Netdev Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Thomas Bogendoerfer <tsbogend@alpha.franken.de>
To: Christoph Hellwig <hch@lst.de>
Cc: alsa-devel@alsa-project.org, linux-ia64@vger.kernel.org,
	linux-doc@vger.kernel.org, nouveau@lists.freedesktop.org,
	linux-nvme@lists.infradead.org, linux-kernel@vger.kernel.org,
	"James E.J. Bottomley" <James.Bottomley@HansenPartnership.com>,
	linux-mm@kvack.org, Marek Szyprowski <m.szyprowski@samsung.com>,
	linux-samsung-soc@vger.kernel.org,
	Joonyoung Shim <jy0922.shim@samsung.com>,
	linux-scsi@vger.kernel.org,
	Kyungmin Park <kyungmin.park@samsung.com>,
	Ben Skeggs <bskeggs@redhat.com>,
	Matt Porter <mporter@kernel.crashing.org>,
	linux-media@vger.kernel.org,
	Tom Lendacky <thomas.lendacky@amd.com>,
	Pawel Osciak <pawel@osciak.com>,
	Mauro Carvalho Chehab <mchehab@kernel.org>,
	linux-arm-kernel@lists.infradead.org,
	linux-parisc@vger.kernel.org, netdev@vger.kernel.org,
	Seung-Woo Kim <sw0312.kim@samsung.com>,
	linux-mips@vger.kernel.org, iommu@lists.linux-foundation.org
Subject: Re: [PATCH 22/28] sgiseeq: convert from dma_cache_sync to dma_sync_single_for_device
Date: Wed, 2 Sep 2020 23:38:09 +0200	[thread overview]
Message-ID: <20200902213809.GA7998@alpha.franken.de> (raw)
In-Reply-To: <20200901173810.GA25282@alpha.franken.de>

On Tue, Sep 01, 2020 at 07:38:10PM +0200, Thomas Bogendoerfer wrote:
> On Tue, Sep 01, 2020 at 07:16:27PM +0200, Christoph Hellwig wrote:
> > Well, if IP22 doesn't speculate (which I'm pretty sure is the case),
> > dma_sync_single_for_cpu should indeeed be a no-op.  But then there
> > also shouldn't be anything in the cache, as the previous
> > dma_sync_single_for_device should have invalidated it.  So it seems like
> > we are missing one (or more) ownership transfers to the device.  I'll
> > try to look at the the ownership management in a little more detail
> > tomorrow.
> 
> this is the problem:
> 
>        /* Always check for received packets. */
>         sgiseeq_rx(dev, sp, hregs, sregs);
> 
> so the driver will look at the rx descriptor on every interrupt, so
> we cache the rx descriptor on the first interrupt and if there was
> $no rx packet, we will only see it, if cache line gets flushed for
> some other reason. kick_tx() does a busy loop checking tx descriptors,
> with just sync_desc_cpu...

the patch below fixes the problem.

Thomas.


diff --git a/drivers/net/ethernet/seeq/sgiseeq.c b/drivers/net/ethernet/seeq/sgiseeq.c
index 8507ff242014..876e3700a0e4 100644
--- a/drivers/net/ethernet/seeq/sgiseeq.c
+++ b/drivers/net/ethernet/seeq/sgiseeq.c
@@ -112,14 +112,18 @@ struct sgiseeq_private {
 
 static inline void dma_sync_desc_cpu(struct net_device *dev, void *addr)
 {
-       dma_cache_sync(dev->dev.parent, addr, sizeof(struct sgiseeq_rx_desc),
-                      DMA_FROM_DEVICE);
+       struct sgiseeq_private *sp = netdev_priv(dev);
+
+       dma_sync_single_for_device(dev->dev.parent, VIRT_TO_DMA(sp, addr),
+                       sizeof(struct sgiseeq_rx_desc), DMA_FROM_DEVICE);
 }
 
 static inline void dma_sync_desc_dev(struct net_device *dev, void *addr)
 {
-       dma_cache_sync(dev->dev.parent, addr, sizeof(struct sgiseeq_rx_desc),
-                      DMA_TO_DEVICE);
+       struct sgiseeq_private *sp = netdev_priv(dev);
+
+       dma_sync_single_for_device(dev->dev.parent, VIRT_TO_DMA(sp, addr),
+                       sizeof(struct sgiseeq_rx_desc), DMA_TO_DEVICE);
 }
 
