksum
func (d *digest) BlockSize() int { return BlockSize }
// Write to digest
func (d *digest) Write(p []byte) (nn int, err error) {
nn = len(p)
d.len += uint64(nn)
if d.nx > 0 {
n := copy(d.x[d.nx:], p)
d.nx += n
if d.nx == chunk {
block(d, d.x[:])
d.nx = 0
}
p = p[n:]
}
if len(p) >= chunk {
n := len(p) &^ (chunk - 1)
block(d, p[:n])
p = p[n:]
}
if len(p) > 0 {
d.nx = copy(d.x[:], p)
}
return
}
// Return sha256 sum in bytes
func (d *digest) Sum(in []byte) []byte {
// Make a copy of d0 so that caller can keep writing and summing.
d0 := *d
hash := d0.checkSum()
return append(in, hash[:]...)
}
// Intermediate checksum function
func (d *digest) checkSum() (digest [Size]byte) {
n := d.nx
var k [64]byte
copy(k[:], d.x[:n])
k[n] = 0x80
if n >= 56 {
block(d, k[:])
// clear block buffer - go compiles this to optimal 1x xorps + 4x movups
// unfortunately expressing this more succinctly results in much worse code
k[0] = 0
k[1] = 0
k[2] = 0
k[3] = 0
k[4] = 0
k[5] = 0
k[6] = 0
k[7] = 0
k[8] = 0
k[9] = 0
k[10] = 0
k[11] = 0
k[12] = 0
k[13] = 0
k[14] = 0
k[15] = 0
k[16] = 0
k[17] = 0
k[18] = 0
k[19] = 0
k[20] = 0
k[21] = 0
k[22] = 0
k[23] = 0
k[24] = 0
k[25] = 0
k[26] = 0
k[27] = 0
k[28] = 0
k[29] = 0
k[30] = 0
k[31] = 0
k[32] = 0
k[33] = 0
k[34] = 0
k[35] = 0
k[36] = 0
k[37] = 0
k[38] = 0
k[39] = 0
k[40] = 0
k[41] = 0
k[42] = 0
k[43] = 0
k[44] = 0
k[45] = 0
k[46] = 0
k[47] = 0
k[48] = 0
k[49] = 0
k[50] = 0
k[51] = 0
k[52] = 0
k[53] = 0
k[54] = 0
k[55] = 0
k[56] = 0
k[57] = 0
k[58] = 0
k[59] = 0
k[60] = 0
k[61] = 0
k[62] = 0
k[63] = 0
}
binary.BigEndian.PutUint64(k[56:64], uint64(d.len)<<3)
block(d, k[:])
{
const i = 0
binary.BigEndian.PutUint32(digest[i*4:i*4+4], d.h[i])
}
{
const i = 1
binary.BigEndian.PutUint32(digest[i*4:i*4+4], d.h[i])
}
{
const i = 2
binary.BigEndian.PutUint32(digest[i*4:i*4+4], d.h[i])
}
{
const i = 3
binary.BigEndian.PutUint32(digest[i*4:i*4+4], d.h[i])
}
{
const i = 4
binary.BigEndian.PutUint32(digest[i*4:i*4+4], d.h[i])
}
{
const i = 5
binary.BigEndian.PutUint32(digest[i*4:i*4+4], d.h[i])
}
{
const i = 6
binary.BigEndian.PutUint32(digest[i*4:i*4+4], d.h[i])
}
{
const i = 7
binary.BigEndian.PutUint32(digest[i*4:i*4+4], d.h[i])
}
return
}
func block(dig *digest, p []byte) {
if blockfunc == blockfuncIntelSha {
blockIntelShaGo(dig, p)
} else if blockfunc == blockfuncArmSha2 {
blockArmSha2Go(dig, p)
} else {
blockGeneric(dig, p)
}
}
func blockGeneric(dig *digest, p []byte) {
var w [64]uint32
h0, h1, h2, h3, h4, h5, h6, h7 := dig.h[0], dig.h[1], dig.h[2], dig.h[3], dig.h[4], dig.h[5], dig.h[6], dig.h[7]
for len(p) >= chunk {
// Can interlace the computation of w with the
// rounds below if needed for speed.
for i := 0; i < 16; i++ {
j := i * 4
w[i] = uint32(p[j])<<24 | uint32(p[j+1])<<16 | uint32(p[j+2])<<8 | uint32(p[j+3])
}
for i := 16; i < 64; i++ {
v1 := w[i-2]
t1 := (v1>>17 | v1<<(32-17)) ^ (v1>>19 | v1<<(32-19)) ^ (v1 >> 10)
v2 := w[i-15]
t2 := (v2>>7 | v2<<(32-7)) ^ (v2>>18 | v2<<(32-18)) ^ (v2 >> 3)
w[i] = t1 + w[i-7] + t2 + w[i-16]
}
a, b, c, d, e, f, g, h := h0, h1, h2, h3, h4, h5, h6, h7
for i := 0; i < 64; i++ {
t1 := h + ((e>>6 | e<<(32-6)) ^ (e>>11 | e<<(32-11)) ^ (e>>25 | e<<(32-25))) + ((e & f) ^ (^e & g)) + _K[i] + w[i]
t2 := ((a>>2 | a<<(32-2)) ^ (a>>13 | a<<(32-13)) ^ (a>>22 | a<<(32-22))) + ((a & b) ^ (a & c) ^ (b & c))
h = g
g = f
f = e
e = d + t1
d = c
c = b
b = a
a = t1 + t2
}
h0 += a
h1 += b
h2 += c
h3 += d
h4 += e
h5 += f
h6 += g
h7 += h
p = p[chunk:]
}
dig.h[0], dig.h[1], dig.h[2], dig.h[3], dig.h[4], dig.h[5], dig.h[6], dig.h[7] = h0, h1, h2, h3, h4, h5, h6, h7
}
var _K = []uint32{
0x428a2f98,
0x71374491,
0xb5c0fbcf,
0xe9b5dba5,
0x3956c25b,
0x59f111f1,
0x923f82a4,
0xab1c5ed5,
0xd807aa98,
0x12835b01,
0x243185be,
0x550c7dc3,
0x72be5d74,
0x80deb1fe,
0x9bdc06a7,
0xc19bf174,
0xe49b69c1,
0xefbe4786,
0x0fc19dc6,
0x240ca1cc,
0x2de92c6f,
0x4a7484aa,
0x5cb0a9dc,
0x76f988da,
0x983e5152,
0xa831c66d,
0xb00327c8,
0xbf597fc7,
0xc6e00bf3,
0xd5a79147,
0x06ca6351,
0x14292967,
0x27b70a85,
0x2e1b2138,
0x4d2c6dfc,
0x53380d13,
0x650a7354,
0x766a0abb,
0x81c2c92e,
0x92722c85,
0xa2bfe8a1,
0xa81a664b,
0xc24b8b70,
0xc76c51a3,
0xd192e819,
0xd6990624,
0xf40e3585,
0x106aa070,
0x19a4c116,
0x1e376c08,
0x2748774c,
0x34b0bcb5,
0x391c0cb3,
0x4ed8aa4a,
0x5b9cca4f,
0x682e6ff3,
0x748f82ee,
0x78a5636f,
0x84c87814,
0x8cc70208,
0x90befffa,
0xa4506ceb,
0xbef9a3f7,
0xc67178f2,
}
const (
magic256 = "sha\x03"
marshaledSize = len(magic256) + 8*4 + chunk + 8
)
func (d *digest) MarshalBinary() ([]byte, error) {
b := make([]byte, 0, marshaledSize)
b = append(b, magic256...)
b = appendUint32(b, d.h[0])
b = appendUint32(b, d.h[1])
b = appendUint32(b, d.h[2])
b = appendUint32(b, d.h[3])
b = appendUint32(b, d.h[4])
b = appendUint32(b, d.h[5])
b = appendUint32(b, d.h[6])
b = appendUint32(b, d.h[7])
b = append(b, d.x[:d.nx]...)
b = b[:len(b)+len(d.x)-d.nx] // already zero
b = appendUint64(b, d.len)
return b, nil
}
func (d *digest) UnmarshalBinary(b []byte) error {
if len(b) < len(magic256) || string(b[:len(magic256)]) != magic256 {
return errors.New("next.orly.dev/pkg/crypto/sha256: invalid hash state identifier")
}
if len(b) != marshaledSize {
return errors.New("next.orly.dev/pkg/crypto/sha256: invalid hash state size")
}
b = b[len(magic256):]
b, d.h[0] = consumeUint32(b)
b, d.h[1] = consumeUint32(b)
b, d.h[2] = consumeUint32(b)
b, d.h[3] = consumeUint32(b)
b, d.h[4] = consumeUint32(b)
b, d.h[5] = consumeUint32(b)
b, d.h[6] = consumeUint32(b)
b, d.h[7] = consumeUint32(b)
b = b[copy(d.x[:], b):]
b, d.len = consumeUint64(b)
d.nx = int(d.len % chunk)
return nil
}
func appendUint32(b []byte, v uint32) []byte {
return append(
b,
byte(v>>24),
byte(v>>16),
byte(v>>8),
byte(v),
)
}
func appendUint64(b []byte, v uint64) []byte {
return append(
b,
byte(v>>56),
byte(v>>48),
byte(v>>40),
byte(v>>32),
byte(v>>24),
byte(v>>16),
byte(v>>8),
byte(v),
)
}
func consumeUint64(b []byte) ([]byte, uint64) {
_ = b[7]
x := uint64(b[7]) | uint64(b[6])<<8 | uint64(b[5])<<16 | uint64(b[4])<<24 |
uint64(b[3])<<32 | uint64(b[2])<<40 | uint64(b[1])<<48 | uint64(b[0])<<56
return b[8:], x
}
func consumeUint32(b []byte) ([]byte, uint32) {
_ = b[3]
x := uint32(b[3]) | uint32(b[2])<<8 | uint32(b[1])<<16 | uint32(b[0])<<24
return b[4:], x
}

