From mboxrd@z Thu Jan 1 00:00:00 1970 From: swarren@wwwdotorg.org (Stephen Warren) Date: Tue, 12 Mar 2013 12:23:32 -0600 Subject: [PATCH v4 0/2] Uncompress debug for multiplatform In-Reply-To: <20130312040242.GF25910@S2101-09.ap.freescale.net> References: <1363001737-8745-1-git-send-email-shawn.guo@linaro.org> <20130311170350.GM4977@n2100.arm.linux.org.uk> <513E149A.2060704@wwwdotorg.org> <20130312040242.GF25910@S2101-09.ap.freescale.net> Message-ID: <513F72A4.8020309@wwwdotorg.org> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On 03/11/2013 10:02 PM, Shawn Guo wrote: > On Mon, Mar 11, 2013 at 11:30:02AM -0600, Stephen Warren wrote: >> On 03/11/2013 11:03 AM, Russell King - ARM Linux wrote: >>> On Mon, Mar 11, 2013 at 07:35:35PM +0800, Shawn Guo wrote: >>>> Instead of giving zero support of uncompress debug for multiplatform >>>> build, the series turns uncompress debug into one part of DEBUG_LL >>>> support. When DEBUG_LL is turned on for a particular platform, >>>> uncompress debug works too for that platform. >>>> >>>> OMAP is an exception here. OMAP low-level debug code places data >>>> in the .data section, and that is allowed in decompressor. That's >>>> why Kconfig symbol DEBUG_UNCOMPRESS that controls multiplatform >>>> uncompress debug support is defined with !DEBUG_OMAP2PLUS_UART. >>>> >>>> Changes since v3: >>>> - Create Kconfig symbol DEBUG_UNCOMPRESS for building in >>>> multiplatform uncompress debug support, so that we can leave >>>> out the platform which is part of multiplatform but not ready >>>> for being part of uncompress debug like OMAP example. >>> >>> Wasn't Tegra also having problems with this as well? >> >> Tegra isn't converted to multi-platform yet. I hope to in 3.10. > > Yeah, that's why I did not touch Tegra as the series is based on > v3.9-rc1 where Tegra hasn't been part of multi-platform yet. > > But I just noticed that debug-macro.S for Tegra has been renamed to > debug/tegra.S, and we already know it will be broken when multi-platform > is enabled together with this series for Tegra. Was that because the code touches .data? Perhaps I can get away with using some hard-coded IRAM location or something.