public inbox for linux-mtd@lists.infradead.org
 help / color / mirror / Atom feed
From: Zhihao Cheng <chengzhihao1@huawei.com>
To: Herbert Xu <herbert@gondor.apana.org.au>,
	Linux Crypto Mailing List <linux-crypto@vger.kernel.org>
Cc: Richard Weinberger <richard@nod.at>,
	<linux-mtd@lists.infradead.org>,
	"Rafael J. Wysocki" <rafael@kernel.org>,
	Pavel Machek <pavel@ucw.cz>, <linux-pm@vger.kernel.org>,
	Steffen Klassert <steffen.klassert@secunet.com>,
	<netdev@vger.kernel.org>
Subject: Re: [v5 PATCH 14/14] ubifs: Pass folios to acomp
Date: Wed, 19 Mar 2025 17:44:17 +0800	[thread overview]
Message-ID: <9f77f2a4-e4ba-813e-f44d-3a42d9637d0f@huawei.com> (raw)
In-Reply-To: <99ae6a15afc1478bab201949dc3dbb2c7634b687.1742034499.git.herbert@gondor.apana.org.au>

在 2025/3/15 18:30, Herbert Xu 写道:
> As the acomp interface supports folios, use that instead of mapping
> the data in ubifs.
> 
> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
> ---
>   fs/ubifs/compress.c | 106 +++++++++++++++++++++++++++++++++++++++++++-
>   fs/ubifs/file.c     |  74 +++++++++++--------------------
>   fs/ubifs/journal.c  |   9 ++--
>   fs/ubifs/ubifs.h    |  11 ++++-
>   4 files changed, 145 insertions(+), 55 deletions(-)


Tested-by: Zhihao Cheng <chengzhihao1@huawei.com> # For xfstests

> 
> diff --git a/fs/ubifs/compress.c b/fs/ubifs/compress.c
> index a241ba01c9a8..ea6f06adcd43 100644
> --- a/fs/ubifs/compress.c
> +++ b/fs/ubifs/compress.c
> @@ -16,6 +16,7 @@
>    */
>   
>   #include <crypto/acompress.h>
> +#include <linux/highmem.h>
>   #include "ubifs.h"
>   
>   /* Fake description object for the "none" compressor */
> @@ -126,7 +127,7 @@ void ubifs_compress(const struct ubifs_info *c, const void *in_buf,
>   	{
>   		ACOMP_REQUEST_ALLOC(req, compr->cc, GFP_NOFS | __GFP_NOWARN);
>   
> -		acomp_request_set_src_nondma(req, in_buf, in_len);
> +		acomp_request_set_src_dma(req, in_buf, in_len);

Why not merging it into patch 13?
>   		err = ubifs_compress_req(c, req, out_buf, out_len, compr->name);
>   	}
>   
> @@ -141,6 +142,58 @@ void ubifs_compress(const struct ubifs_info *c, const void *in_buf,
>   	*compr_type = UBIFS_COMPR_NONE;
>   }
>   
> +/**
> + * ubifs_compress_folio - compress folio.
> + * @c: UBIFS file-system description object
> + * @in_folio: data to compress
> + * @in_offset: offset into @in_folio
> + * @in_len: length of the data to compress
> + * @out_buf: output buffer where compressed data should be stored
> + * @out_len: output buffer length is returned here
> + * @compr_type: type of compression to use on enter, actually used compression
> + *              type on exit
> + *
> + * This function compresses input folio @in_folio of length @in_len and
> + * stores the result in the output buffer @out_buf and the resulting length
> + * in @out_len. If the input buffer does not compress, it is just copied
> + * to the @out_buf. The same happens if @compr_type is %UBIFS_COMPR_NONE
> + * or if compression error occurred.
> + *
> + * Note, if the input buffer was not compressed, it is copied to the output
> + * buffer and %UBIFS_COMPR_NONE is returned in @compr_type.
> + */
> +void ubifs_compress_folio(const struct ubifs_info *c, struct folio *in_folio,
> +			  size_t in_offset, int in_len, void *out_buf,
> +			  int *out_len, int *compr_type)
> +{
> +	int err;
> +	struct ubifs_compressor *compr = ubifs_compressors[*compr_type];
> +
> +	if (*compr_type == UBIFS_COMPR_NONE)
> +		goto no_compr;
> +
> +	/* If the input data is small, do not even try to compress it */
> +	if (in_len < UBIFS_MIN_COMPR_LEN)
> +		goto no_compr;
> +
> +	{
> +		ACOMP_REQUEST_ALLOC(req, compr->cc, GFP_NOFS | __GFP_NOWARN);
> +
> +		acomp_request_set_src_folio(req, in_folio, in_offset, in_len);
> +		err = ubifs_compress_req(c, req, out_buf, out_len, compr->name);
> +	}
> +
> +	if (err)
> +		goto no_compr;
> +
> +	return;
> +
> +no_compr:
> +	memcpy_from_folio(out_buf, in_folio, in_offset, in_len);
> +	*out_len = in_len;
> +	*compr_type = UBIFS_COMPR_NONE;
> +}
> +
>   static int ubifs_decompress_req(const struct ubifs_info *c,
>   				struct acomp_req *req,
>   				const void *in_buf, int in_len, int *out_len,
> @@ -205,7 +258,56 @@ int ubifs_decompress(const struct ubifs_info *c, const void *in_buf,
>   	{
>   		ACOMP_REQUEST_ALLOC(req, compr->cc, GFP_NOFS | __GFP_NOWARN);
>   
> -		acomp_request_set_dst_nondma(req, out_buf, *out_len);
> +		acomp_request_set_dst_dma(req, out_buf, *out_len);

