I only showed the first line. There were many more. Gotta go to work though.
René van de Veerdonk wrote:
Micheal,
Is that all the output you get? Did you notice that the first line with
test-error changed ... the second number went from 31 to 32. We did
something! But not what I expected. BTW. The program does not normally
crash, it just stops after its initial testing is completed and errors
were detected. If you actually went from a gazillion line to just one,
we have hit a nerve. I you just didn't bother to copy the rest of the
output, we made no progress. But, since the output actually changed, it
would still be nice to see a few more lines and see the trend.
Thanks for helping ... debugging for someone else must not be the best
use of your time,
René
On Tue, Aug 25, 2009 at 8:03 PM, Michael Williams
mailto:michaelwilliam...@gmail.com>> wrote:
No. The code is now:
__forceinline
unsigned int bitmap_t::lowest_index () const {
const __m128i comp = _mm_cmpeq_epi32 (_mm_setzero_si128 (), board);
unsigned long m = _mm_movemask_epi8 (comp) ^ 0x;
if (!_BitScanForward (&m, m)) return 127;
/* offending instruction */
//return 8*m + __lzcnt (board.m128i_u32[m/4]);
/* new code follows */
unsigned long n;
_BitScanForward (&n, board.m128i_u32[m/4]);
return 8*m + (32-n);
}
Debug:
testing _BitScanForward (&m, 0): 0 1099227377
testing _BitScanForward (&m, 1): 1 0
testing _BitScanForward (&m, 256): 1 8
testing _BitScanReverse (&m, 0): 0 1103042705
testing _BitScanReverse (&m, 1): 1 0
testing _BitScanReverse (&m, 256): 1 8
test-error [bitmap_t::lowest_index]: 0 32
Release:
testing _BitScanForward (&m, 0): 0 16843009
testing _BitScanForward (&m, 1): 1 0
testing _BitScanForward (&m, 256): 1 8
testing _BitScanReverse (&m, 0): 0 4
testing _BitScanReverse (&m, 1): 1 0
testing _BitScanReverse (&m, 256): 1 8
test-error [bitmap_t::lowest_index]: 0 32
René van de Veerdonk wrote:
Micheal,
Your output is actually correct. The output is undefined when
the second argument to _BitScanForward is zero and that is what
I am explicitly testing for in the method
bitmap_t::lowest_index, which is supposed to return an index to
any set bit (despite the name). Your output shows that it
returns a number that is off in a very predictable way (it seems
that replacing "8*m + n" with "8*m + (32-n)" on the last line
may do the trick for you, but it breaks my local copy). So, now
I am positively baffled.
Help?
René
On Tue, Aug 25, 2009 at 7:25 PM, Michael Williams
mailto:michaelwilliam...@gmail.com>
>> wrote:
Debug build:
testing _BitScanForward (&m, 0): 0 842032512
testing _BitScanForward (&m, 1): 1 0
testing _BitScanForward (&m, 256): 1 8
testing _BitScanReverse (&m, 0): 0 840277132
testing _BitScanReverse (&m, 1): 1 0
testing _BitScanReverse (&m, 256): 1 8
Release build:
testing _BitScanForward (&m, 0): 0 16843009
testing _BitScanForward (&m, 1): 1 0
testing _BitScanForward (&m, 256): 1 8
testing _BitScanReverse (&m, 0): 0 4
testing _BitScanReverse (&m, 1): 1 0
testing _BitScanReverse (&m, 256): 1 8
Sucks when they produce different results.
René van de Veerdonk wrote:
Micheal,
Thanks for trying. Could you try one or two more things?
(1) Replace the _BitScanForward in the new code to
_BitScanReverse. This probably won't help.
(2) Add this to bitmapgo.cpp:
/* sandbox */
const bool sandbox = true;
void test_sandbox () {
unsigned long m = 220;
unsigned char test_it = _BitScanForward (&m, 0);
std::cout << "testing _BitScanForward (&m, 0): "
<< (int) test_it << " " << m << "\n";
test_it = _BitScanForward (&m, 1);
std::cout << "testing _BitScanForward (&m, 1): "
<< (int) test_it << " " << m << "\n";
test_it = _BitScanForward (&m, 256);
std::cout << "testing _BitScanForward (&m, 256): "
<< (int) test_it << " " << m << "\n";
test_it = _BitScanReverse (&m, 0);
std::cout << "testing _BitScanReverse (&m, 0): "
<< (int) test_it << " " << m << "\n";
test_it = _BitScanReverse (&m, 1);
std::cout << "testing _BitScanReverse (&m, 1): "
<< (int) test_it << " " << m << "\n";
test_it = _BitScanReverse (&m, 256);
