dev.dpdk.org archive mirror
 help / color / mirror / Atom feed
From: Bruce Richardson <bruce.richardson-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
To: Neil Horman <nhorman-2XuSBdqkA4R54TAoqtyWWQ@public.gmane.org>
Cc: "dev-VfR2kkLFssw@public.gmane.org" <dev-VfR2kkLFssw@public.gmane.org>
Subject: Re: [PATCH v4 3/5] hash: add fallback to software CRC32 implementation
Date: Wed, 19 Nov 2014 11:38:27 +0000	[thread overview]
Message-ID: <20141119113827.GB2604@bricha3-MOBL3> (raw)
In-Reply-To: <20141119113408.GA10004-B26myB8xz7F8NnZeBjwnZQMhkBWG/bsMQH7oEaQurus@public.gmane.org>

On Wed, Nov 19, 2014 at 06:34:08AM -0500, Neil Horman wrote:
> On Wed, Nov 19, 2014 at 10:16:14AM +0000, Bruce Richardson wrote:
> > On Tue, Nov 18, 2014 at 04:36:24PM -0500, Neil Horman wrote:
> > > On Tue, Nov 18, 2014 at 05:52:27PM +0000, Bruce Richardson wrote:
> > > > On Tue, Nov 18, 2014 at 12:46:19PM -0500, Neil Horman wrote:
> > > > > On Tue, Nov 18, 2014 at 11:13:17PM +0600, Yerden Zhumabekov wrote:
> > > > > > 
> > > > > > 18.11.2014 22:00, Neil Horman пишет:
> > > > > > > On Tue, Nov 18, 2014 at 09:06:35PM +0600, Yerden Zhumabekov wrote:
> > > > > > >> 18.11.2014 20:41, Neil Horman пишет:
> > > > > > >>> On Tue, Nov 18, 2014 at 08:03:40PM +0600, Yerden Zhumabekov wrote:
> > > > > > >>>>  /**
> > > > > > >>>>   * Use single crc32 instruction to perform a hash on a 4 byte value.
> > > > > > >>>> + * Fall back to software crc32 implementation in case SSE4.2 is
> > > > > > >>>> + * not supported
> > > > > > >>>>   *
> > > > > > >>>>   * @param data
> > > > > > >>>>   *   Data to perform hash on.
> > > > > > >>>> @@ -376,11 +413,18 @@ crc32c_2words(uint64_t data, uint32_t init_val)
> > > > > > >>>>  static inline uint32_t
> > > > > > >>>>  rte_hash_crc_4byte(uint32_t data, uint32_t init_val)
> > > > > > >>>>  {
> > > > > > >>>> -	return _mm_crc32_u32(init_val, data);
> > > > > > >>>> +#ifdef RTE_MACHINE_CPUFLAG_SSE4_2
> > > > > > >>>> +	if (likely(crc32_alg == CRC32_SSE42))
> > > > > > >>>> +		return _mm_crc32_u32(init_val, data);
> > > > > > >>>> +#endif
> > > > > > >>> you don't really need these ifdefs here anymore given that you have a
> > > > > > >>> constructor to do the algorithm selection.  In fact you need to remove them, in
> > > > > > >>> the event you build on a system that doesn't support SSE42, but run on a system
> > > > > > >>> that does.
> > > > > > >> Originally, I thought so as well. I wrote the code without these ifdefs,
> > > > > > >> but it didn't compile on my machine which doesn't support SSE4.2. Error
> > > > > > >> was triggered by nmmintrin.h which has a check for respective GCC
> > > > > > >> extension. So I think these ifdefs are indeed required.
> > > > > > >>
> > > > > > > You need to edit the makefile so that the compiler gets passed the option
> > > > > > > -msse42.  That way it will know to emit sse42 instructions. It will also allow
> > > > > > > you to remove the ifdef from the include file
> > > > > > 
> > > > > > In this case, I guess there are two options:
> > > > > > 1) modify all makefiles which use librte_hash
> > > > > > 2) move all function bodies from rte_hash_crc.h to separate module,
> > > > > > leaving prototype definitions there only.
> > > > > > 
> > > > > > Everybody's up for the second option? :)
> > > > > > 
> > > > > Crud, you're right, I didn't think about the header inclusion issue.  Is it
> > > > > worth adding the jump to enable the dynamic hash selection?
> > > > > Neil
> > > > 
> > > > Maybe for cases where SSE4.2 is not currently available, i.e. for generic builds.
