Re: [PATCH v2 06/13] reftable: (de)serialization for the polymorphic record type.

On 01/10/2020 20:23, Junio C Hamano wrote:

"Han-Wen Nienhuys via GitGitGadget" [off-list ref] writes:

quoted

+struct reftable_record reftable_new_record(uint8_t typ)
+{
+	struct reftable_record rec = { NULL };

I can see a lot of "sparse" inspired work went into this, but we
would want to take the previous discussion into account, as nothing
has changed since then.

https://lore.kernel.org/git/41a60e60-d2c0-7d54-5456-e44d106548a4@kdbg.org/ (local)

https://lore.kernel.org/git/1df91aa4-dda5-64da-6ae3-5d65e50a55c5@ramsayjones.plus.com/ (local)

yep, sparse is equally happy with = { 0 }; here - at least any sparse
version after 'v0.6.1-246-g41f651b4'. ;-) So, as I said yesterday, the
last official release v0.6.2 should be fine, and I _think_ Debian Testing
has a suitable version. (or you could build from source).

Unfortunately we didn't reach a clear consensus back then; I thought
we saw a fix to make sparse silent when we use the "= { 0 }" idiom
in recent versions of sparse, but the above threads do not have any
mention of it, so either I am misremembering or there were other
discussions on the same topic where it was also mentioned.

See commit 1c96642326 (sparse: allow '{ 0 }' to be used without warnings,
22-05-2020). We could actually remove that now, since '-Wno-universal-initializer'
is now the default (but it doesn't hurt).

ATB,
Ramsay Jones