[PATCH] crypto: x86/poly1305-intel: Use TEST %reg,%reg instead of CMP $0,%reg

CMP $0,%reg can't set overflow flag, so we can use shorter TEST %reg,%reg instruction when only zero and sign flags are checked (E,L,LE,G,GE conditions). Signed-off-by: Uros Bizjak Cc: Herbert Xu Cc: Borislav Petkov Cc: "H. Peter Anvin" --- arch/x86/crypto/poly1305-x86_64-cr

[PATCH] crypto: x86/sha512-intel: Use TEST %reg,%reg instead of CMP $0,%reg

CMP $0,%reg can't set overflow flag, so we can use shorter TEST %reg,%reg instruction when only zero and sign flags are checked (E,L,LE,G,GE conditions). Signed-off-by: Uros Bizjak Cc: Herbert Xu Cc: Borislav Petkov Cc: "H. Peter Anvin" --- arch/x86/crypto/sha512-avx-asm.S |

[PATCH] crypto: x86/aesni-intel: Use TEST %reg,%reg instead of CMP $0,%reg

CMP $0,%reg can't set overflow flag, so we can use shorter TEST %reg,%reg instruction when only zero and sign flags are checked (E,L,LE,G,GE conditions). Signed-off-by: Uros Bizjak Cc: Herbert Xu Cc: Borislav Petkov Cc: "H. Peter Anvin" --- arch/x86/crypto/aesni-intel_as

Re: [PATCH] crypto/x86: Use XORL r32,32 in curve25519-x86_64.c

On Wed, Sep 2, 2020 at 11:17 AM wrote: > > On Wed, Sep 02, 2020 at 07:50:36AM +0200, Uros Bizjak wrote: > > On Tue, Sep 1, 2020 at 9:12 PM Jason A. Donenfeld wrote: > > > > > > On Tue, Sep 1, 2020 at 8:13 PM Jason A. Donenfeld wrote: > > > > oper

Re: [PATCH] crypto/x86: Use XORL r32,32 in poly1305-x86_64-cryptogams.pl

imization is applied to legacy registers. Uros. > Jason > > On Thu, Aug 27, 2020 at 07:38:31PM +0200, Uros Bizjak wrote: > > x86_64 zero extends 32bit operations, so for 64bit operands, > > XORL r32,r32 is functionally equal to XORQ r64,r64, but avoids > > a REX prefix b

Re: [PATCH] crypto/x86: Use XORL r32,32 in curve25519-x86_64.c

On Tue, Sep 1, 2020 at 9:12 PM Jason A. Donenfeld wrote: > > On Tue, Sep 1, 2020 at 8:13 PM Jason A. Donenfeld wrote: > > operands are the same. Also, have you seen any measurable differences > > when benching this? I can stick it into kbench9000 to see if you > > haven't looked yet. > > On a Sky

[PATCH] crypto/x86: Use XORL r32,32 in poly1305-x86_64-cryptogams.pl

x86_64 zero extends 32bit operations, so for 64bit operands, XORL r32,r32 is functionally equal to XORQ r64,r64, but avoids a REX prefix byte when legacy registers are used. Signed-off-by: Uros Bizjak Cc: Herbert Xu Cc: "David S. Miller" --- arch/x86/crypto/poly1305-x86_64-cryptog

[PATCH] crypto/x86: Use XORL r32,32 in curve25519-x86_64.c

x86_64 zero extends 32bit operations, so for 64bit operands, XORL r32,r32 is functionally equal to XORL r64,r64, but avoids a REX prefix byte when legacy registers are used. Signed-off-by: Uros Bizjak Cc: Herbert Xu Cc: "David S. Miller" --- arch/x86/crypto/curve25519-x86

[PATCH] crypto/x86: Use CRC32 mnemonic in crc32c-intel_glue.c

jne13b As the compiler has some more freedom w.r.t. register allocation, there is also a couple of reg-reg moves removed. There are no hidden states for CRC32 insn, so there is no need to mark assembly as volatile. Signed-off-by: Uros Bizjak CC: Herbert Xu CC: "David S. Miller

[PATCH v2] crypto/x86: Use CRC32 mnemonic in crc32c-intel_glue.c

jne13b As the compiler has some more freedom w.r.t. register allocation, there is also a couple of reg-reg moves removed. There are no hidden states for CRC32 insn, so there is no need to mark assembly as volatile. v2: Introduce CRC32_INST define. Signed-off-by: Uros Bizjak CC

[PATCH] crypto: x86 - Put back integer parts of include/asm/inst.h

Resolves conflict with the tip tree. CC: Herbert Xu CC: Thomas Gleixner CC: Ingo Molnar CC: Borislav Petkov CC: "H. Peter Anvin" CC: Stephen Rothwell , CC: "Chang S. Bae" , CC: Peter Zijlstra , CC: Sasha Levin Signed-off-by: Uros Bizjak --- arch/x86/incl

Re: linux-next: manual merge of the tip tree with the crypto tree

On Mon, Jul 20, 2020 at 6:03 AM Stephen Rothwell wrote: > > Please find attached the incremental patch that puts back integer > > parts of inst.h. This resolves the conflict with the tip tree. > > The tip tree change needs the XMM parts kept as well, sorry. Strange, because I did test my patch w

Re: linux-next: manual merge of the tip tree with the crypto tree

Please find attached the incremental patch that puts back integer parts of inst.h. This resolves the conflict with the tip tree. Uros. On Fri, Jul 17, 2020 at 8:45 AM Herbert Xu wrote: > > On Fri, Jul 17, 2020 at 08:27:27AM +0200, Uros Bizjak wrote: > > > > I will prepar

Re: linux-next: manual merge of the tip tree with the crypto tree

On Fri, Jul 17, 2020 at 6:47 AM Stephen Rothwell wrote: > > Hi all, > > Today's linux-next merge of the tip tree got a conflict in: > > arch/x86/include/asm/inst.h > > between commit: > > d7866e503bdc ("crypto: x86 - Remove include/asm/inst.h") > > from the crypto tree and commit: > > eaad98

[PATCH] crypto/x86: Remove include/asm/inst.h

from x86/crypto directory, and remove now unneeded file. The patch was tested by calculating and comparing sha256sum hashes of stripped object files before and after the patch, to be sure that executable code didn't change. Signed-off-by: Uros Bizjak CC: Herbert Xu CC: "David S. M