Re: [PATCH] fix infinite loop eating all the memory

[Paul Eggert]

Colin Leroy <address@hidden> writes:

> In this patch however, there was no silent truncation but an
> explicit error.

Coreutils tail should also report an explicit error, when you run out
of main memory.  (You evidently didn't wait long enough.  :-)

Running out of memory is not an arbitrary limit, so it's OK for tail
to report an error and exit in that case.  But a 1 MiB limit is
arbitrary (and rather small these days), and it's not OK for tail to
impose such a limit.