mail archive of the barebox mailing list
 help / color / mirror / Atom feed
* [PATCH v2] crypto: crc32: make crc32 available in PBL
@ 2023-08-29 14:38 Johannes Zink
  2023-08-29 15:00 ` Ahmad Fatoum
  2023-09-04  8:19 ` Sascha Hauer
  0 siblings, 2 replies; 6+ messages in thread
From: Johannes Zink @ 2023-08-29 14:38 UTC (permalink / raw)
  To: patchwork-jzi; +Cc: Johannes Zink, Barebox Mailing List

crc32 may be required in PBL for checking data integrity. Add it to PBL
when CONFIG_CRC32 is enabled.

To save some memory use a slower-but-smaller variant of the crc32 algorithm
in the PBL.

Signed-off-by: Johannes Zink <j.zink@pengutronix.de>
---
To: Barebox Mailing List <barebox@lists.infradead.org>
Cc: Johannes Zink <j.zink@pengutronix.de>
Cc: patchwork-jzi@pengutronix.de
---

Changes:

v1->v2: Worked in Ahmad's review (thank you for reviewing my patch):
  - instead of using CRC32_EARLY, always add crc32.o to obj and pbl if
    CONFIG_CRC is set and rely on LTO to remove unused symbols instead
  - use a CRC32 implementation without a prepopulated Lookup Table,
    which trades in speed for code size in the PBL, analogously to what
    Sascha implemented in [1]
  - add hints on some of the #endifs for better readability of nested
    ifdef blocks
  - reword commit message to reflect on the changes made

    [1] 2d13b856604b ("crc: Add PBL variant for crc_itu_t()")
---
 crypto/Makefile |  2 +-
 crypto/crc32.c  | 34 ++++++++++++++++++++++++++++------
 2 files changed, 29 insertions(+), 7 deletions(-)

diff --git a/crypto/Makefile b/crypto/Makefile
index 22035d4f69ee..4a1c7e9615b8 100644
--- a/crypto/Makefile
+++ b/crypto/Makefile
@@ -1,6 +1,6 @@
 # SPDX-License-Identifier: GPL-2.0-only
 
-obj-$(CONFIG_CRC32)	+= crc32.o
+obj-pbl-$(CONFIG_CRC32)	+= crc32.o
 obj-pbl-$(CONFIG_CRC_ITU_T)	+= crc-itu-t.o
 obj-$(CONFIG_CRC7)	+= crc7.o
 obj-$(CONFIG_DIGEST)	+= digest.o
diff --git a/crypto/crc32.c b/crypto/crc32.c
index 95cb2212db2b..2cf13144c1d2 100644
--- a/crypto/crc32.c
+++ b/crypto/crc32.c
@@ -22,7 +22,7 @@
 #define STATIC static inline
 #endif
 
-#ifdef CONFIG_DYNAMIC_CRC_TABLE
+#if defined(CONFIG_DYNAMIC_CRC_TABLE) && !defined(__PBL__)
 
 static uint32_t *crc_table;
 
@@ -73,7 +73,7 @@ static void make_crc_table(void)
     crc_table[n] = c;
   }
 }
-#else
+#elif !defined(__PBL__)
 /* ========================================================================
  * Table of CRC-32's of all single-byte values (made by make_crc_table)
  */
@@ -131,9 +131,30 @@ static const uint32_t crc_table[256] = {
   0x5d681b02L, 0x2a6f2b94L, 0xb40bbe37L, 0xc30c8ea1L, 0x5a05df1bL,
   0x2d02ef8dL
 };
-#endif
+#endif /* defined(CONFIG_DYNAMIC_CRC_TABLE) && !defined(__PBL__) */
+
+
+
+#ifdef __PBL__
+#define CRC32_POLY 0xEDB88320L
+STATIC uint32_t crc32(uint32_t crc, const void *_buf, unsigned int len)
+{
+	int i, j;
+	const unsigned char *buf = _buf;
 
+	crc = crc ^ 0xffffffffL;
 
