870407 - Move CMMSRCS to moz.build

Status: RESOLVED FIXED

(Firefox Build System :: General, defect)

Comment 1

[Joey Armstrong [:joey]]

Attachment: Add CMMSRCS as a moz.build passthrough variable.

Comment 2

[Gregory Szorc [:gps]]

Comment on attachment 749295 [details] [diff] [review]
Add CMMSRCS as a moz.build passthrough variable.

Review of attachment 749295 [details] [diff] [review]:
-----------------------------------------------------------------

::: python/mozbuild/mozbuild/frontend/sandbox_symbols.py
@@ +62,5 @@
>  
>          This variable contains a list of files to invoke the assembler on.
>          """),
>  
> +    'CMMSRCS': (list, [],

Can we just name this CMM_SOURCES immediately? The automatic conversion tool supports mapping to a different variable name and not doing the conversion now just creates more work for everyone later.

@@ +63,5 @@
>          This variable contains a list of files to invoke the assembler on.
>          """),
>  
> +    'CMMSRCS': (list, [],
> +        """ Objective-c++ source files.

Nit: No space between after triple quotes.

Comment 3

[Joey Armstrong [:joey]]

(In reply to Gregory Szorc [:gps] from comment #2)
> Comment on attachment 749295 [details] [diff] [review]
> Add CMMSRCS as a moz.build passthrough variable.
> 
> Review of attachment 749295 [details] [diff] [review]:
> -----------------------------------------------------------------
> 
> ::: python/mozbuild/mozbuild/frontend/sandbox_symbols.py
> @@ +62,5 @@
> >  
> >          This variable contains a list of files to invoke the assembler on.
> >          """),
> >  
> > +    'CMMSRCS': (list, [],
> 
> Can we just name this CMM_SOURCES immediately? The automatic conversion tool
> supports mapping to a different variable name and not doing the conversion
> now just creates more work for everyone later.

Well what is the more important priority around the mozbuild conversions ?  Complete the transition over to mozbuild so Makefile.in can disappear or delay the conversion to work on deploying final answers ?

Renaming variable seems like a simple enough op but will add more work and a potential failure condition with make(var) -> moz(var) -> make::backend.mk(var).

Comment 4

[Joey Armstrong [:joey]]

Attachment: move CMMSRCS to moz.build (logic).

Comment 5

[Joey Armstrong [:joey]]

Comment on attachment 753937 [details] [diff] [review]
move CMMSRCS to moz.build (logic).

Add CMMSRCS as a passthrough variable.
Rebase to pickup sorted order attribute for file lists.

Comment 6

[Joey Armstrong [:joey]]

Comment on attachment 753937 [details] [diff] [review]
move CMMSRCS to moz.build (logic).

ping on the code review

Comment 7

[Ed Morley [:emorley]]

https://hg.mozilla.org/mozilla-central/rev/73ef965a7b66

Comment 8

[Joey Armstrong [:joey]]

[leave open] whiteboard keyword was omitted.
re-open, patch added CMMSOURCS as a mozbuild variable but still need to convert makefiles.

Comment 9

[Joey Armstrong [:joey]]

Attachment: move CMMSRCS to moz.build (file batch #1).

Comment 10

[Joey Armstrong [:joey]]

Comment on attachment 759348 [details] [diff] [review]
move CMMSRCS to moz.build (file batch #1).

First batch of CMMSRC files, tested on mac.  Try coverage pending.

Comment 11

[Michael Shal [:mshal]]

Comment on attachment 759348 [details] [diff] [review]
move CMMSRCS to moz.build (file batch #1).

Looks good!

Comment 12

[Joey Armstrong [:joey]]

Comment on attachment 759348 [details] [diff] [review]
move CMMSRCS to moz.build (file batch #1).

try results: http://tbpl.mozilla.org/?tree=Try&rev=99dcf1ed4faf

Comment 13

[Joey Armstrong [:joey]]

Comment on attachment 759348 [details] [diff] [review]
move CMMSRCS to moz.build (file batch #1).

Inbound push https://hg.mozilla.org/integration/mozilla-inbound/rev/c936da8dd34a

Comment 14

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/c936da8dd34a

Comment 15

[Joey Armstrong [:joey]]

Attachment: move CMMSRCS to moz.build (file batch #2).

Comment 16

[jarmstrong]

Attachment: move CMMSRCS to mozbuild

toolkit/xre/ having problems on mac 10.{6,7}.  Break patch #2 up into smaller chunks for now.

