From: Greg KH Subject: Re: [PATCH -stable] drm/fb_helper: Allow leaking fbdev smem_start Date: Tue, 15 Jan 2019 09:27:16 +0100 Message-ID: <20190115082716.GB29858@kroah.com> References: <20190111150638.21619-1-maxime.ripard@bootlin.com> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Return-path: Content-Disposition: inline In-Reply-To: <20190111150638.21619-1-maxime.ripard@bootlin.com> Sender: stable-owner@vger.kernel.org To: Maxime Ripard Cc: stable@vger.kernel.org, dri-devel@lists.freedesktop.org, Neil Armstrong , Dave Airlie , Daniel Vetter , Bartlomiej Zolnierkiewicz , Noralf =?iso-8859-1?Q?Tr=F8nnes?= , Eric Anholt , Lucas Stach , Rob Clark , Ben Skeggs , Christian =?iso-8859-1?Q?K=F6nig?= List-Id: dri-devel@lists.freedesktop.org On Fri, Jan 11, 2019 at 04:06:38PM +0100, Maxime Ripard wrote: > From: Neil Armstrong > > commit 4be9bd10e22dfc7fc101c5cf5969ef2d3a042d8a upstream. > > Since "drm/fb: Stop leaking physical address", the default behaviour of > the DRM fbdev emulation is to set the smem_base to 0 and pass the new > FBINFO_HIDE_SMEM_START flag. > > The main reason is to avoid leaking physical addresse to user-space, and > it follows a general move over the kernel code to avoid user-space to > manipulate physical addresses and then use some other mechanisms like > dma-buf to transfer physical buffer handles over multiple subsystems. > > But, a lot of devices depends on closed sources binaries to enable > OpenGL hardware acceleration that uses this smem_start value to > pass physical addresses to out-of-tree modules in order to render > into these physical adresses. These should use dma-buf buffers allocated > from the DRM display device instead and stop relying on fbdev overallocation > to gather DMA memory (some HW vendors delivers GBM and Wayland capable > binaries, but older unsupported devices won't have these new binaries > and are doomed until an Open Source solution like Lima finalizes). > > Since these devices heavily depends on this kind of software and because > the smem_start population was available for years, it's a breakage to > stop leaking smem_start without any alternative solutions. > > This patch adds a Kconfig depending on the EXPERT config and an unsafe > kernel module parameter tainting the kernel when enabled. > > A clear comment and Kconfig help text was added to clarify why and when > this patch should be reverted, but in the meantime it's a necessary > feature to keep. > > Cc: Dave Airlie > Cc: Daniel Vetter > Cc: Bartlomiej Zolnierkiewicz > Cc: Noralf Trřnnes > Cc: Maxime Ripard > Cc: Eric Anholt > Cc: Lucas Stach > Cc: Rob Clark > Cc: Ben Skeggs > Cc: Christian König > Signed-off-by: Neil Armstrong > Reviewed-by: Maxime Ripard > Tested-by: Maxime Ripard > Acked-by: Daniel Vetter > Acked-by: Dave Airlie > Link: https://patchwork.freedesktop.org/patch/msgid/1538136355-15383-1-git-send-email-narmstrong@baylibre.com > Signed-off-by: Maxime Ripard > > --- > > Hi, > > This is a backport of a patch fixing a regression introduced in 4.19, and > merged in 4.20. Therefore, it targets 4.19 only. Now queued up, thanks. greg k-h