mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Frank Li <Frank.li@nxp.com>
To: Koichiro Den <den@valinux.co.jp>
Cc: "Manivannan Sadhasivam" <mani@kernel.org>,
	"Vinod Koul" <vkoul@kernel.org>,
	"Gustavo Pimentel" <Gustavo.Pimentel@synopsys.com>,
	"Kees Cook" <kees@kernel.org>,
	"Gustavo A. R. Silva" <gustavoars@kernel.org>,
	"Krzysztof Wilczyński" <kwilczynski@kernel.org>,
	"Kishon Vijay Abraham I" <kishon@kernel.org>,
	"Bjorn Helgaas" <bhelgaas@google.com>,
	"Christoph Hellwig" <hch@lst.de>,
	"Niklas Cassel" <cassel@kernel.org>,
	dmaengine@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-hardening@vger.kernel.org, linux-pci@vger.kernel.org,
	linux-nvme@lists.infradead.org, imx@lists.linux.dev
Subject: Re: [PATCH v2 01/11] dmaengine: dw-edma: Add spinlock to protect DONE_INT_MASK and ABORT_INT_MASK
Date: Thu, 26 Feb 2026 10:23:27 -0500	[thread overview]
Message-ID: <aaBlb42x3qQxOYjk@lizhi-Precision-Tower-5810> (raw)
In-Reply-To: <jjklzlblwbqleshgnhiedqioxskt45awph5rlkh5bymvitk3fq@blisaiwi3zor>

On Thu, Feb 26, 2026 at 11:22:41AM +0900, Koichiro Den wrote:
> On Wed, Feb 25, 2026 at 10:50:41AM -0500, Frank Li wrote:
> > On Wed, Feb 25, 2026 at 05:26:02PM +0900, Koichiro Den wrote:
> > > On Fri, Jan 09, 2026 at 10:28:21AM -0500, Frank Li wrote:
> > > > The DONE_INT_MASK and ABORT_INT_MASK registers are shared by all DMA
> > > > channels, and modifying them requires a read-modify-write sequence.
> > > > Because this operation is not atomic, concurrent calls to
> > > > dw_edma_v0_core_start() can introduce race conditions if two channels
> > > > update these registers simultaneously.
> > > >
> > > > Add a spinlock to serialize access to these registers and prevent race
> > > > conditions.
> > > >
> > > > Signed-off-by: Frank Li <Frank.Li@nxp.com>
> > > > ---
> > > > vc.lock protect should be another problem. This one just fix register
> > > > access for difference DMA channel.
> > > >
> > > > Other improve defer to dynamtic append descriptor works later.
> > > > ---
> > > >  drivers/dma/dw-edma/dw-edma-v0-core.c | 6 ++++++
> > > >  1 file changed, 6 insertions(+)
> > >
> > > Hi Frank,
> > >
> > > I'm very interested in seeing the work toward the "dynamic append" series land,
> > > but in my opinion this one can be submitted independently.
> >
> > This patch serial is actually straight forwards. we can ask vnod pick first
> > one in case have other problems. put together to reduce patch's dependency.
>
> Yes, I see.
>
> My understanding is that the originally planned dependency chain was:
>
>   #1->#2->#3
>
> #1 [PATCH v2 0/8] dmaengine: Add new API to combine onfiguration and descriptor preparation
>    https://lore.kernel.org/dmaengine/20251218-dma_prep_config-v2-0-c07079836128@nxp.com/
> #2 (this series)
> #3 [PATCH RFT 0/5] dmaengine: dw-edma: support dynamtic add link entry during dma engine running
>    https://lore.kernel.org/dmaengine/20260109-edma_dymatic-v1-0-9a98c9c98536@nxp.com/
>
> I'm not sure whether #1 will proceed, as the thread appears to have stalled. I

Vnod said he will review #1 in this week. If not progress, I will create
new one without dependent #1.

