GDB 7.5.91 available for testing

GDB 7.5.91 available for testing

[Joel Brobecker]

Hello,

I have just finished creating the gdb-7.5.91 pre-release.
It is available for download at the following location:

    ftp://sourceware.org/pub/gdb/snapshots/branch/gdb-7.5.91.tar.bz2

A gzip'ed version is also available: gdb-7.5.91.tar.gz.

Please give it a test if you can and report any problems you might find.

On behalf of all the GDB contributors, thank you!

-- 
Joel

sim/erc32/Makefile.in patch - was Re: GDB 7.5.91 available for testing

[Joel Sherrill]

[-- Attachment #1: Type: text/plain, Size: 1310 bytes --]

Hi

The sim/erc32 will not build outside the source tree. It is
missing a -I. to find config.h in the build tree.

mkdir b--gdb
cd b-gdb
../gdb-7.5.91/configure --target=sparc-rtems4.11 \
    --prefix=/home/joel/test-gdb/install/ \
    --enable-sim --enable-sim-hardware  \
     --enable-timebase --enable-sim-trace >c.log 2>&1 && \
     make >b.log 2>&1 && \
    make install >i.log 2>&1

Should be enough to reproduce it.

Can you commit this to the branch and head please?

2013-03-13  Joel Sherrill <joel.sherrill@oarcorp.com>

         * Makefile.in: Include build directory in search path to find
         config.h

Thanks.

On 3/12/2013 6:17 PM, Joel Brobecker wrote:
> Hello,
>
> I have just finished creating the gdb-7.5.91 pre-release.
> It is available for download at the following location:
>
>      ftp://sourceware.org/pub/gdb/snapshots/branch/gdb-7.5.91.tar.bz2
>
> A gzip'ed version is also available: gdb-7.5.91.tar.gz.
>
> Please give it a test if you can and report any problems you might find.
>
> On behalf of all the GDB contributors, thank you!
>


-- 
Joel Sherrill, Ph.D.             Director of Research & Development
joel.sherrill@OARcorp.com        On-Line Applications Research
Ask me about RTEMS: a free RTOS  Huntsville AL 35805
Support Available                (256) 722-9985


[-- Attachment #2: sim_erc32.diff --]
[-- Type: text/plain, Size: 713 bytes --]

2013-03-13  Joel Sherrill <joel.sherrill@oarcorp.com>

	* Makefile.in: Include build directory in search path to find
	config.h

Index: sim/erc32/Makefile.in
===================================================================
RCS file: /cvs/src/src/sim/erc32/Makefile.in,v
retrieving revision 1.12
diff -u -r1.12 Makefile.in
--- sim/erc32/Makefile.in	1 Jan 2013 06:41:35 -0000	1.12
+++ sim/erc32/Makefile.in	13 Mar 2013 14:18:43 -0000
@@ -45,7 +45,7 @@
 # There is AC_C_BIGENDIAN but it doesn't handle float endianness.
 # [Are int/float endians every different on a sparc?]
 end: $(srcdir)/end.c
-	$(CC_FOR_BUILD) $(srcdir)/end.c -o end
+	$(CC_FOR_BUILD) -I. $(srcdir)/end.c -o end
 end.h: end
 	./end > end.h
 

Re: sim/erc32/Makefile.in patch - was Re: GDB 7.5.91 available for testing

[Joel Brobecker]

> The sim/erc32 will not build outside the source tree. It is
> missing a -I. to find config.h in the build tree.
> 
> mkdir b--gdb
> cd b-gdb
> ../gdb-7.5.91/configure --target=sparc-rtems4.11 \
>    --prefix=/home/joel/test-gdb/install/ \
>    --enable-sim --enable-sim-hardware  \
>     --enable-timebase --enable-sim-trace >c.log 2>&1 && \
>     make >b.log 2>&1 && \
>    make install >i.log 2>&1
> 
> Should be enough to reproduce it.

Thanks. Reproduced using configure --target=sparc-elf --enable-sim.

> 2013-03-13  Joel Sherrill <joel.sherrill@oarcorp.com>
> 
>         * Makefile.in: Include build directory in search path to find
>         config.h

After a quick review of the patch, it looks reasonable to me.
But Mike F is the Global Maintainer, so I'll defer to him.

> 2013-03-13  Joel Sherrill <joel.sherrill@oarcorp.com>
> 
> 	* Makefile.in: Include build directory in search path to find
> 	config.h
> 
> Index: sim/erc32/Makefile.in
> ===================================================================
> RCS file: /cvs/src/src/sim/erc32/Makefile.in,v
> retrieving revision 1.12
> diff -u -r1.12 Makefile.in
> --- sim/erc32/Makefile.in	1 Jan 2013 06:41:35 -0000	1.12
> +++ sim/erc32/Makefile.in	13 Mar 2013 14:18:43 -0000
> @@ -45,7 +45,7 @@
>  # There is AC_C_BIGENDIAN but it doesn't handle float endianness.
>  # [Are int/float endians every different on a sparc?]
>  end: $(srcdir)/end.c
> -	$(CC_FOR_BUILD) $(srcdir)/end.c -o end
> +	$(CC_FOR_BUILD) -I. $(srcdir)/end.c -o end
>  end.h: end
>  	./end > end.h
>  


-- 
Joel

Re: sim/erc32/Makefile.in patch - was Re: GDB 7.5.91 available for testing

[Mike Frysinger]

[-- Attachment #1: Type: Text/Plain, Size: 924 bytes --]

On Wednesday 13 March 2013 14:17:59 Joel Brobecker wrote:
> > The sim/erc32 will not build outside the source tree. It is
> > missing a -I. to find config.h in the build tree.
> > 
> > mkdir b--gdb
> > cd b-gdb
> > ../gdb-7.5.91/configure --target=sparc-rtems4.11 \
> >    --prefix=/home/joel/test-gdb/install/ \
> >    --enable-sim --enable-sim-hardware  \
> >     --enable-timebase --enable-sim-trace >c.log 2>&1 && \
> >     make >b.log 2>&1 && \
> >    make install >i.log 2>&1
> > 
> > Should be enough to reproduce it.
> 
> Thanks. Reproduced using configure --target=sparc-elf --enable-sim.
> 
> > 2013-03-13  Joel Sherrill <joel.sherrill@oarcorp.com>
> > 
> >         * Makefile.in: Include build directory in search path to find
> >         config.h
> 
> After a quick review of the patch, it looks reasonable to me.
> But Mike F is the Global Maintainer, so I'll defer to him.

OK
-mike

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

[gdb-7.5.91] mips sim fails to build on mingw32

[Ralf Corsepius]

Hi,

The mips sim fails to build on mingw:
...
x86_64-w64-mingw32-gcc -DHAVE_CONFIG_H     -DPROFILE=1 -DWITH_PROFILE=-1 
   -DWITH_ALIGNMENT=NONSTRICT_ALIGNMENT -DWITH_TARGET_WORD_BITSIZE=32 
-DWITH_TARGET_WORD_MSB=31 -DWITH_DEFAULT_TARGET_BYTE_ORDER=BIG_ENDIAN 
-DWITH_FLOATING_POINT=HARD_FLOATING_POINT 
-DWITH_TARGET_FLOATING_POINT_BITSIZE=32 -DWITH_HW=0 
-DWITH_HOST_BYTE_ORDER=0 -DDEFAULT_INLINE=0   -DWITH_RESERVED_BITS=1 
-DWITH_SMP=0  -Wall -Wdeclaration-after-statement -Wpointer-arith 
-Wformat-nonliteral -Wno-pointer-sign -Wno-unused -Wunused-value 
-Wunused-function -Wno-switch -Wno-char-subscripts -Wmissing-prototypes 
-Wno-format    -DMIPS_MACH_DEFAULT=bfd_mach_mips8000   -I. 
-I../../../gdb-7.5.91/sim/mips -I../common 
-I../../../gdb-7.5.91/sim/mips/../common -I../../include 
-I../../../gdb-7.5.91/sim/mips/../..//include -I../../bfd 
-I../../../gdb-7.5.91/sim/mips/../..//bfd -I../../opcodes 
-I../../../gdb-7.5.91/sim/mips/../..//opcodes  -O2 -g -pipe -Wall 
-Wp,-D_FORTIFY_SOURCE=2 -fexceptions --param=ssp-buffer-size=4 
-mms-bitfields -D__USE_MINGW_ACCESS -c -o interp.o -MT interp.o -MMD -MP 
-MF .deps/interp.Tpo ../../../gdb-7.5.91/sim/mips/interp.c
../../../gdb-7.5.91/sim/mips/../common/dv-sockser.c:44:24: fatal error: 
netinet/in.h: No such file or directory


ATM, I assume this patch to be the cause (sim/mips):

2012-09-03 Joel Sherrill <joel.sherrill@oarcorp.com>

         * configure.ac: Always link against dv-sockser.o.
         * configure: Regenerate.


 From what I can gather from comparing sim/mips/configure.ac to other 
sim/*/configure.ac (e.g. bfin/configure.ac),
dv-sockser.o needs to be suppressed for mingw32-* targets?

Ralf

Re: [gdb-7.5.91] mips sim fails to build on mingw32

[Mike Frysinger]

[-- Attachment #1: Type: Text/Plain, Size: 1977 bytes --]

On Friday 15 March 2013 08:49:28 Ralf Corsepius wrote:
> Hi,
> 
> The mips sim fails to build on mingw:
> ...
> x86_64-w64-mingw32-gcc -DHAVE_CONFIG_H     -DPROFILE=1 -DWITH_PROFILE=-1
>    -DWITH_ALIGNMENT=NONSTRICT_ALIGNMENT -DWITH_TARGET_WORD_BITSIZE=32
> -DWITH_TARGET_WORD_MSB=31 -DWITH_DEFAULT_TARGET_BYTE_ORDER=BIG_ENDIAN
> -DWITH_FLOATING_POINT=HARD_FLOATING_POINT
> -DWITH_TARGET_FLOATING_POINT_BITSIZE=32 -DWITH_HW=0
> -DWITH_HOST_BYTE_ORDER=0 -DDEFAULT_INLINE=0   -DWITH_RESERVED_BITS=1
> -DWITH_SMP=0  -Wall -Wdeclaration-after-statement -Wpointer-arith
> -Wformat-nonliteral -Wno-pointer-sign -Wno-unused -Wunused-value
> -Wunused-function -Wno-switch -Wno-char-subscripts -Wmissing-prototypes
> -Wno-format    -DMIPS_MACH_DEFAULT=bfd_mach_mips8000   -I.
> -I../../../gdb-7.5.91/sim/mips -I../common
> -I../../../gdb-7.5.91/sim/mips/../common -I../../include
> -I../../../gdb-7.5.91/sim/mips/../..//include -I../../bfd
> -I../../../gdb-7.5.91/sim/mips/../..//bfd -I../../opcodes
> -I../../../gdb-7.5.91/sim/mips/../..//opcodes  -O2 -g -pipe -Wall
> -Wp,-D_FORTIFY_SOURCE=2 -fexceptions --param=ssp-buffer-size=4
> -mms-bitfields -D__USE_MINGW_ACCESS -c -o interp.o -MT interp.o -MMD -MP
> -MF .deps/interp.Tpo ../../../gdb-7.5.91/sim/mips/interp.c
> ../../../gdb-7.5.91/sim/mips/../common/dv-sockser.c:44:24: fatal error:
> netinet/in.h: No such file or directory
> 
> 
> ATM, I assume this patch to be the cause (sim/mips):
> 
> 2012-09-03 Joel Sherrill <joel.sherrill@oarcorp.com>
> 
>          * configure.ac: Always link against dv-sockser.o.
>          * configure: Regenerate.
> 
> 
>  From what I can gather from comparing sim/mips/configure.ac to other
> sim/*/configure.ac (e.g. bfin/configure.ac),
> dv-sockser.o needs to be suppressed for mingw32-* targets?

dv-sockser has always been a bit of an ad-hoc hack (at just about every 
layer).  adding a m4 macro to the common/ dir would probably be prudent.
-mike

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: [gdb-7.5.91] mips sim fails to build on mingw32

[Ralf Corsepius]

On 03/15/2013 07:16 PM, Mike Frysinger wrote:
> On Friday 15 March 2013 08:49:28 Ralf Corsepius wrote:
>> Hi,
>>
>> The mips sim fails to build on mingw:
>> ...
>> x86_64-w64-mingw32-gcc -DHAVE_CONFIG_H     -DPROFILE=1 -DWITH_PROFILE=-1
>>     -DWITH_ALIGNMENT=NONSTRICT_ALIGNMENT -DWITH_TARGET_WORD_BITSIZE=32
>> -DWITH_TARGET_WORD_MSB=31 -DWITH_DEFAULT_TARGET_BYTE_ORDER=BIG_ENDIAN
>> -DWITH_FLOATING_POINT=HARD_FLOATING_POINT
>> -DWITH_TARGET_FLOATING_POINT_BITSIZE=32 -DWITH_HW=0
>> -DWITH_HOST_BYTE_ORDER=0 -DDEFAULT_INLINE=0   -DWITH_RESERVED_BITS=1
>> -DWITH_SMP=0  -Wall -Wdeclaration-after-statement -Wpointer-arith
>> -Wformat-nonliteral -Wno-pointer-sign -Wno-unused -Wunused-value
>> -Wunused-function -Wno-switch -Wno-char-subscripts -Wmissing-prototypes
>> -Wno-format    -DMIPS_MACH_DEFAULT=bfd_mach_mips8000   -I.
>> -I../../../gdb-7.5.91/sim/mips -I../common
>> -I../../../gdb-7.5.91/sim/mips/../common -I../../include
>> -I../../../gdb-7.5.91/sim/mips/../..//include -I../../bfd
>> -I../../../gdb-7.5.91/sim/mips/../..//bfd -I../../opcodes
>> -I../../../gdb-7.5.91/sim/mips/../..//opcodes  -O2 -g -pipe -Wall
>> -Wp,-D_FORTIFY_SOURCE=2 -fexceptions --param=ssp-buffer-size=4
>> -mms-bitfields -D__USE_MINGW_ACCESS -c -o interp.o -MT interp.o -MMD -MP
>> -MF .deps/interp.Tpo ../../../gdb-7.5.91/sim/mips/interp.c
>> ../../../gdb-7.5.91/sim/mips/../common/dv-sockser.c:44:24: fatal error:
>> netinet/in.h: No such file or directory
>>
>>
>> ATM, I assume this patch to be the cause (sim/mips):
>>
>> 2012-09-03 Joel Sherrill <joel.sherrill@oarcorp.com>
>>
>>           * configure.ac: Always link against dv-sockser.o.
>>           * configure: Regenerate.
>>
>>
>>   From what I can gather from comparing sim/mips/configure.ac to other
>> sim/*/configure.ac (e.g. bfin/configure.ac),
>> dv-sockser.o needs to be suppressed for mingw32-* targets?
>
> dv-sockser has always been a bit of an ad-hoc hack (at just about every
> layer).  adding a m4 macro to the common/ dir would probably be prudent.
> -mike

Hmm, Joel, which issue did your patch (snippet below) try to address?


diff --git a/sim/mips/configure.ac b/sim/mips/configure.ac
index 5f59637..862e2b5 100644
--- a/sim/mips/configure.ac
+++ b/sim/mips/configure.ac
@@ -409,14 +413,13 @@ case "${target}" in
    mips*tx39*)
         hw_enabled=yes
         hw_extra_devices="tx3904cpu tx3904irc tx3904tmr tx3904sio"
-       mips_extra_objs="dv-sockser.o"
         SIM_SUBTARGET="$SIM_SUBTARGET -DTARGET_TX3904=1"
         ;;
    *)
-       mips_extra_objs=""
         ;;
  esac
  SIM_AC_OPTION_HARDWARE($hw_enabled,$hw_devices,$hw_extra_devices)
+mips_extra_objs="dv-sockser.o"
  AC_SUBST(mips_extra_objs)


IIUC, this lets all mips-sim-variants link against dv-sockser.o, while 
previously only the tx39*-variants did. I don't understand this step.

Ralf

Re: [gdb-7.5.91] mips sim fails to build on mingw32

[Joel Sherrill]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset="utf-8", Size: 3192 bytes --]

My recollection is that it did not link on CentOS 6 and was missing these symbols. Disabling it again should make it obvious.

But I didn't build in any particularly special way and it failed.

--joel

Ralf Corsepius <ralf.corsepius@rtems.org> wrote:


On 03/15/2013 07:16 PM, Mike Frysinger wrote:
> On Friday 15 March 2013 08:49:28 Ralf Corsepius wrote:
>> Hi,
>>
>> The mips sim fails to build on mingw:
>> ...
>> x86_64-w64-mingw32-gcc -DHAVE_CONFIG_H     -DPROFILE=1 -DWITH_PROFILE=-1
>>     -DWITH_ALIGNMENT=NONSTRICT_ALIGNMENT -DWITH_TARGET_WORD_BITSIZE=32
>> -DWITH_TARGET_WORD_MSB=31 -DWITH_DEFAULT_TARGET_BYTE_ORDER=BIG_ENDIAN
>> -DWITH_FLOATING_POINT=HARD_FLOATING_POINT
>> -DWITH_TARGET_FLOATING_POINT_BITSIZE=32 -DWITH_HW=0
>> -DWITH_HOST_BYTE_ORDER=0 -DDEFAULT_INLINE=0   -DWITH_RESERVED_BITS=1
>> -DWITH_SMP=0  -Wall -Wdeclaration-after-statement -Wpointer-arith
>> -Wformat-nonliteral -Wno-pointer-sign -Wno-unused -Wunused-value
>> -Wunused-function -Wno-switch -Wno-char-subscripts -Wmissing-prototypes
>> -Wno-format    -DMIPS_MACH_DEFAULT=bfd_mach_mips8000   -I.
>> -I../../../gdb-7.5.91/sim/mips -I../common
>> -I../../../gdb-7.5.91/sim/mips/../common -I../../include
>> -I../../../gdb-7.5.91/sim/mips/../..//include -I../../bfd
>> -I../../../gdb-7.5.91/sim/mips/../..//bfd -I../../opcodes
>> -I../../../gdb-7.5.91/sim/mips/../..//opcodes  -O2 -g -pipe -Wall
>> -Wp,-D_FORTIFY_SOURCE=2 -fexceptions --param=ssp-buffer-size=4
>> -mms-bitfields -D__USE_MINGW_ACCESS -c -o interp.o -MT interp.o -MMD -MP
>> -MF .deps/interp.Tpo ../../../gdb-7.5.91/sim/mips/interp.c
>> ../../../gdb-7.5.91/sim/mips/../common/dv-sockser.c:44:24: fatal error:
>> netinet/in.h: No such file or directory
>>
>>
>> ATM, I assume this patch to be the cause (sim/mips):
>>
>> 2012-09-03 Joel Sherrill <joel.sherrill@oarcorp.com>
>>
>>           * configure.ac: Always link against dv-sockser.o.
>>           * configure: Regenerate.
>>
>>
>>   From what I can gather from comparing sim/mips/configure.ac to other
>> sim/*/configure.ac (e.g. bfin/configure.ac),
>> dv-sockser.o needs to be suppressed for mingw32-* targets?
>
> dv-sockser has always been a bit of an ad-hoc hack (at just about every
> layer).  adding a m4 macro to the common/ dir would probably be prudent.
> -mike

Hmm, Joel, which issue did your patch (snippet below) try to address?


diff --git a/sim/mips/configure.ac b/sim/mips/configure.ac
index 5f59637..862e2b5 100644
--- a/sim/mips/configure.ac
+++ b/sim/mips/configure.ac
@@ -409,14 +413,13 @@ case "${target}" in
    mips*tx39*)
         hw_enabled=yes
         hw_extra_devices="tx3904cpu tx3904irc tx3904tmr tx3904sio"
-       mips_extra_objs="dv-sockser.o"
         SIM_SUBTARGET="$SIM_SUBTARGET -DTARGET_TX3904=1"
         ;;
    *)
-       mips_extra_objs=""
         ;;
  esac
  SIM_AC_OPTION_HARDWARE($hw_enabled,$hw_devices,$hw_extra_devices)
+mips_extra_objs="dv-sockser.o"
  AC_SUBST(mips_extra_objs)


IIUC, this lets all mips-sim-variants link against dv-sockser.o, while
previously only the tx39*-variants did. I don't understand this step.

Ralf




\x16º&ÖëzÛ«Ÿ}Ó‰b²Ö«r\x18\x1d

Re: [gdb-7.5.91] mips sim fails to build on mingw32

[Ralf Corsepius]

On 03/16/2013 05:33 AM, Ralf Corsepius wrote:
> On 03/15/2013 07:37 PM, Joel Sherrill wrote:
>> My recollection is that it did not link on CentOS 6 and was missing
>> these symbols. Disabling it again should make it obvious.
>
>   Reverting your patch lets build mips-rtems*gdb-7.5.91 build
> successfully on all hosts I am build gdb on [1]
>
>   This is not surprising to me, as reverting the patch pushes
> mips*gdb-7.5.91's configuration into a similar state as gdb-7.5.1 was,
> which built for all of my build hosts.
>
>> But I didn't build in any particularly special way and it failed.
>
> I am suspecting you were building mipstx39-rtems-gdb.

OK, I found the original thread:
http://www.sourceware.org/ml/gdb-patches/2012-09/msg00030.html

All I can say, I can not reproduce the issue you were reporting with 
gdb-7.5.1 and gdb-7.5.91. Reverting this change brings back 
mingw32-w64/gdb-7.5.91.

Ralf

Re: [gdb-7.5.91] mips sim fails to build on mingw32

[Joel Sherrill]

On 3/16/2013 1:52 AM, Ralf Corsepius wrote:
> On 03/16/2013 05:33 AM, Ralf Corsepius wrote:
>> On 03/15/2013 07:37 PM, Joel Sherrill wrote:
>>> My recollection is that it did not link on CentOS 6 and was missing
>>> these symbols. Disabling it again should make it obvious.
>>    Reverting your patch lets build mips-rtems*gdb-7.5.91 build
>> successfully on all hosts I am build gdb on [1]
>>
>>    This is not surprising to me, as reverting the patch pushes
>> mips*gdb-7.5.91's configuration into a similar state as gdb-7.5.1 was,
>> which built for all of my build hosts.
>>
>>> But I didn't build in any particularly special way and it failed.
>> I am suspecting you were building mipstx39-rtems-gdb.
> OK, I found the original thread:
> http://www.sourceware.org/ml/gdb-patches/2012-09/msg00030.html
>
> All I can say, I can not reproduce the issue you were reporting with
> gdb-7.5.1 and gdb-7.5.91. Reverting this change brings back
> mingw32-w64/gdb-7.5.91.
It is still there if you configure correctly. On a completely up to date
CentOS 6.x install. I reversed the patch locally and configured like this:

../gdb-7.5.91/configure --target=mips-rtems4.11 \
   --prefix=/home/joel/test-gdb/install/ --enable-sim \
   --enable-sim-hardware --enable-timebase --enable-sim-trace &&
make

And the build failed like this:

-o run \
       nrun.o libsim.a ../../bfd/libbfd.a ../../opcodes/libopcodes.a 
../../libiberty/libiberty.a -lm -lm -lz -lnsl
libsim.a(interp.o): In function `sim_open':
/home/joel/test-gdb/b-gdb/sim/mips/../../../gdb-7.5.91/sim/mips/interp.c:552: 
undefined reference to `sockser_addr'

Sounds like it needs the same configure magic as the other architectures.



