diff options
| author | Pedro Alves <pedro@palves.net> | 2024-04-19 14:37:56 +0100 |
|---|---|---|
| committer | Pedro Alves <pedro@palves.net> | 2024-04-26 21:22:47 +0100 |
| commit | ef27d39dd2c2e7b6af70884895e6d751872ae305 (patch) | |
| tree | 50c48e1e648a68debcd8971b525ff0b9d8e314b2 | |
| parent | f1fc8dc2dccb4a224cdd1c3973c7a09a752aa95b (diff) | |
| download | gdb-ef27d39dd2c2e7b6af70884895e6d751872ae305.zip gdb-ef27d39dd2c2e7b6af70884895e6d751872ae305.tar.gz gdb-ef27d39dd2c2e7b6af70884895e6d751872ae305.tar.bz2 | |
gdbserver: Fix vAttach response when attaching is not supported
handle_v_attach calls attach_inferior, which says:
"return -1 if attaching is unsupported, 0 if it succeeded, and call
error() otherwise."
So if attach_inferior return != 0, we have the unsupported case,
meaning we should return the empty packet instead of an error.
In practice, this shouldn't trigger, as vAttach support is supposed to
be reported via qSupported. But it doesn't hurt to be pedantic here.
Change-Id: I99cce6fa678f2370571e6bca0657451300956127
Approved-By: Tom Tromey <tom@tromey.com>
| -rw-r--r-- | gdbserver/server.cc | 5 |
1 files changed, 4 insertions, 1 deletions
diff --git a/gdbserver/server.cc b/gdbserver/server.cc index 2633df0..789af36 100644 --- a/gdbserver/server.cc +++ b/gdbserver/server.cc @@ -3318,7 +3318,10 @@ handle_v_attach (char *own_buf) prepare_resume_reply (own_buf, cs.last_ptid, cs.last_status); } else - write_enn (own_buf); + { + /* Not supported. */ + own_buf[0] = 0; + } } catch (const gdb_exception_error &exception) { |