From: Bruce Richardson <bruce.richardson@intel.com>
To: "Yang, Zhiyong" <zhiyong.yang@intel.com>
Cc: "Ananyev, Konstantin" <konstantin.ananyev@intel.com>,
Thomas Monjalon <thomas.monjalon@6wind.com>,
"dev@dpdk.org" <dev@dpdk.org>,
"yuanhan.liu@linux.intel.com" <yuanhan.liu@linux.intel.com>,
"De Lara Guarch, Pablo" <pablo.de.lara.guarch@intel.com>
Subject: Re: [dpdk-dev] [PATCH 1/4] eal/common: introduce rte_memset on IA platform
Date: Thu, 8 Dec 2016 10:27:29 +0000 [thread overview]
Message-ID: <20161208102728.GF55440@bricha3-MOBL3.ger.corp.intel.com> (raw)
In-Reply-To: <E182254E98A5DA4EB1E657AC7CB9BD2A3EB586ED@BGSMSX101.gar.corp.intel.com>
On Thu, Dec 08, 2016 at 09:53:12AM +0000, Yang, Zhiyong wrote:
> Hi, Konstantin:
>
> > -----Original Message-----
> > From: Ananyev, Konstantin
> > Sent: Thursday, December 8, 2016 5:26 PM
> > To: Yang, Zhiyong <zhiyong.yang@intel.com>; Thomas Monjalon
> > <thomas.monjalon@6wind.com>
> > Cc: dev@dpdk.org; yuanhan.liu@linux.intel.com; Richardson, Bruce
> > <bruce.richardson@intel.com>; De Lara Guarch, Pablo
> > <pablo.de.lara.guarch@intel.com>
> > Subject: RE: [dpdk-dev] [PATCH 1/4] eal/common: introduce rte_memset on
> > IA platform
> >
> >
> > Hi Zhiyong,
> >
> > >
> > > HI, Thomas:
> > > Sorry for late reply. I have been being always considering your
> > suggestion.
> > >
> > > > -----Original Message-----
> > > > From: Thomas Monjalon [mailto:thomas.monjalon@6wind.com]
> > > > Sent: Friday, December 2, 2016 6:25 PM
> > > > To: Yang, Zhiyong <zhiyong.yang@intel.com>
> > > > Cc: dev@dpdk.org; yuanhan.liu@linux.intel.com; Richardson, Bruce
> > > > <bruce.richardson@intel.com>; Ananyev, Konstantin
> > > > <konstantin.ananyev@intel.com>; De Lara Guarch, Pablo
> > > > <pablo.de.lara.guarch@intel.com>
> > > > Subject: Re: [dpdk-dev] [PATCH 1/4] eal/common: introduce
> > rte_memset
> > > > on IA platform
> > > >
> > > > 2016-12-05 16:26, Zhiyong Yang:
> > > > > +#ifndef _RTE_MEMSET_X86_64_H_
> > > >
> > > > Is this implementation specific to 64-bit?
> > > >
> > >
> > > Yes.
> > >
> > > > > +
> > > > > +#define rte_memset memset
> > > > > +
> > > > > +#else
> > > > > +
> > > > > +static void *
> > > > > +rte_memset(void *dst, int a, size_t n);
> > > > > +
> > > > > +#endif
> > > >
> > > > If I understand well, rte_memset (as rte_memcpy) is using the most
> > > > recent instructions available (and enabled) when compiling.
> > > > It is not adapting the instructions to the run-time CPU.
> > > > There is no need to downgrade at run-time the instruction set as it
> > > > is obviously not a supported case, but it would be nice to be able
> > > > to upgrade a "default compilation" at run-time as it is done in rte_acl.
> > > > I explain this case more clearly for reference:
> > > >
> > > > We can have AVX512 supported in the compiler but disable it when
> > > > compiling
> > > > (CONFIG_RTE_MACHINE=snb) in order to build a binary running almost
> > > > everywhere.
> > > > When running this binary on a CPU having AVX512 support, it will not
> > > > benefit of the AVX512 improvement.
> > > > Though, we can compile an AVX512 version of some functions and use
> > > > them only if the running CPU is capable.
> > > > This kind of miracle can be achieved in two ways:
> > > >
> > > > 1/ For generic C code compiled with a recent GCC, a function can be
> > > > built for several CPUs thanks to the attribute target_clones.
> > > >
> > > > 2/ For manually optimized functions using CPU-specific intrinsics or
> > > > asm, it is possible to build them with non-default flags thanks to the
> > attribute target.
> > > >
> > > > 3/ For manually optimized files using CPU-specific intrinsics or
> > > > asm, we use specifics flags in the makefile.
> > > >
> > > > The function clone in case 1/ is dynamically chosen at run-time
> > > > through ifunc resolver.
> > > > The specific functions in cases 2/ and 3/ must chosen at run-time by
> > > > initializing a function pointer thanks to rte_cpu_get_flag_enabled().
> > > >
> > > > Note that rte_hash and software crypto PMDs have a run-time check
> > > > with
> > > > rte_cpu_get_flag_enabled() but do not override CFLAGS in the Makefile.
> > > > Next step for these libraries?
> > > >
> > > > Back to rte_memset, I think you should try the solution 2/.
