rockbox/firmware/target/hosted/filesystem-app.c

569 lines
16 KiB
C
Raw Normal View History

Rewrite filesystem code (WIP) This patch redoes the filesystem code from the FAT driver up to the clipboard code in onplay.c. Not every aspect of this is finished therefore it is still "WIP". I don't wish to do too much at once (haha!). What is left to do is get dircache back in the sim and find an implementation for the dircache indicies in the tagcache and playlist code or do something else that has the same benefit. Leaving these out for now does not make anything unusable. All the basics are done. Phone app code should probably get vetted (and app path handling just plain rewritten as environment expansions); the SDL app and Android run well. Main things addressed: 1) Thread safety: There is none right now in the trunk code. Most of what currently works is luck when multiple threads are involved or multiple descriptors to the same file are open. 2) POSIX compliance: Many of the functions behave nothing like their counterparts on a host system. This leads to inconsistent code or very different behavior from native to hosted. One huge offender was rename(). Going point by point would fill a book. 3) Actual running RAM usage: Many targets will use less RAM and less stack space (some more RAM because I upped the number of cache buffers for large memory). There's very little memory lying fallow in rarely-used areas (see 'Key core changes' below). Also, all targets may open the same number of directory streams whereas before those with less than 8MB RAM were limited to 8, not 12 implying those targets will save slightly less. 4) Performance: The test_disk plugin shows markedly improved performance, particularly in the area of (uncached) directory scanning, due partly to more optimal directory reading and to a better sector cache algorithm. Uncached times tend to be better while there is a bit of a slowdown in dircache due to it being a bit heavier of an implementation. It's not noticeable by a human as far as I can say. Key core changes: 1) Files and directories share core code and data structures. 2) The filesystem code knows which descriptors refer to same file. This ensures that changes from one stream are appropriately reflected in every open descriptor for that file (fileobj_mgr.c). 3) File and directory cache buffers are borrowed from the main sector cache. This means that when they are not in use by a file, they are not wasted, but used for the cache. Most of the time, only a few of them are needed. It also means that adding more file and directory handles is less expensive. All one must do in ensure a large enough cache to borrow from. 4) Relative path components are supported and the namespace is unified. It does not support full relative paths to an implied current directory; what is does support is use of "." and "..". Adding the former would not be very difficult. The namespace is unified in the sense that volumes may be specified several times along with relative parts, e.g.: "/<0>/foo/../../<1>/bar" :<=> "/<1>/bar". 5) Stack usage is down due to sharing of data, static allocation and less duplication of strings on the stack. This requires more serialization than I would like but since the number of threads is limited to a low number, the tradoff in favor of the stack seems reasonable. 6) Separates and heirarchicalizes (sic) the SIM and APP filesystem code. SIM path and volume handling is just like the target. Some aspects of the APP file code get more straightforward (e.g. no path hashing is needed). Dircache: Deserves its own section. Dircache is new but pays homage to the old. The old one was not compatible and so it, since it got redone, does all the stuff it always should have done such as: 1) It may be update and used at any time during the build process. No longer has one to wait for it to finish building to do basic file management (create, remove, rename, etc.). 2) It does not need to be either fully scanned or completely disabled; it can be incomplete (i.e. overfilled, missing paths), still be of benefit and be correct. 3) Handles mounting and dismounting of individual volumes which means a full rebuild is not needed just because you pop a new SD card in the slot. Now, because it reuses its freed entry data, may rebuild only that volume. 4) Much more fundamental to the file code. When it is built, it is the keeper of the master file list whether enabled or not ("disabled" is just a state of the cache). Its must always to ready to be started and bind all streams opened prior to being enabled. 5) Maintains any short filenames in OEM format which means that it does not need to be rebuilt when changing the default codepage. Miscellaneous Compatibility: 1) Update any other code that would otherwise not work such as the hotswap mounting code in various card drivers. 