> > > > For builds where we have hardware support confirmed at compile time, just use
> > > > the function from the header file.
> > > > Does that make sense?
> > > > 
> > > I'm not certain of that, as I don't think anything can be 'confirmed' at compile
> > > time.  I.e. just because you have sse42 at compile time doesn't guarantee you
> > > have it at run time with a DSO.  If you have these as macros, you need to enable
> > > sse42 whereever you include the file so that the intrinsic works properly.
> > 
> > Well, if you compile with sse42 at compile time, the compiler is free to insert
> > sse4 instructions at any place it feels like, irrespective of whether or not you
> > use SSE4 intrinsics, so I would never expect such a DSO to work on a system
> > without SSE42 support.
> > 
> > > 
> > > an alternate option would be to not use the intrinsic, and craft some explicit
> > > __asm__ statement that executes the right sse42 instructions.  That way the asm
> > > is directly emitted, without requiring the -msse42 flag at all, and it will just
> > > work in all the files that call it.
> > > 
> > 
> > I really don't like that approach. I think using intrinsics is much more 
> > maintainable.
> > 
> I grant you that using an intrinsic is easier to read, but if the code doesn't
> compile when using the intrinsic unless you have sse42 turned on, I'm not sure
> what choice we have.  and inline asm isn't that hard to maintain.  We're talking
> about three lines of code:
> asm(
>  "mov    %[1],%eax
>  mov    %[2],%edx
>  crc32l %edx,%eax": 
>  [edx] "r" (crc) /*output*/
>  :
>  [1] "r" (crc), /* input */
>  [2] "r" (val)
>  :
>  [eax] "r" /* clobber */
> )
> 
> I don't have the syntax quite right, but its pretty easy to read the intent.
> Its not like we dont have precidence for this, the atomic interface and several
> pmds do this frequently.
> 
> Neil

Fair point. If everyone else is happy enough with it, I'm ok too.

/Bruce

  parent reply	other threads:[~2014-11-19 11:38 UTC|newest]

Thread overview: 98+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-09-03  6:05 [PATCH 0/2] rewritten rte_hash_crc() call Yerden Zhumabekov
     [not found] ` <1409724351-23786-1-git-send-email-e_zhumabekov-8EHiFRVJVgQ@public.gmane.org>
2014-09-03  6:05   ` [PATCH 1/2] hash: add new rte_hash_crc_8byte call Yerden Zhumabekov
2014-09-03  6:05   ` [PATCH 2/2] hash: rte_hash_crc uses 8- and 4-byte CRC32 intrinsics Yerden Zhumabekov
2014-11-13 17:33   ` [PATCH 0/2] rewritten rte_hash_crc() call Thomas Monjalon
2014-11-14  0:52     ` Neil Horman
     [not found]       ` <20141114005211.GC14230-bi+AKbBUZKY6gyzm1THtWbp2dZbC/Bob@public.gmane.org>
2014-11-14  7:15         ` Yerden Zhumabekov
     [not found]           ` <5465AC00.1070602-8EHiFRVJVgQ@public.gmane.org>
2014-11-14 11:33             ` Neil Horman
     [not found]               ` <20141114113327.GA19147-B26myB8xz7F8NnZeBjwnZQMhkBWG/bsMQH7oEaQurus@public.gmane.org>
2014-11-14 11:57                 ` Yerden Zhumabekov
     [not found]                   ` <5465EE3F.2010404-8EHiFRVJVgQ@public.gmane.org>
2014-11-14 13:53                     ` Neil Horman
     [not found]                       ` <20141114135308.GD19147-B26myB8xz7F8NnZeBjwnZQMhkBWG/bsMQH7oEaQurus@public.gmane.org>
2014-11-14 14:33                         ` Thomas Monjalon
2014-11-14 16:43                         ` Yerden Zhumabekov
     [not found]                           ` <5466313B.6040407-8EHiFRVJVgQ@public.gmane.org>
2014-11-14 18:41                             ` Neil Horman
     [not found]                               ` <20141114184157.GF19147-B26myB8xz7F8NnZeBjwnZQMhkBWG/bsMQH7oEaQurus@public.gmane.org>
