make dep-am

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

make dep-am

Alan Modra
HJ,
  What version of gcc are you using to generate makefile dependencies?
I see lots of benign differences when I regenerate, all like the
following.  Your gcc -MM apparently doesn't remove duplicates.

@@ -1088,8 +1088,7 @@ coff-arm.lo: coff-arm.c $(INCDIR)/filena
   libcoff.h $(INCDIR)/bfdlink.h coffcode.h coffswap.h
 coff-aux.lo: coff-aux.c $(INCDIR)/filenames.h $(INCDIR)/coff/aux-coff.h \
   $(INCDIR)/coff/internal.h $(INCDIR)/coff/m68k.h $(INCDIR)/coff/external.h \
-  coff-m68k.c $(INCDIR)/hashtab.h $(INCDIR)/coff/m68k.h \
-  $(INCDIR)/coff/internal.h libcoff.h $(INCDIR)/bfdlink.h \
+  coff-m68k.c $(INCDIR)/hashtab.h libcoff.h $(INCDIR)/bfdlink.h \
   coffcode.h coffswap.h
 coff-h8300.lo: coff-h8300.c $(INCDIR)/filenames.h $(INCDIR)/hashtab.h \
   $(INCDIR)/bfdlink.h genlink.h $(INCDIR)/coff/h8300.h \

--
Alan Modra
IBM OzLabs - Linux Technology Centre
Reply | Threaded
Open this post in threaded view
|

RE: make dep-am

H.J. Lu
I saw those too. I am using gcc version 3.4.4 20050721 (Red Hat 3.4.4-2)
on RHEL 4 U2. I tried gcc version 4.0.3 20051026 (prerelease) and the
result is the same.

H.J. Lu
Intel Corporation


>-----Original Message-----
>From: Alan Modra [mailto:[hidden email]]
>Sent: Sunday, October 30, 2005 4:27 PM
>To: Lu, Hongjiu; [hidden email]
>Subject: make dep-am
>
>HJ,
>  What version of gcc are you using to generate makefile dependencies?
>I see lots of benign differences when I regenerate, all like the
>following.  Your gcc -MM apparently doesn't remove duplicates.
>
>@@ -1088,8 +1088,7 @@ coff-arm.lo: coff-arm.c $(INCDIR)/filena
>   libcoff.h $(INCDIR)/bfdlink.h coffcode.h coffswap.h
> coff-aux.lo: coff-aux.c $(INCDIR)/filenames.h $(INCDIR)/coff/aux-
>coff.h \
>   $(INCDIR)/coff/internal.h $(INCDIR)/coff/m68k.h
>$(INCDIR)/coff/external.h \
>-  coff-m68k.c $(INCDIR)/hashtab.h $(INCDIR)/coff/m68k.h \
>-  $(INCDIR)/coff/internal.h libcoff.h $(INCDIR)/bfdlink.h \
>+  coff-m68k.c $(INCDIR)/hashtab.h libcoff.h $(INCDIR)/bfdlink.h \
>   coffcode.h coffswap.h
> coff-h8300.lo: coff-h8300.c $(INCDIR)/filenames.h $(INCDIR)/hashtab.h
>\
>   $(INCDIR)/bfdlink.h genlink.h $(INCDIR)/coff/h8300.h \
>
>--
>Alan Modra
>IBM OzLabs - Linux Technology Centre
Reply | Threaded
Open this post in threaded view
|

Re: make dep-am

Alan Modra
On Sun, Oct 30, 2005 at 04:32:09PM -0800, H.J. Lu wrote:
> I saw those too. I am using gcc version 3.4.4 20050721 (Red Hat 3.4.4-2)
> on RHEL 4 U2. I tried gcc version 4.0.3 20051026 (prerelease) and the
> result is the same.

Hmm, I guess this counts as a minor gcc regression.  My old gcc-3.3 is
clever enough to remove duplicates.  It seems that gcc-4.1 generates yet
another set of duplicates too, different from your 3.4 commit.  :-(

--
Alan Modra
IBM OzLabs - Linux Technology Centre
Reply | Threaded
Open this post in threaded view
|

RE: make dep-am

H.J. Lu
In reply to this post by Alan Modra
I can open a gcc bug report if necessary.


H.J. Lu
Intel Corporation


>-----Original Message-----
>From: Alan Modra [mailto:[hidden email]]
>Sent: Sunday, October 30, 2005 4:53 PM
>To: Lu, Hongjiu
>Cc: [hidden email]
>Subject: Re: make dep-am
>
>On Sun, Oct 30, 2005 at 04:32:09PM -0800, H.J. Lu wrote:
>> I saw those too. I am using gcc version 3.4.4 20050721 (Red Hat
>3.4.4-2)
>> on RHEL 4 U2. I tried gcc version 4.0.3 20051026 (prerelease) and the
>> result is the same.
>
>Hmm, I guess this counts as a minor gcc regression.  My old gcc-3.3 is
>clever enough to remove duplicates.  It seems that gcc-4.1 generates
>yet
>another set of duplicates too, different from your 3.4 commit.  :-(
>
>--
>Alan Modra
>IBM OzLabs - Linux Technology Centre