linux-kernel - Re: [PATCH v2 09/14] crypto: caam - keep both virtual and dma key addresses

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Fri, 19 Jul 2019 10:08:32 +0000
From:   Horia Geanta <horia.geanta@....com>
To:     Iuliana Prodan <iuliana.prodan@....com>,
        Herbert Xu <herbert@...dor.apana.org.au>,
        Aymen Sghaier <aymen.sghaier@....com>
CC:     "David S. Miller" <davem@...emloft.net>,
        "linux-crypto@...r.kernel.org" <linux-crypto@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        dl-linux-imx <linux-imx@....com>
Subject: Re: [PATCH v2 09/14] crypto: caam - keep both virtual and dma key
 addresses

On 7/19/2019 2:58 AM, Iuliana Prodan wrote:
> From: Horia Geantă <horia.geanta@....com>
> 
> Update alginfo struct to keep both virtual and dma key addresses,
> so that descriptors have them at hand.
> One example where this is needed is in the xcbc(aes) shared descriptors,
> which are updated in current patch.
> Another example is the upcoming fix for DKP.
> 
> Signed-off-by: Horia Geantă <horia.geanta@....com>
> ---
>  drivers/crypto/caam/caamhash.c      | 26 ++++++++++++--------------
>  drivers/crypto/caam/caamhash_desc.c |  5 ++---
>  drivers/crypto/caam/caamhash_desc.h |  2 +-
>  drivers/crypto/caam/desc_constr.h   | 10 ++++------
>  4 files changed, 19 insertions(+), 24 deletions(-)
> 
> diff --git a/drivers/crypto/caam/caamhash.c b/drivers/crypto/caam/caamhash.c
> index 2ec4bad..14fdfa1 100644
> --- a/drivers/crypto/caam/caamhash.c
> +++ b/drivers/crypto/caam/caamhash.c
[...]
> @@ -508,6 +502,10 @@ static int axcbc_setkey(struct crypto_ahash *ahash, const u8 *key,
>  	memcpy(ctx->key, key, keylen);
>  	dma_sync_single_for_device(jrdev, ctx->key_dma, keylen, DMA_TO_DEVICE);
>  	ctx->adata.keylen = keylen;
> +	/* key is loaded from memory for UPDATE and FINALIZE states */
> +	ctx->adata.key_dma = ctx->key_dma;
> +	/* key is immediate data for INIT and INITFINAL states */
> +	ctx->adata.key_virt = ctx->key;
>  
Now that it's possible to store both virtual and dma key addresses,
it would be more efficient to move these assignments .cra_init callback.

I'll submit the changes in v3.

Horia

[...]
> diff --git a/drivers/crypto/caam/desc_constr.h b/drivers/crypto/caam/desc_constr.h
> index 5988a26..8154174 100644
> --- a/drivers/crypto/caam/desc_constr.h
> +++ b/drivers/crypto/caam/desc_constr.h
> @@ -457,8 +457,8 @@ do { \
>   *           functions where it is used.
>   * @keylen: length of the provided algorithm key, in bytes
>   * @keylen_pad: padded length of the provided algorithm key, in bytes
> - * @key: address where algorithm key resides; virtual address if key_inline
> - *       is true, dma (bus) address if key_inline is false.
> + * @key_dma: dma (bus) address where algorithm key resides
> + * @key_virt: virtual address where algorithm key resides
>   * @key_inline: true - key can be inlined in the descriptor; false - key is
>   *              referenced by the descriptor
>   */
> @@ -466,10 +466,8 @@ struct alginfo {
>  	u32 algtype;
>  	unsigned int keylen;
>  	unsigned int keylen_pad;
> -	union {
> -		dma_addr_t key_dma;
> -		const void *key_virt;
> -	};
> +	dma_addr_t key_dma;
> +	const void *key_virt;
>  	bool key_inline;
>  };
>  
>