Frank

> might be missing something, though. In any case, #1 is semantically orthogonal
> to #2, so I believe #2 can be considered independently.
>
> Thanks,
> Koichiro
>
> >
> > Frank
> > >
> > > Even in the current mainline, under concurrent multi-channel load, this race can
> > > already be triggered.
> > >
> > > Also, with this patch, dw->lock is no longer "Only for legacy", so we should
> > > probably update the comment in dw-edma-core.h.
> > >
> > > Best regards,
> > > Koichiro
> > >
> > > >
> > > > diff --git a/drivers/dma/dw-edma/dw-edma-v0-core.c b/drivers/dma/dw-edma/dw-edma-v0-core.c
> > > > index b75fdaffad9a4ea6cd8d15e8f43bea550848b46c..2850a9df80f54d00789144415ed2dfe31dea3965 100644
> > > > --- a/drivers/dma/dw-edma/dw-edma-v0-core.c
> > > > +++ b/drivers/dma/dw-edma/dw-edma-v0-core.c
> > > > @@ -364,6 +364,7 @@ static void dw_edma_v0_core_start(struct dw_edma_chunk *chunk, bool first)
> > > >  {
> > > >  	struct dw_edma_chan *chan = chunk->chan;
> > > >  	struct dw_edma *dw = chan->dw;
> > > > +	unsigned long flags;
> > > >  	u32 tmp;
> > > >
> > > >  	dw_edma_v0_core_write_chunk(chunk);
> > > > @@ -408,6 +409,8 @@ static void dw_edma_v0_core_start(struct dw_edma_chunk *chunk, bool first)
> > > >  			}
> > > >  		}
> > > >  		/* Interrupt unmask - done, abort */
> > > > +		raw_spin_lock_irqsave(&dw->lock, flags);
> > > > +
> > > >  		tmp = GET_RW_32(dw, chan->dir, int_mask);
> > > >  		tmp &= ~FIELD_PREP(EDMA_V0_DONE_INT_MASK, BIT(chan->id));
> > > >  		tmp &= ~FIELD_PREP(EDMA_V0_ABORT_INT_MASK, BIT(chan->id));
> > > > @@ -416,6 +419,9 @@ static void dw_edma_v0_core_start(struct dw_edma_chunk *chunk, bool first)
> > > >  		tmp = GET_RW_32(dw, chan->dir, linked_list_err_en);
> > > >  		tmp |= FIELD_PREP(EDMA_V0_LINKED_LIST_ERR_MASK, BIT(chan->id));
> > > >  		SET_RW_32(dw, chan->dir, linked_list_err_en, tmp);
> > > > +
> > > > +		raw_spin_unlock_irqrestore(&dw->lock, flags);
> > > > +
> > > >  		/* Channel control */
> > > >  		SET_CH_32(dw, chan->dir, chan->id, ch_control1,
> > > >  			  (DW_EDMA_V0_CCS | DW_EDMA_V0_LLE));
> > > >
> > > > --
> > > > 2.34.1
> > > >

  reply	other threads:[~2026-02-26 15:23 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-01-09 15:28 [PATCH v2 00/11] dmaengine: dw-edma: flatten desc structions and simple code Frank Li
2026-01-09 15:28 ` [PATCH v2 01/11] dmaengine: dw-edma: Add spinlock to protect DONE_INT_MASK and ABORT_INT_MASK Frank Li
2026-02-25  8:26   ` Koichiro Den
2026-02-25 15:50     ` Frank Li
2026-02-26  2:22       ` Koichiro Den
2026-02-26 15:23         ` Frank Li [this message]
2026-02-27  6:29           ` Koichiro Den
2026-01-09 15:28 ` [PATCH v2 02/11] dmaengine: dw-edma: Move control field update of DMA link to the last step Frank Li
2026-01-09 15:28 ` [PATCH v2 03/11] dmaengine: dw-edma: Add xfer_sz field to struct dw_edma_chunk Frank Li
2026-01-09 15:28 ` [PATCH v2 04/11] dmaengine: dw-edma: Remove ll_max = -1 in dw_edma_channel_setup() Frank Li
2026-02-25  8:30   ` Koichiro Den
2026-02-25 15:43     ` Frank Li
2026-02-26  2:16       ` Koichiro Den
2026-02-26 15:21         ` Frank Li
2026-02-27  6:30           ` Koichiro Den
2026-01-09 15:28 ` [PATCH v2 05/11] dmaengine: dw-edma: Move ll_region from struct dw_edma_chunk to struct dw_edma_chan Frank Li
2026-01-09 15:28 ` [PATCH v2 06/11] dmaengine: dw-edma: Pass down dw_edma_chan to reduce one level of indirection Frank Li
2026-01-09 15:28 ` [PATCH v2 07/11] dmaengine: dw-edma: Add helper dw_(edma|hdma)_v0_core_ch_enable() Frank Li
2026-01-09 15:28 ` [PATCH v2 08/11] dmaengine: dw-edma: Add callbacks to fill link list entries Frank Li
2026-01-09 15:28 ` [PATCH v2 09/11] dmaengine: dw-edma: Use common dw_edma_core_start() for both eDMA and HDMA Frank Li
2026-01-09 15:28 ` [PATCH v2 10/11] dmaengine: dw-edma: Use burst array instead of linked list Frank Li
2026-01-09 15:28 ` [PATCH v2 11/11] dmaengine: dw-edma: Remove struct dw_edma_chunk Frank Li
2026-01-12 13:07 ` [PATCH v2 00/11] dmaengine: dw-edma: flatten desc structions and simple code Niklas Cassel

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=aaBlb42x3qQxOYjk@lizhi-Precision-Tower-5810 \
    --to=frank.li@nxp.com \
    --cc=Gustavo.Pimentel@synopsys.com \
    --cc=bhelgaas@google.com \
    --cc=cassel@kernel.org \
    --cc=den@valinux.co.jp \
    --cc=dmaengine@vger.kernel.org \
    --cc=gustavoars@kernel.org \
    --cc=hch@lst.de \
    --cc=imx@lists.linux.dev \
    --cc=kees@kernel.org \
    --cc=kishon@kernel.org \
    --cc=kwilczynski@kernel.org \
    --cc=linux-hardening@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-nvme@lists.infradead.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=mani@kernel.org \
    --cc=vkoul@kernel.org \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®