-- 
Crap can work. Given enough thrust pigs will fly, but it's not necessarily a
good idea.                                                [ RFC1925, 2.3 ]

  reply	other threads:[~2020-09-02 21:38 UTC|newest]

Thread overview: 77+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <CGME20200819065610eucas1p2fde88e81917071b1888e7cc01ba0f298@eucas1p2.samsung.com>
2020-08-19  6:55 ` a saner API for allocating DMA addressable pages Christoph Hellwig
2020-08-19  6:55   ` [PATCH 01/28] mm: turn alloc_pages into an inline function Christoph Hellwig
2020-08-19  6:55   ` [PATCH 02/28] drm/exynos: stop setting DMA_ATTR_NON_CONSISTENT Christoph Hellwig
2020-08-19  6:55   ` [PATCH 03/28] drm/nouveau/gk20a: " Christoph Hellwig
2020-08-19  6:55   ` [PATCH 04/28] net/au1000-eth: stop using DMA_ATTR_NON_CONSISTENT Christoph Hellwig
2020-08-19  6:55   ` [PATCH 05/28] media/v4l2: remove V4L2-FLAG-MEMORY-NON-CONSISTENT Christoph Hellwig
2020-08-19 11:16     ` Tomasz Figa
2020-08-19 11:51       ` Robin Murphy
2020-08-19 12:49         ` Tomasz Figa
2020-08-19 13:57           ` Christoph Hellwig
2020-08-19 14:11             ` Tomasz Figa
2020-08-20  4:45               ` Christoph Hellwig
2020-08-20 10:09                 ` Tomasz Figa
2020-08-20 16:51                   ` Christoph Hellwig
2020-08-19 14:07           ` Robin Murphy
2020-08-19 14:22             ` Tomasz Figa
2020-08-20  4:52               ` Christoph Hellwig
2020-08-20  5:02             ` Christoph Hellwig
2020-08-20 10:24               ` Tomasz Figa
2020-08-20 16:52                 ` Christoph Hellwig
2020-08-20 17:41                   ` Tomasz Figa
2020-08-19 13:54       ` Christoph Hellwig
2020-08-19 13:57         ` Tomasz Figa
2020-08-20  4:43           ` Christoph Hellwig
2020-08-20  5:20             ` Christoph Hellwig
2020-08-20 10:05               ` Tomasz Figa
2020-08-20 16:54                 ` Christoph Hellwig
2020-08-20 17:33                   ` Tomasz Figa
2020-09-01 11:06                     ` Christoph Hellwig
2020-09-01 15:02                       ` Tomasz Figa
2020-08-19  6:55   ` [PATCH 06/28] lib82596: move DMA allocation into the callers of i82596_probe Christoph Hellwig
2020-09-01 13:29     ` Thomas Bogendoerfer
2020-08-19  6:55   ` [PATCH 07/28] 53c700: improve non-coherent DMA handling Christoph Hellwig
2020-09-01 14:52     ` James Bottomley
2020-09-01 15:05       ` Matthew Wilcox
2020-09-01 15:22         ` James Bottomley
2020-09-01 16:21           ` Helge Deller
2020-09-01 16:41             ` Helge Deller
2020-09-01 16:53               ` Matthew Wilcox
2020-09-02 15:00                 ` Helge Deller
2020-08-19  6:55   ` [PATCH 08/28] MIPS: make dma_sync_*_for_cpu a little less overzealous Christoph Hellwig
2020-09-01 13:53     ` Thomas Bogendoerfer
2020-08-19  6:55   ` [PATCH 09/28] MIPS/jazzdma: remove the unused vdma_remap function Christoph Hellwig
2020-09-01 13:49     ` Thomas Bogendoerfer
2020-08-19  6:55   ` [PATCH 10/28] MIPS/jazzdma: decouple from dma-direct Christoph Hellwig
2020-09-01 13:49     ` Thomas Bogendoerfer
2020-08-19  6:55   ` [PATCH 11/28] dma-mapping: add (back) arch_dma_mark_clean for ia64 Christoph Hellwig
2020-08-19  6:55   ` [PATCH 12/28] dma-direct: remove dma_direct_{alloc,free}_pages Christoph Hellwig
2020-08-19  6:55   ` [PATCH 13/28] dma-direct: lift gfp_t manipulation out of__dma_direct_alloc_pages Christoph Hellwig