2887
pkg/crypto/sha256/sha256_test.go
View File

File diff suppressed because it is too large Load Diff

686
pkg/crypto/sha256/sha256blockAvx512_amd64.asm
View File

@@ -1,686 +0,0 @@
// 16x Parallel implementation of SHA256 for AVX512
//
// Minio Cloud Storage, (C) 2017 Minio, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
//
// This code is based on the Intel Multi-Buffer Crypto for IPSec library
// and more specifically the following implementation:
// https://github.com/intel/intel-ipsec-mb/blob/master/avx512/sha256_x16_avx512.asm
//
// For Golang it has been converted into Plan 9 assembly with the help of
// github.com/minio/asm2plan9s to assemble the AVX512 instructions
//
// Copyright (c) 2017, Intel Corporation
//
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions are met:
//
// * Redistributions of source code must retain the above copyright notice,
// this list of conditions and the following disclaimer.
// * Redistributions in binary form must reproduce the above copyright
// notice, this list of conditions and the following disclaimer in the
// documentation and/or other materials provided with the distribution.
// * Neither the name of Intel Corporation nor the names of its contributors
// may be used to endorse or promote products derived from this software
// without specific prior written permission.
//
// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
// AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
// IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
// DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE
// FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
// DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
// SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
// CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
// OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
#define SHA256_DIGEST_ROW_SIZE 64
// arg1
#define STATE rdi
#define STATE_P9 DI
// arg2
#define INP_SIZE rsi
#define INP_SIZE_P9 SI
#define IDX rcx
#define TBL rdx
#define TBL_P9 DX
#define INPUT rax
#define INPUT_P9 AX
#define inp0 r9
#define SCRATCH_P9 R12
#define SCRATCH r12
#define maskp r13
#define MASKP_P9 R13
#define mask r14
#define MASK_P9 R14
#define A zmm0
#define B zmm1
#define C zmm2
#define D zmm3
#define E zmm4
#define F zmm5
#define G zmm6
#define H zmm7
#define T1 zmm8
#define TMP0 zmm9
#define TMP1 zmm10
#define TMP2 zmm11
#define TMP3 zmm12
#define TMP4 zmm13
#define TMP5 zmm14
#define TMP6 zmm15
#define W0 zmm16
#define W1 zmm17
#define W2 zmm18
#define W3 zmm19
#define W4 zmm20
#define W5 zmm21
#define W6 zmm22
#define W7 zmm23
#define W8 zmm24
#define W9 zmm25
#define W10 zmm26
#define W11 zmm27
#define W12 zmm28
#define W13 zmm29
#define W14 zmm30
#define W15 zmm31
#define TRANSPOSE16(_r0, _r1, _r2, _r3, _r4, _r5, _r6, _r7, _r8, _r9, _r10, _r11, _r12, _r13, _r14, _r15, _t0, _t1) \
\
\ // input r0 = {a15 a14 a13 a12 a11 a10 a9 a8 a7 a6 a5 a4 a3 a2 a1 a0}
\ // r1 = {b15 b14 b13 b12 b11 b10 b9 b8 b7 b6 b5 b4 b3 b2 b1 b0}
\ // r2 = {c15 c14 c13 c12 c11 c10 c9 c8 c7 c6 c5 c4 c3 c2 c1 c0}
\ // r3 = {d15 d14 d13 d12 d11 d10 d9 d8 d7 d6 d5 d4 d3 d2 d1 d0}
\ // r4 = {e15 e14 e13 e12 e11 e10 e9 e8 e7 e6 e5 e4 e3 e2 e1 e0}
\ // r5 = {f15 f14 f13 f12 f11 f10 f9 f8 f7 f6 f5 f4 f3 f2 f1 f0}
\ // r6 = {g15 g14 g13 g12 g11 g10 g9 g8 g7 g6 g5 g4 g3 g2 g1 g0}
\ // r7 = {h15 h14 h13 h12 h11 h10 h9 h8 h7 h6 h5 h4 h3 h2 h1 h0}
\ // r8 = {i15 i14 i13 i12 i11 i10 i9 i8 i7 i6 i5 i4 i3 i2 i1 i0}
\ // r9 = {j15 j14 j13 j12 j11 j10 j9 j8 j7 j6 j5 j4 j3 j2 j1 j0}
\ // r10 = {k15 k14 k13 k12 k11 k10 k9 k8 k7 k6 k5 k4 k3 k2 k1 k0}
\ // r11 = {l15 l14 l13 l12 l11 l10 l9 l8 l7 l6 l5 l4 l3 l2 l1 l0}
\ // r12 = {m15 m14 m13 m12 m11 m10 m9 m8 m7 m6 m5 m4 m3 m2 m1 m0}
\ // r13 = {n15 n14 n13 n12 n11 n10 n9 n8 n7 n6 n5 n4 n3 n2 n1 n0}
\ // r14 = {o15 o14 o13 o12 o11 o10 o9 o8 o7 o6 o5 o4 o3 o2 o1 o0}
\ // r15 = {p15 p14 p13 p12 p11 p10 p9 p8 p7 p6 p5 p4 p3 p2 p1 p0}
\
\ // output r0 = { p0 o0 n0 m0 l0 k0 j0 i0 h0 g0 f0 e0 d0 c0 b0 a0}
\ // r1 = { p1 o1 n1 m1 l1 k1 j1 i1 h1 g1 f1 e1 d1 c1 b1 a1}
\ // r2 = { p2 o2 n2 m2 l2 k2 j2 i2 h2 g2 f2 e2 d2 c2 b2 a2}
\ // r3 = { p3 o3 n3 m3 l3 k3 j3 i3 h3 g3 f3 e3 d3 c3 b3 a3}
\ // r4 = { p4 o4 n4 m4 l4 k4 j4 i4 h4 g4 f4 e4 d4 c4 b4 a4}
\ // r5 = { p5 o5 n5 m5 l5 k5 j5 i5 h5 g5 f5 e5 d5 c5 b5 a5}
\ // r6 = { p6 o6 n6 m6 l6 k6 j6 i6 h6 g6 f6 e6 d6 c6 b6 a6}
\ // r7 = { p7 o7 n7 m7 l7 k7 j7 i7 h7 g7 f7 e7 d7 c7 b7 a7}
\ // r8 = { p8 o8 n8 m8 l8 k8 j8 i8 h8 g8 f8 e8 d8 c8 b8 a8}
\ // r9 = { p9 o9 n9 m9 l9 k9 j9 i9 h9 g9 f9 e9 d9 c9 b9 a9}
\ // r10 = {p10 o10 n10 m10 l10 k10 j10 i10 h10 g10 f10 e10 d10 c10 b10 a10}
\ // r11 = {p11 o11 n11 m11 l11 k11 j11 i11 h11 g11 f11 e11 d11 c11 b11 a11}
\ // r12 = {p12 o12 n12 m12 l12 k12 j12 i12 h12 g12 f12 e12 d12 c12 b12 a12}
\ // r13 = {p13 o13 n13 m13 l13 k13 j13 i13 h13 g13 f13 e13 d13 c13 b13 a13}
\ // r14 = {p14 o14 n14 m14 l14 k14 j14 i14 h14 g14 f14 e14 d14 c14 b14 a14}
\ // r15 = {p15 o15 n15 m15 l15 k15 j15 i15 h15 g15 f15 e15 d15 c15 b15 a15}
\
\ // process top half
vshufps _t0, _r0, _r1, 0x44 \ // t0 = {b13 b12 a13 a12 b9 b8 a9 a8 b5 b4 a5 a4 b1 b0 a1 a0}
vshufps _r0, _r0, _r1, 0xEE \ // r0 = {b15 b14 a15 a14 b11 b10 a11 a10 b7 b6 a7 a6 b3 b2 a3 a2}
vshufps _t1, _r2, _r3, 0x44 \ // t1 = {d13 d12 c13 c12 d9 d8 c9 c8 d5 d4 c5 c4 d1 d0 c1 c0}
vshufps _r2, _r2, _r3, 0xEE \ // r2 = {d15 d14 c15 c14 d11 d10 c11 c10 d7 d6 c7 c6 d3 d2 c3 c2}
\
vshufps _r3, _t0, _t1, 0xDD \ // r3 = {d13 c13 b13 a13 d9 c9 b9 a9 d5 c5 b5 a5 d1 c1 b1 a1}
vshufps _r1, _r0, _r2, 0x88 \ // r1 = {d14 c14 b14 a14 d10 c10 b10 a10 d6 c6 b6 a6 d2 c2 b2 a2}
vshufps _r0, _r0, _r2, 0xDD \ // r0 = {d15 c15 b15 a15 d11 c11 b11 a11 d7 c7 b7 a7 d3 c3 b3 a3}
vshufps _t0, _t0, _t1, 0x88 \ // t0 = {d12 c12 b12 a12 d8 c8 b8 a8 d4 c4 b4 a4 d0 c0 b0 a0}
\
\ // use r2 in place of t0
vshufps _r2, _r4, _r5, 0x44 \ // r2 = {f13 f12 e13 e12 f9 f8 e9 e8 f5 f4 e5 e4 f1 f0 e1 e0}
vshufps _r4, _r4, _r5, 0xEE \ // r4 = {f15 f14 e15 e14 f11 f10 e11 e10 f7 f6 e7 e6 f3 f2 e3 e2}
vshufps _t1, _r6, _r7, 0x44 \ // t1 = {h13 h12 g13 g12 h9 h8 g9 g8 h5 h4 g5 g4 h1 h0 g1 g0}
vshufps _r6, _r6, _r7, 0xEE \ // r6 = {h15 h14 g15 g14 h11 h10 g11 g10 h7 h6 g7 g6 h3 h2 g3 g2}
\
vshufps _r7, _r2, _t1, 0xDD \ // r7 = {h13 g13 f13 e13 h9 g9 f9 e9 h5 g5 f5 e5 h1 g1 f1 e1}
vshufps _r5, _r4, _r6, 0x88 \ // r5 = {h14 g14 f14 e14 h10 g10 f10 e10 h6 g6 f6 e6 h2 g2 f2 e2}
vshufps _r4, _r4, _r6, 0xDD \ // r4 = {h15 g15 f15 e15 h11 g11 f11 e11 h7 g7 f7 e7 h3 g3 f3 e3}
vshufps _r2, _r2, _t1, 0x88 \ // r2 = {h12 g12 f12 e12 h8 g8 f8 e8 h4 g4 f4 e4 h0 g0 f0 e0}
\
\ // use r6 in place of t0
vshufps _r6, _r8, _r9, 0x44 \ // r6 = {j13 j12 i13 i12 j9 j8 i9 i8 j5 j4 i5 i4 j1 j0 i1 i0}
vshufps _r8, _r8, _r9, 0xEE \ // r8 = {j15 j14 i15 i14 j11 j10 i11 i10 j7 j6 i7 i6 j3 j2 i3 i2}
vshufps _t1, _r10, _r11, 0x44 \ // t1 = {l13 l12 k13 k12 l9 l8 k9 k8 l5 l4 k5 k4 l1 l0 k1 k0}
vshufps _r10, _r10, _r11, 0xEE \ // r10 = {l15 l14 k15 k14 l11 l10 k11 k10 l7 l6 k7 k6 l3 l2 k3 k2}
\
vshufps _r11, _r6, _t1, 0xDD \ // r11 = {l13 k13 j13 113 l9 k9 j9 i9 l5 k5 j5 i5 l1 k1 j1 i1}
vshufps _r9, _r8, _r10, 0x88 \ // r9 = {l14 k14 j14 114 l10 k10 j10 i10 l6 k6 j6 i6 l2 k2 j2 i2}
vshufps _r8, _r8, _r10, 0xDD \ // r8 = {l15 k15 j15 115 l11 k11 j11 i11 l7 k7 j7 i7 l3 k3 j3 i3}
vshufps _r6, _r6, _t1, 0x88 \ // r6 = {l12 k12 j12 112 l8 k8 j8 i8 l4 k4 j4 i4 l0 k0 j0 i0}
\
\ // use r10 in place of t0
vshufps _r10, _r12, _r13, 0x44 \ // r10 = {n13 n12 m13 m12 n9 n8 m9 m8 n5 n4 m5 m4 n1 n0 a1 m0}
vshufps _r12, _r12, _r13, 0xEE \ // r12 = {n15 n14 m15 m14 n11 n10 m11 m10 n7 n6 m7 m6 n3 n2 a3 m2}
vshufps _t1, _r14, _r15, 0x44 \ // t1 = {p13 p12 013 012 p9 p8 09 08 p5 p4 05 04 p1 p0 01 00}
vshufps _r14, _r14, _r15, 0xEE \ // r14 = {p15 p14 015 014 p11 p10 011 010 p7 p6 07 06 p3 p2 03 02}
\
vshufps _r15, _r10, _t1, 0xDD \ // r15 = {p13 013 n13 m13 p9 09 n9 m9 p5 05 n5 m5 p1 01 n1 m1}
vshufps _r13, _r12, _r14, 0x88 \ // r13 = {p14 014 n14 m14 p10 010 n10 m10 p6 06 n6 m6 p2 02 n2 m2}
vshufps _r12, _r12, _r14, 0xDD \ // r12 = {p15 015 n15 m15 p11 011 n11 m11 p7 07 n7 m7 p3 03 n3 m3}
vshufps _r10, _r10, _t1, 0x88 \ // r10 = {p12 012 n12 m12 p8 08 n8 m8 p4 04 n4 m4 p0 00 n0 m0}
\
\ // At this point, the registers that contain interesting data are:
\ // t0, r3, r1, r0, r2, r7, r5, r4, r6, r11, r9, r8, r10, r15, r13, r12
\ // Can use t1 and r14 as scratch registers
LEAQ PSHUFFLE_TRANSPOSE16_MASK1<>(SB), BX \
LEAQ PSHUFFLE_TRANSPOSE16_MASK2<>(SB), R8 \
\
vmovdqu32 _r14, [rbx] \
vpermi2q _r14, _t0, _r2 \ // r14 = {h8 g8 f8 e8 d8 c8 b8 a8 h0 g0 f0 e0 d0 c0 b0 a0}
vmovdqu32 _t1, [r8] \
vpermi2q _t1, _t0, _r2 \ // t1 = {h12 g12 f12 e12 d12 c12 b12 a12 h4 g4 f4 e4 d4 c4 b4 a4}
\
vmovdqu32 _r2, [rbx] \
vpermi2q _r2, _r3, _r7 \ // r2 = {h9 g9 f9 e9 d9 c9 b9 a9 h1 g1 f1 e1 d1 c1 b1 a1}
vmovdqu32 _t0, [r8] \
vpermi2q _t0, _r3, _r7 \ // t0 = {h13 g13 f13 e13 d13 c13 b13 a13 h5 g5 f5 e5 d5 c5 b5 a5}