2014-11-15 21:45                                 ` Yerden Zhumabekov
2014-11-16 17:59   ` [PATCH v2 0/4] rte_hash_crc reworked to be platform-independent Yerden Zhumabekov
     [not found]     ` <1416160760-16087-1-git-send-email-e_zhumabekov-8EHiFRVJVgQ@public.gmane.org>
2014-11-17 11:31       ` Neil Horman
     [not found]         ` <20141117113110.GB17886-B26myB8xz7F8NnZeBjwnZQMhkBWG/bsMQH7oEaQurus@public.gmane.org>
2014-11-17 11:54           ` Yerden Zhumabekov
     [not found]             ` <5469E1ED.4040109-8EHiFRVJVgQ@public.gmane.org>
2014-11-25 17:05               ` Stephen Hemminger
2014-11-18  3:21       ` [PATCH v3 0/5] " Yerden Zhumabekov
     [not found]         ` <cover.1416280649.git.e_zhumabekov-8EHiFRVJVgQ@public.gmane.org>
2014-11-18  3:21           ` [PATCH v3 1/5] hash: add software CRC32 implementation Yerden Zhumabekov
     [not found]             ` <ea79d201b532a9f165396a9197b31d6d331738a6.1416280649.git.e_zhumabekov-8EHiFRVJVgQ@public.gmane.org>
2014-11-25 17:34               ` Stephen Hemminger
2014-11-18  3:21           ` [PATCH v3 2/5] hash: add new rte_hash_crc_8byte call Yerden Zhumabekov
2014-11-18  3:21           ` [PATCH v3 3/5] hash: add fallback to software CRC32 implementation Yerden Zhumabekov
     [not found]             ` <faece334ae2ae58553c0a19c5f8830993ae543f3.1416280649.git.e_zhumabekov-8EHiFRVJVgQ@public.gmane.org>
2014-11-18  4:56               ` Yerden Zhumabekov
     [not found]                 ` <546AD178.9070109-8EHiFRVJVgQ@public.gmane.org>
2014-11-18 13:33                   ` Neil Horman
     [not found]                     ` <20141118133307.GA32375-B26myB8xz7F8NnZeBjwnZQMhkBWG/bsMQH7oEaQurus@public.gmane.org>
2014-11-18 13:37                       ` Yerden Zhumabekov
2014-11-18 13:43                       ` Thomas Monjalon
2014-11-18  3:21           ` [PATCH v3 4/5] hash: rte_hash_crc() slices data into 8-byte pieces Yerden Zhumabekov
2014-11-18  3:25           ` [PATCH v3 5/5] test: remove redundant compile checks Yerden Zhumabekov
2014-11-16 17:59   ` [PATCH v2 1/4] hash: add software CRC32 implementation Yerden Zhumabekov
2014-11-16 17:59   ` [PATCH v2 2/4] hash: add new rte_hash_crc_8byte call Yerden Zhumabekov
2014-11-16 17:59   ` [PATCH v2 3/4] hash: add fallback to software CRC32 implementation Yerden Zhumabekov
     [not found]     ` <1416160760-16087-4-git-send-email-e_zhumabekov-8EHiFRVJVgQ@public.gmane.org>
2014-11-17 12:34       ` Ananyev, Konstantin
     [not found]         ` <2601191342CEEE43887BDE71AB977258213AE40F-kPTMFJFq+rEu0RiL9chJVbfspsVTdybXVpNB7YpNyf8@public.gmane.org>
2014-11-17 12:41           ` Yerden Zhumabekov
2014-11-17 14:06           ` Neil Horman
2014-11-16 17:59   ` [PATCH v2 4/4] hash: rte_hash_crc() slices data into 8-byte pieces Yerden Zhumabekov
2014-11-18 14:03   ` [PATCH v4 0/5] rte_hash_crc reworked to be platform-independent Yerden Zhumabekov
     [not found]     ` <cover.1416318389.git.e_zhumabekov-8EHiFRVJVgQ@public.gmane.org>
2014-11-18 14:03       ` [PATCH v4 1/5] hash: add software CRC32 implementation Yerden Zhumabekov
2014-11-18 14:03       ` [PATCH v4 2/5] hash: add new rte_hash_crc_8byte call Yerden Zhumabekov
2014-11-18 14:03       ` [PATCH v4 3/5] hash: add fallback to software CRC32 implementation Yerden Zhumabekov
     [not found]         ` <a875a94021448514bfbcf1409e4303cde50882f3.1416318389.git.e_zhumabekov-8EHiFRVJVgQ@public.gmane.org>
2014-11-18 14:41           ` Neil Horman
     [not found]             ` <20141118144138.GB32375-B26myB8xz7F8NnZeBjwnZQMhkBWG/bsMQH7oEaQurus@public.gmane.org>
2014-11-18 15:06               ` Yerden Zhumabekov
