TITLE: MITE Micro-ops to IDQISSUE_NAME: MITEUopsToIDQDESCRIPTION: Percentage of micro-ops delivered to the Instruction Decode Queue (IDQ) from the Micro Instruction Translation Engine (MITE), which is the unit that decodes instructions into micro-ops.
TITLE: Bus LockISSUE_NAME: BUS_LOCKDESCRIPTION: Cycles that a BUS LOCK is asserted on the bus. A common atomic lock will not be transferred to a bus lock except in exceptional circumstances of either the memory of the lock residing in uncacheable memory or if the lock extends beyond a cache line boundary splitting cache lines. Atomic lock instructions involve utilizing a lock prefix on the instruction and having the destination operand assigned to a memory address.
TITLE: Short Call ReturnISSUE_NAME: SHORT_CALL_RETURNDESCRIPTION:
TITLE: Zero Length Calls
DESCRIPTION: A “zero length call” uses the attribute of the call instruction to push the immediate instruction pointer on to the stack and then pops off that address into a register. This is accomplished without any matching return on the call. This code construct is also commonly referred to as a zero displacement call.
TITLE: Instruction StarvationISSUE_NAME: InstructionStarvationDESCRIPTION: Execution stalls because the front-end fails to deliver instructions. This metric is the ratio of cycles generated when no instruction was issued by the front-end to all cycles.
TITLE: Instructions Retired Per Clock CycleISSUE_NAME: InstructionsPerClockDESCRIPTION: The ratio of instructions retired per clock cycle (IPC).RELEVANCE:Higher instructions retired per cycle is typically desirable because it represents a larger amount of work being complete in a given unit of time.EXAMPLE: SOLUTION:RELATED_SOURCES:NOTES:
TITLE: Front End Bound
TITLE: x87 Floating Point Store/Restore for RoundingISSUE_NAME: X87_ROUNDINGDESCRIPTION:x87 floating point calculations store/restore from stack to round to 32-bit (single precision) or 64-bit (double precision)RELEVANCE:Negatively impacts most architectures on all OSs but costsAtom's Bonnell, Silverthorne and Saltwell (10-12 cycles) since they cannot store cannot be forwarded to loads in the FP domain.EXAMPLE:Here is an example of how x87 math is done with rounding done by storing/restoring from memory in between calculations:faddfstp [address]fld [address] //The above store is not forward
Does anyone know of workrounds or diagnostics for invalid dump files? The only thing I've seen is a comment to avoid older VTune installation licenses appearing on the "path" and a problem with xed.exe that should now be fixed. On my machine I don't have older vtune information anywhere in the environment or license file, and the data collecting phase proceeded without errors, yet I get an erroras below.
C:\\tmp\\fipde\\pba>java -Xmx1000m xIFJava.Main f "foo.QAG" -jdir "tb5" -NHM
Intel Performance Bottleneck Analyzer - Version 4.0.1 RC3
- Page 6