API tools faq
paste
Guest User

Untitled

a guest
Feb 12th, 2014
575
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 30.94 KB | None | 0 0
raw download clone embed print report
  1.  
  2. -----------------------------------------------------------------
  3. /storage1/Android/cyanogen_mod_bundle11.0/cm-11.0/bionic e2c1351dea669ca2fc49164d386373390aa242d9..f5185cc4f1a0eba96b6fc6dc73c429e5c7dfbb89
  4. commit f5185cc4f1a0eba96b6fc6dc73c429e5c7dfbb89
  5. Author: Dirk Rettschlag <[email protected]>
  6. Date: Tue Feb 11 15:23:42 2014 +0100
  7.  
  8. libm: use neon optimized sin/cos/pow/sqrt/sqrtf
  9.  
  10. This enables the optimized functions for all devices that
  11. support the NEON instruction set.
  12.  
  13. benchmark iterations Normal ns/op NEON ns/op
  14. BM_math_sin 10000000 126 109
  15. BM_math_cos 10000000 112 114
  16. BM_math_pow 5000000 375 328
  17. BM_math_sqrt 50000000 496 38
  18. BM_math_sqrtf 100000000 136 25
  19.  
  20. Change-Id: I2b2ee9ab1769c994edac4a5e56eee2fafa4a37a7
  21. Signed-off-by: Dirk Rettschlag <[email protected]>
  22. -----------------------------------------------------------------
  23. /storage1/Android/cyanogen_mod_bundle11.0/cm-11.0/build 038b7b47a43412c2942c3b316ca130465007bcb1..2c165e5e893828bf1a1fc83c2e6e228cfe93dd90
  24. commit 2c165e5e893828bf1a1fc83c2e6e228cfe93dd90
  25. Author: dhacker29 <[email protected]>
  26. Date: Tue Jan 28 15:51:51 2014 -0600
  27.  
  28. Allow devices to specify certain ro. props via TARGET_UNIFIED_DEVICE
  29.  
  30. This modifies buildinfo.sh to not set model, name, description, and
  31. fingerprint in build.prop. These can then be set via custom init
  32. lib.
  33. This also fixes ota_from_target_files to function without these
  34. props being preset.
  35.  
  36. Change-Id: I945b2c38e60b207d2d0f82b34f4f230ac21b0657
  37. -----------------------------------------------------------------
  38. /storage1/Android/cyanogen_mod_bundle11.0/cm-11.0/frameworks/av 0f4d10e7ea9aa431542a10d73fea372ca081e5c1..c6791bd67e0740c02adfcc3b75f6a22f33ed4d9d
  39. commit c6791bd67e0740c02adfcc3b75f6a22f33ed4d9d
  40. Author: Haynes Mathew George <[email protected]>
  41. Date: Mon Jan 27 11:11:19 2014 -0800
  42.  
  43. audio: Use composition instead of multiple inheritance
  44.  
  45. - AudioTrack is derived from BnDirectTrackClient along
  46. with RefBase to receive the notifications from
  47. AudioFlinger in tunnel/LPA mode. This is breaking
  48. binary compatibility for the clients which use native
  49. AudioTrack directly instead of Java based AudioTrack
  50. API
  51. - Fix is to use composition here instead of inheritance
  52.  
  53. Change-Id: I9fd2c61cff5629924e3b08f63559679e0263bcfd
  54.  
  55. commit a50273c732d068fd91438c0b3348f0944517e947
  56. Author: Pavan Chikkala <[email protected]>
  57. Date: Wed Dec 18 15:06:43 2013 +0530
  58.  
  59. audio: Check the return code of MediaWriter::start
  60.  
  61. - Return code of MediaWriter::start method is not checked
  62. in startRawAudioRecording. Due to this, sound recorder
  63. application is not receiving the error whenever there is
  64. an error in starting of the recording
  65. - Fix is to check the error and return it back to
  66. the client accordingly
  67.  
  68. Change-Id: I15916211da7a2eb0f27fbf9d4151410a8bdfdd42
  69. CRs-Fixed: 591788
  70.  
  71. commit 616521f189409a5d5c7cbd8e2fc822f09b45fc98
  72. Author: Naresh Tanniru <[email protected]>
  73. Date: Mon Dec 16 21:21:10 2013 +0530
  74.  
  75. Audio: Tunnel Exception for ER AAC ELD profile format
  76.  
  77. - some m4a clips are not working properly
  78.  
  79. - ADSP audio decoders don't support ER AAC ELD support
  80.  
  81. - Added tunnel exception for ER AAC ELD format.
  82. Added AACProfile metadata support for MKVExtractor
  83.  
  84. Change-Id: I2f6fe721b05377dfe0f24ea0ad1f6c4a885d576d
  85.  
  86. commit 7be41518c3b60399cade17783497784f65268fb9
  87. Author: Nitesh Bhardwaj <[email protected]>
  88. Date: Fri Dec 13 11:24:37 2013 +0530
  89.  
  90. audioflinger: Add Null pointer check while enabling effects for LPA
  91.  
