|
Packit |
577717 |
2013-11-25
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* a40c96c5 src/components/nvml/linux-nvml.c: nvml component: Add missing }
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 166971ba src/components/nvml/linux-nvml.c: nvml component: modify api
|
|
Packit |
577717 |
checks To check if nvmlDeviceGetEccMode and nvmlDeviceGetPowerUsage are
|
|
Packit |
577717 |
supported, we just call the functions and see if nvml thinks its supported by
|
|
Packit |
577717 |
the card.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-11-21
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 78192de9 delete_before_release.sh: Kill the .gitignore files in
|
|
Packit |
577717 |
delete_before_release
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 60fb1dd4 src/utils/command_line.c: command_line utility: Initialize a
|
|
Packit |
577717 |
variable Initialize data_type to PAPI_DATATYPE_INT64 Addresses a coverity
|
|
Packit |
577717 |
error Error: COMPILER_WARNING: [#def19]
|
|
Packit |
577717 |
papi-5.2.0/src/utils/command_line.c:133:4: warning: 'data_type' may be used
|
|
Packit |
577717 |
uninitialized in this function [-Wmaybe-uninitialized] switch (data_type) { ^
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-11-20
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* da2925f6 src/ctests/data_range.c: Make data_range test use prginfo
|
|
Packit |
577717 |
Coverity complained about prginfo being an unused variable for data_range.c.
|
|
Packit |
577717 |
The code is modified to be stylistically like the code for hw_info in the
|
|
Packit |
577717 |
preceding lines which also is not used elsewhere in the test. This is more
|
|
Packit |
577717 |
to reduce the amount of output in the Coverity scan than to fix this minor
|
|
Packit |
577717 |
issue.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 3386953d src/ctests/data_range.c: Check the return values of PAPI_start()
|
|
Packit |
577717 |
and PAPI_stop() for the data_range test The ia64 data_range test did not
|
|
Packit |
577717 |
check the return values of PAPI_start() or PAPI_stop(). There are propbably
|
|
Packit |
577717 |
few people running this test on ia64 machine, but this is more to eliminate a
|
|
Packit |
577717 |
couple errors noted by a Coverity scan and reduce the clutter in the Coverity
|
|
Packit |
577717 |
scan.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-11-19
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* e704e8f1 src/configure src/configure.in: configure: Build fpapi.h and co
|
|
Packit |
577717 |
for mic When building for mic, set the cross_compiling var in configure to
|
|
Packit |
577717 |
use a native c compiler to build genpapif.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-11-18
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* d32b1dae man/man1/papi_avail.1 man/man1/papi_clockres.1
|
|
Packit |
577717 |
man/man1/papi_command_line.1...: Rebuild the man pages for a 5.3 release
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 4e735d11 doc/Doxyfile-common papi.spec src/Makefile.in...: Bump version
|
|
Packit |
577717 |
numbers for a pending 5.3
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* efe026cd src/Makefile.inc: Makefile.inc: Pass LINKLIB, not SHLIB to the
|
|
Packit |
577717 |
comp_tests
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* f0598acb src/ctests/Makefile.target.in: ctests/Makefile.target.in: Properly
|
|
Packit |
577717 |
catch LINKLIB LINKLIB=$(SHLIB) or $(LIBRARY), so we have to have configure
|
|
Packit |
577717 |
fill in those as well.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 1744c23e src/ctests/Makefile.target.in: ctests/Makefile.target.in: Respect
|
|
Packit |
577717 |
static-tools the --with-static-tools configure flag sets STATIC, not
|
|
Packit |
577717 |
LDFLAGS. This gets passed to the tests' make subprocesses via
|
|
Packit |
