Reduce load time by shifting mul! definition (#1904)

Co-authored-by: Tim Besard <[email protected]>

Author: dkarrasch

.buildkite/pipeline.yml

@@ -34,7 +34,7 @@ steps:
               - "1.6"
               - "1.7"
               - "1.8"
-              - "1.9-nightly"
+              - "1.9"
               - "nightly"
           adjustments:
             - with:

Manifest.toml

@@ -99,15 +99,17 @@ version = "0.1.9"
 
 [[GPUArrays]]
 deps = ["Adapt", "GPUArraysCore", "LLVM", "LinearAlgebra", "Printf", "Random", "Reexport", "Serialization", "Statistics"]
-git-tree-sha1 = "9ade6983c3dbbd492cf5729f865fe030d1541463"
+git-tree-sha1 = "23ac0f43c861cef28338b6129ac7afc6f52b4d2f"
+repo-rev = "master"
+repo-url = "https://github.com/JuliaGPU/GPUArrays.jl.git"
 uuid = "0c68f7d7-f131-5f86-a1c3-88cf8149b2d7"
-version = "8.6.6"
+version = "8.7.0"
 
 [[GPUArraysCore]]
 deps = ["Adapt"]
-git-tree-sha1 = "1cd7f0af1aa58abc02ea1d872953a97359cb87fa"
+git-tree-sha1 = "2d6ca471a6c7b536127afccfa7564b5b39227fe0"
 uuid = "46192b85-c4d5-4398-a991-12ede77f4527"
-version = "0.1.4"
+version = "0.1.5"
 
 [[GPUCompiler]]
 deps = ["ExprTools", "InteractiveUtils", "LLVM", "Libdl", "Logging", "Scratch", "TimerOutputs", "UUIDs"]