selftests/futex: Fix makefile dependencies

PATCH In review Stafford Horne 3 participants 8 messages Jan 6, 2017 – Jan 11, 2017

Subsystems (2)

Kselftest Futex

lore

Share & tools

lore

Stafford Horne

author

2017-01-06 04:27 UTC #0 lore raw

PATCH 1/1 selftests/futex: Fix makefile dependencies

I am working on doing selftests for openrisc and found issues with the futex test is not building after changes to the tests source.

This issue is that the TARGETS variable used in the futex Makefile is also used by the selftests Makefile, so when building from toplevel the TARGETS variable is clobbered and nothing gets built. Changed the variable name to get around that.

Also, the futex makefile did not contain dependencies for all headers, so if we make changes to logging.h rebuild will not happen. Add headers to fix it up.

Signed-off-by Stafford Horne <[email protected]>

---
 tools/testing/selftests/futex/functional/Makefile | 15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)

tools/testing/selftests/futex/functional/Makefile +9 -6

index 9d6b75e..1546cbd 100644
@@ -2,8 +2,11 @@ INCLUDES := -I../include -I../../
 CFLAGS := $(CFLAGS) -g -O2 -Wall -D_GNU_SOURCE -pthread $(INCLUDES)
 LDFLAGS := $(LDFLAGS) -pthread -lrt
 
-HEADERS := ../include/futextest.h
-TARGETS := \
+HEADERS := \
+	../include/futextest.h \
+	../include/atomic.h \
+	../include/logging.h
+TEST_TARGETS := \
 	futex_wait_timeout \
 	futex_wait_wouldblock \
 	futex_requeue_pi \

@@ -12,14 +15,14 @@ TARGETS := \
 	futex_wait_uninitialized_heap \
 	futex_wait_private_mapped_file
 
-TEST_PROGS := $(TARGETS) run.sh
+TEST_PROGS := $(TEST_TARGETS) run.sh
 
 .PHONY: all clean
-all: $(TARGETS)
+all: $(TEST_TARGETS)
 
-$(TARGETS): $(HEADERS)
+$(TEST_TARGETS): $(HEADERS)
 
 include ../../lib.mk
 
 clean:
-	rm -f $(TARGETS)
+	rm -f $(TEST_TARGETS)

Signature · 1 line

2.9.3

Darren Hart

reviewer Reviewed-by on v1 1/1

2017-01-10 22:10 UTC #1 lore raw

Quoted 8 lines · Stafford Horne

I am working on doing selftests for openrisc and found issues with the futex test is not building after changes to the tests source.
This issue is that the TARGETS variable used in the futex Makefile is also used by the selftests Makefile, so when building from toplevel the TARGETS variable is clobbered and nothing gets built. Changed the variable name to get around that.

Quoted 5 lines

Also, the futex makefile did not contain dependencies for all headers, so if we make changes to logging.h rebuild will not happen. Add headers to fix it up.
Signed-off-by Stafford Horne <[email protected]>

Thanks for catching this and the fix.

+Shuah Khan

Note: This appears also to be a problem for intel_pstate/Makefile

Reviewed-by Darren Hart <[email protected]>

Quoted 45 lines

---
 tools/testing/selftests/futex/functional/Makefile | 15 +++++++++------
 1 file changed, 9 insertions(+), 6 deletions(-)

tools/testing/selftests/futex/functional/Makefile +9 -6

index 9d6b75e..1546cbd 100644
@@ -2,8 +2,11 @@ INCLUDES := -I../include -I../../
 CFLAGS := $(CFLAGS) -g -O2 -Wall -D_GNU_SOURCE -pthread $(INCLUDES)
 LDFLAGS := $(LDFLAGS) -pthread -lrt
 
-HEADERS := ../include/futextest.h
-TARGETS := \
+HEADERS := \
+	../include/futextest.h \
+	../include/atomic.h \
+	../include/logging.h
+TEST_TARGETS := \
 	futex_wait_timeout \
 	futex_wait_wouldblock \
 	futex_requeue_pi \

@@ -12,14 +15,14 @@ TARGETS := \
 	futex_wait_uninitialized_heap \
 	futex_wait_private_mapped_file
 
-TEST_PROGS := $(TARGETS) run.sh
+TEST_PROGS := $(TEST_TARGETS) run.sh
 
 .PHONY: all clean
-all: $(TARGETS)
+all: $(TEST_TARGETS)
 
-$(TARGETS): $(HEADERS)
+$(TEST_TARGETS): $(HEADERS)
 
 include ../../lib.mk
 
 clean:
-	rm -f $(TARGETS)
+	rm -f $(TEST_TARGETS)