\
vmovdqu32 _r3, [rbx] \
vpermi2q _r3, _r1, _r5 \ // r3 = {h10 g10 f10 e10 d10 c10 b10 a10 h2 g2 f2 e2 d2 c2 b2 a2}
vmovdqu32 _r7, [r8] \
vpermi2q _r7, _r1, _r5 \ // r7 = {h14 g14 f14 e14 d14 c14 b14 a14 h6 g6 f6 e6 d6 c6 b6 a6}
\
vmovdqu32 _r1, [rbx] \
vpermi2q _r1, _r0, _r4 \ // r1 = {h11 g11 f11 e11 d11 c11 b11 a11 h3 g3 f3 e3 d3 c3 b3 a3}
vmovdqu32 _r5, [r8] \
vpermi2q _r5, _r0, _r4 \ // r5 = {h15 g15 f15 e15 d15 c15 b15 a15 h7 g7 f7 e7 d7 c7 b7 a7}
\
vmovdqu32 _r0, [rbx] \
vpermi2q _r0, _r6, _r10 \ // r0 = {p8 o8 n8 m8 l8 k8 j8 i8 p0 o0 n0 m0 l0 k0 j0 i0}
vmovdqu32 _r4, [r8] \
vpermi2q _r4, _r6, _r10 \ // r4 = {p12 o12 n12 m12 l12 k12 j12 i12 p4 o4 n4 m4 l4 k4 j4 i4}
\
vmovdqu32 _r6, [rbx] \
vpermi2q _r6, _r11, _r15 \ // r6 = {p9 o9 n9 m9 l9 k9 j9 i9 p1 o1 n1 m1 l1 k1 j1 i1}
vmovdqu32 _r10, [r8] \
vpermi2q _r10, _r11, _r15 \ // r10 = {p13 o13 n13 m13 l13 k13 j13 i13 p5 o5 n5 m5 l5 k5 j5 i5}
\
vmovdqu32 _r11, [rbx] \
vpermi2q _r11, _r9, _r13 \ // r11 = {p10 o10 n10 m10 l10 k10 j10 i10 p2 o2 n2 m2 l2 k2 j2 i2}
vmovdqu32 _r15, [r8] \
vpermi2q _r15, _r9, _r13 \ // r15 = {p14 o14 n14 m14 l14 k14 j14 i14 p6 o6 n6 m6 l6 k6 j6 i6}
\
vmovdqu32 _r9, [rbx] \
vpermi2q _r9, _r8, _r12 \ // r9 = {p11 o11 n11 m11 l11 k11 j11 i11 p3 o3 n3 m3 l3 k3 j3 i3}
vmovdqu32 _r13, [r8] \
vpermi2q _r13, _r8, _r12 \ // r13 = {p15 o15 n15 m15 l15 k15 j15 i15 p7 o7 n7 m7 l7 k7 j7 i7}
\
\ // At this point r8 and r12 can be used as scratch registers
vshuff64x2 _r8, _r14, _r0, 0xEE \ // r8 = {p8 o8 n8 m8 l8 k8 j8 i8 h8 g8 f8 e8 d8 c8 b8 a8}
vshuff64x2 _r0, _r14, _r0, 0x44 \ // r0 = {p0 o0 n0 m0 l0 k0 j0 i0 h0 g0 f0 e0 d0 c0 b0 a0}
\
vshuff64x2 _r12, _t1, _r4, 0xEE \ // r12 = {p12 o12 n12 m12 l12 k12 j12 i12 h12 g12 f12 e12 d12 c12 b12 a12}
vshuff64x2 _r4, _t1, _r4, 0x44 \ // r4 = {p4 o4 n4 m4 l4 k4 j4 i4 h4 g4 f4 e4 d4 c4 b4 a4}
\
vshuff64x2 _r14, _r7, _r15, 0xEE \ // r14 = {p14 o14 n14 m14 l14 k14 j14 i14 h14 g14 f14 e14 d14 c14 b14 a14}
vshuff64x2 _t1, _r7, _r15, 0x44 \ // t1 = {p6 o6 n6 m6 l6 k6 j6 i6 h6 g6 f6 e6 d6 c6 b6 a6}
\
vshuff64x2 _r15, _r5, _r13, 0xEE \ // r15 = {p15 o15 n15 m15 l15 k15 j15 i15 h15 g15 f15 e15 d15 c15 b15 a15}
vshuff64x2 _r7, _r5, _r13, 0x44 \ // r7 = {p7 o7 n7 m7 l7 k7 j7 i7 h7 g7 f7 e7 d7 c7 b7 a7}
\
vshuff64x2 _r13, _t0, _r10, 0xEE \ // r13 = {p13 o13 n13 m13 l13 k13 j13 i13 h13 g13 f13 e13 d13 c13 b13 a13}
vshuff64x2 _r5, _t0, _r10, 0x44 \ // r5 = {p5 o5 n5 m5 l5 k5 j5 i5 h5 g5 f5 e5 d5 c5 b5 a5}
\
vshuff64x2 _r10, _r3, _r11, 0xEE \ // r10 = {p10 o10 n10 m10 l10 k10 j10 i10 h10 g10 f10 e10 d10 c10 b10 a10}
vshuff64x2 _t0, _r3, _r11, 0x44 \ // t0 = {p2 o2 n2 m2 l2 k2 j2 i2 h2 g2 f2 e2 d2 c2 b2 a2}
\
vshuff64x2 _r11, _r1, _r9, 0xEE \ // r11 = {p11 o11 n11 m11 l11 k11 j11 i11 h11 g11 f11 e11 d11 c11 b11 a11}
vshuff64x2 _r3, _r1, _r9, 0x44 \ // r3 = {p3 o3 n3 m3 l3 k3 j3 i3 h3 g3 f3 e3 d3 c3 b3 a3}
\
vshuff64x2 _r9, _r2, _r6, 0xEE \ // r9 = {p9 o9 n9 m9 l9 k9 j9 i9 h9 g9 f9 e9 d9 c9 b9 a9}
vshuff64x2 _r1, _r2, _r6, 0x44 \ // r1 = {p1 o1 n1 m1 l1 k1 j1 i1 h1 g1 f1 e1 d1 c1 b1 a1}
\
vmovdqu32 _r2, _t0 \ // r2 = {p2 o2 n2 m2 l2 k2 j2 i2 h2 g2 f2 e2 d2 c2 b2 a2}
vmovdqu32 _r6, _t1 \ // r6 = {p6 o6 n6 m6 l6 k6 j6 i6 h6 g6 f6 e6 d6 c6 b6 a6}
// CH(A, B, C) = (A&B) ^ (~A&C)
// MAJ(E, F, G) = (E&F) ^ (E&G) ^ (F&G)
// SIGMA0 = ROR_2 ^ ROR_13 ^ ROR_22
// SIGMA1 = ROR_6 ^ ROR_11 ^ ROR_25
// sigma0 = ROR_7 ^ ROR_18 ^ SHR_3
// sigma1 = ROR_17 ^ ROR_19 ^ SHR_10
// Main processing loop per round
#define PROCESS_LOOP(_WT, _ROUND, _A, _B, _C, _D, _E, _F, _G, _H) \
\ // T1 = H + SIGMA1(E) + CH(E, F, G) + Kt + Wt
\ // T2 = SIGMA0(A) + MAJ(A, B, C)
\ // H=G, G=F, F=E, E=D+T1, D=C, C=B, B=A, A=T1+T2
\
\ // H becomes T2, then add T1 for A
\ // D becomes D + T1 for E
\
vpaddd T1, _H, TMP3 \ // T1 = H + Kt
vmovdqu32 TMP0, _E \
vprord TMP1, _E, 6 \ // ROR_6(E)
vprord TMP2, _E, 11 \ // ROR_11(E)
vprord TMP3, _E, 25 \ // ROR_25(E)
vpternlogd TMP0, _F, _G, 0xCA \ // TMP0 = CH(E,F,G)
vpaddd T1, T1, _WT \ // T1 = T1 + Wt
vpternlogd TMP1, TMP2, TMP3, 0x96 \ // TMP1 = SIGMA1(E)
vpaddd T1, T1, TMP0 \ // T1 = T1 + CH(E,F,G)
vpaddd T1, T1, TMP1 \ // T1 = T1 + SIGMA1(E)
vpaddd _D, _D, T1 \ // D = D + T1
\
vprord _H, _A, 2 \ // ROR_2(A)
vprord TMP2, _A, 13 \ // ROR_13(A)
vprord TMP3, _A, 22 \ // ROR_22(A)
vmovdqu32 TMP0, _A \
vpternlogd TMP0, _B, _C, 0xE8 \ // TMP0 = MAJ(A,B,C)
vpternlogd _H, TMP2, TMP3, 0x96 \ // H(T2) = SIGMA0(A)
vpaddd _H, _H, TMP0 \ // H(T2) = SIGMA0(A) + MAJ(A,B,C)
vpaddd _H, _H, T1 \ // H(A) = H(T2) + T1
\
vmovdqu32 TMP3, [TBL + ((_ROUND+1)*64)] \ // Next Kt
#define MSG_SCHED_ROUND_16_63(_WT, _WTp1, _WTp9, _WTp14) \
vprord TMP4, _WTp14, 17 \ // ROR_17(Wt-2)
vprord TMP5, _WTp14, 19 \ // ROR_19(Wt-2)
vpsrld TMP6, _WTp14, 10 \ // SHR_10(Wt-2)
vpternlogd TMP4, TMP5, TMP6, 0x96 \ // TMP4 = sigma1(Wt-2)
\
vpaddd _WT, _WT, TMP4 \ // Wt = Wt-16 + sigma1(Wt-2)
vpaddd _WT, _WT, _WTp9 \ // Wt = Wt-16 + sigma1(Wt-2) + Wt-7
\
vprord TMP4, _WTp1, 7 \ // ROR_7(Wt-15)
vprord TMP5, _WTp1, 18 \ // ROR_18(Wt-15)
vpsrld TMP6, _WTp1, 3 \ // SHR_3(Wt-15)
vpternlogd TMP4, TMP5, TMP6, 0x96 \ // TMP4 = sigma0(Wt-15)
\
vpaddd _WT, _WT, TMP4 \ // Wt = Wt-16 + sigma1(Wt-2) +
\ // Wt-7 + sigma0(Wt-15) +
// Note this is reading in a block of data for one lane
// When all 16 are read, the data must be transposed to build msg schedule
#define MSG_SCHED_ROUND_00_15(_WT, OFFSET, LABEL) \
TESTQ $(1<<OFFSET), MASK_P9 \
JE LABEL \
MOVQ OFFSET*24(INPUT_P9), R9 \
vmovups _WT, [inp0+IDX] \
LABEL: \
#define MASKED_LOAD(_WT, OFFSET, LABEL) \
TESTQ $(1<<OFFSET), MASK_P9 \
JE LABEL \
MOVQ OFFSET*24(INPUT_P9), R9 \
vmovups _WT,[inp0+IDX] \
LABEL: \
TEXT ·sha256_x16_avx512(SB), 7, $0
MOVQ digests+0(FP), STATE_P9 //
MOVQ scratch+8(FP), SCRATCH_P9
MOVQ mask_len+32(FP), INP_SIZE_P9 // number of blocks to process
MOVQ mask+24(FP), MASKP_P9
MOVQ (MASKP_P9), MASK_P9
kmovq k1, mask
LEAQ inputs+48(FP), INPUT_P9
// Initialize digests
vmovdqu32 A, [STATE + 0*SHA256_DIGEST_ROW_SIZE]
vmovdqu32 B, [STATE + 1*SHA256_DIGEST_ROW_SIZE]
vmovdqu32 C, [STATE + 2*SHA256_DIGEST_ROW_SIZE]
vmovdqu32 D, [STATE + 3*SHA256_DIGEST_ROW_SIZE]
vmovdqu32 E, [STATE + 4*SHA256_DIGEST_ROW_SIZE]
vmovdqu32 F, [STATE + 5*SHA256_DIGEST_ROW_SIZE]
vmovdqu32 G, [STATE + 6*SHA256_DIGEST_ROW_SIZE]
vmovdqu32 H, [STATE + 7*SHA256_DIGEST_ROW_SIZE]
MOVQ table+16(FP), TBL_P9
xor IDX, IDX
// Read in first block of input data
MASKED_LOAD( W0, 0, skipInput0)
MASKED_LOAD( W1, 1, skipInput1)
MASKED_LOAD( W2, 2, skipInput2)
MASKED_LOAD( W3, 3, skipInput3)
MASKED_LOAD( W4, 4, skipInput4)
MASKED_LOAD( W5, 5, skipInput5)
MASKED_LOAD( W6, 6, skipInput6)
MASKED_LOAD( W7, 7, skipInput7)
MASKED_LOAD( W8, 8, skipInput8)
MASKED_LOAD( W9, 9, skipInput9)
MASKED_LOAD(W10, 10, skipInput10)
MASKED_LOAD(W11, 11, skipInput11)
MASKED_LOAD(W12, 12, skipInput12)
MASKED_LOAD(W13, 13, skipInput13)
MASKED_LOAD(W14, 14, skipInput14)
MASKED_LOAD(W15, 15, skipInput15)
lloop:
LEAQ PSHUFFLE_BYTE_FLIP_MASK<>(SB), TBL_P9
vmovdqu32 TMP2, [TBL]
// Get first K from table
MOVQ table+16(FP), TBL_P9
vmovdqu32 TMP3, [TBL]
// Save digests for later addition
vmovdqu32 [SCRATCH + 64*0], A
vmovdqu32 [SCRATCH + 64*1], B
vmovdqu32 [SCRATCH + 64*2], C
vmovdqu32 [SCRATCH + 64*3], D
vmovdqu32 [SCRATCH + 64*4], E
vmovdqu32 [SCRATCH + 64*5], F
vmovdqu32 [SCRATCH + 64*6], G
vmovdqu32 [SCRATCH + 64*7], H
add IDX, 64
// Transpose input data
TRANSPOSE16(W0, W1, W2, W3, W4, W5, W6, W7, W8, W9, W10, W11, W12, W13, W14, W15, TMP0, TMP1)
vpshufb W0, W0, TMP2
vpshufb W1, W1, TMP2
vpshufb W2, W2, TMP2
vpshufb W3, W3, TMP2
vpshufb W4, W4, TMP2
vpshufb W5, W5, TMP2
vpshufb W6, W6, TMP2
vpshufb W7, W7, TMP2
vpshufb W8, W8, TMP2
vpshufb W9, W9, TMP2
vpshufb W10, W10, TMP2
vpshufb W11, W11, TMP2
vpshufb W12, W12, TMP2
vpshufb W13, W13, TMP2
vpshufb W14, W14, TMP2
vpshufb W15, W15, TMP2
// MSG Schedule for W0-W15 is now complete in registers
// Process first 48 rounds
// Calculate next Wt+16 after processing is complete and Wt is unneeded
PROCESS_LOOP( W0, 0, A, B, C, D, E, F, G, H)
MSG_SCHED_ROUND_16_63( W0, W1, W9, W14)
PROCESS_LOOP( W1, 1, H, A, B, C, D, E, F, G)
MSG_SCHED_ROUND_16_63( W1, W2, W10, W15)
PROCESS_LOOP( W2, 2, G, H, A, B, C, D, E, F)
MSG_SCHED_ROUND_16_63( W2, W3, W11, W0)
PROCESS_LOOP( W3, 3, F, G, H, A, B, C, D, E)
MSG_SCHED_ROUND_16_63( W3, W4, W12, W1)
PROCESS_LOOP( W4, 4, E, F, G, H, A, B, C, D)
MSG_SCHED_ROUND_16_63( W4, W5, W13, W2)
PROCESS_LOOP( W5, 5, D, E, F, G, H, A, B, C)
MSG_SCHED_ROUND_16_63( W5, W6, W14, W3)
PROCESS_LOOP( W6, 6, C, D, E, F, G, H, A, B)
MSG_SCHED_ROUND_16_63( W6, W7, W15, W4)
PROCESS_LOOP( W7, 7, B, C, D, E, F, G, H, A)
MSG_SCHED_ROUND_16_63( W7, W8, W0, W5)
PROCESS_LOOP( W8, 8, A, B, C, D, E, F, G, H)
MSG_SCHED_ROUND_16_63( W8, W9, W1, W6)
PROCESS_LOOP( W9, 9, H, A, B, C, D, E, F, G)
MSG_SCHED_ROUND_16_63( W9, W10, W2, W7)
PROCESS_LOOP(W10, 10, G, H, A, B, C, D, E, F)
MSG_SCHED_ROUND_16_63(W10, W11, W3, W8)
PROCESS_LOOP(W11, 11, F, G, H, A, B, C, D, E)
