All of lore.kernel.org
 help / color / mirror / Atom feed
From: Christoph Hellwig <hch@lst.de>
To: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Christoph Hellwig <hch@lst.de>, Paul Mackerras <paulus@samba.org>,
	Michael Ellerman <mpe@ellerman.id.au>,
	Tony Luck <tony.luck@intel.com>,
	Fenghua Yu <fenghua.yu@intel.com>,
	linuxppc-dev@lists.ozlabs.org, iommu@lists.linux-foundation.org,
	linux-ia64@vger.kernel.org, Robin Murphy <robin.murphy@arm.com>,
	Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Subject: Re: [PATCH 08/20] powerpc/dma: remove the unused dma_nommu_ops export
Date: Wed, 22 Aug 2018 06:45:31 +0000	[thread overview]
Message-ID: <20180822064531.GA19284@lst.de> (raw)
In-Reply-To: <f7daffa5481ab82f345aa5bf9f63f40caa068456.camel@kernel.crashing.org>

On Thu, Aug 09, 2018 at 10:01:16AM +1000, Benjamin Herrenschmidt wrote:
> On Tue, 2018-07-31 at 14:16 +0200, Christoph Hellwig wrote:
> > It turns out cxl actually uses it.  So for now skip this patch,
> > although random code in drivers messing with dma ops will need to
> > be sorted out sooner or later.
> 
> CXL devices are "special", they bypass the classic iommu in favor of
> allowing the device to operate using the main processor page tables
> using an MMU context (so basically the device can use userspace
> addresses directly), akin to ATS.
> 
> I think the code currently uses the nommu ops as a way to do a simple
> kernel mapping for kernel drivers using CXL (not userspace stuff)
> though.

Its still a horrible idea to have this in drivers/, we need some
core API to mediate this behavior.  Also if the device supports
using virtual addresses dma_nommu_ops seems wrong as it won't do
the right thing for e.g. vmalloc addresses not mapped into the
kernel linear mapping (which I guess can't currently happen on
powerpc, but still..)

WARNING: multiple messages have this Message-ID (diff)
From: Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
To: Benjamin Herrenschmidt
	<benh-XVmvHMARGAS8U2dJNN8I7kB+6BGkLq7r@public.gmane.org>
Cc: Fenghua Yu <fenghua.yu-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>,
	Tony Luck <tony.luck-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>,
	linux-ia64-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
	Konrad Rzeszutek Wilk
	<konrad.wilk-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org>,
	Michael Ellerman <mpe-Gsx/Oe8HsFggBc27wqDAHg@public.gmane.org>,
	Robin Murphy <robin.murphy-5wv7dgnIgG8@public.gmane.org>,
	iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org,
	Paul Mackerras <paulus-eUNUBHrolfbYtjvyW6yDsg@public.gmane.org>,
	linuxppc-dev-uLR06cmDAlY/bJ5BZ2RsiQ@public.gmane.org,
	Christoph Hellwig <hch-jcswGhMUV9g@public.gmane.org>
Subject: Re: [PATCH 08/20] powerpc/dma: remove the unused dma_nommu_ops export
Date: Wed, 22 Aug 2018 08:45:31 +0200	[thread overview]
Message-ID: <20180822064531.GA19284@lst.de> (raw)
In-Reply-To: <f7daffa5481ab82f345aa5bf9f63f40caa068456.camel-XVmvHMARGAS8U2dJNN8I7kB+6BGkLq7r@public.gmane.org>

On Thu, Aug 09, 2018 at 10:01:16AM +1000, Benjamin Herrenschmidt wrote:
> On Tue, 2018-07-31 at 14:16 +0200, Christoph Hellwig wrote:
> > It turns out cxl actually uses it.  So for now skip this patch,
> > although random code in drivers messing with dma ops will need to
> > be sorted out sooner or later.
> 
> CXL devices are "special", they bypass the classic iommu in favor of
> allowing the device to operate using the main processor page tables
> using an MMU context (so basically the device can use userspace
> addresses directly), akin to ATS.
> 
> I think the code currently uses the nommu ops as a way to do a simple
> kernel mapping for kernel drivers using CXL (not userspace stuff)
> though.