Comment 17

[jarmstrong]

Attachment: move CMMSRCS to mozbuild (file batch #3).

Comment 18

[Joey Armstrong [:joey]]

Comment on attachment 768369 [details] [diff] [review]
move CMMSRCS to mozbuild (file batch #3).

Try job: https://tbpl.mozilla.org/?tree=Try&rev=c342d4a93877
osx debug failure related to 'leakstats': 887234, 774844

Comment 19

[jarmstrong]

Attachment: move CMMSRCS to mozbuild (file batch #2)

Finally a clean try run for this patch:
http://tbpl.mozilla.org/?tree=Try&rev=3c5b2742a399

Comment 20

[jarmstrong]

Attachment: move CMMSRCS to mozbuild (file batch #3)

Comment 21

[jarmstrong]

Attachment: move CMMSRCS to mozbuild (file batch #3)

Try job: http://tbpl.mozilla.org/?tree=Try&rev=a2862d47ea72

3rd batch of CMMSRCS files.

Comment 22

[Michael Shal [:mshal]]

Comment on attachment 769671 [details] [diff] [review]
move CMMSRCS to mozbuild (file batch #2)

>diff --git a/dom/plugins/ipc/interpose/moz.build b/dom/plugins/ipc/interpose/moz.build
>--- a/dom/plugins/ipc/interpose/moz.build
>+++ b/dom/plugins/ipc/interpose/moz.build
>@@ -1,8 +1,12 @@
> # -*- Mode: python; c-basic-offset: 4; indent-tabs-mode: nil; tab-width: 40 -*-
> # vim: set filetype=python:
> # This Source Code Form is subject to the terms of the Mozilla Public
> # License, v. 2.0. If a copy of the MPL was not distributed with this
> # file, You can obtain one at http://mozilla.org/MPL/2.0/.
> 
> LIBRARY_NAME = 'plugin_child_interpose'
> 
>+CMMSRCS += [
>+    "%s.mm" % (LIBRARY_NAME),
>+    'plugin_child_quirks.mm',
>+]

I think in a previous conversion (can't remember which one) we were going for more explicit definitions where possible. So this could just be:

CMMSRCS += [
    'plugin_child_interpose.mm',
    'plugin_child_quirks.mm',
]

Up to you if you want to change it, but I don't really see a reason to keep CMMSRCS dependent on LIBRARY_NAME.

Comment 23

[jarmstrong]

Attachment: move CMMSRCS to mozbuild (file batch #3)

Ignore but retain last attachment 769677 [details]/file batch #3.  Partial information was recorded by an inadvertent [enter] press which left navigation for the attachment in a bad state.  Attempting to replace patch was not able to correct missing {Diff, Review} links in the attachment action panel.

Same patch content as before.  Move 3rd batch of CMMSRCS files to mozbuild.

Comment 24

[Michael Shal [:mshal]]

Comment on attachment 769688 [details] [diff] [review]
move CMMSRCS to mozbuild (file batch #3)

>diff --git a/ipc/chromium/moz.build b/ipc/chromium/moz.build
>--- a/ipc/chromium/moz.build
>+++ b/ipc/chromium/moz.build
>@@ -207,8 +207,26 @@ if CONFIG['_MSC_VER']:
>         'debug_on_start.cc',
>     ]
> 
> ost = CONFIG['OS_TEST']
> if ost.find('86') == -1 and ost.find('arm') == -1 and ost.find('mips') == -1:
>     CPP_SOURCES += [
>         'atomicops_internals_mutex.cc',
>     ]
>+
>+if CONFIG['OS_ARCH'] == 'Darwin' or CONFIG['OS_MACOSX']:
>+    CMMSRCS += [
>+        'base_paths_mac.mm',
>+        'chrome_application_mac.mm',
>+        'chrome_paths_mac.mm',
>+        'file_util_mac.mm',
>+        'file_version_info_mac.mm',
>+        'mac_util.mm',
>+        'mach_ipc_mac.mm',
>+        'message_pump_mac.mm',
>+        'platform_thread_mac.mm',
>+        'platform_util_mac.mm',
>+        'process_util_mac.mm',
>+        'scoped_nsautorelease_pool.mm',
>+        'sys_string_conversions_mac.mm',
>+        'worker_pool_mac.mm',
>+    ]

Why do we need the OS_ARCH == 'Darwin' test here? The Makefile only has the OS_MACOSX check.

Leaving r? until this is resolved.