  92. - MediaServer is crashing when disabling effects after mediaplayer
  93. object is released.
  94. - EffectChain pointer is de-referenced without NULL check, hence
  95. leading to a mediaserver crash.
  96. - Adding a NULL pointer check before using EffectChain object.
  97.  
  98. Change-Id: If5ff469445246028046ba65e5ddf3300180318c2
  99. CRs-Fixed: 583305
  100.  
  101. commit 5668f2403e055feadb8a89b5277bb1c6ae5d85f5
  102. Author: Phani Kumar Allada <[email protected]>
  103. Date: Thu Sep 27 18:21:54 2012 +0530
  104.  
  105. framework/av: Fix for frame drop in video playback
  106.  
  107. - Breaks in video playback seen in video clips having audio content
  108. with sampling rates less than 16kHz.
  109. - While extrapolating real time with system time, the real time is wrongly
  110. caped.
  111. - Fix this issue by limiting the extrapolated real time only when EOS
  112. is reached.
  113.  
  114. Change-Id: I95b6a8f7f1f1fd561c92955cf52151ae2890c9a7
  115. CRs-Fixed: 400398
  116.  
  117. commit 8bcbb487b7de4c2aaae95d461a39b64f0ea9f590
  118. Author: wjiang <[email protected]>
  119. Date: Tue Oct 29 22:22:16 2013 +0800
  120.  
  121. audio: Supress redundant effect config events in short duration
  122.  
  123. Distortion sound persists for the whole song when re-enabling effect in
  124. LPA playback mode till switch to next song. And pop noise is quite obvious
  125. when enable/disable effects in LPA.
  126.  
  127. As the design of LPA, EffectsThreadEntry() loop is used to handle effect
  128. config changes through applying effects with new config on dsp pcm buffers.
  129. However, it can't handle too many config change events in a short time,
  130. and effect with new configuration will be applied on part of dsp buffers.
  131.  
  132. Filter out redundant effect config events that come too closely to avoid
  133. applying effect on dsp buffer in vain.
  134.  
  135. Change-Id: If8915b1d068724d5976e1519ec57e3c08cf89c9d
  136. CRs-Fixed: 562806, 563418
  137.  
  138. commit 9fc34ac64181a9c1745c6a392bb02285f2a6a047
  139. Author: Sharad Sangle <[email protected]>
  140. Date: Tue Nov 5 18:08:24 2013 +0530
  141.  
  142. audioflinger: SRS effects are not working for LPA
  143.  
  144. -Issue: If SRS effects are applied on LPA playback,
  145. in ARM side there is a delay of 4-5 seconds for effects
  146. application
  147. -Root cause: In applyEffects SRS effects are applied first and
  148. then Android effects are applied.
  149. In case of active android effects, SRS effects are called
  150. first and then android effects are called, so buffer with
  151. SRS effects is getting overwritten with buffer with
  152. android effects.
  153. If there are no android effects applied then all
  154. the SRS effects are overwritten with decoded data.
  155. So until next write call, SRS effects are not applied and hence delay
  156. of 4-5 seconds
  157. -Fix: In applyEffects first apply android effects and then
  158. apply SRS effects
  159.  
  160. Change-Id: I36b2133256d168f06b13c3d4a2bad00312d571ac
  161. CRs-Fixed: 566278
  162.  
  163. commit 1b37c78b7709e6958885e3d2fdeaab45ba64d6dd
  164. Author: Eric Laurent <[email protected]>
  165. Date: Fri Nov 15 12:02:28 2013 -0800
  166.  
  167. audioflinger: do not use raw pointer for tracks
  168.  
  169. Commit 9da3d95 surfaced a problem caused by the use of a raw
  170. pointer to a track in offload thread implementation.
  171.  
  172. Pointers to tracks should always be weak or strong pointers.
  173.  
  174. Bug: 11708529.
  175. Change-Id: Ic48632532d186c9be8261f73cefdf824b9fbbd2b
  176.  
  177. commit 2d844ce6dc3b3558dabe67bd0607042eb0685bb8
  178. Author: Shashank Mittal <[email protected]>
  179. Date: Tue Sep 3 10:58:10 2013 -0700
  180.  
  181. media: Add Ops check before starting audio recording.
  182.  
  183. Confirm user permission before starting audio recording.
  184.  
  185. Change-Id: Ia4e6d8fb6986d4e73fb656ba0efaf42c210e92af
  186.  
  187. commit 4dfe626b6f3e4b80e71c01fc39d9e3c9b16b0ef0
  188. Author: wjiang <[email protected]>
  189. Date: Wed Oct 16 14:41:54 2013 +0800
  190.  
  191. framework/av: Check for NULL pointer of mCblk and zero division error
  192.  
  193. latency calculation happens in two places: AudioTrack::latency() and
  194. AudioTrack::dump(). Need to add null pointer and zero division check on
  195. both of these places.
  196.  
  197. Change-Id: I9cbe8c1c00a390cd626bdbf202124f27e60eb3e3
  198. Fixed-CRs: 548514
  199.  
  200. commit af8b665c1ab2eeb1d0e964684043590522082f3d
  201. Author: Divya Narayanan Poojary <[email protected]>
  202. Date: Fri Nov 15 15:16:48 2013 +0530
  203.  
  204. frameworks/av: Mediacodec initialization check in QCUtils
  205.  
  206. - Mediaserver crash is observed while playing wma10pro clips
  207. on MM stability stress
  208.  
  209. - 8x10 device is not supporting wma10pro format from openmax
  210. which is resulting component creation fail.
  211. However in this case multiaac decoder component is creating
  212. which is resulting crash
  213.  
  214. - On QCUtils::helper_addMediaCodec without intialzing the codec
  215. we are trying to add codec capabilities which is resulting wrong
  216. codec capabilities to assign for multiaac decoder
  217. Added initializarion & flags check to fix the issue
  218.  
  219. Change-Id: I8a61d658bcbebf53a59a30bd6d812b4dafec46c6
  220. CRs-Fixed: 552817
  221. -----------------------------------------------------------------
  222. /storage1/Android/cyanogen_mod_bundle11.0/cm-11.0/frameworks/base d6a6cd6c6fb8ed5fa2c5363881f72a907ce2109c..5127fbb6e37377e121814df20eb5f3ab3f55cda8
  223. commit 5127fbb6e37377e121814df20eb5f3ab3f55cda8
  224. Author: Daniele Di Conza <[email protected]>
  225. Date: Thu Jan 16 10:37:41 2014 +0100
  226.  
  227. SUPL early data connection fix
  228.  
  229. if GPS started before data connection is alive, SUPL will be broken for entire power cycle.
  230.  
  231. Change-Id: I0486ac5e62a2d4a7da14506e3dd8e78da42f405d
  232.  
  233. commit 9f5f243b433010d6d9910dcf52c396d7e5bad7d3
  234. Author: Nilesh Poddar <[email protected]>
  235. Date: Fri Jan 17 11:13:17 2014 -0800
  236.  
  237. Release wakelocks acquired by SyncManager
  238.  
  239. Release the wakelocks internally acquired by SyncManager to improve
  240. power performance.
  241.  
  242. Change-Id: I6d9663292cb4471d92ed9132beb7ed1ba2e8128f
  243. CRs-Fixed: 585825
  244.  
  245. commit b8db689c03949b67b5760263606f7bd53c526531
  246. Author: Guntawee Tiwapong <[email protected]>
  247. Date: Thu Dec 12 12:29:23 2013 -0800
  248.  
  249. server: Handle broadcast intent and receiver
  250.  
  251. Ensure the receivers of broadcast intent is not launched before boot
  252. completion by adding an intent flag, since connectivity events can
  253. happen before boot has completed.
  254.  
  255. Change-Id: Id6fc9e34c89d947d768def38e5c5bb0df54a3a7b
  256.  
  257. commit d3132b3227030f614f3f1cdcad3fe54976a75e74
  258. Author: Jorge Ruesga <[email protected]>
  259. Date: Wed Feb 12 02:05:08 2014 +0100
  260.  
  261. base: fix battery stats
  262.  
  263. * Fix IBatteryStats.aidl posititions. BatteryService.cpp from hardware sensors uses a position token
  264. from the binder. The dock battery patch broke this order, causing sensor stats to be send to
  265. wakelock stats, causing a breakage in the partial wakelock counter.
  266. * Create a separate dock stats implementation to avoid collision in the internal statics variables
  267. of the normal stats implementation.
  268.  
  269. Change-Id: I22c92d5941804449668530715bf5381b00164593
  270. JIRA: CYAN-3127
  271. Issue: https://jira.cyanogenmod.org/browse/CYAN-3127
  272. Signed-off-by: Jorge Ruesga <[email protected]>
  273. -----------------------------------------------------------------
  274. /storage1/Android/cyanogen_mod_bundle11.0/cm-11.0/hardware/libhardware_legacy 107fcb3eeb34b5d4c4a7216dc712a107623533ab..7a133590ff567fe09b7f415371c78480d39a83e7
  275. commit 7a133590ff567fe09b7f415371c78480d39a83e7
  276. Author: Steve Kondik <[email protected]>
  277. Date: Sat Jan 25 22:12:58 2014 -0800
  278.  
  279. audio: Fix check for sonification on digital docks
  280.  
  281. Change-Id: If077bc6336ce57256dd078d0cbe80c3b5e3a0b6e
  282. -----------------------------------------------------------------
  283. /storage1/Android/cyanogen_mod_bundle11.0/cm-11.0/hardware/ril 6ca77e7b9c8c306ab80e574d6d6bacb341760b71..681cd4de0b999eee8fc8d9c52c69ee46c810168d
  284. commit 681cd4de0b999eee8fc8d9c52c69ee46c810168d
  285. Author: Alex Yakavenka <[email protected]>
  286. Date: Wed Dec 4 13:54:37 2013 -0800
  287.  
  288. Send unsol_sim_status on radio_state change
  289.  
  290. Sim status changes based on radio state.
  291. Need to notify new sim state whenever radio state changes
  292.  
  293. Change-Id: I8fef3a6ddf601647107e3059f2b2eb89a67242cf
  294.  
  295. commit 6e803ab428b72f0ad61272b12d9e6104e39dbdf5
  296. Author: Colin Cross <[email protected]>
  297. Date: Fri Jan 24 17:02:19 2014 -0800
  298.  
  299. Set USE_NP based on HAVE_PTHREAD_COND_TIMEDWAIT_RELATIVE
  300.  
  301. 64 bit bionic does not define pthread_cond_timeout_np, use
  302. HAVE_PTHREAD_COND_TIMEDWAIT_RELATIVE to determine whether or
  303. not to use it.
  304.  
  305. Change-Id: Ia94dd76f7ea4b0268f07049c127342a8bd016ec0
  306. -----------------------------------------------------------------
  307. /storage1/Android/cyanogen_mod_bundle11.0/cm-11.0/hardware/ti/wlan 352b51297be9bcf268f9f66eaf5c323cc11327b7..cb0a5534c1cb95772cbd770849198fa68e21c7f3
  308. commit cb0a5534c1cb95772cbd770849198fa68e21c7f3
  309. Author: Michael Bestas <[email protected]>
  310. Date: Wed Nov 20 08:22:06 2013 +0200
  311.  
  312. Use AOSP version of wpa_supplicant_lib
  313.  
  314. Squashed commit of the following changes:
  315.  
  316. Update make file for new location of libnl
  317.  
  318. Change-Id: If93aea9913cfa263fb228feb27137ac16b78cf0a
  319. Signed-off-by: Mike Lockwood <[email protected]>
  320.  
  321. wpa_supplicant_lib: Use common file from hardware_legacy
  322.  
  323. Change-Id: Ibbf60918848ab02dbc1b985005f986ab63d4dc52
  324. Signed-off-by: Dmitry Shmidt <[email protected]>
  325.  
  326. wpa_supplicant_lib: Clean the code and fix the build
  327.  
  328. Change-Id: I6a916c2f9306a49951dbba7bda5793d389899f91
  329. Signed-off-by: Dmitry Shmidt <[email protected]>
  330.  
  331. Also cleans up the files to bring them inline with AOSP
  332.  
  333. Change-Id: I4a1dc7b25abda0254c18677e7e9a6651a920b05f
  334. -----------------------------------------------------------------
  335. /storage1/Android/cyanogen_mod_bundle11.0/cm-11.0/kernel/archos/archos_g9 eb25864ce086ac67dc3e8fb72b9350baed6b7bb3..39bf0849c05d6093897e1b079e2ca0c47dc0bd54
  336. commit 39bf0849c05d6093897e1b079e2ca0c47dc0bd54
  337. Author: Michael Gernoth <[email protected]>
  338. Date: Wed Jan 22 00:01:14 2014 +0100
  339.  
  340. sync struct wl12xx_platform_data to hardware/ti/wlan
  341.  
  342. this stops the wlan-driver from accesing memory outside
  343. of its platform-data
  344.  
  345. Change-Id: I6e0f1920d3f8528cef3528d4989c9626507aa422
  346.  
  347. commit 45634ec9230e9a16790e5c1a8702e785c38cf81f
  348. Author: Matt Wagantall <[email protected]>
  349. Date: Wed Jul 17 06:53:08 2013 +0000
  350.  
  351. net: flow: Prevent bringing up new CPUs during per-CPU initialization
  352.  
  353. In a rare race-condition, it's possible that a CPU will be brought
  354. online between when the for_each_online_cpu() loop executes to
  355. call flow_cache_cpu_prepare(), and when the hotplug notifier for
  356. calling this same function is registered. If this happens,
  357. flow_cache_cpu_prepare() will never be called for the new CPU,
  358. resulting in crashes due to uninitialized per-cpu data.
  359.  
  360. Fix this by preventing CPUs from being added or removed during
  361. this small but sensitive window.
  362.  
  363. Change-Id: Iafbbaa8a50e5c527392d130561874313720849d0
  364. Signed-off-by: Matt Wagantall <[email protected]>
  365.  
  366. Conflicts:
  367. net/core/flow.c
  368.  
  369. commit 43bb6f26e71347c6cd4d5e832b9ba2b6d7ce0ebf
  370. Author: Junxiao Bi <[email protected]>
  371. Date: Wed Sep 11 14:23:04 2013 -0700
  372.  
  373. writeback: fix race that cause writeback hung
  374.  
  375. There is a race between mark inode dirty and writeback thread, see the
  376. following scenario. In this case, writeback thread will not run though
  377. there is dirty_io.
  378.  
  379. __mark_inode_dirty() bdi_writeback_workfn()
  380. ... ...
  381. spin_lock(&inode->i_lock);
  382. ...
