[windows] build error on gdbserver

[windows] build error on gdbserver

[asmwarrior]

Hi, all, I use the gdb cvs.

Now, build gdb under msys+mingw has a build error:

make[4]: Entering directory `/f/build_gdb/gdb/gdbgit/buildNo/gdb/gdbserver'
rm -f version.c-tmp version.c
echo '#include "server.h"' >> version.c-tmp
gcc -c -O0 -g -D__USE_MINGW_ACCESS    -I. -I../../../gdb/gdb/gdbserver -I../../.
./gdb/gdb/gdbserver/../common -I../../../gdb/gdb/gdbserver/../regformats -I../..
/../gdb/gdb/gdbserver/../../include -Wall -Wdeclaration-after-statement -Wpointe
r-arith -Wformat-nonliteral -Wno-char-subscripts -Werror ../../../gdb/gdb/gdbser
ver/mem-break.c
echo 'const char version[] = "'"`sed q ../../../gdb/gdb/gdbserver/../version.in`
"'";' >> version.c-tmp
gcc -c -O0 -g -D__USE_MINGW_ACCESS    -I. -I../../../gdb/gdb/gdbserver -I../../.
./gdb/gdb/gdbserver/../common -I../../../gdb/gdb/gdbserver/../regformats -I../..
/../gdb/gdb/gdbserver/../../include -Wall -Wdeclaration-after-statement -Wpointe
r-arith -Wformat-nonliteral -Wno-char-subscripts -Werror ../../../gdb/gdb/gdbser
ver/hostio.c
gcc -c -O0 -g -D__USE_MINGW_ACCESS    -I. -I../../../gdb/gdb/gdbserver -I../../.
./gdb/gdb/gdbserver/../common -I../../../gdb/gdb/gdbserver/../regformats -I../..
/../gdb/gdb/gdbserver/../../include -Wall -Wdeclaration-after-statement -Wpointe
r-arith -Wformat-nonliteral -Wno-char-subscripts -Werror ../../../gdb/gdb/gdbser
ver/event-loop.c
echo 'const char host_name[] = "mingw32";' >> version.c-tmp
mv version.c-tmp version.c
gcc -c -O0 -g -D__USE_MINGW_ACCESS    -I. -I../../../gdb/gdb/gdbserver -I../../.
./gdb/gdb/gdbserver/../common -I../../../gdb/gdb/gdbserver/../regformats -I../..
/../gdb/gdb/gdbserver/../../include -Wall -Wdeclaration-after-statement -Wpointe
r-arith -Wformat-nonliteral -Wno-char-subscripts -Werror ../../../gdb/gdb/gdbser
ver/tracepoint.c
gcc -c  -O0 -g -D__USE_MINGW_ACCESS    -I. -I../../../gdb/gdb/gdbserver -I../../
../gdb/gdb/gdbserver/../common -I../../../gdb/gdb/gdbserver/../regformats -I../.
./../gdb/gdb/gdbserver/../../include -Wall -Wdeclaration-after-statement -Wpoint
er-arith -Wformat-nonliteral -Wno-char-subscripts -Werror ../../../gdb/gdb/gdbse
rver/../common/xml-utils.c -DGDBSERVER
gcc -c  -O0 -g -D__USE_MINGW_ACCESS    -I. -I../../../gdb/gdb/gdbserver -I../../
../gdb/gdb/gdbserver/../common -I../../../gdb/gdb/gdbserver/../regformats -I../.
./../gdb/gdb/gdbserver/../../include -Wall -Wdeclaration-after-statement -Wpoint
er-arith -Wformat-nonliteral -Wno-char-subscripts -Werror ../../../gdb/gdb/gdbse
rver/../common/common-utils.c -DGDBSERVER
gcc -c  -O0 -g -D__USE_MINGW_ACCESS    -I. -I../../../gdb/gdb/gdbserver -I../../
../gdb/gdb/gdbserver/../common -I../../../gdb/gdb/gdbserver/../regformats -I../.
./../gdb/gdb/gdbserver/../../include -Wall -Wdeclaration-after-statement -Wpoint
er-arith -Wformat-nonliteral -Wno-char-subscripts -Werror ../../../gdb/gdb/gdbse
rver/../common/buffer.c -DGDBSERVER
cc1.exe: warnings being treated as errors
../../../gdb/gdb/gdbserver/tracepoint.c: In function 'cmd_qtstatus':
../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: unknown conversion type c
haracter 'l' in format
../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: unknown conversion type c
haracter 'l' in format
../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: format '%s' expects type
'char *', but argument 12 has type 'LONGEST'
../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: format '%s' expects type
'char *', but argument 13 has type 'LONGEST'
../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: too many arguments for fo
rmat
make[4]: *** [tracepoint.o] Error 1
make[4]: *** Waiting for unfinished jobs....
make[4]: Leaving directory `/f/build_gdb/gdb/gdbgit/buildNo/gdb/gdbserver'
make[3]: *** [subdir_do] Error 1
make[3]: Leaving directory `/f/build_gdb/gdb/gdbgit/buildNo/gdb'
make[2]: *** [all] Error 2
make[2]: Leaving directory `/f/build_gdb/gdb/gdbgit/buildNo/gdb'
make[1]: *** [all-gdb] Error 2
make[1]: Leaving directory `/f/build_gdb/gdb/gdbgit/buildNo'
make: *** [all] Error 2


It looks like something in tracepoint.c, so I CC to Yao.

Thanks.

Re: [windows] build error on gdbserver

[Ralf Corsepius]

On 11/24/2011 02:11 PM, asmwarrior wrote:
> Hi, all, I use the gdb cvs.
>
> Now, build gdb under msys+mingw has a build error:
>
> make[4]: Entering directory `/f/build_gdb/gdb/gdbgit/buildNo/gdb/gdbserver'
> rm -f version.c-tmp version.c
> echo '#include "server.h"' >> version.c-tmp
> gcc -c -O0 -g -D__USE_MINGW_ACCESS -I. -I../../../gdb/gdb/gdbserver
> -I../../.
> ./gdb/gdb/gdbserver/../common -I../../../gdb/gdb/gdbserver/../regformats
> -I../..
> /../gdb/gdb/gdbserver/../../include -Wall -Wdeclaration-after-statement
> -Wpointe
> r-arith -Wformat-nonliteral -Wno-char-subscripts -Werror
> ../../../gdb/gdb/gdbser
> ver/mem-break.c
> echo 'const char version[] = "'"`sed q
> ../../../gdb/gdb/gdbserver/../version.in`
> "'";' >> version.c-tmp
> gcc -c -O0 -g -D__USE_MINGW_ACCESS -I. -I../../../gdb/gdb/gdbserver
> -I../../.
> ./gdb/gdb/gdbserver/../common -I../../../gdb/gdb/gdbserver/../regformats
> -I../..
> /../gdb/gdb/gdbserver/../../include -Wall -Wdeclaration-after-statement
> -Wpointe
> r-arith -Wformat-nonliteral -Wno-char-subscripts -Werror
> ../../../gdb/gdb/gdbser
> ver/hostio.c
> gcc -c -O0 -g -D__USE_MINGW_ACCESS -I. -I../../../gdb/gdb/gdbserver
> -I../../.
> ./gdb/gdb/gdbserver/../common -I../../../gdb/gdb/gdbserver/../regformats
> -I../..
> /../gdb/gdb/gdbserver/../../include -Wall -Wdeclaration-after-statement
> -Wpointe
> r-arith -Wformat-nonliteral -Wno-char-subscripts -Werror
> ../../../gdb/gdb/gdbser
> ver/event-loop.c
> echo 'const char host_name[] = "mingw32";' >> version.c-tmp
> mv version.c-tmp version.c
> gcc -c -O0 -g -D__USE_MINGW_ACCESS -I. -I../../../gdb/gdb/gdbserver
> -I../../.
> ./gdb/gdb/gdbserver/../common -I../../../gdb/gdb/gdbserver/../regformats
> -I../..
> /../gdb/gdb/gdbserver/../../include -Wall -Wdeclaration-after-statement
> -Wpointe
> r-arith -Wformat-nonliteral -Wno-char-subscripts -Werror
> ../../../gdb/gdb/gdbser
> ver/tracepoint.c
> gcc -c -O0 -g -D__USE_MINGW_ACCESS -I. -I../../../gdb/gdb/gdbserver
> -I../../
> ../gdb/gdb/gdbserver/../common
> -I../../../gdb/gdb/gdbserver/../regformats -I../.
> ./../gdb/gdb/gdbserver/../../include -Wall -Wdeclaration-after-statement
