linux-kernel - Re: [PATCH v9 09/38] dma: cirrus: Convert to DT for Cirrus EP93xx

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <ZgUcJByXpLJfG10T@matsya>
Date: Thu, 28 Mar 2024 12:58:36 +0530
From: Vinod Koul <vkoul@...nel.org>
To: Nikita Shubin <nikita.shubin@...uefel.me>
Cc: Alexander Sverdlin <alexander.sverdlin@...il.com>,
	dmaengine@...r.kernel.org, linux-kernel@...r.kernel.org,
	Arnd Bergmann <arnd@...db.de>
Subject: Re: [PATCH v9 09/38] dma: cirrus: Convert to DT for Cirrus EP93xx

On 28-03-24, 10:24, Nikita Shubin wrote:
> Hello Vinod!
> 
> Thank you for looking into this.
> 
> On Thu, 2024-03-28 at 10:01 +0530, Vinod Koul wrote:
> > On 26-03-24, 12:18, Nikita Shubin via B4 Relay wrote:
> > > From: Nikita Shubin <nikita.shubin@...uefel.me>
> > > 
> > > +enum ep93xx_dma_type {
> > > +       M2P_DMA,
> > 
> > Is this missing P2M?
> > 
> > > +       M2M_DMA,
> > > +};
> > > +
> 
> These are internal types used only to distinguish M2P/P2M and M2M
> capable controllers in "of_device_id ep93xx_dma_of_ids[]".
> 
> So M2P_DMA is M2P/P2M, a can rename M2P_DMA to M2P_P2M_DMA to avoid
> confusion.
> 
> 
> > >  struct ep93xx_dma_engine;
> > >  static int ep93xx_dma_slave_config_write(struct dma_chan *chan,
> > >                                          enum
> > > dma_transfer_direction dir,
> > > @@ -129,11 +136,17 @@ struct ep93xx_dma_desc {
> > >         struct list_head                node;
> > >  };
> > >  
> > > +struct ep93xx_dma_chan_cfg {
> > > +       u8                              port;
> > > +       enum dma_transfer_direction     dir;
> > 
> > Why is direction stored here, it should be derived from the prep_xxx
> > call, that has direction as an argument
> > 
> > 
> 
> M2P/P2M channels aren't unidirectional.
> 
> Citing "EP9xx User Guide":
> 
> "Ten fully independent, programmable DMA controller internal M2P/P2M
> channels (5 Tx and 5 Rx)."
> 
> We need to return correct channel based on Device Tree provided data,
> because we need direction in device_alloc_chan_resources() for hardware
> setup before prepping.

Okay it sounds okay in that case...

> 
> May be i am mistaking somewhere.

-- 
~Vinod