577717 |
LDFLAGS="$(LDFLAGS) $(STATIC)" We mimic this in the installed Makefile.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* e9347373 src/ctests/Makefile: ctests/Makefile: Don't clobber value of
|
|
Packit |
577717 |
LIBRARY TOOD: write a better message
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 237219d1 src/Makefile.inc: Makefile.inc: Add enviro vars to fulltest recipe
|
|
Packit |
577717 |
The fulltest target didn't set LD_LIBRARY_PATH and as a result, several
|
|
Packit |
577717 |
tests wouldn't find libpfm and fail to run. The fix is to call our SETPATH
|
|
Packit |
577717 |
command first (as all of the other testing targets do) See
|
|
Packit |
577717 |
------------------------------------------------------------------ icc
|
|
Packit |
577717 |
-diag-disable 188,869,271 -g -g -DSTATIC_PAPI_EVENTS_TABLE
|
|
Packit |
577717 |
-DPEINCLUDE="libpfm4/include/perfmon/perf_event.h" -D_REENTRANT -D_GNU_SOURCE
|
|
Packit |
577717 |
-DUSE_COMPILER_TLS -Ilibpfm4/include -I../../../testlib -I../../.. -I. -o
|
|
Packit |
577717 |
perf_event_offcore_response perf_event_offcore_response.o event_name_lib.o
|
|
Packit |
577717 |
../../../testlib/libtestlib.a ../../../libpapi.so.5.2.0.0 ld: warning:
|
|
Packit |
577717 |
libpfm.so.4, needed by ../../../libpapi.so.5.2.0.0, not found (try using
|
|
Packit |
577717 |
-rpath or -rpath-link) ../../../libpapi.so.5.2.0.0: undefined reference to
|
|
Packit |
577717 |
`pfm_get_event_attr_info' ../../../libpapi.so.5.2.0.0: undefined reference to
|
|
Packit |
577717 |
`pfm_initialize' ../../../libpapi.so.5.2.0.0: undefined reference to
|
|
Packit |
577717 |
`pfm_get_pmu_info' ../../../libpapi.so.5.2.0.0: undefined reference to
|
|
Packit |
577717 |
`pfm_get_version' ../../../libpapi.so.5.2.0.0: undefined reference to
|
|
Packit |
577717 |
`pfm_get_os_event_encoding' ../../../libpapi.so.5.2.0.0: undefined reference
|
|
Packit |
577717 |
to `pfm_get_event_next' ../../../libpapi.so.5.2.0.0: undefined reference to
|
|
Packit |
577717 |
`pfm_get_event_info' ../../../libpapi.so.5.2.0.0: undefined reference to
|
|
Packit |
577717 |
`pfm_strerror' ../../../libpapi.so.5.2.0.0: undefined reference to
|
|
Packit |
577717 |
`pfm_find_event' ../../../libpapi.so.5.2.0.0: undefined reference to
|
|
Packit |
577717 |
`pfm_terminate' make[2]: *** [perf_event_offcore_response] Error 1
|
|
Packit |
577717 |
------------------------------------------------------------------
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-11-17
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* a7f642d2 src/Makefile.inc src/configure src/configure.in: Switch LINKLIB to
|
|
Packit |
577717 |
not have relative pathing
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-11-15
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 91a6fa54 src/components/lustre/tests/Makefile: Fix a typo in the lustre
|
|
Packit |
577717 |
tests' Makefile
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-11-13
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 9a5f9ad4 src/papi_preset.c: papi_preset.c: Fix _papi_load_preset_table func
|
|
Packit |
577717 |
Patch by Gleb Smirnoff ---------------------- The _papi_load_preset_table()
|
|
Packit |
577717 |
loses last entry from a static table. The code in get_event_line() returns
|
|
Packit |
577717 |
value of a char next to the line we are returning. Obviously, for the last
|
|
Packit |
577717 |
entry the char is '\0', so function returns false value and
|
|
Packit |
577717 |
_papi_load_preset_table() ignores the last line. Patch attached. The most
|
|
Packit |
577717 |
important part of my patch is only: - ret =
|
|
Packit |
577717 |
**tmp_perfmon_events_table; + return i; This actually fixes
|
|
Packit |
577717 |
the lost last line. However, I decided to make the entire get_event_line()
|
|
Packit |
577717 |
more robust, protected from bad input, and easier to read.
