This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [PATCH] Add some testcases
- From: Andrew Burgess <andrew dot burgess at embecosm dot com>
- To: Fei Jie <feij dot fnst at cn dot fujitsu dot com>
- Cc: gdb-patches at sourceware dot org
- Date: Tue, 27 Oct 2015 17:24:17 +0000
- Subject: Re: [PATCH] Add some testcases
- Authentication-results: sourceware.org; auth=none
- References: <1445564512-5542-1-git-send-email-feij dot fnst at cn dot fujitsu dot com>
* Fei Jie <feij.fnst@cn.fujitsu.com> [2015-10-23 09:41:52 +0800]:
> * watchpoint working with non-existed variables
> * "x" command working with inaccessible address
> * "backtrace" command with "-n" parameter
> ---
> diff --git a/gdb/testsuite/gdb.base/display.exp b/gdb/testsuite/gdb.base/display.exp
> index 6e21d9e..5ad7ec5 100644
> --- a/gdb/testsuite/gdb.base/display.exp
> +++ b/gdb/testsuite/gdb.base/display.exp
> @@ -188,6 +188,8 @@ gdb_test "print/a main+4" ".*= $hex.*<.*>.*"
> gdb_test "print/a \$pc" ".*= $hex.*<do_vars+.*>.*"
> gdb_test "print/a &&j" ".*A .* error in expression.*"
>
> +gdb_test "x 0xffffff" "0xffffff.*Cannot access memory at address.*"
> +
There's nothing to guarantee that this address will be unavailable on
all targets.
You have a couple of tests similar to this in a different patch, I
wonder if all these tests, that depend on a particular memory layout,
should be grouped together into a single, separate, test file, so it's
easier to skip them on targets where the tests are not appropriate.
Thanks,
Andrew