On Thu, Jan 04, 2024 at 04:47:16PM +0800, Jerry Shih wrote: > On Jan 3, 2024, at 22:50, Eric Biggers <ebiggers@xxxxxxxxxx> wrote: > > On Tue, Jan 02, 2024 at 12:47:33AM -0600, Eric Biggers wrote: > >> diff --git a/arch/riscv/crypto/Makefile b/arch/riscv/crypto/Makefile > >> index dca698c5cba3e..5dd91f34f0d52 100644 > >> --- a/arch/riscv/crypto/Makefile > >> +++ b/arch/riscv/crypto/Makefile > >> @@ -1,7 +1,10 @@ > >> # SPDX-License-Identifier: GPL-2.0-only > >> # > >> # linux/arch/riscv/crypto/Makefile > >> # > >> > >> obj-$(CONFIG_CRYPTO_AES_RISCV64) += aes-riscv64.o > >> aes-riscv64-y := aes-riscv64-glue.o aes-riscv64-zvkned.o > >> + > >> +obj-$(CONFIG_CRYPTO_AES_BLOCK_RISCV64) += aes-block-riscv64.o > >> +aes-block-riscv64-y := aes-riscv64-block-mode-glue.o aes-riscv64-zvkned-zvbb-zvkg.o aes-riscv64-zvkned-zvkb.o > > > > A bug I noticed (which is also present in Jerry's patchset) is that some of the > > code of the aes-block-riscv64 module is located in aes-riscv64-zvkned.S, which > > isn't built into that module but rather into aes-riscv64. This causes a build > > error when both CONFIG_CRYPTO_AES_RISCV64 and CONFIG_CRYPTO_AES_BLOCK_RISCV64 > > are set to 'm': > > > > ERROR: modpost: "aes_cbc_decrypt_zvkned" [arch/riscv/crypto/aes-block-riscv64.ko] undefined! > > ERROR: modpost: "aes_ecb_decrypt_zvkned" [arch/riscv/crypto/aes-block-riscv64.ko] undefined! > > ERROR: modpost: "aes_cbc_encrypt_zvkned" [arch/riscv/crypto/aes-block-riscv64.ko] undefined! > > ERROR: modpost: "aes_ecb_encrypt_zvkned" [arch/riscv/crypto/aes-block-riscv64.ko] undefined! > > > > To fix this, I think we should just merge the two modules and kconfig options > > together so that there is one module that provides both the AES modes and the > > AES single-block cipher. That's how x86's aesni-intel works, for example. > > > > - Eric > > That's a bug in my patchset. > I don't test with all options with `M` settings since I can't boot to qemu with all `M` settings. > > Could we move the cbc and ecb from `aes-riscv64-zvkned` to `aes-block-riscv64` instead of merging > these two modules? > Thus, we could still enable the single aes block cipher without other extensions(e.g. zvbb or zvkg). My proposal to merge the two modules still results in the single-block AES cipher being registered when zvkned alone is present. It just won't be selectable separately via kconfig. I think that's fine. - Eric
