Re: [lttng-dev] babeltrace tests (master/2.0)

Philippe Proulx <eeppelitel...@gmail.com> writes: > On Fri, Sep 15, 2017 at 3:53 PM, Nathan Lynch <nathan_ly...@mentor.com> wrote: >> Nathan Lynch <nathan_ly...@mentor.com> writes: >>> Philippe Proulx <eeppelitel...@gmail.com> writes: >>

Re: [lttng-dev] babeltrace tests (master/2.0)

Nathan Lynch <nathan_ly...@mentor.com> writes: > Philippe Proulx <eeppelitel...@gmail.com> writes: >> >> I asked the Automake mailing list about this >> <http://lists.gnu.org/archive/html/automake/2017-08/msg4.html> and >> we will most pro

Re: [lttng-dev] babeltrace tests (master/2.0)

Philippe Proulx <eeppelitel...@gmail.com> writes: > On Thu, Aug 10, 2017 at 5:40 PM, Nathan Lynch <nathan_ly...@mentor.com> wrote: >> >> Ping? >> > > Sorry. We looked into this. There seems to be no easy way with Automake > to have individual `check` ta

Re: [lttng-dev] babeltrace tests (master/2.0)

Ping? Nathan Lynch <nathan_ly...@mentor.com> writes: > How is the babeltrace test suite on the master and stable-2.0 branches > intended to be used? > > Effectively there are four test suites, one for each definition of TESTS > in the tests/ directory. If one of them has a

[lttng-dev] babeltrace tests (master/2.0)

How is the babeltrace test suite on the master and stable-2.0 branches intended to be used? Effectively there are four test suites, one for each definition of TESTS in the tests/ directory. If one of them has a failure, 'make check' will not proceed to the next unless the -k/--keep-going flag is

Re: [lttng-dev] [RELEASE] Babeltrace 2.0.0-pre3

Michael Jeanson writes: > > This one is my fault, I forgot to stage one file when I submitted > 22e7ae1917a72d89b308345b3bcb97c76362da4f. Jérémie is on vacation for the > week, in the meantime you can apply the attached patch. That works, thanks.

Re: [lttng-dev] [RELEASE] Babeltrace 2.0.0-pre3

(This is with master hence the -pre1 reported, but the code is the same.) I get failures when building with BUILT_IN_PLUGINS as suggested here: https://lists.lttng.org/pipermail/lttng-dev/2017-June/027243.html Is BUILT_IN_PLUGINS intended to accomplish something that libtool cannot? $

[lttng-dev] [PATCH] babeltrace: remove temporary pathnames from test output

a message that won't vary across runs with the same pass/fail status. Signed-off-by: Nathan Lynch <nathan_ly...@mentor.com> --- tests/cli/test_trace_copy.in | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/tests/cli/test_trace_copy.in b/tests/cli/test_trace_copy.in

Re: [lttng-dev] [RELEASE] Babeltrace 2.0.0-pre1

Nathan Lynch <nathan_ly...@mentor.com> writes: > Static linking still needs work, specifically case #3 from this comment: > > https://bugs.lttng.org/issues/1055#note-2 > > Building on Fedora 25 with 'make LDFLAGS=-all-static' gets: > > ../lib/.libs/libbabeltra

[lttng-dev] [PATCH] babeltrace-2.0: do not link with -export-dynamic

It seems that changes in the plugin system make passing -export-dynamic to the linker unnecessary -- a welcome improvement. No regressions observed in the test suite. Signed-off-by: Nathan Lynch <nathan_ly...@mentor.com> --- cli/Makefile.am | 2 +- configure.ac| 4 ++-- 2 files chan

Re: [lttng-dev] [RELEASE] Babeltrace 2.0.0-pre1

Jérémie Galarneau writes: > Known limitations > - > > Being a "pre" release, a number of features are still in active development > and > bugs are being worked on. > > This first release does not expose the Python packages (present in the Git >

[lttng-dev] [PATCH] lttng-tools: remove bogus interpreter line from utils shell library

tests/utils/utils.sh is always sourced, never executed, and /src/bin/bash is not a typical path for a shell interpreter. Just delete it. Signed-off-by: Nathan Lynch <nathan_ly...@mentor.com> --- tests/utils/utils.sh | 2 -- 1 file changed, 2 deletions(-) diff --git a/tests/utils/util

Re: [lttng-dev] lttng upgrade to 2.8.1, need help.

On 11/15/2016 01:14 PM, Aravind HT wrote: > Hi, > > > I was trying to upgrade to lttng-2.8.1 on my yocto build ( using > http://cgit.openembedded.org/cgit.cgi/openembedded-core/tree/meta/recipes-kernel/lttng/lttng-tools_git.bb?h=master) > and I am running into two sets of problems Is there some

Re: [lttng-dev] [PATCH lttng-tools v2] Fix: test cases now rely on explicit workloads

On 11/08/2016 01:49 PM, Francis Deslauriers wrote: > Run a process explicitly in the tracing session to generate the enabled events > rather than relying on the events generated by the lttng CLI. [...] > diff --git a/tests/regression/kernel/test_all_events >

[lttng-dev] [PATCH] lttng-tools: accommodate stricter mktemp implementations in tests

Busybox's mktemp command uses mkstemp(3) which requires the last six characters of the template to be X's. Extend the mktemp templates used in the test scripts. Signed-off-by: Nathan Lynch <nathan_ly...@mentor.com> --- tests/regression/kernel/test_clock_override | 4 ++--

Re: [lttng-dev] [RFC/PATCH 3/3] babeltrace: add simple tests for converter command line options

On 10/26/2016 08:36 AM, Mathieu Desnoyers wrote: > - On Oct 26, 2016, at 3:11 AM, Sebastien Boisvert sboisv...@gydle.com > wrote: > >> Hi Nathan, >> >>> diff --git a/tests/bin/test_clock_seconds b/tests/bin/test_clock_seconds >>> +# [s]+.[n] >>> +timestamp_fmt="\[" >>>

Re: [lttng-dev] [RFC/PATCH 0/3] babeltrace: test harness improvements + new tests

On 10/25/2016 12:43 PM, Michael Jeanson wrote: > On 2016-10-25 12:05, Nathan Lynch wrote: >> On 10/25/2016 10:50 AM, Jonathan Rajotte Julien wrote: >>> >>> On 2016-10-25 11:34 AM, Nathan Lynch wrote: >>>> This is working toward moving from popt to GLib for

Re: [lttng-dev] [RFC/PATCH 0/3] babeltrace: test harness improvements + new tests

On 10/25/2016 10:50 AM, Jonathan Rajotte Julien wrote: > > On 2016-10-25 11:34 AM, Nathan Lynch wrote: >> This is working toward moving from popt to GLib for command-line >> parsing. > > What is the motivation behind this ? Babeltrace already uses GLib extensively

[lttng-dev] [RFC/PATCH 3/3] babeltrace: add simple tests for converter command line options

Add simple tests for the following converter command line options: --no-delta --clock-cycles --clock-seconds --clock-date Signed-off-by: Nathan Lynch <nathan_ly...@mentor.com> --- tests/Makefile.am| 4 tests/bin/Makefile.am| 3 ++- tests/bin/test_clock_cycles

[lttng-dev] [RFC/PATCH 0/3] babeltrace: test harness improvements + new tests

the new tests I've written so far in patch #3. Nathan Lynch (3): babeltrace: use AM_TESTS_ENVIRONMENT for tests babeltrace: drop configure-time substitutions in test scripts babeltrace: add simple tests for converter command line options .gitignore | 10

[lttng-dev] [RFC/PATCH 2/3] babeltrace: drop configure-time substitutions in test scripts

Perform renames of the form test_foo.in => test_foo and rely on the environment as defined by the test harness. Adjust .gitignore accordingly and remove AC_CONFIG_FILES statements. Signed-off-by: Nathan Lynch <nathan_ly...@mentor.com> --- .