2) File management: Clipboard needed updating because of the behavioral changes. Still needs a little more work on some finer points. 3) Remove now-obsolete functionality such as the mutex's "no preempt" flag (which was only for the prior FAT driver). 4) struct dirinfo uses time_t rather than raw FAT directory entry time fields. I plan to follow up on genericizing everything there (i.e. no FAT attributes). 5) unicode.c needed some redoing so that the file code does not try try to load codepages during a scan, which is actually a problem with the current code. The default codepage, if any is required, is now kept in RAM separarately (bufalloced) from codepages specified to iso_decode() (which must not be bufalloced because the conversion may be done by playback threads). Brings with it some additional reusable core code: 1) Revised file functions: Reusable code that does things such as safe path concatenation and parsing without buffer limitations or data duplication. Variants that copy or alter the input path may be based off these. To do: 1) Put dircache functionality back in the sim. Treating it internally as a different kind of file system seems the best approach at this time. 2) Restore use of dircache indexes in the playlist and database or something effectively the same. Since the cache doesn't have to be complete in order to be used, not getting a hit on the cache doesn't unambiguously say if the path exists or not. Change-Id: Ia30f3082a136253e3a0eae0784e3091d138915c8 Reviewed-on: http://gerrit.rockbox.org/566 Reviewed-by: Michael Sevakis <jethead71@rockbox.org> Tested: Michael Sevakis <jethead71@rockbox.org>
2013-08-06 02:02:45 +00:00
/***************************************************************************
* __________ __ ___.
* Open \______ \ ____ ____ | | _\_ |__ _______ ___
* Source | _// _ \_/ ___\| |/ /| __ \ / _ \ \/ /
* Jukebox | | ( <_> ) \___| < | \_\ ( <_> > < <
* Firmware |____|_ /\____/ \___ >__|_ \|___ /\____/__/\_ \
* \/ \/ \/ \/ \/
* $Id$
*
* Copyright (C) 2010 Thomas Martitz
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version 2
* of the License, or (at your option) any later version.
*
* This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
* KIND, either express or implied.
*
****************************************************************************/
#define RB_FILESYSTEM_OS
#include <stdio.h> /* snprintf */
#include <stdlib.h>
#include <stdarg.h>
#include <time.h>
#include <errno.h>
#include <string.h>
#include <limits.h>
#include "config.h"
#include "system.h"
#include "file.h"
#include "dir.h"
#include "file_internal.h"
#include "pathfuncs.h"
#include "string-extra.h"
#include "rbpaths.h"
#include "logf.h"
#if (CONFIG_PLATFORM & PLATFORM_ANDROID)
static const char rbhome[] = "/sdcard";
#elif (CONFIG_PLATFORM & (PLATFORM_SDL|PLATFORM_MAEMO|PLATFORM_PANDORA)) \
&& !defined(__PCTOOL__)
static const char *rbhome;
Rewrite filesystem code (WIP) This patch redoes the filesystem code from the FAT driver up to the clipboard code in onplay.c. Not every aspect of this is finished therefore it is still "WIP". I don't wish to do too much at once (haha!). What is left to do is get dircache back in the sim and find an implementation for the dircache indicies in the tagcache and playlist code or do something else that has the same benefit. Leaving these out for now does not make anything unusable. All the basics are done. Phone app code should probably get vetted (and app path handling just plain rewritten as environment expansions); the SDL app and Android run well. Main things addressed: 1) Thread safety: There is none right now in the trunk code. Most of what currently works is luck when multiple threads are involved or multiple descriptors to the same file are open. 2) POSIX compliance: Many of the functions behave nothing like their counterparts on a host system. This leads to inconsistent code or very different behavior from native to hosted. One huge offender was rename(). Going point by point would fill a book. 3) Actual running RAM usage: Many targets will use less RAM and less stack space (some more RAM because I upped the number of cache buffers for large memory). There's very little memory lying fallow in rarely-used areas (see 'Key core changes' below). Also, all targets may open the same number of directory streams whereas before those with less than 8MB RAM were limited to 8, not 12 implying those targets will save slightly less. 4) Performance: The test_disk plugin shows markedly improved performance, particularly in the area of (uncached) directory scanning, due partly to more optimal directory reading and to a better sector cache algorithm. Uncached times tend to be better while there is a bit of a slowdown in dircache due to it being a bit heavier of an implementation. It's not noticeable by a human as far as I can say. Key core changes: 1) Files and directories share core code and data structures. 