From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:45955) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dG6Rx-0005mp-SK for qemu-devel@nongnu.org; Wed, 31 May 2017 12:26:51 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1dG6Rb-0007uU-M7 for qemu-devel@nongnu.org; Wed, 31 May 2017 12:26:21 -0400 Received: from mail-wm0-x22e.google.com ([2a00:1450:400c:c09::22e]:36346) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1dG6Rb-0007tR-Cx for qemu-devel@nongnu.org; Wed, 31 May 2017 12:25:59 -0400 Received: by mail-wm0-x22e.google.com with SMTP id 7so125023128wmo.1 for ; Wed, 31 May 2017 09:25:59 -0700 (PDT) References: <20170531150933.10156-1-alex.bennee@linaro.org> <20170531150933.10156-3-alex.bennee@linaro.org> <20170531181737.230cbded@p-imbrenda.boeblingen.de.ibm.com> From: Alex =?utf-8?Q?Benn=C3=A9e?= In-reply-to: <20170531181737.230cbded@p-imbrenda.boeblingen.de.ibm.com> Date: Wed, 31 May 2017 17:26:13 +0100 Message-ID: <87y3tdro0q.fsf@linaro.org> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit Subject: Re: [Qemu-devel] [PATCH v1 2/2] gdbstub: don't fail on vCont; C04:0; c packets List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Claudio Imbrenda Cc: pbonzini@redhat.com, doug16k@gmail.com, qemu-devel@nongnu.org Claudio Imbrenda writes: > On Wed, 31 May 2017 16:09:33 +0100 > Alex Bennée wrote: > >> The thread-id of 0 means any CPU but we then ignore the fact we find >> the first_cpu in this case who can have an index of 0. Instead of >> bailing out just test if we have managed to match up thread-id to a >> CPU. >> >> Otherwise you get: >> gdb_handle_packet: command='vCont;C04:0;c' >> put_packet: reply='E22' >> >> Signed-off-by: Alex Bennée >> --- >> gdbstub.c | 4 ++-- >> 1 file changed, 2 insertions(+), 2 deletions(-) >> >> diff --git a/gdbstub.c b/gdbstub.c >> index a249846954..29c9ed3002 100644 >> --- a/gdbstub.c >> +++ b/gdbstub.c >> @@ -934,8 +934,8 @@ static int gdb_handle_vcont(GDBState *s, const >> char *p) >> * CPU first, and only then we can use its index. >> */ >> cpu = find_cpu(idx); >> - /* invalid CPU/thread specified */ >> - if (!idx || !cpu) { >> + /* invalid thread specified, cpu not found. */ >> + if (!cpu) { >> res = -EINVAL; >> goto out; >> } > > This is strange. cpu_index() is defined as: > > static inline int cpu_index(CPUState *cpu) > { > #if defined(CONFIG_USER_ONLY) > return cpu->host_tid; > #else > return cpu->cpu_index + 1; > #endif > } > > therefore it shouldn't return 0 under any circumstance, and > find_cpu(idx) should also fail if idx == 0, because internally it also > uses cpu_index() > > on the other hand, you say that the patch does fix the problem for you, > which really confuses me. Hmm that was me assuming cpu_index(cpu) was the same as cpu->cpu_index. However in the CONFIG_USER_ONLY case we return host_tid which is only set by clone_func() so I think will not be set for the first cpu. > > > > (probably) completely unrelatedly, this: > > res = qemu_strtoul(p + 1, &p, 16, &tmp); > > should be like this instead: > > res = qemu_strtoul(p, &p, 16, &tmp); > > but this shouldn't impact you in any way. -- Alex Bennée