This patch removes the arch-common aese/aesmc and aesd/aesimc fusions (i.e. aes fusion) implemented in the scheduling phase through the aarch_crypto_can_dual function. The reason is due to observing undesired behaviour in cases such as:- when register allocation goes bad (e.g. extra movs)- aes operations with xor and zeroed keys among interleaved operations
A more stable version should be provided by instead doing the aes fusion during the combine pass. As such, new combine patterns have been added to enable this.
The second change is the aese and aesd patterns have been rewritten as encapsulating a xor operation. The purpose is to simplify the need of having additional combine patterns for cases like the ones below:
For AESE (though it also applies to AESD as both have a xor operation):
data = data ^ key; data = vaeseq_u8(data, zero);
veor q1, q0, q1 aese.8 q1, q9
Should mean and generate the same as:
data = vaeseq_u8(data, key);
aese.8 q1, q0
2019-07-09 Sylvia Taylor
- config/arm/crypto.md: (crypto_
- config/arm/arm.c (aarch_macro_fusion_pair_p): Remove aes/aesmc fusion check.
- config/arm/aarch-common-protos.h (aarch_crypto_can_dual_issue): Remove.
- config/arm/aarch-common.c (aarch_crypto_can_dual_issue): Likewise.
- config/arm/exynos-m1.md: Remove aese/aesmc fusion.
- config/arm/cortex-a53.md: Likewise.
- config/arm/cortex-a57.md: Likewise.
- config/arm/iterators.md: (CRYPTO_BINARY): Redefine. (CRYPTO_UNARY): Removed. (CRYPTO_AES, CRYPTO_AESMC): New.
- gcc.target/arm/aes-fuse-1.c: New.
- gcc.target/arm/aes-fuse-2.c: New.
- gcc.target/arm/aes_xor_combine.c: New.
c53fd0cf456 [arm]: redefine aes patterns
gcc/ChangeLog | 21 +++++++
gcc/config/arm/aarch-common-protos.h | 1 -
gcc/config/arm/aarch-common.c | 40 -------------
gcc/config/arm/arm.c | 4 --
gcc/config/arm/cortex-a57.md | 6 --
gcc/config/arm/crypto.md | 83 +++++++++++++++++++-------
gcc/config/arm/exynos-m1.md | 5 --
gcc/config/arm/iterators.md | 7 ++-
gcc/testsuite/ChangeLog | 6 ++
gcc/testsuite/gcc.target/arm/aes-fuse-1.c | 66 ++++++++++++++++++++
gcc/testsuite/gcc.target/arm/aes-fuse-2.c | 66 ++++++++++++++++++++
gcc/testsuite/gcc.target/arm/aes_xor_combine.c | 43 +++++++++++++
12 files changed, 269 insertions(+), 79 deletions(-)