  383. if (bdi_cap_writeback_dirty(bdi)) {
  384. <<< assume wb has dirty_io, so wakeup_bdi is false.
  385. <<< the following inode_dirty also have wakeup_bdi false.
  386. if (!wb_has_dirty_io(&bdi->wb))
  387. wakeup_bdi = true;
  388. }
  389. spin_unlock(&inode->i_lock);
  390. <<< assume last dirty_io is removed here.
  391. pages_written = wb_do_writeback(wb);
  392. ...
  393. <<< work_list empty and wb has no dirty_io,
  394. <<< delayed_work will not be queued.
  395. if (!list_empty(&bdi->work_list) ||
  396. (wb_has_dirty_io(wb) && dirty_writeback_interval))
  397. queue_delayed_work(bdi_wq, &wb->dwork,
  398. msecs_to_jiffies(dirty_writeback_interval * 10));
  399. spin_lock(&bdi->wb.list_lock);
  400. inode->dirtied_when = jiffies;
  401. <<< new dirty_io is added.
  402. list_move(&inode->i_wb_list, &bdi->wb.b_dirty);
  403. spin_unlock(&bdi->wb.list_lock);
  404.  
  405. <<< though there is dirty_io, but wakeup_bdi is false,
  406. <<< so writeback thread will not be waked up and
  407. <<< the new dirty_io will not be flushed.
  408. if (wakeup_bdi)
  409. bdi_wakeup_thread_delayed(bdi);
  410.  
  411. Writeback will run until there is a new flush work queued. This may cause
  412. a lot of dirty pages stay in memory for a long time.
  413.  
  414. Signed-off-by: Junxiao Bi <[email protected]>
  415. Reviewed-by: Jan Kara <[email protected]>
  416. Cc: Fengguang Wu <[email protected]>
  417. Signed-off-by: Andrew Morton <[email protected]>
  418. Signed-off-by: Linus Torvalds <[email protected]>
  419. Signed-off-by: Francisco Franco <[email protected]>
  420.  
  421. Change-Id: I973fcba5381881a003a035ffff48f64348660079
  422. Signed-off-by: Quallenauge <[email protected]>
  423.  
  424. Conflicts:
  425. fs/fs-writeback.c
  426.  
  427. commit 48c726a266853d47454a7f963410d2cc53ae6861
  428. Author: Felix Fietkau <[email protected]>
  429. Date: Fri May 4 21:08:33 2012 -0700
  430.  
  431. timer: optimize apply_slack()
  432.  
  433. __fls(mask) is equivalent to find_last_bit(&mask, BITS_PER_LONG), but cheaper.
  434. find_last_bit was showing up high on the list when I was profiling for stalls
  435. on icache misses on a system with very small cache size (MIPS).
  436.  
  437. Signed-off-by: Felix Fietkau <[email protected]>
  438. Signed-off-by: edoko <[email protected]>
  439.  
  440. Change-Id: I8a5021a2fb2936c00ffd456663a76cb1b23e3100
  441.  
  442. commit 8260d44edd59ee04445d20405dc45c3ae28128f9
  443. Author: Jan Kara <[email protected]>
  444. Date: Fri Jul 12 17:30:07 2013 +0200
  445.  
  446. writeback: Fix occasional slow sync(1)
  447.  
  448. In case when system contains no dirty pages, wakeup_flusher_threads()
  449. will submit WB_SYNC_NONE writeback for 0 pages so wb_writeback() exits
  450. immediately without doing anything. Thus sync(1) will write all the
  451. dirty inodes from a WB_SYNC_ALL writeback pass which is slow.
  452.  
  453. Fix the problem by using get_nr_dirty_pages() in
  454. wakeup_flusher_threads() instead of calculating number of dirty pages
  455. manually. That function also takes number of dirty inodes into account.
  456.  
  457. Change-Id: I458027ae08d9a5a93202a7b97ace1f8da7a18a07
  458. CC: [email protected]
  459. Reported-by: Paul Taysom <[email protected]>
  460. Signed-off-by: Jan Kara <[email protected]>
  461. Signed-off-by: Quallenauge <[email protected]>
  462.  
  463. commit 3ddc205c9aa0c3c19e26611923f516aee8e6cfc7
  464. Author: Namjae Jeon <[email protected]>
  465. Date: Wed Jan 2 20:00:40 2013 -0600
  466.  
  467. writeback: fix writeback cache thrashing
  468.  
  469. From: Namjae Jeon <[email protected]>
  470.  
  471. Consider Process A: huge I/O on sda
  472. doing heavy write operation - dirty memory becomes more
  473. than dirty_background_ratio
  474. on HDD - flusher thread flush-8:0
  475.  
  476. Consider Process B: small I/O on sdb
  477. doing while [1]; read 1024K + rewrite 1024K + sleep 2sec
  478. on Flash device - flusher thread flush-8:16
  479.  
  480. As Process A is a heavy dirtier, dirty memory becomes more
  481. than dirty_background_thresh. Due to this, below check becomes
  482. true(checking global_page_state in over_bground_thresh)
  483. for all bdi devices(even for very small dirtied bdi - sdb):
  484.  
