Re: [PATCH net-next 2/3] net: page_pool: add the possibility to sync DMA memory for non-coherent devices

From: Lorenzo Bianconi <lorenzo@kernel.org>
To: Jonathan Lemon <jonathan.lemon@gmail.com>
Cc: Ilias Apalodimas <ilias.apalodimas@linaro.org>,
	Lorenzo Bianconi <lorenzo.bianconi@redhat.com>,
	Jesper Dangaard Brouer <brouer@redhat.com>,
	netdev@vger.kernel.org, davem@davemloft.net,
	thomas.petazzoni@bootlin.com, matteo.croce@redhat.com
Subject: Re: [PATCH net-next 2/3] net: page_pool: add the possibility to sync DMA memory for non-coherent devices
Date: Fri, 15 Nov 2019 18:53:31 +0200	[thread overview]
Message-ID: <20191115165331.GA2518@localhost.localdomain> (raw)
In-Reply-To: <86A76A45-2CF5-46C4-A7CF-0EC3CB79944B@gmail.com>

[-- Attachment #1: Type: text/plain, Size: 906 bytes --]

[...]

> > > > 
> > > > Okay, then i guess the flag is a better fit for this.
> > > > The only difference would be that the sync semantics will be
> > > > done on 'per
> > > > packet' basis,  instead of 'per pool', but that should be fine
> > > > for our cases.
> > > 
> > > Ack, fine for me.
> > > Do you think when checking for PP_FLAG_DMA_SYNC_DEV we should even
> > > verify
> > > PP_FLAG_DMA_MAP? Something like:
> > > 
> > > if ((pool->p.flags & (PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV)) ==
> > >     (PP_FLAG_DMA_MAP | PP_FLAG_DMA_SYNC_DEV))
> > > 	page_pool_dma_sync_for_device();
> > > 
> > > Regards,
> > > Lorenzo
> > 
> > I think it's better to do the check once on the pool registration and
> > maybe
> > refuse to allocate the pool? Syncing without mapping doesn't really make
> > sense
> 
> +1.

ack, will post v3 soon.

Regards,
Lorenzo

> -- 
> Jonathan

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]