[Gems-users] IFETCHs and the store buffer

["Rubén Titos" <rtitos@xxxxxxxxxxx>]

Hi everyone,


Taking a look at doRequest in ruby/system/Sequencer.C, I just came across this code that totally confused me (after the fast path checking). Assuming TSO consistency, the store buffer is checked to see if the load (LD) can be entirely 
satisfied from it. Since Sequencer::makeRequest is the only one that inserts into the store buffer (by calling insertStore), and it does so only with writes (ST, ST_XACT, LOG_ST, LDX_XACT, ATOMIC), why are IFETCHs also meaningful here?

Cheers,

Rubén

______

bool Sequencer::doRequest(const CacheMsg& request) {

  // Check the fast path
[...]

  if (TSO && (request.getType() == CacheRequestType_LD || 
request.getType() == CacheRequestType_IFETCH)) {

    // See if we can satisfy the load entirely from the store buffer
    SubBlock subblock(line_address(request.getAddress()), request.getSize());
    if (m_chip_ptr->m_L1Cache_storeBuffer_vec[m_version]->trySubBlock(subblock)) {
      DataBlock dummy;
      hitCallback(request, dummy, GenericMachineType_NULL);  // Call with an 'empty' datablock, since the data is in the store buffer
      return true;
    }
  }