kernel/linux.git/tools/perf/util/sort.c, branch v3.16.4

perf tools: Add dcacheline sort

2014-06-09T11:34:49+00:00

In perf's 'mem-mode', one can get access to a whole bunch of details specific to a particular sample instruction. A bunch of those details relate to the data address. One interesting thing you can do with data addresses is to convert them into a unique cacheline they belong too. Organizing these data cachelines into similar groups and sorting them can reveal cache contention. This patch creates an alogorithm based on various sample details that can help group entries together into data cachelines and allows 'perf report' to sort on it. The algorithm relies on having proper mmap2 support in the kernel to help determine if the memory map the data address belongs to is private to a pid or globally shared. The alogortithm is as follows: o group cpumodes together o group entries with discovered maps together o sort on major, minor, inode and inode generation numbers o if userspace anon, then sort on pid o sort on cachelines based on data addresses The 'dcacheline' sort option in 'perf report' only works in 'mem-mode'. Sample output: # # Samples: 206 of event 'cpu/mem-loads/pp' # Total weight : 2534 # Sort order : dcacheline,pid # # Overhead Samples Data Cacheline Command: Pid # ........ ............ ...................................................................... .................. # 13.22% 1 [k] 0xffff88042f08ebc0 swapper: 0 9.27% 1 [k] 0xffff88082e8cea80 swapper: 0 3.59% 2 [k] 0xffffffff819ba180 swapper: 0 0.32% 1 [k] arch_trigger_all_cpu_backtrace_handler_na.23901+0xffffffffffffffe0 swapper: 0 0.32% 1 [k] timekeeper_seq+0xfffffffffffffff8 swapper: 0 Note: Added a '+1' to symlen size in hists__calc_col_len to prevent the next column from prematurely tabbing over and mis-aligning. Not sure what the problem is. Signed-off-by: Don Zickus Link: http://lkml.kernel.org/r/1401208087-181977-8-git-send-email-dzickus@redhat.com Signed-off-by: Jiri Olsa

perf tools: Move elide bool into perf_hpp_fmt struct

2014-06-03T19:34:59+00:00

After output/sort fields refactoring, it's expensive to check the elide bool in its current location inside the 'struct sort_entry'. The perf_hpp__should_skip function gets highly noticable in workloads with high number of output/sort fields, like for: $ perf report -i perf-test.data -F overhead,sample,period,comm,pid,dso,symbol,cpu --stdio Performance report: 9.70% perf [.] perf_hpp__should_skip Moving the elide bool into the 'struct perf_hpp_fmt', which makes the perf_hpp__should_skip just single struct read. Got speedup of around 22% for my test perf.data workload. The change should not harm any other workload types. Performance counter stats for (10 runs): before: 358,319,732,626 cycles ( +- 0.55% ) 467,129,581,515 instructions # 1.30 insns per cycle ( +- 0.00% ) 150.943975206 seconds time elapsed ( +- 0.62% ) now: 278,785,972,990 cycles ( +- 0.12% ) 370,146,797,640 instructions # 1.33 insns per cycle ( +- 0.00% ) 116.416670507 seconds time elapsed ( +- 0.31% ) Acked-by: Namhyung Kim Cc: Arnaldo Carvalho de Melo Cc: Corey Ashford Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/20140601142622.GA9131@krava.brq.redhat.com Signed-off-by: Jiri Olsa

perf tools: Remove elide setup for SORT_MODE__MEMORY mode

2014-06-03T19:34:53+00:00

There's no need to setup elide of sort_dso sort entry again with symbol_conf.dso_list list. The only difference were list names of memory mode data, which does not make much sense to me. Acked-by: Namhyung Kim Cc: Arnaldo Carvalho de Melo Cc: Corey Ashford Cc: David Ahern Cc: Frederic Weisbecker Cc: Ingo Molnar Cc: Namhyung Kim Cc: Paul Mackerras Cc: Peter Zijlstra Cc: Stephane Eranian Link: http://lkml.kernel.org/r/1400858147-7155-2-git-send-email-jolsa@kernel.org Signed-off-by: Jiri Olsa

