Is K12 still alive?By: Heikki Kultala (heikki.kultala.delete@this.tut.fi), 2017-05-12 05:34 UTCRoom: Moderated DiscussionsHas the K12 project been cancelled or have just all chips using K12 been delayed when AMD has been concentrating on getting summit ridge and raven ridge out? Next Post in Thread > Reply to this TopicThread (429 posts)TopicPosted ByPostedIs K12 still alive?Heikki Kultala2017-05-12 05:34 UTC It never made senseSomeone2017-05-12 07:58 UTC It never made sensejuanrga2017-05-12 12:02 UTC It never made senseMichael S2017-05-12 12:47 UTC It never made senseanon.12017-05-12 15:19 UTC It never made sensewumpus2017-05-12 23:57 UTC It never made senseanon.12017-05-13 01:37 UTC It never made sensewumpus2017-05-13 14:52 UTC It never made senseanon.12017-05-14 01:29 UTC It never made senseDavid Kanter2017-05-14 07:41 UTC It never made sensejuanrga2017-05-14 12:23 UTC It never made sensebakaneko2017-05-14 12:56 UTC It never made senseanon.12017-05-14 15:36 UTC Hierofalcon ?Michael S2017-05-14 08:15 UTC Hierofalcon ?anyone2017-05-15 17:05 UTC It never made sensejuanrga2017-05-13 02:11 UTC It never made senseanon.12017-05-13 13:59 UTC It never made sensejuanrga2017-05-14 11:35 UTC It never made senseanon.12017-05-14 16:26 UTC It never made sensejuanrga2017-05-14 23:47 UTC It never made senseanon.12017-05-15 00:49 UTC It never made sensejuanrga2017-05-17 12:10 UTC It never made senseanon.12017-05-18 16:11 UTC It never made sensejuanrga2017-05-20 10:10 UTC It never made senseanon.12017-05-20 16:40 UTC It never made senseBrett2017-05-20 18:08 UTC It never made sensewumpus2017-05-20 19:27 UTC It never made senseMichael S2017-05-20 20:49 UTC It never made senseanon.12017-05-20 23:19 UTC It never made senseBrett2017-05-21 00:44 UTC It never made senseanon.12017-05-21 01:22 UTC It never made senseBrett2017-05-21 02:08 UTC It never made senseanon.12017-05-21 02:35 UTC It never made senseJouni Osmala2017-05-21 15:45 UTC It never made senseBrett2017-05-21 19:28 UTC It never made senseJouni Osmala2017-05-22 08:07 UTC It never made senseMichael S2017-05-22 08:27 UTC It never made senseMaynard Handley2017-05-22 03:09 UTC It never made senseAndreas2017-05-23 12:03 UTC It never made senseMaynard Handley2017-05-23 16:37 UTC It never made senseAndreas2017-05-24 12:11 UTC It never made sensedmcq2017-05-21 00:45 UTC It never made senseanon.12017-05-21 01:24 UTC It never made senseanon.12017-05-21 02:43 UTC It never made sensedmcq2017-05-21 18:34 UTC It never made senseblue2017-05-21 20:29 UTC It never made senseblue2017-05-21 20:30 UTC It never made senseMaynard Handley2017-05-22 03:12 UTC To all! Snip your citations. It's annoying as hell asit is!!! (NT)gallier22017-05-22 07:48 UTC Bogus ICC comparisonWilco2017-05-21 11:06 UTC Bogus ICC comparisonanon.12017-05-21 15:09 UTC Bogus ICC comparisonMichael S2017-05-21 16:11 UTC Bogus ICC comparisonDavid Kanter2017-05-21 19:42 UTC Bogus ICC comparisonAnne O'Nonymous2017-05-22 11:14 UTC Bogus ICC comparisonslacker2017-05-22 12:21 UTC Bogus ICC comparisonAnne O'Nymous2017-05-23 18:26 UTC