[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20240129060947.GC19258@lst.de>
Date: Mon, 29 Jan 2024 07:09:47 +0100
From: Christoph Hellwig <hch@....de>
To: Robin Murphy <robin.murphy@....com>
Cc: Petr Tesařík <petr@...arici.cz>,
Alexander Lobakin <aleksander.lobakin@...el.com>,
"David S. Miller" <davem@...emloft.net>,
Eric Dumazet <edumazet@...gle.com>,
Jakub Kicinski <kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>,
Christoph Hellwig <hch@....de>,
Marek Szyprowski <m.szyprowski@...sung.com>,
Joerg Roedel <joro@...tes.org>, Will Deacon <will@...nel.org>,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
"Rafael J. Wysocki" <rafael@...nel.org>,
Magnus Karlsson <magnus.karlsson@...el.com>,
Maciej Fijalkowski <maciej.fijalkowski@...el.com>,
Alexander Duyck <alexanderduyck@...com>, bpf@...r.kernel.org,
netdev@...r.kernel.org, iommu@...ts.linux.dev,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH net-next 2/7] dma: avoid expensive redundant calls for
sync operations
On Fri, Jan 26, 2024 at 07:13:05PM +0000, Robin Murphy wrote:
>> Can we have a comment that states this assumption along with the flag?
>> Because when it breaks, it will keep someone cursing for days why DMA
>> sometimes fails on their device before they find out it's not synced.
>> And then wondering why the code makes such silly assumptions...
>
> Indeed, apologies if it wasn't totally clear, but I really was implying a
> literal "may skip sync if coherent and not using SWIOTLB (which matches
> dma-direct)" flag, documented as such, and not trying to dress it up as
> anything more generic. I just can't suggest a suitably concise name for
> that of the top of my head... :)
Yes, that seems like the right way to go.
Powered by blists - more mailing lists