From: Andrew Cagney <cagney@highland.com.au>
To: gas2@cygnus.com
Cc: meissner@cygnus.com
Subject: gas-950822/powerpcle-unknown-eabisim - `gas -mbig' still little ...
Date: Wed, 30 Aug 1995 13:36:00 -0000 [thread overview]
Message-ID: <199508301450.AAA08567@puddin.highland.com.au> (raw)
Hello,
This one has puzzled me for a while. Gas, when configured for LE
PowerPC, should still be able to generate BE object files.
Mysteriously it doesn't. (ditto for BE configured gas generating LE
object files). For instance, the assembler:
bash$ cat t.s
.text
.long 0x12345678
bash$ ./as.new -o t.o -mbig -v t.s
GNU assembler version 950822 (powerpcle-unknown-eabisim), using BFD version cygnus-2.5
would give:
bash$ powerpcle-unknown-eabisim-objdump -d t.o
t.o: file format elf32-powerpc
No symbols in "t.o".
Disassembly of section .text:
00000000 rldcr r22,r2,r6,16
instead of:
bash$ powerpcle-unknown-eabisim-objdump -d t.o
t.o: file format elf32-powerpc
No symbols in "t.o".
Disassembly of section .text:
00000000 .long 0x12345678
Seems that md_begin() was clobering md_parse_arg()'s setting of
target_big_endian.
The patch below hopefully fixes this and also elimates what I think
are now redundant header definitions.
Andrew
*** gas-950822/gas/SRC/ChangeLog Mon Aug 28 19:47:32 1995
--- gas-950822/gas/ChangeLog Thu Aug 31 00:10:56 1995
***************
*** 1,3 ****
--- 1,17 ----
+ Thu Aug 31 00:00:38 1995 Andrew Cagney - aka Noid <cagney@kremvax>
+
+ * config/tc-ppc.h: Eliminate redundant definitions of
+ TARGET_BYTES_BIG_ENDIAN, TARGET_BYTES_LITTLE_ENDIAN and
+ PPC_BIG_ENDIAN. The variable target_big_endian now contains this
+ information and is set in read.c/tc-ppc.c.
+
+ Perhaphs even `target_big_endian' should be initialised from
+ stdoutput->xvec->byteorder_big_p.
+
+ * config/tc-ppc.c (md_begin): Delete code incorrectly setting the
+ variable `target_big_endian' to PPC_BIG_ENDIAN. This is now done
+ by read.c/tc-ppc.c.
+
Tue Aug 22 03:00:33 1995 Ken Raeburn <raeburn@kr-laptop.cygnus.com>
Sat Aug 19 18:08:16 1995 Pat Rankin <rankin@eql.caltech.edu>
*** gas-950822/gas/config/SRC/tc-ppc.h Fri Aug 18 19:23:22 1995
--- gas-950822/gas/config/tc-ppc.h Wed Aug 30 23:51:49 1995
***************
*** 63,81 ****
#define DIFF_EXPR_OK /* .-foo gets turned into PC relative relocs */
#endif
- /* Set the endianness we are using. Default to big endian. */
- #ifndef TARGET_BYTES_BIG_ENDIAN
- #ifndef TARGET_BYTES_LITTLE_ENDIAN
- #define TARGET_BYTES_BIG_ENDIAN 1
- #endif
- #endif
-
- #ifdef TARGET_BYTES_BIG_ENDIAN
- #define PPC_BIG_ENDIAN 1
- #else
- #define PPC_BIG_ENDIAN 0
- #endif
-
/* We don't need to handle .word strangely. */
#define WORKING_DOT_WORD
--- 63,68 ----
*** gas-950822/gas/config/SRC/tc-ppc.c Mon Aug 28 19:47:41 1995
--- gas-950822/gas/config/tc-ppc.c Wed Aug 30 23:47:27 1995
***************
*** 436,444 ****
}
}
- /* Tell the main code what the endianness is. */
- target_big_endian = PPC_BIG_ENDIAN;
-
#ifdef OBJ_COFF
ppc_coff_debug_section = coff_section_from_bfd_index (stdoutput, N_DEBUG);
--- 436,441 ----
next reply other threads:[~1995-08-30 13:36 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
1995-08-30 13:36 Andrew Cagney [this message]
1995-08-30 14:14 ` Michael Meissner
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=199508301450.AAA08567@puddin.highland.com.au \
--to=cagney@highland.com.au \
--cc=gas2@cygnus.com \
--cc=meissner@cygnus.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).