public inbox for ecos-discuss@sourceware.org
 help / color / mirror / Atom feed
* [ECOS] building development tools
@ 2000-12-05  3:40 Andrey Panchenko
  2000-12-05 13:39 ` Jonathan Larmour
  0 siblings, 1 reply; 5+ messages in thread
From: Andrey Panchenko @ 2000-12-05  3:40 UTC (permalink / raw)
  To: ecos-discuss

Hi All,

I'm getting problems Building the Intel x86 development tools for
Windows. I've successfully installed binutils -2.10.1, and then I had
patched gcc 2.95.2,puted path to new binutils to the path environment
on the beginning, configured gcc, and when I try to build it I get
following problem:

In file included from function.c:48:
insn-flags.h:425: conflicting types for `gen_beq'
rtl.h:1032: previous declaration of `gen_beq'
insn-flags.h:431: conflicting types for `gen_bge'
rtl.h:1033: previous declaration of `gen_bge'
insn-flags.h:433: conflicting types for `gen_ble'
rtl.h:1034: previous declaration of `gen_ble'
insn-flags.h:435: conflicting types for `gen_jump'
rtl.h:1031: previous declaration of `gen_jump'
function.c: In function `expand_function_end':
function.c:6505: too many arguments to function `gen_jump'
make[1]: *** [function.o] Error 1
make[1]: Leaving directory `/home/andrey/tmp/gcc-2.95.2/gcc'
make: *** [all-gcc] Error 2
make: Leaving directory `/home/andrey/tmp/gcc-2.95.2'


Does anybody familiar with this kind of problem ?
Any suggestions ?
Thanks in advance !

Best regards,
 Andrey                            mailto:ur4ukj@gmx.de


^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [ECOS] building development tools
  2000-12-05  3:40 [ECOS] building development tools Andrey Panchenko
@ 2000-12-05 13:39 ` Jonathan Larmour
  2000-12-06  2:42   ` Re[2]: " Andrey Panchenko
  0 siblings, 1 reply; 5+ messages in thread
From: Jonathan Larmour @ 2000-12-05 13:39 UTC (permalink / raw)
  To: Andrey Panchenko; +Cc: ecos-discuss

Andrey Panchenko wrote:
> 
> I'm getting problems Building the Intel x86 development tools for
> Windows. I've successfully installed binutils -2.10.1, and then I had
> patched gcc 2.95.2,puted path to new binutils to the path environment
> on the beginning, configured gcc, and when I try to build it I get
> following problem:
> 
> In file included from function.c:48:
> insn-flags.h:425: conflicting types for `gen_beq'
> rtl.h:1032: previous declaration of `gen_beq'
> insn-flags.h:431: conflicting types for `gen_bge'
> rtl.h:1033: previous declaration of `gen_bge'
> insn-flags.h:433: conflicting types for `gen_ble'
> rtl.h:1034: previous declaration of `gen_ble'
> insn-flags.h:435: conflicting types for `gen_jump'
> rtl.h:1031: previous declaration of `gen_jump'
> function.c: In function `expand_function_end':
> function.c:6505: too many arguments to function `gen_jump'
> make[1]: *** [function.o] Error 1
> make[1]: Leaving directory `/home/andrey/tmp/gcc-2.95.2/gcc'
> make: *** [all-gcc] Error 2
> make: Leaving directory `/home/andrey/tmp/gcc-2.95.2'
> 
> Does anybody familiar with this kind of problem ?

It's the first time I've seen this. Can you have a look at insn-flags.h (in
the build tree) and rtl.h (in the source tree) to see what the two
prototypes are? In a build tree I have here, they are both:

extern rtx gen_beq                             PROTO((rtx));

Jifl
-- 
Red Hat, 35 Cambridge Place, Cambridge, UK. CB2 1NS  Tel: +44 (1223) 728762
"Plan to be spontaneous tomorrow."  ||  These opinions are all my own fault

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re[2]: [ECOS] building development tools
  2000-12-05 13:39 ` Jonathan Larmour
@ 2000-12-06  2:42   ` Andrey Panchenko
  2000-12-06 12:23     ` Jonathan Larmour
  0 siblings, 1 reply; 5+ messages in thread
From: Andrey Panchenko @ 2000-12-06  2:42 UTC (permalink / raw)
  To: Jonathan Larmour; +Cc: ecos-discuss

Hello Jonathan,

Dienstag, Dienstag, 5. Dezember 2000, you wrote:

>> I'm getting problems Building the Intel x86 development tools for
>> Windows. I've successfully installed binutils -2.10.1, and then I had
>> patched gcc 2.95.2,puted path to new binutils to the path environment
>> on the beginning, configured gcc, and when I try to build it I get
>> following problem:
>> 
>> In file included from function.c:48:
>> insn-flags.h:425: conflicting types for `gen_beq'
>> rtl.h:1032: previous declaration of `gen_beq'
>> insn-flags.h:431: conflicting types for `gen_bge'
>> rtl.h:1033: previous declaration of `gen_bge'
>> insn-flags.h:433: conflicting types for `gen_ble'
>> rtl.h:1034: previous declaration of `gen_ble'
>> insn-flags.h:435: conflicting types for `gen_jump'
>> rtl.h:1031: previous declaration of `gen_jump'
>> function.c: In function `expand_function_end':
>> function.c:6505: too many arguments to function `gen_jump'
>> make[1]: *** [function.o] Error 1
>> make[1]: Leaving directory `/home/andrey/tmp/gcc-2.95.2/gcc'
>> make: *** [all-gcc] Error 2
>> make: Leaving directory `/home/andrey/tmp/gcc-2.95.2'
>> 
>> Does anybody familiar with this kind of problem ?

JL> It's the first time I've seen this. Can you have a look at insn-flags.h (in
JL> the build tree) and rtl.h (in the source tree) to see what the two
JL> prototypes are? In a build tree I have here, they are both:

JL> extern rtx gen_beq                             PROTO((rtx));

Yes,
in  insn-flags.h:      extern rtx gen_beq            PROTO((void));

in rtl.h              extern rtx gen_beq            PROTO((rtx));

So, configure script is not working very properly.
I've tried to change void to rtx but it a moment it failed again:

stmt.c: In function `expand_end_case':
stmt.c:5292: too many arguments to function `gen_casesi'
make[1]: *** [stmt.o] Error 1

And changes in gen_casesi from void to rtx don't help.
Any suggestions ?

JL> Jifl

Best regards,
 Andrey                            mailto:ur4ukj@gmx.de


^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [ECOS] building development tools
  2000-12-06  2:42   ` Re[2]: " Andrey Panchenko
@ 2000-12-06 12:23     ` Jonathan Larmour
  2000-12-06 12:42       ` Re[2]: " Andrey Panchenko
  0 siblings, 1 reply; 5+ messages in thread
From: Jonathan Larmour @ 2000-12-06 12:23 UTC (permalink / raw)
  To: Andrey Panchenko; +Cc: ecos-discuss

[-- Attachment #1: Type: text/plain, Size: 1748 bytes --]

Andrey Panchenko wrote:
> 
> Hello Jonathan,
> 
> Dienstag, Dienstag, 5. Dezember 2000, you wrote:
> 
> JL> It's the first time I've seen this. Can you have a look at insn-flags.h (in
> JL> the build tree) and rtl.h (in the source tree) to see what the two
> JL> prototypes are? In a build tree I have here, they are both:
> 
> JL> extern rtx gen_beq                             PROTO((rtx));
> 
> Yes,
> in  insn-flags.h:      extern rtx gen_beq            PROTO((void));
> 
> in rtl.h              extern rtx gen_beq            PROTO((rtx));
> 
> So, configure script is not working very properly.
> I've tried to change void to rtx but it a moment it failed again:
> 
> stmt.c: In function `expand_end_case':
> stmt.c:5292: too many arguments to function `gen_casesi'
> make[1]: *** [stmt.o] Error 1
> 
> And changes in gen_casesi from void to rtx don't help.
> Any suggestions ?

Erk, not really. It may be worth analysing the build's output carefully to
make sure there isn't something obvious being reported that wasn't treated
as a fatal error. Perhaps just try rebuilding again from an empty build
tree?

The generation of insn-flags.h is handled by the genflags program which is
part of the build. However it's meant to "just work". I just tried it now
for me for the i386-elf target and it worked for example. If this is the
same target you are using, try installing the attached insn-flags.h file in
the gcc subdirectory of your build tree.

Otherwise ask on the gcc@gcc.gnu.org list. You can find details at
http://gcc.gnu.org/

Sorry I can't be more help.

Jifl
-- 
Red Hat, 35 Cambridge Place, Cambridge, UK. CB2 1NS  Tel: +44 (1223) 728762
"Plan to be spontaneous tomorrow."  ||  These opinions are all my own fault
insn-flags.h.gz


[-- Attachment #2: insn-flags.h.gz --]
[-- Type: application/x-gzip, Size: 3193 bytes --]

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re[2]: [ECOS] building development tools
  2000-12-06 12:23     ` Jonathan Larmour
@ 2000-12-06 12:42       ` Andrey Panchenko
  0 siblings, 0 replies; 5+ messages in thread
From: Andrey Panchenko @ 2000-12-06 12:42 UTC (permalink / raw)
  To: Jonathan Larmour; +Cc: ecos-discuss

Hello Jonathan,

Mittwoch, Mittwoch, 6. Dezember 2000, you wrote:

>> Yes,
>> in  insn-flags.h:      extern rtx gen_beq            PROTO((void));
>> 
>> in rtl.h              extern rtx gen_beq            PROTO((rtx));
>> 
>> So, configure script is not working very properly.
>> I've tried to change void to rtx but it a moment it failed again:
>> 
>> stmt.c: In function `expand_end_case':
>> stmt.c:5292: too many arguments to function `gen_casesi'
>> make[1]: *** [stmt.o] Error 1
>> 
>> And changes in gen_casesi from void to rtx don't help.
>> Any suggestions ?

[ ... ]

JL> The generation of insn-flags.h is handled by the genflags program which is
JL> part of the build. However it's meant to "just work". I just tried it now
JL> for me for the i386-elf target and it worked for example. If this is the
JL> same target you are using, try installing the attached insn-flags.h file in
JL> the gcc subdirectory of your build tree.

I've tried your h-file and that step passed, but later:

./genrecog ./config/i386/i386.md > tmp-recog.c
genrecog: Two actions at one point in tree for insns "tstsf_cc" (6) and "tstdf_c
c" (8)
after 31 definitions
make[1]: *** [s-recog] Error 33

:(


Best regards,
 Andrey                            mailto:ur4ukj@gmx.de


^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2000-12-06 12:42 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2000-12-05  3:40 [ECOS] building development tools Andrey Panchenko
2000-12-05 13:39 ` Jonathan Larmour
2000-12-06  2:42   ` Re[2]: " Andrey Panchenko
2000-12-06 12:23     ` Jonathan Larmour
2000-12-06 12:42       ` Re[2]: " Andrey Panchenko

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).