On 11/09/2024 14.51, Richard W.M. Jones wrote:
On Wed, Sep 11, 2024 at 02:46:18PM +0200, Maciej S. Szmigiero wrote:
On 11.09.2024 14:37, Eric Blake wrote:
On Wed, Sep 11, 2024 at 07:33:59AM GMT, Eric Blake wrote:
On Tue, Sep 10, 2024 at 03:15:28PM GMT, Pierrick Bouvier wrote:
Signed-off-by: Pierrick Bouvier <pierrick.bouvier@xxxxxxxxxx>
---
A general suggestion for the entire series: please use a commit
message that explains why this is a good idea. Even something as
boiler-plate as "refer to commit XXX for rationale" that can be
copy-pasted into all the other commits is better than nothing,
although a self-contained message is best. Maybe:
This patch is part of a series that moves towards a consistent use of
g_assert_not_reached() rather than an ad hoc mix of different
assertion mechanisms.
Or summarize your cover letter:
Use of assert(false) can trip spurious control flow warnings from some
versions of gcc:
https://lore.kernel.org/qemu-devel/54bb02a6-1b12-460a-97f6-3f478ef766c6@xxxxxxxxxx/
Solve that by unifying the code base on g_assert_not_reached()
instead.
If using g_assert_not_reached() instead of assert(false) silences
the warning about missing return value in such impossible to reach
locations should we also be deleting the now-unnecessary "return"
statements after g_assert_not_reached()?
Although it's unlikely to be used on any compiler that can also
compile qemu, there is a third implementation of g_assert_not_reached
that does nothing, see:
https://gitlab.gnome.org/GNOME/glib/-/blob/927683ebd94eb66c0d7868b77863f57ce9c5bc76/glib/gtestutils.h#L269
That's only in the #ifdef G_DISABLE_ASSERT case ... and we forbid that in
QEMU, see osdep.h:
#ifdef G_DISABLE_ASSERT
#error building with G_DISABLE_ASSERT is not supported
#endif
So in QEMU, g_assert_not_reached() should always abort.
Thomas