2) The filesystem code knows which descriptors refer to same file. This ensures that changes from one stream are appropriately reflected in every open descriptor for that file (fileobj_mgr.c). 3) File and directory cache buffers are borrowed from the main sector cache. This means that when they are not in use by a file, they are not wasted, but used for the cache. Most of the time, only a few of them are needed. It also means that adding more file and directory handles is less expensive. All one must do in ensure a large enough cache to borrow from. 4) Relative path components are supported and the namespace is unified. It does not support full relative paths to an implied current directory; what is does support is use of "." and "..". Adding the former would not be very difficult. The namespace is unified in the sense that volumes may be specified several times along with relative parts, e.g.: "/<0>/foo/../../<1>/bar" :<=> "/<1>/bar". 5) Stack usage is down due to sharing of data, static allocation and less duplication of strings on the stack. This requires more serialization than I would like but since the number of threads is limited to a low number, the tradoff in favor of the stack seems reasonable. 6) Separates and heirarchicalizes (sic) the SIM and APP filesystem code. SIM path and volume handling is just like the target. Some aspects of the APP file code get more straightforward (e.g. no path hashing is needed). Dircache: Deserves its own section. Dircache is new but pays homage to the old. The old one was not compatible and so it, since it got redone, does all the stuff it always should have done such as: 1) It may be update and used at any time during the build process. No longer has one to wait for it to finish building to do basic file management (create, remove, rename, etc.). 2) It does not need to be either fully scanned or completely disabled; it can be incomplete (i.e. overfilled, missing paths), still be of benefit and be correct. 3) Handles mounting and dismounting of individual volumes which means a full rebuild is not needed just because you pop a new SD card in the slot. Now, because it reuses its freed entry data, may rebuild only that volume. 4) Much more fundamental to the file code. When it is built, it is the keeper of the master file list whether enabled or not ("disabled" is just a state of the cache). Its must always to ready to be started and bind all streams opened prior to being enabled. 5) Maintains any short filenames in OEM format which means that it does not need to be rebuilt when changing the default codepage. Miscellaneous Compatibility: 1) Update any other code that would otherwise not work such as the hotswap mounting code in various card drivers. 2) File management: Clipboard needed updating because of the behavioral changes. Still needs a little more work on some finer points. 3) Remove now-obsolete functionality such as the mutex's "no preempt" flag (which was only for the prior FAT driver). 4) struct dirinfo uses time_t rather than raw FAT directory entry time fields. I plan to follow up on genericizing everything there (i.e. no FAT attributes). 5) unicode.c needed some redoing so that the file code does not try try to load codepages during a scan, which is actually a problem with the current code. The default codepage, if any is required, is now kept in RAM separarately (bufalloced) from codepages specified to iso_decode() (which must not be bufalloced because the conversion may be done by playback threads). Brings with it some additional reusable core code: 1) Revised file functions: Reusable code that does things such as safe path concatenation and parsing without buffer limitations or data duplication. Variants that copy or alter the input path may be based off these. To do: 1) Put dircache functionality back in the sim. Treating it internally as a different kind of file system seems the best approach at this time. 2) Restore use of dircache indexes in the playlist and database or something effectively the same. Since the cache doesn't have to be complete in order to be used, not getting a hit on the cache doesn't unambiguously say if the path exists or not. Change-Id: Ia30f3082a136253e3a0eae0784e3091d138915c8 Reviewed-on: http://gerrit.rockbox.org/566 Reviewed-by: Michael Sevakis <jethead71@rockbox.org> Tested: Michael Sevakis <jethead71@rockbox.org>
2013-08-06 02:02:45 +00:00
#else
/* YPR0, YPR1 */
static const char rbhome[] = HOME_DIR;
#endif
