From mboxrd@z Thu Jan 1 00:00:00 1970 From: arnd@arndb.de (Arnd Bergmann) Date: Fri, 23 Sep 2011 17:45:06 +0200 Subject: DT vs ARM static mappings In-Reply-To: <1316708611.4611.873.camel@hornet.cambridge.arm.com> References: <1316519479.4611.150.camel@hornet.cambridge.arm.com> <201109202128.55517.arnd@arndb.de> <1316708611.4611.873.camel@hornet.cambridge.arm.com> Message-ID: <201109231745.06264.arnd@arndb.de> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Thursday 22 September 2011, Pawel Moll wrote: > Executive summary: > > * I have second map_desc with pfn for RS1 memory map, but using the same > virtual address as the legacy one. The legacy one is used if root of the > tree is compatible with "arm,vexpress-legacy". > > * The devices I need to use in v2m.c have aliases in DTS so I can find > their offsets in the flat tree (the *_find_node_by_alias() function is > rather generic and could be moved to drivers/of/fdt.c if you think it > would be useful for others). > > * There are no more users of MMIO_P2V in v2m.c, next thing I will do is > the same in core tile; then the macro can be killed. Once this happens > the virtual address currently taken from __MMIO_P2V(V2M_PA_CS7) will be > replaced by some kind of "#define V2M_PERIPH_BASE 0xf8000000". > > * Once Nico's changes regarding static maps are in, the manual pointer > operations in v2m_dt_map_io can be replaced with neat ioremap()-s. > > All feedback appreciated, cheers! Look ok to me. Just a comment on the submission: Even when you post something for review instead of inclusion, please add a Signed-off-by: line and a diffstat. Arnd From mboxrd@z Thu Jan 1 00:00:00 1970 From: Arnd Bergmann Subject: Re: DT vs ARM static mappings Date: Fri, 23 Sep 2011 17:45:06 +0200 Message-ID: <201109231745.06264.arnd@arndb.de> References: <1316519479.4611.150.camel@hornet.cambridge.arm.com> <201109202128.55517.arnd@arndb.de> <1316708611.4611.873.camel@hornet.cambridge.arm.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1316708611.4611.873.camel-okZbbLrgpR/YkXV2EHHjLW3o5bpOHsLO@public.gmane.org> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: devicetree-discuss-bounces+gldd-devicetree-discuss=m.gmane.org-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org Sender: devicetree-discuss-bounces+gldd-devicetree-discuss=m.gmane.org-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org To: Pawel Moll Cc: devicetree-discuss-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org, linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org List-Id: devicetree@vger.kernel.org On Thursday 22 September 2011, Pawel Moll wrote: > Executive summary: > > * I have second map_desc with pfn for RS1 memory map, but using the same > virtual address as the legacy one. The legacy one is used if root of the > tree is compatible with "arm,vexpress-legacy". > > * The devices I need to use in v2m.c have aliases in DTS so I can find > their offsets in the flat tree (the *_find_node_by_alias() function is > rather generic and could be moved to drivers/of/fdt.c if you think it > would be useful for others). > > * There are no more users of MMIO_P2V in v2m.c, next thing I will do is > the same in core tile; then the macro can be killed. Once this happens > the virtual address currently taken from __MMIO_P2V(V2M_PA_CS7) will be > replaced by some kind of "#define V2M_PERIPH_BASE 0xf8000000". > > * Once Nico's changes regarding static maps are in, the manual pointer > operations in v2m_dt_map_io can be replaced with neat ioremap()-s. > > All feedback appreciated, cheers! Look ok to me. Just a comment on the submission: Even when you post something for review instead of inclusion, please add a Signed-off-by: line and a diffstat. Arnd