Bogus ICC comparisondmcq2017-05-22 12:55 UTC Bogus ICC comparisonanon.12017-05-22 18:59 UTC Bogus ICC comparisonWilco2017-05-22 20:15 UTC Bogus ICC comparisonanon.12017-05-22 18:44 UTC Bogus ICC comparisonWilco2017-05-22 19:55 UTC Just look at the 403.gcc resultsDoug S2017-05-21 19:24 UTC Just look at the 403.gcc resultsMaynard Handley2017-05-22 03:17 UTC Just look at the 403.gcc resultsDoug S2017-05-22 05:14 UTC Just look at the 403.gcc resultsdmcq2017-05-22 13:08 UTC It never made sensejuanrga2017-05-21 12:46 UTC It never made senseanon.12017-05-21 14:57 UTC It never made senseanon.12017-05-21 15:32 UTC It never made senseAnne O'Nonymous2017-05-22 11:11 UTC required PRF sizeHeikki Kultala2017-05-15 03:59 UTC required PRF sizeWilco2017-05-15 09:18 UTC required PRF sizeMichael S2017-05-15 10:05 UTC required PRF sizeanon.12017-05-15 13:57 UTC required PRF sizeWilco2017-05-15 21:46 UTC required PRF sizeanon.12017-05-16 01:30 UTC required PRF sizeWilco2017-05-16 10:50 UTC required PRF sizeMichael S2017-05-16 11:23 UTC required PRF sizeanon.12017-05-16 13:57 UTC required PRF sizeRicardo B2017-05-16 16:10 UTC required PRF sizeanon.12017-05-16 18:56 UTC Thanks! (NT)Ricardo B2017-05-16 22:51 UTC required PRF sizeJouni Osmala2017-05-17 05:03 UTC required PRF sizeanon.12017-05-17 07:04 UTC required PRF sizeMaynard Handley2017-05-16 23:56 UTC required PRF sizeanon.12017-05-16 15:21 UTC required PRF sizeLinus B Torvalds2017-05-15 17:11 UTC required PRF sizeMichael S2017-05-15 18:20 UTC required PRF sizeLinus B Torvalds2017-05-15 22:49 UTC required PRF sizeJouni Osmala2017-05-17 13:04 UTC Load-op usageWilco2017-05-15 23:29 UTC Load-op usageanon52017-05-16 01:05 UTC Load-op usageWilco2017-05-17 00:15 UTC Load-op usageMichael S2017-05-17 08:00 UTC Load-op usageWilco2017-05-17 10:02 UTC could it be C vs C++? (NT)Michael S2017-05-17 10:46 UTC Load-op usageGabriele Svelto2017-05-17 12:27 UTC Load-op usageGian-Carlo Pascutto2017-05-17 15:53 UTC Use perf top?Travis2017-05-17 20:21 UTC Use perf top?Wilco2017-05-17 23:23 UTC Use perf top?Travis2017-05-18 01:12 UTC Use perf top?Seni2017-05-18 04:13 UTC Use perf top?Wilco2017-05-18 10:37 UTC Compiled on Skylake? (NT)Michael S2017-05-18 11:16 UTC Use perf top?Gabriele Svelto2017-05-18 12:19 UTC Use perf top?octoploid2017-05-18 12:48 UTC Use perf top?Gabriele Svelto2017-05-18 16:33 UTC Use perf top?octoploid2017-05-18 17:51 UTC Use perf top?Gabriele Svelto2017-05-18 20:12 UTC Use perf top?octoploid2017-05-18 20:29 UTC Use perf top?Gian-Carlo Pascutto2017-05-22 15:21 UTC Use perf top?octoploid2017-05-22 16:01 UTC Use perf top?Gian-Carlo Pascutto2017-05-22 17:21 UTC Use perf top?octoploid2017-05-22 17:34 UTC Use perf top?Gian-Carlo Pascutto2017-05-22 17:53 UTC Use perf top?octoploid2017-05-23 10:54 UTC Use perf top?rwessel2017-05-23 15:58 UTC Use perf top?octoploid2017-05-23 16:09 UTC Use perf top?Megol2017-05-24 12:04 UTC Use perf top?octoploid2017-05-24 12:24 UTC Use perf top?Gian-Carlo Pascutto2017-05-24 13:53 UTC Use perf top?octoploid2017-05-24 14:01 UTC Use perf top?Megol2017-05-25 