Signature · 1 line

2.9.3

Signature · 2 lines

Darren Hart
Intel Open Source Technology Center

Darren Hart

reviewer

2017-01-10 22:17 UTC #2 lore raw

Quoted 13 lines · Darren Hart

Quoted 12 lines · Stafford Horne
I am working on doing selftests for openrisc and found issues with the futex test is not building after changes to the tests source.
This issue is that the TARGETS variable used in the futex Makefile is also used by the selftests Makefile, so when building from toplevel the TARGETS variable is clobbered and nothing gets built. Changed the variable name to get around that.
Also, the futex makefile did not contain dependencies for all headers, so if we make changes to logging.h rebuild will not happen. Add headers to fix it up.

Quoted 7 lines

Quoted 1 line
Signed-off-by Stafford Horne <[email protected]>
Thanks for catching this and the fix.
+Shuah Khan
Note: This appears also to be a problem for intel_pstate/Makefile

Or, it would be if intel_pstate was listed in selftests/Makefile :-)

Signature · 2 lines

Darren Hart
Intel Open Source Technology Center

Stafford Horne

author

2017-01-11 04:37 UTC #3 lore raw

Quoted 16 lines · Darren Hart

Quoted 15 lines · Darren Hart
Quoted 13 lines · Stafford Horne
I am working on doing selftests for openrisc and found issues with the futex test is not building after changes to the tests source.
This issue is that the TARGETS variable used in the futex Makefile is also used by the selftests Makefile, so when building from toplevel the TARGETS variable is clobbered and nothing gets built. Changed the variable name to get around that.
Also, the futex makefile did not contain dependencies for all headers, so if we make changes to logging.h rebuild will not happen. Add headers to fix it up.
Signed-off-by Stafford Horne <[email protected]>

Quoted 7 lines

Quoted 5 lines
Thanks for catching this and the fix.
+Shuah Khan
Note: This appears also to be a problem for intel_pstate/Makefile
Or, it would be if intel_pstate was listed in selftests/Makefile :-)

Right, thanks for having a look. I actually have a small patch to also add <stdio.h> to logging.h. If you have not already fixed the intel_pstate test I'll send them all together.

Although I am working on testing openrisc, doing a intel_pstate test on my box shouldnt be much of a problem :)

-Stafford

Darren Hart

reviewer

2017-01-11 08:08 UTC #4 lore raw

Quoted 21 lines · Stafford Horne

Quoted 20 lines · Darren Hart
Quoted 19 lines · Darren Hart
Quoted 8 lines · Stafford Horne
I am working on doing selftests for openrisc and found issues with the futex test is not building after changes to the tests source.
This issue is that the TARGETS variable used in the futex Makefile is also used by the selftests Makefile, so when building from toplevel the TARGETS variable is clobbered and nothing gets built. Changed the variable name to get around that.
Quoted 5 lines
Also, the futex makefile did not contain dependencies for all headers, so if we make changes to logging.h rebuild will not happen. Add headers to fix it up.
Signed-off-by Stafford Horne <[email protected]>
Thanks for catching this and the fix.
+Shuah Khan

Quoted 7 lines

Quoted 3 lines
Quoted 1 line
Note: This appears also to be a problem for intel_pstate/Makefile
Or, it would be if intel_pstate was listed in selftests/Makefile :-)
Right, thanks for having a look. I actually have a small patch to also add <stdio.h> to logging.h. If you have not already fixed the intel_pstate test I'll send them all together.

Please do, thanks!

Signature · 2 lines

Darren Hart
Intel Open Source Technology Center

Stafford Horne

author

2017-01-11 08:48 UTC #5 lore raw

Quoted 24 lines · Darren Hart

Quoted 23 lines · Stafford Horne
Quoted 22 lines · Darren Hart
Quoted 20 lines · Darren Hart
Quoted 8 lines · Stafford Horne
I am working on doing selftests for openrisc and found issues with the futex test is not building after changes to the tests source.
This issue is that the TARGETS variable used in the futex Makefile is also used by the selftests Makefile, so when building from toplevel the TARGETS variable is clobbered and nothing gets built. Changed the variable name to get around that.
Quoted 5 lines
Also, the futex makefile did not contain dependencies for all headers, so if we make changes to logging.h rebuild will not happen. Add headers to fix it up.
Signed-off-by Stafford Horne <[email protected]>
Thanks for catching this and the fix.
+Shuah Khan
Note: This appears also to be a problem for intel_pstate/Makefile

Quoted 7 lines

Quoted 5 lines
Quoted 1 line
Or, it would be if intel_pstate was listed in selftests/Makefile :-)
Right, thanks for having a look. I actually have a small patch to also add <stdio.h> to logging.h. If you have not already fixed the intel_pstate test I'll send them all together.
Please do, thanks!