+	for (i=0; i< len; i++) {
+		crc ^= buf[i];
+		for (j = 0; j < 8; j++) {
+			crc = (crc >> 1) ^ ((crc & 1) ? CRC32_POLY : 0);
+		}
+	}
+
+	crc = crc ^ 0xffffffffL;
+	return crc;
+}
+#else
 /* ========================================================================= */
 #define DO1(buf) crc = crc_table[((int)crc ^ (*buf++)) & 0xff] ^ (crc >> 8);
 #define DO2(buf)  DO1(buf); DO1(buf);
@@ -148,7 +169,7 @@ STATIC uint32_t crc32(uint32_t crc, const void *_buf, unsigned int len)
 #ifdef CONFIG_DYNAMIC_CRC_TABLE
 	if (!crc_table)
 		make_crc_table();
-#endif
+#endif /* CONFIG_DYNAMIC_CRC_TABLE */
     crc = crc ^ 0xffffffffL;
     while (len >= 8)
     {
@@ -162,7 +183,7 @@ STATIC uint32_t crc32(uint32_t crc, const void *_buf, unsigned int len)
 }
 #ifdef __BAREBOX__
 EXPORT_SYMBOL(crc32);
-#endif
+#endif /* __BAREBOX__ */
 
 /* No ones complement version. JFFS2 (and other things ?)
  * don't use ones compliment in their CRC calculations.
@@ -174,7 +195,7 @@ STATIC uint32_t crc32_no_comp(uint32_t crc, const void *_buf, unsigned int len)
 #ifdef CONFIG_DYNAMIC_CRC_TABLE
 	if (!crc_table)
 		make_crc_table();
-#endif
+#endif /* CONFIG_DYNAMIC_CRC_TABLE */
     while (len >= 8)
     {
       DO8(buf);
@@ -186,6 +207,7 @@ STATIC uint32_t crc32_no_comp(uint32_t crc, const void *_buf, unsigned int len)
 
     return crc;
 }
+#endif /* __PBL__ */
 
 STATIC uint32_t crc32_be(uint32_t crc, const void *_buf, unsigned int len)
 {

---
base-commit: bef38b18eeb5d2f1fac334fb8b831e47261e099c
change-id: 20230829-crc32_in_pbl-4d824629d4e2

Best regards,
-- 
Johannes Zink <j.zink@pengutronix.de>




^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH v2] crypto: crc32: make crc32 available in PBL
  2023-08-29 14:38 [PATCH v2] crypto: crc32: make crc32 available in PBL Johannes Zink
@ 2023-08-29 15:00 ` Ahmad Fatoum
  2023-09-04  8:19 ` Sascha Hauer
  1 sibling, 0 replies; 6+ messages in thread
From: Ahmad Fatoum @ 2023-08-29 15:00 UTC (permalink / raw)
  To: Johannes Zink, patchwork-jzi; +Cc: Barebox Mailing List

On 29.08.23 16:38, Johannes Zink wrote:
> crc32 may be required in PBL for checking data integrity. Add it to PBL
> when CONFIG_CRC32 is enabled.
> 
> To save some memory use a slower-but-smaller variant of the crc32 algorithm
> in the PBL.
> 
> Signed-off-by: Johannes Zink <j.zink@pengutronix.de>
> ---
> To: Barebox Mailing List <barebox@lists.infradead.org>
> Cc: Johannes Zink <j.zink@pengutronix.de>
> Cc: patchwork-jzi@pengutronix.de
> ---
> 
> Changes:
> 
> v1->v2: Worked in Ahmad's review (thank you for reviewing my patch):
>   - instead of using CRC32_EARLY, always add crc32.o to obj and pbl if
>     CONFIG_CRC is set and rely on LTO to remove unused symbols instead

fyi, LTO goes beyond linker garbage collection and actively rewrites
code, while --gc-sections, just checks which data/code sections are unused
and drops the dead code/data. We don't (yet) do LTO in barebox.

Patch looks good otherwise:
Reviewed-by: Ahmad Fatoum <a.fatoum@pengutronix.de>

>   - use a CRC32 implementation without a prepopulated Lookup Table,
>     which trades in speed for code size in the PBL, analogously to what
>     Sascha implemented in [1]
>   - add hints on some of the #endifs for better readability of nested
>     ifdef blocks
>   - reword commit message to reflect on the changes made
> 
>     [1] 2d13b856604b ("crc: Add PBL variant for crc_itu_t()")
> ---
>  crypto/Makefile |  2 +-
>  crypto/crc32.c  | 34 ++++++++++++++++++++++++++++------
>  2 files changed, 29 insertions(+), 7 deletions(-)
> 
> diff --git a/crypto/Makefile b/crypto/Makefile
> index 22035d4f69ee..4a1c7e9615b8 100644
> --- a/crypto/Makefile
> +++ b/crypto/Makefile
> @@ -1,6 +1,6 @@
>  # SPDX-License-Identifier: GPL-2.0-only
>  
> -obj-$(CONFIG_CRC32)	+= crc32.o
> +obj-pbl-$(CONFIG_CRC32)	+= crc32.o
>  obj-pbl-$(CONFIG_CRC_ITU_T)	+= crc-itu-t.o
>  obj-$(CONFIG_CRC7)	+= crc7.o
>  obj-$(CONFIG_DIGEST)	+= digest.o
> diff --git a/crypto/crc32.c b/crypto/crc32.c
> index 95cb2212db2b..2cf13144c1d2 100644
> --- a/crypto/crc32.c
> +++ b/crypto/crc32.c
> @@ -22,7 +22,7 @@
>  #define STATIC static inline
>  #endif
>  
> -#ifdef CONFIG_DYNAMIC_CRC_TABLE
> +#if defined(CONFIG_DYNAMIC_CRC_TABLE) && !defined(__PBL__)
>  
>  static uint32_t *crc_table;
>  
> @@ -73,7 +73,7 @@ static void make_crc_table(void)
>      crc_table[n] = c;
>    }
>  }
> -#else
> +#elif !defined(__PBL__)
>  /* ========================================================================
>   * Table of CRC-32's of all single-byte values (made by make_crc_table)
>   */
> @@ -131,9 +131,30 @@ static const uint32_t crc_table[256] = {
>    0x5d681b02L, 0x2a6f2b94L, 0xb40bbe37L, 0xc30c8ea1L, 0x5a05df1bL,
>    0x2d02ef8dL
>  };
> -#endif
> +#endif /* defined(CONFIG_DYNAMIC_CRC_TABLE) && !defined(__PBL__) */
> +
> +
> +
> +#ifdef __PBL__
> +#define CRC32_POLY 0xEDB88320L
> +STATIC uint32_t crc32(uint32_t crc, const void *_buf, unsigned int len)
> +{
> +	int i, j;
> +	const unsigned char *buf = _buf;
>  
> +	crc = crc ^ 0xffffffffL;
>  
> +	for (i=0; i< len; i++) {
> +		crc ^= buf[i];
> +		for (j = 0; j < 8; j++) {
> +			crc = (crc >> 1) ^ ((crc & 1) ? CRC32_POLY : 0);
> +		}
> +	}
> +
> +	crc = crc ^ 0xffffffffL;
> +	return crc;
> +}
> +#else
>  /* ========================================================================= */
>  #define DO1(buf) crc = crc_table[((int)crc ^ (*buf++)) & 0xff] ^ (crc >> 8);
>  #define DO2(buf)  DO1(buf); DO1(buf);
> @@ -148,7 +169,7 @@ STATIC uint32_t crc32(uint32_t crc, const void *_buf, unsigned int len)
>  #ifdef CONFIG_DYNAMIC_CRC_TABLE
>  	if (!crc_table)
>  		make_crc_table();
> -#endif
> +#endif /* CONFIG_DYNAMIC_CRC_TABLE */
>      crc = crc ^ 0xffffffffL;
>      while (len >= 8)
>      {
> @@ -162,7 +183,7 @@ STATIC uint32_t crc32(uint32_t crc, const void *_buf, unsigned int len)
>  }
>  #ifdef __BAREBOX__
>  EXPORT_SYMBOL(crc32);
> -#endif
> +#endif /* __BAREBOX__ */
>  
>  /* No ones complement version. JFFS2 (and other things ?)
>   * don't use ones compliment in their CRC calculations.
> @@ -174,7 +195,7 @@ STATIC uint32_t crc32_no_comp(uint32_t crc, const void *_buf, unsigned int len)
>  #ifdef CONFIG_DYNAMIC_CRC_TABLE
>  	if (!crc_table)
>  		make_crc_table();
> -#endif
> +#endif /* CONFIG_DYNAMIC_CRC_TABLE */
>      while (len >= 8)
>      {
>        DO8(buf);
> @@ -186,6 +207,7 @@ STATIC uint32_t crc32_no_comp(uint32_t crc, const void *_buf, unsigned int len)
>  
>      return crc;
>  }
> +#endif /* __PBL__ */
>  
>  STATIC uint32_t crc32_be(uint32_t crc, const void *_buf, unsigned int len)
>  {
> 
> ---
> base-commit: bef38b18eeb5d2f1fac334fb8b831e47261e099c
> change-id: 20230829-crc32_in_pbl-4d824629d4e2
> 
> Best regards,