20:24 UTC Use perf top?Wilco2017-05-18 10:20 UTC Use perf top?Travis2017-05-18 21:24 UTC Use perf top?Wilco2017-05-18 23:50 UTC Use perf top?Travis2017-05-19 02:34 UTC Load-op usageMichael S2017-05-17 08:21 UTC Load-op usageWilco2017-05-17 10:20 UTC Load-op usageLinus B Torvalds2017-05-17 16:29 UTC Load-op usageLinus B Torvalds2017-05-17 21:45 UTC Load-op usageanon.12017-05-16 01:36 UTC Load-op usageMichael S2017-05-16 08:27 UTC Load-op usageanon.12017-05-16 14:52 UTC Load-op usageanon.12017-05-16 14:58 UTC Load-op usageMichael S2017-05-17 07:52 UTC Load-op usageanon.12017-05-17 14:03 UTC Load-op usageMichael S2017-05-17 14:24 UTC Load-op usageanon.12017-05-18 06:53 UTC Load-op usageMichael S2017-05-18 07:48 UTC Load-op usageLinus B Torvalds2017-05-16 16:01 UTC Load-op usageLinus B Torvalds2017-05-16 16:17 UTC Load-op usage_Arthur2017-05-18 00:11 UTC Load-op usageMichael S2017-05-18 09:50 UTC Load-op usageLinus B Torvalds2017-05-18 17:03 UTC Load-op usageoctoploid2017-05-18 18:45 UTC Load-op usageLinus B Torvalds2017-05-18 19:28 UTC required PRF sizeanon.12017-05-15 14:44 UTC required PRF sizeslacker2017-05-16 00:20 UTC required PRF sizeanon.12017-05-16 02:48 UTC required PRF sizeslacker2017-05-16 04:52 UTC Fixed linkslacker2017-05-16 04:54 UTC required PRF sizeanon.12017-05-16 14:56 UTC It never made senseanon.12017-05-13 15:03 UTC It never made senseanon.12017-05-13 15:31 UTC It never made sensenobody in particular2017-05-13 16:02 UTC It never made senseGabriele Svelto2017-05-13 16:05 UTC It never made senseanon.12017-05-13 18:07 UTC It never made senseAaron Spink2017-05-14 00:18 UTC It never made senseDavid Hess2017-05-14 02:28 UTC It never made senseBrett2017-05-14 05:25 UTC It never made senseanon.12017-05-14 06:44 UTC It never made senseNiels Jørgen Kruse2017-05-14 09:37 UTC It never made senseanon.12017-05-14 16:45 UTC It never made senseNiels Jørgen Kruse2017-05-14 20:06 UTC It never made senseMaynard Handley2017-05-16 11:46 UTC It never made senseNiels Jørgen Kruse2017-05-17 05:24 UTC It never made sensejuanrga2017-05-14 12:02 UTC It never made sensenobody in particular2017-05-14 12:31 UTC It never made sensejuanrga2017-05-14 21:36 UTC It never made sensenobody in particular2017-05-14 22:50 UTC It never made sensejuanrga2017-05-15 00:36 UTC You're discussing two dead-in-the-water architecturesdefault2017-05-15 21:52 UTC You're discussing two dead-in-the-water architecturesblue2017-05-16 02:14 UTC You're discussing two dead-in-the-water architecturesjuanrga2017-05-17 11:52 UTC It never made senseanon.12017-05-14 15:27 UTC It never made senseMichael S2017-05-14 15:54 UTC It never made senseanon.12017-05-14 16:40 UTC It never made sensejuanrga2017-05-14 22:09 UTC It never made sensenobody in particular2017-05-14 22:51 UTC It never made senseMichael S2017-05-14 22:56 UTC It never made senseanon.12017-05-15 00:54 UTC It never made senseDavid Hess2017-05-14 18:02 UTC It never made senseBrett2017-05-14 20:24 UTC It never made senseMichael S2017-05-15 11:55 UTC It never made senseAnon2017-05-15 23:14 UTC It