|
|
Packit |
577717 |
----------------------
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-11-12
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 579139a6 src/utils/hybrid_native_avail.c: more doxygen xml tag cleanup
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 952bb621 src/components/bgpm/CNKunit/linux-CNKunit.c
|
|
Packit |
577717 |
src/components/bgpm/CNKunit/linux-CNKunit.h
|
|
Packit |
577717 |
src/components/bgpm/IOunit/linux-IOunit.c...: Fix doxygen Unsupported
|
|
Packit |
577717 |
xml/html tag warnings
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 0c161015 src/components/micpower/linux-micpower.h: micpower: fix doxygen
|
|
Packit |
577717 |
warning
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* b187f065 src/components/host_micpower/README: host_micpower: update docs
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-11-11
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 4d379c6f src/ctests/p4_lst_ins.c: ctests/p4_lst_ins: Narrow scope of test
|
|
Packit |
577717 |
This test attempted to ensure that it was running on a P4, the test missed
|
|
Packit |
577717 |
for all non intel systems.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-11-10
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* ee1c7967 .../host_micpower/utils/host_micpower_plot.c: Added energy
|
|
Packit |
577717 |
consumption to host_micpower utility.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-11-08
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* eee49912 src/ctests/shlib.c: shlib.c: Check for NULL Thanks to Will Cohen
|
|
Packit |
577717 |
for reporting. Coverity picked up an instance of a value that could be NULL
|
|
Packit |
577717 |
and strlen would barf on it. Error: FORWARD_NULL (CWE-476):
|
|
Packit |
577717 |
papi-5.2.0/src/ctests/shlib.c:70: var_compare_op: Comparing "shinfo->map" to
|
|
Packit |
577717 |
null implies that "shinfo->map" might be null.
|
|
Packit |
577717 |
papi-5.2.0/src/ctests/shlib.c:74: var_deref_model: Passing "shinfo" to
|
|
Packit |
577717 |
function "print_shlib_info_map(PAPI_shlib_info_t const *)", which
|
|
Packit |
577717 |
dereferences null "shinfo->map". papi-5.2.0/src/ctests/shlib.c:13:26:
|
|
Packit |
577717 |
var_assign_parm: Assigning: "map" = "shinfo->map".
|
|
Packit |
577717 |
papi-5.2.0/src/ctests/shlib.c:24:3: deref_var_in_call: Function "strlen(char
|
|
Packit |
577717 |
const *)" dereferences an offset off "map" (which is a copy of
|
|
Packit |
577717 |
"shinfo->map").
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 83c31e25 src/components/perf_event/perf_event.c: perf_event.c: Check return
|
|
Packit |
577717 |
value of ioctl Thanks to Will Cohen for reporting based upon output of
|
|
Packit |
577717 |
coverity.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* e5b33574 src/utils/multiplex_cost.c: multiplex_cost: check return value on
|
|
Packit |
577717 |
PAPI_set_opt Thanks to Will Cohen for reporting based upon output of
|
|
Packit |
577717 |
coverity.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 04f95b14 src/components/.gitignore: Ignore component target makefile
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* cbf7c1a8 src/components/rapl/linux-rapl.c
|
|
Packit |
577717 |
src/components/rapl/tests/Makefile src/components/rapl/tests/rapl_basic.c:
|
|
Packit |
577717 |
Modify linux-rapl to support one wrap-around of the 32-bit registers for
|
|
Packit |
577717 |
reading energy. This insures availability of the full 32-bit dynamic range.
|
|
Packit |
577717 |
However, it does not protect against two wrap-arounds. Care must be taken not
|
|
Packit |
577717 |
to exceed the expected dynamic range, or to check reasonableness of results
|
|
Packit |
577717 |
on completion. Modifications were also made to report rapl events as unscaled
|
|
Packit |
577717 |
binary values in order to compute dynamic ranges. Modify rapl-basic to add a
|
|
Packit |
577717 |
test (rapl_wraparound) to estimate maximum measurement time for a naive gemm.
|
|
Packit |
577717 |
With a -w option, measurement for this amount of time will be performed. The
|
|
Packit |
577717 |
gemm can be replaced with a user kernel for more accurate time estimates.
