Re: [PATCH 3/5] jffs2: Add a "favourlzo" compression mode to jffs2

From: Randy Dunlap
Date: Tue May 01 2007 - 16:02:51 EST


On Tue, 01 May 2007 15:47:12 +0100 Richard Purdie wrote:

> Add a "favourlzo" compression mode to jffs2 which tries to
> optimise by size but gives lzo an advantage when comparing sizes.
> This means the faster lzo algorithm can be preferred when there
> isn't much difference in compressed size (the exact threshold can
> be changed).
>
> Signed-off-by: Richard Purdie <rpurdie@xxxxxxxxxxxxxx>
> ---
> fs/Kconfig | 7 +++++++
> fs/jffs2/compr.c | 45 ++++++++++++++++++++++++++++++++++++++++-----
> fs/jffs2/compr.h | 3 +++
> 3 files changed, 50 insertions(+), 5 deletions(-)
>
> diff --git a/fs/Kconfig b/fs/Kconfig
> index 1645dfa..328a62f 100644
> --- a/fs/Kconfig
> +++ b/fs/Kconfig
> @@ -1359,6 +1359,13 @@ config JFFS2_CMODE_SIZE
> Tries all compressors and chooses the one which has the smallest
> result.
>
> +config JFFS2_CMODE_FAVOURLZO
> + bool "Favour LZO"
> + help
> + Tries all compressors and chooses the one which has the smallest
> + result but gives some preference to LZO (which has faster
> + decompression) at the expense of size.

Use tabs instead of spaces on "bool" and "help".
Use <tab><space><space> on help text lines.

> endchoice
>
> config CRAMFS
> diff --git a/fs/jffs2/compr.c b/fs/jffs2/compr.c
> index 6a23408..fdb0efd 100644
> --- a/fs/jffs2/compr.c
> +++ b/fs/jffs2/compr.c
> @@ -24,6 +24,34 @@ static int jffs2_compression_mode = JFFS2_COMPR_MODE_PRIORITY;
> /* Statistics for blocks stored without compression */
> static uint32_t none_stat_compr_blocks=0,none_stat_decompr_blocks=0,none_stat_compr_size=0;
>
> +
> +/*
> + * Return 1 to use this compression
> + */
> +static int jffs2_is_best_compression(struct jffs2_compressor *this,
> + struct jffs2_compressor *best, uint32_t size, uint32_t bestsize)
> +{
> + switch (jffs2_compression_mode) {

Place case statements at same indent level as 'switch'.

> + case JFFS2_COMPR_MODE_SIZE:
> + if (bestsize > size)
> + return 1;
> + return 0;
> + case JFFS2_COMPR_MODE_FAVOURLZO:
> + if ((this->compr == JFFS2_COMPR_LZO) && (bestsize > size))
> + return 1;
> + if ((best->compr != JFFS2_COMPR_LZO) && (bestsize > size))
> + return 1;
> + if ((this->compr == JFFS2_COMPR_LZO) && (bestsize > (size * FAVOUR_LZO_PERCENT / 100)))

line too long.

> + return 1;
> + if ((bestsize * FAVOUR_LZO_PERCENT / 100) > size)
> + return 1;
> +
> + return 0;
> + }
> + /* Shouldn't happen */
> + return 0;
> +}
> +
> /* jffs2_compress:
> * @data: Pointer to uncompressed data
> * @cdata: Pointer to returned pointer to buffer for compressed data
> @@ -106,15 +135,15 @@ uint16_t jffs2_compress(struct jffs2_sb_info *c, struct jffs2_inode_info *f,
> }
> if (!this->compr_buf) {
> spin_unlock(&jffs2_compressor_list_lock);
> - tmp_buf = kmalloc(orig_dlen,GFP_KERNEL);
> + tmp_buf = kmalloc(orig_slen,GFP_KERNEL);

space after comma

> spin_lock(&jffs2_compressor_list_lock);
> if (!tmp_buf) {
> - printk(KERN_WARNING "JFFS2: No memory for compressor allocation. (%d bytes)\n",orig_dlen);
> + printk(KERN_WARNING "JFFS2: No memory for compressor allocation. (%d bytes)\n",orig_slen);

space after comma. Line too long (except for David :).

> continue;
> }
> else {
> this->compr_buf = tmp_buf;
> - this->compr_buf_size = orig_dlen;
> + this->compr_buf_size = orig_slen;
> }
> }
> this->usecount++;
> @@ -125,7 +154,8 @@ uint16_t jffs2_compress(struct jffs2_sb_info *c, struct jffs2_inode_info *f,
> spin_lock(&jffs2_compressor_list_lock);
> this->usecount--;
> if (!compr_ret) {
> - if ((!best_dlen)||(best_dlen>*cdatalen)) {
> + if (((!best_dlen) || jffs2_is_best_compression(this, best, *cdatalen, best_dlen))

line too long

> + && (*cdatalen < *datalen)) {
> best_dlen = *cdatalen;
> best_slen = *datalen;
> best = this;

---
~Randy
*** Remember to use Documentation/SubmitChecklist when testing your code ***
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/