This is the mail archive of the
gdb@sourceware.org
mailing list for the GDB project.
Re: Minimum set of actions to support vCont
- From: Simon Marchi <simon dot marchi at polymtl dot ca>
- To: Mathieu Tarral <mathieu dot tarral at protonmail dot com>
- Cc: gdb at sourceware dot org
- Date: Tue, 11 Dec 2018 12:01:16 -0500
- Subject: Re: Minimum set of actions to support vCont
- References: <mpwybaShdcjqwDYecd26eDmMXanC6zWs3o_dZD4vcWWEsbmbsAtsQihuSZFku7knyjcH0UCFl43cDQpHrCUEgtPlIPmACZ_MgralRTzzqws=@protonmail.com>
On 2018-11-23 05:16, Mathieu Tarral wrote:
Hi,
I'm implementing the vCont packet in my stub,
and so far, when I receive a vCont? packet, I reply with
"vCont;c;s" since I support singlestep and continue.
This implementation works in radare2, but it doesn't work with GDB:
Sending packet: $vCont?#49...Packet received: vCont;c;s
Packet vCont (verbose-resume) is NOT supported
What is the minimal set of actions required to support
vCont in GDB RSP ?
Thanks.
--
Mathieu Tarral
Hi Mathieu,
Sorry for the late reply, but I thought it would be better than no reply
at all.
I don't think there is a clear spec about that. The implementation of
GDB is the closest thing to a source of truth that you'll get.
In remote.c, in remote_target::remote_vcont_probe (), I see
/* If c, and C are not all supported, we can't use vCont.
Clearing
BUF will make packet_ok disable the packet. */
if (!support_c || !support_C)
buf[0] = 0;
So I guess GDB wants you to support C (or claim you support C) in order
to use vCont.
Simon