It should be part of the 9.4.1 point release. Peter
On Thursday, 23 February 2023 at 15:20:14 UTC+1 yy.wayne wrote: > I'm using deal.II 9.4.0. It seems bug are fixed... > 在2023年2月23日星期四 UTC+8 22:16:47<peterr...@gmail.com> 写道： > >> What version of deal.II are you using. I thought we fixed the >> vector-valued version: >> https://github.com/dealii/dealii/commit/6351f255d16c3cf465752e2ce943726df6bf4ea4 >> >> Peter >> >> On Thursday, 23 February 2023 at 15:13:42 UTC+1 yy.wayne wrote: >> >>> I think the problem exists for vector-valued problem. >>> I modified the vectorization length of step-37 and it works fine. >>> But for vector value problem, on line 6090 of >>> deal.II/matrix_free/fe_evaluation.h: >>> >>> - val_in[comp] * weight * jac[comp][comp] >>> >>> The jac is defined as const Tensor<2, dim, >>> dealii::VectorizedArray<Number>>, which doesn't >>> match customized VectorizedArrayType. >>> >>> Similar on line 6101 >>> >>> - transpose(invert(inv_t_jac)) >>> >>> where inv_t_jac is also dealii::VectorizedArray<Number> type data. >>> 在2023年2月23日星期四 UTC+8 18:43:07<yy.wayne> 写道： >>> >>>> Thanks you Peter, >>>> >>>> The error message is quite long and tricky (beyond my terminal can >>>> show) when I set >>>> VectorizedArray<double, width> with width=1. >>>> Knowing that deal.II do support this function is great. I‘ll try debug >>>> the code first and came back with >>>> further progress or questions :) >>>> >>>> Best, >>>> Wayne >>>> >>>> 在2023年2月23日星期四 UTC+8 16:00:43<peterr...@gmail.com> 写道： >>>> >>>>> Hi Wayne, >>>>> >>>>> What is you error message? >>>>> >>>>> > I replace all FEEvaluation, MatrixFree and VectorizedArray >>>>> construction with a VectorizedArray<number, width>, but make returns some >>>>> intricate error. Is specify vectorization length possible? I think I've >>>>> seen one in dealii/test ... >>>>> >>>>> Yes, it is possible and the to specify the additional template >>>>> argument is the right approach. >>>>> >>>>> I have introduced this template parameter for the hyper.deal project: >>>>> https://github.com/hyperdeal/hyperdeal, where we create a tensor >>>>> product of MatrixFree instances (where one is vectorized and the other is >>>>> not). >>>>> >>>>> Take a look at the test: >>>>> https://github.com/dealii/dealii/blob/3b785178375a8ea233fada96cf987aed6d53583e/tests/matrix_free/ecl.h#L52 >>>>> >>>>> The important aspect is that the vector length has to match the ones >>>>> supported by given ISA. See table 1 in the release paper of deal.II 9.2: >>>>> https://dealii.org/deal92-preprint.pdf >>>>> >>>>> Peter >>>>> >>>>> On Thursday, February 23, 2023 at 7:09:07 AM UTC+1 yy.wayne wrote: >>>>> >>>>>> Hi, >>>>>> is it possible to specify vectorization length(<deafult)? >>>>>> I'm running same code on local computer and server, whose cpu >>>>>> supports SSE2+4.10GHz >>>>>> and AVX512+2.60GHz respectively. The speed up for iteration time is >>>>>> only 2 on server, which should be 8*2.6 / 2*4.1 > 2.5 times. Therefore >>>>>> I'd >>>>>> like to run code with different >>>>>> vectorization length on same computer to test Vectorization's power. >>>>>> >>>>>> I replace all FEEvaluation, MatrixFree and VectorizedArray >>>>>> construction with a VectorizedArray<number, width>, but make returns >>>>>> some >>>>>> intricate error. Is specify vectorization length possible? I think I've >>>>>> seen one in dealii/test ... >>>>>> >>>>> -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to dealii+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/dealii/e04071d1-1143-4413-9053-f53762a61be1n%40googlegroups.com.
