linux-kernel - Re: [PATCH v5 1/3] dmaengine: dw-axi-dmac: Support hardware quirks

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: Windows password security audit tool. GUI, reports in PDF.

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

Date: Fri, 28 Jun 2024 13:09:15 +0530
From: Vinod Koul <vkoul@...nel.org>
To: JiaJie Ho <jiajie.ho@...rfivetech.com>
Cc: Herbert Xu <herbert@...dor.apana.org.au>,
	"David S . Miller" <davem@...emloft.net>,
	Eugeniy Paltsev <Eugeniy.Paltsev@...opsys.com>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	"linux-crypto@...r.kernel.org" <linux-crypto@...r.kernel.org>,
	"dmaengine@...r.kernel.org" <dmaengine@...r.kernel.org>
Subject: Re: [PATCH v5 1/3] dmaengine: dw-axi-dmac: Support hardware quirks


Hi JiaJie

On 12-06-24, 10:13, JiaJie Ho wrote:
> > On 30-05-24, 11:11, Jia Jie Ho wrote:
> > 
> > > +
> > > +struct dw_axi_peripheral_config {
> > > +#define DWAXIDMAC_STARFIVE_SM_ALGO	BIT(0)
> > 
> > what does this quirk mean?
> > 
> > > +	u32 quirks;
> > 
> > Can you explain why you need this to be exposed. I would prefer we use
> > existing interfaces and not define a new one...
> > 
> 
> Hi Vinod,
> Thanks for reviewing this.
> This is a dedicated dma controller for the crypto engine.
> I am adding this quirk to:
> 1. Select the src and dest AXI master for transfers between mem and dev. 
>     Driver currently only uses AXI0 for both.

why cant this information be passed thru DT, that is typically done by
most controllers?

> 2. Workaround a hardware limitation on the crypto engine to
>      transfer data > 256B by incrementing the peripheral FIFO offset.

Dont set the transfer data > 256B, you should ideally use maxburst for
configuring the FIFO...

> 
> What is the recommended way to handle such cases besides using 
> peripheral_config in dma_slave_config?

First use dma_slave_config() fields, if they are not appropriate, lets
discuss why before we add custom methods or use peripheral_config

-- 
~Vinod