public inbox for linux-riscv@lists.infradead.org
 help / color / mirror / Atom feed
From: Eric Biggers <ebiggers@kernel.org>
To: Jerry Shih <jerry.shih@sifive.com>
Cc: paul.walmsley@sifive.com, palmer@dabbelt.com,
	aou@eecs.berkeley.edu, herbert@gondor.apana.org.au,
	davem@davemloft.net, andy.chiu@sifive.com,
	greentime.hu@sifive.com, conor.dooley@microchip.com,
	guoren@kernel.org, bjorn@rivosinc.com, heiko@sntech.de,
	ardb@kernel.org, phoebe.chen@sifive.com, hongrong.hsu@sifive.com,
	linux-riscv@lists.infradead.org, linux-kernel@vger.kernel.org,
	linux-crypto@vger.kernel.org
Subject: Re: [PATCH 07/12] RISC-V: crypto: add Zvkg accelerated GCM GHASH implementation
Date: Tue, 21 Nov 2023 17:42:16 -0800	[thread overview]
Message-ID: <20231122014216.GI2172@sol.localdomain> (raw)
In-Reply-To: <20231025183644.8735-8-jerry.shih@sifive.com>

On Thu, Oct 26, 2023 at 02:36:39AM +0800, Jerry Shih wrote:
> +struct riscv64_ghash_context {
> +	be128 key;
> +};
> +
> +struct riscv64_ghash_desc_ctx {
> +	be128 shash;
> +	u8 buffer[GHASH_BLOCK_SIZE];
> +	u32 bytes;
> +};

I recommend calling the first struct 'riscv64_ghash_tfm_ctx', and making the
pointers to it be named 'tctx'.  That would more clearly distinguish it from the
desc_ctx / dctx.

> +
> +typedef void (*ghash_func)(be128 *Xi, const be128 *H, const u8 *inp,
> +			   size_t len);
> +
> +static inline void ghash_blocks(const struct riscv64_ghash_context *ctx,
> +				struct riscv64_ghash_desc_ctx *dctx,
> +				const u8 *src, size_t srclen, ghash_func func)
> +	if (crypto_simd_usable()) {
> +		kernel_vector_begin();
> +		func(&dctx->shash, &ctx->key, src, srclen);
> +		kernel_vector_end();

The indirection to ghash_func is unnecessary, since the only value is
gcm_ghash_rv64i_zvkg.

This also means that ghash_update() should be folded into ghash_update_zvkg(),
and ghash_final() into ghash_final_zvkg().

> +	} else {
> +		while (srclen >= GHASH_BLOCK_SIZE) {
> +			crypto_xor((u8 *)&dctx->shash, src, GHASH_BLOCK_SIZE);
> +			gf128mul_lle(&dctx->shash, &ctx->key);
> +			srclen -= GHASH_BLOCK_SIZE;
> +			src += GHASH_BLOCK_SIZE;
> +		}
> +	}

The assembly code uses the equivalent of the following do-while loop instead:

        do {
                srclen -= GHASH_BLOCK_SIZE;
        } while (srclen);

I.e., it assumes the length here is nonzero and a multiple of 16, which it is.

To avoid confusion, I recommend making the C code use the same do-while loop.


>        const struct riscv64_ghash_context *ctx =
>               crypto_tfm_ctx(crypto_shash_tfm(desc->tfm));

crypto_tfm_ctx(crypto_shash_tfm(tfm)) should be crypto_shash_ctx(tfm)

> +static int ghash_final(struct shash_desc *desc, u8 *out, ghash_func func)
> +{
> +	const struct riscv64_ghash_context *ctx =
> +		crypto_tfm_ctx(crypto_shash_tfm(desc->tfm));
> +	struct riscv64_ghash_desc_ctx *dctx = shash_desc_ctx(desc);
> +	int i;
> +
> +	if (dctx->bytes) {
> +		for (i = dctx->bytes; i < GHASH_BLOCK_SIZE; i++)
> +			dctx->buffer[i] = 0;
> +
> +		ghash_blocks(ctx, dctx, dctx->buffer, GHASH_BLOCK_SIZE, func);
> +		dctx->bytes = 0;
> +	}
> +

Setting dctx->bytes above is unnecessary.

> +static int ghash_init(struct shash_desc *desc)
> +{
> +	struct riscv64_ghash_desc_ctx *dctx = shash_desc_ctx(desc);
> +
> +	*dctx = (struct riscv64_ghash_desc_ctx){};
> +
> +	return 0;
> +}
> +
> +static int ghash_update_zvkg(struct shash_desc *desc, const u8 *src,
> +			     unsigned int srclen)
> +{
> +	return ghash_update(desc, src, srclen, gcm_ghash_rv64i_zvkg);
> +}
> +
> +static int ghash_final_zvkg(struct shash_desc *desc, u8 *out)
> +{
> +	return ghash_final(desc, out, gcm_ghash_rv64i_zvkg);
> +}
> +
> +static int ghash_setkey(struct crypto_shash *tfm, const u8 *key,
> +			unsigned int keylen)
> +{
> +	struct riscv64_ghash_context *ctx =
> +		crypto_tfm_ctx(crypto_shash_tfm(tfm));
> +
> +	if (keylen != GHASH_BLOCK_SIZE)
> +		return -EINVAL;
> +
> +	memcpy(&ctx->key, key, GHASH_BLOCK_SIZE);
> +
> +	return 0;
> +}
> +
> +static struct shash_alg riscv64_ghash_alg_zvkg = {
> +	.digestsize = GHASH_DIGEST_SIZE,
> +	.init = ghash_init,
> +	.update = ghash_update_zvkg,
> +	.final = ghash_final_zvkg,
> +	.setkey = ghash_setkey,

IMO it's helpful to order the shash functions as follows, both in their
definitions and their fields in struct shash_alg:

    setkey
    init
    update
    final

That matches the order in which they're called.