Hello,
Actually I just had a look at Shuah's linux-kselftest repo on kernel.org. It looks like Bamvor Jian has actually fixed a lot of these issues on the next branch.

I will see what is missing if anything.

-Stafford

Shuah Khan

maintainer

2017-01-11 14:41 UTC #6 lore raw

Quoted 33 lines · Stafford Horne

Quoted 24 lines · Darren Hart
Quoted 23 lines · Stafford Horne
Quoted 22 lines · Darren Hart
Quoted 20 lines · Darren Hart
Quoted 8 lines · Stafford Horne
I am working on doing selftests for openrisc and found issues with the futex test is not building after changes to the tests source.
This issue is that the TARGETS variable used in the futex Makefile is also used by the selftests Makefile, so when building from toplevel the TARGETS variable is clobbered and nothing gets built. Changed the variable name to get around that.
Quoted 5 lines
Also, the futex makefile did not contain dependencies for all headers, so if we make changes to logging.h rebuild will not happen. Add headers to fix it up.
Signed-off-by Stafford Horne <[email protected]>
Thanks for catching this and the fix.
+Shuah Khan
Note: This appears also to be a problem for intel_pstate/Makefile
Quoted 7 lines
Quoted 5 lines
Quoted 1 line
Or, it would be if intel_pstate was listed in selftests/Makefile :-)
Right, thanks for having a look. I actually have a small patch to also add <stdio.h> to logging.h. If you have not already fixed the intel_pstate test I'll send them all together.
Please do, thanks!

Quoted 10 lines

Hello,
Actually I just had a look at Shuah's linux-kselftest repo on kernel.org. It looks like Bamvor Jian has actually fixed a lot of these issues on the next branch.
I will see what is missing if anything.
-Stafford

Hi Stafford,

Thanks for looking into the issues. Yes Bemovar fixed several issues. Could you please send me the patches you alre

If you think [PATCH] selftests/futex: Fix makefile dependencies should be pulled in, please send it to me. I just Darren's response, don't have the original patch email in my Inbox.

thanks,
-- Shuah

Stafford Horne

author

2017-01-11 15:30 UTC #7 lore raw

Quoted 45 lines · Shuah Khan

Quoted 33 lines · Stafford Horne
Quoted 24 lines · Darren Hart
Quoted 23 lines · Stafford Horne
Quoted 22 lines · Darren Hart
Quoted 20 lines · Darren Hart
Quoted 8 lines · Stafford Horne
I am working on doing selftests for openrisc and found issues with the futex test is not building after changes to the tests source.
This issue is that the TARGETS variable used in the futex Makefile is also used by the selftests Makefile, so when building from toplevel the TARGETS variable is clobbered and nothing gets built. Changed the variable name to get around that.
Quoted 5 lines
Also, the futex makefile did not contain dependencies for all headers, so if we make changes to logging.h rebuild will not happen. Add headers to fix it up.
Signed-off-by Stafford Horne <[email protected]>
Thanks for catching this and the fix.
+Shuah Khan
Note: This appears also to be a problem for intel_pstate/Makefile
Quoted 7 lines
Quoted 5 lines
Quoted 1 line
Or, it would be if intel_pstate was listed in selftests/Makefile :-)
Right, thanks for having a look. I actually have a small patch to also add <stdio.h> to logging.h. If you have not already fixed the intel_pstate test I'll send them all together.
Please do, thanks!
Quoted 10 lines
Hello,
Actually I just had a look at Shuah's linux-kselftest repo on kernel.org. It looks like Bamvor Jian has actually fixed a lot of these issues on the next branch.
I will see what is missing if anything.
-Stafford

Quoted 8 lines

Hi Stafford,
Thanks for looking into the issues. Yes Bemovar fixed several issues. Could you please send me the patches you alre
If you think [PATCH] selftests/futex: Fix makefile dependencies should be pulled in, please send it to me. I just Darren's response, don't have the original patch email in my Inbox.

Hello Shuah,

There was a conflict with my patch and Bamvor's. As you make have seen, I have resolved those conflicts and sent a few more patches in new series.

2/6 is what has become of the original patch that Darren reviewed. I didn't keep Reviewed-by since its completely different now, I will wait for review of the others.

-Stafford