> -Wpoint
> er-arith -Wformat-nonliteral -Wno-char-subscripts -Werror
> ../../../gdb/gdb/gdbse
> rver/../common/xml-utils.c -DGDBSERVER
> gcc -c -O0 -g -D__USE_MINGW_ACCESS -I. -I../../../gdb/gdb/gdbserver
> -I../../
> ../gdb/gdb/gdbserver/../common
> -I../../../gdb/gdb/gdbserver/../regformats -I../.
> ./../gdb/gdb/gdbserver/../../include -Wall -Wdeclaration-after-statement
> -Wpoint
> er-arith -Wformat-nonliteral -Wno-char-subscripts -Werror
> ../../../gdb/gdb/gdbse
> rver/../common/common-utils.c -DGDBSERVER
> gcc -c -O0 -g -D__USE_MINGW_ACCESS -I. -I../../../gdb/gdb/gdbserver
> -I../../
> ../gdb/gdb/gdbserver/../common
> -I../../../gdb/gdb/gdbserver/../regformats -I../.
> ./../gdb/gdb/gdbserver/../../include -Wall -Wdeclaration-after-statement
> -Wpoint
> er-arith -Wformat-nonliteral -Wno-char-subscripts -Werror
> ../../../gdb/gdb/gdbse
> rver/../common/buffer.c -DGDBSERVER
> cc1.exe: warnings being treated as errors
> ../../../gdb/gdb/gdbserver/tracepoint.c: In function 'cmd_qtstatus':
> ../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: unknown
> conversion type c
> haracter 'l' in format
> ../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: unknown
> conversion type c
> haracter 'l' in format
> ../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: format '%s'
> expects type
> 'char *', but argument 12 has type 'LONGEST'
> ../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: format '%s'
> expects type
> 'char *', but argument 13 has type 'LONGEST'
> ../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: too many
> arguments for fo
> rmat
> make[4]: *** [tracepoint.o] Error 1
> make[4]: *** Waiting for unfinished jobs....
> make[4]: Leaving directory `/f/build_gdb/gdb/gdbgit/buildNo/gdb/gdbserver'
> make[3]: *** [subdir_do] Error 1
> make[3]: Leaving directory `/f/build_gdb/gdb/gdbgit/buildNo/gdb'
> make[2]: *** [all] Error 2
> make[2]: Leaving directory `/f/build_gdb/gdb/gdbgit/buildNo/gdb'
> make[1]: *** [all-gdb] Error 2
> make[1]: Leaving directory `/f/build_gdb/gdb/gdbgit/buildNo'
> make: *** [all] Error 2
>
>
> It looks like something in tracepoint.c, so I CC to Yao.

Use "configure --disable-werror ..."

Ralf

Re: [windows] build error on gdbserver

[Eli Zaretskii]