>diff --git a/netwerk/wifi/moz.build b/netwerk/wifi/moz.build
>--- a/netwerk/wifi/moz.build
>+++ b/netwerk/wifi/moz.build
>@@ -42,8 +42,13 @@ elif CONFIG['OS_ARCH'] == 'SunOS':
> 
> if CONFIG['MOZ_ENABLE_DBUS']:
>     CPP_SOURCES += [
>         'nsWifiScannerDBus.cpp',
>     ]
> 
> LIBRARY_NAME = 'neckowifi_s'
> 
>+
>+if CONFIG['OS_ARCH'] == 'Darwin':
>+    CMMSRCS += [
>+        'osx_corewlan.mm',
>+    ]

Please move this in with the same Darwin if-block earlier in the moz.build file:

if CONFIG['OS_ARCH'] == 'Darwin':
    CPP_SOURCES += [
        'nsWifiScannerMac.cpp',
    ]
+    CMMSRCS += [
+        'osx_corewlan.mm',
+    ]

Comment 25

[jarmstrong]

Comment on attachment 769671 [details] [diff] [review]
move CMMSRCS to mozbuild (file batch #2)

Inbound push for batch #2
https://hg.mozilla.org/integration/mozilla-inbound/rev/932964100cd2
committed changeset 137035:932964100cd2

Comment 26

[jarmstrong]

(In reply to Michael Shal [:mshal] from comment #24)
> Comment on attachment 769688 [details] [diff] [review]
> move CMMSRCS to mozbuild (file batch #3)
> 
> >diff --git a/ipc/chromium/moz.build b/ipc/chromium/moz.build
> >--- a/ipc/chromium/moz.build
> >+++ b/ipc/chromium/moz.build
> >@@ -207,8 +207,26 @@ if CONFIG['_MSC_VER']:
> >         'debug_on_start.cc',
> >     ]
> > 
> >+if CONFIG['OS_ARCH'] == 'Darwin' or CONFIG['OS_MACOSX']:
> >+    CMMSRCS += [
> >+        'base_paths_mac.mm',
> >+    ]
> 
> Why do we need the OS_ARCH == 'Darwin' test here? The Makefile only has the
> OS_MACOSX check.

ipc/chromium/chromium-config.mk
===============================
ifeq ($(OS_ARCH),Darwin) # {

OS_MACOSX = 1
DEFINES += \
  -DOS_MACOSX=1 \
  $(NULL)

For consistency with the other mac mozbuild conditionals.  OS_MACOSX is only defined for use in ipc/chromium/ and the value is a redeclaration of OS_ARCH=='Darwin' -- likely to match a compiler define.


> >diff --git a/netwerk/wifi/moz.build b/netwerk/wifi/moz.build
> >--- a/netwerk/wifi/moz.build
> >+++ b/netwerk/wifi/moz.build
> >@@ -42,8 +42,13 @@ elif CONFIG['OS_ARCH'] == 'SunOS':
> > 
> > if CONFIG['MOZ_ENABLE_DBUS']:
> >     CPP_SOURCES += [
> >         'nsWifiScannerDBus.cpp',
> >     ]
> > 
> > LIBRARY_NAME = 'neckowifi_s'
> > 
> >+
> >+if CONFIG['OS_ARCH'] == 'Darwin':
> >+    CMMSRCS += [
> >+        'osx_corewlan.mm',
> >+    ]
> 
> Please move this in with the same Darwin if-block earlier in the moz.build
> file:
> 
> if CONFIG['OS_ARCH'] == 'Darwin':
>     CPP_SOURCES += [
>         'nsWifiScannerMac.cpp',
>     ]
> +    CMMSRCS += [
> +        'osx_corewlan.mm',
> +    ]


There may be more of these on the way.  The conversion is not attempting to merge into existing conditional blocks so some like this may slip through.

Comment 27

[Michael Shal [:mshal]]

(In reply to jarmstrong from comment #26)
> (In reply to Michael Shal [:mshal] from comment #24)
> > Comment on attachment 769688 [details] [diff] [review]
> > move CMMSRCS to mozbuild (file batch #3)
> > 
> > >diff --git a/ipc/chromium/moz.build b/ipc/chromium/moz.build
> > >--- a/ipc/chromium/moz.build
> > >+++ b/ipc/chromium/moz.build
> > >@@ -207,8 +207,26 @@ if CONFIG['_MSC_VER']:
> > >         'debug_on_start.cc',
> > >     ]
> > > 
> > >+if CONFIG['OS_ARCH'] == 'Darwin' or CONFIG['OS_MACOSX']:
> > >+    CMMSRCS += [
> > >+        'base_paths_mac.mm',
> > >+    ]
> > 
> > Why do we need the OS_ARCH == 'Darwin' test here? The Makefile only has the
> > OS_MACOSX check.
> 
> ipc/chromium/chromium-config.mk
> ===============================
> ifeq ($(OS_ARCH),Darwin) # {
> 
> OS_MACOSX = 1
> DEFINES += \
>   -DOS_MACOSX=1 \
>   $(NULL)
> 
> For consistency with the other mac mozbuild conditionals.  OS_MACOSX is only
> defined for use in ipc/chromium/ and the value is a redeclaration of
> OS_ARCH=='Darwin' -- likely to match a compiler define.