MSG_SCHED_ROUND_16_63(W11, W12, W4, W9)
PROCESS_LOOP(W12, 12, E, F, G, H, A, B, C, D)
MSG_SCHED_ROUND_16_63(W12, W13, W5, W10)
PROCESS_LOOP(W13, 13, D, E, F, G, H, A, B, C)
MSG_SCHED_ROUND_16_63(W13, W14, W6, W11)
PROCESS_LOOP(W14, 14, C, D, E, F, G, H, A, B)
MSG_SCHED_ROUND_16_63(W14, W15, W7, W12)
PROCESS_LOOP(W15, 15, B, C, D, E, F, G, H, A)
MSG_SCHED_ROUND_16_63(W15, W0, W8, W13)
PROCESS_LOOP( W0, 16, A, B, C, D, E, F, G, H)
MSG_SCHED_ROUND_16_63( W0, W1, W9, W14)
PROCESS_LOOP( W1, 17, H, A, B, C, D, E, F, G)
MSG_SCHED_ROUND_16_63( W1, W2, W10, W15)
PROCESS_LOOP( W2, 18, G, H, A, B, C, D, E, F)
MSG_SCHED_ROUND_16_63( W2, W3, W11, W0)
PROCESS_LOOP( W3, 19, F, G, H, A, B, C, D, E)
MSG_SCHED_ROUND_16_63( W3, W4, W12, W1)
PROCESS_LOOP( W4, 20, E, F, G, H, A, B, C, D)
MSG_SCHED_ROUND_16_63( W4, W5, W13, W2)
PROCESS_LOOP( W5, 21, D, E, F, G, H, A, B, C)
MSG_SCHED_ROUND_16_63( W5, W6, W14, W3)
PROCESS_LOOP( W6, 22, C, D, E, F, G, H, A, B)
MSG_SCHED_ROUND_16_63( W6, W7, W15, W4)
PROCESS_LOOP( W7, 23, B, C, D, E, F, G, H, A)
MSG_SCHED_ROUND_16_63( W7, W8, W0, W5)
PROCESS_LOOP( W8, 24, A, B, C, D, E, F, G, H)
MSG_SCHED_ROUND_16_63( W8, W9, W1, W6)
PROCESS_LOOP( W9, 25, H, A, B, C, D, E, F, G)
MSG_SCHED_ROUND_16_63( W9, W10, W2, W7)
PROCESS_LOOP(W10, 26, G, H, A, B, C, D, E, F)
MSG_SCHED_ROUND_16_63(W10, W11, W3, W8)
PROCESS_LOOP(W11, 27, F, G, H, A, B, C, D, E)
MSG_SCHED_ROUND_16_63(W11, W12, W4, W9)
PROCESS_LOOP(W12, 28, E, F, G, H, A, B, C, D)
MSG_SCHED_ROUND_16_63(W12, W13, W5, W10)
PROCESS_LOOP(W13, 29, D, E, F, G, H, A, B, C)
MSG_SCHED_ROUND_16_63(W13, W14, W6, W11)
PROCESS_LOOP(W14, 30, C, D, E, F, G, H, A, B)
MSG_SCHED_ROUND_16_63(W14, W15, W7, W12)
PROCESS_LOOP(W15, 31, B, C, D, E, F, G, H, A)
MSG_SCHED_ROUND_16_63(W15, W0, W8, W13)
PROCESS_LOOP( W0, 32, A, B, C, D, E, F, G, H)
MSG_SCHED_ROUND_16_63( W0, W1, W9, W14)
PROCESS_LOOP( W1, 33, H, A, B, C, D, E, F, G)
MSG_SCHED_ROUND_16_63( W1, W2, W10, W15)
PROCESS_LOOP( W2, 34, G, H, A, B, C, D, E, F)
MSG_SCHED_ROUND_16_63( W2, W3, W11, W0)
PROCESS_LOOP( W3, 35, F, G, H, A, B, C, D, E)
MSG_SCHED_ROUND_16_63( W3, W4, W12, W1)
PROCESS_LOOP( W4, 36, E, F, G, H, A, B, C, D)
MSG_SCHED_ROUND_16_63( W4, W5, W13, W2)
PROCESS_LOOP( W5, 37, D, E, F, G, H, A, B, C)
MSG_SCHED_ROUND_16_63( W5, W6, W14, W3)
PROCESS_LOOP( W6, 38, C, D, E, F, G, H, A, B)
MSG_SCHED_ROUND_16_63( W6, W7, W15, W4)
PROCESS_LOOP( W7, 39, B, C, D, E, F, G, H, A)
MSG_SCHED_ROUND_16_63( W7, W8, W0, W5)
PROCESS_LOOP( W8, 40, A, B, C, D, E, F, G, H)
MSG_SCHED_ROUND_16_63( W8, W9, W1, W6)
PROCESS_LOOP( W9, 41, H, A, B, C, D, E, F, G)
MSG_SCHED_ROUND_16_63( W9, W10, W2, W7)
PROCESS_LOOP(W10, 42, G, H, A, B, C, D, E, F)
MSG_SCHED_ROUND_16_63(W10, W11, W3, W8)
PROCESS_LOOP(W11, 43, F, G, H, A, B, C, D, E)
MSG_SCHED_ROUND_16_63(W11, W12, W4, W9)
PROCESS_LOOP(W12, 44, E, F, G, H, A, B, C, D)
MSG_SCHED_ROUND_16_63(W12, W13, W5, W10)
PROCESS_LOOP(W13, 45, D, E, F, G, H, A, B, C)
MSG_SCHED_ROUND_16_63(W13, W14, W6, W11)
PROCESS_LOOP(W14, 46, C, D, E, F, G, H, A, B)
MSG_SCHED_ROUND_16_63(W14, W15, W7, W12)
PROCESS_LOOP(W15, 47, B, C, D, E, F, G, H, A)
MSG_SCHED_ROUND_16_63(W15, W0, W8, W13)
// Check if this is the last block
sub INP_SIZE, 1
JE lastLoop
// Load next mask for inputs
ADDQ $8, MASKP_P9
MOVQ (MASKP_P9), MASK_P9
// Process last 16 rounds
// Read in next block msg data for use in first 16 words of msg sched
PROCESS_LOOP( W0, 48, A, B, C, D, E, F, G, H)
MSG_SCHED_ROUND_00_15( W0, 0, skipNext0)
PROCESS_LOOP( W1, 49, H, A, B, C, D, E, F, G)
MSG_SCHED_ROUND_00_15( W1, 1, skipNext1)
PROCESS_LOOP( W2, 50, G, H, A, B, C, D, E, F)
MSG_SCHED_ROUND_00_15( W2, 2, skipNext2)
PROCESS_LOOP( W3, 51, F, G, H, A, B, C, D, E)
MSG_SCHED_ROUND_00_15( W3, 3, skipNext3)
PROCESS_LOOP( W4, 52, E, F, G, H, A, B, C, D)
MSG_SCHED_ROUND_00_15( W4, 4, skipNext4)
PROCESS_LOOP( W5, 53, D, E, F, G, H, A, B, C)
MSG_SCHED_ROUND_00_15( W5, 5, skipNext5)
PROCESS_LOOP( W6, 54, C, D, E, F, G, H, A, B)
MSG_SCHED_ROUND_00_15( W6, 6, skipNext6)
PROCESS_LOOP( W7, 55, B, C, D, E, F, G, H, A)
MSG_SCHED_ROUND_00_15( W7, 7, skipNext7)
PROCESS_LOOP( W8, 56, A, B, C, D, E, F, G, H)
MSG_SCHED_ROUND_00_15( W8, 8, skipNext8)
PROCESS_LOOP( W9, 57, H, A, B, C, D, E, F, G)
MSG_SCHED_ROUND_00_15( W9, 9, skipNext9)
PROCESS_LOOP(W10, 58, G, H, A, B, C, D, E, F)
MSG_SCHED_ROUND_00_15(W10, 10, skipNext10)
PROCESS_LOOP(W11, 59, F, G, H, A, B, C, D, E)
MSG_SCHED_ROUND_00_15(W11, 11, skipNext11)
PROCESS_LOOP(W12, 60, E, F, G, H, A, B, C, D)
MSG_SCHED_ROUND_00_15(W12, 12, skipNext12)
PROCESS_LOOP(W13, 61, D, E, F, G, H, A, B, C)
MSG_SCHED_ROUND_00_15(W13, 13, skipNext13)
PROCESS_LOOP(W14, 62, C, D, E, F, G, H, A, B)
MSG_SCHED_ROUND_00_15(W14, 14, skipNext14)
PROCESS_LOOP(W15, 63, B, C, D, E, F, G, H, A)
MSG_SCHED_ROUND_00_15(W15, 15, skipNext15)
// Add old digest
vmovdqu32 TMP2, A
vmovdqu32 A, [SCRATCH + 64*0]
vpaddd A{k1}, A, TMP2
vmovdqu32 TMP2, B
vmovdqu32 B, [SCRATCH + 64*1]
vpaddd B{k1}, B, TMP2
vmovdqu32 TMP2, C
vmovdqu32 C, [SCRATCH + 64*2]
vpaddd C{k1}, C, TMP2
vmovdqu32 TMP2, D
vmovdqu32 D, [SCRATCH + 64*3]
vpaddd D{k1}, D, TMP2
vmovdqu32 TMP2, E
vmovdqu32 E, [SCRATCH + 64*4]
vpaddd E{k1}, E, TMP2
vmovdqu32 TMP2, F
vmovdqu32 F, [SCRATCH + 64*5]
vpaddd F{k1}, F, TMP2
vmovdqu32 TMP2, G
vmovdqu32 G, [SCRATCH + 64*6]
vpaddd G{k1}, G, TMP2
vmovdqu32 TMP2, H
vmovdqu32 H, [SCRATCH + 64*7]
vpaddd H{k1}, H, TMP2
kmovq k1, mask
JMP lloop
lastLoop:
// Process last 16 rounds
PROCESS_LOOP( W0, 48, A, B, C, D, E, F, G, H)
PROCESS_LOOP( W1, 49, H, A, B, C, D, E, F, G)
PROCESS_LOOP( W2, 50, G, H, A, B, C, D, E, F)
PROCESS_LOOP( W3, 51, F, G, H, A, B, C, D, E)
PROCESS_LOOP( W4, 52, E, F, G, H, A, B, C, D)
PROCESS_LOOP( W5, 53, D, E, F, G, H, A, B, C)
PROCESS_LOOP( W6, 54, C, D, E, F, G, H, A, B)
PROCESS_LOOP( W7, 55, B, C, D, E, F, G, H, A)
PROCESS_LOOP( W8, 56, A, B, C, D, E, F, G, H)
PROCESS_LOOP( W9, 57, H, A, B, C, D, E, F, G)
PROCESS_LOOP(W10, 58, G, H, A, B, C, D, E, F)
PROCESS_LOOP(W11, 59, F, G, H, A, B, C, D, E)
PROCESS_LOOP(W12, 60, E, F, G, H, A, B, C, D)
PROCESS_LOOP(W13, 61, D, E, F, G, H, A, B, C)
PROCESS_LOOP(W14, 62, C, D, E, F, G, H, A, B)
PROCESS_LOOP(W15, 63, B, C, D, E, F, G, H, A)
// Add old digest
vmovdqu32 TMP2, A
vmovdqu32 A, [SCRATCH + 64*0]
vpaddd A{k1}, A, TMP2
vmovdqu32 TMP2, B
vmovdqu32 B, [SCRATCH + 64*1]
vpaddd B{k1}, B, TMP2
vmovdqu32 TMP2, C
vmovdqu32 C, [SCRATCH + 64*2]
vpaddd C{k1}, C, TMP2
vmovdqu32 TMP2, D
vmovdqu32 D, [SCRATCH + 64*3]
vpaddd D{k1}, D, TMP2
vmovdqu32 TMP2, E
vmovdqu32 E, [SCRATCH + 64*4]
vpaddd E{k1}, E, TMP2
vmovdqu32 TMP2, F
vmovdqu32 F, [SCRATCH + 64*5]
vpaddd F{k1}, F, TMP2
vmovdqu32 TMP2, G
vmovdqu32 G, [SCRATCH + 64*6]
vpaddd G{k1}, G, TMP2
vmovdqu32 TMP2, H
vmovdqu32 H, [SCRATCH + 64*7]
vpaddd H{k1}, H, TMP2
// Write out digest
vmovdqu32 [STATE + 0*SHA256_DIGEST_ROW_SIZE], A
vmovdqu32 [STATE + 1*SHA256_DIGEST_ROW_SIZE], B
vmovdqu32 [STATE + 2*SHA256_DIGEST_ROW_SIZE], C
vmovdqu32 [STATE + 3*SHA256_DIGEST_ROW_SIZE], D
vmovdqu32 [STATE + 4*SHA256_DIGEST_ROW_SIZE], E
vmovdqu32 [STATE + 5*SHA256_DIGEST_ROW_SIZE], F
vmovdqu32 [STATE + 6*SHA256_DIGEST_ROW_SIZE], G
vmovdqu32 [STATE + 7*SHA256_DIGEST_ROW_SIZE], H
VZEROUPPER
RET
//
// Tables
//
DATA PSHUFFLE_BYTE_FLIP_MASK<>+0x000(SB)/8, $0x0405060700010203
DATA PSHUFFLE_BYTE_FLIP_MASK<>+0x008(SB)/8, $0x0c0d0e0f08090a0b
DATA PSHUFFLE_BYTE_FLIP_MASK<>+0x010(SB)/8, $0x0405060700010203
DATA PSHUFFLE_BYTE_FLIP_MASK<>+0x018(SB)/8, $0x0c0d0e0f08090a0b
DATA PSHUFFLE_BYTE_FLIP_MASK<>+0x020(SB)/8, $0x0405060700010203
DATA PSHUFFLE_BYTE_FLIP_MASK<>+0x028(SB)/8, $0x0c0d0e0f08090a0b
DATA PSHUFFLE_BYTE_FLIP_MASK<>+0x030(SB)/8, $0x0405060700010203
DATA PSHUFFLE_BYTE_FLIP_MASK<>+0x038(SB)/8, $0x0c0d0e0f08090a0b
GLOBL PSHUFFLE_BYTE_FLIP_MASK<>(SB), 8, $64
DATA PSHUFFLE_TRANSPOSE16_MASK1<>+0x000(SB)/8, $0x0000000000000000
DATA PSHUFFLE_TRANSPOSE16_MASK1<>+0x008(SB)/8, $0x0000000000000001
DATA PSHUFFLE_TRANSPOSE16_MASK1<>+0x010(SB)/8, $0x0000000000000008
DATA PSHUFFLE_TRANSPOSE16_MASK1<>+0x018(SB)/8, $0x0000000000000009
DATA PSHUFFLE_TRANSPOSE16_MASK1<>+0x020(SB)/8, $0x0000000000000004
DATA PSHUFFLE_TRANSPOSE16_MASK1<>+0x028(SB)/8, $0x0000000000000005
DATA PSHUFFLE_TRANSPOSE16_MASK1<>+0x030(SB)/8, $0x000000000000000C
DATA PSHUFFLE_TRANSPOSE16_MASK1<>+0x038(SB)/8, $0x000000000000000D
GLOBL PSHUFFLE_TRANSPOSE16_MASK1<>(SB), 8, $64
DATA PSHUFFLE_TRANSPOSE16_MASK2<>+0x000(SB)/8, $0x0000000000000002
DATA PSHUFFLE_TRANSPOSE16_MASK2<>+0x008(SB)/8, $0x0000000000000003
DATA PSHUFFLE_TRANSPOSE16_MASK2<>+0x010(SB)/8, $0x000000000000000A
DATA PSHUFFLE_TRANSPOSE16_MASK2<>+0x018(SB)/8, $0x000000000000000B
DATA PSHUFFLE_TRANSPOSE16_MASK2<>+0x020(SB)/8, $0x0000000000000006
DATA PSHUFFLE_TRANSPOSE16_MASK2<>+0x028(SB)/8, $0x0000000000000007
DATA PSHUFFLE_TRANSPOSE16_MASK2<>+0x030(SB)/8, $0x000000000000000E
DATA PSHUFFLE_TRANSPOSE16_MASK2<>+0x038(SB)/8, $0x000000000000000F
GLOBL PSHUFFLE_TRANSPOSE16_MASK2<>(SB), 8, $64

