|
From: | Paolo Bonzini |
Subject: | Re: [PATCH v2 09/12] qapi/meson: Restrict qdev code to system-mode emulation |
Date: | Tue, 26 Jan 2021 16:46:33 +0100 |
User-agent: | Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Thunderbird/78.6.0 |
On 24/01/21 01:32, Philippe Mathieu-Daudé wrote:
The alternative is ugly: -- >8 -- diff --git a/hw/core/qdev.c b/hw/core/qdev.c index cefc5eaa0a9..d09f32e38d6 100644 --- a/hw/core/qdev.c +++ b/hw/core/qdev.c @@ -950,7 +950,9 @@ static void device_finalize(Object *obj) if (dev->pending_deleted_event) { g_assert(dev->canonical_path); +#ifdef CONFIG_USER_ONLY qapi_event_send_device_deleted(!!dev->id, dev->id, dev->canonical_path); +#endif g_free(dev->canonical_path); dev->canonical_path = NULL; } ---
Well, the alternative is also to not do anything. There's already a stub for qapi_event_emit, does it really hurt to have the not-stubbed-out code for qapi_event_send_device_deleted?
Paolo
[Prev in Thread] | Current Thread | [Next in Thread] |