So should it just be:

if CONFIG['OS_ARCH'] == 'Darwin':
    CMMSRCS += [
        'base_paths_mac.mm',
    ]

I don't see CONFIG['OS_MACOSX'] anywhere else in moz.build - if OS_MACOSX is just from chromium-config.mk then I don't think we can use it.

Comment 28

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

(In reply to Michael Shal [:mshal] from comment #27)
> (In reply to jarmstrong from comment #26)
> > (In reply to Michael Shal [:mshal] from comment #24)
> > > Comment on attachment 769688 [details] [diff] [review]
> > > move CMMSRCS to mozbuild (file batch #3)
> > > 
> > > >diff --git a/ipc/chromium/moz.build b/ipc/chromium/moz.build
> > > >--- a/ipc/chromium/moz.build
> > > >+++ b/ipc/chromium/moz.build
> > > >@@ -207,8 +207,26 @@ if CONFIG['_MSC_VER']:
> > > >         'debug_on_start.cc',
> > > >     ]
> > > > 
> > > >+if CONFIG['OS_ARCH'] == 'Darwin' or CONFIG['OS_MACOSX']:
> > > >+    CMMSRCS += [
> > > >+        'base_paths_mac.mm',
> > > >+    ]
> > > 
> > > Why do we need the OS_ARCH == 'Darwin' test here? The Makefile only has the
> > > OS_MACOSX check.
> > 
> > ipc/chromium/chromium-config.mk
> > ===============================
> > ifeq ($(OS_ARCH),Darwin) # {
> > 
> > OS_MACOSX = 1
> > DEFINES += \
> >   -DOS_MACOSX=1 \
> >   $(NULL)
> > 
> > For consistency with the other mac mozbuild conditionals.  OS_MACOSX is only
> > defined for use in ipc/chromium/ and the value is a redeclaration of
> > OS_ARCH=='Darwin' -- likely to match a compiler define.
> 
> So should it just be:
> 
> if CONFIG['OS_ARCH'] == 'Darwin':
>     CMMSRCS += [
>         'base_paths_mac.mm',
>     ]
> 
> I don't see CONFIG['OS_MACOSX'] anywhere else in moz.build - if OS_MACOSX is
> just from chromium-config.mk then I don't think we can use it.

Or use os_macosx which is defined in ipc/chromium/moz.build.

Comment 29

[Michael Shal [:mshal]]

(In reply to :Ms2ger from comment #28)
> (In reply to Michael Shal [:mshal] from comment #27)
> > I don't see CONFIG['OS_MACOSX'] anywhere else in moz.build - if OS_MACOSX is
> > just from chromium-config.mk then I don't think we can use it.
> 
> Or use os_macosx which is defined in ipc/chromium/moz.build.

Oh, yeah - this makes the most sense to me :)

Comment 30

[Joey Armstrong [:joey]]

(In reply to Michael Shal [:mshal] from comment #29)
> (In reply to :Ms2ger from comment #28)
> > (In reply to Michael Shal [:mshal] from comment #27)
> > > I don't see CONFIG['OS_MACOSX'] anywhere else in moz.build - if OS_MACOSX is
> > > just from chromium-config.mk then I don't think we can use it.
> > 
> > Or use os_macosx which is defined in ipc/chromium/moz.build.
> 
> Oh, yeah - this makes the most sense to me :)


Looking at the conditional block I'm staring to think it may be cleaner/clearer to drop most the local os_* vars (except os_posix) and inline platform config variable tests where needed:

os_freebsd=, os_netbsd= & os_openbsd= are currently unused.
There is a 1-to-1 mapping between $OS_ARCH and os_${arch}:
    os_bsd => if CONFIG['OS_ARCH'] in ('FreeBSD', 'NetBSD', 'OpenBSD')

If the locals are replaced we retain consistent syntax for testing platforms in mozbuild.
Longer term should add functions like isbsd(), ismac(), etc to avoid typo problems while testing for platform name but that could be a bug for later on.

