Thread: Fix overflow hazard in interval rounding

Joseph Koshakow <koshy44@gmail.com> writes:
> Attached is a patch that fixes some overflow/underflow hazards that I
> discovered in the interval rounding code.

I think you need to use ereturn not ereport here; see other error
cases in AdjustIntervalForTypmod.

(We'd need ereport in back branches, but this problem seems to
me to probably not be worth back-patching.)

            regards, tom lane

Hi,

On 2024-02-13 13:31:22 -0500, Joseph Koshakow wrote:
> Attached is a patch that fixes some overflow/underflow hazards that I
> discovered in the interval rounding code.

Random, mildly related thought: I wonder if it's time to, again, look at
enabling -ftrapv in assert enabled builds.  I had looked at that a few years
back, and fixed a number of instances, but not all I think. But I think we are
a lot closer to avoiding signed overflows everywhere, and it'd be nice to find
overflow hazards more easily. Many places are broken even with -fwrapv
semantics (which we don't have on all compilers!). Trapping on such overflows
makes it far easier to find problems with tools like sqlsmith.

Greetings,

Andres Freund

Joseph Koshakow <koshy44@gmail.com> writes:
> On Tue, Feb 13, 2024 at 1:46 PM Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> (We'd need ereport in back branches, but this problem seems to
>> me to probably not be worth back-patching.)

> Agreed, this seems like a pretty rare overflow/underflow.

OK, pushed to HEAD only.  I converted the second steps to be like
"a -= a%b" instead of "a = (a/b)*b" to make it a little clearer
that they don't have their own risks of overflow.  Maybe it's a
shade faster that way too, not sure.

            regards, tom lane