From kernel@kolivas.org Wed Nov 9 23:51:41 2005 From: Con Kolivas X-KMail-Identity: 593142621 To: linux kernel mailing list Subject: 2.6.14-ck4 (staircase v13) Date: Wed, 9 Nov 2005 23:51:41 +1100 User-Agent: KMail/1.8.3 Cc: ck@vds.kolivas.org MIME-Version: 1.0 Status: RO X-Status: RSC Content-Type: multipart/signed; boundary="nextPart113161070.ImMU2gikkl"; protocol="application/pgp-signature"; micalg=pgp-sha1 Content-Transfer-Encoding: 7bit Message-Id: <200511092351.44024.kernel@kolivas.org> X-KMail-EncryptionState: N X-KMail-SignatureState: N X-KMail-MDN-Sent: --nextPart113161070.ImMU2gikkl Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline These are patches designed to improve system responsiveness and interactivi= ty.=20 It is configurable to any workload but the default ck patch is aimed at the= =20 desktop and cks is available with more emphasis on serverspace. Apply to 2.6.14 http://ck.kolivas.org/patches/2.6/2.6.14/2.6.14-ck4/patch-2.6.14-ck4.bz2 or server version http://ck.kolivas.org/patches/cks/patch-2.6.14-cks4.bz2 web: http://kernel.kolivas.org all patches: http://ck.kolivas.org/patches/ Split patches available. Changes: Added: +patch-2.6.14.1 Latest stable patch +net-fix_zero-size_datagram_reception.patch Bugfix queued for 2.6.14.2 +sched-staircase12.2_13.patch Major update in the staircase code. This now makes the interactive mode wor= k=20 out a simple mathematical calculation to determine the cpu percentage of=20 entitled cpu a task uses and use that to determine what best dynamic priori= ty=20 (and thus how low the latency is). This means that a low cpu using task can= =20 have very low latency even in the presence of much less niced tasks. For=20 example a nice 0 task in the presence of a nice -20 task can still have low= =20 latency if it uses very little cpu. This means that 'nice' is a determinant= =20 of cpu entitlement, but no longer the prime determinant of latency. It is=20 thus possible to even run low cpu using audio apps at nice 19. Running audi= o=20 at nice 19 would be unusual, though. However this does help for the opposit= e=20 as well - there are kernel threads that run nice 0, -5 and even -20, when=20 normal user tasks run at nice 0. Some of these kernel threads and workqueue= s=20 can use extraordinary amounts of cpu as well. These are real world examples= =20 where it would help on normal desktops now. Interbench benchmarks demonstra= te=20 this below. -2614ck3-version.diff +2614ck4-version.diff Version update =2D-- If you are new to interbench results, the order of importance is from right= to=20 left, where deadlines met and desired cpu should be as close to 100% as=20 possible, amd max latency and average latency as low as possible.=20 Asterisks have been placed to the right to indicate where one kernel=20 outperforms the other. Use a fixed font to be able to read this table! Interbench 0.29 results on 3Ghz P4 uniprocessor: Here is mainline 2.6.14, with the benchmarked interactive process running a= t=20 nice 19: =2D-- Benchmarking simulated cpu of Audio nice 19 in the presence of simula= ted Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met None 0.004 +/- 0.00442 0.008 100 100 Video 0.004 +/- 0.00491 0.009 100 100 X 36.4 +/- 62.7 233 96 63.8 Burn 673 +/- 699 701 0.568 0.568 Write 1.39 +/- 9.22 118 99.5 99.5 Read 0.01 +/- 0.0133 0.105 100 100 Compile 774 +/- 803 1003 0.559 0 Memload 4.78 +/- 27.9 225 95.5 95.5 =2D-- Benchmarking simulated cpu of Video nice 19 in the presence of simula= ted=20 Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met None 0.004 +/- 0.00493 0.026 100 100 * X 21.9 +/- 50 266 54.2 48.8 Burn 741 +/- 770 771 0.174 0.174 Write 2.58 +/- 11.2 140 96.1 90.1 Read 0.009 +/- 0.0176 0.182 100 100 Compile 758 +/- 783 877 0.173 0 Memload 2.38 +/- 20.3 269 93.9 92.9 =2D-- Benchmarking simulated cpu of X nice 19 in the presence of simulated = =2D-- Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met None 0.009 +/- 0.1 1 100 99 * Video 13 +/- 23.7 60 69.8 53 Burn 607 +/- 634 750 1.15 1.15 Write 6.48 +/- 17.4 90 80.5 66.7 Read 0.227 +/- 1.05 6 92.7 89.6 Compile 680 +/- 720 957 1.14 1.14 Memload 7.84 +/- 31.7 272 76.6 66.4 Note the massive dropped deadlines even with something using just 5% cpu wh= en=20 run at nice 19 Here is 2.6.14-ck4 running staircase v13: =2D-- Benchmarking simulated cpu of Audio nice 19 in the presence of simula= ted=20 Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met None 0.004 +/- 0.00419 0.009 100 100 Video 0.004 +/- 0.00457 0.008 100 100 X 0.054 +/- 0.388 4 100 100 * Burn 3.27 +/- 18.8 136 98 97.4 * Write 1.3 +/- 11.7 120 99 99 * Read 0.011 +/- 0.0119 0.056 100 100 Compile 2.81 +/- 16.6 126 98 98 * Memload 3.39 +/- 22 206 97.5 97.5 * =2D-- Benchmarking simulated cpu of Video nice 19 in the presence of simula= ted=20 Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met None 0.031 +/- 0.682 16.7 100 99.8 X 9.87 +/- 25.5 133 74.8 67.4 * Burn 6.33 +/- 31.7 483 85.8 79.6 * Write 1.69 +/- 11.7 172 97.1 94.6 * Read 0.011 +/- 0.0142 0.107 100 100 Compile 7.17 +/- 46.4 617 78.3 75.7 * Memload 2.46 +/- 20.8 245 92.9 92.9 =2D-- Benchmarking simulated cpu of X nice 19 in the presence of simulated = =2D-- Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met None 0.237 +/- 2.3 23 100 98 Video 13 +/- 23.6 60 69.8 53 Burn 327 +/- 659 1398 20.5 16.4 * Write 2.9 +/- 13.4 108 73.3 69.9 * Read 0.217 +/- 1.01 6 92.7 89.6 Compile 365 +/- 691 1519 19.3 14.3 * Memload 8.42 +/- 19.8 92 80.8 73.2 * You can see here that the audio benchmark meets more than 97% of the deadli= nes=20 even in extreme loads when run at nice 19, and interbench ignores the effec= t=20 of buffering. With some buffering it is unlikely to skip sound even at nice= =20 19. Now if you move on to video and then X you can see that the improvement= =20 over mainline gets progressively less as each load uses more cpu. This is=20 because the latency is determined by cpu% of entitlement. Even so, the=20 deadlines met are significantly higher (especially for video which doesnt=20 drop below 67% compared to mainline which bottoms out at 1%).=20 =2D-- Moving on to the more "real world" example of nice 0 interactive task in th= e=20 presence of background loads running at nice -5: Mainline 2.6.14: =2D-- Benchmarking simulated cpu of Audio in the presence of simulated nice= -5=20 Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met None 0.004 +/- 0.00526 0.029 100 100 Video 0.004 +/- 0.00484 0.007 100 100 X 36 +/- 59.8 113 75.5 63.3 Burn 6.78 +/- 90.5 1200 88.1 88 Write 2.21 +/- 13.5 141 99 99 Read 0.01 +/- 0.0146 0.111 100 100 Compile 22.1 +/- 183 1600 69.7 69.3 Memload 0.191 +/- 0.957 7.07 100 100 =2D-- Benchmarking simulated cpu of Video in the presence of simulated nice= -5=20 Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met None 0.004 +/- 0.00415 0.01 100 100 * X 21.4 +/- 51.1 417 53.7 48.2 Burn 23.6 +/- 186 1467 41.4 41 Write 3.11 +/- 16.9 315 93.7 88.8 Read 0.009 +/- 0.0145 0.12 100 100 Compile 21.5 +/- 171 1400 43.7 43.1 Memload 0.292 +/- 2.08 28.3 100 99 =2D-- Benchmarking simulated cpu of X in the presence of simulated nice -5 = =2D-- Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met None 0.009 +/- 0.0995 1 100 99 Video 13 +/- 23.7 60 69.8 53 * Burn 77.5 +/- 306 1404 45.6 43.8 Write 9.71 +/- 28.4 224 76.2 64.9 Read 0.245 +/- 1.13 6 91.1 88.9 Compile 97.6 +/- 365 1564 41.1 39.2 Memload 4.31 +/- 12.7 72 57.6 51.2 You can see that without any buffering, it is possible even at nice 0 for=20 audio to miss almost 40% of the deadlines under a nice -5 X load (or=20 equivalent load from kernel threads etc). And here is 2.6.14-ck4: =2D-- Benchmarking simulated cpu of Audio in the presence of simulated nice= -5=20 Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met None 0.003 +/- 0.00351 0.009 100 100 Video 0.004 +/- 0.00451 0.009 100 100 X 0.287 +/- 1.43 10 100 100 * Burn 0.004 +/- 0.00457 0.008 100 100 * Write 0.009 +/- 0.0102 0.018 100 100 * Read 0.011 +/- 0.0128 0.069 100 100 Compile 0.009 +/- 0.0107 0.039 100 100 * Memload 0.037 +/- 0.167 2 100 100 =2D-- Benchmarking simulated cpu of Video in the presence of simulated nice= -5=20 Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met None 0.031 +/- 0.682 16.7 100 99.8 X 3.43 +/- 8.46 27.7 100 83 * Burn 0.032 +/- 0.681 16.7 100 99.8 * Write 0.006 +/- 0.00846 0.07 100 100 * Read 0.011 +/- 0.0115 0.04 100 100 Compile 0.04 +/- 0.683 16.7 100 99.8 * Memload 0.062 +/- 0.694 16.7 100 99.8 * =2D-- Benchmarking simulated cpu of X in the presence of simulated nice -5 = =2D-- Load Latency +/- SD (ms) Max Latency % Desired CPU % Deadlines Met None 0.009 +/- 0.0995 1 100 99 Video 13.5 +/- 24.1 60 67.7 50.8 Burn 0.117 +/- 0.783 6 98.1 96.1 * Write 0.147 +/- 0.751 4 94.4 92.4 * Read 0.009 +/- 0.0995 1 100 99 * Compile 0.147 +/- 0.751 4 94.4 92.4 * Memload 0.176 +/- 1.69 17 100 99 * Whereas here audio does not drop a single deadline even without any bufferi= ng.=20 Video and even X also show significant improvement. Cheers, Con --nextPart113161070.ImMU2gikkl Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.1 (GNU/Linux) iD8DBQBDcfDfZUg7+tp6mRURAvLmAJ94OpEFyS3Ths3kdYp6MIoygXP14QCfdqAg PS4YrBej9AfKNPHrhsQ/Blk= =K+e+ -----END PGP SIGNATURE----- --nextPart113161070.ImMU2gikkl--