|
|
Packit |
577717 |
Makefile was modified to support the new test case.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-11-07
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 7784de21 src/ctests/data_range.c src/ctests/zero_shmem.c: Modernize some
|
|
Packit |
577717 |
ctests Add tests_quiet check to data_range and zero_shmem
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-11-06
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 7c953490 src/configure src/configure.in: More MPICC checking Have
|
|
Packit |
577717 |
configure check for mpcc on AIX, in addition to mpicc.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 5c8d2ce0 src/ctests/zero_shmem.c: zero_shmem.c: Fix compiler warning The
|
|
Packit |
577717 |
worker threads in the test print an ID, the test was setup to call
|
|
Packit |
577717 |
pthread_self(), this is problematic. Since each thread is started with a
|
|
Packit |
577717 |
unique work load, use this to lable threads.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 993a6e96 src/ctests/Makefile.recipies src/ctests/Makefile.target.in:
|
|
Packit |
577717 |
ctests/Makefile.recipies: conditionally build the MPI test
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* b29d5f56 src/Makefile.inc src/configure src/configure.in: Check for mpicc
|
|
Packit |
577717 |
at configure time configure[.in]: look for mpicc Makefile.inc: Pass MPICC to
|
|
Packit |
577717 |
ctests' make
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-11-05
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* b2d643df src/papi_events.csv: Add floating point events for IvyBridge Now
|
|
Packit |
577717 |
that Intel has documented them and libpfm4 supports them, PAPI can use them.
|
|
Packit |
577717 |
We just use the same events as on sandybridge. Tested on an ivybridge
|
|
Packit |
577717 |
system.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-11-01
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* c5be5e26 src/components/micpower/linux-micpower.c: micpower: check return
|
|
Packit |
577717 |
of fopen before use Issue reported by Will Cohen from results of Coverity
|
|
Packit |
577717 |
run.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 5c1405ab src/components/host_micpower/utils/Makefile
|
|
Packit |
577717 |
src/components/host_micpower/utils/README
|
|
Packit |
577717 |
.../host_micpower/utils/host_micpower_plot.c: Add host_micpower utility to
|
|
Packit |
577717 |
gather power (and voltage) measurements on Intel Xeon Phi chips using
|
|
Packit |
577717 |
MicAccessAPI.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 46b9bdf5 src/components/host_micpower/linux-host_micpower.c: Added more
|
|
Packit |
577717 |
detailed event description and correct units to host mic power events.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* b97c0126 src/components/host_micpower/linux-host_micpower.c: host_micpower:
|
|
Packit |
577717 |
Better error reporting grab output of dlerror on library load failure
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-10-31
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 84da7fd3 src/components/host_micpower/Rules.host_micpower
|
|
Packit |
577717 |
src/components/host_micpower/tests/Makefile: host_micpower: Fix some makefile
|
|
Packit |
577717 |
bits tests/Makefile needed to define a target to work with the
|
|
Packit |
577717 |
Make_comp_tests install machinery. Rules.host_micpower had a typo
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-10-30
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 14f3e4c4 src/components/host_micpower/linux-host_micpower.c: host_micpower:
|
|
Packit |
577717 |
fix function signature shutdown_thread took wrong arguments.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-10-28
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* a4cc1113 release_procedure.txt: Update release_procedure.txt Bug in the
|
|
Packit |
577717 |
version of doxygen we were using to produce the documentation led to some of
|
|
Packit |
577717 |
the Fortran functions being left out in the cold. We now proscribe 1.8.5
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* a1d6ae34 src/components/host_micpower/README: host_micpower: Add a README
|
|
Packit |
577717 |
file.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-10-25
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 859dbc2c src/Makefile.inc src/components/Makefile_comp_tests
|
|
Packit |
577717 |
src/components/Makefile_comp_tests.target.in...: Make the testsuite as a
|
|
Packit |
577717 |
stand-alone copy-able directory of code These changes to the Makefiles
|
|
Packit |
577717 |
allows the testsuite to be compiled separately from the papi sources. This
|
|
Packit |
577717 |
is useful for people wanting to experiment with the tests and verify that the
|
|
Packit |
577717 |
existing installation of papi works. We put absolute paths to the installed
|
|
Packit |
577717 |
library and include files into the installed makefile for the tests.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* c307ad18 src/ctests/Makefile src/ctests/attach_target.c
|
|
Packit |
577717 |
src/testlib/do_loops.c: Refactor the driver in do_loops.c into its own file.
|
|
Packit |
577717 |
(ctests/Makefile, ctests/attach_target.c testlib/do_loops.c)
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-10-23
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* ace71699 src/components/bgpm/CNKunit/linux-CNKunit.c
|
|
Packit |
577717 |
src/components/bgpm/IOunit/linux-IOunit.c
|
|
Packit |
577717 |
src/components/bgpm/L2unit/linux-L2unit.c...: Passing BGPM errors up to PAPI.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-10-22
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 2ee090ec src/components/bgpm/NWunit/linux-NWunit.c
|
|
Packit |
577717 |
src/components/bgpm/NWunit/linux-NWunit.h: Fixed the behavior in BGQ's NWunit
|
|
Packit |
577717 |
component after attaching an event set to a specific thread that owns the
|
|
Packit |
577717 |
target recourse counting
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 8ab071ee src/components/cuda/linux-cuda.c: CUDA component: Set the number
|
|
Packit |
577717 |
of native events Patch by Steve Kaufmann When running papi_component_avail I
|
|
Packit |
577717 |
notice that the number of CUDA events was always zero when the component was
|
|
Packit |
577717 |
available. The following change correctly sets the number of native events
|
|
Packit |
577717 |
for the component:
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-10-11
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 071943b6 src/configure src/configure.in src/linux-context.h...: add
|
|
Packit |
577717 |
preliminary aarch64 (arm64) support There has been some work to build fedora
|
|
Packit |
577717 |
19 on 64-bit arm armv8 machines (aarch64). I took a look that the why the
|
|
Packit |
577717 |
papi build was failing. The attached is a set of minimal patches to get papi
|
|
Packit |
577717 |
to build. The patch is just a step toward getting aarch64 support for papi.
|
|
Packit |
577717 |
Things are not all there for papi to work in that environment. Still need
|
|
Packit |
577717 |
libpfm to support aarch64 and papi_events.csv describing mappings to machine
|
|
Packit |
577717 |
specific events.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-10-01
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 096eb7fc src/ctests/zero_shmem.c: zero_shmem: cleanup compiler warnings
|
|
Packit |
577717 |
Remove unused variables.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* d9669053 src/ctests/earprofile.c: ctests/earprofile.c: Fix compiler warning
|
|
Packit |
577717 |
Both PAPI_get_hardware_info and PAPI_get_executable_info expect const
|
|
Packit |
577717 |
pointers, (get_executable_info is called by prof_init in profile_utils).
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-09-30
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 87e7e387 src/ctests/p4_lst_ins.c: ctests/p4_lst_ins.c: Fix the P4 load
|
|
Packit |
577717 |
test. This test relied upon a removed symbol to decide if it should run. The
|
|
Packit |
577717 |
symbol unsigned char PENTIUM4 was removed in 2011, update the logic.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 737d91ff src/ctests/zero_shmem.c: ctests/zero_shmem: Update the test *
|
|
Packit |
577717 |
add_test_events expects another argument, update the zero_shmem test's
|
|
Packit |
577717 |
invocation * Protect[Hide] OpenSHMEM calls with ifdefs
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-09-27
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 86c11829 src/ctests/zero_shmem.c: zero_shmem: Include pthread.h
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 2d0e666c src/ctests/zero_smp.c: zero_smp: Change a compile time error to a
|
|
Packit |
577717 |
test_skip In 8d1f2c1, we changed the default assumption to be that all
|
|
Packit |
577717 |
ctests are build. This change allows the test to gracefully skip if it does
|
|
Packit |
577717 |
not have 'native SMP' threads.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-09-26
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 8d1f2c16 src/ctests/Makefile: ctests/Makefile: Default to building
|
|
Packit |
577717 |
everything Set target all to depend upon ALL
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* ffd051cf src/ftests/Makefile src/testlib/Makefile: testlib, ftests
|
|
Packit |
577717 |
Makefiles: cleanup ifort generated files ifort produces mod and f90
|
|
Packit |
577717 |
intermediate files which clean does not cleanup
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* c720bb59 src/components/coretemp/tests/coretemp_basic.c
|
|
Packit |
577717 |
src/components/coretemp/tests/coretemp_pretty.c: Coretemp tests: Fix skipping
|
|
Packit |
577717 |
logic The coretemp_basic test was failing if coretemp was disabled, skip
|
|
Packit |
577717 |
seems more appropriate. Add this logic to the coretemp_pretty test.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* af7f7508 src/configure src/configure.in: configure: refactor CTEST_TARGETS
|
|
Packit |
577717 |
Problem: The set of ctests to build is determined at configure time, in
|
|
Packit |
577717 |
CTEST_TARGETS. This is set in each OS detection section and suffers from
|
|
Packit |
577717 |
neglect. Solution: Try to push the decisions about which tests to build out
|
|
Packit |
577717 |
of configure, ask for them all. Idealy the tests will be written in such a
|
|
Packit |
577717 |
way as to fail/skip gracefully if they lack functionallity, teething problems
|
|
Packit |
577717 |
are expected initially.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 14421695 src/testlib/Makefile: testlib: Fix the Makefile variable
|
|
Packit |
577717 |
assignment Consider: src=a.c b.c c.F obj=$(src:.c=.o) c.o After this
|
|
Packit |
577717 |
substution, obj is {a.o b.o c.F c.o}, not quite the nut. Change the logic to
|
|
Packit |
577717 |
correct that.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-09-17
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 05a4e17b .../perf_event_uncore/perf_event_uncore.c: perf_event_uncore:
|
|
Packit |
577717 |
cleanup a compiler warning _peu_read does not use the hwd_context argument.
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* f2056857 src/papi_events.csv: papi_events.csv: Add PAPI_L1_ICM for Haswell
|
|
Packit |
577717 |
Thanks to Maurice Marks of Unisys for the contribution ------------- I've
|
|
Packit |
577717 |
continued testing on Haswell. By comparison with Vtune and Emon on Haswell I
|
|
Packit |
577717 |
found that we can use the counter L2_RQSTS:ALL_CODE_RD for PAPI_L1_ICM, which
|
|
Packit |
577717 |
is a very useful measure. Attached is my current version of papi_events.csv
|
|
Packit |
577717 |
with Haswell fixes. -------------
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-08-28
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* efe3533d src/Makefile.inc src/components/Makefile_comp_tests
|
|
Packit |
577717 |
src/ctests/Makefile...: testlib: library-ify testlib * Move ftests_util to
|
|
Packit |
577717 |
testlib * Naively create libtestlib.a * utils link to the testlib library *
|
|
Packit |
577717 |
[c|f]tests Switch the tests over to linking libtestlib.a * Component tests
|
|
Packit |
577717 |
link libtestlib.a
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-08-26
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* d2a76dde src/configure src/configure.in src/utils/hybrid_native_avail.c:
|
|
Packit |
577717 |
Gabrial's mic with icc changes to configure. Specify --with-mic at configure
|
|
Packit |
577717 |
time and upon finding icc as the C compiler, it adds --mmic
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 4c0349c0 src/papi_events.csv: papi_events.csv: First draft preset events on
|
|
Packit |
577717 |
HSW Contributed by Nils Smeds ------------------------- Here is a suggestion
|
|
Packit |
577717 |
for addition to Hsw counters. These are not rigorously tested. It compiles
|
|
Packit |
577717 |
and loads. I'm rather uncertain on many of the events so I am hoping that
|
|
Packit |
577717 |
adding events like this will get some useful feedback from the community so
|
|
Packit |
577717 |
that we can improve. -------------------------
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-08-20
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 1b8ff589 src/utils/command_line.c: command_line util: Fix skipping event
|
|
Packit |
577717 |
bug. The command line utility had an extranious index increment which
|
|
Packit |
577717 |
resulted in skipping the reporting of event counts. Remove the increment.
