This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [PATCH v5] Implement the ability to set/unset environment variables to GDBserver when starting the inferior
- From: Simon Marchi <simon dot marchi at ericsson dot com>
- To: Thomas Preudhomme <thomas dot preudhomme at foss dot arm dot com>, Sergio Durigan Junior <sergiodj at redhat dot com>
- Cc: GDB Patches <gdb-patches at sourceware dot org>
- Date: Tue, 5 Sep 2017 18:25:51 +0200
- Subject: Re: [PATCH v5] Implement the ability to set/unset environment variables to GDBserver when starting the inferior
- Authentication-results: sourceware.org; auth=none
- Authentication-results: spf=none (sender IP is ) smtp.mailfrom=simon dot marchi at ericsson dot com;
- References: <20170629194106.23070-1-sergiodj@redhat.com> <20170831204932.25219-1-sergiodj@redhat.com> <f2ac2852747e02594f28641f8d1eefff@polymtl.ca> <87o9qv5sza.fsf@redhat.com> <2aea3170-004a-bfe1-3e57-f7e3238167fb@foss.arm.com>
- Spamdiagnosticmetadata: NSPM
- Spamdiagnosticoutput: 1:99
On 2017-09-05 05:32 PM, Thomas Preudhomme wrote:
> Hi Sergio,
>
> I noticed that the share-env-with-gdbserver.exp test you added is skipped if using a gdb stub. However the comment says the test is meaningless for native-gdbserver. Shouldn't the check also look whether this is a cross debugger?
>
> Best regards,
>
> Thomas
Why would a cross debugger make this test irrelevant? If you are debugging
your ARM board from your x86 PC, you can still set environment variables
for new processes that are spawned by gdbserver. I have the feeling that
you are confused by the meanings of the various terms (native/remote/
extended-remote/cross/etc), which is understandable because these terms
have different meanings depending on the context. Can you expand on why
you think this is wrong, and maybe we can find where the possible confusion
comes from?
Simon