linux-kselftest.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Greg Thelen <gthelen@google.com>
To: Shuah Khan <shuah@kernel.org>
Cc: linux-kernel@vger.kernel.org, linux-kselftest@vger.kernel.org,
	Greg Thelen <gthelen@google.com>
Subject: [PATCH] selftests: more general make nesting support
Date: Tue, 28 Jul 2020 00:32:41 -0700	[thread overview]
Message-ID: <20200728073241.3625015-1-gthelen@google.com> (raw)

selftests can be built from the toplevel kernel makefile (e.g. make
kselftest-all) or directly (make -C tools/testing/selftests all).

The toplevel kernel makefile explicitly disables implicit rules with
"MAKEFLAGS += -rR", which is passed to tools/testing/selftests.  Some
selftest makefiles require implicit make rules, which is why
commit 67d8712dcc70 ("selftests: Fix build failures when invoked from
kselftest target") reenables implicit rules by clearing MAKEFLAGS if
MAKELEVEL=1.

So far so good.  However, if the toplevel makefile is called from an
outer makefile then MAKELEVEL will be elevated, which breaks the
MAKELEVEL equality test.
Example wrapped makefile error:
  $ cat ~/Makefile
  all:
  	$(MAKE) defconfig
  	$(MAKE) kselftest-all
  $ make -sf ~/Makefile
    futex_wait_timeout.c /src/tools/testing/selftests/kselftest_harness.h   /src/tools/testing/selftests/kselftest.h ../include/futextest.h ../include/atomic.h ../include/logging.h -lpthread -lrt -o /src/tools/testing/selftests/futex/functional/futex_wait_timeout
  make[4]: futex_wait_timeout.c: Command not found

Rather than checking $(MAKELEVEL), check for $(LINK.c), which is a more
direct side effect of "make -R".  This enables arbitrary makefile
nesting.

Signed-off-by: Greg Thelen <gthelen@google.com>
---
 tools/testing/selftests/Makefile | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/tools/testing/selftests/Makefile b/tools/testing/selftests/Makefile
index 1195bd85af38..289a2e4b3f6f 100644
--- a/tools/testing/selftests/Makefile
+++ b/tools/testing/selftests/Makefile
@@ -84,10 +84,10 @@ endif
 # of the targets gets built.
 FORCE_TARGETS ?=
 
-# Clear LDFLAGS and MAKEFLAGS if called from main
-# Makefile to avoid test build failures when test
-# Makefile doesn't have explicit build rules.
-ifeq (1,$(MAKELEVEL))
+# Clear LDFLAGS and MAKEFLAGS when implicit rules are missing.  This provides
+# implicit rules to sub-test Makefiles which avoids build failures in test
+# Makefile that don't have explicit build rules.
+ifeq (,$(LINK.c))
 override LDFLAGS =
 override MAKEFLAGS =
 endif
-- 
2.28.0.rc0.142.g3c755180ce-goog


             reply	other threads:[~2020-07-28  7:32 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-07-28  7:32 Greg Thelen [this message]
2020-08-05 19:36 ` [PATCH] selftests: more general make nesting support Greg Thelen
2020-08-05 19:50   ` Shuah Khan
2020-08-05 20:40     ` Greg Thelen

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20200728073241.3625015-1-gthelen@google.com \
    --to=gthelen@google.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-kselftest@vger.kernel.org \
    --cc=shuah@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).