This is the mail archive of the
binutils@sources.redhat.com
mailing list for the binutils project.
Re: PATCH: Fix gas/mips/vr4120.d
- From: "H. J. Lu" <hjl at lucon dot org>
- To: cgd at broadcom dot com
- Cc: binutils at sources dot redhat dot com
- Date: Mon, 30 Sep 2002 17:14:21 -0700
- Subject: Re: PATCH: Fix gas/mips/vr4120.d
- References: <20020930140049.A4494@lucon.org> <mailpost.1033419988.3181@news-sj1-1> <yov5zntzoxre.fsf@broadcom.com>
On Mon, Sep 30, 2002 at 04:36:53PM -0700, cgd@broadcom.com wrote:
> At Mon, 30 Sep 2002 21:06:28 +0000 (UTC), "H. J. Lu" wrote:
> > FYI, I checked in the following patch to fix gas/mips/vr4120.d.
> >
> > 2002-09-30 H.J. Lu <hjl@gnu.org>
> >
> > * gas/mips/vr4120.d: Pass -m mips:4120 to objdump.
>
> Err, uh, I don't understand, why is this change necessary?
>
> (I don't really care, I just want to understand.)
>
> Assuming ELF, the assemble w/ -march=vr4120 causes the 4120 'machine
> type' bits to be set in the ELF header (for better or for worse).
>
> That in turn seems to cause obdump to dump the correct disassembly,
> even without the specific flag.
>
> (This test could be seen as testing two things: disassembly, and ELF
> machine type handling. You effectively removed the test for the
> latter.)
>
You are right.
>
> (FWIW, it passes just fine on mips-linux w/o the additional flag to
> objdump. Also, you changed more than that according to your patch...
> 8-)
I checked in this patch instead.
H.J.
---
2002-09-30 H.J. Lu <hjl@gnu.org>
* gas/mips/vr4120.d: Use "#pass" instead of "#..." to skip the
rest of output.
--- vr4120.d.mips Mon Sep 30 09:13:32 2002
+++ vr4120.d Mon Sep 30 17:09:53 2002
@@ -26,4 +26,4 @@ Disassembly of section \.text:
+4c: 00a62428 maccs a0,a1,a2
+50: 00a62068 maccu a0,a1,a2
+54: 00a62468 maccus a0,a1,a2
-#...
+#pass