     [not found]                 ` <546B607B.9030808-8EHiFRVJVgQ@public.gmane.org>
2014-11-18 16:00                   ` Neil Horman
     [not found]                     ` <20141118160005.GC32375-B26myB8xz7F8NnZeBjwnZQMhkBWG/bsMQH7oEaQurus@public.gmane.org>
2014-11-18 16:04                       ` Bruce Richardson
2014-11-18 16:08                         ` Bruce Richardson
2014-11-18 16:38                         ` Neil Horman
2014-11-18 17:13                       ` Yerden Zhumabekov
     [not found]                         ` <546B7E2D.7050705-8EHiFRVJVgQ@public.gmane.org>
2014-11-18 17:29                           ` Wang, Shawn
     [not found]                             ` <12C2AAD9525203489F7B523D670129D91CCF3615-CFWIZfY9kHm7lkz2zA3lWedPMyk3kRfBmjCW/i4Lttk@public.gmane.org>
2014-11-19  4:07                               ` Yerden Zhumabekov
2014-11-18 17:46                           ` Neil Horman
     [not found]                             ` <20141118174619.GE32375-B26myB8xz7F8NnZeBjwnZQMhkBWG/bsMQH7oEaQurus@public.gmane.org>
2014-11-18 17:52                               ` Bruce Richardson
2014-11-18 21:36                                 ` Neil Horman
     [not found]                                   ` <20141118213624.GF32375-B26myB8xz7F8NnZeBjwnZQMhkBWG/bsMQH7oEaQurus@public.gmane.org>
2014-11-19  3:51                                     ` Yerden Zhumabekov
2014-11-19 10:16                                     ` Bruce Richardson
2014-11-19 11:34                                       ` Neil Horman
     [not found]                                         ` <20141119113408.GA10004-B26myB8xz7F8NnZeBjwnZQMhkBWG/bsMQH7oEaQurus@public.gmane.org>
2014-11-19 11:38                                           ` Bruce Richardson [this message]
2014-11-19 11:50                                             ` Ananyev, Konstantin
     [not found]                                               ` <2601191342CEEE43887BDE71AB977258213B6C4D-kPTMFJFq+rEu0RiL9chJVbfspsVTdybXVpNB7YpNyf8@public.gmane.org>
2014-11-19 11:59                                                 ` Yerden Zhumabekov
2014-11-19 15:05                                                 ` Neil Horman
     [not found]                                                   ` <20141119150533.GA28013-bi+AKbBUZKY6gyzm1THtWbp2dZbC/Bob@public.gmane.org>
2014-11-19 16:51                                                     ` Ananyev, Konstantin
2014-11-19 11:35                                       ` Yerden Zhumabekov
     [not found]                                         ` <546C8097.6000509-8EHiFRVJVgQ@public.gmane.org>
2014-11-19 15:07                                           ` Neil Horman
     [not found]                                             ` <20141119150725.GB28013-bi+AKbBUZKY6gyzm1THtWbp2dZbC/Bob@public.gmane.org>
2014-11-20  3:04                                               ` Yerden Zhumabekov
2014-11-18 17:58                               ` Yerden Zhumabekov
2014-11-18 14:03       ` [PATCH v4 4/5] hash: rte_hash_crc() slices data into 8-byte pieces Yerden Zhumabekov
2014-11-18 14:05       ` [PATCH v4 5/5] test: remove redundant compile checks Yerden Zhumabekov
2014-11-20  5:15   ` [PATCH v5 0/7] rte_hash_crc reworked to be platform-independent Yerden Zhumabekov
     [not found]     ` <cover.1416459284.git.e_zhumabekov-8EHiFRVJVgQ@public.gmane.org>
2014-11-20  5:16       ` [PATCH v5 1/7] hash: add software CRC32 implementation Yerden Zhumabekov
2014-11-20  5:16       ` [PATCH v5 2/7] hash: add assembly implementation of CRC32 intrinsics Yerden Zhumabekov
2014-11-20  5:16       ` [PATCH v5 3/7] hash: replace built-in functions implementing SSE4.2 Yerden Zhumabekov
