Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1514898

Re: [PATCH] poly1305: generic C can be faster on chips with slow unaligned access

Show all headers | View raw

Hi David,

On Thu, Nov 3, 2016 at 6:08 PM, David Miller <davem@davemloft.net> wrote:
> In any event no piece of code should be doing 32-bit word reads from
> addresses like "x + 3" without, at a very minimum, going through the
> kernel unaligned access handlers.

Excellent point. In otherwords,

    ctx->r[0] = (le32_to_cpuvp(key +  0) >> 0) & 0x3ffffff;
    ctx->r[1] = (le32_to_cpuvp(key +  3) >> 2) & 0x3ffff03;
    ctx->r[2] = (le32_to_cpuvp(key +  6) >> 4) & 0x3ffc0ff;
    ctx->r[3] = (le32_to_cpuvp(key +  9) >> 6) & 0x3f03fff;
    ctx->r[4] = (le32_to_cpuvp(key + 12) >> 8) & 0x00fffff;

should change to:

    ctx->r[0] = (le32_to_cpuvp(key +  0) >> 0) & 0x3ffffff;
    ctx->r[1] = (get_unaligned_le32(key +  3) >> 2) & 0x3ffff03;
    ctx->r[2] = (get_unaligned_le32(key +  6) >> 4) & 0x3ffc0ff;
    ctx->r[3] = (get_unaligned_le32(key +  9) >> 6) & 0x3f03fff;
    ctx->r[4] = (le32_to_cpuvp(key + 12) >> 8) & 0x00fffff;

> We know explicitly that these offsets will not be 32-bit aligned, so
> it is required that we use the helpers, or alternatively do things to
> avoid these unaligned accesses such as using temporary storage when
> the HAVE_EFFICIENT_UNALIGNED_ACCESS kconfig value is not set.

So the question is: is the clever avoidance of unaligned accesses of
the original patch faster or slower than changing the unaligned
accesses to use the helper function?

I've put a little test harness together for playing with this:

    $ git clone git://git.zx2c4.com/polybench
    $ cd polybench
    $ make run

To test with one method, do as normal. To test with the other, remove
"#define USE_FIRST_METHOD" from the source code.

@René: do you think you could retest on your MIPS32r2 hardware and
report back which is faster?

And if anybody else has other hardware and would like to try, this
could be nice.

Regards,
Jason

Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

[PATCH] poly1305: generic C can be faster on chips with slow unaligned access "Jason A. Donenfeld" <Jason@zx2c4.com> - 2016-11-02 19:00 +0100
  Re: [PATCH] poly1305: generic C can be faster on chips with slow  unaligned access Herbert Xu <herbert@gondor.apana.org.au> - 2016-11-02 21:20 +0100
    Re: [PATCH] poly1305: generic C can be faster on chips with slow  unaligned access Sandy Harris <sandyinchina@gmail.com> - 2016-11-02 21:50 +0100
    Re: [PATCH] poly1305: generic C can be faster on chips with slow  unaligned access "Jason A. Donenfeld" <Jason@zx2c4.com> - 2016-11-02 22:10 +0100
      Re: [PATCH] poly1305: generic C can be faster on chips with slow  unaligned access Herbert Xu <herbert@gondor.apana.org.au> - 2016-11-02 22:10 +0100
        Re: [PATCH] poly1305: generic C can be faster on chips with slow  unaligned access "Jason A. Donenfeld" <Jason@zx2c4.com> - 2016-11-02 22:30 +0100
          Re: [PATCH] poly1305: generic C can be faster on chips with slow  unaligned access Herbert Xu <herbert@gondor.apana.org.au> - 2016-11-02 22:30 +0100
            Re: [PATCH] poly1305: generic C can be faster on chips with slow  unaligned access "Jason A. Donenfeld" <Jason@zx2c4.com> - 2016-11-02 23:10 +0100
              Re: [PATCH] poly1305: generic C can be faster on chips with slow  unaligned access Herbert Xu <herbert@gondor.apana.org.au> - 2016-11-03 01:50 +0100
                Re: [PATCH] poly1305: generic C can be faster on chips with slow  unaligned access "Jason A. Donenfeld" <Jason@zx2c4.com> - 2016-11-03 08:30 +0100
                Re: [PATCH] poly1305: generic C can be faster on chips with slow  unaligned access David Miller <davem@davemloft.net> - 2016-11-03 18:10 +0100
                Re: [PATCH] poly1305: generic C can be faster on chips with slow  unaligned access "Jason A. Donenfeld" <Jason@zx2c4.com> - 2016-11-03 23:30 +0100
                Re: [PATCH] poly1305: generic C can be faster on chips with slow  unaligned access Eric Biggers <ebiggers@google.com> - 2016-11-04 18:40 +0100
                Re: [PATCH] poly1305: generic C can be faster on chips with slow  unaligned access "Jason A. Donenfeld" <Jason@zx2c4.com> - 2016-11-07 19:10 +0100
                Re: [PATCH] poly1305: generic C can be faster on chips with slow  unaligned access "Jason A. Donenfeld" <Jason@zx2c4.com> - 2016-11-07 19:30 +0100
                Re: [PATCH] poly1305: generic C can be faster on chips with slow  unaligned access Eric Biggers <ebiggers@google.com> - 2016-11-07 19:40 +0100
                Re: [PATCH] poly1305: generic C can be faster on chips with slow  unaligned access "Jason A. Donenfeld" <Jason@zx2c4.com> - 2016-11-07 20:10 +0100
                Re: [PATCH] poly1305: generic C can be faster on chips with slow  unaligned access Eric Biggers <ebiggers@google.com> - 2016-11-07 20:30 +0100
                Re: [PATCH] poly1305: generic C can be faster on chips with slow  unaligned access "Jason A. Donenfeld" <Jason@zx2c4.com> - 2016-11-07 20:50 +0100
  [PATCH v2] poly1305: generic C can be faster on chips with slow unaligned access "Jason A. Donenfeld" <Jason@zx2c4.com> - 2016-11-07 20:20 +0100
    [PATCH v3] poly1305: generic C can be faster on chips with slow unaligned access "Jason A. Donenfeld" <Jason@zx2c4.com> - 2016-11-07 20:50 +0100
    [PATCH v4] poly1305: generic C can be faster on chips with slow unaligned access "Jason A. Donenfeld" <Jason@zx2c4.com> - 2016-11-07 21:00 +0100
      Re: [PATCH v4] poly1305: generic C can be faster on chips with slow  unaligned access Eric Biggers <ebiggers@google.com> - 2016-11-07 21:50 +0100
      Re: [PATCH v4] poly1305: generic C can be faster on chips with slow  unaligned access Martin Willi <martin@strongswan.org> - 2016-11-08 09:10 +0100
        Re: [PATCH v4] poly1305: generic C can be faster on chips with slow  unaligned access Eric Biggers <ebiggers@google.com> - 2016-11-08 18:30 +0100

csiph-web