Jens Steube
|
fd2cb59d26
|
AMD GPUs: On Apple OpenCL platform, we ask for the preferred kernel thread size rather than hard-coding 32
ECC secp256k1: Removed the inline assembly code for AMD GPUs because the latest JIT compilers optimize it with the same efficiency
|
2021-07-27 09:37:31 +02:00 |
|
Jens Steube
|
cf512faa53
|
Update large switch() cases in inc_common.cl and some inline assembly common functions for devices managed with HIP backend
|
2021-07-14 17:06:20 +02:00 |
|
Royce Williams
|
1e88990b46
|
minor whitespace fixes, per #2841
|
2021-06-21 07:47:22 -08:00 |
|
Bernard Ladenthin
|
b85ffd63fb
|
Refactoring: Extract transform_public, point_mul_xy and set_precomputed_basepoint_g. Add constants and documentation.
|
2021-02-08 23:02:47 +01:00 |
|
Bernard Ladenthin
|
19f4b44840
|
Refactoring: Extract convert_to_window_naf and add some documentation.
|
2020-11-17 21:33:51 +01:00 |
|
philsmd
|
53d2e45795
|
fixes #2341: electrum 4/5 mod_512 () infinite loop fix
|
2020-03-31 11:01:47 +02:00 |
|
Jens Steube
|
0378a01422
|
Fix more rocm compiler warning
|
2020-01-12 11:22:26 +01:00 |
|
philsmd
|
4338f100e9
|
remove condition which is always true
|
2019-12-27 11:50:02 +01:00 |
|
philsmd
|
7ef92379d8
|
Electrum 4/5: speedup by using w-NAF (Non-Adjacent Form)
|
2019-12-27 09:12:22 +01:00 |
|
Jens Steube
|
75b4164498
|
Use a different code for mod_512() to help some NV GPU to not hang
|
2019-12-07 11:29:39 +01:00 |
|
Jens Steube
|
53254b45aa
|
Backport inc_ecc_secp256k1 inline assembly code for AMD ISA
|
2019-12-05 15:43:01 +01:00 |
|
Jens Steube
|
cb24236067
|
Inline assembly optimization for 256 bit ADD and SUB in inc_ecc_secp256k1.cl
|
2019-12-05 14:49:51 +01:00 |
|
philsmd
|
6d822e04a1
|
fix minor typos in inc_ecc_secp256k1.cl
|
2019-12-05 12:23:54 +01:00 |
|
philsmd
|
d07f002337
|
electrum 4/5: improve speed (rm hook)
|
2019-12-05 10:43:42 +01:00 |
|