From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ivan Shapovalov Subject: Re: Kernel config option which causes reiser4 to be instable Date: Tue, 11 Dec 2012 22:49:47 +0400 Message-ID: <3128977.locbVvMWgS@intelfx-laptop> References: <2612159.zxK0OqJdpP@intelfx-laptop> <50C77C83.4080402@gmail.com> Mime-Version: 1.0 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=from:to:cc:subject:date:message-id:user-agent:in-reply-to :references:mime-version:content-transfer-encoding:content-type; bh=OTOVKpgSI6o/AmJ4h0LGWHISVvuSQrsFF8tBEDS7vIM=; b=BEkwn3p79+a5tJGwTcJAtD9FpxpEqoQRHjFrdQZ3lDpNB69/Jr7Bjx4n7oYdb5mYdY HoOcSEQLLwdzqdYM3R4DpGQL9wdkqgy85yFOcsHKOMcyKerdusa9XPkbOIr/5Jk1BZCp eMuVwsHgHCk+hrI5MjcNs50KicpwQrio6SbHA9f8aYcFId5IiOYHjrGuKzcPSjAkg7Pv BiOhGeBTVDlh3BsOTHcIKnURk8ViDxLmzV/y/09JKbu8AazUt/cWcqS4E37z/f9OOx7s imZm+cp3zl96eniN9vY1DbJBb26cjpeCSQdgwHEwRYmG/WXLrAQxzJHp/23LdPZvmrDc Qbcg== In-Reply-To: <50C77C83.4080402@gmail.com> Sender: reiserfs-devel-owner@vger.kernel.org List-ID: Content-Type: text/plain; charset="utf-8" To: Edward Shishkin Cc: reiserfs-devel , =?utf-8?B?RHXFoWFuIMSMb2xpxIc=?= On 11 December 2012 19:33:39 Edward Shishkin wrote: > On 12/11/2012 04:08 PM, Ivan Shapovalov wrote: > > Hello! >=20 > Hello. >=20 > > With help of Du=C5=A1an =C4=8Coli=C4=87 who prov= ided his kernel config > > diff I've found a kernel option which, when disabled, greatly reduc= es > > (hopefully to zero, but need time to verify it) corruption rate in > > reiser4. > >=20 > > It's CONFIG_TRANSPARENT_HUGEPAGE (or something which is used by it = like > > CONFIG_COMPACTION or CONFIG_MIGRATION). > > For now I'm testing it with CONFIG_TRANSPARENT_HUGEPAGE disabled >=20 > How long? 12 hours of indexing, scanning, compiling, repeated execution of "find -type f -exec grep wtf {} \;" and so on. >=20 > > on kernel > >=20 > > 3.6.10, and everything seems to be OK so far (so the workaround is > > version- > > agnostic). > >=20 > > Edward, are there any guesses on what can make reiser4 choke on > > hugepages/compaction/migration? >=20 > TBH, no ideas. They (hugepages) are _transparent_. > It means we shouldn't suffer in theory ;) Maybe it's actually migration who does the damage? If we don't lock the= pages=20 properly and they are "stolen" by the migration code... If this is the = case, I=20 shall eventually get corruptions with current setup (since=20 migration/compaction is not disabled). If I get them, I'll rebuild without migration at all and will see if=20 corruptions disappear completely. (Then they should disappear, if the=20 prediction is true.) >=20 > > I'm not even barely familiar with the kernel > >=20 > > internals. > >=20 > > Thanks, > > Ivan. -- To unsubscribe from this list: send the line "unsubscribe reiserfs-deve= l" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html