Comment 31

[Michael Shal [:mshal]]

(In reply to Joey Armstrong [:joey] from comment #30)
> (In reply to Michael Shal [:mshal] from comment #29)
> > (In reply to :Ms2ger from comment #28)
> > > (In reply to Michael Shal [:mshal] from comment #27)
> > > > I don't see CONFIG['OS_MACOSX'] anywhere else in moz.build - if OS_MACOSX is
> > > > just from chromium-config.mk then I don't think we can use it.
> > > 
> > > Or use os_macosx which is defined in ipc/chromium/moz.build.
> > 
> > Oh, yeah - this makes the most sense to me :)
> 
> 
> Looking at the conditional block I'm staring to think it may be
> cleaner/clearer to drop most the local os_* vars (except os_posix) and
> inline platform config variable tests where needed:
> 
> os_freebsd=, os_netbsd= & os_openbsd= are currently unused.
> There is a 1-to-1 mapping between $OS_ARCH and os_${arch}:
>     os_bsd => if CONFIG['OS_ARCH'] in ('FreeBSD', 'NetBSD', 'OpenBSD')
> 
> If the locals are replaced we retain consistent syntax for testing platforms
> in mozbuild.
> Longer term should add functions like isbsd(), ismac(), etc to avoid typo
> problems while testing for platform name but that could be a bug for later
> on.

Some of the other vars (freebsd and netbsd) are still used in the Makefile.in, so once we port DEFINES over they will be used in the moz.build file. We should probably just use os_macosx for CMMSRCS now, and file a cleanup bug to make this more general. I don't think that needs to be solved to get CMMSRCS done, though.

Comment 32

[Joey Armstrong [:joey]]

Attachment: move CMMSRCS to mozbuild (file batch #3).

Comment 33

[Joey Armstrong [:joey]]

Attachment: move CMMSRCS to mozbuild (file batch #3)

Moved 2 nit CMMSRCS assignments up into earlier Darwin conditional blocks.

Comment 34

[Michael Shal [:mshal]]

Comment on attachment 770213 [details] [diff] [review]
move CMMSRCS to mozbuild (file batch #3)

Looks ready to me!

Comment 35

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/932964100cd2

Comment 36

[Joey Armstrong [:joey]]

Comment on attachment 770213 [details] [diff] [review]
move CMMSRCS to mozbuild (file batch #3)

Inbound push, eyes slighty crossed as these are starting to all look the same.
https://hg.mozilla.org/integration/mozilla-inbound/rev/de57cc8ba195
committed changeset 137217:de57cc8ba195

Comment 37

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/de57cc8ba195

Comment 38

[Joey Armstrong [:joey]]

Attachment: cleanup bug.

Comment 39

[Joey Armstrong [:joey]]

Comment on attachment 770978 [details] [diff] [review]
cleanup bug.

Cleanup for patch #1-3

Comment 40

[Joey Armstrong [:joey]]

Attachment: move CMMSRCS to mozbuild (file batch #4).

Comment 41

[Joey Armstrong [:joey]]

Comment on attachment 770996 [details] [diff] [review]
move CMMSRCS to mozbuild (file batch #4).

Final batch of CMMSRCS files

Comment 42

[Michael Shal [:mshal]]

Comment on attachment 770978 [details] [diff] [review]
cleanup bug.

Looks good!

Comment 43

[Michael Shal [:mshal]]

Comment on attachment 770996 [details] [diff] [review]
move CMMSRCS to mozbuild (file batch #4).

>diff --git a/uriloader/exthandler/moz.build b/uriloader/exthandler/moz.build
>--- a/uriloader/exthandler/moz.build
>+++ b/uriloader/exthandler/moz.build
>@@ -59,18 +59,21 @@ CPP_SOURCES += [
>     'ExternalHelperAppParent.cpp',
>     'nsExternalHelperAppService.cpp',
>     'nsExternalProtocolHandler.cpp',
>     'nsLocalHandlerApp.cpp',
>     'nsMIMEInfoImpl.cpp',
> ]
> 
> if CONFIG['MOZ_WIDGET_TOOLKIT'] == 'cocoa':
>-    # TODO: CMMSRCS go here
>-    pass
>+    CMMSRCS = [
>+        'nsLocalHandlerAppMac.mm',
>+        'nsMIMEInfoMac.mm',
>+        'nsOSHelperAppService.mm',
>+    ]

nit: Prefer 'CMMSRCS +=' over 'CMMSRCS ='

Looks ready to go - always nice to see a TODO comment go away :)

Comment 44

[Joey Armstrong [:joey]]

Comment on attachment 770996 [details] [diff] [review]
move CMMSRCS to mozbuild (file batch #4).