2020-08-19  6:55   ` [PATCH 14/28] dma-direct: use phys_to_dma_direct in dma_direct_alloc Christoph Hellwig
2020-08-19  6:55   ` [PATCH 15/28] dma-direct: remove __dma_to_phys Christoph Hellwig
2020-08-19  6:55   ` [PATCH 16/28] dma-direct: rename and cleanup __phys_to_dma Christoph Hellwig
2020-08-19  6:55   ` [PATCH 17/28] dma-mapping: move dma_common_{mmap,get_sgtable} out of mapping.c Christoph Hellwig
2020-08-19  6:55   ` [PATCH 18/28] dma-mapping: move the dma_declare_coherent_memory documentation Christoph Hellwig
2020-08-19  6:55   ` [PATCH 19/28] dma-mapping: replace DMA_ATTR_NON_CONSISTENT with dma_{alloc,free}_pages Christoph Hellwig
2020-08-19 15:03     ` [PATCH 19/28] dma-mapping: replace DMA_ATTR_NON_CONSISTENT with dma_{alloc, free}_pages Tomasz Figa
2020-08-20  5:15       ` Christoph Hellwig
2020-08-19  6:55   ` [PATCH 20/28] sgiwd93: convert from dma_cache_sync to dma_sync_single_for_device Christoph Hellwig
2020-08-19  6:55   ` [PATCH 21/28] hal2: " Christoph Hellwig
2020-08-19  6:55   ` [PATCH 22/28] sgiseeq: " Christoph Hellwig
2020-09-01 15:22     ` Thomas Bogendoerfer
2020-09-01 17:12       ` Thomas Bogendoerfer
2020-09-01 17:16         ` Christoph Hellwig
2020-09-01 17:38           ` Thomas Bogendoerfer
2020-09-02 21:38             ` Thomas Bogendoerfer [this message]
2020-09-03  8:42               ` Christoph Hellwig
2020-09-03  8:43             ` Christoph Hellwig
2020-09-03  8:46               ` Christoph Hellwig
2020-08-19  6:55   ` [PATCH 23/28] lib82596: " Christoph Hellwig
2020-08-19  6:55   ` [PATCH 24/28] 53c700: " Christoph Hellwig
2020-08-19  6:55   ` [PATCH 25/28] dma-mapping: remove dma_cache_sync Christoph Hellwig
2020-08-19  6:55   ` [PATCH 26/28] dmapool: add dma_alloc_pages support Christoph Hellwig
2020-08-19  6:55   ` [PATCH 27/28] nvme-pci: fix PRP pool size Christoph Hellwig
2020-08-19  6:55   ` [PATCH 28/28] nvme-pci: use dma_alloc_pages backed dmapools Christoph Hellwig
2020-08-25 11:30   ` a saner API for allocating DMA addressable pages Marek Szyprowski
2020-08-25 13:26     ` Christoph Hellwig
2020-08-29  9:46   ` Helge Deller

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=20200902213809.GA7998@alpha.franken.de \
    --to=tsbogend@alpha.franken.de \
    --cc=James.Bottomley@HansenPartnership.com \
    --cc=alsa-devel@alsa-project.org \
    --cc=bskeggs@redhat.com \
    --cc=hch@lst.de \
    --cc=iommu@lists.linux-foundation.org \
    --cc=jy0922.shim@samsung.com \
    --cc=kyungmin.park@samsung.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-ia64@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-media@vger.kernel.org \
    --cc=linux-mips@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=linux-nvme@lists.infradead.org \
    --cc=linux-parisc@vger.kernel.org \
    --cc=linux-samsung-soc@vger.kernel.org \
    --cc=linux-scsi@vger.kernel.org \
    --cc=m.szyprowski@samsung.com \
    --cc=mchehab@kernel.org \
    --cc=mporter@kernel.crashing.org \
    --cc=netdev@vger.kernel.org \
    --cc=nouveau@lists.freedesktop.org \
    --cc=pawel@osciak.com \
    --cc=sw0312.kim@samsung.com \
    --cc=thomas.lendacky@amd.com \
    --subject='Re: [PATCH 22/28] sgiseeq: convert from dma_cache_sync to dma_sync_single_for_device' \
    /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

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).