Problem compiling latest CVS with gcc-2.95.3

Problem compiling latest CVS with gcc-2.95.3

[Panagiotis Papadakos]

I get the following error:

gcc -M -D__KERNEL__ -DMODULE=1
-I/mnt/disk/CVS/ALSA-CVS/alsa-driver/include
-I/lib/modules/2.4.21-pre5-ac2/build/include -O2
-mpreferred-stack-boundary=2 -march=i686 -falign-functions=4 -DLINUX -Wall
-Wstrict-prototypes -fomit-frame-pointer -Wno-trigraphs -O2
-fno-strict-aliasing -fno-common -pipe -DALSA_BUILD  usbaudio.c usbmidi.c
usbmixer.c > .depend
make[1]: Leaving directory `/mnt/disk/CVS/ALSA-CVS/alsa-driver/usb'
make[1]: Entering directory `/mnt/disk/CVS/ALSA-CVS/alsa-driver/acore'
gcc -D__KERNEL__ -DMODULE=1 -I/mnt/disk/CVS/ALSA-CVS/alsa-driver/include
-I/lib/modules/2.4.21-pre5-ac2/build/include -O2
-mpreferred-stack-boundary=2 -march=i686 -falign-functions=4 -DLINUX -Wall
-Wstrict-prototypes -fomit-frame-pointer -Wno-trigraphs -O2
-fno-strict-aliasing -fno-common -pipe -DALSA_BUILD   -DEXPORT_SYMTAB -c
hwdep.c
cc1: Invalid option `-falign-functions=4'
make[1]: *** [hwdep.o] Error 1
make[1]: Leaving directory `/mnt/disk/CVS/ALSA-CVS/alsa-driver/acore'
make: *** [compile] Error 1
-(root@Santorini:0)-(12 files:18k@ALSA-CVS)-(0 jobs)-(10:48)-
-(/mnt/disk/CVS/ALSA-CVS:#)->

	Panagiotis Papadakos


-------------------------------------------------------
This SF.net email is sponsored by: Etnus, makers of TotalView, The debugger 
for complex code. Debugging C/C++ programs can leave you feeling lost and 
disoriented. TotalView can help you find your way. Available on major UNIX 
and Linux platforms. Try it free. www.etnus.com

Re: Problem compiling latest CVS with gcc-2.95.3

[Martin Langer]

On Sat, Mar 08, 2003 at 10:51:29AM +0200, Panagiotis Papadakos wrote:
> I get the following error:

same with gcc 2.95.4!

martin


> 
> gcc -M -D__KERNEL__ -DMODULE=1
> -I/mnt/disk/CVS/ALSA-CVS/alsa-driver/include
> -I/lib/modules/2.4.21-pre5-ac2/build/include -O2
> -mpreferred-stack-boundary=2 -march=i686 -falign-functions=4 -DLINUX -Wall
                                           ^^^^^^^^^^^^^^^^^^^
> -Wstrict-prototypes -fomit-frame-pointer -Wno-trigraphs -O2
> -fno-strict-aliasing -fno-common -pipe -DALSA_BUILD  usbaudio.c usbmidi.c
> usbmixer.c > .depend
> make[1]: Leaving directory `/mnt/disk/CVS/ALSA-CVS/alsa-driver/usb'
> make[1]: Entering directory `/mnt/disk/CVS/ALSA-CVS/alsa-driver/acore'
> gcc -D__KERNEL__ -DMODULE=1 -I/mnt/disk/CVS/ALSA-CVS/alsa-driver/include
> -I/lib/modules/2.4.21-pre5-ac2/build/include -O2
> -mpreferred-stack-boundary=2 -march=i686 -falign-functions=4 -DLINUX -Wall
> -Wstrict-prototypes -fomit-frame-pointer -Wno-trigraphs -O2
> -fno-strict-aliasing -fno-common -pipe -DALSA_BUILD   -DEXPORT_SYMTAB -c
> hwdep.c
> cc1: Invalid option `-falign-functions=4'
                       ^^^^^^^^^^^^^^^^^^^
> make[1]: *** [hwdep.o] Error 1
> make[1]: Leaving directory `/mnt/disk/CVS/ALSA-CVS/alsa-driver/acore'
> make: *** [compile] Error 1
> -(root@Santorini:0)-(12 files:18k@ALSA-CVS)-(0 jobs)-(10:48)-
> -(/mnt/disk/CVS/ALSA-CVS:#)->


