From mboxrd@z Thu Jan 1 00:00:00 1970 From: Loic Dachary Subject: Re: gf_unit 16 A -1 -m SPLIT 16 4 -r ALTMAP - failure Date: Sat, 10 Jan 2015 18:58:23 +0100 Message-ID: <54B1683F.7080004@dachary.org> References: <54B16763.7070602@dachary.org> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="5dRl25Ccw4qU5pj170g1Vx80cde2jmixb" Return-path: Received: from mail2.dachary.org ([91.121.57.175]:36519 "EHLO smtp.dmail.dachary.org" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1751201AbbAJR60 (ORCPT ); Sat, 10 Jan 2015 12:58:26 -0500 In-Reply-To: <54B16763.7070602@dachary.org> Sender: ceph-devel-owner@vger.kernel.org List-ID: To: Kevin Greenan , Janne Grunau Cc: Ceph Development This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --5dRl25Ccw4qU5pj170g1Vx80cde2jmixb Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable I should add that this failure only happens on ARMv7 Ubuntu with gcc vers= ion 4.8.1 (Ubuntu/Linaro 4.8.1-10ubuntu7) On 10/01/2015 18:54, Loic Dachary wrote: > Hi Kevin & Janne, >=20 > The test gf_unit 16 A -1 -m SPLIT 16 4 -r ALTMAP - fails on the current= gf-complete master. The first commit where it fails is >=20 > commit 474010a91d35fef5ca7dea77205b6a5c7e68c3e9 > Author: Janne Grunau > Date: Wed Sep 17 16:10:25 2014 +0200 >=20 > arm: NEON optimisations for gf_w16 >=20 > Optimisations for the 4,16 split table region multiplications. >=20 > Selected time_tool.sh 16 -A -B results for a 1.7 GHz cortex-a9: > Region Best (MB/s): 532.14 W-Method: 16 -m SPLIT 16 4 -r SIMD -= > Region Best (MB/s): 212.34 W-Method: 16 -m SPLIT 16 4 -r NOSIMD= - > Region Best (MB/s): 801.36 W-Method: 16 -m SPLIT 16 4 -r SIMD -= r ALTMAP - > Region Best (MB/s): 93.20 W-Method: 16 -m SPLIT 16 4 -r NOSIMD= -r ALTMAP - > Region Best (MB/s): 273.99 W-Method: 16 -m SPLIT 16 8 - > Region Best (MB/s): 270.81 W-Method: 16 -m SPLIT 8 8 - > Region Best (MB/s): 70.42 W-Method: 16 -m COMPOSITE 2 - - > Region Best (MB/s): 393.54 W-Method: 16 -m COMPOSITE 2 - -r ALT= MAP - >=20 > but the test did exit(0) on error instead of exit(1) and we failed to n= otice. >=20 > gf_unit 16 A -1 -m SPLIT 16 4 -r ALTMAP - > Args: 16 A -1 -m SPLIT 16 4 -r ALTMAP - / size (bytes): 524428 > Problem with region multiply (all values in hex): > Target address base: 0x8fd08e. Word 0x1 of 0x1fee. Xor: 0 > Value: 2 > Original source word: d00a > Product word: a000 > It should be: b01f >=20 > Do you have an idea why this happens ? For the record here is the bisec= t command I used: >=20 > git bisect start 6fdd8bc3d32cb2f7fa55d2de9dc7cc5bb2f885aa 36e75c3efec08= b1e9bdb9c1f69a5b0018abd8ac7 > git bisect run try.sh >=20 > #!/bin/bash > log=3D$(git rev-parse HEAD) > echo $log.log > make distclean > $log.log 2>&1 > ./autogen.sh >> $log.log 2>&1 > ./configure >> $log.log 2>&1 > make -j4 >> $log.log 2>&1 > ! test/gf_unit 16 A -1 -m SPLIT 16 4 -r ALTMAP - 2>&1 | grep 'It should= be' >=20 > Note that b87c41f282dccc1b3649e3ea3fb80d19f820310 fails the test for di= fferent reasons: >=20 > Args: 16 A -1 -m SPLIT 16 4 -r ALTMAP - / size (bytes): 524428 > *** Error in `/home/ubuntu/f/gf-complete/test/.libs/lt-gf_unit': free()= : invalid pointer: 0x00ce7070 *** > try.sh: line 8: 12193 Aborted test/gf_unit 16 A -1 -m S= PLIT 16 4 -r ALTMAP - >=20 > Cheers >=20 --=20 Lo=C3=AFc Dachary, Artisan Logiciel Libre --5dRl25Ccw4qU5pj170g1Vx80cde2jmixb Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iEYEARECAAYFAlSxaD8ACgkQ8dLMyEl6F230JgCZAT2CON5k3NLz00qUWrV//gyq ms4An2KGY7wG5jEx4ZXzm5EwVuch783B =7w0r -----END PGP SIGNATURE----- --5dRl25Ccw4qU5pj170g1Vx80cde2jmixb--