[lttng-dev] [RFC/PATCH 1/3] babeltrace: use AM_TESTS_ENVIRONMENT for tests

supplement or override settings in AM_TESTS_ENVIRONMENT by using TESTS_ENVIRONMENT on the command line. Signed-off-by: Nathan Lynch <nathan_ly...@mentor.com> --- tests/Makefile.am | 12 1 file changed, 12 insertions(+) diff --git a/tests/Makefile.am b/tests/Makefile.am index eda9f2

[lttng-dev] [PATCH] babeltrace: address warnings and suggestions from autoscan

that the correct source directory is used at configure time. It also suggests adding checks for a handful of other headers. Signed-off-by: Nathan Lynch <nathan_ly...@mentor.com> --- configure.ac | 8 1 file changed, 8 insertions(+) diff --git a/configure.ac b/configure.ac index 450e22

Re: [lttng-dev] Problem with Babeltrace

On 10/24/2016 03:30 AM, Tim Tannert wrote: > Hey, > > everytime when I start babeltrace: > *babeltrace --version* > > or > *lttng view > > *I'll get following message: > babeltrace: symbol lookup error: babeltrace: undefined symbol: > opt_debug_info_full_path > Seems your babeltrace

Re: [lttng-dev] lttng-tools backport request: Fix: error.h -> common/error.h

On 10/03/2016 10:36 AM, Nathan Lynch wrote: > Can this be backported to the stable-2.8 branch? > > commit 6712db617fe9 > Author: Michael Jeanson <mjean...@efficios.com> > Date: Wed Jun 15 17:18:03 2016 -0400 > > Fix: error.h -> common/error.h > > OE

[lttng-dev] lttng-tools backport request: Fix: error.h -> common/error.h

Can this be backported to the stable-2.8 branch? commit 6712db617fe9 Author: Michael Jeanson Date: Wed Jun 15 17:18:03 2016 -0400 Fix: error.h -> common/error.h OE is carrying this patch for its lttng-tools 2.8 package in order to build against musl libc:

Re: [lttng-dev] [PATCH] babeltrace: add test for babeltrace --list

On 08/18/2016 12:10 PM, Nathan Lynch wrote: > This test simply checks the converter's list of supported formats. > > Signed-off-by: Nathan Lynch <nathan_ly...@mentor.com> > --- > > Currently I expect this to fail for babeltrace configured with > --disable-shared;

[lttng-dev] [PATCH] babeltrace: add test for babeltrace --list

This test simply checks the converter's list of supported formats. Signed-off-by: Nathan Lynch <nathan_ly...@mentor.com> --- Currently I expect this to fail for babeltrace configured with --disable-shared; see https://bugs.lttng.org/issues/1055 .gitignore| 1 + config

[lttng-dev] [PATCH] babeltrace: remove useless PACKAGE_LIBS substitution

PACKAGE_LIBS isn't referred to anywhere in the project so remove this dead code from configure.ac. Signed-off-by: Nathan Lynch <nathan_ly...@mentor.com> --- configure.ac | 1 - 1 file changed, 1 deletion(-) diff --git a/configure.ac b/configure.ac index 39973296ba1d..e600e252d9ce

[lttng-dev] [PATCH] babeltrace: do not use g_quark_from_static_string in unloadable code

cally loaded modules." https://developer.gnome.org/glib/stable/glib-Quarks.html#g-quark-from-static-string Convert call sites of g_quark_from_static_string to g_quark_from_string in code that is potentially unloadable. Call sites in the babeltrace main program remain unchanged. Signed-off-

Re: [lttng-dev] [PATCH] use AC_CHECK_TOOL for objcopy and strip

On 12/11/2015 10:43 AM, Nathan Lynch wrote: > The build steps for some tests refer directly to objcopy and strip, > which when cross-compiling can result in: > > objcopy --only-keep-debug prog prog.debug > objcopy: Unable to recognise the format of the input file `prog' > &

[lttng-dev] [PATCH] use AC_CHECK_TOOL for objcopy and strip

-by: Nathan Lynch <nathan_ly...@mentor.com> --- configure.ac | 2 ++ tests/regression/ust/baddr-statedump/Makefile.am | 6 +++--- tests/regression/ust/ust-dl/Makefile.am | 6 +++--- 3 files changed, 8 insertions(+), 6 deletions(-) diff --git a/config

Re: [lttng-dev] lttng-tracer missing

On 02/19/2015 03:02 AM, Neil Bryan wrote: Hi Nathan, Thank you for the prompt response. I have fixed the kernel tracing problem and indeed your diagnosis was accurate. I'm glad to have helped, but... Step 4) Within the recipe ~/yocto/meta/recipes-kernel/lttng-2.0/lttng-modules_2.0.5.bb

