search for: char_offset

...)) static inline __m128i sse_maddubs_epi16(__m128i a, __m128i b) { } + +/* + a simple 32 bit checksum that can be updated from either end + (inspired by Mark Adler's Adler-32 checksum) + */ +/* + Original loop per 4 bytes: + s2 += 4*(s1 + buf[i]) + 3*buf[i+1] + 2*buf[i+2] + buf[i+3] + 10*CHAR_OFFSET; + s1 += buf[i] + buf[i+1] + buf[i+2] + buf[i+3] + 4*CHAR_OFFSET; + + SSE2/SSSE3 loop per 32 bytes: + int16 t1[8]; + int16 t2[8]; + for (int j = 0; j < 8; j++) { + t1[j] = buf[j*4 + i] + buf[j*4 + i+1] + buf[j*4 + i+2] + buf[j*4 + i+3]; + t2[j] = 4*buf[j*4 + i] + 3*buf[j*4...

...even_epi8(a, b), + sse_mulu_odd_epi8(a, b) + ); +#endif +} + +/* + a simple 32 bit checksum that can be updated from either end + (inspired by Mark Adler's Adler-32 checksum) + */ +/* + Original loop per 4 bytes: + s2 += 4*(s1 + buf[i]) + 3*buf[i+1] + 2*buf[i+2] + buf[i+3] + 10*CHAR_OFFSET; + s1 += buf[i] + buf[i+1] + buf[i+2] + buf[i+3] + 4*CHAR_OFFSET; + + SSE2/SSSE3 loop per 32 bytes: + int16 t1[8]; + int16 t2[8]; + for (int j = 0; j < 8; j++) { + t1[j] = buf[j*4 + i] + buf[j*4 + i+1] + buf[j*4 + i+2] + buf[j*4 + i+3]; + t2[j] = 4*buf[j*4 + i] + 3*buf[j*4...

...); > +#endif > +} > + > +/* > + a simple 32 bit checksum that can be updated from either end > + (inspired by Mark Adler's Adler-32 checksum) > + */ > +/* > + Original loop per 4 bytes: > + s2 += 4*(s1 + buf[i]) + 3*buf[i+1] + 2*buf[i+2] + buf[i+3] + > 10*CHAR_OFFSET; > + s1 += buf[i] + buf[i+1] + buf[i+2] + buf[i+3] + 4*CHAR_OFFSET; > + > + SSE2/SSSE3 loop per 32 bytes: > + int16 t1[8]; > + int16 t2[8]; > + for (int j = 0; j < 8; j++) { > + t1[j] = buf[j*4 + i] + buf[j*4 + i+1] + buf[j*4 + i+2] + buf[j*4 + > i+3]; &...

...m128i a, __m128i b) { } > + > +/* > + a simple 32 bit checksum that can be updated from either end > + (inspired by Mark Adler's Adler-32 checksum) > + */ > +/* > + Original loop per 4 bytes: > + s2 += 4*(s1 + buf[i]) + 3*buf[i+1] + 2*buf[i+2] + buf[i+3] + > 10*CHAR_OFFSET; > + s1 += buf[i] + buf[i+1] + buf[i+2] + buf[i+3] + 4*CHAR_OFFSET; > + > + SSE2/SSSE3 loop per 32 bytes: > + int16 t1[8]; > + int16 t2[8]; > + for (int j = 0; j < 8; j++) { > + t1[j] = buf[j*4 + i] + buf[j*4 + i+1] + buf[j*4 + i+2] + buf[j*4 + i+3]; > +...

...gt;> +/* >> + a simple 32 bit checksum that can be updated from either end >> + (inspired by Mark Adler's Adler-32 checksum) >> + */ >> +/* >> + Original loop per 4 bytes: >> + s2 += 4*(s1 + buf[i]) + 3*buf[i+1] + 2*buf[i+2] + buf[i+3] + >> 10*CHAR_OFFSET; >> + s1 += buf[i] + buf[i+1] + buf[i+2] + buf[i+3] + 4*CHAR_OFFSET; >> + >> + SSE2/SSSE3 loop per 32 bytes: >> + int16 t1[8]; >> + int16 t2[8]; >> + for (int j = 0; j < 8; j++) { >> + t1[j] = buf[j*4 + i] + buf[j*4 + i+1] + buf[j*4 + i...

What do you base this on? Per https://gcc.gnu.org/onlinedocs/gcc/x86-Options.html : "For the x86-32 compiler, you must use -march=cpu-type, -msse or -msse2 switches to enable SSE extensions and make this option effective. For the x86-64 compiler, these extensions are enabled by default." That reads to me like we're fine for SSE2. As stated in my comments, SSSE3 support must be

Hi all, I am writing a xdelta-like application as a personal experiment and am busy implementing the rsync protocol, so far so good. I am using C++ templates and creating the algorithms so that operate on any stream, array, etc. through iterators. All seems well except that I am getting a lot of false hits with the weak checksum. When generating checksums of blocksize 1024 on the RedHat 7.1

Hi, I'm curious if anybody knows the exact reason why the weak checksum calculation is slightly different to the standard adler-32 checksum as seen for example here http://en.wikipedia.org/wiki/Adler-32 ? Thanks Julian -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.samba.org/pipermail/rsync/attachments/20090924/c4ed210b/attachment.html>

Attached is a patch that makes window strides constant when files are walked with a constant block size. In these cases, it completely avoids all memmoves. In my simple local test of rsyncing 57MB of 10 local files, memmoved bytes went from 18MB to zero. I haven't tested this for a big variety of file cases. I think that this will always reduce the memmoves involved with walking a large

...uf,offset,s->n); + k = buf->m_len; + sum = get_checksum1(buf->m_ptr, k); + s1 = sum & 0xFFFF; s2 = sum >> 16; matches++; break; } null_tag: /* Trim off the first byte from the checksum */ - map = (schar *)map_ptr(buf,offset,k+1); - s1 -= map[0] + CHAR_OFFSET; - s2 -= k * (map[0]+CHAR_OFFSET); + map_ptr(buf,offset,k+1); + if (buf->m_len==0) break; /* encountered EOF */ + s1 -= buf->m_ptr[0] + CHAR_OFFSET; + s2 -= k * (buf->m_ptr[0]+CHAR_OFFSET); /* Add on the next byte (if there is one) to the checksum */ - if (k < (len-offset))...

...are identical on +both systems\&. .IP .IP "\fB--ignore-errors\fP" Tells --delete to go ahead and delete files Index: rsync.h --- rsync.h 15 Jan 2004 07:42:27 -0000 1.173 +++ rsync.h 17 Jan 2004 05:04:56 -0000 @@ -39,6 +39,7 @@ incompatible with older versions :-( */ #define CHAR_OFFSET 0 +/* These flags are used during the flist transfer. */ #define FLAG_DELETE (1<<0) #define SAME_MODE (1<<1) @@ -54,9 +55,14 @@ #define HAS_INODE_DATA (1<<9) #define SAME_DEV (1<<10) -/* What flags are relevant after the transfer of the flist is complete? */ +/* Wh...