never made senseMichael S2017-05-16 09:21 UTC It never made sensehobel2017-05-16 15:42 UTC It never made senseDavid Hess2017-05-15 13:33 UTC It never made sensewumpus2017-05-14 22:08 UTC It never made senseDavid Hess2017-05-15 13:23 UTC It never made sensejuanrga2017-05-14 11:49 UTC It never made senseAaron Spink2017-05-14 11:58 UTC It never made senseHeikki Kultala2017-05-12 18:47 UTC It never made senseAaron Spink2017-05-14 00:20 UTC It never made senseWes Felter2017-05-12 20:18 UTC It never made senseanon.12017-05-13 01:32 UTC Is K12 still alive?juanrga2017-05-12 11:49 UTC Is K12 still alive?Heikki Kultala2017-05-12 18:31 UTC Is K12 still alive?who me?2017-05-18 02:39 UTC Is K12 still alive?juanrga2017-05-18 09:44 UTC Is K12 still alive?dmcq2017-05-22 13:19 UTC Is K12 still alive?Foo_2017-05-22 14:56 UTC Is K12 still alive?David Kanter2017-05-22 21:42 UTC Is K12 still alive?Linus B Torvalds2017-05-23 02:45 UTC Is K12 still alive?Michael_S2017-05-23 06:34 UTC Is K12 still alive?David Kanter2017-05-23 16:17 UTC Is K12 still alive?Linus B Torvalds2017-05-23 17:29 UTC Is K12 still alive?octoploid2017-05-23 18:25 UTC slow AVX-512 memcpy/memsetEric Bron2017-05-23 19:48 UTC slow AVX-512 memcpy/memsetLinus B Torvalds2017-05-23 20:51 UTC slow AVX-512 memcpy/memsetEric Bron2017-05-23 21:05 UTC slow AVX-512 memcpy/memsetLinus B Torvalds2017-05-23 21:43 UTC slow AVX-512 memcpy/memsetEric Bron2017-05-23 21:59 UTC KNL code generator vs 2014Michael S2017-05-24 07:57 UTC KNL code generator vs 2014Eric Bron2017-05-24 11:21 UTC KNL code generator vs 2014anon.5122017-05-24 23:03 UTC KNL code generator vs 2014Michael S2017-05-25 15:32 UTC food for thoughtEric Bron2017-05-24 23:57 UTC icc 17 on godbolt disagreeMichael S2017-05-25 08:45 UTC Sorry, I posted SKX code twiceMichael S2017-05-25 08:48 UTC stall 2 - are KNL VPUs really OoO?Michael S2017-05-25 09:27 UTC which version of icc 17 ? (NT)Eric Bron2017-05-25 10:50 UTC 17.0.0Michael S2017-05-25 10:52 UTC 17.0.0Eric Bron2017-05-25 11:13 UTC 17.0.0Eric Bron2017-05-25 11:24 UTC 17.0.0Michael S2017-05-25 12:29 UTC 17.0.0Eric Bron2017-05-25 12:43 UTC 17.0.0Michael S2017-05-25 15:40 UTC strange 256-bit code with icc v7.0.4Eric Bron2017-05-25 17:51 UTC 17.0.0Eric Bron2017-05-25 12:54 UTC fixed exampleEric Bron2017-05-25 11:57 UTC slow AVX-512 memcpy/memsetTravis2017-05-23 22:57 UTC correction: has NOT been the caseTravis2017-05-23 22:58 UTC slow AVX-512 memcpy/memsetanon2017-05-24 13:00 UTC slow AVX-512 memcpy/memsetTravis2017-05-24 21:27 UTC slow AVX-512 memcpy/memsetanon2017-05-25 09:16 UTC slow AVX-512 memcpy/memsetTravis2017-05-26 00:02 UTC slow AVX-512 memcpy/memsetGabriele Svelto2017-05-24 12:12 UTC slow AVX-512 memcpy/memsetDoug S2017-05-23 21:35 UTC slow AVX-512 memcpy/memsetLinus B Torvalds2017-05-23 22:07 UTC Dedicated mem* instructionsDoug S2017-05-24 06:17 UTC Dedicated mem* instructionsLinus Torvalds2017-05-24 08:21 UTC Dedicated mem* instructionsLinus Torvalds2017-05-24 15:16 UTC Dedicated mem* instructionsanon2017-05-24 16:52 UTC Dedicated mem* instructionsLinus Torvalds2017-05-24 18:31 UTC Should mem copy/fill/move be an instruction or a coprocessor with asychronous instructions? (NT)TEMLIB2017-05-24 19:52 UTC asynchronous co-processors are evil (NT)Michael S2017-05-24 19:57 UTC Should mem copy/fill/move be an instruction or a coprocessor with asychronous instructions?David Hess2017-05-24 22:52 UTC Should mem copy/fill/move be an instruction or a coprocessor with asychronous instructions?Travis2017-05-24 22:55 UTC Should mem copy/fill/move be an instruction or a coprocessor with asychronous instructions?TEMLIB2017-05-24 23:29 UTC Dedicated mem* instructionsanon2017-05-25 03:39 UTC AVX-512 and XOPYuhong Bao2017-05-25 06:19 UTC 128-bit vs 256-bit vectors in cryptoYuhong Bao2017-05-31 18:37 UTC Dedicated mem* instructionsDoug S2017-05-24 19:37 UTC Dedicated mem* instructionsMichael S2017-05-24 19:55 UTC Dedicated mem* instructionsDoug S2017-05-24 21:35 UTC Dedicated mem* instructionsLinus Torvalds2017-05-24 22:41 UTC Dedicated mem* instructionsTravis2017-05-24 23:20 UTC Dedicated mem* instructionsLinus Torvalds2017-05-25 17:54 UTC Dedicated mem* instructionsGabriele Svelto2017-05-25 23:05 UTC Immediate lengths for mem* instructionsPaul A. Clayton2017-05-26 11:55 UTC slow AVX-512 memcpy/memsetTravis2017-05-24 22:41 UTC ucode branch predictionDavid Kanter2017-05-25 00:45 UTC Then why use even AVX2 for memcpy?Mark Roulo2017-05-23 23:30 UTC Then why use even AVX2 for memcpy?Linus B Torvalds2017-05-24 05:08 UTC Danke (NT).Mark Roulo2017-05-24 18:52 UTC It's all about the length of the memcpy.Heikki Kultala2017-05-24 05:18 UTC It's all about the length of the memcpy.Heikki Kultala2017-05-24 05:26 UTC It's all about the length of the memcpy.Yoav2017-05-24 08:08 UTC It's all about the length of the memcpy.Michael S2017-05-24 08:37 UTC It's all about the length of the memcpy.Megol2017-05-24 10:39 UTC It's all about the length of the memcpy.Gabriele Svelto2017-05-24 12:17 UTC It's all about the length of the memcpy.Travis2017-05-24 21:46 UTC It's all about the length of the memcpy.Gabriele Svelto2017-05-25 11:24 UTC It's all about the length of the memcpy.octoploid2017-05-25 11:45 UTC Forgot , but you get the idea (NT)octoploid2017-05-25 12:12 UTC Forgot to add a pre tag but you get the idea (NT)octoploid2017-05-25 12:14 UTC It's all about the length of the memcpy.Gabriele Svelto2017-05-25 22:37 UTC It's all about the length of the memcpy.Wilco2017-05-25 22:48 UTC It's all about the length of the memcpy.Gabriele Svelto2017-05-25 23:07 UTC It's all about the length of the memcpy.Wilco2017-05-26 09:47 UTC "manual memcpy" and modern compilersHeikki Kultala2017-05-28 06:27 UTC "manual memcpy" and modern compilersLinus Torvalds2017-05-30 03:30 UTC "manual memcpy" and modern compilersTravis2017-05-30 04:32 UTC "manual memcpy" and modern compilersLinus Torvalds2017-05-30 17:54 UTC "manual memcpy" and modern compilersJason Creighton2017-05-30 19:33 UTC "manual memcpy" and modern compilersWilco2017-05-31 03:29 UTC "manual memcpy" and modern compilersTravis2017-05-31 03:23 UTC "manual memcpy" and modern