From nobody Mon Jul 11 20:04:38 2022 X-Original-To: freebsd-current@mlmmj.nyi.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2610:1c1:1:606c::19:1]) by mlmmj.nyi.freebsd.org (Postfix) with ESMTP id 2BB0B17FA2F4 for ; Mon, 11 Jul 2022 20:04:44 +0000 (UTC) (envelope-from evgeniy@khramtsov.org) Received: from mxa.khramtsov.org (mxa.khramtsov.org [IPv6:2a0a:e5c0:2:10f::f]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 4LhZbv2PPnz3mGV for ; Mon, 11 Jul 2022 20:04:43 +0000 (UTC) (envelope-from evgeniy@khramtsov.org) Received: from mxa.khramtsov.org (mxa.khramtsov.org [IPv6:2a0a:e5c0:2:10f::f]) by mxa.khramtsov.org (Postfix) with ESMTP id 23745125EB7 for ; Mon, 11 Jul 2022 19:57:25 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=khramtsov.org; s=rsa; t=1657569445; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:mime-version:mime-version:content-type:content-type: in-reply-to:in-reply-to:references:references; bh=lCQ7wPSV3XT9BeMG9SQS0rxBOwnqdJx7CVBiUQmhRMc=; b=nfbJVJ/stiYltXT8N7XOn8JeqGXFMZKtxPOx9PJ+aO3gP5MdIbHzAIY+fnCMTtHPWfIdHK ywfztlXbWol6Hp4ZDWe+azOakZoXUUz787heKWo2zAHuo1pbeNKyKx1tZTZRjDFxOqVm6t C78gvel64kaB3uFUDz6q5GpCKpErZ1rsnCKglKWo9vthD+NYAY40DVMBZ0MnTR9ckGbzBt 54iUOq2uXcIQ3UCUS1guAEigAh3KDlmF2o3G1CVx5Cm7ZGZ5h3KP/qhKl6cH7scIOaMwWb P/iTBLiJjIQiJUeD4fRxa61BcWmVIcAFxlmh4ReN1DKpe2VgYkC5lZ4xiaZDhQ== Date: Mon, 11 Jul 2022 20:04:38 +0000 From: Evgeniy Khramtsov To: FreeBSD -CURRENT Subject: Re: BLAKE3 unstability? Message-ID: <20220711200438.qsmpfcs2bi23ng7z@vax.khramtsov.org> References: <20220709162640.7my2bq6rax5npdhf@vax.khramtsov.org> <20220709175605.ofkoft2mglrkaqpf@vax.khramtsov.org> <20220711200212.h74fo2bmkb22olkj@vax.khramtsov.org> List-Id: Discussions about the use of FreeBSD-current List-Archive: https://lists.freebsd.org/archives/freebsd-current List-Help: List-Post: List-Subscribe: List-Unsubscribe: Sender: owner-freebsd-current@freebsd.org MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20220711200212.h74fo2bmkb22olkj@vax.khramtsov.org> X-Rspamd-Queue-Id: 4LhZbv2PPnz3mGV X-Spamd-Bar: -- Authentication-Results: mx1.freebsd.org; dkim=pass header.d=khramtsov.org header.s=rsa header.b="nfbJVJ/s"; dmarc=pass (policy=reject) header.from=khramtsov.org; spf=pass (mx1.freebsd.org: domain of evgeniy@khramtsov.org designates 2a0a:e5c0:2:10f::f as permitted sender) smtp.mailfrom=evgeniy@khramtsov.org X-Spamd-Result: default: False [-2.80 / 15.00]; SUBJECT_ENDS_QUESTION(1.00)[]; NEURAL_HAM_MEDIUM(-1.00)[-1.000]; NEURAL_HAM_LONG(-1.00)[-1.000]; NEURAL_HAM_SHORT(-1.00)[-1.000]; DMARC_POLICY_ALLOW(-0.50)[khramtsov.org,reject]; R_DKIM_ALLOW(-0.20)[khramtsov.org:s=rsa]; R_SPF_ALLOW(-0.20)[+ip6:2a0a:e5c0:2:10f::f:c]; RCVD_NO_TLS_LAST(0.10)[]; MIME_GOOD(-0.10)[text/plain]; ONCE_RECEIVED(0.10)[]; FROM_EQ_ENVFROM(0.00)[]; MLMMJ_DEST(0.00)[FreeBSD-CURRENT]; RCVD_COUNT_ONE(0.00)[1]; MIME_TRACE(0.00)[0:+]; TO_DN_ALL(0.00)[]; ASN(0.00)[asn:207996, ipnet:2a0a:e5c0:2::/48, country:CH]; TO_MATCH_ENVRCPT_ALL(0.00)[]; FROM_HAS_DN(0.00)[]; DKIM_TRACE(0.00)[khramtsov.org:+]; PREVIOUSLY_DELIVERED(0.00)[freebsd-current@freebsd.org]; MID_RHS_MATCH_FROMTLD(0.00)[]; RCPT_COUNT_ONE(0.00)[1]; ARC_NA(0.00)[] X-ThisMailContainsUnwantedMimeParts: N (forgot to CC -CURRENT, so replying to the list) > It did not reproduce for me with in-tree ZFS on main@3c9ad9398fcd either. > > Could you share sysctl kstat.zfs.misc.chksum_bench, maybe we are using > different implementations? > I do see that blake3 went in with only a Linux module parameter for the > implementation selection, so I'll have to fix that. For now we can at least > see which was fastest, which should be the one selected. You just won't be > able to manually change it to see if that helps. > > - Ryan $ sysctl kstat.zfs.misc.chksum_bench kstat.zfs.misc.chksum_bench: implementation 1k 4k 16k 64k 256k 1m 4m edonr-generic 1358 1580 1642 1642 1621 1560 1525 skein-generic 238 252 256 256 256 256 254 sha256-generic 242 263 269 271 271 271 271 sha512-generic 373 416 428 430 431 431 431 blake3-generic 482 478 477 474 473 473 473 blake3-sse2 338 1403 1505 1526 1519 1504 1503 blake3-sse41 350 1602 1725 1758 1753 1747 1747 blake3-avx2 350 1874 3336 3550 3527 3492 3485 Could it be due to SIMD? I can try a patch. I can also try with GENERIC kernel tomorrow as it is the only local modification left. I thought it also could be damaged hardware, but AVX2 torture tests and memtest runs fine. Thanks for investigating this issue.