-------------------------------------------------------
This SF.net email is sponsored by: Etnus, makers of TotalView, The debugger 
for complex code. Debugging C/C++ programs can leave you feeling lost and 
disoriented. TotalView can help you find your way. Available on major UNIX 
and Linux platforms. Try it free. www.etnus.com

Re: Problem compiling latest CVS with gcc-2.95.3

[Martin Langer]

On Sat, Mar 08, 2003 at 10:11:24AM +0100, Martin Langer wrote:
> On Sat, Mar 08, 2003 at 10:51:29AM +0200, Panagiotis Papadakos wrote:
>
> same with gcc 2.95.4!
> > 
> > gcc -M -D__KERNEL__ -DMODULE=1
> > -I/mnt/disk/CVS/ALSA-CVS/alsa-driver/include
> > -I/lib/modules/2.4.21-pre5-ac2/build/include -O2
> > -mpreferred-stack-boundary=2 -march=i686 -falign-functions=4 -DLINUX -Wall
>                                            ^^^^^^^^^^^^^^^^^^^
> > -Wstrict-prototypes -fomit-frame-pointer -Wno-trigraphs -O2
> > -fno-strict-aliasing -fno-common -pipe -DALSA_BUILD  usbaudio.c usbmidi.c
> > usbmixer.c > .depend
> > make[1]: Leaving directory `/mnt/disk/CVS/ALSA-CVS/alsa-driver/usb'
> > make[1]: Entering directory `/mnt/disk/CVS/ALSA-CVS/alsa-driver/acore'
> > gcc -D__KERNEL__ -DMODULE=1 -I/mnt/disk/CVS/ALSA-CVS/alsa-driver/include
> > -I/lib/modules/2.4.21-pre5-ac2/build/include -O2
> > -mpreferred-stack-boundary=2 -march=i686 -falign-functions=4 -DLINUX -Wall
> > -Wstrict-prototypes -fomit-frame-pointer -Wno-trigraphs -O2
> > -fno-strict-aliasing -fno-common -pipe -DALSA_BUILD   -DEXPORT_SYMTAB -c
> > hwdep.c
> > cc1: Invalid option `-falign-functions=4'
>                        ^^^^^^^^^^^^^^^^^^^
> > make[1]: *** [hwdep.o] Error 1
> > make[1]: Leaving directory `/mnt/disk/CVS/ALSA-CVS/alsa-driver/acore'
> > make: *** [compile] Error 1
> > -(root@Santorini:0)-(12 files:18k@ALSA-CVS)-(0 jobs)-(10:48)-
> > -(/mnt/disk/CVS/ALSA-CVS:#)->
> 

this are messages the of the following machine:

processor       : 0
vendor_id       : AuthenticAMD
cpu family      : 6
model           : 4
model name      : AMD Athlon(tm) processor
stepping        : 2
cpu MHz         : 1102.523
cache size      : 256 KB
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid level     : 1
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 sep mtrr pge mca cmov
pat pse36 mmx fxsr syscall mmxext 3dnowext 3dnow
bogomips        : 2195.45




Another machine (same Alsa-CVS, both debian woody) shows a different problem:

gcc -D__KERNEL__ -DMODULE=1 -I/usr/local/src/alsa-cvs/alsa-driver/include
-I/lib/modules/2.4.20/build/include -O2 -mpreferred-stack-boundary=2
-march=i586 -DLINUX -Wall -Wstrict-prototypes -fomit-frame-pointer
-Wno-trigraphs -O2 -fno-strict-aliasing -fno-common -pipe -DALSA_BUILD
-DEXPORT_SYMTAB -c ainstr_iw.c
rm -f snd-ainstr-fm.o
ld  -r -o snd-ainstr-fm.o ainstr_fm.o
rm -f snd-ainstr-gf1.o
ld  -r -o snd-ainstr-gf1.o ainstr_gf1.o
rm -f snd-ainstr-iw.o
ld  -r -o snd-ainstr-iw.o ainstr_iw.o
rm -f snd-ainstr-simple.o
ld  -r -o snd-ainstr-simple.o ainstr_simple.o
ainstr_simple.o: file not recognized: File truncated
make[4]: *** [snd-ainstr-simple.o] Error 1
make[4]: Leaving directory
/usr/local/src/alsa-cvs/alsa-driver/acore/seq/instr'
make[3]: *** [_modsubdir_instr] Error 2
make[3]: Leaving directory /usr/local/src/alsa-cvs/alsa-driver/acore/seq'
make[2]: *** [_modsubdir_seq] Error 2
make[2]: Leaving directory /usr/local/src/alsa-cvs/alsa-driver/acore'

and that one is such a damned slow compiling.....

processor       : 0
vendor_id       : CyrixInstead
cpu family      : 5
model           : 7
model name      : Cyrix MediaGXtm MMXtm Enhanced
stepping        : 4
cpu MHz         : 200.456
cache size      : 16 KB
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid level     : 2
wp              : yes
flags           : fpu tsc msr cx8 cmov mmx cxmmx
bogomips        : 399.76



-------------------------------------------------------
This SF.net email is sponsored by: Etnus, makers of TotalView, The debugger 
for complex code. Debugging C/C++ programs can leave you feeling lost and 
disoriented. TotalView can help you find your way. Available on major UNIX 
and Linux platforms. Try it free. www.etnus.com

Re: Problem compiling latest CVS with gcc-2.95.3

[Martin Langer]

On Sat, Mar 08, 2003 at 10:40:52AM +0100, Martin Langer wrote:
> On Sat, Mar 08, 2003 at 10:11:24AM +0100, Martin Langer wrote:
> > On Sat, Mar 08, 2003 at 10:51:29AM +0200, Panagiotis Papadakos wrote:
> >
> > > cc1: Invalid option `-falign-functions=4'


Removing that option really helps in my case. Don't know if it should be
really applied to the CVS tree.

martin


--- alsa-driver/configure.in.ORIGINAL	Sat Mar  8 23:20:05 2003
+++ alsa-driver/configure.in	Sat Mar  8 23:20:54 2003
@@ -430,7 +430,7 @@
     if $CC -march=athlon -S -o /dev/null -xc /dev/null >/dev/null 2>&1; then
       c_opts="-march=athlon"
     else
-      c_opts="-march=i686 -falign-functions=4"
+      c_opts="-march=i686"
     fi
     ;;
   crusoe)


> Another machine (same Alsa-CVS, both debian woody) shows a different problem:
> 
> ainstr_simple.o: file not recognized: File truncated

latest CVS has no longer that problem. It compiles fine.


-------------------------------------------------------
This SF.net email is sponsored by: Etnus, makers of TotalView, The debugger 
for complex code. Debugging C/C++ programs can leave you feeling lost and 
disoriented. TotalView can help you find your way. Available on major UNIX 
and Linux platforms. Try it free. www.etnus.com

Re: Problem compiling latest CVS with gcc-2.95.3

[Takashi Iwai]

[-- Attachment #1: Type: text/plain, Size: 589 bytes --]

At Sat, 8 Mar 2003 23:44:27 +0100,
Martin Langer wrote:
> 
> On Sat, Mar 08, 2003 at 10:40:52AM +0100, Martin Langer wrote:
> > On Sat, Mar 08, 2003 at 10:11:24AM +0100, Martin Langer wrote:
> > > On Sat, Mar 08, 2003 at 10:51:29AM +0200, Panagiotis Papadakos wrote:
> > >
> > > > cc1: Invalid option `-falign-functions=4'
> 
> 
> Removing that option really helps in my case. Don't know if it should be
> really applied to the CVS tree.

IIRC, the -falign-xxx is a new option and the older compiler had
-malign-xxx instead.  could you check the attached patch whether it
works?


Takashi

[-- Attachment #2: config-fix.dif --]
[-- Type: application/octet-stream, Size: 1828 bytes --]

Index: alsa-driver/configure.in
===================================================================
RCS file: /suse/tiwai/cvs/alsa/alsa-driver/configure.in,v
retrieving revision 1.99
diff -u -r1.99 configure.in
--- alsa-driver/configure.in	10 Mar 2003 13:02:39 -0000	1.99
+++ alsa-driver/configure.in	10 Mar 2003 16:29:24 -0000
@@ -335,8 +335,7 @@
 	fprintf(file, "i386");
 #elif defined(CONFIG_M486)
 	fprintf(file, "i486");
-#elif defined(CONFIG_M586) || defined(CONFIG_MWINCHIPC6) || defined(CONFIG_MWINCHIP2)\
-	|| defined(CONFIG_MWINCHIP3D) || defined(CONFIG_MCYRIXIII)
+#elif defined(CONFIG_M586)
 	fprintf(file, "i586");
 #elif defined(CONFIG_M586TSC)
 	fprintf(file, "i586tsc");
@@ -434,11 +433,19 @@
     if $CC -march=athlon -S -o /dev/null -xc /dev/null >/dev/null 2>&1; then
       c_opts="-march=athlon"
     else
-      c_opts="-march=i686 -malign-functions=4"
+      if $CC -falign-functions=0 -S -o /dev/null -xc /dev/null >/dev/null 2>&1; then
+        c_opts="-march=i686 -falign-functions=4"
+      else
+	c_opts="-march=i686 -malign-functions=4"
+      fi
     fi
     ;;
   crusoe)
-    c_opts="-march=i686 -malign-functions=0 -malign-jumps=0 -malign-loops=0"
+    if $CC -falign-functions=0 -S -o /dev/null -xc /dev/null >/dev/null 2>&1; then
+      c_opts="-march=i686 -falign-functions=0 -falign-jumps=0 -falign-loops=0"
+    else
+      c_opts="-march=i686 -malign-functions=0 -malign-jumps=0 -malign-loops=0"
+    fi
     ;;
   mcyrixiii)
     if $CC -march=c3 -S -o /dev/null -xc /dev/null >/dev/null 2>&1; then
@@ -573,7 +580,7 @@
 
 dnl set ia32 (X86)
 case "$processor" in
-  i?86*|k?|crusoe)
+  i?86*|k?|crusoe|mcyrixiii|mwinchip)
     if $CC -mpreferred-stack-boundary=2 -S -o /dev/null -xc /dev/null >/dev/null 2>&1; then
       c_opts="-mpreferred-stack-boundary=2 $c_opts"
       CONFIG_X86=y

Re: Problem compiling latest CVS with gcc-2.95.3

[Martin Langer]

On Tue, Mar 11, 2003 at 11:14:43AM +0100, Takashi Iwai wrote:
> At Sat, 8 Mar 2003 23:44:27 +0100,
> Martin Langer wrote:
> > 
> > On Sat, Mar 08, 2003 at 10:40:52AM +0100, Martin Langer wrote:
> > > On Sat, Mar 08, 2003 at 10:11:24AM +0100, Martin Langer wrote:
> > > > On Sat, Mar 08, 2003 at 10:51:29AM +0200, Panagiotis Papadakos wrote:
> > > >
> > > > > cc1: Invalid option `-falign-functions=4'
> > 
> > 
> > Removing that option really helps in my case. Don't know if it should be
> > really applied to the CVS tree.
> 
> IIRC, the -falign-xxx is a new option and the older compiler had
> -malign-xxx instead.  could you check the attached patch whether it
> works?
> 

ATM I'm using CVS without any patch again. It compiles fine. I haven't made
anything except an update of the CVS tree. This patch is no longer necessary
in my case.
I think it was fixed by Jaroslav from version 1.145 to 1.146 of configure.

I'm ready for .9 stable!

martin


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf

Re: Problem compiling latest CVS with gcc-2.95.3

[Takashi Iwai]

At Tue, 11 Mar 2003 12:12:54 +0100,
Martin Langer wrote:
> 
> On Tue, Mar 11, 2003 at 11:14:43AM +0100, Takashi Iwai wrote:
> > At Sat, 8 Mar 2003 23:44:27 +0100,
> > Martin Langer wrote:
> > > 
> > > On Sat, Mar 08, 2003 at 10:40:52AM +0100, Martin Langer wrote:
> > > > On Sat, Mar 08, 2003 at 10:11:24AM +0100, Martin Langer wrote:
> > > > > On Sat, Mar 08, 2003 at 10:51:29AM +0200, Panagiotis Papadakos wrote:
> > > > >
> > > > > > cc1: Invalid option `-falign-functions=4'
> > > 
> > > 
> > > Removing that option really helps in my case. Don't know if it should be
> > > really applied to the CVS tree.
> > 
> > IIRC, the -falign-xxx is a new option and the older compiler had
> > -malign-xxx instead.  could you check the attached patch whether it
> > works?
> > 
> 
> ATM I'm using CVS without any patch again. It compiles fine. I haven't made
> anything except an update of the CVS tree. This patch is no longer necessary
> in my case.
> I think it was fixed by Jaroslav from version 1.145 to 1.146 of configure.

yes, the current cvs version uses -malign-xxxx, so it should work for
you as it is.  but it's not good for some compliers which prefer
-falign-xxxx options.

my patch works like 2.5 kernel does; checks whether -falign is
accepted and use -malign if not.  so, it would be appreciated if you
can check that before putting it to the cvs tree.


ciao,

Takashi


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf

Re: Problem compiling latest CVS with gcc-2.95.3

[Panagiotis Papadakos]

I works for me too...

	Panagiotis Papadakos

On Tue, 11 Mar 2003, Martin Langer wrote:

> On Tue, Mar 11, 2003 at 11:14:43AM +0100, Takashi Iwai wrote:
> > At Sat, 8 Mar 2003 23:44:27 +0100,
> > Martin Langer wrote:
> > >
> > > On Sat, Mar 08, 2003 at 10:40:52AM +0100, Martin Langer wrote:
> > > > On Sat, Mar 08, 2003 at 10:11:24AM +0100, Martin Langer wrote:
> > > > > On Sat, Mar 08, 2003 at 10:51:29AM +0200, Panagiotis Papadakos wrote:
> > > > >
> > > > > > cc1: Invalid option `-falign-functions=4'
> > >
> > >
> > > Removing that option really helps in my case. Don't know if it should be
> > > really applied to the CVS tree.
> >
> > IIRC, the -falign-xxx is a new option and the older compiler had
> > -malign-xxx instead.  could you check the attached patch whether it
> > works?
> >
>
> ATM I'm using CVS without any patch again. It compiles fine. I haven't made
> anything except an update of the CVS tree. This patch is no longer necessary
> in my case.
> I think it was fixed by Jaroslav from version 1.145 to 1.146 of configure.
>
> I'm ready for .9 stable!
>
> martin
>
>
> -------------------------------------------------------
> This sf.net email is sponsored by:ThinkGeek
> Welcome to geek heaven.
> http://thinkgeek.com/sf
> _______________________________________________
> Alsa-devel mailing list
> Alsa-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/alsa-devel
>


-------------------------------------------------------
This SF.net email is sponsored by:Crypto Challenge is now open! 
Get cracking and register here for some mind boggling fun and 
the chance of winning an Apple iPod:
http://ads.sourceforge.net/cgi-bin/redirect.pl?thaw0031en

Re: Problem compiling latest CVS with gcc-2.95.3

[Martin Langer]

On Tue, Mar 11, 2003 at 02:02:51PM +0100, Takashi Iwai wrote:
> > On Tue, Mar 11, 2003 at 11:14:43AM +0100, Takashi Iwai wrote:
> > > 
> > > IIRC, the -falign-xxx is a new option and the older compiler had
> > > -malign-xxx instead.  could you check the attached patch whether it
> > > works?
> 
> my patch works like 2.5 kernel does; checks whether -falign is
> accepted and use -malign if not.  so, it would be appreciated if you
> can check that before putting it to the cvs tree.
> 

No problems. That patch compiles fine using -malign. 
So, everything is ok.


martin


-------------------------------------------------------
This SF.net email is sponsored by:Crypto Challenge is now open! 
Get cracking and register here for some mind boggling fun and 
the chance of winning an Apple iPod:
http://ads.sourceforge.net/cgi-bin/redirect.pl?thaw0031en

Re: Problem compiling latest CVS with gcc-2.95.3

[Takashi Iwai]

At Tue, 11 Mar 2003 15:30:24 +0100,
Martin Langer wrote:
> 
> On Tue, Mar 11, 2003 at 02:02:51PM +0100, Takashi Iwai wrote:
> > > On Tue, Mar 11, 2003 at 11:14:43AM +0100, Takashi Iwai wrote:
> > > > 
> > > > IIRC, the -falign-xxx is a new option and the older compiler had
> > > > -malign-xxx instead.  could you check the attached patch whether it
> > > > works?
> > 
> > my patch works like 2.5 kernel does; checks whether -falign is
> > accepted and use -malign if not.  so, it would be appreciated if you
> > can check that before putting it to the cvs tree.
> > 
> 
> No problems. That patch compiles fine using -malign. 
> So, everything is ok.

thanks for testing.
the new configure is on cvs.


Takashi


-------------------------------------------------------
This SF.net email is sponsored by:Crypto Challenge is now open! 
Get cracking and register here for some mind boggling fun and 
the chance of winning an Apple iPod:
http://ads.sourceforge.net/cgi-bin/redirect.pl?thaw0031en