Hi, On Tue, Feb 24, 2026 at 11:19:50AM -0600, Nathan Bossart wrote: > On Tue, Feb 24, 2026 at 05:08:09PM +0000, Bertrand Drouvot wrote: > > This patch makes use of unvolatize() in vac_truncate_clog(). > > > > Note that it does not remove the warning but moves it to c.h (where > > unvolatize() > > is defined) but that's consistent with what 481018f2804 did too. > > Why is this preferable to marking the function parameter as volatile
I think that that would sound misleading for the other callers that don't really need the volatile qualification. > or removing the volatile qualifier from the variable? That looks mandatory according to 2d2e40e3bef. Regards, -- Bertrand Drouvot PostgreSQL Contributors Team RDS Open Source Databases Amazon Web Services: https://aws.amazon.com
