File vector_math_impl-Wstrict-aliasing.patch of Package nodejs-electron.v3

Overview Repositories Revisions Requests Users Attributes Meta

File vector_math_impl-Wstrict-aliasing.patch of Package nodejs-electron.v3

--- src/third_party/blink/renderer/platform/audio/cpu/x86/vector_math_impl.h.old	2022-10-20 19:00:30.057387900 +0200
+++ src/third_party/blink/renderer/platform/audio/cpu/x86/vector_math_impl.h	2022-10-29 23:03:21.785150600 +0200
@@ -13,6 +13,7 @@
 
 #include <algorithm>
 #include <cmath>
+#include <cstring>
 
 #include "base/check_op.h"
 #include "third_party/blink/renderer/platform/audio/audio_array.h"
@@ -208,13 +209,14 @@
 // source_max = max(abs(source[k])) for all k
 void Vmaxmgv(const float* source_p, float* max_p, uint32_t frames_to_process) {
   constexpr uint32_t kMask = 0x7FFFFFFFu;
-
+  float kMask_float;
+  std::memcpy(&kMask_float, &kMask, 4);
   const float* const source_end_p = source_p + frames_to_process;
 
   DCHECK(IsAligned(source_p));
   DCHECK_EQ(0u, frames_to_process % kPackedFloatsPerRegister);
 
-  MType m_mask = MM_PS(set1)(*reinterpret_cast<const float*>(&kMask));
+  MType m_mask = MM_PS(set1)(kMask_float);
   MType m_max = MM_PS(setzero)();
 
   while (source_p < source_end_p) {

Places

File vector_math_impl-Wstrict-aliasing.patch of Package nodejs-electron.v3

Places