[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: qemu-pr-helper -v suppresses errors, isn't that weird?
From: |
Paolo Bonzini |
Subject: |
Re: qemu-pr-helper -v suppresses errors, isn't that weird? |
Date: |
Thu, 18 Jun 2020 14:15:21 +0200 |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101 Thunderbird/68.6.0 |
On 18/06/20 07:32, Markus Armbruster wrote:
> prh_co_entry() reports reports errors reading requests / writing
> responses only when @verbose (command line -v); relevant code appended
> for you convenience.
>
> Sure these are *errors*? The program recovers and continues, and this
> is deemed normal enough to inform the user only when he specifically
> asks for it. Yet when we inform, we format it as an error. Should we
> tune it down to warnings?
They are errors, but they're errors in the client rather than in
qemu-pr-helper.c itself.
Paolo
>
> static void coroutine_fn prh_co_entry(void *opaque)
> {
> [...]
> while (atomic_read(&state) == RUNNING) {
> [...]
> sz = prh_read_request(client, &req, &resp, &local_err);
> if (sz < 0) {
> break;
> }
> [...]
> if (prh_write_response(client, &req, &resp, &local_err) < 0) {
> break;
> }
> }
> if (local_err) {
> if (verbose == 0) {
> error_free(local_err);
> } else {
> error_report_err(local_err);
> }
> }
>
> out:
> qio_channel_detach_aio_context(QIO_CHANNEL(client->ioc));
> object_unref(OBJECT(client->ioc));
> g_free(client);
> }
>