  485. In this case, even small cached data on 'sdb' is forced to flush
  486. and writeback cache thrashing happens.
  487.  
  488. When we added debug prints inside above 'if' condition and ran
  489. above Process A(heavy dirtier on bdi with flush-8:0) and
  490. Process B(1024K frequent read/rewrite on bdi with flush-8:16)
  491. we got below prints:
  492.  
  493. [Test setup: ARM dual core CPU, 512 MB RAM]
  494.  
  495. [over_bground_thresh]: wakeup flush-8:0 : BDI_RECLAIMABLE = 56064 KB
  496. [over_bground_thresh]: wakeup flush-8:0 : BDI_RECLAIMABLE = 56704 KB
  497. [over_bground_thresh]: wakeup flush-8:0 : BDI_RECLAIMABLE = 84720 KB
  498. [over_bground_thresh]: wakeup flush-8:0 : BDI_RECLAIMABLE = 94720 KB
  499. [over_bground_thresh]: wakeup flush-8:16 : BDI_RECLAIMABLE = 384 KB
  500. [over_bground_thresh]: wakeup flush-8:16 : BDI_RECLAIMABLE = 960 KB
  501. [over_bground_thresh]: wakeup flush-8:16 : BDI_RECLAIMABLE = 64 KB
  502. [over_bground_thresh]: wakeup flush-8:0 : BDI_RECLAIMABLE = 92160 KB
  503. [over_bground_thresh]: wakeup flush-8:16 : BDI_RECLAIMABLE = 256 KB
  504. [over_bground_thresh]: wakeup flush-8:16 : BDI_RECLAIMABLE = 768 KB
  505. [over_bground_thresh]: wakeup flush-8:16 : BDI_RECLAIMABLE = 64 KB
  506. [over_bground_thresh]: wakeup flush-8:16 : BDI_RECLAIMABLE = 256 KB
  507. [over_bground_thresh]: wakeup flush-8:16 : BDI_RECLAIMABLE = 320 KB
  508. [over_bground_thresh]: wakeup flush-8:16 : BDI_RECLAIMABLE = 0 KB
  509. [over_bground_thresh]: wakeup flush-8:0 : BDI_RECLAIMABLE = 92032 KB
  510. [over_bground_thresh]: wakeup flush-8:0 : BDI_RECLAIMABLE = 91968 KB
  511. [over_bground_thresh]: wakeup flush-8:16 : BDI_RECLAIMABLE = 192 KB
  512. [over_bground_thresh]: wakeup flush-8:16 : BDI_RECLAIMABLE = 1024 KB
  513. [over_bground_thresh]: wakeup flush-8:16 : BDI_RECLAIMABLE = 64 KB
  514. [over_bground_thresh]: wakeup flush-8:16 : BDI_RECLAIMABLE = 192 KB
  515. [over_bground_thresh]: wakeup flush-8:16 : BDI_RECLAIMABLE = 576 KB
  516. [over_bground_thresh]: wakeup flush-8:16 : BDI_RECLAIMABLE = 0 KB
  517. [over_bground_thresh]: wakeup flush-8:0 : BDI_RECLAIMABLE = 84352 KB
  518. [over_bground_thresh]: wakeup flush-8:16 : BDI_RECLAIMABLE = 192 KB
  519. [over_bground_thresh]: wakeup flush-8:16 : BDI_RECLAIMABLE = 512 KB
  520. [over_bground_thresh]: wakeup flush-8:16 : BDI_RECLAIMABLE = 0 KB
  521. [over_bground_thresh]: wakeup flush-8:0 : BDI_RECLAIMABLE = 92608 KB
  522. [over_bground_thresh]: wakeup flush-8:0 : BDI_RECLAIMABLE = 92544 KB
  523.  
  524. As mentioned in above log, when global dirty memory > global background_thresh
  525. small cached data is also forced to flush by flush-8:16.
  526. If removing global background_thresh checking code, we can reduce cache
  527. thrashing of frequently used small data.
  528. And It will be great if we can reserve a portion of writeback cache using
  529. min_ratio.
  530.  
  531. After applying patch:
  532. $ echo 5 > /sys/block/sdb/bdi/min_ratio
  533. $ cat /sys/block/sdb/bdi/min_ratio
  534. 5
  535. [over_bground_thresh]: wakeup flush-8:0 : BDI_RECLAIMABLE = 56064 KB
  536. [over_bground_thresh]: wakeup flush-8:0 : BDI_RECLAIMABLE = 56704 KB
  537. [over_bground_thresh]: wakeup flush-8:0 : BDI_RECLAIMABLE = 84160 KB
  538. [over_bground_thresh]: wakeup flush-8:0 : BDI_RECLAIMABLE = 96960 KB
  539. [over_bground_thresh]: wakeup flush-8:0 : BDI_RECLAIMABLE = 94080 KB
  540. [over_bground_thresh]: wakeup flush-8:0 : BDI_RECLAIMABLE = 93120 KB
  541. [over_bground_thresh]: wakeup flush-8:0 : BDI_RECLAIMABLE = 93120 KB
  542. [over_bground_thresh]: wakeup flush-8:0 : BDI_RECLAIMABLE = 91520 KB
  543. [over_bground_thresh]: wakeup flush-8:0 : BDI_RECLAIMABLE = 89600 KB
  544. [over_bground_thresh]: wakeup flush-8:0 : BDI_RECLAIMABLE = 93696 KB
  545. [over_bground_thresh]: wakeup flush-8:0 : BDI_RECLAIMABLE = 93696 KB
  546. [over_bground_thresh]: wakeup flush-8:0 : BDI_RECLAIMABLE = 72960 KB
  547. [over_bground_thresh]: wakeup flush-8:0 : BDI_RECLAIMABLE = 90624 KB
  548. [over_bground_thresh]: wakeup flush-8:0 : BDI_RECLAIMABLE = 90624 KB
  549. [over_bground_thresh]: wakeup flush-8:0 : BDI_RECLAIMABLE = 90688 KB
  550.  
  551. As mentioned in the above logs, once cache is reserved for Process B,
  552. and patch is applied there is less writeback cache thrashing on sdb
  553. by frequent forced writeback by flush-8:16 in over_bground_thresh.
  554.  
  555. After all, small cached data will be flushed by periodic writeback
  556. once every dirty_writeback_interval.
  557.  
  558. Change-Id: I678b30a5a28ed99bed0e782e38926ba4626b86ee
  559. Suggested-by: Wanpeng Li <[email protected]>
  560. Signed-off-by: Namjae Jeon <[email protected]>
  561. Signed-off-by: Vivek Trivedi <[email protected]>
  562. Signed-off-by: Quallenauge <[email protected]>
  563.  
  564. Conflicts:
  565. fs/fs-writeback.c
  566.  
  567. commit 25d434fb1ef7f64dde1de297f36eef680a364d7e
  568. Author: Quallenauge <[email protected]>
  569. Date: Tue Feb 11 23:59:13 2014 +0100
  570.  
  571. Let the twl driver know, that there is a specific archos battery driver available.
  572. -----------------------------------------------------------------
  573. /storage1/Android/cyanogen_mod_bundle11.0/cm-11.0/packages/apps/Camera2 42067bbce2203088e09039169b0262691dd07e97..33a19d784bc8b051ccef05ce56a85d7ca1088b9c
  574. commit 33a19d784bc8b051ccef05ce56a85d7ca1088b9c
  575. Author: Lars Greiss <[email protected]>
  576. Date: Sun Feb 2 03:09:22 2014 +0100
  577.  
  578. Camera2: tweak volume key zoom and cleanup
  579.  
  580. - cleanup unused code
  581. - cleanup codestyle
  582. - add more zoom steps for smother zooming
  583. - show zoom circle indicator during zoom
  584.  
  585. Change-Id: I5214fc2620a0c5167bfdc8e85f7115966c3b8bcf
  586.  
  587. commit 942890667aa08cc3c1305d7ba64c1340df1219db
  588. Author: Lars Greiss <[email protected]>
  589. Date: Sat Feb 1 05:16:28 2014 +0100
  590.  
  591. Camera2: add option to prevend autofocus cancel if preview stopped
  592.  
  593. There are camera HALs which cannot handle the autofocus cancel call if the preview
  594. stopped in startPreview(). Add an option for this HAL's to do not cancel autofocus in this cases.
  595.  
  596. Change-Id: I0767cc7279405d17afc0b00120cb43e5bb309220
  597.  
  598. commit 8c66953d26640d205a974d1f100a1cea98057326
  599. Author: Lars Greiss <[email protected]>
  600. Date: Wed Jan 15 05:08:26 2014 +0100
  601.  
  602. Camera2: fix possible NPE on configurationchange
  603.  
  604. Step to reproduce:
  605. - open camera from lockscreen
  606. - during it opens turn your phone
  607. - result: crash
  608.  
  609. Due that the camera is in middle of the initialization mParameters
  610. is null at this point. Just check against null to prevend the crash.
  611. The correct resizing happens so or so during the ongoing initialization
  612.  
  613. Thanks TJKV from xda for the report
  614.  
  615. Change-Id: I0aab385172d7ff571adf525004305fd3b34d499c
  616.  
  617. commit 2fae069ed775750fde89f6556dedac8a1bfe3df9
  618. Author: Lars Greiss <[email protected]>
  619. Date: Sat Jan 4 15:30:19 2014 +0100
  620.  
  621. Camera2: actually fix ArrayAdapter on MoreSettingsPopup
  622.  
  623. Like CAF already discovered googles attempt to handle the
  624. ArrayAdapter listview does not work as expected if the scroll view
  625. is larger then the screen. CAF tried to fix it with this commit on landscape
  626. view
  627.  
  628. https://github.com/SlimRoms/packages_apps_Camera2/commit/c5b0bfcf2340f02ca65ad3d8c08b5c7d74776ea1
  629.  