Same as the suggestion above.
> +		return ubifs_decompress_req(c, req, in_buf, in_len, out_len,
> +					    compr->name);
> +	}
> +}
> +
> +/**
> + * ubifs_decompress_folio - decompress folio.
> + * @c: UBIFS file-system description object
> + * @in_buf: data to decompress
> + * @in_len: length of the data to decompress
> + * @out_folio: output folio where decompressed data should
> + * @out_offset: offset into @out_folio
> + * @out_len: output length is returned here
> + * @compr_type: type of compression
> + *
> + * This function decompresses data from buffer @in_buf into folio
> + * @out_folio.  The length of the uncompressed data is returned in
> + * @out_len.  This functions returns %0 on success or a negative error
> + * code on failure.
> + */
> +int ubifs_decompress_folio(const struct ubifs_info *c, const void *in_buf,
> +			   int in_len, struct folio *out_folio,
> +			   size_t out_offset, int *out_len, int compr_type)
> +{
> +	struct ubifs_compressor *compr;
> +
> +	if (unlikely(compr_type < 0 || compr_type >= UBIFS_COMPR_TYPES_CNT)) {
> +		ubifs_err(c, "invalid compression type %d", compr_type);
> +		return -EINVAL;
> +	}
> +
> +	compr = ubifs_compressors[compr_type];
> +
> +	if (unlikely(!compr->capi_name)) {
> +		ubifs_err(c, "%s compression is not compiled in", compr->name);
> +		return -EINVAL;
> +	}
> +
> +	if (compr_type == UBIFS_COMPR_NONE) {
> +		memcpy_to_folio(out_folio, out_offset, in_buf, in_len);
> +		*out_len = in_len;
> +		return 0;
> +	}
> +
> +	{
> +		ACOMP_REQUEST_ALLOC(req, compr->cc, GFP_NOFS | __GFP_NOWARN);
> +
> +		acomp_request_set_dst_folio(req, out_folio, out_offset,
> +					    *out_len);
>   		return ubifs_decompress_req(c, req, in_buf, in_len, out_len,
>   					    compr->name);
>   	}

______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

  reply	other threads:[~2025-03-19  9:45 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-03-15 10:30 [v5 PATCH 00/14] crypto: acomp - Add virtual address and folio support Herbert Xu
2025-03-15 10:30 ` [v5 PATCH 01/14] xfrm: ipcomp: Call pskb_may_pull in ipcomp_input Herbert Xu
2025-03-15 10:30 ` [v5 PATCH 02/14] crypto: scomp - Remove support for some non-trivial SG lists Herbert Xu
2025-03-15 10:30 ` [v5 PATCH 03/14] crypto: iaa - Remove dst_null support Herbert Xu
2025-03-15 10:30 ` [v5 PATCH 04/14] crypto: qat " Herbert Xu
2025-03-15 10:30 ` [v5 PATCH 05/14] crypto: acomp - Remove dst_free Herbert Xu
2025-03-15 10:30 ` [v5 PATCH 06/14] crypto: scomp - Add chaining and virtual address support Herbert Xu
2025-03-15 10:30 ` [v5 PATCH 07/14] crypto: acomp - Add ACOMP_REQUEST_ALLOC and acomp_request_alloc_extra Herbert Xu
2025-03-15 10:30 ` [v5 PATCH 08/14] crypto: iaa - Use acomp stack fallback Herbert Xu
2025-03-15 10:30 ` [v5 PATCH 09/14] crypto: acomp - Add async nondma fallback Herbert Xu
2025-03-15 10:30 ` [v5 PATCH 10/14] crypto: acomp - Add support for folios Herbert Xu
2025-03-19  9:57   ` Herbert Xu
2025-03-15 10:30 ` [v5 PATCH 11/14] xfrm: ipcomp: Use crypto_acomp interface Herbert Xu
2025-03-20  9:33   ` Herbert Xu
2025-04-25 12:11   ` Sabrina Dubroca
2025-04-25 12:20     ` Herbert Xu
2025-04-25 12:28       ` Sabrina Dubroca
2025-03-15 10:30 ` [v5 PATCH 12/14] PM: hibernate: " Herbert Xu
2025-03-15 10:30 ` [v5 PATCH 13/14] ubifs: " Herbert Xu
2025-03-19  9:39   ` Zhihao Cheng
2025-03-15 10:30 ` [v5 PATCH 14/14] ubifs: Pass folios to acomp Herbert Xu
2025-03-19  9:44   ` Zhihao Cheng [this message]
2025-03-19  9:46     ` Herbert Xu
2025-03-19 12:28       ` Zhihao Cheng

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=9f77f2a4-e4ba-813e-f44d-3a42d9637d0f@huawei.com \
    --to=chengzhihao1@huawei.com \
    --cc=herbert@gondor.apana.org.au \
    --cc=linux-crypto@vger.kernel.org \
    --cc=linux-mtd@lists.infradead.org \
    --cc=linux-pm@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=pavel@ucw.cz \
    --cc=rafael@kernel.org \
    --cc=richard@nod.at \
    --cc=steffen.klassert@secunet.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