> Date: Thu, 24 Nov 2011 15:11:11 +0100
> From: Ralf Corsepius <ralf.corsepius@rtems.org>
> 
> > ../../../gdb/gdb/gdbserver/tracepoint.c: In function 'cmd_qtstatus':
> > ../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: unknown
> > conversion type character 'l' in format
> > ../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: unknown
> > conversion type character 'l' in format
> > ../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: format '%s'
> > expects type 'char *', but argument 12 has type 'LONGEST'
> > ../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: format '%s'
> > expects type 'char *', but argument 13 has type 'LONGEST'
> > ../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: too many
> > arguments for format
> > make[4]: *** [tracepoint.o] Error 1
> > make[4]: *** Waiting for unfinished jobs....
> > make[4]: Leaving directory `/f/build_gdb/gdb/gdbgit/buildNo/gdb/gdbserver'
> > make[3]: *** [subdir_do] Error 1
> > make[3]: Leaving directory `/f/build_gdb/gdb/gdbgit/buildNo/gdb'
> > make[2]: *** [all] Error 2
> > make[2]: Leaving directory `/f/build_gdb/gdb/gdbgit/buildNo/gdb'
> > make[1]: *** [all-gdb] Error 2
> > make[1]: Leaving directory `/f/build_gdb/gdb/gdbgit/buildNo'
> > make: *** [all] Error 2
> >
> >
> > It looks like something in tracepoint.c, so I CC to Yao.
> 
> Use "configure --disable-werror ..."

And get a buggy binary?  Isn't it clear that the problem is real?

Re: [windows] build error on gdbserver

[asmwarrior]

On 2011-11-24 22:11, Ralf Corsepius wrote:
> Use "configure --disable-werror ..."
For me, I think this is only a work around, the code snippet looks like below:

static void
cmd_qtstatus (char *packet)
{
   char *stop_reason_rsp = NULL;
   char *buf1, *buf2, *buf3, *str;
   int slen;

   /* Translate the plain text of the notes back into hex for
      transmission.  */

   str = (tracing_user_name ? tracing_user_name : "");
   slen = strlen (str);
   buf1 = (char *) alloca (slen * 2 + 1);
   hexify (buf1, str, slen);

   str = (tracing_notes ? tracing_notes : "");
   slen = strlen (str);
   buf2 = (char *) alloca (slen * 2 + 1);
   hexify (buf2, str, slen);

   str = (tracing_stop_note ? tracing_stop_note : "");
   slen = strlen (str);
   buf3 = (char *) alloca (slen * 2 + 1);
   hexify (buf3, str, slen);

   trace_debug ("Returning trace status as %d, stop reason %s",
	       tracing, tracing_stop_reason);

   if (in_process_agent_loaded ())
     {
       pause_all (1);

       upload_fast_traceframes ();

       unpause_all (1);
    }

   stop_reason_rsp = (char *) tracing_stop_reason;

   /* The user visible error string in terror needs to be hex encoded.
      We leave it as plain string in `tracing_stop_reason' to ease
      debugging.  */
   if (strncmp (stop_reason_rsp, "terror:", strlen ("terror:")) == 0)
     {
       const char *result_name;
       int hexstr_len;
       char *p;

       result_name = stop_reason_rsp + strlen ("terror:");
       hexstr_len = strlen (result_name) * 2;
       p = stop_reason_rsp = alloca (strlen ("terror:") + hexstr_len + 1);
       strcpy (p, "terror:");
       p += strlen (p);
       convert_int_to_ascii ((gdb_byte *) result_name, p, strlen (result_name));
     }

   /* If this was a forced stop, include any stop note that was supplied.  */
   if (strcmp (stop_reason_rsp, "tstop") == 0)
     {
       stop_reason_rsp = alloca (strlen ("tstop:") + strlen (buf3) + 1);
       strcpy (stop_reason_rsp, "tstop:");
       strcat (stop_reason_rsp, buf3);
     }

   sprintf (packet,
	   "T%d;"
	   "%s:%x;"
	   "tframes:%x;tcreated:%x;"
	   "tfree:%x;tsize:%s;"
	   "circular:%d;"
	   "disconn:%d;"
	   "starttime:%llx;stoptime:%llx;"
	   "username:%s:;notes:%s:",
	   tracing ? 1 : 0,
	   stop_reason_rsp, tracing_stop_tpnum,
	   traceframe_count, traceframes_created,
	   free_space (), phex_nz (trace_buffer_hi - trace_buffer_lo, 0),
	   circular_trace_buffer,
	   disconnected_tracing,
	   tracing_start_time, tracing_stop_time,
	   buf1, buf2);
}


There is a sprintf call with many parameters. I have no idea about %llx ????

Re: [windows] build error on gdbserver

[Ralf Corsepius]

On 11/24/2011 03:32 PM, Eli Zaretskii wrote:
>> Date: Thu, 24 Nov 2011 15:11:11 +0100
>> From: Ralf Corsepius<ralf.corsepius@rtems.org>
>>
>>> ../../../gdb/gdb/gdbserver/tracepoint.c: In function 'cmd_qtstatus':
>>> ../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: unknown
>>> conversion type character 'l' in format
>>> ../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: unknown
>>> conversion type character 'l' in format
>>> ../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: format '%s'
>>> expects type 'char *', but argument 12 has type 'LONGEST'
>>> ../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: format '%s'
>>> expects type 'char *', but argument 13 has type 'LONGEST'
>>> ../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: too many
>>> arguments for format
>>> make[4]: *** [tracepoint.o] Error 1
>>> make[4]: *** Waiting for unfinished jobs....
>>> make[4]: Leaving directory `/f/build_gdb/gdb/gdbgit/buildNo/gdb/gdbserver'
>>> make[3]: *** [subdir_do] Error 1
>>> make[3]: Leaving directory `/f/build_gdb/gdb/gdbgit/buildNo/gdb'
>>> make[2]: *** [all] Error 2
>>> make[2]: Leaving directory `/f/build_gdb/gdb/gdbgit/buildNo/gdb'
>>> make[1]: *** [all-gdb] Error 2
>>> make[1]: Leaving directory `/f/build_gdb/gdb/gdbgit/buildNo'
>>> make: *** [all] Error 2
>>>
>>>
>>> It looks like something in tracepoint.c, so I CC to Yao.
>> Use "configure --disable-werror ..."
> And get a buggy binary?
You are right, I didn't look carefully enough and missed there was a 
real error.

I had thought this was one of the usual false build errors noise mails 
gdb's -Werror is causing.

>    Isn't it clear that the problem is real?
Apologies, sorry.

Ralf

Re: [windows] build error on gdbserver

[Pedro Alves]

On Thursday 24 November 2011 14:50:47, asmwarrior wrote:
> There is a sprintf call with many parameters. I have no idea about %llx ????

%llx is C99, not C90, so we shouldn't use it in common code; we should
use plongest instead to print LONGEST's.

-- 
Pedro Alves

Re: [windows] build error on gdbserver

[Stan Shebs]

On 11/24/11 5:11 AM, asmwarrior wrote:
> ../../../gdb/gdb/gdbserver/tracepoint.c: In function 'cmd_qtstatus':
> ../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: unknown 
> conversion type c
> haracter 'l' in format
> ../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: unknown 
> conversion type c
> haracter 'l' in format
> ../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: format '%s' 
> expects type
> 'char *', but argument 12 has type 'LONGEST'
> ../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: format '%s' 
> expects type
> 'char *', but argument 13 has type 'LONGEST'
> ../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: too many 
> arguments for fo
> rmat

Ugh, that's me actually, I fixed some of the ll's before committing, but 
missed
those...

Stan

Re: [windows] build error on gdbserver

[asmwarrior]

On 2011-11-25 1:38, Stan Shebs wrote:
> On 11/24/11 5:11 AM, asmwarrior wrote:
>> ../../../gdb/gdb/gdbserver/tracepoint.c: In function 'cmd_qtstatus':
>> ../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: unknown
>> conversion type c
>> haracter 'l' in format
>> ../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: unknown
>> conversion type c
>> haracter 'l' in format
>> ../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: format '%s'
>> expects type
>> 'char *', but argument 12 has type 'LONGEST'
>> ../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: format '%s'
>> expects type
>> 'char *', but argument 13 has type 'LONGEST'
>> ../../../gdb/gdb/gdbserver/tracepoint.c:3460:5: error: too many
>> arguments for fo
>> rmat
>
> Ugh, that's me actually, I fixed some of the ll's before committing, but
> missed
> those...
>
> Stan
>
>
I have seen that you fixed this in the cvs HEAD, thanks!

Now, There is no build error under MinGW.

asmwarrior