a DfI@sddlmZddlmZmZddlmZddlmZddl m Z ddl m Z ddl mZddlmZd d Zed d Zed dZeddZe eedddddZeddZddZeddZeddZeddZd S)!)ir)cudatypes)cgutils)RequireLiteralValue) signature)overload_attribute) nvvmutils) intrinsiccCsB|j}|dkrtj}n |dvr.ttj|}ntdt|tjS)N)zargument can only be 1, 2, 3)Z literal_valuerint64UniTuple ValueErrorrint32)ndimvalrestyperb/nfs/NAS7/SABIOD/METHODE/ermites/ermites_venv/lib/python3.9/site-packages/numba/cuda/intrinsics.py_type_grid_functionsrcCs,t|tjst|t|}dd}||fS)agrid(ndim) Return the absolute position of the current thread in the entire grid of blocks. *ndim* should correspond to the number of dimensions declared when instantiating the kernel. If *ndim* is 1, a single integer is returned. If *ndim* is 2 or 3, a tuple of the given number of integers is returned. Computation of the first integer is as follows:: cuda.threadIdx.x + cuda.blockIdx.x * cuda.blockDim.x and is similar for the other two indices, but using the ``y`` and ``z`` attributes. cSsJ|j}|tjkrtj|ddSt|tjrFtj||jd}t ||SdS)Nr )dim) return_typerrr Z get_global_id isinstancercountr pack_array)contextbuildersigargsridsrrrcodegen1s   zgrid..codegenrrZIntegerLiteralrr typingctxrrr"rrrgrids  r&cs8t|tjst|t|}ddfdd}||fS)agridsize(ndim) Return the absolute size (or shape) in threads of the entire grid of blocks. *ndim* should correspond to the number of dimensions declared when instantiating the kernel. If *ndim* is 1, a single integer is returned. If *ndim* is 2 or 3, a tuple of the given number of integers is returned. Computation of the first integer is as follows:: cuda.blockDim.x * cuda.gridDim.x and is similar for the other two indices, but using the ``y`` and ``z`` attributes. cSsJtd}t|d|}t|d|}|||||||S)N@zntid.znctaid.)rIntTyper call_sregmulZsext)rrZi64ZntidZnctaidrrr_nthreads_for_dimRs z#gridsize.._nthreads_for_dimcsx|j}|d}|tjkr|St|tjrt|d}|jdkrNt|||fS|jdkrt|d}t||||fSdS)Nxyr r z)rrrrrrrr)rrrr rZnxnyZnzr+rrr"Xs       zgridsize..codegenr#r$rr0rgridsize<s   r1cCsttj}dd}||fS)NcSs t|dS)Nwarpsize)r r))rrrr rrrr"nsz_warpsize..codegen)rrrr%rr"rrr _warpsizejs r4r2r)targetcCs dd}|S)z_ The size of a warp. All architectures implemented to date have a warp size of 32. cSstS)N)r4)modrrrgetzszcuda_warpsize..getr)r6r7rrr cuda_warpsizetsr8cCsttj}dd}||fS)a Synchronize all threads in the same thread block. This function implements the same pattern as barriers in traditional multi-threaded programming: this function waits until all threads in the block call it, at which point it returns control to all its callers. cSs<d}|j}ttd}t|||}||d|S)Nzllvm.nvvm.barrier0r)moduler FunctionTypeZVoidTyperget_or_insert_functioncallZget_dummy_value)rrrr fnameZlmodfntysyncrrrr"s  zsyncthreads..codegen)rrnoner3rrr syncthreadss rAcs2t|tjsdSttjtj}fdd}||fS)Ncs6ttdtdf}t|j|}|||S)N )rr:r(rr;r9r<)rrrr r>r?r=rrr"sz'_syncthreads_predicate..codegen)rrZIntegerri4)r% predicater=rr"rrCr_syncthreads_predicates   rFcCsd}t|||S)z syncthreads_count(predicate) An extension to numba.cuda.syncthreads where the return value is a count of the threads where predicate is true. zllvm.nvvm.barrier0.popcrFr%rEr=rrrsyncthreads_countsrIcCsd}t|||S)z syncthreads_and(predicate) An extension to numba.cuda.syncthreads where 1 is returned if predicate is true for all threads or 0 otherwise. zllvm.nvvm.barrier0.andrGrHrrrsyncthreads_andsrJcCsd}t|||S)z syncthreads_or(predicate) An extension to numba.cuda.syncthreads where 1 is returned if predicate is true for any thread or 0 otherwise. zllvm.nvvm.barrier0.orrGrHrrrsyncthreads_orsrKN)ZllvmliterZnumbarrZ numba.corerZnumba.core.errorsrZnumba.core.typingrZnumba.core.extendingrZ numba.cudar Znumba.cuda.extendingr rr&r1r4Moduler8rArFrIrJrKrrrrs2          -