This is the mail archive of the glibc-bugs@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[Bug libc/19920] RFE: test whether a block of memory is all 0


https://sourceware.org/bugzilla/show_bug.cgi?id=19920

Peter Maydell <peter.maydell at linaro dot org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |peter.maydell at linaro dot org

--- Comment #1 from Peter Maydell <peter.maydell at linaro dot org> ---
(In reply to Dr. David Alan Gilbert from comment #0)
>   The current qemu implementation is:
>     size_t buffer_find_nonzero_offset(const void *buf, size_t len)
> 
>     that returns an index to the first !0 byte;

This isn't quite right -- it returns an index which is definitely
earlier than the first nonzero byte, and this is actually where
it gets a lot of the speedup from: you test the buffer 64 bytes
at a time (say), and your loop body only needs to check "is any of
these 64 bytes nonzero?" rather than "find the first", so it
can just do a SIMD logical OR of everything together (and then
if something came out nonzero you return the index of the start
of this 64-byte chunk).

> however all of our current
> uses are just for testing whether the whole buffer is zero.

A straightforward boolean return is probably a cleaner API in
any case, though.

-- 
You are receiving this mail because:
You are on the CC list for the bug.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]