iBasso DX50/DX90: Major code cleanup and reorganization. Reorganization - Separated iBasso devices from PLATFORM_ANDROID. These are now standlone hosted targets. Most device specific code is in the firmware/target/hosted/ibasso directory. - No dependency on Android SDK, only the Android NDK is needed. 32 bit Android NDK and Android API Level 16. - Separate implementation for each device where feasible. Code cleanup - Rewrite of existing code, from simple reformat to complete reimplementation. - New backlight interface, seperating backlight from touchscreen. - Rewrite of device button handler, removing unneeded code and fixing memory leaks. - New Debug messages interface logging to Android adb logcat (DEBUGF, panicf, logf). - Rewrite of lcd device handler, removing unneeded code and fixing memory leaks. - Rewrite of audiohw device handler/pcm interface, removing unneeded code and fixing memory leaks, enabling 44.1/48kHz pthreaded playback. - Rewrite of power and powermng, proper shutdown, using batterylog results (see http://gerrit.rockbox.org/r/#/c/1047/). - Rewrite of configure (Android NDK) and device specific config. - Rewrite of the Android NDK specific Makefile. Misc - All plugins/games/demos activated. - Update tinyalsa to latest from https://github.com/tinyalsa/tinyalsa. Includes - http://gerrit.rockbox.org/r/#/c/993/ - http://gerrit.rockbox.org/r/#/c/1010/ - http://gerrit.rockbox.org/r/#/c/1035/ Does not include http://gerrit.rockbox.org/r/#/c/1007/ due to new backlight interface and new option for hold switch, touchscreen, physical button interaction. Rockbox needs the iBasso DX50/DX90 loader for startup, see http://gerrit.rockbox.org/r/#/c/1099/ The loader expects Rockbox to be installed in /mnt/sdcard/.rockbox/. If /mnt/sdcard/ is accessed as USB mass storage device, Rockbox will exit gracefully and the loader will restart Rockbox on USB disconnect. Tested on iBasso DX50. Compiled (not tested) for iBasso DX90. Compiled (not tested) for PLATFORM_ANDROID. Change-Id: I5f5e22e68f5b4cf29c28e2b40b2c265f2beb7ab7
2015-02-02 20:44:29 +00:00
#if !(defined(SAMSUNG_YPR0) || defined(SAMSUNG_YPR1) || defined(DX50) || defined(DX90)) && !defined(__PCTOOL__)
Rewrite filesystem code (WIP) This patch redoes the filesystem code from the FAT driver up to the clipboard code in onplay.c. Not every aspect of this is finished therefore it is still "WIP". I don't wish to do too much at once (haha!). What is left to do is get dircache back in the sim and find an implementation for the dircache indicies in the tagcache and playlist code or do something else that has the same benefit. Leaving these out for now does not make anything unusable. All the basics are done. Phone app code should probably get vetted (and app path handling just plain rewritten as environment expansions); the SDL app and Android run well. Main things addressed: 1) Thread safety: There is none right now in the trunk code. Most of what currently works is luck when multiple threads are involved or multiple descriptors to the same file are open. 2) POSIX compliance: Many of the functions behave nothing like their counterparts on a host system. This leads to inconsistent code or very different behavior from native to hosted. One huge offender was rename(). Going point by point would fill a book. 3) Actual running RAM usage: Many targets will use less RAM and less stack space (some more RAM because I upped the number of cache buffers for large memory). There's very little memory lying fallow in rarely-used areas (see 'Key core changes' below). Also, all targets may open the same number of directory streams whereas before those with less than 8MB RAM were limited to 8, not 12 implying those targets will save slightly less. 4) Performance: The test_disk plugin shows markedly improved performance, particularly in the area of (uncached) directory scanning, due partly to more optimal directory reading and to a better sector cache algorithm. Uncached times tend to be better while there is a bit of a slowdown in dircache due to it being a bit heavier of an implementation. It's not noticeable by a human as far as I can say. Key core changes: 1) Files and directories share core code and data structures. 2) The filesystem code knows which descriptors refer to same file. This ensures that changes from one stream are appropriately reflected in every open descriptor for that file (fileobj_mgr.c). 3) File and directory cache buffers are borrowed from the main sector cache. This means that when they are not in use by a file, they are not wasted, but used for the cache. Most of the time, only a few of them are needed. It also means that adding more file and directory handles is less expensive. All one must do in ensure a large enough cache to borrow from. 4) Relative path components are supported and the namespace is unified. It does not support full relative paths to an implied current directory; what is does support is use of "." and "..". Adding the former would not be very difficult. The namespace is unified in the sense that volumes may be specified several times along with relative parts, e.g.: "/<0>/foo/../../<1>/bar" :<=> "/<1>/bar". 