Its still a horrible idea to have this in drivers/, we need some
core API to mediate this behavior.  Also if the device supports
using virtual addresses dma_nommu_ops seems wrong as it won't do
the right thing for e.g. vmalloc addresses not mapped into the
kernel linear mapping (which I guess can't currently happen on
powerpc, but still..)

WARNING: multiple messages have this Message-ID (diff)
From: Christoph Hellwig <hch@lst.de>
To: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Christoph Hellwig <hch@lst.de>, Paul Mackerras <paulus@samba.org>,
	Michael Ellerman <mpe@ellerman.id.au>,
	Tony Luck <tony.luck@intel.com>,
	Fenghua Yu <fenghua.yu@intel.com>,
	linuxppc-dev@lists.ozlabs.org, iommu@lists.linux-foundation.org,
	linux-ia64@vger.kernel.org, Robin Murphy <robin.murphy@arm.com>,
	Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Subject: Re: [PATCH 08/20] powerpc/dma: remove the unused dma_nommu_ops export
Date: Wed, 22 Aug 2018 08:45:31 +0200	[thread overview]
Message-ID: <20180822064531.GA19284@lst.de> (raw)
In-Reply-To: <f7daffa5481ab82f345aa5bf9f63f40caa068456.camel@kernel.crashing.org>

On Thu, Aug 09, 2018 at 10:01:16AM +1000, Benjamin Herrenschmidt wrote:
> On Tue, 2018-07-31 at 14:16 +0200, Christoph Hellwig wrote:
> > It turns out cxl actually uses it.  So for now skip this patch,
> > although random code in drivers messing with dma ops will need to
> > be sorted out sooner or later.
> 
> CXL devices are "special", they bypass the classic iommu in favor of
> allowing the device to operate using the main processor page tables
> using an MMU context (so basically the device can use userspace
> addresses directly), akin to ATS.
> 
> I think the code currently uses the nommu ops as a way to do a simple
> kernel mapping for kernel drivers using CXL (not userspace stuff)
> though.

Its still a horrible idea to have this in drivers/, we need some
core API to mediate this behavior.  Also if the device supports
using virtual addresses dma_nommu_ops seems wrong as it won't do
the right thing for e.g. vmalloc addresses not mapped into the
kernel linear mapping (which I guess can't currently happen on
powerpc, but still..)

  reply	other threads:[~2018-08-22  6:45 UTC|newest]

