-------------------------------------------- On Sat, 9/30/17, Laurent Vivier <[email protected]> wrote: > Could you change the NR_read too, for consistency? The NR_read case for length 0 doesn't check if the buffer is NULL, unlike this patch, and I'm not sure if switching that to make a real read() syscall with a NULL buffer even if the original buffer is non-null would break anything.
Thanks Zhuowei