compilersWilco2017-05-31 03:34 UTC "manual memcpy" and modern compilersoctoploid2017-05-31 04:46 UTC "manual memcpy" and modern compilersWilco2017-05-31 09:28 UTC "manual memcpy" and modern compilersoctoploid2017-05-31 10:14 UTC "manual memcpy" and modern compilersWilco2017-05-31 21:42 UTC "manual memcpy" and modern compilersTravis2017-06-01 01:40 UTC "manual memcpy" and modern compilersJouni Osmala2017-06-01 06:42 UTC "manual memcpy" and modern compilersLinus Torvalds2017-06-01 17:39 UTC "manual memcpy" and modern compilersTravis2017-06-01 23:30 UTC "manual memcpy" and modern compilersoctoploid2017-06-02 08:26 UTC "manual memcpy" and modern compilersoctoploid2017-06-02 08:27 UTC "manual memcpy" and modern compilersTravis2017-06-02 19:18 UTC "manual memcpy" and modern compilersTravis2017-06-02 19:40 UTC "manual memcpy" and modern compilersoctoploid2017-06-02 10:29 UTC "manual memcpy" and modern compilersGiGNiC2017-06-02 12:23 UTC "manual memcpy" and modern compilersTravis2017-06-03 02:56 UTC "manual memcpy" and modern compilersTravis2017-06-02 21:05 UTC "manual memcpy" and modern compilersLinus Torvalds2017-06-02 22:48 UTC "manual memcpy" and modern compilersTravis2017-06-02 23:50 UTC "manual memcpy" and modern compilersgiovanni deretta2017-06-03 20:43 UTC "manual memcpy" and modern compilersDavid Kanter2017-06-04 17:04 UTC "manual memcpy" and modern compilersTravis2017-06-04 20:53 UTC "manual memcpy" and modern compilersDavid Kanter2017-06-05 04:03 UTC memory renamingTravis2017-06-06 18:52 UTC memory renaminganon.12017-06-08 03:06 UTC memory renaminganon.12017-06-08 03:54 UTC "manual memcpy" and modern compilersTravis2017-06-03 03:21 UTC "manual memcpy" and modern compilersoctoploid2017-06-03 04:31 UTC "manual memcpy" and modern compilersoctoploid2017-06-03 09:19 UTC "manual memcpy" and modern compilersTravis2017-06-03 18:38 UTC "manual memcpy" and modern compilersLinus Torvalds2017-06-04 17:57 UTC "manual memcpy" and modern compilersTravis2017-06-04 21:11 UTC "manual memcpy" and modern compilersMichael S2017-06-05 11:47 UTC "manual memcpy" and modern compilersLinus Torvalds2017-06-02 16:21 UTC "manual memcpy" and modern compilersYuhong Bao2017-06-03 01:02 UTC "manual memcpy" and modern compilersLinus Torvalds2017-06-03 05:27 UTC "manual memcpy" and modern compilersYuhong Bao2017-06-04 05:26 UTC "manual memcpy" and modern compilersLinus Torvalds2017-06-04 18:12 UTC "manual memcpy" and modern compilersgiovanni deretta2017-06-05 08:22 UTC "manual memcpy" and modern compilersLinus Torvalds2017-06-05 16:49 UTC "manual memcpy" and modern compilersBrett2017-05-31 05:07 UTC "manual memcpy" and modern compilersWilco2017-05-31 09:37 UTC "manual memcpy" and modern compilersBrett2017-06-01 05:28 UTC "manual memcpy" and modern compilersTravis2017-06-01 01:29 UTC "manual memcpy" and modern compilersTravis2017-06-01 01:30 UTC "manual memcpy" and modern compilersWilco2017-06-01 09:06 UTC "manual memcpy" and modern compilersTravis2017-06-01 19:32 UTC "manual memcpy" and modern compilersWilco2017-06-01 20:51 UTC It's all about