GOSTSec/sgminer
1
0
Fork 0
mirror of https://github.com/GOSTSec/sgminer synced 2025-01-22 20:44:19 +00:00
Code Issues Releases Wiki Activity

Remove the rounding-up of the scrypt padbuffer which was not effectual and counter-productive on devices with lots of ram, limiting thread concurrencies and intensities.

Browse Source
This commit is contained in:
Con Kolivas 2013-03-15 21:48:48 +11:00
parent 1c6d8a36d8
commit 3c3fbdce1c
1 changed files with 1 additions and 20 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

21
ocl.c
View File

@ -475,9 +475,6 @@ _clState *initCl(unsigned int gpu, char *name, size_t nameSize)
#ifdef USE_SCRYPT #ifdef USE_SCRYPT
if (opt_scrypt) { if (opt_scrypt) {
cl_ulong ma = cgpu->max_alloc, mt;
int pow2 = 0;
if (!cgpu->opt_lg) { if (!cgpu->opt_lg) {
applog(LOG_DEBUG, "GPU %d: selecting lookup gap of 2", gpu); applog(LOG_DEBUG, "GPU %d: selecting lookup gap of 2", gpu);
cgpu->lookup_gap = 2; cgpu->lookup_gap = 2;
@ -485,31 +482,15 @@ _clState *initCl(unsigned int gpu, char *name, size_t nameSize)
cgpu->lookup_gap = cgpu->opt_lg; cgpu->lookup_gap = cgpu->opt_lg;
if (!cgpu->opt_tc) { if (!cgpu->opt_tc) {
cgpu->thread_concurrency = ma / 32768 / cgpu->lookup_gap; cgpu->thread_concurrency = cgpu->max_alloc / 32768 / cgpu->lookup_gap;
if (cgpu->shaders && cgpu->thread_concurrency > cgpu->shaders) { if (cgpu->shaders && cgpu->thread_concurrency > cgpu->shaders) {
cgpu->thread_concurrency -= cgpu->thread_concurrency % cgpu->shaders; cgpu->thread_concurrency -= cgpu->thread_concurrency % cgpu->shaders;
if (cgpu->thread_concurrency > cgpu->shaders * 5) if (cgpu->thread_concurrency > cgpu->shaders * 5)
cgpu->thread_concurrency = cgpu->shaders * 5; cgpu->thread_concurrency = cgpu->shaders * 5;
} }
applog(LOG_DEBUG, "GPU %d: selecting thread concurrency of %u",gpu, cgpu->thread_concurrency); applog(LOG_DEBUG, "GPU %d: selecting thread concurrency of %u",gpu, cgpu->thread_concurrency);
} else } else
cgpu->thread_concurrency = cgpu->opt_tc; cgpu->thread_concurrency = cgpu->opt_tc;
/* If we have memory to spare, try to find a power of 2 value
* >= required amount to map nicely to an intensity */
mt = cgpu->thread_concurrency * 32768 * cgpu->lookup_gap;
if (ma > mt) {
while (ma >>= 1)
pow2++;
ma = 1;
while (--pow2 && ma < mt)
ma <<= 1;
if (ma >= mt) {
cgpu->max_alloc = ma;
applog(LOG_DEBUG, "Max alloc decreased to %lu", cgpu->max_alloc);
}
}
} }
#endif #endif