RE: [PATCH v5 6/6] vhost: optimize memcpy routines when cc memcpy is used

2024-07-29 Thread Morten Brørup
> From: Mattias Rönnblom [mailto:hof...@lysator.liu.se] > Sent: Monday, 29 July 2024 21.27 > > On 2024-07-29 13:00, Morten Brørup wrote: > >> From: Mattias Rönnblom [mailto:mattias.ronnb...@ericsson.com] > >> Sent: Wednesday, 24 July 2024 09.54 > > > > Which packet mix was used for your tests? Syn

Re: [PATCH v5 6/6] vhost: optimize memcpy routines when cc memcpy is used

2024-07-29 Thread Mattias Rönnblom
On 2024-07-29 13:00, Morten Brørup wrote: From: Mattias Rönnblom [mailto:mattias.ronnb...@ericsson.com] Sent: Wednesday, 24 July 2024 09.54 Which packet mix was used for your tests? Synthetic IMIX, or some live data? I used the same test as was being done when the performance regression was

RE: [PATCH v5 6/6] vhost: optimize memcpy routines when cc memcpy is used

2024-07-29 Thread Morten Brørup
> From: Mattias Rönnblom [mailto:mattias.ronnb...@ericsson.com] > Sent: Wednesday, 24 July 2024 09.54 Which packet mix was used for your tests? Synthetic IMIX, or some live data? > +/* The code generated by GCC (and to a lesser extent, clang) with just > + * a straight memcpy() to copy packets is

[PATCH v5 6/6] vhost: optimize memcpy routines when cc memcpy is used

2024-07-24 Thread Mattias Rönnblom
In build where use_cc_memcpy is set to true, the vhost user PMD suffers a large performance drop on Intel P-cores for small packets, at least when built by GCC and (to a much lesser extent) clang. This patch addresses that issue by using a custom virtio memcpy()-based packet copying routine. Perf