Vineeth, On Wed, Jul 14, 2010 at 7:39 PM, Vineeth Mekkat <mekka...@umn.edu> wrote: > Hi, > > This is more of a Nehalem question than perfmon one. Apologies if I am > completely out of place. > > I am trying to analyze some benchmarks on Intel Nehalem processor and > see how much of their stall cycles are related to memory access. I > looked at the documents: "Intel® 64 and IA-32 Architectures Optimization > Reference Manual" and "Performance Analysis Guide for Intel® Core™ i7 > Processor and Intel® Xeon™ 5500 processors". > > I could gather that: “Total Cycles“ = UOPS_EXECUTED.CORE_STALLS_CYCLES + > UOPS_EXECUTED.CORE_ACTIVE_CYCLES where Total Cycles is > CPU_CLK_UNHALTED.THREAD. I also understand that memory related accesses > are through ports 2,3 and 4; where as, ALU related operations are > through ports 0, 1 and 5. > > I could find UOPS_EXECUTED.PORT015_STALL_CYCLES counter to get ALU > related stalls but no counter to get memory related stalls. Counter > UOPS_EXECUTED.PORT234_CORE seems to be overall memory UOPS and not stall > cycles. Could anyone suggest how to identify memory related stalls? > You have to read the fine print in the Intel document:
B1H 40H UOPS_EXECUTED.PORT015 Counts number of Uops executed that where issued on port 0, 1, or 5. use cmask=1, invert=1 to count stall cycles In libpfm, I have not created a another unit mask to measure the stall cycles, but you can obtain this easily with pfmon using the info in the Intel document: pfmon -e uops_executed.port015 --counter-mask=1 --inv-mask=1 ....... counter-mask is equivalent to cmask inv-mask is equivalent to inv > Also, for the programs I ran, UOPS_EXECUTED.PORT015_STALL_CYCLES was > greater than UOPS_EXECUTED.CORE_STALLS_CYCLES. Does that make sense? > > Thanks, > Vineeth > > ------------------------------------------------------------------------------ > This SF.net email is sponsored by Sprint > What will you do first with EVO, the first 4G phone? > Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first > _______________________________________________ > perfmon2-devel mailing list > perfmon2-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/perfmon2-devel > ------------------------------------------------------------------------------ This SF.net email is sponsored by Sprint What will you do first with EVO, the first 4G phone? Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first _______________________________________________ perfmon2-devel mailing list perfmon2-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/perfmon2-devel