- Eric

_______________________________________________
linux-riscv mailing list
linux-riscv@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-riscv

  reply	other threads:[~2023-11-22  1:42 UTC|newest]

Thread overview: 50+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-10-25 18:36 [PATCH 00/12] RISC-V: provide some accelerated cryptography implementations using vector extensions Jerry Shih
2023-10-25 18:36 ` [PATCH 01/12] RISC-V: add helper function to read the vector VLEN Jerry Shih
2023-10-25 18:36 ` [PATCH 02/12] RISC-V: hook new crypto subdir into build-system Jerry Shih
2023-10-25 18:36 ` [PATCH 03/12] RISC-V: crypto: add OpenSSL perl module for vector instructions Jerry Shih
2023-10-25 18:36 ` [PATCH 04/12] RISC-V: crypto: add Zvkned accelerated AES implementation Jerry Shih
2023-11-02  4:51   ` Eric Biggers
2023-11-20  2:53     ` Jerry Shih
2023-10-25 18:36 ` [PATCH 05/12] crypto: scatterwalk - Add scatterwalk_next() to get the next scatterlist in scatter_walk Jerry Shih
2023-10-25 18:36 ` [PATCH 06/12] RISC-V: crypto: add accelerated AES-CBC/CTR/ECB/XTS implementations Jerry Shih
2023-11-02  5:16   ` Eric Biggers
2023-11-07  8:53     ` Jerry Shih
2023-11-09  7:16       ` Eric Biggers
2023-11-10  3:58         ` Jerry Shih
2023-11-10  4:34           ` Eric Biggers
2023-11-10  4:58         ` Andy Chiu
2023-11-10  5:44           ` Eric Biggers
2023-11-11 11:08             ` Ard Biesheuvel
2023-11-11 17:52               ` Eric Biggers
2023-11-20  2:47     ` Jerry Shih
2023-11-20 19:28       ` Eric Biggers
2023-11-22  1:14     ` Eric Biggers
2023-11-27  2:52       ` Jerry Shih
2023-11-09  8:05   ` Eric Biggers
2023-11-10  4:06     ` Jerry Shih
2023-11-20  2:36       ` Jerry Shih
2023-10-25 18:36 ` [PATCH 07/12] RISC-V: crypto: add Zvkg accelerated GCM GHASH implementation Jerry Shih
2023-11-22  1:42   ` Eric Biggers [this message]
2023-11-27  2:49     ` Jerry Shih
2023-10-25 18:36 ` [PATCH 08/12] RISC-V: crypto: add Zvknha/b accelerated SHA224/256 implementations Jerry Shih
2023-10-25 18:36 ` [PATCH 09/12] RISC-V: crypto: add Zvknhb accelerated SHA384/512 implementations Jerry Shih
2023-11-22  1:32   ` Eric Biggers
2023-11-27  2:50     ` Jerry Shih
2023-10-25 18:36 ` [PATCH 10/12] RISC-V: crypto: add Zvksed accelerated SM4 implementation Jerry Shih
2023-11-02  5:58   ` Eric Biggers
2023-11-20  2:55     ` Jerry Shih
2023-10-25 18:36 ` [PATCH 11/12] RISC-V: crypto: add Zvksh accelerated SM3 implementation Jerry Shih
2023-10-25 18:36 ` [PATCH 12/12] RISC-V: crypto: add Zvkb accelerated ChaCha20 implementation Jerry Shih
2023-11-02  5:43   ` Eric Biggers
2023-11-20  2:55     ` Jerry Shih
2023-11-20 19:18       ` Eric Biggers
2023-11-21 10:55         ` Jerry Shih
2023-11-21 13:14           ` Conor Dooley
2023-11-21 23:37             ` Eric Biggers
2023-11-22  0:39               ` Conor Dooley
2023-11-22 17:37             ` Jerry Shih
2023-11-22 18:05               ` Palmer Dabbelt
2023-11-22 18:20               ` Conor Dooley
2023-11-22 19:05                 ` Jerry Shih
2023-11-22  1:29   ` Eric Biggers
2023-11-27  2:14     ` Jerry Shih

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20231122014216.GI2172@sol.localdomain \
    --to=ebiggers@kernel.org \
    --cc=andy.chiu@sifive.com \
    --cc=aou@eecs.berkeley.edu \
    --cc=ardb@kernel.org \
    --cc=bjorn@rivosinc.com \
    --cc=conor.dooley@microchip.com \
    --cc=davem@davemloft.net \
    --cc=greentime.hu@sifive.com \
    --cc=guoren@kernel.org \
    --cc=heiko@sntech.de \
    --cc=herbert@gondor.apana.org.au \
    --cc=hongrong.hsu@sifive.com \
    --cc=jerry.shih@sifive.com \
    --cc=linux-crypto@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-riscv@lists.infradead.org \
    --cc=palmer@dabbelt.com \
    --cc=paul.walmsley@sifive.com \
    --cc=phoebe.chen@sifive.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox