Skip to contentSkip to conversation

crc32: Make crc32 build on non Linux too

PATCH New Joakim Tjernlund 4 participants 14 messages Nov 17, 2009 – Dec 2, 2009

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 crc32: Make crc32 build on non Linux too

The #include <asm/byteorder.h> is Linux specific. Use compiler.h and adjust the code to cope.

---

Mike, how does this work for you?

 lib_generic/crc32.c |   14 ++++++--------
 1 files changed, 6 insertions(+), 8 deletions(-)
lib_generic/crc32.c +6 -8
index 770ea38..468b397 100644
@@ -8,12 +8,10 @@
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
-#ifndef USE_HOSTCC	/* Shut down "ANSI does not permit..." warnings */
+#ifndef USE_HOSTCC
 #include <common.h>
-#else
-#include <stdint.h>
 #endif
-#include <asm/byteorder.h>
+#include <compiler.h>
 #include <u-boot/crc.h>
 
 #if defined(CONFIG_HW_WATCHDOG) || defined(CONFIG_WATCHDOG)
@@ -24,6 +22,8 @@
 #define local static
 #define ZEXPORT	/* empty */
 
+#define tole(x) cpu_to_le32(x)
+
 #ifdef DYNAMIC_CRC_TABLE
 
 local int crc_table_empty = 1;
@@ -54,7 +54,6 @@ local void make_crc_table OF((void));
   the information needed to generate CRC's on data a byte at a time for all
   combinations of CRC register values and incoming bytes.
 */
-#define tole(x) __cpu_to_le32(x)
 local void make_crc_table()
 {
   uint32_t c;
@@ -81,7 +80,6 @@ local void make_crc_table()
 /* ========================================================================
  * Table of CRC-32's of all single-byte values (made by make_crc_table)
  */
-#define tole(x) __constant_cpu_to_le32(x)
 
 local const uint32_t crc_table[256] = {
 tole(0x00000000L), tole(0x77073096L), tole(0xee0e612cL), tole(0x990951baL),
@@ -185,7 +183,7 @@ uint32_t ZEXPORT crc32_no_comp(uint32_t crc, const Bytef *buf, uInt len)
     if (crc_table_empty)
       make_crc_table();
 #endif
-    crc = __cpu_to_le32(crc);
+    crc = cpu_to_le32(crc);
     /* Align it */
     if (((long)b) & 3 && len) {
 	 uint8_t *p = (uint8_t *)b;
@@ -214,7 +212,7 @@ uint32_t ZEXPORT crc32_no_comp(uint32_t crc, const Bytef *buf, uInt len)
 	 } while (--len);
     }
 
-    return __le32_to_cpu(crc);
+    return le32_to_cpu(crc);
 }
 #undef DO_CRC
 
Signature · 1 line
1.6.4.4
#1 lore raw

PATCH 1/1 crc32: Make crc32 build on non Linux too

Quoted 7 lines · On Tuesday 17 November 2009 03:26:14 Joakim Tjernlund wrote:

The #include <asm/byteorder.h> is Linux specific. Use compiler.h and adjust the code to cope.

---

Mike, how does this work for you?

this is pretty much the change i committed locally

-mike
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: This is a digitally signed message part.
Url : http://lists.denx.de/pipermail/u-boot/attachments/20091117/566c1e77/attachment.pgp

#2 lore raw

PATCH 1/1 crc32: Make crc32 build on non Linux too

In reply to Joakim Tjernlund

Dear Joakim Tjernlund,

Quoted 7 lines · In message <[email protected]>…

The #include <asm/byteorder.h> is Linux specific. Use compiler.h and adjust the code to cope.

---

Mike, how does this work for you?

Could you please do me a favour and mark your patches in some way so everybody knows what these are?

Is this for inclusion into mainline (I guess not) or intended for testing - you should use something like "[PATCH RFC]" in the Subject then.

Also, when reposting new versions of this stuff then please mark the version in the Subject (like '[PATCH RFC v5]"), and provide a history of changes between the versions below the "---" line.

And PLEASE keep the In-reply-to: and References: headers so threading of the messages works.

Wolfgang Denk

Signature · 6 lines
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
Administration: An ingenious abstraction  in  politics,  designed  to
receive the kicks and cuffs due to the premier or president.
- Ambrose Bierce
#3 lore raw

PATCH 1/1 crc32: Make crc32 build on non Linux too

Wolfgang Denk <[email protected]> wrote on 17/11/2009 22:06:02:

Quoted 7 lines

Dear Joakim Tjernlund,

Quoted 3 lines · In message <[email protected]>…

The #include <asm/byteorder.h> is Linux specific. Use compiler.h and adjust the code to cope.

Quoted 7 lines
Quoted 4 lines

---

Mike, how does this work for you?

Could you please do me a favour and mark your patches in some way so everybody knows what these are?

This was a "reply" for Mikes comment on
[PATCH v2] crc32: Impl. linux optimized crc32() Would it been better for you if I just included the patch in a normal email reply?

Quoted 4 lines

Is this for inclusion into mainline (I guess not) or intended for testing - you should use something like "[PATCH RFC]" in the Subject then.

I think the crc32 patch(es) are ready to go in after the pending release.

Quoted 7 lines

Also, when reposting new versions of this stuff then please mark the version in the Subject (like '[PATCH RFC v5]"), and provide a history of changes between the versions below the "---" line.

And PLEASE keep the In-reply-to: and References: headers so threading of the messages works.

Since this was a new email sent with git, I don't In-reply-to: and References:
should be there?

Anyhow, I can send a V3 version with everything in it to sum it up?

Jocke

#4 lore raw

PATCH 1/1 crc32: Make crc32 build on non Linux too

Dear Joakim Tjernlund,

Quoted 3 lines · In message <OFFB1B0130.660000EA-ONC1257672.00315799-C1257672.003A0CCD@transmode…

This was a "reply" for Mikes comment on
[PATCH v2] crc32: Impl. linux optimized crc32()

If it was a reply to "[PATCH v2] crc32: ..." then why didn't the Subject read "Re: [PATCH v2] crc32: " ?

Quoted 2 lines

Would it been better for you if I just included the patch in a normal email reply?

Sorry, I fail to understand what you mean here.

Quoted 5 lines
Quoted 2 lines

And PLEASE keep the In-reply-to: and References: headers so threading of the messages works.

Since this was a new email sent with git, I don't In-reply-to: and References:
should be there?

You do reply to this thread. And "git send-email" asks you to provide the message Id's:

Message-ID to be used as In-Reply-To for the first email?

Please do not ignore this.

Quoted 1 line

Anyhow, I can send a V3 version with everything in it to sum it up?

Yes, please do, with clean commit message, s-o-b etc.

Best regards,

Wolfgang Denk

Signature · 6 lines
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
Our universe is a fragile house of atoms, held together by the mortar
of cause-and-effect. One magician would be two too many.
                        - Terry Pratchett, _The Dark Side of the Sun_
#5 lore raw

PATCH 1/1 crc32: Make crc32 build on non Linux too

Wolfgang Denk <[email protected]> wrote on 18/11/2009 23:45:56:

Quoted 9 lines

Dear Joakim Tjernlund,

In message <OFFB1B0130.660000EA-ONC1257672.00315799-C1257672.

Quoted 3 lines · 003A0CCD at transmode.se> you wrote:

This was a "reply" for Mikes comment on
[PATCH v2] crc32: Impl. linux optimized crc32()

Quoted 7 lines

If it was a reply to "[PATCH v2] crc32: ..." then why didn't the Subject read "Re: [PATCH v2] crc32: " ?

Quoted 2 lines

Would it been better for you if I just included the patch in a normal email reply?

Sorry, I fail to understand what you mean here.

Hit the reply button for Mike's mail and just include the patch last in the reply.

Quoted 7 lines
Quoted 6 lines
Quoted 3 lines

And PLEASE keep the In-reply-to: and References: headers so threading of the messages works.

Since this was a new email sent with git, I don't In-reply-to: and References:
should be there?

Quoted 6 lines

You do reply to this thread. And "git send-email" asks you to provide the message Id's:

Message-ID to be used as In-Reply-To for the first email?

Please do not ignore this.

I familiar on when it is appropriate to use In-reply-to:/References:
You mean even if a send a new mail(a git patch in this case) with a new subject, I should still add those two headers?

Getting those headers is another story, our crappy mail system don't let me see headers easily :(

Quoted 4 lines
Quoted 2 lines

Anyhow, I can send a V3 version with everything in it to sum it up?

Yes, please do, with clean commit message, s-o-b etc.

Will do.

#6 lore raw

PATCH 1/1 crc32: Make crc32 build on non Linux too

Quoted 11 lines · On Wednesday 18 November 2009 17:59:59 Joakim Tjernlund wrote:

Wolfgang Denk wrote on 18/11/2009 23:45:56:

Quoted 10 lines
Quoted 5 lines · Joakim Tjernlund
Quoted 2 lines

And PLEASE keep the In-reply-to: and References: headers so threading of the messages works.

Since this was a new email sent with git, I don't In-reply-to: and
References: should be there?

You do reply to this thread. And "git send-email" asks you to provide the message Id's:

Quoted 7 lines
Quoted 3 lines

Message-ID to be used as In-Reply-To for the first email?

Please do not ignore this.

I familiar on when it is appropriate to use In-reply-to:/References:
You mean even if a send a new mail(a git patch in this case) with a new subject, I should still add those two headers?

yes

Quoted 2 lines

Getting those headers is another story, our crappy mail system don't let me see headers easily :(

it should be relatively easy to obtain via gmane. but i'm pretty sure every MS based e-mail system has support for 'view source', and certainly every open source e-mail client does.
-mike
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: This is a digitally signed message part.
Url : http://lists.denx.de/pipermail/u-boot/attachments/20091118/114bb46a/attachment.pgp

#7 lore raw

PATCH 1/1 crc32: Make crc32 build on non Linux too

Mike Frysinger <[email protected]> wrote on 19/11/2009 00:45:16:

Quoted 15 lines · On Wednesday 18 November 2009 17:59:59 Joakim Tjernlund wrote:
Quoted 13 lines

Wolfgang Denk wrote on 18/11/2009 23:45:56:

Quoted 12 lines
Quoted 5 lines · Joakim Tjernlund
Quoted 2 lines

And PLEASE keep the In-reply-to: and References: headers so threading of the messages works.

Since this was a new email sent with git, I don't In-reply-to: and
References: should be there?

You do reply to this thread. And "git send-email" asks you to provide the message Id's:

Message-ID to be used as In-Reply-To for the first email?

Quoted 7 lines
Quoted 5 lines
Quoted 1 line

Please do not ignore this.

I familiar on when it is appropriate to use In-reply-to:/References:
You mean even if a send a new mail(a git patch in this case) with a new subject, I should still add those two headers?

yes

I see, thanks.

Quoted 5 lines
Quoted 3 lines

Getting those headers is another story, our crappy mail system don't let me see headers easily :(

it should be relatively easy to obtain via gmane. but i'm pretty sure every

I will not go to such length to get these.

Quoted 2 lines

MS based e-mail system has support for 'view source', and certainly every open source e-mail client does.

If I only had one of those :(
However, Lotus/Domino does have a view source too and sometimes it even works :)

#8 lore raw

PATCH v3 1/1 crc32: Impl. linux optimized crc32()

In reply to Wolfgang Denk

Ported over the more efficient linux crc32() function. A quick comparsion on ppc:
After changing the old crc32 to do 4 bytes in the inner loop to be able to compare with new version one can note:
- old inner loop has 61 insn, new has 19 insn.
- new crc32 does one 32 bit load of data to crc while the old does four 8 bits loads.
- size is bit bigger for the new crc32:
1392(old) 1416(new) of text. The is because the new version shares code with crc32_no_comp() instead of duplicating code.
- about 33% faster on ppc:
New > crc 0 0xfffffff -> 39 secs
Old > crc 0 0xfffffff -> 60 secs

---

V3: squashed "crc32: Make crc32 build on non Linux too"
into this patch to form one clean patch.

I think this patch is ready for general use.

 lib_generic/crc32.c |  200 ++++++++++++++++++++++++++++----------------------
 1 files changed, 112 insertions(+), 88 deletions(-)
lib_generic/crc32.c +112 -88
index b27048c..468b397 100644
@@ -8,11 +8,11 @@
  * For conditions of distribution and use, see copyright notice in zlib.h
  */
 
-#ifndef USE_HOSTCC	/* Shut down "ANSI does not permit..." warnings */
+#ifndef USE_HOSTCC
 #include <common.h>
-#else
-#include <stdint.h>
 #endif
+#include <compiler.h>
+#include <u-boot/crc.h>
 
 #if defined(CONFIG_HW_WATCHDOG) || defined(CONFIG_WATCHDOG)
 #include <watchdog.h>
@@ -22,6 +22,8 @@
 #define local static
 #define ZEXPORT	/* empty */
 
+#define tole(x) cpu_to_le32(x)
+
 #ifdef DYNAMIC_CRC_TABLE
 
 local int crc_table_empty = 1;
@@ -70,7 +72,7 @@ local void make_crc_table()
     c = (uLong)n;
     for (k = 0; k < 8; k++)
       c = c & 1 ? poly ^ (c >> 1) : c >> 1;
-    crc_table[n] = c;
+    crc_table[n] = tole(c);
   }
   crc_table_empty = 0;
 }
@@ -78,59 +80,72 @@ local void make_crc_table()
 /* ========================================================================
  * Table of CRC-32's of all single-byte values (made by make_crc_table)
  */
+
 local const uint32_t crc_table[256] = {
-  0x00000000L, 0x77073096L, 0xee0e612cL, 0x990951baL, 0x076dc419L,
-  0x706af48fL, 0xe963a535L, 0x9e6495a3L, 0x0edb8832L, 0x79dcb8a4L,
-  0xe0d5e91eL, 0x97d2d988L, 0x09b64c2bL, 0x7eb17cbdL, 0xe7b82d07L,
-  0x90bf1d91L, 0x1db71064L, 0x6ab020f2L, 0xf3b97148L, 0x84be41deL,
-  0x1adad47dL, 0x6ddde4ebL, 0xf4d4b551L, 0x83d385c7L, 0x136c9856L,
-  0x646ba8c0L, 0xfd62f97aL, 0x8a65c9ecL, 0x14015c4fL, 0x63066cd9L,
-  0xfa0f3d63L, 0x8d080df5L, 0x3b6e20c8L, 0x4c69105eL, 0xd56041e4L,
-  0xa2677172L, 0x3c03e4d1L, 0x4b04d447L, 0xd20d85fdL, 0xa50ab56bL,
-  0x35b5a8faL, 0x42b2986cL, 0xdbbbc9d6L, 0xacbcf940L, 0x32d86ce3L,
-  0x45df5c75L, 0xdcd60dcfL, 0xabd13d59L, 0x26d930acL, 0x51de003aL,
-  0xc8d75180L, 0xbfd06116L, 0x21b4f4b5L, 0x56b3c423L, 0xcfba9599L,
-  0xb8bda50fL, 0x2802b89eL, 0x5f058808L, 0xc60cd9b2L, 0xb10be924L,
-  0x2f6f7c87L, 0x58684c11L, 0xc1611dabL, 0xb6662d3dL, 0x76dc4190L,
-  0x01db7106L, 0x98d220bcL, 0xefd5102aL, 0x71b18589L, 0x06b6b51fL,
-  0x9fbfe4a5L, 0xe8b8d433L, 0x7807c9a2L, 0x0f00f934L, 0x9609a88eL,
-  0xe10e9818L, 0x7f6a0dbbL, 0x086d3d2dL, 0x91646c97L, 0xe6635c01L,
-  0x6b6b51f4L, 0x1c6c6162L, 0x856530d8L, 0xf262004eL, 0x6c0695edL,
-  0x1b01a57bL, 0x8208f4c1L, 0xf50fc457L, 0x65b0d9c6L, 0x12b7e950L,
-  0x8bbeb8eaL, 0xfcb9887cL, 0x62dd1ddfL, 0x15da2d49L, 0x8cd37cf3L,
-  0xfbd44c65L, 0x4db26158L, 0x3ab551ceL, 0xa3bc0074L, 0xd4bb30e2L,
-  0x4adfa541L, 0x3dd895d7L, 0xa4d1c46dL, 0xd3d6f4fbL, 0x4369e96aL,
-  0x346ed9fcL, 0xad678846L, 0xda60b8d0L, 0x44042d73L, 0x33031de5L,
-  0xaa0a4c5fL, 0xdd0d7cc9L, 0x5005713cL, 0x270241aaL, 0xbe0b1010L,
-  0xc90c2086L, 0x5768b525L, 0x206f85b3L, 0xb966d409L, 0xce61e49fL,
-  0x5edef90eL, 0x29d9c998L, 0xb0d09822L, 0xc7d7a8b4L, 0x59b33d17L,
-  0x2eb40d81L, 0xb7bd5c3bL, 0xc0ba6cadL, 0xedb88320L, 0x9abfb3b6L,
-  0x03b6e20cL, 0x74b1d29aL, 0xead54739L, 0x9dd277afL, 0x04db2615L,
-  0x73dc1683L, 0xe3630b12L, 0x94643b84L, 0x0d6d6a3eL, 0x7a6a5aa8L,
-  0xe40ecf0bL, 0x9309ff9dL, 0x0a00ae27L, 0x7d079eb1L, 0xf00f9344L,
-  0x8708a3d2L, 0x1e01f268L, 0x6906c2feL, 0xf762575dL, 0x806567cbL,
-  0x196c3671L, 0x6e6b06e7L, 0xfed41b76L, 0x89d32be0L, 0x10da7a5aL,
-  0x67dd4accL, 0xf9b9df6fL, 0x8ebeeff9L, 0x17b7be43L, 0x60b08ed5L,
-  0xd6d6a3e8L, 0xa1d1937eL, 0x38d8c2c4L, 0x4fdff252L, 0xd1bb67f1L,
-  0xa6bc5767L, 0x3fb506ddL, 0x48b2364bL, 0xd80d2bdaL, 0xaf0a1b4cL,
-  0x36034af6L, 0x41047a60L, 0xdf60efc3L, 0xa867df55L, 0x316e8eefL,
-  0x4669be79L, 0xcb61b38cL, 0xbc66831aL, 0x256fd2a0L, 0x5268e236L,
-  0xcc0c7795L, 0xbb0b4703L, 0x220216b9L, 0x5505262fL, 0xc5ba3bbeL,
-  0xb2bd0b28L, 0x2bb45a92L, 0x5cb36a04L, 0xc2d7ffa7L, 0xb5d0cf31L,
-  0x2cd99e8bL, 0x5bdeae1dL, 0x9b64c2b0L, 0xec63f226L, 0x756aa39cL,
-  0x026d930aL, 0x9c0906a9L, 0xeb0e363fL, 0x72076785L, 0x05005713L,
-  0x95bf4a82L, 0xe2b87a14L, 0x7bb12baeL, 0x0cb61b38L, 0x92d28e9bL,
-  0xe5d5be0dL, 0x7cdcefb7L, 0x0bdbdf21L, 0x86d3d2d4L, 0xf1d4e242L,
-  0x68ddb3f8L, 0x1fda836eL, 0x81be16cdL, 0xf6b9265bL, 0x6fb077e1L,
-  0x18b74777L, 0x88085ae6L, 0xff0f6a70L, 0x66063bcaL, 0x11010b5cL,
-  0x8f659effL, 0xf862ae69L, 0x616bffd3L, 0x166ccf45L, 0xa00ae278L,
-  0xd70dd2eeL, 0x4e048354L, 0x3903b3c2L, 0xa7672661L, 0xd06016f7L,
-  0x4969474dL, 0x3e6e77dbL, 0xaed16a4aL, 0xd9d65adcL, 0x40df0b66L,
-  0x37d83bf0L, 0xa9bcae53L, 0xdebb9ec5L, 0x47b2cf7fL, 0x30b5ffe9L,
-  0xbdbdf21cL, 0xcabac28aL, 0x53b39330L, 0x24b4a3a6L, 0xbad03605L,
-  0xcdd70693L, 0x54de5729L, 0x23d967bfL, 0xb3667a2eL, 0xc4614ab8L,
-  0x5d681b02L, 0x2a6f2b94L, 0xb40bbe37L, 0xc30c8ea1L, 0x5a05df1bL,
-  0x2d02ef8dL
+tole(0x00000000L), tole(0x77073096L), tole(0xee0e612cL), tole(0x990951baL),
+tole(0x076dc419L), tole(0x706af48fL), tole(0xe963a535L), tole(0x9e6495a3L),
+tole(0x0edb8832L), tole(0x79dcb8a4L), tole(0xe0d5e91eL), tole(0x97d2d988L),
+tole(0x09b64c2bL), tole(0x7eb17cbdL), tole(0xe7b82d07L), tole(0x90bf1d91L),
+tole(0x1db71064L), tole(0x6ab020f2L), tole(0xf3b97148L), tole(0x84be41deL),
+tole(0x1adad47dL), tole(0x6ddde4ebL), tole(0xf4d4b551L), tole(0x83d385c7L),
+tole(0x136c9856L), tole(0x646ba8c0L), tole(0xfd62f97aL), tole(0x8a65c9ecL),
+tole(0x14015c4fL), tole(0x63066cd9L), tole(0xfa0f3d63L), tole(0x8d080df5L),
+tole(0x3b6e20c8L), tole(0x4c69105eL), tole(0xd56041e4L), tole(0xa2677172L),
+tole(0x3c03e4d1L), tole(0x4b04d447L), tole(0xd20d85fdL), tole(0xa50ab56bL),
+tole(0x35b5a8faL), tole(0x42b2986cL), tole(0xdbbbc9d6L), tole(0xacbcf940L),
+tole(0x32d86ce3L), tole(0x45df5c75L), tole(0xdcd60dcfL), tole(0xabd13d59L),
+tole(0x26d930acL), tole(0x51de003aL), tole(0xc8d75180L), tole(0xbfd06116L),
+tole(0x21b4f4b5L), tole(0x56b3c423L), tole(0xcfba9599L), tole(0xb8bda50fL),
+tole(0x2802b89eL), tole(0x5f058808L), tole(0xc60cd9b2L), tole(0xb10be924L),
+tole(0x2f6f7c87L), tole(0x58684c11L), tole(0xc1611dabL), tole(0xb6662d3dL),
+tole(0x76dc4190L), tole(0x01db7106L), tole(0x98d220bcL), tole(0xefd5102aL),
+tole(0x71b18589L), tole(0x06b6b51fL), tole(0x9fbfe4a5L), tole(0xe8b8d433L),
+tole(0x7807c9a2L), tole(0x0f00f934L), tole(0x9609a88eL), tole(0xe10e9818L),
+tole(0x7f6a0dbbL), tole(0x086d3d2dL), tole(0x91646c97L), tole(0xe6635c01L),
+tole(0x6b6b51f4L), tole(0x1c6c6162L), tole(0x856530d8L), tole(0xf262004eL),
+tole(0x6c0695edL), tole(0x1b01a57bL), tole(0x8208f4c1L), tole(0xf50fc457L),
+tole(0x65b0d9c6L), tole(0x12b7e950L), tole(0x8bbeb8eaL), tole(0xfcb9887cL),
+tole(0x62dd1ddfL), tole(0x15da2d49L), tole(0x8cd37cf3L), tole(0xfbd44c65L),
+tole(0x4db26158L), tole(0x3ab551ceL), tole(0xa3bc0074L), tole(0xd4bb30e2L),
+tole(0x4adfa541L), tole(0x3dd895d7L), tole(0xa4d1c46dL), tole(0xd3d6f4fbL),
+tole(0x4369e96aL), tole(0x346ed9fcL), tole(0xad678846L), tole(0xda60b8d0L),
+tole(0x44042d73L), tole(0x33031de5L), tole(0xaa0a4c5fL), tole(0xdd0d7cc9L),
+tole(0x5005713cL), tole(0x270241aaL), tole(0xbe0b1010L), tole(0xc90c2086L),
+tole(0x5768b525L), tole(0x206f85b3L), tole(0xb966d409L), tole(0xce61e49fL),
+tole(0x5edef90eL), tole(0x29d9c998L), tole(0xb0d09822L), tole(0xc7d7a8b4L),
+tole(0x59b33d17L), tole(0x2eb40d81L), tole(0xb7bd5c3bL), tole(0xc0ba6cadL),
+tole(0xedb88320L), tole(0x9abfb3b6L), tole(0x03b6e20cL), tole(0x74b1d29aL),
+tole(0xead54739L), tole(0x9dd277afL), tole(0x04db2615L), tole(0x73dc1683L),
+tole(0xe3630b12L), tole(0x94643b84L), tole(0x0d6d6a3eL), tole(0x7a6a5aa8L),
+tole(0xe40ecf0bL), tole(0x9309ff9dL), tole(0x0a00ae27L), tole(0x7d079eb1L),
+tole(0xf00f9344L), tole(0x8708a3d2L), tole(0x1e01f268L), tole(0x6906c2feL),
+tole(0xf762575dL), tole(0x806567cbL), tole(0x196c3671L), tole(0x6e6b06e7L),
+tole(0xfed41b76L), tole(0x89d32be0L), tole(0x10da7a5aL), tole(0x67dd4accL),
+tole(0xf9b9df6fL), tole(0x8ebeeff9L), tole(0x17b7be43L), tole(0x60b08ed5L),
+tole(0xd6d6a3e8L), tole(0xa1d1937eL), tole(0x38d8c2c4L), tole(0x4fdff252L),
+tole(0xd1bb67f1L), tole(0xa6bc5767L), tole(0x3fb506ddL), tole(0x48b2364bL),
+tole(0xd80d2bdaL), tole(0xaf0a1b4cL), tole(0x36034af6L), tole(0x41047a60L),
+tole(0xdf60efc3L), tole(0xa867df55L), tole(0x316e8eefL), tole(0x4669be79L),
+tole(0xcb61b38cL), tole(0xbc66831aL), tole(0x256fd2a0L), tole(0x5268e236L),
+tole(0xcc0c7795L), tole(0xbb0b4703L), tole(0x220216b9L), tole(0x5505262fL),
+tole(0xc5ba3bbeL), tole(0xb2bd0b28L), tole(0x2bb45a92L), tole(0x5cb36a04L),
+tole(0xc2d7ffa7L), tole(0xb5d0cf31L), tole(0x2cd99e8bL), tole(0x5bdeae1dL),
+tole(0x9b64c2b0L), tole(0xec63f226L), tole(0x756aa39cL), tole(0x026d930aL),
+tole(0x9c0906a9L), tole(0xeb0e363fL), tole(0x72076785L), tole(0x05005713L),
+tole(0x95bf4a82L), tole(0xe2b87a14L), tole(0x7bb12baeL), tole(0x0cb61b38L),
+tole(0x92d28e9bL), tole(0xe5d5be0dL), tole(0x7cdcefb7L), tole(0x0bdbdf21L),
+tole(0x86d3d2d4L), tole(0xf1d4e242L), tole(0x68ddb3f8L), tole(0x1fda836eL),
+tole(0x81be16cdL), tole(0xf6b9265bL), tole(0x6fb077e1L), tole(0x18b74777L),
+tole(0x88085ae6L), tole(0xff0f6a70L), tole(0x66063bcaL), tole(0x11010b5cL),
+tole(0x8f659effL), tole(0xf862ae69L), tole(0x616bffd3L), tole(0x166ccf45L),
+tole(0xa00ae278L), tole(0xd70dd2eeL), tole(0x4e048354L), tole(0x3903b3c2L),
+tole(0xa7672661L), tole(0xd06016f7L), tole(0x4969474dL), tole(0x3e6e77dbL),
+tole(0xaed16a4aL), tole(0xd9d65adcL), tole(0x40df0b66L), tole(0x37d83bf0L),
+tole(0xa9bcae53L), tole(0xdebb9ec5L), tole(0x47b2cf7fL), tole(0x30b5ffe9L),
+tole(0xbdbdf21cL), tole(0xcabac28aL), tole(0x53b39330L), tole(0x24b4a3a6L),
+tole(0xbad03605L), tole(0xcdd70693L), tole(0x54de5729L), tole(0x23d967bfL),
+tole(0xb3667a2eL), tole(0xc4614ab8L), tole(0x5d681b02L), tole(0x2a6f2b94L),
+tole(0xb40bbe37L), tole(0xc30c8ea1L), tole(0x5a05df1bL), tole(0x2d02ef8dL)
 };
 #endif
 
@@ -148,54 +163,63 @@ const uint32_t * ZEXPORT get_crc_table()
 #endif
 
 /* ========================================================================= */
-#define DO1(buf) crc = crc_table[((int)crc ^ (*buf++)) & 0xff] ^ (crc >> 8);
-#define DO2(buf)  DO1(buf); DO1(buf);
-#define DO4(buf)  DO2(buf); DO2(buf);
-#define DO8(buf)  DO4(buf); DO4(buf);
+# ifdef __LITTLE_ENDIAN
+#  define DO_CRC(x) crc = tab[(crc ^ (x)) & 255] ^ (crc >> 8)
+# else
+#  define DO_CRC(x) crc = tab[((crc >> 24) ^ (x)) & 255] ^ (crc << 8)
+# endif
 
 /* ========================================================================= */
-uint32_t ZEXPORT crc32 (uint32_t crc, const Bytef *buf, uInt len)
-{
-#ifdef DYNAMIC_CRC_TABLE
-    if (crc_table_empty)
-      make_crc_table();
-#endif
-    crc = crc ^ 0xffffffffL;
-    while (len >= 8)
-    {
-      DO8(buf);
-      len -= 8;
-    }
-    if (len) do {
-      DO1(buf);
-    } while (--len);
-    return crc ^ 0xffffffffL;
-}
-
-#if defined(CONFIG_CMD_JFFS2) || defined(CONFIG_CMD_NAND)
 
 /* No ones complement version. JFFS2 (and other things ?)
  * don't use ones compliment in their CRC calculations.
  */
 uint32_t ZEXPORT crc32_no_comp(uint32_t crc, const Bytef *buf, uInt len)
 {
+    const uint32_t *tab = crc_table;
+    const uint32_t *b =(const uint32_t *)buf;
+    size_t rem_len;
 #ifdef DYNAMIC_CRC_TABLE
     if (crc_table_empty)
       make_crc_table();
 #endif
-    while (len >= 8)
-    {
-      DO8(buf);
-      len -= 8;
+    crc = cpu_to_le32(crc);
+    /* Align it */
+    if (((long)b) & 3 && len) {
+	 uint8_t *p = (uint8_t *)b;
+	 do {
+	      DO_CRC(*p++);
+	 } while ((--len) && ((long)p)&3);
+	 b = (uint32_t *)p;
+    }
+
+    rem_len = len & 3;
+    len = len >> 2;
+    for (--b; len; --len) {
+	 /* load data 32 bits wide, xor data 32 bits wide. */
+	 crc ^= *++b; /* use pre increment for speed */
+	 DO_CRC(0);
+	 DO_CRC(0);
+	 DO_CRC(0);
+	 DO_CRC(0);
+    }
+    len = rem_len;
+    /* And the last few bytes */
+    if (len) {
+	 uint8_t *p = (uint8_t *)(b + 1) - 1;
+	 do {
+	      DO_CRC(*++p); /* use pre increment for speed */
+	 } while (--len);
     }
-    if (len) do {
-      DO1(buf);
-    } while (--len);
 
-    return crc;
+    return le32_to_cpu(crc);
 }
+#undef DO_CRC
 
-#endif
+uint32_t ZEXPORT crc32 (uint32_t crc, const Bytef *p, uInt len)
+{
+     return crc32_no_comp(crc ^ 0xffffffffL, p, len) ^ 0xffffffffL;
+}
 
 /*
  * Calculate the crc32 checksum triggering the watchdog every 'chunk_sz' bytes
Signature · 1 line
1.6.4.4
#9 lore raw

PATCH v3 1/1 crc32: Impl. linux optimized crc32()

Hi,

lib_generic/crc32.c line 26

Quoted 3 lines · Joakim Tjernlund
lib_generic/crc32.c +2 -0
+#define tole(x) cpu_to_le32(x)
+

This define is only used once and at the constants table, after that, cpu_to_le32 is called directly. Perhaps this could be removed and cpu_to_le32 used instead, or change everything at that file to tole(). Two "symbols" to do the same thing doesn't seem right.

Personally I think cpu_to_le32 is more readable, at the expense of a few more keyboard strokes.

Just my 2c :)

Resubmiting is a lot of trouble (at least to a git newbie like me), so I think it could be a second patch to fix that anyway, assuming you decide it's broken.

