[PATCH] fs/udf: re-use hex_asc_upper_{hi,lo} macros

[PATCH] fs/udf: re-use hex_asc_upper_{hi,lo} macros

[Andy Shevchenko]

This patch cleans up udf_translate_to_linux() a bit by using globally defined
macros instead of custom code.

We can use sprintf(buf, "%04X", ...) there as well, but this one faster.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
---
 fs/udf/unicode.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

diff --git a/fs/udf/unicode.c b/fs/udf/unicode.c
index 44b815e..afd470e 100644
--- a/fs/udf/unicode.c
+++ b/fs/udf/unicode.c
@@ -412,7 +412,6 @@ static int udf_translate_to_linux(uint8_t *newName, uint8_t *udfName,
 	int extIndex = 0, newExtIndex = 0, hasExt = 0;
 	unsigned short valueCRC;
 	uint8_t curr;
-	const uint8_t hexChar[] = "0123456789ABCDEF";
 
 	if (udfName[0] == '.' &&
 	    (udfLen == 1 || (udfLen == 2 && udfName[1] == '.'))) {
@@ -477,10 +476,10 @@ static int udf_translate_to_linux(uint8_t *newName, uint8_t *udfName,
 			newIndex = 250;
 		newName[newIndex++] = CRC_MARK;
 		valueCRC = crc_itu_t(0, fidName, fidNameLen);
-		newName[newIndex++] = hexChar[(valueCRC & 0xf000) >> 12];
-		newName[newIndex++] = hexChar[(valueCRC & 0x0f00) >> 8];
-		newName[newIndex++] = hexChar[(valueCRC & 0x00f0) >> 4];
-		newName[newIndex++] = hexChar[(valueCRC & 0x000f)];
+		newName[newIndex++] = hex_asc_upper_hi(valueCRC >> 8);
+		newName[newIndex++] = hex_asc_upper_lo(valueCRC >> 8);
+		newName[newIndex++] = hex_asc_upper_hi(valueCRC);
+		newName[newIndex++] = hex_asc_upper_lo(valueCRC);
 
 		if (hasExt) {
 			newName[newIndex++] = EXT_MARK;
-- 
2.0.1

Re: [PATCH] fs/udf: re-use hex_asc_upper_{hi,lo} macros

[Jan Kara]

On Wed 09-07-14 15:35:30, Andy Shevchenko wrote:
> This patch cleans up udf_translate_to_linux() a bit by using globally defined
> macros instead of custom code.
> 
> We can use sprintf(buf, "%04X", ...) there as well, but this one faster.
> 
> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
  Thanks. I've merged the patch.

								Honza

> ---
>  fs/udf/unicode.c | 9 ++++-----
>  1 file changed, 4 insertions(+), 5 deletions(-)
> 
> diff --git a/fs/udf/unicode.c b/fs/udf/unicode.c
> index 44b815e..afd470e 100644
> --- a/fs/udf/unicode.c
> +++ b/fs/udf/unicode.c
> @@ -412,7 +412,6 @@ static int udf_translate_to_linux(uint8_t *newName, uint8_t *udfName,
>  	int extIndex = 0, newExtIndex = 0, hasExt = 0;
>  	unsigned short valueCRC;
>  	uint8_t curr;
> -	const uint8_t hexChar[] = "0123456789ABCDEF";
>  
>  	if (udfName[0] == '.' &&
>  	    (udfLen == 1 || (udfLen == 2 && udfName[1] == '.'))) {
> @@ -477,10 +476,10 @@ static int udf_translate_to_linux(uint8_t *newName, uint8_t *udfName,
>  			newIndex = 250;
>  		newName[newIndex++] = CRC_MARK;
>  		valueCRC = crc_itu_t(0, fidName, fidNameLen);
> -		newName[newIndex++] = hexChar[(valueCRC & 0xf000) >> 12];
> -		newName[newIndex++] = hexChar[(valueCRC & 0x0f00) >> 8];
> -		newName[newIndex++] = hexChar[(valueCRC & 0x00f0) >> 4];
> -		newName[newIndex++] = hexChar[(valueCRC & 0x000f)];
> +		newName[newIndex++] = hex_asc_upper_hi(valueCRC >> 8);
> +		newName[newIndex++] = hex_asc_upper_lo(valueCRC >> 8);
> +		newName[newIndex++] = hex_asc_upper_hi(valueCRC);
> +		newName[newIndex++] = hex_asc_upper_lo(valueCRC);
>  
>  		if (hasExt) {
>  			newName[newIndex++] = EXT_MARK;
> -- 
> 2.0.1
> 
-- 
Jan Kara <jack@suse.cz>
SUSE Labs, CR