perf tools: Reset output/sort order to default

2014-06-01T12:35:10+00:00

When reset_output_field() is called, also reset field/sort order to NULL so that it can have the default values. It's needed for testing. Signed-off-by: Namhyung Kim CC: Arun Sharma Cc: Frederic Weisbecker Link: http://lkml.kernel.org/r/1401335910-16832-26-git-send-email-namhyung@kernel.org Signed-off-by: Jiri Olsa

perf ui/hist: Add support to accumulated hist stat

2014-06-01T12:35:00+00:00

Print accumulated stat of a hist entry if requested. To do that, add new HPP_PERCENT_ACC_FNS macro and generate a perf_hpp_fmt using it. The __hpp__sort_acc() function sorts entries by accumulated period value. When accumulated periods of two entries are same (i.e. single path callchain) put the caller above since accumulation tends to put callers on higher position for obvious reason. Also add "overhead_children" output field to be selected by user. Signed-off-by: Namhyung Kim Tested-by: Arun Sharma Tested-by: Rodrigo Campos Cc: Frederic Weisbecker Link: http://lkml.kernel.org/r/1401335910-16832-11-git-send-email-namhyung@kernel.org Signed-off-by: Jiri Olsa

perf tools: Introduce reset_output_field()

2014-05-21T09:45:37+00:00

The reset_output_field() function is for clearing output field settings and will be used for test code in later patch. Signed-off-by: Namhyung Kim Link: http://lkml.kernel.org/r/1400480762-22852-19-git-send-email-namhyung@kernel.org Signed-off-by: Jiri Olsa

perf tools: Get rid of obsolete hist_entry__sort_list

2014-05-21T09:45:37+00:00

Now we moved to the perf_hpp_[_sort]_list so no need to keep the old hist_entry__sort_list and sort__first_dimension. Also the hist_entry__sort_snprintf() can be gone as hist_entry__snprintf() provides the functionality. Signed-off-by: Namhyung Kim Link: http://lkml.kernel.org/r/1400480762-22852-18-git-send-email-namhyung@kernel.org Signed-off-by: Jiri Olsa

perf hists: Reset width of output fields with header length

2014-05-21T09:45:36+00:00

Some fields missed to set default column length so it broke align in --stdio output. Add perf_hpp__reset_width() to set it to a sane default value. Note that this change will ignore -w/--column-widths option for now. Before: $ perf report -F cpu,comm,overhead --stdio ... # CPU Command Overhead # ............... ........ # 0 firefox 2.65% 0 kworker/0:0 1.45% 0 swapper 5.52% 0 synergys 0.92% 1 firefox 4.54% After: # CPU Command Overhead # ... ............... ........ # 0 firefox 2.65% 0 kworker/0:0 1.45% 0 swapper 5.52% 0 synergys 0.92% 1 firefox 4.54% Signed-off-by: Namhyung Kim Link: http://lkml.kernel.org/r/1400480762-22852-17-git-send-email-namhyung@kernel.org Signed-off-by: Jiri Olsa

perf tools: Skip elided sort entries

2014-05-21T09:45:36+00:00

When it converted sort entries to hpp formats, it missed se->elide handling, so add it for compatibility. Signed-off-by: Namhyung Kim Link: http://lkml.kernel.org/r/1400480762-22852-16-git-send-email-namhyung@kernel.org Signed-off-by: Jiri Olsa

perf tools: Add ->sort() member to struct sort_entry

2014-05-21T09:45:35+00:00

Currently, what the sort_entry does is just identifying hist entries so that they can be grouped properly. However, with -F option support, it indeed needs to sort entries appropriately to be shown to users. So add ->sort() member to do it. Signed-off-by: Namhyung Kim Acked-by: Ingo Molnar Link: http://lkml.kernel.org/r/1400480762-22852-13-git-send-email-namhyung@kernel.org Signed-off-by: Jiri Olsa