5) Stack usage is down due to sharing of data, static allocation and less duplication of strings on the stack. This requires more serialization than I would like but since the number of threads is limited to a low number, the tradoff in favor of the stack seems reasonable. 6) Separates and heirarchicalizes (sic) the SIM and APP filesystem code. SIM path and volume handling is just like the target. Some aspects of the APP file code get more straightforward (e.g. no path hashing is needed). Dircache: Deserves its own section. Dircache is new but pays homage to the old. The old one was not compatible and so it, since it got redone, does all the stuff it always should have done such as: 1) It may be update and used at any time during the build process. No longer has one to wait for it to finish building to do basic file management (create, remove, rename, etc.). 2) It does not need to be either fully scanned or completely disabled; it can be incomplete (i.e. overfilled, missing paths), still be of benefit and be correct. 3) Handles mounting and dismounting of individual volumes which means a full rebuild is not needed just because you pop a new SD card in the slot. Now, because it reuses its freed entry data, may rebuild only that volume. 4) Much more fundamental to the file code. When it is built, it is the keeper of the master file list whether enabled or not ("disabled" is just a state of the cache). Its must always to ready to be started and bind all streams opened prior to being enabled. 5) Maintains any short filenames in OEM format which means that it does not need to be rebuilt when changing the default codepage. Miscellaneous Compatibility: 1) Update any other code that would otherwise not work such as the hotswap mounting code in various card drivers. 2) File management: Clipboard needed updating because of the behavioral changes. Still needs a little more work on some finer points. 3) Remove now-obsolete functionality such as the mutex's "no preempt" flag (which was only for the prior FAT driver). 4) struct dirinfo uses time_t rather than raw FAT directory entry time fields. I plan to follow up on genericizing everything there (i.e. no FAT attributes). 5) unicode.c needed some redoing so that the file code does not try try to load codepages during a scan, which is actually a problem with the current code. The default codepage, if any is required, is now kept in RAM separarately (bufalloced) from codepages specified to iso_decode() (which must not be bufalloced because the conversion may be done by playback threads). Brings with it some additional reusable core code: 1) Revised file functions: Reusable code that does things such as safe path concatenation and parsing without buffer limitations or data duplication. Variants that copy or alter the input path may be based off these. To do: 1) Put dircache functionality back in the sim. Treating it internally as a different kind of file system seems the best approach at this time. 2) Restore use of dircache indexes in the playlist and database or something effectively the same. Since the cache doesn't have to be complete in order to be used, not getting a hit on the cache doesn't unambiguously say if the path exists or not. Change-Id: Ia30f3082a136253e3a0eae0784e3091d138915c8 Reviewed-on: http://gerrit.rockbox.org/566 Reviewed-by: Michael Sevakis <jethead71@rockbox.org> Tested: Michael Sevakis <jethead71@rockbox.org>
2013-08-06 02:02:45 +00:00
/* Special dirs are user-accessible (and user-writable) dirs which take priority
* over the ones where Rockbox is installed to. Classic example would be
* $HOME/.config/rockbox.org vs /usr/share/rockbox */
#define HAVE_SPECIAL_DIRS
#endif
#ifdef HAVE_MULTIDRIVE
/* This is to compare any opened directories with the home directory so that
the special drive links may be returned for it only */
static int rbhome_fildes = -1;
/* A special link is created under e.g. HOME_DIR/<microSD1>, e.g. to access
* external storage in a convenient location, much similar to the mount
* point on our native targets. Here they are treated as symlink (one which
* doesn't actually exist in the filesystem and therefore we have to override
* readlink() */
static const char *handle_special_links(const char* link, unsigned flags,
char *buf, const size_t bufsize)
{
(void) flags;
char vol_string[VOL_MAX_LEN + 1];
int len;