Kind Regards,
Thiago A. Correa

#10 lore raw

PATCH v3 1/1 crc32: Impl. linux optimized crc32()

Thiago A. Corr?a <[email protected]> wrote on 19/11/2009 16:41:05:

Quoted 14 lines

Hi,

lib_generic/crc32.c line 26

Quoted 3 lines · Joakim Tjernlund
lib_generic/crc32.c +2 -0
+#define tole(x) cpu_to_le32(x)
+

This define is only used once and at the constants table, after that, cpu_to_le32 is called directly. Perhaps this could be removed and cpu_to_le32 used instead, or change everything at that file to tole(). Two "symbols" to do the same thing doesn't seem right.

Quoted 2 lines

Personally I think cpu_to_le32 is more readable, at the expense of a few more keyboard strokes.

The whole point of the tole define is to make the table more readable. Try and replace tole in the table and then look at it.

Jocke

#11 lore raw

PATCH v3 1/1 crc32: Impl. linux optimized crc32()

Hi,

2009/11/19 Joakim Tjernlund <[email protected]>:

Quoted 5 lines

The whole point of the tole define is to make the table more readable. Try and replace tole in the table and then look at it.

Ah, ok. Then I think this one at make_crc_table should use the __cpu_to_le32() as tole only porpuse is a more readable table, then you could lose the first tole define as __cpu_to_le32 and keep only the one defined as __constant_cpu_to_le32. Also, I don't know how you are not getting a warning from the preprocessor about the redefined tole macro:

+#define tole(x) __cpu_to_le32(x)
local void make_crc_table()
{
uint32_t c;
@@ -70,7 +73,7 @@ local void make_crc_table()

    c = (uLong)n;
    for (k = 0; k < 8; k++)
      c = c & 1 ? poly ^ (c >> 1) : c >> 1;

- crc_table[n] = c;
+ crc_table[n] = tole(c);
}
crc_table_empty = 0;
}
@@ -78,59 +81,73 @@ local void make_crc_table()
/* ========================================================================
* Table of CRC-32's of all single-byte values (made by make_crc_table) */
+#define tole(x) __constant_cpu_to_le32(x)
+

I'm not looking at the whole file but since you introduced in this diff a +#define tole, and there is no +#undef tole, you should be getting warnings. Except for possibly the warning, my suggestion was very cosmetic. On the other hand, I didn't notice the redefinition, that usually causes a lot of maintenance trouble latter on.

Kind Regards,
Thiago A. Correa

#12 lore raw

PATCH v3 1/1 crc32: Impl. linux optimized crc32()

Quoted 11 lines · On Thursday 19 November 2009 13:22:43 Thiago A. Corr?a wrote:
Quoted 3 lines · 2009/11/19 Joakim Tjernlund wrote:

The whole point of the tole define is to make the table more readable. Try and replace tole in the table and then look at it.

Ah, ok. Then I think this one at make_crc_table should use the __cpu_to_le32() as tole only porpuse is a more readable table, then you could lose the first tole define as __cpu_to_le32 and keep only the one defined as __constant_cpu_to_le32. Also, I don't know how you are not getting a warning from the preprocessor about the redefined tole macro:

except then you revert the fix i posted and restore the bug i referred to earlier about cross-compiling.

since the define needs to exist to keep the table understandable, keeping the two table code paths similar makes sense too.
-mike
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: This is a digitally signed message part.
Url : http://lists.denx.de/pipermail/u-boot/attachments/20091119/a99d2192/attachment.pgp

#13 lore raw

PATCH v3 1/1 crc32: Impl. linux optimized crc32()

In reply to Joakim Tjernlund

Dear Joakim Tjernlund,

Quoted 17 lines · In message <[email protected]>…

Ported over the more efficient linux crc32() function. A quick comparsion on ppc:
After changing the old crc32 to do 4 bytes in the inner loop to be able to compare with new version one can note:
- old inner loop has 61 insn, new has 19 insn.
- new crc32 does one 32 bit load of data to crc while the old does four 8 bits loads.
- size is bit bigger for the new crc32:
1392(old) 1416(new) of text. The is because the new version shares code with crc32_no_comp() instead of duplicating code.
- about 33% faster on ppc:
New > crc 0 0xfffffff -> 39 secs
Old > crc 0 0xfffffff -> 60 secs

---

Quoted 7 lines

V3: squashed "crc32: Make crc32 build on non Linux too"
into this patch to form one clean patch.

I think this patch is ready for general use.

 lib_generic/crc32.c |  200 ++++++++++++++++++++++++++++----------------------
 1 files changed, 112 insertions(+), 88 deletions(-)

Applied to "next", thanks.

Best regards,

Wolfgang Denk

Signature · 5 lines
DENX Software Engineering GmbH,     MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de
There are no data that cannot be plotted on a straight  line  if  the
axis are chosen correctly.