This is the mail archive of the
binutils@sourceware.org
mailing list for the binutils project.
Re: RFC: Allow the suppression of expensive linker and gas tests
- From: Alan Modra <amodra at bigpond dot net dot au>
- To: Nick Clifton <nickc at redhat dot com>
- Cc: binutils at sourceware dot org
- Date: Mon, 12 Nov 2007 09:32:53 +1030
- Subject: Re: RFC: Allow the suppression of expensive linker and gas tests
- References: <m3lk99q9w6.fsf@redhat.com>
On Thu, Nov 08, 2007 at 11:00:57AM +0000, Nick Clifton wrote:
> * gas/macros/macros.exp: Suppress the execution of the purge
> test if the environment variable NO_EXPENSIVE_GAS_TESTS
> exists.
I'll admit to having this one commented out in my local trees. How
about we remove the irpc loops at the end, presumably there to test
that memory is reclaimed?
a) It doesn't make sense to invoke a little used feature of the gas
macro processor almost half a million times.
b) We are unlikely to regress this feature.
c) "Out of memory" might not occur on a large system even if macro
memory was not reclaimed.
* gas/macros/purge.s: Delete irpc loops.
Index: gas/testsuite/gas/macros/purge.s
===================================================================
RCS file: /cvs/src/src/gas/testsuite/gas/macros/purge.s,v
retrieving revision 1.3
diff -u -p -r1.3 purge.s
--- gas/testsuite/gas/macros/purge.s 25 Nov 2005 02:11:40 -0000 1.3
+++ gas/testsuite/gas/macros/purge.s 11 Nov 2007 22:53:06 -0000
@@ -25,18 +25,3 @@
macro2
.purgem MACRO1
.purgem macro2
-
- .irpc a,ABCDEFGHIJKLMNOPQRSTUVWXYZ
- .irpc b,ABCDEFGHIJKLMNOPQRSTUVWXYZ
- .irpc c,ABCDEFGHIJKLMNOPQRSTUVWXYZ
- .irpc d,ABCDEFGHIJKLMNOPQRSTUVWXYZ
- .macro _\a\b\c\d arg1=0, arg2=0
- .if \arg1 + \arg2
- .purgem _\a\b\c\d
- .endif
- .endm
- _\a\b\c\d 1, 2
- .endr
- .endr
- .endr
- .endr
--
Alan Modra
Australia Development Lab, IBM