From patchwork Fri Jun 16 14:57:50 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Masahiro Yamada X-Patchwork-Id: 13282867 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 0D8DAEB64D7 for ; Fri, 16 Jun 2023 14:58:01 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1345746AbjFPO6A (ORCPT ); Fri, 16 Jun 2023 10:58:00 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:55874 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S229706AbjFPO57 (ORCPT ); Fri, 16 Jun 2023 10:57:59 -0400 Received: from dfw.source.kernel.org (dfw.source.kernel.org [IPv6:2604:1380:4641:c500::1]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 2CEB830C5; Fri, 16 Jun 2023 07:57:58 -0700 (PDT) Received: from smtp.kernel.org (relay.kernel.org [52.25.139.140]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits)) (No client certificate requested) by dfw.source.kernel.org (Postfix) with ESMTPS id AE4376336C; Fri, 16 Jun 2023 14:57:57 +0000 (UTC) Received: by smtp.kernel.org (Postfix) with ESMTPSA id 356D7C433C0; Fri, 16 Jun 2023 14:57:55 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1686927476; bh=MPjQQ3JtAC+4df4LxJ4VAYqSVq149IdkIDQqJgcx6ms=; h=From:To:Cc:Subject:Date:From; b=cin4XiA8F5ZrvkEgAT9yCstY6jWTY4WQgdfjRhVHvNTEV8fdJk09mFVVX3xCL7f5R hcioGNhqMxMv0U/BP+D0OP9hCZ/NNav+wpI+5qAdE2rvRS2TYLv5jBtMvazOGR/hMZ IRGQqv408IoiGCY/os9eA/HvFsgLxiIE63aBgAVsJKhhZJ1W4vosbQgoDlTW5Wh+jO oU5wHGj/AC3K4C4jhGQG6YsUqpygPqP5B1oRoErskfYHjRxdGE7Icrxomhp3++l3Pu ZLzB7bqS8oRvhYGzb/2iHO6s3DbJQpl/b+YOA1LqqNvHPgUM9DFzekfmFK4RWPsSWJ KHFbM24MgslJg== From: Masahiro Yamada To: linux-kbuild@vger.kernel.org Cc: linux-kernel@vger.kernel.org, Nathan Chancellor , Nick Desaulniers , Nicolas Schier , Masahiro Yamada , David Howells Subject: [PATCH v2 1/2] kbuild: revive "Entering directory" for Make >= 4.4.1 Date: Fri, 16 Jun 2023 23:57:50 +0900 Message-Id: <20230616145751.945864-1-masahiroy@kernel.org> X-Mailer: git-send-email 2.39.2 MIME-Version: 1.0 Precedence: bulk List-ID: X-Mailing-List: linux-kbuild@vger.kernel.org With commit 9da0763bdd82 ("kbuild: Use relative path when building in a subdir of the source tree"), compiler messages in out-of-tree builds include relative paths, which are relative to the build directory, not the directory where make was started. To help IDEs/editors find the source files, Kbuild lets GNU Make print "Entering directory ..." when it changes the working directory. It has been working fine for a long time, but David reported it is broken with the latest GNU Make. The behavior was changed by GNU Make commit 8f9e7722ff0f ("[SV 63537] Fix setting -w in makefiles"). Previously, setting --no-print-directory to MAKEFLAGS only affected child makes, but it is now interpreted in the current make as soon as it is set. [test code] $ cat /tmp/Makefile ifneq ($(SUBMAKE),1) MAKEFLAGS += --no-print-directory all: ; $(MAKE) SUBMAKE=1 else all: ; : endif [before 8f9e7722ff0f] $ make -C /tmp make: Entering directory '/tmp' make SUBMAKE=1 : make: Leaving directory '/tmp' [after 8f9e7722ff0f] $ make -C /tmp make SUBMAKE=1 : Previously, the effect of --no-print-directory was delayed until Kbuild started the directory descending, but it is no longer true with GNU Make 4.4.1. This commit adds one more recursion to cater to GNU Make >= 4.4.1. When Kbuild needs to change the working directory, __submake will be executed twice. __submake without --no-print-directory --> show "Entering directory ..." __submake with --no-print-directory --> parse the rest of Makefile We end up with one more recursion than needed for GNU Make < 4.4.1, but I do not want to complicate the version check. Reported-by: David Howells Closes: https://lore.kernel.org/all/2427604.1686237298@warthog.procyon.org.uk/ Signed-off-by: Masahiro Yamada Tested-by: Nicolas Schier --- Changes in v2: - Fix the code. V1 did not work. - Remove the version check to simplify the code Makefile | 36 ++++++++++++++++++++---------------- 1 file changed, 20 insertions(+), 16 deletions(-) diff --git a/Makefile b/Makefile index cc3fe09c4dec..916c1a7984b0 100644 --- a/Makefile +++ b/Makefile @@ -189,13 +189,6 @@ else abs_objtree := $(CURDIR) endif # ifneq ($(KBUILD_OUTPUT),) -ifeq ($(abs_objtree),$(CURDIR)) -# Suppress "Entering directory ..." unless we are changing the work directory. -MAKEFLAGS += --no-print-directory -else -need-sub-make := 1 -endif - this-makefile := $(lastword $(MAKEFILE_LIST)) abs_srctree := $(realpath $(dir $(this-makefile))) @@ -214,6 +207,23 @@ endif export abs_srctree abs_objtree export sub_make_done := 1 +endif # sub_make_done + +ifeq ($(abs_objtree),$(CURDIR)) +# Suppress "Entering directory ..." if we are at the final work directory. +no-print-directory := --no-print-directory +else +# Recursion to show "Entering directory ..." +need-sub-make := 1 +endif + +ifeq ($(filter --no-print-directory, $(MAKEFLAGS)),) +# If --no-print-directory is unset, recurse once again to set it. +# You may end up with recursing into __sub-make twice. This is due to the +# behavior change for GNU Make 4.4.1. +need-sub-make := 1 +endif + ifeq ($(need-sub-make),1) PHONY += $(MAKECMDGOALS) __sub-make @@ -223,18 +233,12 @@ $(filter-out $(this-makefile), $(MAKECMDGOALS)) __all: __sub-make # Invoke a second make in the output directory, passing relevant variables __sub-make: - $(Q)$(MAKE) -C $(abs_objtree) -f $(abs_srctree)/Makefile $(MAKECMDGOALS) + $(Q)$(MAKE) $(no-print-directory) -C $(abs_objtree) \ + -f $(abs_srctree)/Makefile $(MAKECMDGOALS) -endif # need-sub-make -endif # sub_make_done +else # need-sub-make # We process the rest of the Makefile if this is the final invocation of make -ifeq ($(need-sub-make),) - -# Do not print "Entering directory ...", -# but we want to display it when entering to the output directory -# so that IDEs/editors are able to understand relative filenames. -MAKEFLAGS += --no-print-directory ifeq ($(abs_srctree),$(abs_objtree)) # building in the source tree