Michael Wang

The following occurs when attempting to run almost any algorithm on hashcat ``` [CLVK] clSetKernelArg: passing a null pointer to clSetKernelArg is only supported for local arguments clSetKernelArg(): CL_INVALID_ARG_VALUE ```

using the env variables ``` CLVK_SPIRV_ARCH=spir64 CLVK_PHYSICAL_ADDRESSING=1 ``` still results in ``` [CLVK] clSetKernelArg: passing a null pointer to clSetKernelArg is only supported for local arguments clSetKernelArg(): CL_INVALID_ARG_VALUE ```

certain algorithms produce errors similar to the following ``` [CLVK] Loaded SPIR-V binary from "clvk-K8YQe9/compiled.spv", size = 31348 words error: 3548: Expected input to have different bit width from Result...

With #1211 and #1213: ``` unzip a.zip CLVK_SPIRV_ARCH=spir64 CLVK_PHYSICAL_ADDRESSING=1 clspv -cl-mad-enable -DDEVICE_INFO=32 -D__SIZEOF_HOST_SIZE_T__=8 -DDEV_VER_MAJOR=3 -DDEV_VER_MINOR=0 -D_OPENCL_COMPILER -DHASH_LOOPS=250 -DPLAINTEXT_LENGTH=110 -DPBKDF2_64_MAX_SALT_SIZE=107 -cl-single-precision-constant -cl-kernel-arg-info -rounding-mode-rte=16,32,64 -rewrite-packed-structs -std430-ubo-layout -decorate-nonuniform -arch=spir64 -physical-storage-buffers --use-native-builtins=acos,acosh,acospi,asin,asinh,asinpi,atan,atan2,atan2pi,atanh,atanpi,ceil,copysign,fabs,fdim,floor,fma,fmax,fmin,frexp,half_rsqrt,half_sqrt,isequal,isfinite,isgreater,isgreaterequal,isinf,isless,islessequal,islessgreater,isnan,isnormal,isnotequal,isordered,isunordered,ldexp,mad,rint,round,rsqrt,signbit,sqrt,tanh,trunc, -spv-version=1.6 -max-pushconstant-size=256...

This appears to be fixed as of #1245 although clvk still outputs the following ``` error: 28312: Opcode ConvertPtrToU requires one of these capabilities: Addresses PhysicalStorageBufferAddresses %27424 = OpConvertPtrToU %uint...

#### With env variables: ``` unzip env.zip CLVK_SPIRV_ARCH=spir64 CLVK_PHYSICAL_ADDRESSING=1 clspv -cl-mad-enable -DDEVICE_INFO=32 -D__SIZEOF_HOST_SIZE_T__=8 -DDEV_VER_MAJOR=3 -DDEV_VER_MINOR=0 -D_OPENCL_COMPILER -DHASH_LOOPS=105 -DITERATIONS=4095 -DMAX_OUTLEN=32 -DPLAINTEXT_LENGTH=64 -DV_WIDTH=1 -cl-single-precision-constant -cl-kernel-arg-info -rounding-mode-rte=16,32,64 -rewrite-packed-structs -std430-ubo-layout -decorate-nonuniform -arch=spir64 -physical-storage-buffers --use-native-builtins=acos,acosh,acospi,asin,asinh,asinpi,atan,atan2,atan2pi,atanh,atanpi,ceil,copysign,fabs,fdim,floor,fma,fmax,fmin,frexp,half_rsqrt,half_sqrt,isequal,isfinite,isgreater,isgreaterequal,isinf,isless,islessequal,islessgreater,isnan,isnormal,isnotequal,isordered,isunordered,ldexp,mad,rint,round,rsqrt,signbit,sqrt,tanh,trunc,...

Perhaps the macro could take in a compilation target at compile time (like `rustc -C target-cpu=native`)?