From: Jonathan Austin Subject: Re: [PATCHv4] arm: Preserve the user r/w register TPIDRURW on context switch and fork Date: Mon, 20 May 2013 12:03:41 +0100 Message-ID: <519A030D.1030809@arm.com> References: <518AA19B.2020209@dawncrow.de> <51979801.7010903@dawncrow.de> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: In-Reply-To: <51979801.7010903@dawncrow.de> Sender: linux-kernel-owner@vger.kernel.org To: =?UTF-8?B?QW5kcsOpIEhlbnRzY2hlbA==?= Cc: "linux-arch@vger.kernel.org" , Russell King - ARM Linux , "linux-arm-kernel@lists.infradead.org" , "linux-kernel@vger.kernel.org" , "gregkh@linuxfoundation.org" , Will Deacon List-Id: linux-arch.vger.kernel.org Hi Andr=C3=A9 On 18/05/13 16:02, Andr=C3=A9 Hentschel wrote: > Am 08.05.2013 21:03, schrieb Andr=C3=A9 Hentschel: >> From: =3D?UTF-8?q?Andr=3DC3=3DA9=3D20Hentschel?=3D >> This is strangely formatted for me too, and I use a different client=20 from Will so I'm not sure that the problem is just at our end... (Also see that the list archive has weird formatting: http://lists.infradead.org/pipermail/linux-arm-kernel/2013-May/167325.h= tml ) Your first couple of patches didn't come out strangely formatted, so I'= m=20 not really sure what's going on. >> Since commit 6a1c53124aa1 the user writeable TLS register was zeroed= to >> prevent it from being used as a covert channel between two tasks. >> >> There are more and more applications coming to WinRT, Wine could sup= port them, >> but mostly they expect to have the thread environment block (TEB) in= TPIDRURW. >> >> This patch preserves that register per thread instead of clearing it= =2E >> Unlike the TPIDRURO, which is already switched, the TPIDRURW >> can be updated from userspace so needs careful treatment in the case= that we >> modify TPIDRURW and call fork(). To avoid this we must always read >> TPIDRURW in copy_thread. >> >> Signed-off-by: Andr=C3=A9 Hentschel >> Signed-off-by: Will Deacon >> Signed-off-by: Jonathan Austin >> > > Hi, > I'm not yet very familiar with the development process here, > am i getting no feedback on v4 because of the mergewindow being close= d? > Or is there another reason? Sry for being impatient. > This is a feature, not a fix, so most likely it'll be included at the=20 next merge window. For that to happen it should be in Russell's tree=20 around the middle of this cycle. Can you please rebase on 3.10-rc2 (when it happens) and post one more=20 version? After that, assuming nobody else has any final comments, you=20 could put it in to Russell's patch system... Just as a hint - one thing you might have done to increase the chances=20 of getting comments to clarify what's different between v3 and v4 - as = a=20 way to make life easier for reviewers you can highlight the differences= =20 between versions after the "---" (where you currently have the=20 description of why there are so many S-o-Bs). Hope that helps, Jonny