Thread overview: 156+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-07-30 16:38 use generic DMA mapping code in powerpc Christoph Hellwig
2018-07-30 16:38 ` Christoph Hellwig
2018-07-30 16:38 ` Christoph Hellwig
2018-07-30 16:38 ` [PATCH 01/20] kernel/dma/direct: take DMA offset into account in dma_direct_supported Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-08-08 23:44   ` Benjamin Herrenschmidt
2018-08-08 23:44     ` Benjamin Herrenschmidt
2018-08-08 23:44     ` Benjamin Herrenschmidt
2018-08-22  6:53     ` Christoph Hellwig
2018-08-22  6:53       ` Christoph Hellwig
2018-08-22  6:53       ` Christoph Hellwig
2018-08-22 23:59       ` Benjamin Herrenschmidt
2018-08-22 23:59         ` Benjamin Herrenschmidt
2018-08-22 23:59         ` Benjamin Herrenschmidt
2018-08-23  5:24         ` Christoph Hellwig
2018-08-23  5:24           ` Christoph Hellwig
2018-08-23  5:24           ` Christoph Hellwig
2018-08-23  5:24           ` Benjamin Herrenschmidt
2018-08-23  5:24             ` Benjamin Herrenschmidt
2018-08-23  5:24             ` Benjamin Herrenschmidt
2018-07-30 16:38 ` [PATCH 02/20] kernel/dma/direct: refine dma_direct_alloc zone selection Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-08-08 23:54   ` Benjamin Herrenschmidt
2018-08-08 23:54     ` Benjamin Herrenschmidt
2018-08-08 23:54     ` Benjamin Herrenschmidt
2018-08-22  6:58     ` Christoph Hellwig
2018-08-22  6:58       ` Christoph Hellwig
2018-08-22  6:58       ` Christoph Hellwig
2018-08-23  0:01       ` Benjamin Herrenschmidt
2018-08-23  0:01         ` Benjamin Herrenschmidt
2018-08-23  0:01         ` Benjamin Herrenschmidt
2018-08-23  5:26         ` Christoph Hellwig
2018-08-23  5:26           ` Christoph Hellwig
2018-08-23  5:26           ` Christoph Hellwig
2018-07-30 16:38 ` [PATCH 03/20] dma-mapping: make the get_required_mask method available unconditionally Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-07-30 16:38 ` [PATCH 04/20] ia64: remove get_required_mask implementation Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-07-30 16:38 ` [PATCH 05/20] swiotlb: allow the architecture to provide a get_required_mask hook Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-08-27 16:06   ` Konrad Rzeszutek Wilk
2018-08-27 16:06     ` Konrad Rzeszutek Wilk
2018-08-27 16:06     ` Konrad Rzeszutek Wilk
2018-07-30 16:38 ` [PATCH 06/20] dma-noncoherent: add an optional arch hook for ->get_required_mask Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-07-30 16:38 ` [PATCH 07/20] powerpc/dma: remove the unused ARCH_HAS_DMA_MMAP_COHERENT define Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-08-08 23:56   ` Benjamin Herrenschmidt
2018-08-08 23:56     ` Benjamin Herrenschmidt
2018-08-08 23:56     ` Benjamin Herrenschmidt
2018-07-30 16:38 ` [PATCH 08/20] powerpc/dma: remove the unused dma_nommu_ops export Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-07-31 12:16   ` Christoph Hellwig
2018-07-31 12:16     ` Christoph Hellwig
2018-07-31 12:16     ` Christoph Hellwig
2018-08-09  0:01     ` Benjamin Herrenschmidt
2018-08-09  0:01       ` Benjamin Herrenschmidt
2018-08-09  0:01       ` Benjamin Herrenschmidt
2018-08-22  6:45       ` Christoph Hellwig [this message]
2018-08-22  6:45         ` Christoph Hellwig
2018-08-22  6:45         ` Christoph Hellwig
2018-08-22 23:50         ` Benjamin Herrenschmidt
2018-08-22 23:50           ` Benjamin Herrenschmidt
2018-08-22 23:50           ` Benjamin Herrenschmidt
2018-07-30 16:38 ` [PATCH 09/20] powerpc/dma: remove the unused ISA_DMA_THRESHOLD export Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-08-09  0:14   ` Benjamin Herrenschmidt
2018-08-09  0:14     ` Benjamin Herrenschmidt
2018-08-09  0:14     ` Benjamin Herrenschmidt
2018-07-30 16:38 ` [PATCH 10/20] powerpc/dma-noncoherent: don't disable irqs over kmap_atomic Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-08-09  0:27   ` Benjamin Herrenschmidt