2014-11-20  5:16       ` [PATCH v5 4/7] hash: add rte_hash_crc_8byte function Yerden Zhumabekov
     [not found]         ` <5bc163e39a2e8176f7e0b424c70bb051d2f5a0a8.1416459284.git.e_zhumabekov-8EHiFRVJVgQ@public.gmane.org>
2014-11-21 11:22           ` Neil Horman
     [not found]             ` <20141121112242.GA20661-B26myB8xz7F8NnZeBjwnZQMhkBWG/bsMQH7oEaQurus@public.gmane.org>
2014-11-21 11:26               ` Yerden Zhumabekov
2014-11-20  5:17       ` [PATCH v5 6/7] hash: rte_hash_crc() slices data into 8-byte pieces Yerden Zhumabekov
2014-11-20  5:17       ` [PATCH v5 7/7] test: remove redundant compile checks Yerden Zhumabekov
2014-11-20  5:17       ` [PATCH v5 5/7] hash: add fallback to software CRC32 implementation Yerden Zhumabekov
2014-11-27 21:04       ` [PATCH v5 0/7] rte_hash_crc reworked to be platform-independent Thomas Monjalon
2014-11-28  3:28         ` Yerden Zhumabekov
2015-01-29  8:48   ` [PATCH v6 " Yerden Zhumabekov
     [not found]     ` <cover.1417092208.git.e_zhumabekov-8EHiFRVJVgQ@public.gmane.org>
2015-01-29  8:48       ` [PATCH v6 1/7] hash: add software CRC32 implementation Yerden Zhumabekov
2015-01-29  8:48       ` [PATCH v6 2/7] hash: add assembly implementation of CRC32 intrinsics Yerden Zhumabekov
     [not found]         ` <3fdc339aa3f0206ae2200937b9791af69e83409e.1417092208.git.e_zhumabekov-8EHiFRVJVgQ@public.gmane.org>
2015-02-02  5:15           ` Liang, Cunming
     [not found]             ` <54CF0805.8040103-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
2015-02-02  5:34               ` Yerden Zhumabekov
     [not found]                 ` <54CF0C49.5000002-8EHiFRVJVgQ@public.gmane.org>
2015-02-02  5:59                   ` Liang, Cunming
2015-01-29  8:49       ` [PATCH v6 3/7] hash: replace built-in functions implementing SSE4.2 Yerden Zhumabekov
2015-01-29  8:49       ` [PATCH v6 4/7] hash: add rte_hash_crc_8byte function Yerden Zhumabekov
2015-01-29  8:50       ` [PATCH v6 5/7] hash: add fallback to software CRC32 implementation Yerden Zhumabekov
2015-01-29  8:50       ` [PATCH v6 6/7] hash: rte_hash_crc() slices data into 8-byte pieces Yerden Zhumabekov
2015-01-29  8:50       ` [PATCH v6 7/7] test: remove redundant compile checks Yerden Zhumabekov
2015-02-01 14:13       ` [PATCH v6 0/7] rte_hash_crc reworked to be platform-independent Neil Horman
     [not found]         ` <20150201141356.GA3141-bi+AKbBUZKY6gyzm1THtWbp2dZbC/Bob@public.gmane.org>
2015-02-02  3:07           ` Yerden Zhumabekov
     [not found]             ` <54CEEA01.3090600-8EHiFRVJVgQ@public.gmane.org>
2015-02-02  3:31               ` Neil Horman
     [not found]                 ` <20150202033122.GA18728-0o1r3XBGOEbbgkc5XkKeNuvMHUBZFtU3YPYVAmT7z5s@public.gmane.org>
2015-02-02  5:18                   ` HA: " Жумабеков Ерден Мирзагулович
2015-02-02  5:39                   ` Yerden Zhumabekov
     [not found]                     ` <54CF0D86.3090507-8EHiFRVJVgQ@public.gmane.org>
2015-02-19 15:21                       ` Bruce Richardson
2015-02-23 17:36                         ` Thomas Monjalon
2015-02-24  3:00                           ` Yerden Zhumabekov
     [not found]                             ` <54EBE934.8080707-8EHiFRVJVgQ@public.gmane.org>
2015-02-24  3:10                               ` Thomas Monjalon
2015-02-24  9:12                                 ` Bruce Richardson

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=20141119113827.GB2604@bricha3-MOBL3 \
    --to=bruce.richardson-ral2jqcrhueavxtiumwx3w@public.gmane.org \
    --cc=dev-VfR2kkLFssw@public.gmane.org \
    --cc=nhorman-2XuSBdqkA4R54TAoqtyWWQ@public.gmane.org \
    /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).