Rewrite filesystem code (WIP) This patch redoes the filesystem code from the FAT driver up to the clipboard code in onplay.c. Not every aspect of this is finished therefore it is still "WIP". I don't wish to do too much at once (haha!). What is left to do is get dircache back in the sim and find an implementation for the dircache indicies in the tagcache and playlist code or do something else that has the same benefit. Leaving these out for now does not make anything unusable. All the basics are done. Phone app code should probably get vetted (and app path handling just plain rewritten as environment expansions); the SDL app and Android run well. Main things addressed: 1) Thread safety: There is none right now in the trunk code. Most of what currently works is luck when multiple threads are involved or multiple descriptors to the same file are open. 2) POSIX compliance: Many of the functions behave nothing like their counterparts on a host system. This leads to inconsistent code or very different behavior from native to hosted. One huge offender was rename(). Going point by point would fill a book. 3) Actual running RAM usage: Many targets will use less RAM and less stack space (some more RAM because I upped the number of cache buffers for large memory). There's very little memory lying fallow in rarely-used areas (see 'Key core changes' below). Also, all targets may open the same number of directory streams whereas before those with less than 8MB RAM were limited to 8, not 12 implying those targets will save slightly less. 4) Performance: The test_disk plugin shows markedly improved performance, particularly in the area of (uncached) directory scanning, due partly to more optimal directory reading and to a better sector cache algorithm. Uncached times tend to be better while there is a bit of a slowdown in dircache due to it being a bit heavier of an implementation. It's not noticeable by a human as far as I can say. Key core changes: 1) Files and directories share core code and data structures. 2) The filesystem code knows which descriptors refer to same file. This ensures that changes from one stream are appropriately reflected in every open descriptor for that file (fileobj_mgr.c). 3) File and directory cache buffers are borrowed from the main sector cache. This means that when they are not in use by a file, they are not wasted, but used for the cache. Most of the time, only a few of them are needed. It also means that adding more file and directory handles is less expensive. All one must do in ensure a large enough cache to borrow from. 4) Relative path components are supported and the namespace is unified. It does not support full relative paths to an implied current directory; what is does support is use of "." and "..". Adding the former would not be very difficult. The namespace is unified in the sense that volumes may be specified several times along with relative parts, e.g.: "/<0>/foo/../../<1>/bar" :<=> "/<1>/bar". 5) Stack usage is down due to sharing of data, static allocation and less duplication of strings on the stack. This requires more serialization than I would like but since the number of threads is limited to a low number, the tradoff in favor of the stack seems reasonable. 6) Separates and heirarchicalizes (sic) the SIM and APP filesystem code. SIM path and volume handling is just like the target. Some aspects of the APP file code get more straightforward (e.g. no path hashing is needed). Dircache: Deserves its own section. Dircache is new but pays homage to the old. The old one was not compatible and so it, since it got redone, does all the stuff it always should have done such as: 1) It may be update and used at any time during the build process. No longer has one to wait for it to finish building to do basic file management (create, remove, rename, etc.). 2) It does not need to be either fully scanned or completely disabled; it can be incomplete (i.e. overfilled, missing paths), still be of benefit and be correct. 3) Handles mounting and dismounting of individual volumes which means a full rebuild is not needed just because you pop a new SD card in the slot. Now, because it reuses its freed entry data, may rebuild only that volume. 4) Much more fundamental to the file code. When it is built, it is the keeper of the master file list whether enabled or not ("disabled" is just a state of the cache). Its must always to ready to be started and bind all streams opened prior to being enabled. 5) Maintains any short filenames in OEM format which means that it does not need to be rebuilt when changing the default codepage. Miscellaneous Compatibility: 1) Update any other code that would otherwise not work such as the hotswap mounting code in various card drivers. 2) File management: Clipboard needed updating because of the behavioral changes. Still needs a little more work on some finer points. 3) Remove now-obsolete functionality such as the mutex's "no preempt" flag (which was only for the prior FAT driver). 