Fixed =/+= assignment {byproduct of a manual edit}.

Try job: http://tbpl.mozilla.org/?tree=Try&rev=3352acc891e0
todo: check on mac failures.

Comment 45

[Joey Armstrong [:joey]]

Attachment: move CMMSRCS to mozbuild (file batch #4).

Comment 46

[Ryan VanderMeulen [:RyanVM]]

https://hg.mozilla.org/mozilla-central/rev/73a7afa43d5a

Comment 47

[Joey Armstrong [:joey]]

Comment on attachment 772123 [details] [diff] [review]
move CMMSRCS to mozbuild (file batch #4).

Converted file batch #4 for CMMSRCS.

Try results: https://tbpl.mozilla.org/?tree=Try&rev=83e9324021f1
  o oth failure on osx 10.6 was due to timeout waiting for an hg command.

Comment 48

[Michael Shal [:mshal]]

Comment on attachment 772123 [details] [diff] [review]
move CMMSRCS to mozbuild (file batch #4).

># HG changeset patch
># User Joey Armstrong <[e-mail]>
># Date 1372883135 14400
># Node ID 2a50bc40f534d22b81dbe00241350e9f8d186192
># Parent  c0830a5933e8a7d1c744acea9b9183d0d72d01ab
>bug 870407: move CMMSRCS to mozbuild (file batch #4).
>
>diff --git a/toolkit/xre/Makefile.in b/toolkit/xre/Makefile.in
>--- a/toolkit/xre/Makefile.in
>+++ b/toolkit/xre/Makefile.in
>@@ -35,25 +35,19 @@ ifdef MOZ_UPDATER
> ifneq (android,$(MOZ_WIDGET_TOOLKIT))
> DEFINES += -DMOZ_UPDATER
> endif
> endif
> 
> ifeq ($(MOZ_WIDGET_TOOLKIT),windows)
> DEFINES += -DWIN32_LEAN_AND_MEAN -DUNICODE -D_UNICODE
> else
>-ifeq ($(MOZ_WIDGET_TOOLKIT),cocoa)
>-CMMSRCS = nsNativeAppSupportCocoa.mm
>-endif
> endif

nit: Please remove the 'else' statement for ifeq ($(MOZ_WIDGET_TOOLKIT),windows) since that block is now empty.

Looks good to go!

Comment 49

[Joey Armstrong [:joey]]

Attachment: move CMMSRCS to mozbuild (file batch #4)

Nit fixed + rebase.  r=mshal carried forward.

Comment 50

[Joey Armstrong [:joey]]

Comment on attachment 774836 [details] [diff] [review]
move CMMSRCS to mozbuild (file batch #4)

committed changeset 138390:e6bda3437a7c
remote:   https://hg.mozilla.org/integration/mozilla-inbound/rev/e6bda3437a7c
inbound push

Comment 51

[Wes Kocher (:KWierso) (Not reading bugmail; email directly if needed)]

(In reply to Joey Armstrong [:joey] from comment #50)
> Comment on attachment 774836 [details] [diff] [review]
> move CMMSRCS to mozbuild (file batch #4)
> 
> committed changeset 138390:e6bda3437a7c
> remote:   https://hg.mozilla.org/integration/mozilla-inbound/rev/e6bda3437a7c
> inbound push

Touched CLOBBER because it looks like this needed clobbering to work on m-i:
http://hg.mozilla.org/integration/mozilla-inbound/rev/92fa0b21fbb8

Comment 52

[Wes Kocher (:KWierso) (Not reading bugmail; email directly if needed)]

(In reply to Wes Kocher (:KWierso) from comment #51)
And the clobber came back busted, too.

Backed out in http://hg.mozilla.org/integration/mozilla-inbound/rev/417682db13bd

The try run was green, so I guess the rebase and nits fixed broke things?

Comment 53

[Ed Morley [:emorley]]

https://hg.mozilla.org/mozilla-central/rev/92fa0b21fbb8

Comment 54

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

Attachment: Final batch + disallow in makefiles

This seems to work on try...

Comment 55

[:Ms2ger (he/him; ⌚ UTC+1/+2)]

https://hg.mozilla.org/mozilla-central/rev/8beb34543e7b