This is the mail archive of the gdb-patches@sourceware.org mailing list for the GDB project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: [SH] gdb.dwarf2 tests: use .byte4 instead of .long

From: Jan Kratochvil <jan dot kratochvil at redhat dot com>
To: Pedro Alves <palves at redhat dot com>
Cc: Thomas Schwinge <thomas at codesourcery dot com>, gdb-patches at sourceware dot org, Kevin Buettner <kevinb at redhat dot com>
Date: Wed, 14 Mar 2012 21:25:54 +0100
Subject: Re: [SH] gdb.dwarf2 tests: use .byte4 instead of .long
References: <87d38gl7r2.fsf@schwinge.name> <20120314193509.GA30809@host2.jankratochvil.net> <4F60FCC3.1010304@redhat.com>

On Wed, 14 Mar 2012 21:17:07 +0100, Pedro Alves wrote:
> Is this due to the test having been originally written by adjusting gcc output?

No, amd64 gcc just produces .long and .value.

> Maybe gcc could be changed to emit .byte4 instead?

I was looking at it once already but there are some concerns with
compatibility with various non-GNU assemblers I was unaware of at all.
As the assembler directives are chosen according to the target arch GCC can
always adjust it accordingly.  Multi-arch GDB testfiles cannot.

It is true this .4byte directive probably breaks compatibility with some
non-GNU assemblers but I believe easy cross-arch compatibility is of a more
concern to GDB than such non-GNU assemblers compatibility.  Or is it?

Thanks,
Jan

Follow-Ups:
- Re: [SH] gdb.dwarf2 tests: use .byte4 instead of .long
  - From: Pedro Alves

References:
- [SH] gdb.dwarf2 tests: use .byte4 instead of .long
  - From: Thomas Schwinge
- Re: [SH] gdb.dwarf2 tests: use .byte4 instead of .long
  - From: Jan Kratochvil
- Re: [SH] gdb.dwarf2 tests: use .byte4 instead of .long
  - From: Pedro Alves

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]