663
pkg/crypto/sha256/sha256blockAvx512_amd64.go
View File

@@ -1,663 +0,0 @@
//go:build !noasm && !appengine && gc
// +build !noasm,!appengine,gc
/*
* Minio Cloud Storage, (C) 2017 Minio, Inc.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package sha256
import (
"encoding/binary"
"errors"
"hash"
"sort"
"sync/atomic"
"time"
)
//go:noescape
func sha256X16Avx512(
digests *[512]byte, scratch *[512]byte, table *[512]uint64, mask []uint64,
inputs [16][]byte,
)
// Avx512ServerUID - Do not start at 0 but next multiple of 16 so as to be able to
// differentiate with default initialiation value of 0
const Avx512ServerUID = 16
var uidCounter uint64
// NewAvx512 - initialize sha256 Avx512 implementation.
func NewAvx512(a512srv *Avx512Server) hash.Hash {
uid := atomic.AddUint64(&uidCounter, 1)
return &Avx512Digest{uid: uid, a512srv: a512srv}
}
// Avx512Digest - Type for computing SHA256 using Avx512
type Avx512Digest struct {
uid uint64
a512srv *Avx512Server
x [chunk]byte
nx int
len uint64
final bool
result [Size]byte
}
// Size - Return size of checksum
func (d *Avx512Digest) Size() int { return Size }
// BlockSize - Return blocksize of checksum
func (d Avx512Digest) BlockSize() int { return BlockSize }
// Reset - reset sha digest to its initial values
func (d *Avx512Digest) Reset() {
d.a512srv.blocksCh <- blockInput{uid: d.uid, reset: true}
d.nx = 0
d.len = 0
d.final = false
}
// Write to digest
func (d *Avx512Digest) Write(p []byte) (nn int, err error) {
if d.final {
return 0, errors.New("Avx512Digest already finalized. Reset first before writing again")
}
nn = len(p)
d.len += uint64(nn)
if d.nx > 0 {
n := copy(d.x[d.nx:], p)
d.nx += n
if d.nx == chunk {
d.a512srv.blocksCh <- blockInput{uid: d.uid, msg: d.x[:]}
d.nx = 0
}
p = p[n:]
}
if len(p) >= chunk {
n := len(p) &^ (chunk - 1)
d.a512srv.blocksCh <- blockInput{uid: d.uid, msg: p[:n]}
p = p[n:]
}
if len(p) > 0 {
d.nx = copy(d.x[:], p)
}
return
}
// Sum - Return sha256 sum in bytes
func (d *Avx512Digest) Sum(in []byte) (result []byte) {
if d.final {
return append(in, d.result[:]...)
}
trail := make([]byte, 0, 128)
trail = append(trail, d.x[:d.nx]...)
len := d.len
// Padding. Add a 1 bit and 0 bits until 56 bytes mod 64.
var tmp [64]byte
tmp[0] = 0x80
if len%64 < 56 {
trail = append(trail, tmp[0:56-len%64]...)
} else {
trail = append(trail, tmp[0:64+56-len%64]...)
}
d.nx = 0
// Length in bits.
len <<= 3
for i := uint(0); i < 8; i++ {
tmp[i] = byte(len >> (56 - 8*i))
}
trail = append(trail, tmp[0:8]...)
sumCh := make(chan [Size]byte)
d.a512srv.blocksCh <- blockInput{
uid: d.uid, msg: trail, final: true, sumCh: sumCh,
}
d.result = <-sumCh
d.final = true
return append(in, d.result[:]...)
}
var table = [512]uint64{
0x428a2f98428a2f98, 0x428a2f98428a2f98, 0x428a2f98428a2f98,
0x428a2f98428a2f98,
0x428a2f98428a2f98, 0x428a2f98428a2f98, 0x428a2f98428a2f98,
0x428a2f98428a2f98,
0x7137449171374491, 0x7137449171374491, 0x7137449171374491,
0x7137449171374491,
0x7137449171374491, 0x7137449171374491, 0x7137449171374491,
0x7137449171374491,
0xb5c0fbcfb5c0fbcf, 0xb5c0fbcfb5c0fbcf, 0xb5c0fbcfb5c0fbcf,
0xb5c0fbcfb5c0fbcf,
0xb5c0fbcfb5c0fbcf, 0xb5c0fbcfb5c0fbcf, 0xb5c0fbcfb5c0fbcf,
0xb5c0fbcfb5c0fbcf,
0xe9b5dba5e9b5dba5, 0xe9b5dba5e9b5dba5, 0xe9b5dba5e9b5dba5,
0xe9b5dba5e9b5dba5,
0xe9b5dba5e9b5dba5, 0xe9b5dba5e9b5dba5, 0xe9b5dba5e9b5dba5,
0xe9b5dba5e9b5dba5,
0x3956c25b3956c25b, 0x3956c25b3956c25b, 0x3956c25b3956c25b,
0x3956c25b3956c25b,
0x3956c25b3956c25b, 0x3956c25b3956c25b, 0x3956c25b3956c25b,
0x3956c25b3956c25b,
0x59f111f159f111f1, 0x59f111f159f111f1, 0x59f111f159f111f1,
0x59f111f159f111f1,
0x59f111f159f111f1, 0x59f111f159f111f1, 0x59f111f159f111f1,
0x59f111f159f111f1,
0x923f82a4923f82a4, 0x923f82a4923f82a4, 0x923f82a4923f82a4,
0x923f82a4923f82a4,
0x923f82a4923f82a4, 0x923f82a4923f82a4, 0x923f82a4923f82a4,
0x923f82a4923f82a4,
0xab1c5ed5ab1c5ed5, 0xab1c5ed5ab1c5ed5, 0xab1c5ed5ab1c5ed5,
0xab1c5ed5ab1c5ed5,
0xab1c5ed5ab1c5ed5, 0xab1c5ed5ab1c5ed5, 0xab1c5ed5ab1c5ed5,
0xab1c5ed5ab1c5ed5,
0xd807aa98d807aa98, 0xd807aa98d807aa98, 0xd807aa98d807aa98,
0xd807aa98d807aa98,
0xd807aa98d807aa98, 0xd807aa98d807aa98, 0xd807aa98d807aa98,
0xd807aa98d807aa98,
0x12835b0112835b01, 0x12835b0112835b01, 0x12835b0112835b01,
0x12835b0112835b01,
0x12835b0112835b01, 0x12835b0112835b01, 0x12835b0112835b01,
0x12835b0112835b01,
0x243185be243185be, 0x243185be243185be, 0x243185be243185be,
0x243185be243185be,
0x243185be243185be, 0x243185be243185be, 0x243185be243185be,
0x243185be243185be,
0x550c7dc3550c7dc3, 0x550c7dc3550c7dc3, 0x550c7dc3550c7dc3,
0x550c7dc3550c7dc3,
0x550c7dc3550c7dc3, 0x550c7dc3550c7dc3, 0x550c7dc3550c7dc3,
0x550c7dc3550c7dc3,
0x72be5d7472be5d74, 0x72be5d7472be5d74, 0x72be5d7472be5d74,
0x72be5d7472be5d74,
0x72be5d7472be5d74, 0x72be5d7472be5d74, 0x72be5d7472be5d74,
0x72be5d7472be5d74,
0x80deb1fe80deb1fe, 0x80deb1fe80deb1fe, 0x80deb1fe80deb1fe,
0x80deb1fe80deb1fe,
0x80deb1fe80deb1fe, 0x80deb1fe80deb1fe, 0x80deb1fe80deb1fe,
0x80deb1fe80deb1fe,
0x9bdc06a79bdc06a7, 0x9bdc06a79bdc06a7, 0x9bdc06a79bdc06a7,
0x9bdc06a79bdc06a7,
0x9bdc06a79bdc06a7, 0x9bdc06a79bdc06a7, 0x9bdc06a79bdc06a7,
0x9bdc06a79bdc06a7,
0xc19bf174c19bf174, 0xc19bf174c19bf174, 0xc19bf174c19bf174,
0xc19bf174c19bf174,
0xc19bf174c19bf174, 0xc19bf174c19bf174, 0xc19bf174c19bf174,
0xc19bf174c19bf174,
0xe49b69c1e49b69c1, 0xe49b69c1e49b69c1, 0xe49b69c1e49b69c1,
0xe49b69c1e49b69c1,
0xe49b69c1e49b69c1, 0xe49b69c1e49b69c1, 0xe49b69c1e49b69c1,
0xe49b69c1e49b69c1,
0xefbe4786efbe4786, 0xefbe4786efbe4786, 0xefbe4786efbe4786,
0xefbe4786efbe4786,
0xefbe4786efbe4786, 0xefbe4786efbe4786, 0xefbe4786efbe4786,
0xefbe4786efbe4786,
0x0fc19dc60fc19dc6, 0x0fc19dc60fc19dc6, 0x0fc19dc60fc19dc6,
0x0fc19dc60fc19dc6,
0x0fc19dc60fc19dc6, 0x0fc19dc60fc19dc6, 0x0fc19dc60fc19dc6,
0x0fc19dc60fc19dc6,
0x240ca1cc240ca1cc, 0x240ca1cc240ca1cc, 0x240ca1cc240ca1cc,
0x240ca1cc240ca1cc,
0x240ca1cc240ca1cc, 0x240ca1cc240ca1cc, 0x240ca1cc240ca1cc,
0x240ca1cc240ca1cc,
0x2de92c6f2de92c6f, 0x2de92c6f2de92c6f, 0x2de92c6f2de92c6f,
0x2de92c6f2de92c6f,
0x2de92c6f2de92c6f, 0x2de92c6f2de92c6f, 0x2de92c6f2de92c6f,
0x2de92c6f2de92c6f,
0x4a7484aa4a7484aa, 0x4a7484aa4a7484aa, 0x4a7484aa4a7484aa,
0x4a7484aa4a7484aa,
0x4a7484aa4a7484aa, 0x4a7484aa4a7484aa, 0x4a7484aa4a7484aa,
0x4a7484aa4a7484aa,
0x5cb0a9dc5cb0a9dc, 0x5cb0a9dc5cb0a9dc, 0x5cb0a9dc5cb0a9dc,
0x5cb0a9dc5cb0a9dc,
0x5cb0a9dc5cb0a9dc, 0x5cb0a9dc5cb0a9dc, 0x5cb0a9dc5cb0a9dc,
0x5cb0a9dc5cb0a9dc,
0x76f988da76f988da, 0x76f988da76f988da, 0x76f988da76f988da,
0x76f988da76f988da,
0x76f988da76f988da, 0x76f988da76f988da, 0x76f988da76f988da,
0x76f988da76f988da,
0x983e5152983e5152, 0x983e5152983e5152, 0x983e5152983e5152,
0x983e5152983e5152,
0x983e5152983e5152, 0x983e5152983e5152, 0x983e5152983e5152,
0x983e5152983e5152,
0xa831c66da831c66d, 0xa831c66da831c66d, 0xa831c66da831c66d,
0xa831c66da831c66d,
0xa831c66da831c66d, 0xa831c66da831c66d, 0xa831c66da831c66d,
0xa831c66da831c66d,
0xb00327c8b00327c8, 0xb00327c8b00327c8, 0xb00327c8b00327c8,
0xb00327c8b00327c8,
0xb00327c8b00327c8, 0xb00327c8b00327c8, 0xb00327c8b00327c8,
0xb00327c8b00327c8,
0xbf597fc7bf597fc7, 0xbf597fc7bf597fc7, 0xbf597fc7bf597fc7,
0xbf597fc7bf597fc7,
0xbf597fc7bf597fc7, 0xbf597fc7bf597fc7, 0xbf597fc7bf597fc7,
0xbf597fc7bf597fc7,
0xc6e00bf3c6e00bf3, 0xc6e00bf3c6e00bf3, 0xc6e00bf3c6e00bf3,
0xc6e00bf3c6e00bf3,
0xc6e00bf3c6e00bf3, 0xc6e00bf3c6e00bf3, 0xc6e00bf3c6e00bf3,
0xc6e00bf3c6e00bf3,
0xd5a79147d5a79147, 0xd5a79147d5a79147, 0xd5a79147d5a79147,
0xd5a79147d5a79147,
0xd5a79147d5a79147, 0xd5a79147d5a79147, 0xd5a79147d5a79147,
0xd5a79147d5a79147,
0x06ca635106ca6351, 0x06ca635106ca6351, 0x06ca635106ca6351,
0x06ca635106ca6351,
0x06ca635106ca6351, 0x06ca635106ca6351, 0x06ca635106ca6351,
0x06ca635106ca6351,
0x1429296714292967, 0x1429296714292967, 0x1429296714292967,
0x1429296714292967,
0x1429296714292967, 0x1429296714292967, 0x1429296714292967,
0x1429296714292967,
0x27b70a8527b70a85, 0x27b70a8527b70a85, 0x27b70a8527b70a85,
0x27b70a8527b70a85,
0x27b70a8527b70a85, 0x27b70a8527b70a85, 0x27b70a8527b70a85,
0x27b70a8527b70a85,
0x2e1b21382e1b2138, 0x2e1b21382e1b2138, 0x2e1b21382e1b2138,
0x2e1b21382e1b2138,
0x2e1b21382e1b2138, 0x2e1b21382e1b2138, 0x2e1b21382e1b2138,
0x2e1b21382e1b2138,
0x4d2c6dfc4d2c6dfc, 0x4d2c6dfc4d2c6dfc, 0x4d2c6dfc4d2c6dfc,
0x4d2c6dfc4d2c6dfc,
0x4d2c6dfc4d2c6dfc, 0x4d2c6dfc4d2c6dfc, 0x4d2c6dfc4d2c6dfc,
0x4d2c6dfc4d2c6dfc,
0x53380d1353380d13, 0x53380d1353380d13, 0x53380d1353380d13,
0x53380d1353380d13,
0x53380d1353380d13, 0x53380d1353380d13, 0x53380d1353380d13,
0x53380d1353380d13,
0x650a7354650a7354, 0x650a7354650a7354, 0x650a7354650a7354,
0x650a7354650a7354,
0x650a7354650a7354, 0x650a7354650a7354, 0x650a7354650a7354,
0x650a7354650a7354,
0x766a0abb766a0abb, 0x766a0abb766a0abb, 0x766a0abb766a0abb,
0x766a0abb766a0abb,
0x766a0abb766a0abb, 0x766a0abb766a0abb, 0x766a0abb766a0abb,
0x766a0abb766a0abb,
0x81c2c92e81c2c92e, 0x81c2c92e81c2c92e, 0x81c2c92e81c2c92e,
0x81c2c92e81c2c92e,
0x81c2c92e81c2c92e, 0x81c2c92e81c2c92e, 0x81c2c92e81c2c92e,
0x81c2c92e81c2c92e,
0x92722c8592722c85, 0x92722c8592722c85, 0x92722c8592722c85,
0x92722c8592722c85,
0x92722c8592722c85, 0x92722c8592722c85, 0x92722c8592722c85,
0x92722c8592722c85,
0xa2bfe8a1a2bfe8a1, 0xa2bfe8a1a2bfe8a1, 0xa2bfe8a1a2bfe8a1,
0xa2bfe8a1a2bfe8a1,
0xa2bfe8a1a2bfe8a1, 0xa2bfe8a1a2bfe8a1, 0xa2bfe8a1a2bfe8a1,
0xa2bfe8a1a2bfe8a1,
0xa81a664ba81a664b, 0xa81a664ba81a664b, 0xa81a664ba81a664b,
0xa81a664ba81a664b,
0xa81a664ba81a664b, 0xa81a664ba81a664b, 0xa81a664ba81a664b,
0xa81a664ba81a664b,
0xc24b8b70c24b8b70, 0xc24b8b70c24b8b70, 0xc24b8b70c24b8b70,
0xc24b8b70c24b8b70,
0xc24b8b70c24b8b70, 0xc24b8b70c24b8b70, 0xc24b8b70c24b8b70,
0xc24b8b70c24b8b70,
0xc76c51a3c76c51a3, 0xc76c51a3c76c51a3, 0xc76c51a3c76c51a3,
0xc76c51a3c76c51a3,
0xc76c51a3c76c51a3, 0xc76c51a3c76c51a3, 0xc76c51a3c76c51a3,
0xc76c51a3c76c51a3,
0xd192e819d192e819, 0xd192e819d192e819, 0xd192e819d192e819,
0xd192e819d192e819,
0xd192e819d192e819, 0xd192e819d192e819, 0xd192e819d192e819,
0xd192e819d192e819,
0xd6990624d6990624, 0xd6990624d6990624, 0xd6990624d6990624,
0xd6990624d6990624,
0xd6990624d6990624, 0xd6990624d6990624, 0xd6990624d6990624,
0xd6990624d6990624,
0xf40e3585f40e3585, 0xf40e3585f40e3585, 0xf40e3585f40e3585,
0xf40e3585f40e3585,
0xf40e3585f40e3585, 0xf40e3585f40e3585, 0xf40e3585f40e3585,
0xf40e3585f40e3585,
0x106aa070106aa070, 0x106aa070106aa070, 0x106aa070106aa070,
0x106aa070106aa070,
0x106aa070106aa070, 0x106aa070106aa070, 0x106aa070106aa070,
0x106aa070106aa070,
0x19a4c11619a4c116, 0x19a4c11619a4c116, 0x19a4c11619a4c116,
0x19a4c11619a4c116,
0x19a4c11619a4c116, 0x19a4c11619a4c116, 0x19a4c11619a4c116,
0x19a4c11619a4c116,
0x1e376c081e376c08, 0x1e376c081e376c08, 0x1e376c081e376c08,
0x1e376c081e376c08,
0x1e376c081e376c08, 0x1e376c081e376c08, 0x1e376c081e376c08,
0x1e376c081e376c08,
0x2748774c2748774c, 0x2748774c2748774c, 0x2748774c2748774c,
0x2748774c2748774c,
0x2748774c2748774c, 0x2748774c2748774c, 0x2748774c2748774c,
0x2748774c2748774c,
0x34b0bcb534b0bcb5, 0x34b0bcb534b0bcb5, 0x34b0bcb534b0bcb5,
0x34b0bcb534b0bcb5,
0x34b0bcb534b0bcb5, 0x34b0bcb534b0bcb5, 0x34b0bcb534b0bcb5,
0x34b0bcb534b0bcb5,
0x391c0cb3391c0cb3, 0x391c0cb3391c0cb3, 0x391c0cb3391c0cb3,
0x391c0cb3391c0cb3,
0x391c0cb3391c0cb3, 0x391c0cb3391c0cb3, 0x391c0cb3391c0cb3,
0x391c0cb3391c0cb3,
0x4ed8aa4a4ed8aa4a, 0x4ed8aa4a4ed8aa4a, 0x4ed8aa4a4ed8aa4a,
0x4ed8aa4a4ed8aa4a,
0x4ed8aa4a4ed8aa4a, 0x4ed8aa4a4ed8aa4a, 0x4ed8aa4a4ed8aa4a,
0x4ed8aa4a4ed8aa4a,
0x5b9cca4f5b9cca4f, 0x5b9cca4f5b9cca4f, 0x5b9cca4f5b9cca4f,
0x5b9cca4f5b9cca4f,
0x5b9cca4f5b9cca4f, 0x5b9cca4f5b9cca4f, 0x5b9cca4f5b9cca4f,
0x5b9cca4f5b9cca4f,
0x682e6ff3682e6ff3, 0x682e6ff3682e6ff3, 0x682e6ff3682e6ff3,
0x682e6ff3682e6ff3,
0x682e6ff3682e6ff3, 0x682e6ff3682e6ff3, 0x682e6ff3682e6ff3,
0x682e6ff3682e6ff3,
0x748f82ee748f82ee, 0x748f82ee748f82ee, 0x748f82ee748f82ee,
0x748f82ee748f82ee,
0x748f82ee748f82ee, 0x748f82ee748f82ee, 0x748f82ee748f82ee,
0x748f82ee748f82ee,
0x78a5636f78a5636f, 0x78a5636f78a5636f, 0x78a5636f78a5636f,
0x78a5636f78a5636f,
0x78a5636f78a5636f, 0x78a5636f78a5636f, 0x78a5636f78a5636f,
0x78a5636f78a5636f,
0x84c8781484c87814, 0x84c8781484c87814, 0x84c8781484c87814,
0x84c8781484c87814,
0x84c8781484c87814, 0x84c8781484c87814, 0x84c8781484c87814,
0x84c8781484c87814,
0x8cc702088cc70208, 0x8cc702088cc70208, 0x8cc702088cc70208,
0x8cc702088cc70208,
0x8cc702088cc70208, 0x8cc702088cc70208, 0x8cc702088cc70208,
0x8cc702088cc70208,
0x90befffa90befffa, 0x90befffa90befffa, 0x90befffa90befffa,
0x90befffa90befffa,
0x90befffa90befffa, 0x90befffa90befffa, 0x90befffa90befffa,
0x90befffa90befffa,
0xa4506ceba4506ceb, 0xa4506ceba4506ceb, 0xa4506ceba4506ceb,
0xa4506ceba4506ceb,
0xa4506ceba4506ceb, 0xa4506ceba4506ceb, 0xa4506ceba4506ceb,
0xa4506ceba4506ceb,
0xbef9a3f7bef9a3f7, 0xbef9a3f7bef9a3f7, 0xbef9a3f7bef9a3f7,
0xbef9a3f7bef9a3f7,
0xbef9a3f7bef9a3f7, 0xbef9a3f7bef9a3f7, 0xbef9a3f7bef9a3f7,
0xbef9a3f7bef9a3f7,
0xc67178f2c67178f2, 0xc67178f2c67178f2, 0xc67178f2c67178f2,
0xc67178f2c67178f2,
0xc67178f2c67178f2, 0xc67178f2c67178f2, 0xc67178f2c67178f2,
0xc67178f2c67178f2,
}
// Interface function to assembly ode
func blockAvx512(
digests *[512]byte, input [16][]byte, mask []uint64,
) [16][Size]byte {
scratch := [512]byte{}
sha256X16Avx512(digests, &scratch, &table, mask, input)
output := [16][Size]byte{}
for i := 0; i < 16; i++ {
output[i] = getDigest(i, digests[:])
}
return output
}
func getDigest(index int, state []byte) (sum [Size]byte) {
for j := 0; j < 16; j += 2 {
for i := index*4 + j*Size; i < index*4+(j+1)*Size; i += Size {
binary.BigEndian.PutUint32(
sum[j*2:], binary.LittleEndian.Uint32(state[i:i+4]),
)
}
}
return
}
// Message to send across input channel
type blockInput struct {
uid uint64
msg []byte
reset bool
final bool
sumCh chan [Size]byte
}
// Avx512Server - Type to implement 16x parallel handling of SHA256 invocations
type Avx512Server struct {
blocksCh chan blockInput // Input channel
totalIn int // Total number of inputs waiting to be processed
lanes [16]Avx512LaneInfo // Array with info per lane (out of 16)
digests map[uint64][Size]byte // Map of uids to (interim) digest results
}
// Avx512LaneInfo - Info for each lane
type Avx512LaneInfo struct {
uid uint64 // unique identification for this SHA processing
block []byte // input block to be processed
outputCh chan [Size]byte // channel for output result
}
// NewAvx512Server - Create new object for parallel processing handling
func NewAvx512Server() *Avx512Server {
a512srv := &Avx512Server{}
a512srv.digests = make(map[uint64][Size]byte)
a512srv.blocksCh = make(chan blockInput)
// Start a single thread for reading from the input channel
go a512srv.Process()
return a512srv
}
// Process - Sole handler for reading from the input channel
func (a512srv *Avx512Server) Process() {
for {
select {
case block := <-a512srv.blocksCh:
if block.reset {
a512srv.reset(block.uid)
continue
}
index := block.uid & 0xf
// fmt.Println("Adding message:", block.uid, index)
if a512srv.lanes[index].block != nil { // If slot is already filled, process all inputs
// fmt.Println("Invoking Blocks()")
a512srv.blocks()
}
a512srv.totalIn++
a512srv.lanes[index] = Avx512LaneInfo{
uid: block.uid, block: block.msg,
}
if block.final {
a512srv.lanes[index].outputCh = block.sumCh
}
if a512srv.totalIn == len(a512srv.lanes) {
// fmt.Println("Invoking Blocks() while FULL: ")
a512srv.blocks()
}
// TODO: test with larger timeout
case <-time.After(1 * time.Microsecond):
for _, lane := range a512srv.lanes {
if lane.block != nil { // check if there is any input to process
// fmt.Println("Invoking Blocks() on TIMEOUT: ")
a512srv.blocks()
break // we are done
}
}
}
}
}
// Do a reset for this calculation
func (a512srv *Avx512Server) reset(uid uint64) {
// Check if there is a message still waiting to be processed (and remove if so)
for i, lane := range a512srv.lanes {
if lane.uid == uid {
if lane.block != nil {
a512srv.lanes[i] = Avx512LaneInfo{} // clear message
a512srv.totalIn--
}
}
}
// Delete entry from hash map
delete(a512srv.digests, uid)
}
// Invoke assembly and send results back
func (a512srv *Avx512Server) blocks() {
inputs := [16][]byte{}
for i := range inputs {
inputs[i] = a512srv.lanes[i].block
}
mask := expandMask(genMask(inputs))
outputs := blockAvx512(a512srv.getDigests(), inputs, mask)
a512srv.totalIn = 0
for i := 0; i < len(outputs); i++ {
uid, outputCh := a512srv.lanes[i].uid, a512srv.lanes[i].outputCh
a512srv.digests[uid] = outputs[i]
a512srv.lanes[i] = Avx512LaneInfo{}
if outputCh != nil {
// Send back result
outputCh <- outputs[i]
delete(a512srv.digests, uid) // Delete entry from hashmap
}
}
}
func (a512srv *Avx512Server) Write(uid uint64, p []byte) (nn int, err error) {
a512srv.blocksCh <- blockInput{uid: uid, msg: p}
return len(p), nil
}
// Sum - return sha256 sum in bytes for a given sum id.
func (a512srv *Avx512Server) Sum(uid uint64, p []byte) [32]byte {
sumCh := make(chan [32]byte)
a512srv.blocksCh <- blockInput{uid: uid, msg: p, final: true, sumCh: sumCh}
return <-sumCh
}
func (a512srv *Avx512Server) getDigests() *[512]byte {
digests := [512]byte{}
for i, lane := range a512srv.lanes {
a, ok := a512srv.digests[lane.uid]
if ok {
binary.BigEndian.PutUint32(
digests[(i+0*16)*4:], binary.LittleEndian.Uint32(a[0:4]),
)
binary.BigEndian.PutUint32(
digests[(i+1*16)*4:], binary.LittleEndian.Uint32(a[4:8]),
)
binary.BigEndian.PutUint32(
digests[(i+2*16)*4:],
binary.LittleEndian.Uint32(a[8:12]),
)
binary.BigEndian.PutUint32(
digests[(i+3*16)*4:],
binary.LittleEndian.Uint32(a[12:16]),
)
binary.BigEndian.PutUint32(
digests[(i+4*16)*4:],
binary.LittleEndian.Uint32(a[16:20]),
)
binary.BigEndian.PutUint32(
digests[(i+5*16)*4:],
binary.LittleEndian.Uint32(a[20:24]),
)
binary.BigEndian.PutUint32(
digests[(i+6*16)*4:],
binary.LittleEndian.Uint32(a[24:28]),
)
binary.BigEndian.PutUint32(
digests[(i+7*16)*4:],
binary.LittleEndian.Uint32(a[28:32]),
)
} else {
binary.LittleEndian.PutUint32(digests[(i+0*16)*4:], init0)
binary.LittleEndian.PutUint32(digests[(i+1*16)*4:], init1)
binary.LittleEndian.PutUint32(digests[(i+2*16)*4:], init2)
binary.LittleEndian.PutUint32(digests[(i+3*16)*4:], init3)
binary.LittleEndian.PutUint32(digests[(i+4*16)*4:], init4)
binary.LittleEndian.PutUint32(digests[(i+5*16)*4:], init5)
binary.LittleEndian.PutUint32(digests[(i+6*16)*4:], init6)
binary.LittleEndian.PutUint32(digests[(i+7*16)*4:], init7)
}
}
return &digests
}
// Helper struct for sorting blocks based on length
type lane struct {
len uint
pos uint
}
type lanes []lane
func (lns lanes) Len() int { return len(lns) }
func (lns lanes) Swap(i, j int) { lns[i], lns[j] = lns[j], lns[i] }
func (lns lanes) Less(i, j int) bool { return lns[i].len < lns[j].len }
// Helper struct for
type maskRounds struct {
mask uint64
rounds uint64
}
func genMask(input [16][]byte) [16]maskRounds {
// Sort on blocks length small to large
var sorted [16]lane
for c, inpt := range input {
sorted[c] = lane{uint(len(inpt)), uint(c)}
}
sort.Sort(lanes(sorted[:]))
// Create mask array including 'rounds' between masks
m, round, index := uint64(0xffff), uint64(0), 0
var mr [16]maskRounds
for _, s := range sorted {
if s.len > 0 {
if uint64(s.len)>>6 > round {
mr[index] = maskRounds{m, (uint64(s.len) >> 6) - round}
index++
}
round = uint64(s.len) >> 6
}
m = m & ^(1 << uint(s.pos))
}
return mr
}
// TODO: remove function
func expandMask(mr [16]maskRounds) []uint64 {
size := uint64(0)
for _, r := range mr {
size += r.rounds
}
result, index := make([]uint64, size), 0
for _, r := range mr {
for j := uint64(0); j < r.rounds; j++ {
result[index] = r.mask
index++
}
}
return result
}

