Skip to content

Add SSE2 dec_loop.c #46

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 1 commit into from
Closed

Add SSE2 dec_loop.c #46

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
133 changes: 133 additions & 0 deletions lib/arch/sse2/dec_loop.c
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,133 @@
// If we have SSE2 support, pick off 16 bytes at a time for as long as we can,
// but make sure that we quit before seeing any == markers at the end of the
// string. Also, because we write four zeroes at the end of the output, ensure
// that there are at least 6 valid bytes of input data remaining to close the
// gap. 16 + 2 + 6 = 24 bytes:
while (srclen >= 24)
{
const __m128i digit_top = _mm_set1_epi32( 0x46464646 ); // 0x46 = 0x7F - '9'
const __m128i lowercase_top = _mm_set1_epi32( 0x05050505 ); // 0x05 = 0x7F - 'z'
const __m128i uppercase_top = _mm_set1_epi32( 0x25252525 ); // 0x25 = 0x7F - 'Z'
const __m128i digit_bottom = _mm_set1_epi32( 0x76767676 ); // 0x76 = 0x7F - 9 ('9'-'0')
const __m128i alpha_bottom = _mm_set1_epi32( 0x66666666 ); // 0x66 = 0x7F - 25 ('Z'-'A') ('z'-'a')
const __m128i num_letters = _mm_set1_epi32( 0x1A1A1A1A ); // 0x1A = 26
const __m128i detect_plus = _mm_set1_epi32( 0xFBFBFBFB ); // 0xFB = plus sign value after '0-9' roll down
const __m128i adj_plus = _mm_set1_epi32( 0xD4D4D4D4 ); // 0xD4 = plus sign shift after 'A-Z' roll down
const __m128i detect_slash = _mm_set1_epi32( 0xFFFFFFFF ); // 0xFF = after 0-9 roll down

__m128i m0, m1, m2, m3, m4;

// Load string:
m0 = _mm_loadu_si128((__m128i *)c);

// Map input bytes to 6-bit values:
m1 = _mm_add_epi8(m0, digit_top); // "0-9" roll up
m2 = _mm_add_epi8(m0, lowercase_top); // "a-z" roll up
m1 = _mm_subs_epi8(m1, digit_bottom); // "0-9" roll down
m2 = _mm_subs_epi8(m2, alpha_bottom); // "a-z" roll down
m4 = _mm_cmpeq_epi8(m1, detect_slash); // match '/' ( foward slash )
m3 = _mm_cmpeq_epi8(m1, detect_plus); // match '+' ( plus sign )
m1 = _mm_adds_epu8(m1, num_letters); // ('0'=26 thru '9'=35) else signed
m4 = _mm_subs_epu8(m4, uppercase_top); // 0xDA = 0xFF - 0x25
m0 = _mm_add_epi8(m0, uppercase_top); // "A-Z" roll up
m1 = _mm_xor_si128(m1, m4); // ( '/' = 37 ) = FF ^ DA
m3 = _mm_and_si128(m3, adj_plus); // = 0xFF & 0xD4
m0 = _mm_subs_epi8(m0, alpha_bottom); // "A-Z" roll down
m1 = _mm_min_epu8(m1, m2); // merge
m0 = _mm_xor_si128(m0, m3); // ( '+' = 62 )
m1 = _mm_adds_epu8(m1, num_letters); // ('a'=26 thru '9'=61, '/'=63) else signed
m0 = _mm_min_epu8(m0, m1); // merge

// Fall back on bytewise code to do error checking and reporting:
unsigned int mask = _mm_movemask_epi8(m0);
if(mask != 0){
break;
}

// Pack 16 bytes into the 12-byte output format and store back:
_mm_storeu_si128((__m128i *)o, dec_reshuffle(out));
Copy link
Owner

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The dec_reshuffle function appears to be undefined.

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

This was meant for atom CPUs
so the ssse3 dec_reshuffle() would have been used.

I tried to do a sse2 packer here
but it is not too great.

Copy link
Owner

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I tried to get the code working with the SSSE3 decoder to get an idea of its performance, but hit a small bug: the function call should be dec_reshuffle(m0).

If the assumption is that this code would be used in conjunction with the SSSE3 reshuffler, then why not use the full SSSE3 codec?

One possible answer comes when testing this on a 32-bit Atom N270. There is a large improvement over the stock SSSE3 decoder: 591 MB/s for the SSE2 code against 427 MB/s for the SSSE3 code.

It's significantly slower on my desktop i5-4590S though: 6150 MB/s against 6800 MB/s for the stock SSSE3 codec. This makes it hard to justify merging this code for common use.

Maybe GCC's __builtin_cpu_is("atom") builtin would be useful here...


c += 16;
o += 12;
outl += 12;
srclen -= 16;
}

/*
bool Test_SSE2_Decode()
{
const unsigned char base64_table_dec[] =
{
255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 62, 255, 255, 255, 63,
52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 255, 255, 255, 255, 255, 255,
255, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14,
15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 255, 255, 255, 255, 255,
255, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40,
41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 255, 255, 255, 255, 255,
255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255, 255,
};

const __m128i digit_top = _mm_set1_epi32( 0x46464646 ); // 0x46 = 0x7F - '9'
const __m128i lowercase_top = _mm_set1_epi32( 0x05050505 ); // 0x05 = 0x7F - 'z'
const __m128i uppercase_top = _mm_set1_epi32( 0x25252525 ); // 0x25 = 0x7F - 'Z'
const __m128i digit_bottom = _mm_set1_epi32( 0x76767676 ); // 0x76 = 0x7F - 9 ('9'-'0')
const __m128i alpha_bottom = _mm_set1_epi32( 0x66666666 ); // 0x66 = 0x7F - 25 ('Z'-'A') ('z'-'a')
const __m128i num_letters = _mm_set1_epi32( 0x1A1A1A1A ); // 0x1A = 26
const __m128i detect_plus = _mm_set1_epi32( 0xFBFBFBFB ); // 0xFB = plus sign value after '0-9' roll down
const __m128i adj_plus = _mm_set1_epi32( 0xD4D4D4D4 ); // 0xD4 = plus sign shift after 'A-Z' roll down
const __m128i detect_slash = _mm_set1_epi32( 0xFFFFFFFF ); // 0xFF = after 0-9 roll down

// check all possible values of 16 bits
for(int i = 0; i < 0x10000; i++)
{
__m128i m0, m1, m2, m3, m4;

m0 = _mm_cvtsi32_si128(i); // load string

m1 = _mm_add_epi8(m0, digit_top); // "0-9" roll up
m2 = _mm_add_epi8(m0, lowercase_top); // "a-z" roll up
m1 = _mm_subs_epi8(m1, digit_bottom); // "0-9" roll down
m2 = _mm_subs_epi8(m2, alpha_bottom); // "a-z" roll down
m4 = _mm_cmpeq_epi8(m1, detect_slash); // match '/' ( foward slash )
m3 = _mm_cmpeq_epi8(m1, detect_plus); // match '+' ( plus sign )
m1 = _mm_adds_epu8(m1, num_letters); // ('0'=26 thru '9'=35) else signed
m4 = _mm_subs_epu8(m4, uppercase_top); // 0xDA = 0xFF - 0x25
m0 = _mm_add_epi8(m0, uppercase_top); // "A-Z" roll up
m1 = _mm_xor_si128(m1, m4); // ( '/' = 37 ) = FF ^ DA
m3 = _mm_and_si128(m3, adj_plus); // = 0xFF & 0xD4
m0 = _mm_subs_epi8(m0, alpha_bottom); // "A-Z" roll down
m1 = _mm_min_epu8(m1, m2); // merge
m0 = _mm_xor_si128(m0, m3); // ( '+' = 62 )
m1 = _mm_adds_epu8(m1, num_letters); // ('a'=26 thru '9'=61, '/'=63) else signed
m0 = _mm_min_epu8(m0, m1); // merge

// set signed bytes to 255
__m128i res;
res = _mm_cmpgt_epi8(_mm_setzero_si128(), m0);
res = _mm_or_si128(res, m0);

// get low 16 bits
unsigned int r = _mm_cvtsi128_si32(res) & 0x0000FFFF;

// check that byte_0 decoded correctly
if((base64_table_dec[i & 0x00FF]) != (r & 0x00FF)){
return false;
}

// check that byte_1 decoded correctly
if((base64_table_dec[i >> 8]) != ((r >> 8) & 0x00FF)){
return false;
}
}
return true;
}
*/