-- 
Pengutronix e.K.                           |                             |
Steuerwalder Str. 21                       | http://www.pengutronix.de/  |
31137 Hildesheim, Germany                  | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |




^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH v2] crypto: crc32: make crc32 available in PBL
  2023-08-29 14:38 [PATCH v2] crypto: crc32: make crc32 available in PBL Johannes Zink
  2023-08-29 15:00 ` Ahmad Fatoum
@ 2023-09-04  8:19 ` Sascha Hauer
  2023-09-05  7:57   ` Johannes Zink
  1 sibling, 1 reply; 6+ messages in thread
From: Sascha Hauer @ 2023-09-04  8:19 UTC (permalink / raw)
  To: Johannes Zink; +Cc: patchwork-jzi, Barebox Mailing List

On Tue, Aug 29, 2023 at 04:38:32PM +0200, Johannes Zink wrote:
> crc32 may be required in PBL for checking data integrity. Add it to PBL
> when CONFIG_CRC32 is enabled.
> 
> To save some memory use a slower-but-smaller variant of the crc32 algorithm
> in the PBL.

I've just sent an alternative patch series. Please check if that works
for you.

Sascha


> 
> Signed-off-by: Johannes Zink <j.zink@pengutronix.de>
> ---
> To: Barebox Mailing List <barebox@lists.infradead.org>
> Cc: Johannes Zink <j.zink@pengutronix.de>
> Cc: patchwork-jzi@pengutronix.de
> ---
> 
> Changes:
> 
> v1->v2: Worked in Ahmad's review (thank you for reviewing my patch):
>   - instead of using CRC32_EARLY, always add crc32.o to obj and pbl if
>     CONFIG_CRC is set and rely on LTO to remove unused symbols instead
>   - use a CRC32 implementation without a prepopulated Lookup Table,
>     which trades in speed for code size in the PBL, analogously to what
>     Sascha implemented in [1]
>   - add hints on some of the #endifs for better readability of nested
>     ifdef blocks
>   - reword commit message to reflect on the changes made
> 
>     [1] 2d13b856604b ("crc: Add PBL variant for crc_itu_t()")
> ---
>  crypto/Makefile |  2 +-
>  crypto/crc32.c  | 34 ++++++++++++++++++++++++++++------
>  2 files changed, 29 insertions(+), 7 deletions(-)
> 
> diff --git a/crypto/Makefile b/crypto/Makefile
> index 22035d4f69ee..4a1c7e9615b8 100644
> --- a/crypto/Makefile
> +++ b/crypto/Makefile
> @@ -1,6 +1,6 @@
>  # SPDX-License-Identifier: GPL-2.0-only
>  
> -obj-$(CONFIG_CRC32)	+= crc32.o
> +obj-pbl-$(CONFIG_CRC32)	+= crc32.o
>  obj-pbl-$(CONFIG_CRC_ITU_T)	+= crc-itu-t.o
>  obj-$(CONFIG_CRC7)	+= crc7.o
>  obj-$(CONFIG_DIGEST)	+= digest.o
> diff --git a/crypto/crc32.c b/crypto/crc32.c
> index 95cb2212db2b..2cf13144c1d2 100644
> --- a/crypto/crc32.c
> +++ b/crypto/crc32.c
> @@ -22,7 +22,7 @@
>  #define STATIC static inline
>  #endif
>  
> -#ifdef CONFIG_DYNAMIC_CRC_TABLE
> +#if defined(CONFIG_DYNAMIC_CRC_TABLE) && !defined(__PBL__)
>  
>  static uint32_t *crc_table;
>  
> @@ -73,7 +73,7 @@ static void make_crc_table(void)
>      crc_table[n] = c;
>    }
>  }
> -#else
> +#elif !defined(__PBL__)
>  /* ========================================================================
>   * Table of CRC-32's of all single-byte values (made by make_crc_table)
>   */
> @@ -131,9 +131,30 @@ static const uint32_t crc_table[256] = {
>    0x5d681b02L, 0x2a6f2b94L, 0xb40bbe37L, 0xc30c8ea1L, 0x5a05df1bL,
>    0x2d02ef8dL
>  };
> -#endif
> +#endif /* defined(CONFIG_DYNAMIC_CRC_TABLE) && !defined(__PBL__) */
> +
> +
> +
> +#ifdef __PBL__
> +#define CRC32_POLY 0xEDB88320L
> +STATIC uint32_t crc32(uint32_t crc, const void *_buf, unsigned int len)
> +{
> +	int i, j;
> +	const unsigned char *buf = _buf;
>  
> +	crc = crc ^ 0xffffffffL;
>  
> +	for (i=0; i< len; i++) {
> +		crc ^= buf[i];
> +		for (j = 0; j < 8; j++) {
> +			crc = (crc >> 1) ^ ((crc & 1) ? CRC32_POLY : 0);
> +		}
> +	}
> +
> +	crc = crc ^ 0xffffffffL;
> +	return crc;
> +}
> +#else
>  /* ========================================================================= */
>  #define DO1(buf) crc = crc_table[((int)crc ^ (*buf++)) & 0xff] ^ (crc >> 8);
>  #define DO2(buf)  DO1(buf); DO1(buf);
> @@ -148,7 +169,7 @@ STATIC uint32_t crc32(uint32_t crc, const void *_buf, unsigned int len)
>  #ifdef CONFIG_DYNAMIC_CRC_TABLE
>  	if (!crc_table)
>  		make_crc_table();
> -#endif
> +#endif /* CONFIG_DYNAMIC_CRC_TABLE */
>      crc = crc ^ 0xffffffffL;
>      while (len >= 8)
>      {
> @@ -162,7 +183,7 @@ STATIC uint32_t crc32(uint32_t crc, const void *_buf, unsigned int len)
>  }
>  #ifdef __BAREBOX__
>  EXPORT_SYMBOL(crc32);
> -#endif
> +#endif /* __BAREBOX__ */
>  
>  /* No ones complement version. JFFS2 (and other things ?)
>   * don't use ones compliment in their CRC calculations.
> @@ -174,7 +195,7 @@ STATIC uint32_t crc32_no_comp(uint32_t crc, const void *_buf, unsigned int len)
>  #ifdef CONFIG_DYNAMIC_CRC_TABLE
>  	if (!crc_table)
>  		make_crc_table();
> -#endif
> +#endif /* CONFIG_DYNAMIC_CRC_TABLE */
>      while (len >= 8)
>      {
>        DO8(buf);
> @@ -186,6 +207,7 @@ STATIC uint32_t crc32_no_comp(uint32_t crc, const void *_buf, unsigned int len)
>  
>      return crc;
>  }
> +#endif /* __PBL__ */
>  
>  STATIC uint32_t crc32_be(uint32_t crc, const void *_buf, unsigned int len)
>  {
> 
> ---
> base-commit: bef38b18eeb5d2f1fac334fb8b831e47261e099c
> change-id: 20230829-crc32_in_pbl-4d824629d4e2
> 
> Best regards,
> -- 
> Johannes Zink <j.zink@pengutronix.de>
> 
> 
> 

-- 
Pengutronix e.K.                           |                             |
Steuerwalder Str. 21                       | http://www.pengutronix.de/  |
31137 Hildesheim, Germany                  | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |



^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH v2] crypto: crc32: make crc32 available in PBL
  2023-09-04  8:19 ` Sascha Hauer
@ 2023-09-05  7:57   ` Johannes Zink
  2023-09-05 12:26     ` Sascha Hauer
  0 siblings, 1 reply; 6+ messages in thread
From: Johannes Zink @ 2023-09-05  7:57 UTC (permalink / raw)
  To: Sascha Hauer; +Cc: patchwork-jzi, Barebox Mailing List

Hi Sascha,

On 9/4/23 10:19, Sascha Hauer wrote:
> On Tue, Aug 29, 2023 at 04:38:32PM +0200, Johannes Zink wrote:
>> crc32 may be required in PBL for checking data integrity. Add it to PBL
>> when CONFIG_CRC32 is enabled.
>>
>> To save some memory use a slower-but-smaller variant of the crc32 algorithm
>> in the PBL.
> 
> I've just sent an alternative patch series. Please check if that works
> for you.

What did you base the series on? Maybe I am holding it wrong, but it does 
neither apply on v2023.07.1 which I use in my customer project nor on 
upstream/next.

Johannes

> 
> Sascha
> 
> 
>>
>> Signed-off-by: Johannes Zink <j.zink@pengutronix.de>
>> ---
>> To: Barebox Mailing List <barebox@lists.infradead.org>
>> Cc: Johannes Zink <j.zink@pengutronix.de>
>> Cc: patchwork-jzi@pengutronix.de
>> ---
>>
>> Changes:
>>
>> v1->v2: Worked in Ahmad's review (thank you for reviewing my patch):
>>    - instead of using CRC32_EARLY, always add crc32.o to obj and pbl if
>>      CONFIG_CRC is set and rely on LTO to remove unused symbols instead
>>    - use a CRC32 implementation without a prepopulated Lookup Table,
>>      which trades in speed for code size in the PBL, analogously to what
>>      Sascha implemented in [1]
>>    - add hints on some of the #endifs for better readability of nested
>>      ifdef blocks
>>    - reword commit message to reflect on the changes made
>>
>>      [1] 2d13b856604b ("crc: Add PBL variant for crc_itu_t()")
>> ---
>>   crypto/Makefile |  2 +-
>>   crypto/crc32.c  | 34 ++++++++++++++++++++++++++++------
>>   2 files changed, 29 insertions(+), 7 deletions(-)
>>
>> diff --git a/crypto/Makefile b/crypto/Makefile
>> index 22035d4f69ee..4a1c7e9615b8 100644
>> --- a/crypto/Makefile
>> +++ b/crypto/Makefile
>> @@ -1,6 +1,6 @@
>>   # SPDX-License-Identifier: GPL-2.0-only
>>   
>> -obj-$(CONFIG_CRC32)	+= crc32.o
>> +obj-pbl-$(CONFIG_CRC32)	+= crc32.o
>>   obj-pbl-$(CONFIG_CRC_ITU_T)	+= crc-itu-t.o
>>   obj-$(CONFIG_CRC7)	+= crc7.o
>>   obj-$(CONFIG_DIGEST)	+= digest.o
>> diff --git a/crypto/crc32.c b/crypto/crc32.c
>> index 95cb2212db2b..2cf13144c1d2 100644
>> --- a/crypto/crc32.c
>> +++ b/crypto/crc32.c
>> @@ -22,7 +22,7 @@
>>   #define STATIC static inline
>>   #endif
>>   
>> -#ifdef CONFIG_DYNAMIC_CRC_TABLE
>> +#if defined(CONFIG_DYNAMIC_CRC_TABLE) && !defined(__PBL__)
>>   
>>   static uint32_t *crc_table;
>>   
>> @@ -73,7 +73,7 @@ static void make_crc_table(void)
>>       crc_table[n] = c;
>>     }
>>   }
>> -#else
>> +#elif !defined(__PBL__)
>>   /* ========================================================================
>>    * Table of CRC-32's of all single-byte values (made by make_crc_table)
>>    */
>> @@ -131,9 +131,30 @@ static const uint32_t crc_table[256] = {
>>     0x5d681b02L, 0x2a6f2b94L, 0xb40bbe37L, 0xc30c8ea1L, 0x5a05df1bL,
>>     0x2d02ef8dL
>>   };
>> -#endif
>> +#endif /* defined(CONFIG_DYNAMIC_CRC_TABLE) && !defined(__PBL__) */
>> +
>> +
>> +
>> +#ifdef __PBL__
>> +#define CRC32_POLY 0xEDB88320L
>> +STATIC uint32_t crc32(uint32_t crc, const void *_buf, unsigned int len)
>> +{
>> +	int i, j;
>> +	const unsigned char *buf = _buf;
>>   
>> +	crc = crc ^ 0xffffffffL;
>>   
>> +	for (i=0; i< len; i++) {
>> +		crc ^= buf[i];
>> +		for (j = 0; j < 8; j++) {
>> +			crc = (crc >> 1) ^ ((crc & 1) ? CRC32_POLY : 0);
>> +		}
>> +	}
>> +
>> +	crc = crc ^ 0xffffffffL;
>> +	return crc;
>> +}
>> +#else
>>   /* ========================================================================= */
>>   #define DO1(buf) crc = crc_table[((int)crc ^ (*buf++)) & 0xff] ^ (crc >> 8);
>>   #define DO2(buf)  DO1(buf); DO1(buf);
>> @@ -148,7 +169,7 @@ STATIC uint32_t crc32(uint32_t crc, const void *_buf, unsigned int len)
>>   #ifdef CONFIG_DYNAMIC_CRC_TABLE
>>   	if (!crc_table)
>>   		make_crc_table();
>> -#endif
>> +#endif /* CONFIG_DYNAMIC_CRC_TABLE */
>>       crc = crc ^ 0xffffffffL;
>>       while (len >= 8)
>>       {
>> @@ -162,7 +183,7 @@ STATIC uint32_t crc32(uint32_t crc, const void *_buf, unsigned int len)
>>   }
>>   #ifdef __BAREBOX__
>>   EXPORT_SYMBOL(crc32);
>> -#endif
>> +#endif /* __BAREBOX__ */
>>   
>>   /* No ones complement version. JFFS2 (and other things ?)
>>    * don't use ones compliment in their CRC calculations.
>> @@ -174,7 +195,7 @@ STATIC uint32_t crc32_no_comp(uint32_t crc, const void *_buf, unsigned int len)
>>   #ifdef CONFIG_DYNAMIC_CRC_TABLE
>>   	if (!crc_table)
>>   		make_crc_table();
>> -#endif
>> +#endif /* CONFIG_DYNAMIC_CRC_TABLE */
>>       while (len >= 8)
>>       {
>>         DO8(buf);
>> @@ -186,6 +207,7 @@ STATIC uint32_t crc32_no_comp(uint32_t crc, const void *_buf, unsigned int len)
>>   
>>       return crc;
>>   }
>> +#endif /* __PBL__ */
>>   
>>   STATIC uint32_t crc32_be(uint32_t crc, const void *_buf, unsigned int len)
>>   {
>>
>> ---
>> base-commit: bef38b18eeb5d2f1fac334fb8b831e47261e099c
>> change-id: 20230829-crc32_in_pbl-4d824629d4e2
>>
>> Best regards,
>> -- 
>> Johannes Zink <j.zink@pengutronix.de>
>>
>>
>>
> 

-- 
Pengutronix e.K.                | Johannes Zink                  |
Steuerwalder Str. 21            | https://www.pengutronix.de/    |
31137 Hildesheim, Germany       | Phone: +49-5121-206917-0       |
Amtsgericht Hildesheim, HRA 2686| Fax:   +49-5121-206917-5555    |




