diff options
| author | Andy Polyakov <appro@openssl.org> | 2017-11-15 12:25:02 +0100 |
|---|---|---|
| committer | Andy Polyakov <appro@openssl.org> | 2017-11-16 13:57:55 +0100 |
| commit | a78324d95bd4568ce2c3b34bfa1d6f14cddf92ef (patch) | |
| tree | 86648764e99579b852aaaf2bab96da8261e33acc /crypto/bn | |
| parent | 8e4ec5b2e7c9a5d12c348b2a3430c4b51ab041c5 (diff) | |
bn/bn_add.c: address performance regression.
Performance regression was reported for EC key generation between
1.0.2 and 1.1.x [in GH#2891]. It naturally depends on platform,
values between 6 and 9% were observed.
Reviewed-by: Richard Levitte <levitte@openssl.org>
(Merged from https://github.com/openssl/openssl/pull/4743)
Diffstat (limited to 'crypto/bn')
| -rw-r--r-- | crypto/bn/asm/x86_64-gcc.c | 14 | ||||
| -rw-r--r-- | crypto/bn/bn_add.c | 6 |
2 files changed, 13 insertions, 7 deletions
diff --git a/crypto/bn/asm/x86_64-gcc.c b/crypto/bn/asm/x86_64-gcc.c index 7a3b1a12b2..d38f337164 100644 --- a/crypto/bn/asm/x86_64-gcc.c +++ b/crypto/bn/asm/x86_64-gcc.c @@ -225,9 +225,10 @@ BN_ULONG bn_add_words(BN_ULONG *rp, const BN_ULONG *ap, const BN_ULONG *bp, " adcq (%5,%2,8),%0 \n" " movq %0,(%3,%2,8) \n" " lea 1(%2),%2 \n" - " loop 1b \n" - " sbbq %0,%0 \n":"=&r" (ret), "+c"(n), - "+r"(i) + " dec %1 \n" + " jnz 1b \n" + " sbbq %0,%0 \n" + :"=&r" (ret), "+c"(n), "+r"(i) :"r"(rp), "r"(ap), "r"(bp) :"cc", "memory"); @@ -251,9 +252,10 @@ BN_ULONG bn_sub_words(BN_ULONG *rp, const BN_ULONG *ap, const BN_ULONG *bp, " sbbq (%5,%2,8),%0 \n" " movq %0,(%3,%2,8) \n" " lea 1(%2),%2 \n" - " loop 1b \n" - " sbbq %0,%0 \n":"=&r" (ret), "+c"(n), - "+r"(i) + " dec %1 \n" + " jnz 1b \n" + " sbbq %0,%0 \n" + :"=&r" (ret), "+c"(n), "+r"(i) :"r"(rp), "r"(ap), "r"(bp) :"cc", "memory"); diff --git a/crypto/bn/bn_add.c b/crypto/bn/bn_add.c index 6479650afb..7cdefa77a1 100644 --- a/crypto/bn/bn_add.c +++ b/crypto/bn/bn_add.c @@ -141,9 +141,13 @@ int BN_usub(BIGNUM *r, const BIGNUM *a, const BIGNUM *b) borrow &= (t1 == 0); } + while (max && *--rp == 0) + max--; + r->top = max; r->neg = 0; - bn_correct_top(r); + bn_pollute(r); + return 1; } |