public inbox for openembedded-core@lists.openembedded.org
 help / color / mirror / Atom feed
From: Richard Purdie <richard.purdie@linuxfoundation.org>
To: openembedded-core <openembedded-core@lists.openembedded.org>
Subject: [PATCH] autotools: Correctly filter m4 files view to setscene dependencies
Date: Tue, 13 Sep 2016 10:53:40 +0100	[thread overview]
Message-ID: <1473760420.7207.30.camel@linuxfoundation.org> (raw)

Currently when you run builds from sstate, you can see warnings like:

WARNING: systemd-1_230+gitAUTOINC+3a74d4fc90-r0 do_configure: /data/poky-master/tmp-glibc/sstate-control/manifest-intel-corei7-64-glibc-initial.populate_sysroot not found
WARNING: systemd-1_230+gitAUTOINC+3a74d4fc90-r0 do_configure: /data/poky-master/tmp-glibc/sstate-control/manifest-intel-corei7-64-libgcc-initial.populate_sysroot not found

This is due to do_configure wanting to copy a limited number of m4 macros,
only listed in a recipes DEPENDS but that set is still larger than the set of
recipes which get restored from sstate.

For build determinism and to avoid these warnings, we need to make this
function match what the sstate code does. We really don't want to duplicate
the functionality since keeping things in sync would be hard so we create
a data structure which can be passed into the same underlying function,
setscene_depvalid().

[YOCTO #10030]

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

diff --git a/meta/classes/autotools.bbclass b/meta/classes/autotools.bbclass
index 5fda03d..92aa738 100644
--- a/meta/classes/autotools.bbclass
+++ b/meta/classes/autotools.bbclass
@@ -137,6 +137,8 @@ do_configure[postfuncs] += "autotools_postconfigure"
 ACLOCALDIR = "${WORKDIR}/aclocal-copy"
 
 python autotools_copy_aclocals () {
+    import copy
+
     s = d.getVar("AUTOTOOLS_SCRIPT_PATH", True)
     if not os.path.exists(s + "/configure.in") and not os.path.exists(s + "/configure.ac"):
         if not d.getVar("AUTOTOOLS_COPYACLOCAL", False):
@@ -165,36 +167,63 @@ python autotools_copy_aclocals () {
     if start is None:
         bb.fatal("Couldn't find ourself in BB_TASKDEPDATA?")
 
-    # We need to find configure tasks which are either from <target> -> <target>
-    # or <native> -> <native> but not <target> -> <native> unless they're direct
-    # dependencies. This mirrors what would get restored from sstate.
-    done = [start]
-    next = [start]
+    # We need to figure out which m4 files we need to expose to this do_configure task.
+    # This needs to match what would get restored from sstate, which is controlled 
+    # ultimately by calls from bitbake to setscene_depvalid().
+    # That function expects a setscene dependency tree. We add a dummy "do_populate_sysroot"
+    # task for ourselves (depending on do_configure) and then build a dependency tree 
+    # condensed to do_populate_sysroot -> do_populate_sysroot dependencies, similar to 
+    # that used by setscene tasks. We can then call into setscene_depvalid() and decide
+    # which dependencies we can "see" and should expose the m4 files for.
+    setscenedeps = copy.deepcopy(taskdepdata)
+
+    # Add dummy do_populate_sysroot task depending on the real do_configure
+    dummystart = start.replace("do_configure", "do_populate_sysroot")
+    setscenedeps[dummystart] = copy.copy(setscenedeps[start])
+    setscenedeps[dummystart][1] = "do_populate_sysroot"
+    setscenedeps[dummystart][3] = set()
+    setscenedeps[dummystart][3].add(start)
+
+    # Create collapsed do_populate_sysroot -> do_populate_sysroot tree
+    for dep in taskdepdata:
+        data = setscenedeps[dep]        
+        if data[1] != "do_populate_sysroot":
+            for dep2 in setscenedeps:
+                data2 = setscenedeps[dep2]
+                if dep in data2[3]:
+                    data2[3].update(setscenedeps[dep][3])
+                    data2[3].remove(dep)
+            del setscenedeps[dep]
+
+    # Remove circular references
+    for dep in setscenedeps:
+        if dep in setscenedeps[dep][3]:
+            setscenedeps[dep][3].remove(dep)
+
+    # Call into setscene_depvalid for each dependency and only copy m4 files
+    # for ones that would be restored from sstate.
+    done = [dummystart]
+    next = [dummystart]
     while next:
         new = []
         for dep in next:
-            data = taskdepdata[dep]
+            data = setscenedeps[dep]
             for datadep in data[3]:
                 if datadep in done:
                     continue
-                if (not data[0].endswith("-native")) and taskdepdata[datadep][0].endswith("-native") and dep != start:
-                    continue
                 done.append(datadep)
+
+                taskdeps = {}
+                taskdeps[dep] = setscenedeps[dep][:2]
+                taskdeps[datadep] = setscenedeps[datadep][:2]
+                retval = setscene_depvalid(dep, taskdeps, [], d)
+                if retval:
+                    bb.note("Skipping setscene dependency %s for m4 macro copying" % datadep)
+                    continue
                 new.append(datadep)
-                if taskdepdata[datadep][1] == "do_configure":
-                    configuredeps.append(taskdepdata[datadep][0])
+                configuredeps.append(setscenedeps[datadep][0])
         next = new
 
-    #configuredeps2 = []
-    #for dep in taskdepdata:
-    #    data = taskdepdata[dep]
-    #    if data[1] == "do_configure" and data[0] != pn:
-    #        configuredeps2.append(data[0])
-    #configuredeps.sort()
-    #configuredeps2.sort()
-    #bb.warn(str(configuredeps))
-    #bb.warn(str(configuredeps2))
-
     cp = []
     if nodeps:
         bb.warn("autotools: Unable to find task dependencies, -b being used? Pulling in all m4 files")




             reply	other threads:[~2016-09-13  9:53 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-09-13  9:53 Richard Purdie [this message]
2016-09-13 13:56 ` [PATCH v2] autotools: Correctly filter m4 files view to setscene dependencies Richard Purdie

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=1473760420.7207.30.camel@linuxfoundation.org \
    --to=richard.purdie@linuxfoundation.org \
    --cc=openembedded-core@lists.openembedded.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