267
pkg/crypto/sha256/sha256blockAvx512_amd64.s
View File

File diff suppressed because one or more lines are too long

545
pkg/crypto/sha256/sha256blockAvx512_amd64_test.go
View File

@@ -1,545 +0,0 @@
//go:build !noasm && !appengine && gc
// +build !noasm,!appengine,gc
/*
* Minio Cloud Storage, (C) 2017 Minio, Inc.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package sha256
import (
"bytes"
"encoding/binary"
"encoding/hex"
"fmt"
"hash"
"reflect"
"sync"
"testing"
)
func TestGoldenAVX512(t *testing.T) {
if !hasAvx512 {
// t.SkipNow()
return
}
server := NewAvx512Server()
h512 := NewAvx512(server)
for _, g := range golden {
h512.Reset()
h512.Write([]byte(g.in))
digest := h512.Sum([]byte{})
s := fmt.Sprintf("%x", digest)
if !reflect.DeepEqual(digest, g.out[:]) {
t.Fatalf(
"Sum256 function: sha256(%s) = %s want %s", g.in, s,
hex.EncodeToString(g.out[:]),
)
}
}
}
func createInputs(size int) [16][]byte {
input := [16][]byte{}
for i := 0; i < 16; i++ {
input[i] = make([]byte, size)
}
return input
}
func initDigests() *[512]byte {
digests := [512]byte{}
for i := 0; i < 16; i++ {
binary.LittleEndian.PutUint32(digests[(i+0*16)*4:], init0)
binary.LittleEndian.PutUint32(digests[(i+1*16)*4:], init1)
binary.LittleEndian.PutUint32(digests[(i+2*16)*4:], init2)
binary.LittleEndian.PutUint32(digests[(i+3*16)*4:], init3)
binary.LittleEndian.PutUint32(digests[(i+4*16)*4:], init4)
binary.LittleEndian.PutUint32(digests[(i+5*16)*4:], init5)
binary.LittleEndian.PutUint32(digests[(i+6*16)*4:], init6)
binary.LittleEndian.PutUint32(digests[(i+7*16)*4:], init7)
}
return &digests
}
func testSha256Avx512(t *testing.T, offset, padding int) [16][]byte {
if !hasAvx512 {
// t.SkipNow()
return [16][]byte{}
}
l := uint(len(golden[offset].in))
extraBlock := uint(0)
if padding == 0 {
extraBlock += 9
} else {
extraBlock += 64
}
input := createInputs(int(l + extraBlock))
for i := 0; i < 16; i++ {
copy(input[i], golden[offset+i].in)
input[i][l] = 0x80
copy(input[i][l+1:], bytes.Repeat([]byte{0}, padding))
// Length in bits.
len := uint64(l)
len <<= 3
for ii := uint(0); ii < 8; ii++ {
input[i][l+1+uint(padding)+ii] = byte(len >> (56 - 8*ii))
}
}
mask := make([]uint64, len(input[0])>>6)
for m := range mask {
mask[m] = 0xffff
}
output := blockAvx512(initDigests(), input, mask)
for i := 0; i < 16; i++ {
if bytes.Compare(output[i][:], golden[offset+i].out[:]) != 0 {
t.Fatalf(
"Sum256 function: sha256(%s) = %s want %s", golden[offset+i].in,
hex.EncodeToString(output[i][:]),
hex.EncodeToString(golden[offset+i].out[:]),
)
}
}
return input
}
func TestAvx512_1Block(t *testing.T) { testSha256Avx512(t, 31, 0) }
func TestAvx512_3Blocks(t *testing.T) { testSha256Avx512(t, 47, 55) }
func TestAvx512_MixedBlocks(t *testing.T) {
if !hasAvx512 {
// t.SkipNow()
return
}
inputSingleBlock := testSha256Avx512(t, 31, 0)
inputMultiBlock := testSha256Avx512(t, 47, 55)
input := [16][]byte{}
for i := range input {
if i%2 == 0 {
input[i] = inputMultiBlock[i]
} else {
input[i] = inputSingleBlock[i]
}
}
mask := [3]uint64{0xffff, 0x5555, 0x5555}
output := blockAvx512(initDigests(), input, mask[:])
var offset int
for i := 0; i < len(output); i++ {
if i%2 == 0 {
offset = 47
} else {
offset = 31
}
if bytes.Compare(output[i][:], golden[offset+i].out[:]) != 0 {
t.Fatalf(
"Sum256 function: sha256(%s) = %s want %s", golden[offset+i].in,
hex.EncodeToString(output[i][:]),
hex.EncodeToString(golden[offset+i].out[:]),
)
}
}
}
func TestAvx512_MixedWithNilBlocks(t *testing.T) {
if !hasAvx512 {
// t.SkipNow()
return
}
inputSingleBlock := testSha256Avx512(t, 31, 0)
inputMultiBlock := testSha256Avx512(t, 47, 55)
input := [16][]byte{}
for i := range input {
if i%3 == 0 {
input[i] = inputMultiBlock[i]
} else if i%3 == 1 {
input[i] = inputSingleBlock[i]
} else {
input[i] = nil
}
}
mask := [3]uint64{0xb6db, 0x9249, 0x9249}
output := blockAvx512(initDigests(), input, mask[:])
var offset int
for i := 0; i < len(output); i++ {
if i%3 == 2 { // for nil inputs
initvec := [32]byte{
0x6a, 0x09, 0xe6, 0x67, 0xbb, 0x67, 0xae, 0x85,
0x3c, 0x6e, 0xf3, 0x72, 0xa5, 0x4f, 0xf5, 0x3a,
0x51, 0x0e, 0x52, 0x7f, 0x9b, 0x05, 0x68, 0x8c,
0x1f, 0x83, 0xd9, 0xab, 0x5b, 0xe0, 0xcd, 0x19,
}
if bytes.Compare(output[i][:], initvec[:]) != 0 {
t.Fatalf(
"Sum256 function: sha256 for nil vector = %s want %s",
hex.EncodeToString(output[i][:]),
hex.EncodeToString(initvec[:]),
)
}
continue
}
if i%3 == 0 {
offset = 47
} else {
offset = 31
}
if bytes.Compare(output[i][:], golden[offset+i].out[:]) != 0 {
t.Fatalf(
"Sum256 function: sha256(%s) = %s want %s", golden[offset+i].in,
hex.EncodeToString(output[i][:]),
hex.EncodeToString(golden[offset+i].out[:]),
)
}
}
}
func TestAvx512Server(t *testing.T) {
if !hasAvx512 {
// t.SkipNow()
return
}
const offset = 31 + 16
server := NewAvx512Server()
// First block of 64 bytes
for i := 0; i < 16; i++ {
input := make([]byte, 64)
copy(input, golden[offset+i].in)
server.Write(uint64(Avx512ServerUID+i), input)
}
// Second block of 64 bytes
for i := 0; i < 16; i++ {
input := make([]byte, 64)
copy(input, golden[offset+i].in[64:])
server.Write(uint64(Avx512ServerUID+i), input)
}
wg := sync.WaitGroup{}
wg.Add(16)
// Third and final block
for i := 0; i < 16; i++ {
input := make([]byte, 64)
input[0] = 0x80
copy(input[1:], bytes.Repeat([]byte{0}, 63-8))
// Length in bits.
len := uint64(128)
len <<= 3
for ii := uint(0); ii < 8; ii++ {
input[63-8+1+ii] = byte(len >> (56 - 8*ii))
}
go func(i int, uid uint64, input []byte) {
output := server.Sum(uid, input)
if bytes.Compare(output[:], golden[offset+i].out[:]) != 0 {
t.Fatalf(
"Sum256 function: sha256(%s) = %s want %s",
golden[offset+i].in,
hex.EncodeToString(output[:]),
hex.EncodeToString(golden[offset+i].out[:]),
)
}
wg.Done()
}(i, uint64(Avx512ServerUID+i), input)
}
wg.Wait()
}
func TestAvx512Digest(t *testing.T) {
if !hasAvx512 {
// t.SkipNow()
return
}
server := NewAvx512Server()
const tests = 16
h512 := [16]hash.Hash{}
for i := 0; i < tests; i++ {
h512[i] = NewAvx512(server)
}
const offset = 31 + 16
for i := 0; i < tests; i++ {
input := make([]byte, 64)
copy(input, golden[offset+i].in)
h512[i].Write(input)
}
for i := 0; i < tests; i++ {
input := make([]byte, 64)
copy(input, golden[offset+i].in[64:])
h512[i].Write(input)
}
for i := 0; i < tests; i++ {
output := h512[i].Sum([]byte{})
if bytes.Compare(output[:], golden[offset+i].out[:]) != 0 {
t.Fatalf(
"Sum256 function: sha256(%s) = %s want %s", golden[offset+i].in,
hex.EncodeToString(output[:]),
hex.EncodeToString(golden[offset+i].out[:]),
)
}
}
}
func benchmarkAvx512SingleCore(h512 []hash.Hash, body []byte) {
for i := 0; i < len(h512); i++ {
h512[i].Write(body)
}
for i := 0; i < len(h512); i++ {
_ = h512[i].Sum([]byte{})
}
}
func benchmarkAvx512(b *testing.B, size int) {
if !hasAvx512 {
b.SkipNow()
return
}
server := NewAvx512Server()
const tests = 16
body := make([]byte, size)
b.SetBytes(int64(len(body) * tests))
b.ResetTimer()
for i := 0; i < b.N; i++ {
h512 := make([]hash.Hash, tests)
for i := 0; i < tests; i++ {
h512[i] = NewAvx512(server)
}
benchmarkAvx512SingleCore(h512, body)
}
}
func BenchmarkAvx512_05M(b *testing.B) { benchmarkAvx512(b, 512*1024) }
func BenchmarkAvx512_1M(b *testing.B) { benchmarkAvx512(b, 1*1024*1024) }
func BenchmarkAvx512_5M(b *testing.B) { benchmarkAvx512(b, 5*1024*1024) }
func BenchmarkAvx512_10M(b *testing.B) { benchmarkAvx512(b, 10*1024*1024) }
func benchmarkAvx512MultiCore(b *testing.B, size, cores int) {
if !hasAvx512 {
b.SkipNow()
return
}
servers := make([]*Avx512Server, cores)
for c := 0; c < cores; c++ {
servers[c] = NewAvx512Server()
}
const tests = 16
body := make([]byte, size)
h512 := make([]hash.Hash, tests*cores)
for i := 0; i < tests*cores; i++ {
h512[i] = NewAvx512(servers[i>>4])
}
b.SetBytes(int64(size * 16 * cores))
b.ResetTimer()
var wg sync.WaitGroup
for i := 0; i < b.N; i++ {
wg.Add(cores)
for c := 0; c < cores; c++ {
go func(c int) {
benchmarkAvx512SingleCore(
h512[c*tests:(c+1)*tests],
body,
)
wg.Done()
}(c)
}
wg.Wait()
}
}
func BenchmarkAvx512_5M_2Cores(b *testing.B) {
benchmarkAvx512MultiCore(
b, 5*1024*1024, 2,
)
}
func BenchmarkAvx512_5M_4Cores(b *testing.B) {
benchmarkAvx512MultiCore(
b, 5*1024*1024, 4,
)
}
func BenchmarkAvx512_5M_6Cores(b *testing.B) {
benchmarkAvx512MultiCore(
b, 5*1024*1024, 6,
)
}
type maskTest struct {
in [16]int
out [16]maskRounds
}
var goldenMask = []maskTest{
{[16]int{0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0}, [16]maskRounds{}},
{
[16]int{64, 0, 64, 0, 64, 0, 64, 0, 64, 0, 64, 0, 64, 0, 64, 0},
[16]maskRounds{{0x5555, 1}},
},
{
[16]int{0, 64, 0, 64, 0, 64, 0, 64, 0, 64, 0, 64, 0, 64, 0, 64},
[16]maskRounds{{0xaaaa, 1}},
},
{
[16]int{64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64, 64},
[16]maskRounds{{0xffff, 1}},
},
{
[16]int{
128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128, 128,
128, 128, 128,
},
[16]maskRounds{{0xffff, 2}},
},
{
[16]int{
64, 128, 64, 128, 64, 128, 64, 128, 64, 128, 64, 128, 64, 128, 64,
128,
},
[16]maskRounds{{0xffff, 1}, {0xaaaa, 1}},
},
{
[16]int{
128, 64, 128, 64, 128, 64, 128, 64, 128, 64, 128, 64, 128, 64, 128,
64,
},
[16]maskRounds{{0xffff, 1}, {0x5555, 1}},
},
{
[16]int{
64, 192, 64, 192, 64, 192, 64, 192, 64, 192, 64, 192, 64, 192, 64,
192,
},
[16]maskRounds{{0xffff, 1}, {0xaaaa, 2}},
},
//
// >= 64 0110=6 1011=b 1101=d 0110=6
// >=128 0100=4 0010=2 1001=9 0100=4
{
[16]int{0, 64, 128, 0, 64, 128, 0, 64, 128, 0, 64, 128, 0, 64, 128, 0},
[16]maskRounds{{0x6db6, 1}, {0x4924, 1}},
},
{
[16]int{
1 * 64, 2 * 64, 3 * 64, 4 * 64, 5 * 64, 6 * 64, 7 * 64, 8 * 64,
9 * 64, 10 * 64,
11 * 64, 12 * 64, 13 * 64, 14 * 64, 15 * 64, 16 * 64,
},
[16]maskRounds{
{0xffff, 1}, {0xfffe, 1}, {0xfffc, 1}, {0xfff8, 1}, {0xfff0, 1},
{0xffe0, 1}, {0xffc0, 1}, {0xff80, 1},
{0xff00, 1}, {0xfe00, 1}, {0xfc00, 1}, {0xf800, 1}, {0xf000, 1},
{0xe000, 1},
{0xc000, 1}, {0x8000, 1},
},
},
{
[16]int{
2 * 64, 1 * 64, 3 * 64, 4 * 64, 5 * 64, 6 * 64, 7 * 64, 8 * 64,
9 * 64, 10 * 64,
11 * 64, 12 * 64, 13 * 64, 14 * 64, 15 * 64, 16 * 64,
},
[16]maskRounds{
{0xffff, 1}, {0xfffd, 1}, {0xfffc, 1}, {0xfff8, 1}, {0xfff0, 1},
{0xffe0, 1}, {0xffc0, 1}, {0xff80, 1},
{0xff00, 1}, {0xfe00, 1}, {0xfc00, 1}, {0xf800, 1}, {0xf000, 1},
{0xe000, 1},
{0xc000, 1}, {0x8000, 1},
},
},
{
[16]int{
10 * 64, 20 * 64, 30 * 64, 40 * 64, 50 * 64, 60 * 64, 70 * 64,
80 * 64, 90 * 64,
100 * 64, 110 * 64, 120 * 64, 130 * 64, 140 * 64, 150 * 64,
160 * 64,
},
[16]maskRounds{
{0xffff, 10}, {0xfffe, 10}, {0xfffc, 10}, {0xfff8, 10},
{0xfff0, 10},
{0xffe0, 10}, {0xffc0, 10}, {0xff80, 10},
{0xff00, 10}, {0xfe00, 10}, {0xfc00, 10}, {0xf800, 10},
{0xf000, 10}, {0xe000, 10},
{0xc000, 10}, {0x8000, 10},
},
},
{
[16]int{
10 * 64, 19 * 64, 27 * 64, 34 * 64, 40 * 64, 45 * 64, 49 * 64,
52 * 64, 54 * 64,
55 * 64, 57 * 64, 60 * 64, 64 * 64, 69 * 64, 75 * 64, 82 * 64,
},
[16]maskRounds{
{0xffff, 10}, {0xfffe, 9}, {0xfffc, 8}, {0xfff8, 7}, {0xfff0, 6},
{0xffe0, 5}, {0xffc0, 4}, {0xff80, 3},
{0xff00, 2}, {0xfe00, 1}, {0xfc00, 2}, {0xf800, 3}, {0xf000, 4},
{0xe000, 5},
{0xc000, 6}, {0x8000, 7},
},
},
}
func TestMaskGen(t *testing.T) {
input := [16][]byte{}
for gcase, g := range goldenMask {
for i, l := range g.in {
buf := make([]byte, l)
input[i] = buf[:]
}
mr := genMask(input)
if !reflect.DeepEqual(mr, g.out) {
t.Fatalf(
"case %d: got %04x\n want %04x", gcase, mr,
g.out,
)
}
}
}