4) struct dirinfo uses time_t rather than raw FAT directory entry time fields. I plan to follow up on genericizing everything there (i.e. no FAT attributes). 5) unicode.c needed some redoing so that the file code does not try try to load codepages during a scan, which is actually a problem with the current code. The default codepage, if any is required, is now kept in RAM separarately (bufalloced) from codepages specified to iso_decode() (which must not be bufalloced because the conversion may be done by playback threads). Brings with it some additional reusable core code: 1) Revised file functions: Reusable code that does things such as safe path concatenation and parsing without buffer limitations or data duplication. Variants that copy or alter the input path may be based off these. To do: 1) Put dircache functionality back in the sim. Treating it internally as a different kind of file system seems the best approach at this time. 2) Restore use of dircache indexes in the playlist and database or something effectively the same. Since the cache doesn't have to be complete in order to be used, not getting a hit on the cache doesn't unambiguously say if the path exists or not. Change-Id: Ia30f3082a136253e3a0eae0784e3091d138915c8 Reviewed-on: http://gerrit.rockbox.org/566 Reviewed-by: Michael Sevakis <jethead71@rockbox.org> Tested: Michael Sevakis <jethead71@rockbox.org>
2013-08-06 02:02:45 +00:00
for (int i = 1; i < NUM_VOLUMES; i++)
Rewrite filesystem code (WIP) This patch redoes the filesystem code from the FAT driver up to the clipboard code in onplay.c. Not every aspect of this is finished therefore it is still "WIP". I don't wish to do too much at once (haha!). What is left to do is get dircache back in the sim and find an implementation for the dircache indicies in the tagcache and playlist code or do something else that has the same benefit. Leaving these out for now does not make anything unusable. All the basics are done. Phone app code should probably get vetted (and app path handling just plain rewritten as environment expansions); the SDL app and Android run well. Main things addressed: 1) Thread safety: There is none right now in the trunk code. Most of what currently works is luck when multiple threads are involved or multiple descriptors to the same file are open. 2) POSIX compliance: Many of the functions behave nothing like their counterparts on a host system. This leads to inconsistent code or very different behavior from native to hosted. One huge offender was rename(). Going point by point would fill a book. 3) Actual running RAM usage: Many targets will use less RAM and less stack space (some more RAM because I upped the number of cache buffers for large memory). There's very little memory lying fallow in rarely-used areas (see 'Key core changes' below). Also, all targets may open the same number of directory streams whereas before those with less than 8MB RAM were limited to 8, not 12 implying those targets will save slightly less. 4) Performance: The test_disk plugin shows markedly improved performance, particularly in the area of (uncached) directory scanning, due partly to more optimal directory reading and to a better sector cache algorithm. Uncached times tend to be better while there is a bit of a slowdown in dircache due to it being a bit heavier of an implementation. It's not noticeable by a human as far as I can say. Key core changes: 1) Files and directories share core code and data structures. 2) The filesystem code knows which descriptors refer to same file. This ensures that changes from one stream are appropriately reflected in every open descriptor for that file (fileobj_mgr.c). 3) File and directory cache buffers are borrowed from the main sector cache. This means that when they are not in use by a file, they are not wasted, but used for the cache. Most of the time, only a few of them are needed. It also means that adding more file and directory handles is less expensive. All one must do in ensure a large enough cache to borrow from. 4) Relative path components are supported and the namespace is unified. It does not support full relative paths to an implied current directory; what is does support is use of "." and "..". Adding the former would not be very difficult. The namespace is unified in the sense that volumes may be specified several times along with relative parts, e.g.: "/<0>/foo/../../<1>/bar" :<=> "/<1>/bar". 5) Stack usage is down due to sharing of data, static allocation and less duplication of strings on the stack. This requires more serialization than I would like but since the number of threads is limited to a low number, the tradoff in favor of the stack seems reasonable. 