  630. which is only a dirty workaround fix for the actual root cause and only works
  631. on landscape partly and if the complete layout is loaded new completly not.
  632.  
  633. Whatever to fix it we need to override and set getItemViewType(int)
  634. and getViewTypeCount() to actually allow to handle propper two row
  635. layouts like it is needed in MoreSettingsPopup (either checkbox layout
  636. or menu layout). The actual detection which type of preference
  637. we need is handled in getItemViewType.
  638.  
  639. Change-Id: I6b48ca83daa5f10a7783dfa29c63e9c0bc562167
  640. -----------------------------------------------------------------
  641. /storage1/Android/cyanogen_mod_bundle11.0/cm-11.0/packages/apps/Contacts 8310831e5de6894ebc1b3176c211ab593cd4b618..74bbd34794b6a455d9a4b4f27f67dfed10c08bd6
  642. commit 74bbd34794b6a455d9a4b4f27f67dfed10c08bd6
  643. Author: Jorge Ruesga <[email protected]>
  644. Date: Sun Feb 9 03:25:06 2014 +0100
  645.  
  646. contacts: Add support for import contacts to local phone storage
  647.  
  648. This change allow to select the local storage account to add/import contacts
  649. when others cloud/corporate accounts exists. If one or more account exists just
  650. ask the user to select the destination account
  651.  
  652. Depends on: http://review.cyanogenmod.org/#/c/59353/
  653.  
  654. Change-Id: Ic055bdb09d037c2f2cdff06cd9a0159df87e2896
  655. JIRA: CYAN-391
  656. Issue: https://jira.cyanogenmod.org/browse/CYAN-391
  657. Signed-off-by: Jorge Ruesga <[email protected]>
  658. -----------------------------------------------------------------
  659. /storage1/Android/cyanogen_mod_bundle11.0/cm-11.0/packages/apps/ContactsCommon 65994eb578d656252c749aaa86b3198a23022b0b..162a1345f14753648063bdb76fc335f321e9bdd7
  660. commit 162a1345f14753648063bdb76fc335f321e9bdd7
  661. Author: Jorge Ruesga <[email protected]>
  662. Date: Sun Feb 9 03:23:42 2014 +0100
  663.  
  664. contacts-common: Add support for import contacts to local phone storage
  665.  
  666. This change allow to select the local storage account to add/import contacts
  667. when others cloud/corporate accounts exists. If one or more account exists just
  668. ask the user to select the destination account
  669.  
  670. Depends on: http://review.cyanogenmod.org/#/c/59352/
  671.  
  672. Change-Id: I0120861479a4dcced11a430803513fc21a2e130b
  673. JIRA: CYAN-391
  674. Issue: https://jira.cyanogenmod.org/browse/CYAN-391
  675. Signed-off-by: Jorge Ruesga <[email protected]>
  676. -----------------------------------------------------------------
  677. /storage1/Android/cyanogen_mod_bundle11.0/cm-11.0/packages/apps/Settings ae175cfe7f198d544b424a9e43cf21c97be93e44..651bd3bc65294f85e023068a7664c3f411e678ef
  678. commit 651bd3bc65294f85e023068a7664c3f411e678ef
  679. Author: Jorge Ruesga <[email protected]>
  680. Date: Wed Feb 12 01:04:24 2014 +0100
  681.  
  682. settings: fix battery stats
  683.  
  684. * Use the new dock stats parcelable creator
  685.  
  686. Change-Id: I4f1ac0b2b7181f0c9c023795acf1671f7b9e9820
  687. JIRA: CYAN-3127
  688. Issue: https://jira.cyanogenmod.org/browse/CYAN-3127
  689. Signed-off-by: Jorge Ruesga <[email protected]>
  690. -----------------------------------------------------------------
  691. /storage1/Android/cyanogen_mod_bundle11.0/cm-11.0/vendor/cm ae9bff6bc2d5bd698c06414b0677158370cf64d4..ed80aaf56f3f0bae7743f96b98ad0854df29a6b1
  692. commit ed80aaf56f3f0bae7743f96b98ad0854df29a6b1
  693. Author: Arne Coucheron <[email protected]>
  694. Date: Sun Feb 9 04:31:46 2014 +0100
  695.  
  696. Cleanup and consolidate APNs for providers using Telenor NO network
  697.  
  698. * Remove "Telenor MMS" - Was merged into "Telenor" in a previous patch
  699. * Rename Ludo to Ventelo - Ludo doesn't exist anymore
  700. * Merge "Mobitalk Internet" and "Mobitalk MMS" into a single APN
  701.  
  702. Change-Id: I77ff06d6c8d6021cd1887468bbeb7ac9315e68bb
Advertisement
Add Comment
Please, Sign In to add comment
Public Pastes
We use cookies for various purposes including analytics. By continuing to use Pastebin, you agree to our use of cookies as described in the Cookies Policy.  OK, I Understand
Not a member of Pastebin yet?
Sign Up, it unlocks many cool features!