31
pkg/crypto/sha256/sha256block_amd64.go
View File

@@ -1,31 +0,0 @@
//go:build !noasm && !appengine && gc
// +build !noasm,!appengine,gc
/*
* Minio Cloud Storage, (C) 2016 Minio, Inc.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package sha256
func blockArmSha2Go(dig *digest, p []byte) {
panic("blockArmSha2Go called unexpectedly")
}
//go:noescape
func blockIntelSha(h *[8]uint32, message []uint8)
func blockIntelShaGo(dig *digest, p []byte) {
blockIntelSha(&dig.h, p)
}

266
pkg/crypto/sha256/sha256block_amd64.s
View File

@@ -1,266 +0,0 @@
//+build !noasm,!appengine,gc
// SHA intrinsic version of SHA256
// Kristofer Peterson, (C) 2018.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
//
#include "textflag.h"
DATA K<>+0x00(SB)/4, $0x428a2f98
DATA K<>+0x04(SB)/4, $0x71374491
DATA K<>+0x08(SB)/4, $0xb5c0fbcf
DATA K<>+0x0c(SB)/4, $0xe9b5dba5
DATA K<>+0x10(SB)/4, $0x3956c25b
DATA K<>+0x14(SB)/4, $0x59f111f1
DATA K<>+0x18(SB)/4, $0x923f82a4
DATA K<>+0x1c(SB)/4, $0xab1c5ed5
DATA K<>+0x20(SB)/4, $0xd807aa98
DATA K<>+0x24(SB)/4, $0x12835b01
DATA K<>+0x28(SB)/4, $0x243185be
DATA K<>+0x2c(SB)/4, $0x550c7dc3
DATA K<>+0x30(SB)/4, $0x72be5d74
DATA K<>+0x34(SB)/4, $0x80deb1fe
DATA K<>+0x38(SB)/4, $0x9bdc06a7
DATA K<>+0x3c(SB)/4, $0xc19bf174
DATA K<>+0x40(SB)/4, $0xe49b69c1
DATA K<>+0x44(SB)/4, $0xefbe4786
DATA K<>+0x48(SB)/4, $0x0fc19dc6
DATA K<>+0x4c(SB)/4, $0x240ca1cc
DATA K<>+0x50(SB)/4, $0x2de92c6f
DATA K<>+0x54(SB)/4, $0x4a7484aa
DATA K<>+0x58(SB)/4, $0x5cb0a9dc
DATA K<>+0x5c(SB)/4, $0x76f988da
DATA K<>+0x60(SB)/4, $0x983e5152
DATA K<>+0x64(SB)/4, $0xa831c66d
DATA K<>+0x68(SB)/4, $0xb00327c8
DATA K<>+0x6c(SB)/4, $0xbf597fc7
DATA K<>+0x70(SB)/4, $0xc6e00bf3
DATA K<>+0x74(SB)/4, $0xd5a79147
DATA K<>+0x78(SB)/4, $0x06ca6351
DATA K<>+0x7c(SB)/4, $0x14292967
DATA K<>+0x80(SB)/4, $0x27b70a85
DATA K<>+0x84(SB)/4, $0x2e1b2138
DATA K<>+0x88(SB)/4, $0x4d2c6dfc
DATA K<>+0x8c(SB)/4, $0x53380d13
DATA K<>+0x90(SB)/4, $0x650a7354
DATA K<>+0x94(SB)/4, $0x766a0abb
DATA K<>+0x98(SB)/4, $0x81c2c92e
DATA K<>+0x9c(SB)/4, $0x92722c85
DATA K<>+0xa0(SB)/4, $0xa2bfe8a1
DATA K<>+0xa4(SB)/4, $0xa81a664b
DATA K<>+0xa8(SB)/4, $0xc24b8b70
DATA K<>+0xac(SB)/4, $0xc76c51a3
DATA K<>+0xb0(SB)/4, $0xd192e819
DATA K<>+0xb4(SB)/4, $0xd6990624
DATA K<>+0xb8(SB)/4, $0xf40e3585
DATA K<>+0xbc(SB)/4, $0x106aa070
DATA K<>+0xc0(SB)/4, $0x19a4c116
DATA K<>+0xc4(SB)/4, $0x1e376c08
DATA K<>+0xc8(SB)/4, $0x2748774c
DATA K<>+0xcc(SB)/4, $0x34b0bcb5
DATA K<>+0xd0(SB)/4, $0x391c0cb3
DATA K<>+0xd4(SB)/4, $0x4ed8aa4a
DATA K<>+0xd8(SB)/4, $0x5b9cca4f
DATA K<>+0xdc(SB)/4, $0x682e6ff3
DATA K<>+0xe0(SB)/4, $0x748f82ee
DATA K<>+0xe4(SB)/4, $0x78a5636f
DATA K<>+0xe8(SB)/4, $0x84c87814
DATA K<>+0xec(SB)/4, $0x8cc70208
DATA K<>+0xf0(SB)/4, $0x90befffa
DATA K<>+0xf4(SB)/4, $0xa4506ceb
DATA K<>+0xf8(SB)/4, $0xbef9a3f7
DATA K<>+0xfc(SB)/4, $0xc67178f2
GLOBL K<>(SB), RODATA|NOPTR, $256
DATA SHUF_MASK<>+0x00(SB)/8, $0x0405060700010203
DATA SHUF_MASK<>+0x08(SB)/8, $0x0c0d0e0f08090a0b
GLOBL SHUF_MASK<>(SB), RODATA|NOPTR, $16
// Register Usage
// BX base address of constant table (constant)
// DX hash_state (constant)
// SI hash_data.data
// DI hash_data.data + hash_data.length - 64 (constant)
// X0 scratch
// X1 scratch
// X2 working hash state // ABEF
// X3 working hash state // CDGH
// X4 first 16 bytes of block
// X5 second 16 bytes of block
// X6 third 16 bytes of block
// X7 fourth 16 bytes of block
// X12 saved hash state // ABEF
// X13 saved hash state // CDGH
// X15 data shuffle mask (constant)
TEXT ·blockIntelSha(SB), NOSPLIT, $0-32
MOVQ h+0(FP), DX
MOVQ message_base+8(FP), SI
MOVQ message_len+16(FP), DI
LEAQ -64(SI)(DI*1), DI
MOVOU (DX), X2
MOVOU 16(DX), X1
MOVO X2, X3
PUNPCKLLQ X1, X2
PUNPCKHLQ X1, X3
PSHUFD $0x27, X2, X2
PSHUFD $0x27, X3, X3
MOVO SHUF_MASK<>(SB), X15
LEAQ K<>(SB), BX
JMP TEST
LOOP:
MOVO X2, X12
MOVO X3, X13
// load block and shuffle
MOVOU (SI), X4
MOVOU 16(SI), X5
MOVOU 32(SI), X6
MOVOU 48(SI), X7
PSHUFB X15, X4
PSHUFB X15, X5
PSHUFB X15, X6
PSHUFB X15, X7
#define ROUND456 \
PADDL X5, X0 \
LONG $0xdacb380f \ // SHA256RNDS2 XMM3, XMM2
MOVO X5, X1 \
LONG $0x0f3a0f66; WORD $0x04cc \ // PALIGNR XMM1, XMM4, 4
PADDL X1, X6 \
LONG $0xf5cd380f \ // SHA256MSG2 XMM6, XMM5
PSHUFD $0x4e, X0, X0 \
LONG $0xd3cb380f \ // SHA256RNDS2 XMM2, XMM3
LONG $0xe5cc380f // SHA256MSG1 XMM4, XMM5
#define ROUND567 \
PADDL X6, X0 \
LONG $0xdacb380f \ // SHA256RNDS2 XMM3, XMM2
MOVO X6, X1 \
LONG $0x0f3a0f66; WORD $0x04cd \ // PALIGNR XMM1, XMM5, 4
PADDL X1, X7 \
LONG $0xfecd380f \ // SHA256MSG2 XMM7, XMM6
PSHUFD $0x4e, X0, X0 \
LONG $0xd3cb380f \ // SHA256RNDS2 XMM2, XMM3
LONG $0xeecc380f // SHA256MSG1 XMM5, XMM6
#define ROUND674 \
PADDL X7, X0 \
LONG $0xdacb380f \ // SHA256RNDS2 XMM3, XMM2
MOVO X7, X1 \
LONG $0x0f3a0f66; WORD $0x04ce \ // PALIGNR XMM1, XMM6, 4
PADDL X1, X4 \
LONG $0xe7cd380f \ // SHA256MSG2 XMM4, XMM7
PSHUFD $0x4e, X0, X0 \
LONG $0xd3cb380f \ // SHA256RNDS2 XMM2, XMM3
LONG $0xf7cc380f // SHA256MSG1 XMM6, XMM7
#define ROUND745 \
PADDL X4, X0 \
LONG $0xdacb380f \ // SHA256RNDS2 XMM3, XMM2
MOVO X4, X1 \
LONG $0x0f3a0f66; WORD $0x04cf \ // PALIGNR XMM1, XMM7, 4
PADDL X1, X5 \
LONG $0xeccd380f \ // SHA256MSG2 XMM5, XMM4
PSHUFD $0x4e, X0, X0 \
LONG $0xd3cb380f \ // SHA256RNDS2 XMM2, XMM3
LONG $0xfccc380f // SHA256MSG1 XMM7, XMM4
// rounds 0-3
MOVO (BX), X0
PADDL X4, X0
LONG $0xdacb380f // SHA256RNDS2 XMM3, XMM2
PSHUFD $0x4e, X0, X0
LONG $0xd3cb380f // SHA256RNDS2 XMM2, XMM3
// rounds 4-7
MOVO 1*16(BX), X0
PADDL X5, X0
LONG $0xdacb380f // SHA256RNDS2 XMM3, XMM2
PSHUFD $0x4e, X0, X0
LONG $0xd3cb380f // SHA256RNDS2 XMM2, XMM3
LONG $0xe5cc380f // SHA256MSG1 XMM4, XMM5
// rounds 8-11
MOVO 2*16(BX), X0
PADDL X6, X0
LONG $0xdacb380f // SHA256RNDS2 XMM3, XMM2
PSHUFD $0x4e, X0, X0
LONG $0xd3cb380f // SHA256RNDS2 XMM2, XMM3
LONG $0xeecc380f // SHA256MSG1 XMM5, XMM6
MOVO 3*16(BX), X0; ROUND674 // rounds 12-15
MOVO 4*16(BX), X0; ROUND745 // rounds 16-19
MOVO 5*16(BX), X0; ROUND456 // rounds 20-23
MOVO 6*16(BX), X0; ROUND567 // rounds 24-27
MOVO 7*16(BX), X0; ROUND674 // rounds 28-31
MOVO 8*16(BX), X0; ROUND745 // rounds 32-35
MOVO 9*16(BX), X0; ROUND456 // rounds 36-39
MOVO 10*16(BX), X0; ROUND567 // rounds 40-43
MOVO 11*16(BX), X0; ROUND674 // rounds 44-47
MOVO 12*16(BX), X0; ROUND745 // rounds 48-51
// rounds 52-55
MOVO 13*16(BX), X0
PADDL X5, X0
LONG $0xdacb380f // SHA256RNDS2 XMM3, XMM2
MOVO X5, X1
LONG $0x0f3a0f66; WORD $0x04cc // PALIGNR XMM1, XMM4, 4
PADDL X1, X6
LONG $0xf5cd380f // SHA256MSG2 XMM6, XMM5
PSHUFD $0x4e, X0, X0
LONG $0xd3cb380f // SHA256RNDS2 XMM2, XMM3
// rounds 56-59
MOVO 14*16(BX), X0
PADDL X6, X0
LONG $0xdacb380f // SHA256RNDS2 XMM3, XMM2
MOVO X6, X1
LONG $0x0f3a0f66; WORD $0x04cd // PALIGNR XMM1, XMM5, 4
PADDL X1, X7
LONG $0xfecd380f // SHA256MSG2 XMM7, XMM6
PSHUFD $0x4e, X0, X0
LONG $0xd3cb380f // SHA256RNDS2 XMM2, XMM3
// rounds 60-63
MOVO 15*16(BX), X0
PADDL X7, X0
LONG $0xdacb380f // SHA256RNDS2 XMM3, XMM2
PSHUFD $0x4e, X0, X0
LONG $0xd3cb380f // SHA256RNDS2 XMM2, XMM3
PADDL X12, X2
PADDL X13, X3
ADDQ $64, SI
TEST:
CMPQ SI, DI
JBE LOOP
PSHUFD $0x4e, X3, X0
LONG $0x0e3a0f66; WORD $0xf0c2 // PBLENDW XMM0, XMM2, 0xf0
PSHUFD $0x4e, X2, X1
LONG $0x0e3a0f66; WORD $0x0fcb // PBLENDW XMM1, XMM3, 0x0f
PSHUFD $0x1b, X0, X0
PSHUFD $0x1b, X1, X1
MOVOU X0, (DX)
MOVOU X1, 16(DX)
RET

