From mboxrd@z Thu Jan 1 00:00:00 1970 From: Marek Szyprowski Subject: Re: [Linux-c6x-dev] [PATCH 3/9] c6x: Provide dma_mmap_coherent() and dma_get_sgtable() Date: Thu, 24 Jan 2013 11:49:52 +0100 Message-ID: <510111D0.4000501@samsung.com> References: <1358073890-3610-1-git-send-email-geert@linux-m68k.org> <1358073890-3610-3-git-send-email-geert@linux-m68k.org> <1358177872.4357.53.camel@t520.localdomain> <50F4D83A.7020803@synopsys.com> <50F56286.8070200@samsung.com> <1358269008.10591.11.camel@dabdike.int.hansenpartnership.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from mailout4.w1.samsung.com ([210.118.77.14]:55019 "EHLO mailout4.w1.samsung.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752782Ab3AXKt4 (ORCPT ); Thu, 24 Jan 2013 05:49:56 -0500 In-reply-to: Sender: linux-arch-owner@vger.kernel.org List-ID: To: Geert Uytterhoeven Cc: James Bottomley , Mark Salter , Vineet Gupta , linux-arch@vger.kernel.org, linux-c6x-dev@linux-c6x.org, linux-kernel@vger.kernel.org On 1/21/2013 9:00 PM, Geert Uytterhoeven wrote: > On Tue, Jan 15, 2013 at 5:56 PM, James Bottomley > wrote: > > On Tue, 2013-01-15 at 15:07 +0100, Marek Szyprowski wrote: > >> On 1/15/2013 10:13 AM, Geert Uytterhoeven wrote: > >> > Marek? > >> > > >> > On Tue, Jan 15, 2013 at 5:16 AM, Vineet Gupta > >> > wrote: > >> > > On Monday 14 January 2013 09:07 PM, Mark Salter wrote: > >> > >> On Sun, 2013-01-13 at 11:44 +0100, Geert Uytterhoeven wrote: > >> > >>> c6x/allmodconfig (assumed): > >> > >>> > >> > >>> drivers/media/v4l2-core/videobuf2-dma-contig.c: In function = =E2=80=98vb2_dc_mmap=E2=80=99: > >> > >>> drivers/media/v4l2-core/videobuf2-dma-contig.c:204: error: i= mplicit declaration of function =E2=80=98dma_mmap_coherent=E2=80=99 > >> > >>> drivers/media/v4l2-core/videobuf2-dma-contig.c: In function = =E2=80=98vb2_dc_get_base_sgt=E2=80=99: > >> > >>> drivers/media/v4l2-core/videobuf2-dma-contig.c:387: error: i= mplicit declaration of function =E2=80=98dma_get_sgtable=E2=80=99 > >> > >>> > >> > >>> For architectures using dma_map_ops, dma_mmap_coherent() and > >> > >>> dma_get_sgtable() are provided in . > >> > >>> > >> > >>> C6x does not use dma_map_ops, hence it should implement them= as inline > >> > >>> stubs using dma_common_mmap() and dma_common_get_sgtable(). > >> > >>> > >> > >> So are dma_mmap_coherent() and dma_get_sgtable() part of the = DMA API > >> > >> now? I don't them in Documentation/DMA*.txt anywhere. > >> > >> > >> > >> Why does the default dma_common_mmap() for !CONFIG_MMU return= an > >> > >> error? > >> > >> > >> > >> Wouldn't it be better to provide default implementations that= an arch > >> > >> could override rather than having to patch all "no dma_map_op= s" > >> > >> architectures? > >> > >> > >> > > Speaking for the still-reviewed ARC Port, I completely agree w= ith Mark. > >> > >> dma_mmap_coherent() was partially in the DMA mapping API for some = time, but > >> it was available only on a few architectures (afair ARM, powerpc a= nd avr32). > >> This caused significant problems for writing unified device driver= s or some > >> device helper modules which were aimed to work on more than one > >> architecture. > >> > >> dma_get_sgtable() is an extension discussed during the Linaro meet= ings. It > >> is required to correctly implement buffer sharing between device d= river > >> without hacks or any assumptions about memory layout in the device= drivers. > >> > >> I have implemented some generic code for both of those two functio= ns, > >> keeping > >> in mind that on some hardware architectures (like already mentione= d VIVT) > >> it might be not possible to provide coherent mapping to userspace.= It is > >> perfectly fine for those functions to return an error in such case= =2E > > > > It's not possible on VIPT either. This means that the API is unusa= ble > > on quite a large number of architectures. Surely, if we're startin= g to > > write drivers using this, we need to fix the API before more people= try > > to use it. > > > > For PA-RISC (and all other VIPT, I assume) I need an API which allo= ws me > > to remap the virtual address of the kernel component (probably usin= g the > > kmap area) so the user space and kernel space addresses are congrue= nt. > > So what are we gonna do for 3.8? I'd like to get my allmodconfig buil= d > green again ;-) > > Change the API? > > Keep the API but do a best-effort fix to unbreak the builds? > - Apply my patches that got acks (avr32/blackfin/cris/m68k), > - Use static inlines that return -EINVAL for the rest > (c6x/frv/mn10300/parisc/xtensa). > I still have a few m68k fixes queued for 3.8, for which I've been pos= tponing the > pull request to get this sorted out, so I could include the above. I think this would be the best solution for fixing v3.8. I would like t= o put those patches in linux-next asap and then after a week or so send a pul= l request for merging them to v3.8-rc6. EINVAL approach can be later fixed/extended with real implementation (i= f there are any driver candidates, which will use such interface). Best regards --=20 Marek Szyprowski Samsung Poland R&D Center