|
|
Packit |
577717 |
Reported by Steve Kaufmann -------------------------- I am getting some funny
|
|
Packit |
577717 |
results when I use papi_command_line with the RAPL events. If I request them
|
|
Packit |
577717 |
all: $ papi_command_line THERMAL_SPEC:PACKAGE0 MINIMUM_POWER:PACKAGE0
|
|
Packit |
577717 |
MAXIMUM_POWER:PACKAGE0 MAXIMUM_TIME_WINDOW:PACKAGE0 PACKAGE_ENERGY:PACKAGE0
|
|
Packit |
577717 |
DRAM_ENERGY:PACKAGE0 PP0_ENERGY:PACKAGE0 Successfully added:
|
|
Packit |
577717 |
THERMAL_SPEC:PACKAGE0 Successfully added: MINIMUM_POWER:PACKAGE0 Successfully
|
|
Packit |
577717 |
added: MAXIMUM_POWER:PACKAGE0 Successfully added:
|
|
Packit |
577717 |
MAXIMUM_TIME_WINDOW:PACKAGE0 Successfully added: PACKAGE_ENERGY:PACKAGE0
|
|
Packit |
577717 |
Successfully added: DRAM_ENERGY:PACKAGE0 Successfully added:
|
|
Packit |
577717 |
PP0_ENERGY:PACKAGE0 THERMAL_SPEC:PACKAGE0 : 115.000 W <<<<<
|
|
Packit |
577717 |
MINIMUM_POWER:PACKAGE0 ?? MAXIMUM_POWER:PACKAGE0 : 180.000 W
|
|
Packit |
577717 |
PACKAGE_ENERGY:PACKAGE0 : 2003784180(u) nJ DRAM_ENERGY:PACKAGE0
|
|
Packit |
577717 |
: 438751220(u) nJ PP0_ENERGY:PACKAGE0 : 1248748779(u) nJ
|
|
Packit |
577717 |
---------------------------------- Verification: Checks for valid event name.
|
|
Packit |
577717 |
This utility lets you add events from the command line interface to see if
|
|
Packit |
577717 |
they work. command_line.c PASSED Note that a value for
|
|
Packit |
577717 |
MINIMUM_POWER:PACKAGE0 is not displayed even though it was successfully added
|
|
Packit |
577717 |
to the event set. In fact, if combined with other events, the value for this
|
|
Packit |
577717 |
event is never displayed. If you specifiy it on its own it is displayed:
|
|
Packit |
577717 |
------------------------------------
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-08-16
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 0cb63d6e src/components/lustre/linux-lustre.c: lustre component: fix memory
|
|
Packit |
577717 |
leak
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-08-13
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* c810cd0d src/components/micpower/linux-micpower.c src/linux-memory.c
|
|
Packit |
577717 |
src/papi_preset.c: Close resource leaks User dcb reported several resource
|
|
Packit |
577717 |
leaks in trac bug #184. -------------------- I just ran the static analysis
|
|
Packit |
577717 |
checker "cppcheck" over the source code of papi-5.2.0 It said 1.
|
|
Packit |
577717 |
[linux-memory.c:711]: (error) Resource leak: sys_cpu 2. [papi_preset.c:735]:
|
|
Packit |
577717 |
(error) Resource leak: fp 3. [components/micpower/linux-micpower.c:166]:
|
|
Packit |
577717 |
(error) Resource leak: fp I've checked them all and they all look like
|
|
Packit |
577717 |
resource leaks to me. Suggest code rework.
|
|
Packit |
577717 |
----------------------------------
|
|
Packit |
577717 |
|
|
Packit |
577717 |
2013-08-07
|
|
Packit |
577717 |
|
|
Packit |
577717 |
* 8d479895 doc/Makefile: Doxygen makefile: update dependencies The manpages
|
|
Packit |
577717 |
are generated from comments in papi.h, papi.c, papi_hl.c and
|
|
Packit |
577717 |
papi_fwrappers.c; update the make dependencies to reflect this.
|