^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH v2] crypto: crc32: make crc32 available in PBL
  2023-09-05  7:57   ` Johannes Zink
@ 2023-09-05 12:26     ` Sascha Hauer
  2023-09-05 12:27       ` Johannes Zink
  0 siblings, 1 reply; 6+ messages in thread
From: Sascha Hauer @ 2023-09-05 12:26 UTC (permalink / raw)
  To: Johannes Zink; +Cc: patchwork-jzi, Barebox Mailing List

On Tue, Sep 05, 2023 at 09:57:21AM +0200, Johannes Zink wrote:
> Hi Sascha,
> 
> On 9/4/23 10:19, Sascha Hauer wrote:
> > On Tue, Aug 29, 2023 at 04:38:32PM +0200, Johannes Zink wrote:
> > > crc32 may be required in PBL for checking data integrity. Add it to PBL
> > > when CONFIG_CRC32 is enabled.
> > > 
> > > To save some memory use a slower-but-smaller variant of the crc32 algorithm
> > > in the PBL.
> > 
> > I've just sent an alternative patch series. Please check if that works
> > for you.
> 
> What did you base the series on? Maybe I am holding it wrong, but it does
> neither apply on v2023.07.1 which I use in my customer project nor on
> upstream/next.

It's already included in next. I had to re-order the patches though
because my original order caused some compile failures in between.

Sascha


-- 
Pengutronix e.K.                           |                             |
Steuerwalder Str. 21                       | http://www.pengutronix.de/  |
31137 Hildesheim, Germany                  | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |



^ permalink raw reply	[flat|nested] 6+ messages in thread

* Re: [PATCH v2] crypto: crc32: make crc32 available in PBL
  2023-09-05 12:26     ` Sascha Hauer
@ 2023-09-05 12:27       ` Johannes Zink
  0 siblings, 0 replies; 6+ messages in thread
From: Johannes Zink @ 2023-09-05 12:27 UTC (permalink / raw)
  To: Sascha Hauer; +Cc: patchwork-jzi, Barebox Mailing List

On 9/5/23 14:26, Sascha Hauer wrote:
> On Tue, Sep 05, 2023 at 09:57:21AM +0200, Johannes Zink wrote:
>> Hi Sascha,
>>
>> On 9/4/23 10:19, Sascha Hauer wrote:
>>> On Tue, Aug 29, 2023 at 04:38:32PM +0200, Johannes Zink wrote:
>>>> crc32 may be required in PBL for checking data integrity. Add it to PBL
>>>> when CONFIG_CRC32 is enabled.
>>>>
>>>> To save some memory use a slower-but-smaller variant of the crc32 algorithm
>>>> in the PBL.
>>>
>>> I've just sent an alternative patch series. Please check if that works
>>> for you.
>>
>> What did you base the series on? Maybe I am holding it wrong, but it does
>> neither apply on v2023.07.1 which I use in my customer project nor on
>> upstream/next.
> 
> It's already included in next. I had to re-order the patches though
> because my original order caused some compile failures in between.
> 

I see, thanks. I will pick and test the patches on my customer hardware in the 
next few days.

Johannes

> Sascha
> 
> 

-- 
Pengutronix e.K.                | Johannes Zink                  |
Steuerwalder Str. 21            | https://www.pengutronix.de/    |
31137 Hildesheim, Germany       | Phone: +49-5121-206917-0       |
Amtsgericht Hildesheim, HRA 2686| Fax:   +49-5121-206917-5555    |




^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2023-09-05 12:29 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-08-29 14:38 [PATCH v2] crypto: crc32: make crc32 available in PBL Johannes Zink
2023-08-29 15:00 ` Ahmad Fatoum
2023-09-04  8:19 ` Sascha Hauer
2023-09-05  7:57   ` Johannes Zink
2023-09-05 12:26     ` Sascha Hauer
2023-09-05 12:27       ` Johannes Zink

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox