Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
* Re: [committed testsuite patch] auxv.exp cascading failures
@ 2004-06-04 23:44 Michael Elizabeth Chastain
  2004-06-05  0:11 ` Roland McGrath
  0 siblings, 1 reply; 3+ messages in thread
From: Michael Elizabeth Chastain @ 2004-06-04 23:44 UTC (permalink / raw)
  To: gdb-patches, roland; +Cc: mec.gnu

It's PR gdb/1647, sorry if the number got detached somewhere.

I have a concern about this patch.  Code like this:

	-re "The program has no auxiliary information now.*$" {
 	    set bad 1
	    exp_continue
 	}
	...

It might consume an arbitrary number of characters with ".*",
maybe including a $gdb_prompt, maybe not.  Maybe even consuming
only part of the prompt!  We're at the mercy of OS buffering
for how many characters get delivered to expect at once.
Jim B has seen a lot of this lately.  So I don't think the ".*$"
is robust.

Michael C


^ permalink raw reply	[flat|nested] 3+ messages in thread
* [committed testsuite patch] auxv.exp cascading failures
@ 2004-06-04 21:37 Roland McGrath
  0 siblings, 0 replies; 3+ messages in thread
From: Roland McGrath @ 2004-06-04 21:37 UTC (permalink / raw)
  To: gdb-patches; +Cc: Michael Elizabeth Chastain

I've committed this obvious fix to the test case I added a while back.
This fixes a report from Michael, but the message I got contains no PR number.


Thanks,
Roland


2004-06-04  Roland McGrath  <roland@redhat.com>

	* gdb.base/auxv.exp (fetch_auxv): Consume output fully through
	next gdb prompt.


--- auxv.exp.~1.1.~	2004-03-16 13:39:57.000000000 -0800
+++ auxv.exp	2004-06-04 14:33:06.000000000 -0700
@@ -80,14 +80,17 @@ proc fetch_auxv {test} {
 	-re "info auxv\[\r\n\]+" {
 	    exp_continue
 	}
-	-ex "The program has no auxiliary information now" {
+	-re "The program has no auxiliary information now.*$" {
 	    set bad 1
+	    exp_continue
 	}
-	-ex "Auxiliary vector is empty" {
+	-re "Auxiliary vector is empty.*$" {
 	    set bad 1
+	    exp_continue
 	}
-	-ex "No auxiliary vector found" {
+	-re "No auxiliary vector found.*$" {
 	    set bad 1
+	    exp_continue
 	}
 	-re "^\[0-9\]+\[ \t\]+(AT_\[^ \t\]+)\[^\r\n\]+\[\r\n\]+" {
 	    lappend auxv_lines $expect_out(0,string)
@@ -99,12 +102,13 @@ proc fetch_auxv {test} {
 	    lappend auxv_lines $expect_out(0,string)
 	    exp_continue
 	}
-	-re ".*$gdb_prompt $" {
-	    incr bad
-	}
 	-re "^\[^\r\n\]+\[\r\n\]+" {
 	    warning "Unrecognized output: $expect_out(0,string)"
 	    set bad 1
+	    exp_continue
+	}
+	-re ".*$gdb_prompt $" {
+	    incr bad
 	}
     }] != 0} {
 	return {}


^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2004-06-05  0:11 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2004-06-04 23:44 [committed testsuite patch] auxv.exp cascading failures Michael Elizabeth Chastain
2004-06-05  0:11 ` Roland McGrath
  -- strict thread matches above, loose matches on Subject: below --
2004-06-04 21:37 Roland McGrath

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox