On Thu, Nov 01, 2012 at 05:38:01PM -0300, Arnaldo Carvalho de Melo wrote: > Em Tue, Oct 30, 2012 at 11:01:41PM +0100, Jiri Olsa escreveu: > > hi, > > this code tests the link between perf stat/record command line > > options and final perf_event_attr struct values. Also it tests > > the group fd linkage. > > > > Plus moving test related objects under 'test' directory. > > Ok, so I applied this to my local perf/core branch with a few changes, > namely to move 24/25 to after 5/25 so that I could do, for each > subsequent tests: > > $ make -j8 -C tools/perf/ O=/home/acme/git/build/perf install > $ perf test -v perf_event_attr > > And see if each one passed. > > So at the end I changed > /home/acme/libexec/perf-core/tests/attr/base-record flipping > "watermark=0" to "watermark=1" and got: > > [root@sandy ~]# perf test perf_event_attr > 13: struct perf_event_attr setup :FAILED > '/home/acme/libexec/perf-core/tests/attr/test-record-count' - match failure > FAILED! > [root@sandy ~]# > > But I wanted it to show _what_ didn't match, so I tried: > > [root@sandy ~]# perf test -v perf_event_attr > 13: struct perf_event_attr setup : > --- start --- > running '/home/acme/libexec/perf-core/tests/attr/test-record-count' > FAILED '/home/acme/libexec/perf-core/tests/attr/test-record-count' - match > failure > ---- end ---- > struct perf_event_attr setup: FAILED! > [root@sandy ~]#
yep, you'd need to run: python ./tests/attr.py -d ./tests/attr/ -p ./perf -t test-record-count -vvv to see what really did not match 'perf test attr' takes just single -v option into account could be changed ;) > > Also is there a strict need for setting PERF_TEST_ATTR? I thought that if it > found the /home/acme/libexec/perf-core/tests/attr/ dirs it would test it, but > haven't dug deep, care to ellaborate here? PERF_TEST_ATTR instructs perf to store each event details into PERF_TEST_ATTR directory and python script is running 'PERF_TEST_ATTR=/tmp/temp perf ...' to get events > > Ah, I just pushed perf/core with this patchset, thanks! thanks, jirka -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
