hashcat/include/kernel_vendor.h

/**
 * Author......: Jens Steube <jens.steube@gmail.com>
 * License.....: MIT
 */

#pragma OPENCL EXTENSION cl_khr_byte_addressable_store : enable

/**
 * vendor specific
 */

#if VENDOR_ID == 4098
#define IS_AMD
#endif

#if VENDOR_ID == 4318
#define IS_NV
#endif

#if VENDOR_ID == 9998 // temporary for dev
#define IS_UNKNOWN
#endif

#if VENDOR_ID == 9999
#define IS_UNKNOWN
#endif

/**
 * AMD specific
 */

#ifdef IS_AMD
#pragma OPENCL EXTENSION cl_amd_media_ops  : enable
#pragma OPENCL EXTENSION cl_amd_media_ops2 : enable
#endif

/**
 * NV specific
 */

#ifdef IS_NV
#endif
Initial commit 2015-12-04 14:47:52 +00:00			`/**`
			`* Author......: Jens Steube <jens.steube@gmail.com>`
			`* License.....: MIT`
			`*/`

- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor 2015-12-15 11:04:22 +00:00			`#pragma OPENCL EXTENSION cl_khr_byte_addressable_store : enable`

Initial commit 2015-12-04 14:47:52 +00:00			`/**`
			`* vendor specific`
			`*/`

Fix invalid VENDOR_ID for AMD 2015-12-16 09:00:33 +00:00			`#if VENDOR_ID == 4098`
Initial commit 2015-12-04 14:47:52 +00:00			`#define IS_AMD`
			`#endif`

- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor 2015-12-15 11:04:22 +00:00			`#if VENDOR_ID == 4318`
Initial commit 2015-12-04 14:47:52 +00:00			`#define IS_NV`
			`#endif`

Prepare for allow other OpenCL platforms thans AMD and NVidia 2016-01-04 12:17:20 +00:00			`#if VENDOR_ID == 9998 // temporary for dev`
			`#define IS_UNKNOWN`
			`#endif`

			`#if VENDOR_ID == 9999`
			`#define IS_UNKNOWN`
			`#endif`

Initial commit 2015-12-04 14:47:52 +00:00			`/**`
			`* AMD specific`
			`*/`

			`#ifdef IS_AMD`
Prepare for allow other OpenCL platforms thans AMD and NVidia 2016-01-04 12:17:20 +00:00			`#pragma OPENCL EXTENSION cl_amd_media_ops : enable`
Initial commit 2015-12-04 14:47:52 +00:00			`#pragma OPENCL EXTENSION cl_amd_media_ops2 : enable`
			`#endif`

- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor 2015-12-15 11:04:22 +00:00			`/**`
			`* NV specific`
			`*/`
Initial commit 2015-12-04 14:47:52 +00:00
			`#ifdef IS_NV`
			`#endif`