qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH] build: further refine build.ninja rules


From: Mark Cave-Ayland
Subject: Re: [PATCH] build: further refine build.ninja rules
Date: Thu, 8 Jun 2023 20:21:05 +0100
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101 Thunderbird/102.10.0

On 08/06/2023 14:43, Paolo Bonzini wrote:

In commit b0fcc6fc7fc1 ("build: rebuild build.ninja using
"meson setup --reconfigure"", 2023-05-19) I changed the build.ninja
rule in the Makefile to use "meson setup" so that the Makefile would
pick up a changed path to the meson binary.

However, there was a reason why build.ninja was rebuilt using $(NINJA)
itself.  Namely, ninja has its own cache of file modification times,
and if it does not know about the modification that was done outside
its control, it will *also* try to regenerate build.ninja.  This can be
simply by running "make" on a fresh tree immediately after "configure";
that will trigger an unnecessary meson run.

So, apply a refinement to the rule in order to cover both cases:

- track the meson binary that was used (and that is embedded in
   build.ninja's reconfigure rules); to do this, write build.ninja.stamp
   right after executing meson successfully

- if it changed, force usage of "$(MESON) setup --reconfigure" to
   update the path in the reconfigure rule

- if it didn't change, use "$(NINJA) build.ninja" just like before
   commit b0fcc6fc7fc1.

Reported-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
  Makefile  | 17 +++++++++++++----
  configure |  1 +
  2 files changed, 14 insertions(+), 4 deletions(-)

diff --git a/Makefile b/Makefile
index b22bf6fba12..804a5681e0a 100644
--- a/Makefile
+++ b/Makefile
@@ -83,16 +83,17 @@ config-host.mak: $(SRC_PATH)/configure 
$(SRC_PATH)/scripts/meson-buildoptions.sh
        @if test -f meson-private/coredata.dat; then \
          ./config.status --skip-meson; \
        else \
-         ./config.status && touch build.ninja.stamp; \
+         ./config.status; \
        fi
# 2. meson.stamp exists if meson has run at least once (so ninja reconfigure
  # works), but otherwise never needs to be updated
+
  meson-private/coredata.dat: meson.stamp
  meson.stamp: config-host.mak
        @touch meson.stamp
-# 3. ensure generated build files are up-to-date
+# 3. ensure meson-generated build files are up-to-date
ifneq ($(NINJA),)
  Makefile.ninja: build.ninja
@@ -106,11 +107,19 @@ Makefile.ninja: build.ninja
  endif
ifneq ($(MESON),)
-# A separate rule is needed for Makefile dependencies to avoid -n
+# The path to meson always points to pyvenv/bin/meson, but the absolute
+# paths could change.  In that case, force a regeneration of build.ninja.
+# Note that this invocation of $(NINJA), just like when Make rebuilds
+# Makefiles, does not include -n.
  build.ninja: build.ninja.stamp
  $(build-files):
  build.ninja.stamp: meson.stamp $(build-files)
-       $(MESON) setup --reconfigure $(SRC_PATH) && touch $@
+       @if test "$$(cat build.ninja.stamp)" = "$(MESON)" && test -n 
"$(NINJA)"; then \
+         $(NINJA) build.ninja; \
+       else \
+         echo "$(MESON) setup --reconfigure $(SRC_PATH)"; \
+         $(MESON) setup --reconfigure $(SRC_PATH); \
+       fi && echo "$(MESON)" > $@
Makefile.mtest: build.ninja scripts/mtest2make.py
        $(MESON) introspect --targets --tests --benchmarks | $(PYTHON) 
scripts/mtest2make.py > $@
diff --git a/configure b/configure
index 8a638dd82ae..cbdb389dc95 100755
--- a/configure
+++ b/configure
@@ -1894,6 +1894,7 @@ if test "$skip_meson" = no; then
    if test "$?" -ne 0 ; then
        error_exit "meson setup failed"
    fi
+  echo "$meson" > build.ninja.stamp
  else
    if test -f meson-private/cmd_line.txt; then
      # Adjust old command line options that were removed

Thanks Paolo!

I just tried a complete fresh build, make, a few changes followed by another make and everything built without running reconfigure so:

Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>


ATB,

Mark.




reply via email to

[Prev in Thread] Current Thread [Next in Thread]