the length of the memcpy.Travis2017-05-26 00:19 UTC It's all about the length of the memcpy.Michael S2017-05-26 10:07 UTC It's all about the length of the memcpy.Linus Torvalds2017-05-26 21:01 UTC It's all about the length of the memcpy.Linus Torvalds2017-05-26 19:34 UTC It's all about the length of the memcpy.Travis2017-05-27 00:13 UTC It's all about the length of the memcpy.Travis2017-05-27 00:16 UTC It's all about the length of the memcpy.Brett2017-05-27 03:25 UTC It's all about the length of the memcpy.Travis2017-05-27 21:56 UTC It's all about the length of the memcpy.Linus Torvalds2017-05-27 15:50 UTC big.LITTLE ???Michael S2017-05-27 18:09 UTC big.LITTLE ???Linus Torvalds2017-05-27 18:56 UTC may be, Mongoose core ?Michael S2017-05-27 19:43 UTC big.LITTLE ???Travis2017-05-27 22:18 UTC big.LITTLE ???Linus Torvalds2017-05-29 00:18 UTC big.LITTLE ???Travis2017-05-29 04:31 UTC In *theory* this is fixable with better benchmarks ...Mark Roulo2017-05-30 17:22 UTC In *theory* this is fixable with better benchmarks ...Linus Torvalds2017-05-30 18:12 UTC It's all about the length of the memcpy.Travis2017-05-27 21:49 UTC NT stores are an issueHeikki Kultala2017-05-28 06:25 UTC NT stores are an issueTravis2017-05-28 07:38 UTC NT stores are an issue (Ryzen result)octoploid2017-05-28 07:57 UTC NT stores are an issue (Ryzen result)octoploid2017-05-28 07:59 UTC Bogus extra newline when using code,preoctoploid2017-05-28 08:03 UTC Bogus extra newline when using code,preMichael S2017-05-28 08:35 UTC NT stores are an issue (Ryzen result)Travis2017-05-28 08:30 UTC NT stores are an issue (Ryzen result)Travis2017-05-28 08:35 UTC NT stores are an issue (Ryzen result)Michael S2017-05-28 08:45 UTC NT stores are an issue (Ryzen result)Travis2017-05-28 09:20 UTC NT stores are an issue (Ryzen result)Travis2017-05-28 09:22 UTC NT stores are an issue (Ryzen result)octoploid2017-05-28 09:30 UTC NT stores are an issue (Ryzen result)Travis2017-05-28 20:10 UTC It's all about the length of the memcpy.Doug S2017-05-28 15:55 UTC It's all about the length of the memcpy.Gabriele Svelto2017-05-26 22:33 UTC It's all about the length of the memcpy.Travis2017-05-27 01:51 UTC It's all about the length of the memcpy.Seni2017-05-28 22:14 UTC It's all about the length of the memcpy.Travis2017-05-28 22:26 UTC It's all about the length of the memcpy.Gabriele Svelto2017-05-29 12:53 UTC It's all about the length of the memcpy.Travis2017-05-29 21:04 UTC It's all about the length of the memcpy.Seni2017-05-30 00:06 UTC It's all about the length of the memcpy.Travis2017-05-30 02:45 UTC It's all about the length of the memcpy.Brett2017-05-30 04:36 UTC Real code, real data from a real workloadGabriele Svelto2017-05-30 22:59 UTC Real code, real data from a real workloadTravis2017-05-31 03:01 UTC Real code, real data from a real workloadGabriele Svelto2017-05-31 16:31 UTC Real code, real data from a real workloadgallier22017-05-31 17:02 UTC Real code, real data from a real workloadSymmetry2017-05-31 17:17 UTC Real code, real data from a real workloadTravis2017-06-01 