Skip to content

8329538: Accelerate P256 on x86_64 using Montgomery intrinsic #18583

New issue

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

Sign up for GitHub

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

Already on GitHub? Sign in to your account

Closed
wants to merge 17 commits into from
+2,253 −316
Closed

8329538: Accelerate P256 on x86_64 using Montgomery intrinsic #18583

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
17 commits
Select commit Hold shift + click to select a range
e36a552
rebase
vpaprotsk Sep 26, 2023
6285a33
conditionalAssign intrinsic
vpaprotsk Oct 10, 2023
c431528
Address comments from Anas and thorough cleanup
vpaprotsk Feb 19, 2024
bc8d611
Fix rebase typo
vpaprotsk Mar 26, 2024
7bf06b7
Remeasure performance
vpaprotsk Mar 29, 2024
dbe6cd3
remove trailing whitespace
vpaprotsk Apr 2, 2024
82b6dae
remove use of jdk.crypto.ec
vpaprotsk Apr 2, 2024
6f9ac04
Comments from Jatin and Tony
vpaprotsk Apr 15, 2024
c93a71f
Comments from Tony and Jatin
vpaprotsk Apr 24, 2024
a198450
whitespace
vpaprotsk Apr 25, 2024
8ff243a
Use AffinePoint to exit Montgomery domain
vpaprotsk May 6, 2024
1ecfdc4
whitespace
vpaprotsk May 9, 2024
83b2131
add message back
vpaprotsk May 10, 2024
8cd095d
whitespace
vpaprotsk May 10, 2024
5c360e3
comments from Sandhya
vpaprotsk May 17, 2024
df4fe6f
shenandoah verifier
vpaprotsk May 17, 2024
b1a3300
Merge remote-tracking branch 'origin/master' into ecc-montgomery
vpaprotsk May 21, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion src/hotspot/cpu/x86/stubGenerator_x86_64_poly_mont.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -167,7 +167,7 @@ void montgomeryMultiply(const Register aLimbs, const Register bLimbs, const Regi
// M = load(*modulus_p256)
__ evmovdquq(modulus, allLimbs, ExternalAddress(modulus_p256()), false, Assembler::AVX_512bit, rscratch);

// A = load(*aLimbs) // masked evmovdquq() can be slow. Instead load full 256bit, and compbine with 64bit
// A = load(*aLimbs); masked evmovdquq() can be slow. Instead load full 256bit, and compbine with 64bit
__ evmovdquq(A, Address(aLimbs, 8), Assembler::AVX_256bit);
__ evpermq(A, allLimbs, shift1L, A, false, Assembler::AVX_512bit);
__ movq(T, Address(aLimbs, 0));
Expand Down
6 changes: 6 additions & 0 deletions src/hotspot/share/gc/shenandoah/c2/shenandoahSupport.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -463,6 +463,12 @@ void ShenandoahBarrierC2Support::verify(RootNode* root) {
"decodeBlock",
{ { TypeFunc::Parms, ShenandoahLoad }, { TypeFunc::Parms+3, ShenandoahStore }, { -1, ShenandoahNone },
{ -1, ShenandoahNone}, { -1, ShenandoahNone}, { -1, ShenandoahNone} },
"intpoly_montgomeryMult_P256",
{ { TypeFunc::Parms, ShenandoahLoad }, { TypeFunc::Parms+1, ShenandoahLoad }, { TypeFunc::Parms+2, ShenandoahStore },
{ -1, ShenandoahNone}, { -1, ShenandoahNone}, { -1, ShenandoahNone} },
"intpoly_assign",
{ { TypeFunc::Parms+1, ShenandoahStore }, { TypeFunc::Parms+2, ShenandoahLoad }, { -1, ShenandoahNone },
{ -1, ShenandoahNone}, { -1, ShenandoahNone}, { -1, ShenandoahNone} },
};

if (call->is_call_to_arraycopystub()) {
Expand Down