78
pkg/crypto/sha256/sha256block_amd64_test.go
View File

@@ -1,78 +0,0 @@
//go:build !noasm && !appengine && gc
// +build !noasm,!appengine,gc
package sha256
import (
"crypto/sha256"
"encoding/binary"
"testing"
)
func sha256hash(m []byte) (r [32]byte) {
var h [8]uint32
h[0] = 0x6a09e667
h[1] = 0xbb67ae85
h[2] = 0x3c6ef372
h[3] = 0xa54ff53a
h[4] = 0x510e527f
h[5] = 0x9b05688c
h[6] = 0x1f83d9ab
h[7] = 0x5be0cd19
blockIntelSha(&h, m)
l0 := len(m)
l := l0 & (BlockSize - 1)
m = m[l0-l:]
var k [64]byte
copy(k[:], m)
k[l] = 0x80
if l >= 56 {
blockIntelSha(&h, k[:])
binary.LittleEndian.PutUint64(k[0:8], 0)
binary.LittleEndian.PutUint64(k[8:16], 0)
binary.LittleEndian.PutUint64(k[16:24], 0)
binary.LittleEndian.PutUint64(k[24:32], 0)
binary.LittleEndian.PutUint64(k[32:40], 0)
binary.LittleEndian.PutUint64(k[40:48], 0)
binary.LittleEndian.PutUint64(k[48:56], 0)
}
binary.BigEndian.PutUint64(k[56:64], uint64(l0)<<3)
blockIntelSha(&h, k[:])
binary.BigEndian.PutUint32(r[0:4], h[0])
binary.BigEndian.PutUint32(r[4:8], h[1])
binary.BigEndian.PutUint32(r[8:12], h[2])
binary.BigEndian.PutUint32(r[12:16], h[3])
binary.BigEndian.PutUint32(r[16:20], h[4])
binary.BigEndian.PutUint32(r[20:24], h[5])
binary.BigEndian.PutUint32(r[24:28], h[6])
binary.BigEndian.PutUint32(r[28:32], h[7])
return
}
func runTestSha(hashfunc func([]byte) [32]byte) bool {
var m = []byte("This is a message. This is a message. This is a message. This is a message.")
ar := hashfunc(m)
br := sha256.Sum256(m)
return ar == br
}
func TestSha0(t *testing.T) {
if !runTestSha(Sum256) {
t.Errorf("FAILED")
}
}
func TestSha1(t *testing.T) {
if hasIntelSha && !runTestSha(sha256hash) {
t.Errorf("FAILED")
}
}

40
pkg/crypto/sha256/sha256block_arm64.go
View File

@@ -1,40 +0,0 @@
//go:build !noasm && !appengine && gc
// +build !noasm,!appengine,gc
/*
* Minio Cloud Storage, (C) 2016 Minio, Inc.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package sha256
func blockIntelShaGo(dig *digest, p []byte) {
panic("blockIntelShaGo called unexpectedly")
}
//go:noescape
func blockArmSha2(h []uint32, message []uint8)
func blockArmSha2Go(dig *digest, p []byte) {
h := []uint32{
dig.h[0], dig.h[1], dig.h[2], dig.h[3], dig.h[4], dig.h[5], dig.h[6],
dig.h[7],
}
blockArmSha2(h[:], p[:])
dig.h[0], dig.h[1], dig.h[2], dig.h[3], dig.h[4], dig.h[5], dig.h[6], dig.h[7] = h[0], h[1], h[2], h[3], h[4],
h[5], h[6], h[7]
}

192
pkg/crypto/sha256/sha256block_arm64.s
View File

@@ -1,192 +0,0 @@
//+build !noasm,!appengine,gc
// ARM64 version of SHA256
//
// Minio Cloud Storage, (C) 2016 Minio, Inc.
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
//
//
// Based on implementation as found in https://github.com/jocover/sha256-armv8
//
// Use github.com/minio/asm2plan9s on this file to assemble ARM instructions to
// their Plan9 equivalents
//
TEXT ·blockArmSha2(SB), 7, $0
MOVD h+0(FP), R0
MOVD message+24(FP), R1
MOVD message_len+32(FP), R2 // length of message
SUBS $64, R2
BMI complete
// Load constants table pointer
MOVD $·constants(SB), R3
// Cache constants table in registers v16 - v31
WORD $0x4cdf2870 // ld1 {v16.4s-v19.4s}, [x3], #64
WORD $0x4cdf7800 // ld1 {v0.4s}, [x0], #16
WORD $0x4cdf2874 // ld1 {v20.4s-v23.4s}, [x3], #64
WORD $0x4c407801 // ld1 {v1.4s}, [x0]
WORD $0x4cdf2878 // ld1 {v24.4s-v27.4s}, [x3], #64
WORD $0xd1004000 // sub x0, x0, #0x10
WORD $0x4cdf287c // ld1 {v28.4s-v31.4s}, [x3], #64
loop:
// Main loop
WORD $0x4cdf2025 // ld1 {v5.16b-v8.16b}, [x1], #64
WORD $0x4ea01c02 // mov v2.16b, v0.16b
WORD $0x4ea11c23 // mov v3.16b, v1.16b
WORD $0x6e2008a5 // rev32 v5.16b, v5.16b
WORD $0x6e2008c6 // rev32 v6.16b, v6.16b
WORD $0x4eb084a9 // add v9.4s, v5.4s, v16.4s
WORD $0x6e2008e7 // rev32 v7.16b, v7.16b
WORD $0x4eb184ca // add v10.4s, v6.4s, v17.4s
WORD $0x4ea21c44 // mov v4.16b, v2.16b
WORD $0x5e094062 // sha256h q2, q3, v9.4s
WORD $0x5e095083 // sha256h2 q3, q4, v9.4s
WORD $0x5e2828c5 // sha256su0 v5.4s, v6.4s
WORD $0x6e200908 // rev32 v8.16b, v8.16b
WORD $0x4eb284e9 // add v9.4s, v7.4s, v18.4s
WORD $0x4ea21c44 // mov v4.16b, v2.16b
WORD $0x5e0a4062 // sha256h q2, q3, v10.4s
WORD $0x5e0a5083 // sha256h2 q3, q4, v10.4s
WORD $0x5e2828e6 // sha256su0 v6.4s, v7.4s
WORD $0x5e0860e5 // sha256su1 v5.4s, v7.4s, v8.4s
WORD $0x4eb3850a // add v10.4s, v8.4s, v19.4s
WORD $0x4ea21c44 // mov v4.16b, v2.16b
WORD $0x5e094062 // sha256h q2, q3, v9.4s
WORD $0x5e095083 // sha256h2 q3, q4, v9.4s
WORD $0x5e282907 // sha256su0 v7.4s, v8.4s
WORD $0x5e056106 // sha256su1 v6.4s, v8.4s, v5.4s
WORD $0x4eb484a9 // add v9.4s, v5.4s, v20.4s
WORD $0x4ea21c44 // mov v4.16b, v2.16b
WORD $0x5e0a4062 // sha256h q2, q3, v10.4s
WORD $0x5e0a5083 // sha256h2 q3, q4, v10.4s
WORD $0x5e2828a8 // sha256su0 v8.4s, v5.4s
WORD $0x5e0660a7 // sha256su1 v7.4s, v5.4s, v6.4s
WORD $0x4eb584ca // add v10.4s, v6.4s, v21.4s
WORD $0x4ea21c44 // mov v4.16b, v2.16b
WORD $0x5e094062 // sha256h q2, q3, v9.4s
WORD $0x5e095083 // sha256h2 q3, q4, v9.4s
WORD $0x5e2828c5 // sha256su0 v5.4s, v6.4s
WORD $0x5e0760c8 // sha256su1 v8.4s, v6.4s, v7.4s
WORD $0x4eb684e9 // add v9.4s, v7.4s, v22.4s
WORD $0x4ea21c44 // mov v4.16b, v2.16b
WORD $0x5e0a4062 // sha256h q2, q3, v10.4s
WORD $0x5e0a5083 // sha256h2 q3, q4, v10.4s
WORD $0x5e2828e6 // sha256su0 v6.4s, v7.4s
WORD $0x5e0860e5 // sha256su1 v5.4s, v7.4s, v8.4s
WORD $0x4eb7850a // add v10.4s, v8.4s, v23.4s
WORD $0x4ea21c44 // mov v4.16b, v2.16b
WORD $0x5e094062 // sha256h q2, q3, v9.4s
WORD $0x5e095083 // sha256h2 q3, q4, v9.4s
WORD $0x5e282907 // sha256su0 v7.4s, v8.4s
WORD $0x5e056106 // sha256su1 v6.4s, v8.4s, v5.4s
WORD $0x4eb884a9 // add v9.4s, v5.4s, v24.4s
WORD $0x4ea21c44 // mov v4.16b, v2.16b
WORD $0x5e0a4062 // sha256h q2, q3, v10.4s
WORD $0x5e0a5083 // sha256h2 q3, q4, v10.4s
WORD $0x5e2828a8 // sha256su0 v8.4s, v5.4s
WORD $0x5e0660a7 // sha256su1 v7.4s, v5.4s, v6.4s
WORD $0x4eb984ca // add v10.4s, v6.4s, v25.4s
WORD $0x4ea21c44 // mov v4.16b, v2.16b
WORD $0x5e094062 // sha256h q2, q3, v9.4s
WORD $0x5e095083 // sha256h2 q3, q4, v9.4s
WORD $0x5e2828c5 // sha256su0 v5.4s, v6.4s
WORD $0x5e0760c8 // sha256su1 v8.4s, v6.4s, v7.4s
WORD $0x4eba84e9 // add v9.4s, v7.4s, v26.4s
WORD $0x4ea21c44 // mov v4.16b, v2.16b
WORD $0x5e0a4062 // sha256h q2, q3, v10.4s
WORD $0x5e0a5083 // sha256h2 q3, q4, v10.4s
WORD $0x5e2828e6 // sha256su0 v6.4s, v7.4s
WORD $0x5e0860e5 // sha256su1 v5.4s, v7.4s, v8.4s
WORD $0x4ebb850a // add v10.4s, v8.4s, v27.4s
WORD $0x4ea21c44 // mov v4.16b, v2.16b
WORD $0x5e094062 // sha256h q2, q3, v9.4s
WORD $0x5e095083 // sha256h2 q3, q4, v9.4s
WORD $0x5e282907 // sha256su0 v7.4s, v8.4s
WORD $0x5e056106 // sha256su1 v6.4s, v8.4s, v5.4s
WORD $0x4ebc84a9 // add v9.4s, v5.4s, v28.4s
WORD $0x4ea21c44 // mov v4.16b, v2.16b
WORD $0x5e0a4062 // sha256h q2, q3, v10.4s
WORD $0x5e0a5083 // sha256h2 q3, q4, v10.4s
WORD $0x5e2828a8 // sha256su0 v8.4s, v5.4s
WORD $0x5e0660a7 // sha256su1 v7.4s, v5.4s, v6.4s
WORD $0x4ebd84ca // add v10.4s, v6.4s, v29.4s
WORD $0x4ea21c44 // mov v4.16b, v2.16b
WORD $0x5e094062 // sha256h q2, q3, v9.4s
WORD $0x5e095083 // sha256h2 q3, q4, v9.4s
WORD $0x5e0760c8 // sha256su1 v8.4s, v6.4s, v7.4s
WORD $0x4ebe84e9 // add v9.4s, v7.4s, v30.4s
WORD $0x4ea21c44 // mov v4.16b, v2.16b
WORD $0x5e0a4062 // sha256h q2, q3, v10.4s
WORD $0x5e0a5083 // sha256h2 q3, q4, v10.4s
WORD $0x4ebf850a // add v10.4s, v8.4s, v31.4s
WORD $0x4ea21c44 // mov v4.16b, v2.16b
WORD $0x5e094062 // sha256h q2, q3, v9.4s
WORD $0x5e095083 // sha256h2 q3, q4, v9.4s
WORD $0x4ea21c44 // mov v4.16b, v2.16b
WORD $0x5e0a4062 // sha256h q2, q3, v10.4s
WORD $0x5e0a5083 // sha256h2 q3, q4, v10.4s
WORD $0x4ea38421 // add v1.4s, v1.4s, v3.4s
WORD $0x4ea28400 // add v0.4s, v0.4s, v2.4s
SUBS $64, R2
BPL loop
// Store result
WORD $0x4c00a800 // st1 {v0.4s, v1.4s}, [x0]
complete:
RET
// Constants table
DATA ·constants+0x0(SB)/8, $0x71374491428a2f98
DATA ·constants+0x8(SB)/8, $0xe9b5dba5b5c0fbcf
DATA ·constants+0x10(SB)/8, $0x59f111f13956c25b
DATA ·constants+0x18(SB)/8, $0xab1c5ed5923f82a4
DATA ·constants+0x20(SB)/8, $0x12835b01d807aa98
DATA ·constants+0x28(SB)/8, $0x550c7dc3243185be
DATA ·constants+0x30(SB)/8, $0x80deb1fe72be5d74
DATA ·constants+0x38(SB)/8, $0xc19bf1749bdc06a7
DATA ·constants+0x40(SB)/8, $0xefbe4786e49b69c1
DATA ·constants+0x48(SB)/8, $0x240ca1cc0fc19dc6
DATA ·constants+0x50(SB)/8, $0x4a7484aa2de92c6f
DATA ·constants+0x58(SB)/8, $0x76f988da5cb0a9dc
DATA ·constants+0x60(SB)/8, $0xa831c66d983e5152
DATA ·constants+0x68(SB)/8, $0xbf597fc7b00327c8
DATA ·constants+0x70(SB)/8, $0xd5a79147c6e00bf3
DATA ·constants+0x78(SB)/8, $0x1429296706ca6351
DATA ·constants+0x80(SB)/8, $0x2e1b213827b70a85
DATA ·constants+0x88(SB)/8, $0x53380d134d2c6dfc
DATA ·constants+0x90(SB)/8, $0x766a0abb650a7354
DATA ·constants+0x98(SB)/8, $0x92722c8581c2c92e
DATA ·constants+0xa0(SB)/8, $0xa81a664ba2bfe8a1
DATA ·constants+0xa8(SB)/8, $0xc76c51a3c24b8b70
DATA ·constants+0xb0(SB)/8, $0xd6990624d192e819
DATA ·constants+0xb8(SB)/8, $0x106aa070f40e3585
DATA ·constants+0xc0(SB)/8, $0x1e376c0819a4c116
DATA ·constants+0xc8(SB)/8, $0x34b0bcb52748774c
DATA ·constants+0xd0(SB)/8, $0x4ed8aa4a391c0cb3
DATA ·constants+0xd8(SB)/8, $0x682e6ff35b9cca4f
DATA ·constants+0xe0(SB)/8, $0x78a5636f748f82ee
DATA ·constants+0xe8(SB)/8, $0x8cc7020884c87814
DATA ·constants+0xf0(SB)/8, $0xa4506ceb90befffa
DATA ·constants+0xf8(SB)/8, $0xc67178f2bef9a3f7
GLOBL ·constants(SB), 8, $256

29
pkg/crypto/sha256/sha256block_other.go
View File

@@ -1,29 +0,0 @@
//go:build appengine || noasm || (!amd64 && !arm64) || !gc
// +build appengine noasm !amd64,!arm64 !gc
/*
* Minio Cloud Storage, (C) 2019 Minio, Inc.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package sha256
func blockIntelShaGo(dig *digest, p []byte) {
panic("blockIntelShaGo called unexpectedly")
}
func blockArmSha2Go(dig *digest, p []byte) {
panic("blockArmSha2Go called unexpectedly")
}

15
pkg/crypto/sha256/test-architectures.sh
View File

@@ -1,15 +0,0 @@
#!/bin/sh
set -e
go tool dist list | while IFS=/ read os arch; do
echo "Checking $os/$arch..."
echo " normal"
GOARCH=$arch GOOS=$os go build -o /dev/null ./...
echo " noasm"
GOARCH=$arch GOOS=$os go build -tags noasm -o /dev/null ./...
echo " appengine"
GOARCH=$arch GOOS=$os go build -tags appengine -o /dev/null ./...
echo " noasm,appengine"
GOARCH=$arch GOOS=$os go build -tags 'appengine noasm' -o /dev/null ./...
done