oneDNN Files

This is a patch release containing the following changes to v3.8.1: * Fixed performance regression for f32 convolution primitive on processors with Intel AVX-512 instruction set support (5f3af68b6beb52377d40082f9c5a9a47b16c3753) * Introduced support for f16 destination in int8 matmul and int8 inner product on x64 CPUs (53fd12abd898d2540ab203036bd8980ca4d1b383, 22e252c8e4f6a9e0242758fe4a06f482f453e8ae, f5b2d7f7cf943ed4e02858bb2ebf849b6c3b3499, e4e2f1cba595fbb593a29755c39d447eb6489ea9) * Improved RNN primitive performance on processors with Intel AVX2 instruction set support (71e5d81b5a0a673ecdba39fad54bdc734f469b83, eb27db21e6c22f81ffdd7362ce1c5f1624ccbac8, dd4e6272877c83cdd1c76da6da4521252f486c13, ff134e0c22775d5c10b8c6b5f00bec8574dd184d, 5a86c1f0c4ede1ca714538d667a0490ca7adea1f, e9395ae868d3fe35ab30edc4f47ddd0b70fe988c) * Improved fp32 matmul performance on processors with Intel AVX-512 instruction set support (11193399a8c2b8b1b8578cabb47fe986a19e3623) * Fixed segmentation fault in f32 binary primitive with broadcast on x64 processors (2082e9879e7db168ca4eb93a8d2dfe823b79d0c0) * Fixed correctness issue in f64 convolution weight gradient with bias on Intel Arc GPUs (a00bfabdd89afff020de91c8f68e3366b64135f4) * Updated spdlog component to version 1.15.3 (dbb362929a7aa5769b58855a857da34db336148e) * Fixed potential undefined behavior in convolution on Intel GPUs (5ac3e31d61fb0184a12f97221d55cb707b406952) * Fixed segmentation fault in convolution implementation with trivial filter on Intel CPUs (908c5fcb5579fd133931311c18fd28ceca7adc75, f0a0eee94ed0a600c859f37d82e70792fe0d2f69) * Fixed segmentation fault in f16 convolution with odd dimensions on processors with Intel AVX10.1 instruction set support (78d683508d1ac89e60e24c0681f7658a4a1d2558) * Improved convolution primitive descriptor creation time on x64 processors (e9c5366361c9c9db55cdeba4108b5151c5ab2836, fd9dc580d0af14763d06088a567e70c4ca6d4c25, f1d038e9e2292c6cc5e4f93f5fd59a410e23e368) * Fixed performance regression in f16 matmul with int4 weights on Intel Arc Graphics B-series (38d761b1c630a117078ebdd3952b97736f33fe61) * Improved bf16 matmul performance on processors with Intel AMX instruction set support (0887aec3af524cd8d8b0518544253019118744ca) * Fixed correctness issue in f32 RNN primitive on processors with Intel AMX instruction set support (460a014c9400d8d43f78e466594b5306b4313a8f)