Re: [lttng-dev] lttng-tracer missing

On 02/17/2015 11:02 AM, Neil Bryan wrote: Following the bitbake build I see these kernel modules in the build output directory: ./altera-image-1.0-r0/rootfs/lib/modules/3.7.0/kernel/lttng-modules/lib/lib-ring-buffer.ko

[lttng-dev] lttng-tools 2.6-rc vs. lttng-ust master

By accident I attempted to build lttng-tools stable-2.6 branch against current lttng-ust master, and got: trace-ust.c: In function 'trace_ust_create_event': trace-ust.c:430:11: error: 'struct lttng_ust_event' has no member named 'disabled' lue-attr.disabled = 1; ^ I found the commit

Re: [lttng-dev] [PATCH lttng-tools] Fix: mi test: possible race between listing UST events and testapp start

On 10/26/2014 08:21 PM, Jonathan Rajotte wrote: Signed-off-by: Jonathan Rajotte jonathan.r.jul...@gmail.com More description of the change is needed IMO. --- tests/regression/tools/mi/test_mi | 5 + 1 file changed, 5 insertions(+) diff --git a/tests/regression/tools/mi/test_mi

[lttng-dev] [PATCH] lttng-tools: remove bogus argument from prove invocation

. Signed-off-by: Nathan Lynch nathan_ly...@mentor.com --- tests/run.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tests/run.sh b/tests/run.sh index ce51639f7626..c6c50fd9cff9 100755 --- a/tests/run.sh +++ b/tests/run.sh @@ -19,4 +19,4 @@ [ -z $1 ] echo Error: No testlist

Re: [lttng-dev] [PATCH] lttng-tools: remove bogus argument from prove invocation

On 10/27/2014 09:38 PM, Jonathan Rajotte wrote: Hey On Mon, Oct 27, 2014 at 10:00 PM, Nathan Lynch nathan_ly...@mentor.com mailto:nathan_ly...@mentor.com wrote: An argument of $2 was added to the prove command line in tests/run.sh by 68270f0f604e Mi test: Basic test structure

[lttng-dev] lttng-modules master build failure with ARM kernels

Bisect says that commit 644d6e9c2807041db2b4fd9c3350d4fa243060b1 Remove sys_ prefix from syscall names broke the build for ARM kernels (I've tried 3.16 and 3.17): CC [M] /home/nathanl/src/lttng-modules.git/lttng-syscalls.o /home/nathanl/src/lttng-modules.git/lttng-syscalls.c:281:11:

Re: [lttng-dev] [RFC PATCH 1/1] check if set_schedaffinity works in configure.ac

On 06/08/2014 03:18 PM, Pranith Kumar wrote: On 06/08/2014 04:15 PM, Pranith Kumar wrote: On some systems set_schedaffinity exists but return -1 when called. Tests fail in such cases (test_urcu_hash) Check if we can call set_schedaffinity and define accordingly I just realized that this is

Re: [lttng-dev] [RFC PATCH 1/1] check if set_schedaffinity works in configure.ac

On 06/08/2014 03:31 PM, Pranith Kumar wrote: +AS_IF([test x$sched_set_affinity_args = x3], [ +# test if sched_setaffinity works +AC_MSG_CHECKING([if sched_setaffinity works]) +AC_RUN_IFELSE([AC_LANG_SOURCE([[ +#include sched.h +int

Re: [lttng-dev] might_sleep warnings in overwrite mode

On 11/10/2013 08:18 PM, Nathan Lynch wrote: At this point I cannot trigger the issue without overwrite mode on 3.11.6, but on a 3.8-based vendor kernel I can recreate it regardless of the mode. Some updates on this: - It's not specific to overwrite mode; I was able to provoke it on 3.11.6

Re: [lttng-dev] might_sleep warnings in overwrite mode

On 11/10/2013 08:18 PM, Nathan Lynch wrote: At this point I cannot trigger the issue without overwrite mode on 3.11.6, but on a 3.8-based vendor kernel I can recreate it regardless of the mode. Is it possible that the lttng module code is messing up the current task's preempt count somehow

Re: [lttng-dev] [RFC PATCH lttng-tools 2/2] Implement dynamic instrumentation with dyninst

Hello, On 09/04/2013 06:00 AM, Zifei Tong wrote: This patch provides a *very* basic implementation of dynamic instrumentation support using dyninst library. Looking through dyninst's source and web site, it appears to support only x86 and PowerPC architectures; is this correct?