[patch] Fix build if using --with-system-readline

[patch] Fix build if using --with-system-readline

[Jan Kratochvil]

Hi,

./configure --enable-targets=all --with-system-readline; make
->
gcc -g -O2   -I. -I. -I./common -I./config -DLOCALEDIR="\"/usr/local/share/locale\"" -DHAVE_CONFIG_H -I./../include/opcode  -I../bfd -I./../bfd -I./../include -I../libdecnumber -I./../libdecnumber  -I./gnulib -Ignulib  -DMI_OUT=1 -DTUI=1   -Wall -Wdeclaration-after-statement -Wpointer-arith -Wformat-nonliteral -Wno-pointer-sign -Wno-unused -Wunused-value -Wno-switch -Wno-char-subscripts -Werror -c -o frv-tdep.o -MT frv-tdep.o -MMD -MP -MF .deps/frv-tdep.Tpo frv-tdep.c
frv-tdep.c:35:60: error: opcodes/frv-desc.h: No such file or directory

(it is not `make -j' race)

because with --with-system-readline there is no:
	-I./../readline/..

OK to check-in?


Thanks,
Jan


gdb/
2009-11-29  Jan Kratochvil  <jan.kratochvil@redhat.com>

	Fix build if using --with-system-readline.
	* Makefile.in (OPCODES_CFLAGS): Add `-I$(OPCODES_DIR)/..'.

--- a/gdb/Makefile.in
+++ b/gdb/Makefile.in
@@ -301,7 +301,7 @@ OPCODES = $(OPCODES_DIR)/libopcodes.a
 # Where are the other opcode tables which only have header file
 # versions?
 OP_INCLUDE = $(INCLUDE_DIR)/opcode
-OPCODES_CFLAGS = -I$(OP_INCLUDE)
+OPCODES_CFLAGS = -I$(OP_INCLUDE) -I$(OPCODES_DIR)/..
 
 # The simulator is usually nonexistent; targets that include one
 # should set this to list all the .o or .a files to be linked in.

Re: [patch] Fix build if using --with-system-readline

[Doug Evans]

On Sun, Nov 29, 2009 at 12:57 PM, Jan Kratochvil
<jan.kratochvil@redhat.com> wrote:
> Hi,
>
> ./configure --enable-targets=all --with-system-readline; make
> ->
> gcc -g -O2   -I. -I. -I./common -I./config -DLOCALEDIR="\"/usr/local/share/locale\"" -DHAVE_CONFIG_H -I./../include/opcode  -I../bfd -I./../bfd -I./../include -I../libdecnumber -I./../libdecnumber  -I./gnulib -Ignulib  -DMI_OUT=1 -DTUI=1   -Wall -Wdeclaration-after-statement -Wpointer-arith -Wformat-nonliteral -Wno-pointer-sign -Wno-unused -Wunused-value -Wno-switch -Wno-char-subscripts -Werror -c -o frv-tdep.o -MT frv-tdep.o -MMD -MP -MF .deps/frv-tdep.Tpo frv-tdep.c
> frv-tdep.c:35:60: error: opcodes/frv-desc.h: No such file or directory
>
> (it is not `make -j' race)
>
> because with --with-system-readline there is no:
>        -I./../readline/..
>
> OK to check-in?
>
>
> Thanks,
> Jan
>
>
> gdb/
> 2009-11-29  Jan Kratochvil  <jan.kratochvil@redhat.com>
>
>        Fix build if using --with-system-readline.
>        * Makefile.in (OPCODES_CFLAGS): Add `-I$(OPCODES_DIR)/..'.
>
> --- a/gdb/Makefile.in
> +++ b/gdb/Makefile.in
> @@ -301,7 +301,7 @@ OPCODES = $(OPCODES_DIR)/libopcodes.a
>  # Where are the other opcode tables which only have header file
>  # versions?
>  OP_INCLUDE = $(INCLUDE_DIR)/opcode
> -OPCODES_CFLAGS = -I$(OP_INCLUDE)
> +OPCODES_CFLAGS = -I$(OP_INCLUDE) -I$(OPCODES_DIR)/..
>
>  # The simulator is usually nonexistent; targets that include one
>  # should set this to list all the .o or .a files to be linked in.
>


That is the easiest patch, and it doesn't actually change anything
since we're already building with (effectively) -I$(srcdir).
So it's ok with me.
I'd wait a bit to see if anyone else wants to speak up.

Having said that, -I$(srcdir) "feels" less clean than
-I$(OPCODES_DIR).  We have -$(BFD_SRC) already, it seems like opcodes
should be treated no different.
I realize going this route involves more work, fixing all the files
that do #include "opcodes/mumble.h", there's only a handful of them
though.
I'd skip putting in the effort to go this route until someone else agrees.