2018-08-09  0:27     ` Benjamin Herrenschmidt
2018-08-09  0:27     ` Benjamin Herrenschmidt
2018-08-22  7:02     ` Christoph Hellwig
2018-08-22  7:02       ` Christoph Hellwig
2018-08-22  7:02       ` Christoph Hellwig
2018-08-22 23:45       ` Benjamin Herrenschmidt
2018-08-22 23:45         ` Benjamin Herrenschmidt
2018-08-22 23:45         ` Benjamin Herrenschmidt
2018-07-30 16:38 ` [PATCH 11/20] powerpc/dma: split the two __dma_alloc_coherent implementations Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-08-09  0:40   ` Benjamin Herrenschmidt
2018-08-09  0:40     ` Benjamin Herrenschmidt
2018-08-09  0:40     ` Benjamin Herrenschmidt
2018-07-30 16:38 ` [PATCH 12/20] powerpc/dma: use phys_to_dma instead of get_dma_offset Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-08-09  0:43   ` Benjamin Herrenschmidt
2018-08-09  0:43     ` Benjamin Herrenschmidt
2018-08-09  0:43     ` Benjamin Herrenschmidt
2018-07-30 16:38 ` [PATCH 13/20] powerpc/dma: remove get_dma_offset Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-08-09  0:45   ` Benjamin Herrenschmidt
2018-08-09  0:45     ` Benjamin Herrenschmidt
2018-08-09  0:45     ` Benjamin Herrenschmidt
2018-07-30 16:38 ` [PATCH 14/20] powerpc/dma: replace dma_nommu_dma_supported with dma_direct_supported Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-08-09  0:49   ` Benjamin Herrenschmidt
2018-08-09  0:49     ` Benjamin Herrenschmidt
2018-08-09  0:49     ` Benjamin Herrenschmidt
2018-07-30 16:38 ` [PATCH 15/20] powerpc/dma: remove the unused unmap_page and unmap_sg methods Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-08-09  0:49   ` Benjamin Herrenschmidt
2018-08-09  0:49     ` Benjamin Herrenschmidt
2018-08-09  0:49     ` Benjamin Herrenschmidt
2018-07-30 16:38 ` [PATCH 16/20] powerpc/dma: use dma_direct_{alloc,free} Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-08-09  0:52   ` Benjamin Herrenschmidt
2018-08-09  0:52     ` Benjamin Herrenschmidt
2018-08-09  0:52     ` Benjamin Herrenschmidt
2018-08-27  8:51     ` Scott Wood
2018-08-27  8:51       ` Scott Wood
2018-08-27  8:51       ` Scott Wood
2018-07-30 16:38 ` [PATCH 17/20] powerpc/dma-swiotlb: use generic swiotlb_dma_ops Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-08-09  0:54   ` Benjamin Herrenschmidt
2018-08-09  0:54     ` Benjamin Herrenschmidt
2018-08-09  0:54     ` Benjamin Herrenschmidt
2018-08-09  1:57     ` Benjamin Herrenschmidt
2018-08-09  1:57       ` Benjamin Herrenschmidt
2018-08-09  1:57       ` Benjamin Herrenschmidt
2018-08-22  7:04       ` Christoph Hellwig
2018-08-22  7:04         ` Christoph Hellwig
2018-08-22  7:04         ` Christoph Hellwig
2018-07-30 16:38 ` [PATCH 18/20] powerpc/dma-noncoherent: use generic dma_noncoherent_ops Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-08-09  1:00   ` Benjamin Herrenschmidt
2018-08-09  1:00     ` Benjamin Herrenschmidt
2018-08-09  1:00     ` Benjamin Herrenschmidt
2018-07-30 16:38 ` [PATCH 19/20] powerpc/dma: use the generic dma-direct map_page and map_sg routines Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-07-30 16:38 ` [PATCH 20/20] powerpc/dma: remove dma_nommu_mmap_coherent Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-07-30 16:38   ` Christoph Hellwig
2018-08-09  1:05   ` Benjamin Herrenschmidt
2018-08-09  1:05     ` Benjamin Herrenschmidt
2018-08-09  1:05     ` Benjamin Herrenschmidt

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20180822064531.GA19284@lst.de \
    --to=hch@lst.de \
    --cc=benh@kernel.crashing.org \
    --cc=fenghua.yu@intel.com \
    --cc=iommu@lists.linux-foundation.org \
    --cc=konrad.wilk@oracle.com \
    --cc=linux-ia64@vger.kernel.org \
    --cc=linuxppc-dev@lists.ozlabs.org \
    --cc=mpe@ellerman.id.au \
    --cc=paulus@samba.org \
    --cc=robin.murphy@arm.com \
    --cc=tony.luck@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.