SWDEV-312822 - Revert "SWDEV-310187 - Change flag to keep track of aligned sizes instead of expanded patterns."

This reverts commit 8307886644.

Change-Id: I022c2a8375f9929e9723cec66e1e0b960263fc39

rocclr/device/blit.cpp

@@ -763,8 +763,8 @@ bool HostBlitManager::FillBufferInfo::PackInfo(const device::Memory& memory, siz
                           pattern, pattern_size);
         return false;
       }
+      fill_info.pattern_expanded_ = true;
     }
-    fill_info.pattern_aligned_ = true;
     fill_info.fill_size_ = aligned_size;
     packed_info.push_back(fill_info);
   }

rocclr/device/blit.hpp

@@ -354,7 +354,7 @@ class HostBlitManager : public device::BlitManager {
   // Packed Fill Buffer
   class FillBufferInfo {
   public:
-    FillBufferInfo(): fill_size_(0), expanded_pattern_(0), pattern_aligned_(false) {}
+    FillBufferInfo(): fill_size_(0), expanded_pattern_(0), pattern_expanded_(false) {}
 
     static bool PackInfo(const device::Memory& memory, size_t fill_size,
                   size_t fill_origin, const void* pattern, size_t pattern_size,
@@ -371,13 +371,13 @@ class HostBlitManager : public device::BlitManager {
     void clearInfo () {
       fill_size_ = 0;
       expanded_pattern_ = 0;
-      pattern_aligned_ = false;
+      pattern_expanded_ = false;
     }
 
   public:
     size_t fill_size_;          // Fill size for this command
     uint64_t expanded_pattern_; // Pattern for this command
-    bool pattern_aligned_;     // Boolean to check if pattern is expanded
+    bool pattern_expanded_;     // Boolean to check if pattern is expanded
   };
 
 

rocclr/device/rocm/rocblit.cpp

@@ -1966,18 +1966,17 @@ bool KernelBlitManager::fillBuffer(device::Memory& memory, const void* pattern,
     FillBufferInfo::PackInfo(memory, size[0], origin[0], pattern, patternSize, packed_vector);
 
     size_t overall_offset = origin[0];
-    uint fillType = FillBufferAligned;
     for (auto& packed_obj: packed_vector) {
+      uint fillType = FillBufferAligned;
+      size_t globalWorkOffset[3] = {0, 0, 0};
+      size_t globalWorkSize = amd::alignUp(packed_obj.fill_size_, 256);
+      size_t localWorkSize = 256;
 
-      uint32_t kpattern_size32 = (packed_obj.pattern_aligned_) ? sizeof(size_t) : patternSize;
+      uint32_t kpattern_size32 = (packed_obj.pattern_expanded_) ? sizeof(size_t) : patternSize;
       size_t kfill_size = packed_obj.fill_size_/kpattern_size32;
       size_t koffset = overall_offset;
       overall_offset += packed_obj.fill_size_;
 
-      size_t globalWorkOffset[3] = {0, 0, 0};
-      size_t globalWorkSize = amd::alignUp(kfill_size, 256);
-      size_t localWorkSize = 256;
-
       uint32_t alignment = (kpattern_size32 & 0x7) == 0 ?
                             sizeof(uint64_t) :
                             (kpattern_size32 & 0x3) == 0 ?
@@ -2019,7 +2018,7 @@ bool KernelBlitManager::fillBuffer(device::Memory& memory, const void* pattern,
       auto constBuf = reinterpret_cast<address>(constantBuffer_->getHostMem()) + constBufOffset;
 
       // If pattern has been expanded, use the expanded pattern, otherwise use the default pattern.
-      if (packed_obj.pattern_aligned_) {
+      if (packed_obj.pattern_expanded_) {
         memcpy(constBuf, &packed_obj.expanded_pattern_, kpattern_size32);
       } else {
         memcpy(constBuf, pattern, kpattern_size32);
@@ -2028,8 +2027,8 @@ bool KernelBlitManager::fillBuffer(device::Memory& memory, const void* pattern,
       mem = as_cl<amd::Memory>(gpuCB->owner());
       setArgument(kernels_[fillType], 4, sizeof(cl_mem), &mem, constBufOffset);
 
-      kpattern_size32 /= alignment;
       koffset /= alignment;
+      kpattern_size32 /= alignment;
 
       setArgument(kernels_[fillType], 5, sizeof(uint32_t), &kpattern_size32);
       setArgument(kernels_[fillType], 6, sizeof(koffset), &koffset);