https://bugzilla.kernel.org/show_bug.cgi?id=207959 Linus Torvalds (torvalds@xxxxxxxxxxxxxxxxxxxx) changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |torvalds@linux-foundation.o | |rg --- Comment #5 from Linus Torvalds (torvalds@xxxxxxxxxxxxxxxxxxxx) --- That said, I'm not sure the kernel cares. If sparse makes '{ 0 }' be eqivalent to '{ }' and doesn't warn for it, it's not like it's a huge deal. The problem with using 0 instead of NULL (or vice versa, which is a crime, and which is why NULL should never have been defined to plain 0) comes when it is actually confusing. For something like a silly zero struct initializer it's not like it's the end of the world. I do find the whole "0 could be a pointer, and NULL could be used for an integer or float" to be very distasteful, and the C++ people in particular were in denial about their brokenness for much much too long. So I'd prefer the "0 for NULL" warning, even if this may not be the most important case for it. -- You are receiving this mail because: You are watching the assignee of the bug.