01:49 UTC Real code, real data from a real workloadTravis2017-06-01 01:27 UTC Real code, real data from a real workloadMichael S2017-06-01 09:38 UTC Real code, real data from a real workloadWilco2017-06-01 18:06 UTC fixed indeedMichael S2017-06-01 19:23 UTC Real code, real data from a real workloadGabriele Svelto2017-06-02 04:44 UTC Real code, real data from a real workloadTravis2017-06-02 21:38 UTC Real code, real data from a real workloadmeh2017-06-03 13:22 UTC Real code, real data from a real workloadTravis2017-06-03 18:50 UTC Real code, real data from a real workloadSeni2017-06-02 23:34 UTC Real code, real data from a real workloadBrendan2017-06-03 06:09 UTC Real code, real data from a real workloadSeni2017-06-03 10:49 UTC Real code, real data from a real workloadrwessel2017-06-03 18:40 UTC Real code, real data from a real workloadTravis2017-06-03 20:40 UTC Real code, real data from a real workloadTravis2017-06-03 20:20 UTC Real code, real data from a real workloadRicardo B2017-06-04 21:47 UTC Real code, real data from a real workloadTravis2017-06-05 00:15 UTC correctionTravis2017-06-05 00:17 UTC Real code, real data from a real workloadRicardo B2017-06-05 02:03 UTC Real code, real data from a real workloadTravis2017-06-06 19:33 UTC Real code, real data from a real workloadEtienne2017-06-05 10:40 UTC It's all about the length of the memcpy.Megol2017-05-25 15:08 UTC rep movsb is still slowWilco2017-05-25 22:43 UTC 4K is not small... (NT)iz2017-05-26 20:10 UTC Random copies are < 256 bytes (NT)Wilco2017-05-26 21:38 UTC rep movsb is still slowBrendan2017-05-28 02:50 UTC rep movsb is still slowTravis2017-05-28 04:27 UTC Then why use even AVX2 for memcpy?Eric Bron2017-05-24 07:22 UTC Is K12 still alive?Ronald Maas2017-05-24 04:27 UTC Is K12 still alive?dmcq2017-05-24 10:37 UTC Wide registersLaurent2017-05-24 15:53 UTC It's called Amdahl's law (NT)Gabriele Svelto2017-05-25 23:09 UTC Wide registersMichael S2017-05-26 10:24 UTC Wide registersEric Bron2017-05-26 12:47 UTC Ivan Godard (NT)Michael S2017-05-27 18:11 UTC Wide registersLaurent2017-05-26 15:44 UTC Is K12 still alive?dmcq2017-05-23 11:47 UTC Is K12 still alive?juanrga2017-05-23 12:29 UTC the whole post makes no sense at all (NT)Michael S2017-05-23 13:03 UTC did you expect different?blue2017-05-23 15:07 UTC did you expect different?dmcq2017-05-24 10:35 UTC did you expect juanrga post to make sense? (NT) (clarified?)blue2017-05-27 10:44 UTC did you follow the discussion?Michael S2017-05-28 08:30 UTC did you follow the discussion?dmcq2017-05-28 10:05 UTC did you follow the discussion?juanrga2017-05-28 19:24 UTC did you follow the discussion?anon.12017-05-28 20:57 UTC did you follow the discussion?dmcq2017-05-28 22:18 UTC Reply to this TopicName:Email:Topic:Body: No TextHeikki Kultala (heikki.kultala.delete@this.tut.fi) on 2017-05-12 05:34 UTC wrote: > Has the K12 project been cancelled or have just all chips using K12 been delayed > when AMD has been concentrating on getting summit ridge and raven ridge out? > > Explain 🐈🐕: (no spaces, 6 letters, lowercase)