>
> Ralf
>
>


-- 
Joel Sherrill, Ph.D.             Director of Research & Development
joel.sherrill@OARcorp.com        On-Line Applications Research
Ask me about RTEMS: a free RTOS  Huntsville AL 35805
Support Available                (256) 722-9985

Re: [gdb-7.5.91] mips sim fails to build on mingw32

[Joel Brobecker]

Joel,

> It is still there if you configure correctly. On a completely up to date
> CentOS 6.x install. I reversed the patch locally and configured like this:
> 
> ../gdb-7.5.91/configure --target=mips-rtems4.11 \
>   --prefix=/home/joel/test-gdb/install/ --enable-sim \
>   --enable-sim-hardware --enable-timebase --enable-sim-trace &&
> make
> 
> And the build failed like this:
> 
> -o run \
>       nrun.o libsim.a ../../bfd/libbfd.a ../../opcodes/libopcodes.a
> ../../libiberty/libiberty.a -lm -lm -lz -lnsl
> libsim.a(interp.o): In function `sim_open':
> /home/joel/test-gdb/b-gdb/sim/mips/../../../gdb-7.5.91/sim/mips/interp.c:552:
> undefined reference to `sockser_addr'
> 
> Sounds like it needs the same configure magic as the other architectures.

Do I understand correctly that the patch you added is too broad and
affects more targets than needed? If that's the case, I agree with
Ralf that the best course is to revert your patch first, and then
implement something that only affects CentOS.

Scheduled release time is 10 days away (Mar 26). It would be nice
to have this fixed by then. Otherwise, we'll have to delay the release.
Since you're the only one who seems to be able to reproduce, would
you mind working on this?

-- 
Joel

Re: [gdb-7.5.91] mips sim fails to build on mingw32

[Joel Sherrill]

On 3/16/2013 1:10 PM, Joel Brobecker wrote:
> Joel,
>
>> It is still there if you configure correctly. On a completely up to date
>> CentOS 6.x install. I reversed the patch locally and configured like this:
>>
>> ../gdb-7.5.91/configure --target=mips-rtems4.11 \
>>    --prefix=/home/joel/test-gdb/install/ --enable-sim \
>>    --enable-sim-hardware --enable-timebase --enable-sim-trace &&
>> make
>>
>> And the build failed like this:
>>
>> -o run \
>>        nrun.o libsim.a ../../bfd/libbfd.a ../../opcodes/libopcodes.a
>> ../../libiberty/libiberty.a -lm -lm -lz -lnsl
>> libsim.a(interp.o): In function `sim_open':
>> /home/joel/test-gdb/b-gdb/sim/mips/../../../gdb-7.5.91/sim/mips/interp.c:552:
>> undefined reference to `sockser_addr'
>>
>> Sounds like it needs the same configure magic as the other architectures.
> Do I understand correctly that the patch you added is too broad and
> affects more targets than needed? If that's the case, I agree with
> Ralf that the best course is to revert your patch first, and then
> implement something that only affects CentOS.
This is not CentOS specific. The file is required when you configure with
--enable-sim-hardware.  It is just a matter of tripping the configure
command correctly.

The only workaround going on here is that we need to account for
mingw not being able to build dv-sockser.c.

FWIW I see dv-sockser.o included by a multiple targets but only but
only the bfin has logic to avoid it on mingw.  Are any of the following
buildable on mingw?

cris
frv
iq2000
m32r
m68hc11
mn10300
sh64

It would be nice to know if this is a general problem on mingw.
> Scheduled release time is 10 days away (Mar 26). It would be nice
> to have this fixed by then. Otherwise, we'll have to delay the release.
> Since you're the only one who seems to be able to reproduce, would
> you mind working on this?
>

Sure. I have a patch written based on the logic in sim/bfin and just need to
test it.

But I don't know how to regenerate the configure script in gdb. I see
I need autoconf 2.64 but what's the magic incantation.

-- 
Joel Sherrill, Ph.D.             Director of Research & Development
joel.sherrill@OARcorp.com        On-Line Applications Research
Ask me about RTEMS: a free RTOS  Huntsville AL 35805
Support Available                (256) 722-9985

Re: [gdb-7.5.91] mips sim fails to build on mingw32

[Mike Frysinger]

[-- Attachment #1: Type: Text/Plain, Size: 230 bytes --]

On Saturday 16 March 2013 14:48:12 Joel Sherrill wrote:
> But I don't know how to regenerate the configure script in gdb. I see
> I need autoconf 2.64 but what's the magic incantation.

i do:
	cd sim/mips
	aclocal
	autoconf
-mike

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: [gdb-7.5.91] mips sim fails to build on mingw32

[Joel Brobecker]

> On Saturday 16 March 2013 14:48:12 Joel Sherrill wrote:
> > But I don't know how to regenerate the configure script in gdb. I see
> > I need autoconf 2.64 but what's the magic incantation.
> 
> i do:
> 	cd sim/mips
> 	aclocal
> 	autoconf

This might not be needed in this context, but I usually add a call
to "autoheader" as well, for when config.in needs to get regenerated.

-- 
Joel

Re: [gdb-7.5.91] mips sim fails to build on mingw32

[Mike Frysinger]

[-- Attachment #1: Type: Text/Plain, Size: 589 bytes --]

On Saturday 16 March 2013 21:03:20 Joel Brobecker wrote:
> > On Saturday 16 March 2013 14:48:12 Joel Sherrill wrote:
> > > But I don't know how to regenerate the configure script in gdb. I see
> > > I need autoconf 2.64 but what's the magic incantation.
> > 
> > i do:
> > 	cd sim/mips
> > 	aclocal
> > 	autoconf
> 
> This might not be needed in this context, but I usually add a call
> to "autoheader" as well, for when config.in needs to get regenerated.

yeah, mips doesn't need it.  my script will look for AC_CONFIG_HEADERS and 
call autoheader when it's found.
-mike

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: [gdb-7.5.91] mips sim fails to build on mingw32

[Joel Sherrill]

[-- Attachment #1: Type: text/plain, Size: 959 bytes --]

Hi

Thanks for the pointers.

I have done some minimal testing on this. I built it on
CentOS 6.x with and without the --enable-sim-hardware
option. The patch is against the 7.5.91 snapshot which
should be close enough.

It is based off of bfin.

2013-03-16  Joel Sherrill <joel.sherrill@oarcorp.com>

     * configure.ac: Avoid use of dv-sockser on *mingw32*
     * aclocal.m4: Regenerated.
     * configure: Regenerated.

--joel



On 3/16/2013 6:27 PM, Mike Frysinger wrote:
> On Saturday 16 March 2013 14:48:12 Joel Sherrill wrote:
>> But I don't know how to regenerate the configure script in gdb. I see
>> I need autoconf 2.64 but what's the magic incantation.
> i do:
> 	cd sim/mips
> 	aclocal
> 	autoconf
> -mike


-- 
Joel Sherrill, Ph.D.             Director of Research & Development
joel.sherrill@OARcorp.com        On-Line Applications Research
Ask me about RTEMS: a free RTOS  Huntsville AL 35805
Support Available                (256) 722-9985


[-- Attachment #2: mips_diff.txt --]
[-- Type: text/plain, Size: 993 bytes --]

2013-03-16  Joel Sherrill <joel.sherrill@oarcorp.com>

	* configure.ac: Avoid use of dv-sockser on *mingw32*
	* aclocal.m4: Regenerated.
	* configure: Regenerated.

diff -uNr gdb-7.5.91-virgin/sim/mips/configure.ac gdb-7.5.91/sim/mips/configure.ac
--- gdb-7.5.91-virgin/sim/mips/configure.ac	2013-01-01 00:41:39.000000000 -0600
+++ gdb-7.5.91/sim/mips/configure.ac	2013-03-16 13:38:42.800403652 -0500
@@ -418,9 +418,23 @@
   *)
 	;;
 esac
+
 SIM_AC_OPTION_HARDWARE($hw_enabled,$hw_devices,$hw_extra_devices)
-mips_extra_objs="dv-sockser.o"
-AC_SUBST(mips_extra_objs)
+
+mips_extra_objs=
+dnl make sure the dv-sockser code can be supported (i.e. windows)
+case ${host} in
+  *mingw32*) ;;
+  *)
+    AC_DEFINE_UNQUOTED([HAVE_DV_SOCKSER], 1, [Define if dv-sockser is usable.])
+    mips_extra_objs="${mips_extra_objs} dv-sockser.o"
+    ;;
+esac
+
+AC_SUBST([mips_extra_objs], ${mips_extra_objs})
+
+## mips_extra_objs="dv-sockser.o"
+## AC_SUBST(mips_extra_objs)
 
 
 # Choose simulator engine

Re: [gdb-7.5.91] mips sim fails to build on mingw32

[Mike Frysinger]

[-- Attachment #1: Type: Text/Plain, Size: 554 bytes --]

On Saturday 16 March 2013 21:13:29 Joel Sherrill wrote:
> Hi
> 
> Thanks for the pointers.
> 
> I have done some minimal testing on this. I built it on
> CentOS 6.x with and without the --enable-sim-hardware
> option. The patch is against the 7.5.91 snapshot which
> should be close enough.
> 
> It is based off of bfin.
> 
> 2013-03-16  Joel Sherrill <joel.sherrill@oarcorp.com>
> 
>      * configure.ac: Avoid use of dv-sockser on *mingw32*
>      * aclocal.m4: Regenerated.
>      * configure: Regenerated.

seems reasonable
-mike

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: [gdb-7.5.91] mips sim fails to build on mingw32

[Joel Brobecker]

> 2013-03-16  Joel Sherrill <joel.sherrill@oarcorp.com>
> 
>     * configure.ac: Avoid use of dv-sockser on *mingw32*
>     * aclocal.m4: Regenerated.
>     * configure: Regenerated.
[...]
> +
>  SIM_AC_OPTION_HARDWARE($hw_enabled,$hw_devices,$hw_extra_devices)
> -mips_extra_objs="dv-sockser.o"
> -AC_SUBST(mips_extra_objs)
> +
> +mips_extra_objs=
> +dnl make sure the dv-sockser code can be supported (i.e. windows)
> +case ${host} in
> +  *mingw32*) ;;

I would change that to *mingw*. We might get mingw64 someday...


> +  *)
> +    AC_DEFINE_UNQUOTED([HAVE_DV_SOCKSER], 1, [Define if dv-sockser is usable.])
> +    mips_extra_objs="${mips_extra_objs} dv-sockser.o"
> +    ;;
> +esac
> +
> +AC_SUBST([mips_extra_objs], ${mips_extra_objs})
> +
> +## mips_extra_objs="dv-sockser.o"
> +## AC_SUBST(mips_extra_objs)

Can you also remove the commented-out code?

-- 
Joel

Re: [gdb-7.5.91] mips sim fails to build on mingw32

[Joel Sherrill]

On 3/17/2013 11:02 AM, Joel Brobecker wrote:
>> 2013-03-16  Joel Sherrill <joel.sherrill@oarcorp.com>
>>
>>      * configure.ac: Avoid use of dv-sockser on *mingw32*
>>      * aclocal.m4: Regenerated.
>>      * configure: Regenerated.
> [...]
>> +
>>   SIM_AC_OPTION_HARDWARE($hw_enabled,$hw_devices,$hw_extra_devices)
>> -mips_extra_objs="dv-sockser.o"
>> -AC_SUBST(mips_extra_objs)
>> +
>> +mips_extra_objs=
>> +dnl make sure the dv-sockser code can be supported (i.e. windows)
>> +case ${host} in
>> +  *mingw32*) ;;
> I would change that to *mingw*. We might get mingw64 someday...
:) The patch I posted simultaneously to address this in common/acinclude.m4
also has mingw32. I will fix it. But that patch needs more review.

cd sim
grep dv-sockser.o */Makefile.in
grep dv-sockser.o */configure.ac

And you will see what I need advice on. The patch in the next email
fixes bfin and mips I think. Scratches problem on other targets.
>
>> +  *)
>> +    AC_DEFINE_UNQUOTED([HAVE_DV_SOCKSER], 1, [Define if dv-sockser is usable.])
>> +    mips_extra_objs="${mips_extra_objs} dv-sockser.o"
>> +    ;;
>> +esac
>> +
>> +AC_SUBST([mips_extra_objs], ${mips_extra_objs})
>> +
>> +## mips_extra_objs="dv-sockser.o"
>> +## AC_SUBST(mips_extra_objs)
> Can you also remove the commented-out code?
>

The mips/configure.ac in the new patch relies on common/acinclude.m4 to
tell it whether or to use dv-sockser.o. This centralizes the test for 
which hosts
it supports.

-- 
Joel Sherrill, Ph.D.             Director of Research & Development
joel.sherrill@OARcorp.com        On-Line Applications Research
Ask me about RTEMS: a free RTOS  Huntsville AL 35805
Support Available                (256) 722-9985

Re: [gdb-7.5.91] mips sim fails to build on mingw32

[Ralf Corsepius]

On 03/16/2013 04:08 PM, Joel Sherrill wrote:
> On 3/16/2013 1:52 AM, Ralf Corsepius wrote:
>> On 03/16/2013 05:33 AM, Ralf Corsepius wrote:
>>> On 03/15/2013 07:37 PM, Joel Sherrill wrote:
>>>> My recollection is that it did not link on CentOS 6 and was missing
>>>> these symbols. Disabling it again should make it obvious.
>>>    Reverting your patch lets build mips-rtems*gdb-7.5.91 build
>>> successfully on all hosts I am build gdb on [1]
>>>
>>>    This is not surprising to me, as reverting the patch pushes
>>> mips*gdb-7.5.91's configuration into a similar state as gdb-7.5.1 was,
>>> which built for all of my build hosts.
>>>
>>>> But I didn't build in any particularly special way and it failed.
>>> I am suspecting you were building mipstx39-rtems-gdb.
>> OK, I found the original thread:
>> http://www.sourceware.org/ml/gdb-patches/2012-09/msg00030.html
>>
>> All I can say, I can not reproduce the issue you were reporting with
>> gdb-7.5.1 and gdb-7.5.91. Reverting this change brings back
>> mingw32-w64/gdb-7.5.91.
> It is still there if you configure correctly.
"Correctly" is a bold term!

> On a completely up to date
> CentOS 6.x install. I reversed the patch locally and configured like this:
>
> ../gdb-7.5.91/configure --target=mips-rtems4.11 \
>    --prefix=/home/joel/test-gdb/install/ --enable-sim \
>    --enable-sim-hardware --enable-timebase --enable-sim-trace &&
> make
>
> And the build failed like this:
>
> -o run \
>        nrun.o libsim.a ../../bfd/libbfd.a ../../opcodes/libopcodes.a
> ../../libiberty/libiberty.a -lm -lm -lz -lnsl
> libsim.a(interp.o): In function `sim_open':
> /home/joel/test-gdb/b-gdb/sim/mips/../../../gdb-7.5.91/sim/mips/interp.c:552:
> undefined reference to `sockser_addr'

This fails on all hosts for me.

But ... this provides further insights:

I configure using "default" simulator flags:
--enable-sim

You configure using "exotic" simulator flags:
--enable-sim --enable-sim-hardware --enable-timebase --enable-sim-trace

Why you're using them, I don't know.


Further trial and error with your patch reversed shows:
--enable-sim-hardware triggers the link-error your reported on all hosts.
All other permutations of --enable-sim --enable-sim-{timebase,sim-trace} 
build.

I conclude from this:
sim/mips only needs dv-sockser.o for tx39* targets or when
--enable-sim-hardware is used.

=> The patch is wrong and should be reverted. The configuration magic to 
pull in dv-socksers.o should be reworked and likely needs to be keyed to 
-enable-sim-hardware.


Ralf

Re: [gdb-7.5.91] mips sim fails to build on mingw32

[Joel Sherrill]

[-- Attachment #1: Type: text/plain, Size: 5057 bytes --]

Comments interspersed but the bottom line is that this is
an ugly issue where I believe a number of the simulators
have hard-coded dependencies on dv-sockser.o and do
not link without it even on CentOS. These sims are:

frv/Makefile.in:CONFIG_DEVICES = dv-sockser.o
iq2000/Makefile.in:CONFIG_DEVICES = dv-sockser.o
m32r/Makefile.in:CONFIG_DEVICES = dv-sockser.o
mn10300/Makefile.in:    dv-sockser.o
sh64/Makefile.in:CONFIG_DEVICES = dv-sockser.o

Plus m68hc11 which does not honor --enable-sim-hardware
in its configure.ac. Based on the target name, it just enables
hardware. It appears to be broken in another way.

I have attached a patch which addresses common, mips, bfin
and makes m68hc11 use the conditional for dv-sockser.o.  It
doesn't make m68hc11 honor --enable-sim-hardware.

mips and bfin build fine with this patch with or without
--enable-sim-hardware.

I don't know what to do about the other targets. My recommendation
would be:

+ m68hc11 -minimum  honor --enable-sim-hardware
+ others - if dependency on dv-sockser.o is hard-coded and unavoidable,
the entire simulator is unsupported on mingw32.

I am certainly open for suggestions.




On 3/16/2013 11:14 PM, Ralf Corsepius wrote:
> On 03/16/2013 04:08 PM, Joel Sherrill wrote:
>> On 3/16/2013 1:52 AM, Ralf Corsepius wrote:
>>> On 03/16/2013 05:33 AM, Ralf Corsepius wrote:
>>>> On 03/15/2013 07:37 PM, Joel Sherrill wrote:
>>>>> My recollection is that it did not link on CentOS 6 and was missing
>>>>> these symbols. Disabling it again should make it obvious.
>>>>     Reverting your patch lets build mips-rtems*gdb-7.5.91 build
>>>> successfully on all hosts I am build gdb on [1]
>>>>
>>>>     This is not surprising to me, as reverting the patch pushes
>>>> mips*gdb-7.5.91's configuration into a similar state as gdb-7.5.1 was,
>>>> which built for all of my build hosts.
>>>>
>>>>> But I didn't build in any particularly special way and it failed.
>>>> I am suspecting you were building mipstx39-rtems-gdb.
>>> OK, I found the original thread:
>>> http://www.sourceware.org/ml/gdb-patches/2012-09/msg00030.html
>>>
>>> All I can say, I can not reproduce the issue you were reporting with
>>> gdb-7.5.1 and gdb-7.5.91. Reverting this change brings back
>>> mingw32-w64/gdb-7.5.91.
>> It is still there if you configure correctly.
> "Correctly" is a bold term!
>
>> On a completely up to date
>> CentOS 6.x install. I reversed the patch locally and configured like this:
>>
>> ../gdb-7.5.91/configure --target=mips-rtems4.11 \
>>     --prefix=/home/joel/test-gdb/install/ --enable-sim \
>>     --enable-sim-hardware --enable-timebase --enable-sim-trace &&
>> make
>>
>> And the build failed like this:
>>
>> -o run \
>>         nrun.o libsim.a ../../bfd/libbfd.a ../../opcodes/libopcodes.a
>> ../../libiberty/libiberty.a -lm -lm -lz -lnsl
>> libsim.a(interp.o): In function `sim_open':
>> /home/joel/test-gdb/b-gdb/sim/mips/../../../gdb-7.5.91/sim/mips/interp.c:552:
>> undefined reference to `sockser_addr'
> This fails on all hosts for me.
FWIW the host pattern is mingw32 for the work around and
mingw64 may have this as well. I have no way to test this.
> But ... this provides further insights:
>
> I configure using "default" simulator flags:
> --enable-sim
>
> You configure using "exotic" simulator flags:
> --enable-sim --enable-sim-hardware --enable-timebase --enable-sim-trace
>
> Why you're using them, I don't know.
Very simple.  configure options are supposed to be used or ignored
by each component.  That is by definition.

--enable-sim-hardware
====================
It is a generic option to indicate that if the simulator has optional
devices, enable them all.

Simulators that don't have anything to enable should just work or ignore
the flag.

The PowerPC simulator does something meaningful with this. All simulators
using common/acinclude.m4 are apparently broken when this is enabled.

--enable-sim-timease
==================
sim/ppc specific. Correctly ignored everywhere else. No issue.

--enable-sim-trace
================
This enables execution trace capabilities which are useful in debug.
This option is present in 26 subdirectories in sim so appears to be
universal or close.
>
> Further trial and error with your patch reversed shows:
> --enable-sim-hardware triggers the link-error your reported on all hosts.
> All other permutations of --enable-sim --enable-sim-{timebase,sim-trace}
> build.
>
> I conclude from this:
> sim/mips only needs dv-sockser.o for tx39* targets or when
> --enable-sim-hardware is used.
>
> => The patch is wrong and should be reverted. The configuration magic to
> pull in dv-socksers.o should be reworked and likely needs to be keyed to
> -enable-sim-hardware.
Agreed. This is a bug in common/acinclude.m4  and (possibly) the Makefile.in
for the actual simulators.
>   
>
> Ralf
>


-- 
Joel Sherrill, Ph.D.             Director of Research & Development
joel.sherrill@OARcorp.com        On-Line Applications Research
Ask me about RTEMS: a free RTOS  Huntsville AL 35805
Support Available                (256) 722-9985


[-- Attachment #2: sockser_diff_v1.txt --]
[-- Type: text/plain, Size: 3286 bytes --]

common/ChangeLog:
2013-03-17  Joel Sherrill <joel.sherrill@oarcorp.com>

	* acinclude.m4: Add SIM_DV_SOCKSER_O which is empty on hosts
	which do not support dv-sockser.o.

bfin/ChangeLog:
2013-03-17  Joel Sherrill <joel.sherrill@oarcorp.com>

	* configure.ac: Use $SIM_DV_SOCKSER_O
	* configure: Regenerated.

m68hc11/ChangeLog:
2013-03-17  Joel Sherrill <joel.sherrill@oarcorp.com>

	* configure.ac: Use $SIM_DV_SOCKSER_O
	* configure: Regenerated.

mips/ChangeLog:
2013-03-17  Joel Sherrill <joel.sherrill@oarcorp.com>

	* configure.ac: Use $SIM_DV_SOCKSER_O
	* configure: Regenerated.

diff -uNr gdb-7.5.91-virgin/sim/bfin/configure.ac gdb-7.5.91/sim/bfin/configure.ac
--- gdb-7.5.91-virgin/sim/bfin/configure.ac	2012-04-09 01:13:06.000000000 -0500
+++ gdb-7.5.91/sim/bfin/configure.ac	2013-03-17 10:29:12.831880985 -0500
@@ -56,17 +56,7 @@
 	sys/mman.h \
 ])
 
-BFIN_SIM_EXTRA_OBJS=
-
-dnl make sure the dv-sockser code can be supported (i.e. windows)
-case ${host} in
-  *mingw32*) ;;
-  *)
-    AC_DEFINE_UNQUOTED([HAVE_DV_SOCKSER], 1, [Define if dv-sockser is usable.])
-    BFIN_SIM_EXTRA_OBJS="${BFIN_SIM_EXTRA_OBJS} dv-sockser.o"
-    ;;
-esac
-
+BFIN_SIM_EXTRA_OBJS="${BFIN_SIM_EXTRA_OBJS} ${SIM_DV_SOCKSER_O}"
 AC_SUBST([BFIN_SIM_EXTRA_OBJS], ${BFIN_SIM_EXTRA_OBJS})
 
 PKG_PROG_PKG_CONFIG
Binary files gdb-7.5.91-virgin/sim/bfin/.configure.ac.swp and gdb-7.5.91/sim/bfin/.configure.ac.swp differ
diff -uNr gdb-7.5.91-virgin/sim/common/acinclude.m4 gdb-7.5.91/sim/common/acinclude.m4
--- gdb-7.5.91-virgin/sim/common/acinclude.m4	2012-11-20 08:41:08.000000000 -0600
+++ gdb-7.5.91/sim/common/acinclude.m4	2013-03-17 10:24:21.757167294 -0500
@@ -647,6 +647,16 @@
       *) sim_hw="$sim_hw $i" ; sim_hw_objs="$sim_hw_objs dv-$i.o";;
     esac
   done
+  # mingw32 does not support sockser
+ SIM_DV_SOCKSER_O=""
+  case ${host} in
+    *mingw32*) ;;
+    *) SIM_DV_SOCKSER_O="dv-sockser.o"
+       AC_DEFINE_UNQUOTED(
+         [HAVE_DV_SOCKSER], 1, [Define if dv-sockser is usable.])
+       ;;
+  esac
+  AC_SUBST(SIM_DV_SOCKSER_O)
 fi
 if test x"$silent" != x"yes" && test "$sim_hw_p" = "yes"; then
   echo "Setting hardware to $sim_hw_cflags, $sim_hw, $sim_hw_objs"
diff -uNr gdb-7.5.91-virgin/sim/m68hc11/configure.ac gdb-7.5.91/sim/m68hc11/configure.ac
--- gdb-7.5.91-virgin/sim/m68hc11/configure.ac	2011-10-17 19:30:58.000000000 -0500
+++ gdb-7.5.91/sim/m68hc11/configure.ac	2013-03-17 10:17:21.532456675 -0500
@@ -19,7 +19,7 @@
   m68hc11-*-*|m6811-*-*)
 	hw_enabled=yes
 	hw_extra_devices="m68hc11 m68hc11sio m68hc11eepr m68hc11tim m68hc11spi nvram"
-	m68hc11_extra_objs="dv-sockser.o"
+	m68hc11_extra_objs="$SIM_SV_SOCKSER_O"
 	SIM_SUBTARGET="$SIM_SUBTARGET -DTARGET_M68HC11=1"
 	;;
   *)
diff -uNr gdb-7.5.91-virgin/sim/mips/configure.ac gdb-7.5.91/sim/mips/configure.ac
--- gdb-7.5.91-virgin/sim/mips/configure.ac	2013-01-01 00:41:39.000000000 -0600
+++ gdb-7.5.91/sim/mips/configure.ac	2013-03-17 10:36:33.251716513 -0500
@@ -414,12 +414,12 @@
 	hw_enabled=yes
 	hw_extra_devices="tx3904cpu tx3904irc tx3904tmr tx3904sio" 
 	SIM_SUBTARGET="$SIM_SUBTARGET -DTARGET_TX3904=1"
+	mips_extra_objs="$SIM_DV_SOCKSER_O"
 	;;
   *)
 	;;
 esac
 SIM_AC_OPTION_HARDWARE($hw_enabled,$hw_devices,$hw_extra_devices)
-mips_extra_objs="dv-sockser.o"
 AC_SUBST(mips_extra_objs)
 
 

Re: [gdb-7.5.91] mips sim fails to build on mingw32

[Joel Sherrill]

On 3/18/2013 1:05 PM, Mike Frysinger wrote:
> On Sunday 17 March 2013 12:05:03 Joel Sherrill wrote:
>> Comments interspersed but the bottom line is that this is
>> an ugly issue where I believe a number of the simulators
>> have hard-coded dependencies on dv-sockser.o and do
>> not link without it even on CentOS. These sims are:
>>
>> frv/Makefile.in:CONFIG_DEVICES = dv-sockser.o
>> iq2000/Makefile.in:CONFIG_DEVICES = dv-sockser.o
>> m32r/Makefile.in:CONFIG_DEVICES = dv-sockser.o
>> mn10300/Makefile.in:    dv-sockser.o
>> sh64/Makefile.in:CONFIG_DEVICES = dv-sockser.o
>>
>> Plus m68hc11 which does not honor --enable-sim-hardware
>> in its configure.ac. Based on the target name, it just enables
>> hardware. It appears to be broken in another way.
> i'm not sure this is true.  the m68hc11 code:
> 	hw_enabled=no
> 	case "${target}" in
> 	  m68hc11-*-*|m6811-*-*)
> 	    hw_enabled=yes
> 	....
> 	SIM_AC_OPTION_HARDWARE($hw_enabled,$hw_devices,$hw_extra_devices)
>
> the common code:
> 	dnl --enable-sim-hardware is for users of the simulator
> 	dnl arg[1] Enable sim-hw by default? ("yes" or "no")
> 	...
> 	AC_DEFUN([SIM_AC_OPTION_HARDWARE],
>
> the first arg is the *default* behavior, not the hardcoded behavior.  so if you
> want to test w/out sim hardware, you have to pass --disable-sim-hardware.
Got it. Thanks.
>> I have attached a patch which addresses common, mips, bfin
>> and makes m68hc11 use the conditional for dv-sockser.o.  It
>> doesn't make m68hc11 honor --enable-sim-hardware.
>>
>> mips and bfin build fine with this patch with or without
>> --enable-sim-hardware.
> well, the default for Blackfin at least is to enable sim hardware :).  so
> really you have to test with --enable-sim-hardware and --disable-sim-hardware.
> 	SIM_AC_OPTION_HARDWARE(yes,,\
>
> for mips, only tx39 defaults to --enable-sim-hardware.  the other mips targets
> default to --disable-sim-hardware.
Thanks for the clarification.
>> I don't know what to do about the other targets. My recommendation
>> would be:
>>
>> + m68hc11 -minimum  honor --enable-sim-hardware
>> + others - if dependency on dv-sockser.o is hard-coded and unavoidable,
>> the entire simulator is unsupported on mingw32.
> i made the Blackfin sim build on Windows because it was a use case i cared
> about :).  all of the device models should work fine there too.
All I changed was replace the case on mingw to avoid using dv-sockser.o 
to use
the variable from common/acinclude.m4.

FYI you have to have bfin as installed to compile the simulator. I 
suppose this
has never been noticed.
> it should be easy to update the various sims configure scripts that have
> hardware device models that unconditionally use sockser to check the state.
> after the call to SIM_AC_OPTION_HARDWARE, you can do:
> 	if test "$sim_hw_p" = yes -a -z "$SIM_DV_SOCKSER_O"; then
> 		case " $sim_hw " in
> 		*" tx3904sio "*)
> 			AC_MSG_ERROR([Sorry, but tx3904sio hardware support is
> unavailable for your target.  Please use --disable-sim-hardware, or pass a
> list of devices to enable that does not include that.])
> 		esac
> 	fi
Done and tested for mips.

Can I put a generic message for other targets that "dv-sockser hardware 
support is
unavailable for your target" and let it go at that? Or do I specifically 
need to list the
devices when dv-sockser.o is hard-coded in the Makefile.in?
> and then add a TODO comment to each of these failing device modules with
> something like "should change code to respect HAVE_DV_SOCKSER".
>
What would such a message look like and where in the file would it go?

> i'd also highlight my opinion that the way the mips sim puts itself together
> isn't the right way to go about things.  the fact that you can't build a
> single mips sim and, at runtime, select the actual target is bad.  in glancing
> through the code though, i'm not sure this is something that'd be easy to fix
> since it seems to be fundamentally baked in :/.
I'm not pulling that thread but for RTEMS we build both mips-rtems and 
mipstx39-rtems
so this is pretty obvious to us. :)
>> FWIW the host pattern is mingw32 for the work around and
>> mingw64 may have this as well. I have no way to test this.
> i've never tested mingw64 myself.  i think it's safe to assume that the
> failure is common to all Windows targets, so using *mingw* as the pattern is
> sane.
Got it.
>>> But ... this provides further insights:
>>>
>>> I configure using "default" simulator flags:
>>> --enable-sim
>>>
>>> You configure using "exotic" simulator flags:
>>> --enable-sim --enable-sim-hardware --enable-timebase --enable-sim-trace
>>>
>>> Why you're using them, I don't know.
>> Very simple.  configure options are supposed to be used or ignored
>> by each component.  That is by definition.
> yeah, i didn't see anything unusual in your configure flag selection.  seemed
> perfectly reasonable to me :).
Now I am testing with explicit --enable-sim-hardware and 
--disable-sim-hardware.
>>> => The patch is wrong and should be reverted. The configuration magic to
>>> pull in dv-socksers.o should be reworked and likely needs to be keyed to
>>> -enable-sim-hardware.
>> Agreed. This is a bug in common/acinclude.m4  and (possibly) the
>> Makefile.in for the actual simulators.
> i don't generally buy into the "revert revert revert" mentality.  if you're
> working on a real fix, let's go for it.
I am getting close. Something should show up shortly.
>> --- gdb-7.5.91-virgin/sim/common/acinclude.m4
>> +++ gdb-7.5.91/sim/common/acinclude.m4
>> @@ -647,6 +647,16 @@
>>         *) sim_hw="$sim_hw $i" ; sim_hw_objs="$sim_hw_objs dv-$i.o";;
>>       esac
>>     done
>> +  # mingw32 does not support sockser
>> + SIM_DV_SOCKSER_O=""
> indentation is off by one space here
Fixed
> -mike


-- 
Joel Sherrill, Ph.D.             Director of Research & Development
joel.sherrill@OARcorp.com        On-Line Applications Research
Ask me about RTEMS: a free RTOS  Huntsville AL 35805
Support Available                (256) 722-9985

Re: [gdb-7.5.91] mips sim fails to build on mingw32

[Mike Frysinger]

[-- Attachment #1: Type: Text/Plain, Size: 1335 bytes --]

On Wednesday 20 March 2013 13:51:08 Joel Sherrill wrote:
> FYI you have to have bfin as installed to compile the simulator. I
> suppose this
> has never been noticed.

i'm guessing you're referring to the assembly of some fixed code file ?  that 
happens when timestamps get out of whack.  i originally just wanted to add a 
.h, but people requested i also add the source .s.

not sure if there's a way i could put this logic behind "if (file does not 
exist) || (file is out of date && maintainer mode)" so it doesn't bite people 
randomly ...

> Can I put a generic message for other targets that "dv-sockser hardware
> support is
> unavailable for your target" and let it go at that? Or do I specifically
> need to list the
> devices when dv-sockser.o is hard-coded in the Makefile.in?

since we don't have a large body of people clamoring for sim device support 
under Windows, let's start with the simple version and avoid hardcoded lists.

> > and then add a TODO comment to each of these failing device modules with
> > something like "should change code to respect HAVE_DV_SOCKSER".
> 
> What would such a message look like and where in the file would it go?

i was thinking of a /* TODO: ... */ in each device model file that 
unconditionally uses dv-socker.  e.g. sim/mips/dv-tx3904sio.c.
-mike

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: [gdb-7.5.91] mips sim fails to build on mingw32

[Joel Sherrill]

On 3/20/2013 1:33 PM, Mike Frysinger wrote:
> On Wednesday 20 March 2013 13:51:08 Joel Sherrill wrote:
>> FYI you have to have bfin as installed to compile the simulator. I
>> suppose this
>> has never been noticed.
> i'm guessing you're referring to the assembly of some fixed code file ?  that
> happens when timestamps get out of whack.  i originally just wanted to add a
> .h, but people requested i also add the source .s.
>
> not sure if there's a way i could put this logic behind "if (file does not
> exist) || (file is out of date && maintainer mode)" so it doesn't bite people
> randomly ...
I think it may have been with the gdb-7.5.91 snapshot. Maybe the 
timestamp wasn't
right. I am in a virtual machine and maybe my time of day was messed up 
at the time.

>> Can I put a generic message for other targets that "dv-sockser hardware
>> support is
>> unavailable for your target" and let it go at that? Or do I specifically
>> need to list the
>> devices when dv-sockser.o is hard-coded in the Makefile.in?
> since we don't have a large body of people clamoring for sim device support
> under Windows, let's start with the simple version and avoid hardcoded lists.
OK. Separate message with proposed addition to those configure.ac's.
>>> and then add a TODO comment to each of these failing device modules with
>>> something like "should change code to respect HAVE_DV_SOCKSER".
>> What would such a message look like and where in the file would it go?
> i was thinking of a /* TODO: ... */ in each device model file that
> unconditionally uses dv-socker.  e.g. sim/mips/dv-tx3904sio.c.
OK. This is at the bottom of my list.

I think I am getting run over by the squeaky wheel that I reported. :)

> -mike


-- 
Joel Sherrill, Ph.D.             Director of Research & Development
joel.sherrill@OARcorp.com        On-Line Applications Research
Ask me about RTEMS: a free RTOS  Huntsville AL 35805
Support Available                (256) 722-9985

Re: [gdb-7.5.91] mips sim fails to build on mingw32

[Pedro Alves]

On 03/17/2013 04:05 PM, Joel Sherrill wrote:
> +  # mingw32 does not support sockser
> + SIM_DV_SOCKSER_O=""
> +  case ${host} in
> +    *mingw32*) ;;
> +    *) SIM_DV_SOCKSER_O="dv-sockser.o"
> +       AC_DEFINE_UNQUOTED(
> +         [HAVE_DV_SOCKSER], 1, [Define if dv-sockser is usable.])
> +       ;;
> +  esac
> +  AC_SUBST(SIM_DV_SOCKSER_O)
>  fi

OOC, is there a fundamental reason dv-sockser.o couldn't
be made to work on mingw, or is it that nobody has bothered yet?

I skimmed the file and it just looked like straight socket stuff, which
usually means that porting to mingw just revolves around including
windows.h/winsock2.h and using send/recv/closesocket instead of bsd
style read/write/close, etc.

-- 
Pedro Alves

Re: [gdb-7.5.91] mips sim fails to build on mingw32

[Mike Frysinger]

[-- Attachment #1: Type: Text/Plain, Size: 899 bytes --]

On Wednesday 27 March 2013 07:20:23 Pedro Alves wrote:
> On 03/17/2013 04:05 PM, Joel Sherrill wrote:
> > +  # mingw32 does not support sockser
> > + SIM_DV_SOCKSER_O=""
> > +  case ${host} in
> > +    *mingw32*) ;;
> > +    *) SIM_DV_SOCKSER_O="dv-sockser.o"
> > +       AC_DEFINE_UNQUOTED(
> > +         [HAVE_DV_SOCKSER], 1, [Define if dv-sockser is usable.])
> > +       ;;
> > +  esac
> > +  AC_SUBST(SIM_DV_SOCKSER_O)
> > 
> >  fi
> 
> OOC, is there a fundamental reason dv-sockser.o couldn't
> be made to work on mingw, or is it that nobody has bothered yet?

the last part one

> I skimmed the file and it just looked like straight socket stuff, which
> usually means that porting to mingw just revolves around including
> windows.h/winsock2.h and using send/recv/closesocket instead of bsd
> style read/write/close, etc.

is there a gnulib module already ?
-mike

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 836 bytes --]