Re: RFC: using NIC DMA ring as tracing buffer?
Alexander Aring <[email protected]> Wed, 17 Apr 2024 18:51:26 -0400
| Newsgroups | org.kernel.vger.linux-trace-users,org.kernel.vger.linux-trace-devel |
|---|---|
| Message-ID | <CAK-6q+h=5LcjtTxJ0i3BcaVZRGfyWpQJy6Htxjh_zFfOgUpwfA@mail.gmail.com> |
Hi Mathieu, On Tue, Apr 16, 2024 at 10:49=E2=80=AFAM Mathieu Desnoyers <[email protected]> wrote: > > On 2024-04-16 10:41, Alexander Aring wrote: > > Hi all, > > > > I have an idea about using the DMA rings of a networking card as a > > tracing buffer. > > The first thing I'd be worried about is overhead caused by > false-sharing if you trace a large SMP/many-core system all > into the same global NIC buffer. The cache lines will bounce > left and right which is bad performance-wise. > modern NICs have several tx and rx dma rings. There is an ethtool command to query the number: sudo ethtool -g $IFACE so it depends on the hardware if you can pin one or more rings on a per-cpu buffer basis. I hope that somehow helps to make this idea less worrisome. Thanks. - Alex