> > >
> > > I have read the ACL code, if I understand well , for complex algo
> > > implementation, it is good idea, but Choosing functions at run time
> > > will bring some overhead. For frequently called function Which
> > > consumes small cycles, the overhead maybe is more than the gains
> > optimizations brings For example, for most applications in dpdk, memset only
> > set N = 10 or 12bytes. It consumes fewer cycles.
> >
> > But then what the point to have an rte_memset() using vector instructions at
> > all?
> > From what you are saying the most common case is even less then SSE
> > register size.
> > Konstantin
>
> For most cases, memset is used such as memset(address, 0, sizeof(struct xxx));
> The use case here is small by accident, I only give an example here.
> but rte_memset is introduced to need consider generic case.
> sizeof(struct xxx) is not limited to very small size, such as less than SSE register size.
> I just want to say that the size for the most use case is not very large, So cycles consumed
> Is not large. It is not suited to choose function at run-time since overhead is considered.
>
For small copies with sizes specified at compile time, do compilers not
fully inline the memset call with a fixed-size equivalent. I believe
some compilers used to do so with memcpy - which is why we had a macro
for it in DPDK, so that compile-time constant copies would use regular
memcpy. If that is also the case for memset, then we should perhaps
specify that rte_memset is only for relatively large copies, e.g. >64
bytes. In that case, run-time detection may be worthwhile.
/Bruce
next prev parent reply other threads:[~2016-12-08 10:27 UTC|newest]
Thread overview: 44+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-12-05 8:26 [dpdk-dev] [PATCH 0/4] eal/common: introduce rte_memset and related test Zhiyong Yang
2016-12-02 10:00 ` Maxime Coquelin
2016-12-06 6:33 ` Yang, Zhiyong
2016-12-06 8:29 ` Maxime Coquelin
2016-12-07 9:28 ` Yang, Zhiyong
2016-12-07 9:37 ` Yuanhan Liu
2016-12-07 9:43 ` Yang, Zhiyong
2016-12-07 9:48 ` Yuanhan Liu
2016-12-05 8:26 ` [dpdk-dev] [PATCH 1/4] eal/common: introduce rte_memset on IA platform Zhiyong Yang
2016-12-02 10:25 ` Thomas Monjalon
2016-12-08 7:41 ` Yang, Zhiyong
2016-12-08 9:26 ` Ananyev, Konstantin
2016-12-08 9:53 ` Yang, Zhiyong
2016-12-08 10:27 ` Bruce Richardson [this message]
2016-12-08 10:30 ` Ananyev, Konstantin
2016-12-11 12:32 ` Yang, Zhiyong
2016-12-15 6:51 ` Yang, Zhiyong
2016-12-15 10:12 ` Bruce Richardson
2016-12-16 10:19 ` Yang, Zhiyong
2016-12-19 6:27 ` Yuanhan Liu
2016-12-20 2:41 ` Yao, Lei A
2016-12-15 10:53 ` Ananyev, Konstantin
2016-12-16 2:15 ` Yang, Zhiyong
2016-12-16 11:47 ` Ananyev, Konstantin
2016-12-20 9:31 ` Yang, Zhiyong
2016-12-08 15:09 ` Thomas Monjalon
2016-12-11 12:04 ` Yang, Zhiyong
2016-12-27 10:04 ` [dpdk-dev] [PATCH v2 0/4] eal/common: introduce rte_memset and related test Zhiyong Yang
2016-12-27 10:04 ` [dpdk-dev] [PATCH v2 1/4] eal/common: introduce rte_memset on IA platform Zhiyong Yang
2016-12-27 10:04 ` [dpdk-dev] [PATCH v2 2/4] app/test: add functional autotest for rte_memset Zhiyong Yang
2016-12-27 10:04 ` [dpdk-dev] [PATCH v2 3/4] app/test: add performance " Zhiyong Yang
2016-12-27 10:04 ` [dpdk-dev] [PATCH v2 4/4] lib/librte_vhost: improve vhost perf using rte_memset Zhiyong Yang
2017-01-09 9:48 ` [dpdk-dev] [PATCH v2 0/4] eal/common: introduce rte_memset and related test Yang, Zhiyong
2017-01-17 6:24 ` Yang, Zhiyong
2017-01-17 20:14 ` Thomas Monjalon
2017-01-18 0:15 ` Vincent JARDIN
2017-01-18 2:42 ` Yang, Zhiyong
2017-01-18 7:42 ` Thomas Monjalon
2017-01-19 1:36 ` Yang, Zhiyong
2016-12-05 8:26 ` [dpdk-dev] [PATCH 2/4] app/test: add functional autotest for rte_memset Zhiyong Yang
2016-12-05 8:26 ` [dpdk-dev] [PATCH 3/4] app/test: add performance " Zhiyong Yang
2016-12-05 8:26 ` [dpdk-dev] [PATCH 4/4] lib/librte_vhost: improve vhost perf using rte_memset Zhiyong Yang
2016-12-02 9:46 ` Thomas Monjalon
2016-12-06 8:04 ` Yang, Zhiyong
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20161208102728.GF55440@bricha3-MOBL3.ger.corp.intel.com \
--to=bruce.richardson@intel.com \
--cc=dev@dpdk.org \
--cc=konstantin.ananyev@intel.com \
--cc=pablo.de.lara.guarch@intel.com \
--cc=thomas.monjalon@6wind.com \
--cc=yuanhan.liu@linux.intel.com \
--cc=zhiyong.yang@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).