6) Separates and heirarchicalizes (sic) the SIM and APP filesystem code. SIM path and volume handling is just like the target. Some aspects of the APP file code get more straightforward (e.g. no path hashing is needed). Dircache: Deserves its own section. Dircache is new but pays homage to the old. The old one was not compatible and so it, since it got redone, does all the stuff it always should have done such as: 1) It may be update and used at any time during the build process. No longer has one to wait for it to finish building to do basic file management (create, remove, rename, etc.). 2) It does not need to be either fully scanned or completely disabled; it can be incomplete (i.e. overfilled, missing paths), still be of benefit and be correct. 3) Handles mounting and dismounting of individual volumes which means a full rebuild is not needed just because you pop a new SD card in the slot. Now, because it reuses its freed entry data, may rebuild only that volume. 4) Much more fundamental to the file code. When it is built, it is the keeper of the master file list whether enabled or not ("disabled" is just a state of the cache). Its must always to ready to be started and bind all streams opened prior to being enabled. 5) Maintains any short filenames in OEM format which means that it does not need to be rebuilt when changing the default codepage. Miscellaneous Compatibility: 1) Update any other code that would otherwise not work such as the hotswap mounting code in various card drivers. 2) File management: Clipboard needed updating because of the behavioral changes. Still needs a little more work on some finer points. 3) Remove now-obsolete functionality such as the mutex's "no preempt" flag (which was only for the prior FAT driver). 4) struct dirinfo uses time_t rather than raw FAT directory entry time fields. I plan to follow up on genericizing everything there (i.e. no FAT attributes). 5) unicode.c needed some redoing so that the file code does not try try to load codepages during a scan, which is actually a problem with the current code. The default codepage, if any is required, is now kept in RAM separarately (bufalloced) from codepages specified to iso_decode() (which must not be bufalloced because the conversion may be done by playback threads). Brings with it some additional reusable core code: 1) Revised file functions: Reusable code that does things such as safe path concatenation and parsing without buffer limitations or data duplication. Variants that copy or alter the input path may be based off these. To do: 1) Put dircache functionality back in the sim. Treating it internally as a different kind of file system seems the best approach at this time. 2) Restore use of dircache indexes in the playlist and database or something effectively the same. Since the cache doesn't have to be complete in order to be used, not getting a hit on the cache doesn't unambiguously say if the path exists or not. Change-Id: Ia30f3082a136253e3a0eae0784e3091d138915c8 Reviewed-on: http://gerrit.rockbox.org/566 Reviewed-by: Michael Sevakis <jethead71@rockbox.org> Tested: Michael Sevakis <jethead71@rockbox.org>
2013-08-06 02:02:45 +00:00
{
len = get_volume_name(i, vol_string);
/* link might be passed with or without HOME_DIR expanded. To handle
* both perform substring matching (VOL_NAMES is unique enough) */
const char *begin = strstr(link, vol_string);
if (begin)
{
/* begin now points to the start of vol_string within link,
* we want to copy the remainder of the paths, prefixed by
* the actual mount point (the remainder might be "") */
snprintf(buf, bufsize, MULTIDRIVE_DIR"%s", begin + len);
return buf;
}
Rewrite filesystem code (WIP) This patch redoes the filesystem code from the FAT driver up to the clipboard code in onplay.c. Not every aspect of this is finished therefore it is still "WIP". I don't wish to do too much at once (haha!). What is left to do is get dircache back in the sim and find an implementation for the dircache indicies in the tagcache and playlist code or do something else that has the same benefit. Leaving these out for now does not make anything unusable. All the basics are done. Phone app code should probably get vetted (and app path handling just plain rewritten as environment expansions); the SDL app and Android run well. Main things addressed: 1) Thread safety: There is none right now in the trunk code. Most of what currently works is luck when multiple threads are involved or multiple descriptors to the same file are open. 2) POSIX compliance: Many of the functions behave nothing like their counterparts on a host system. This leads to inconsistent code or very different behavior from native to hosted. One huge offender was rename(). Going point by point would fill a book. 3) Actual running RAM usage: Many targets will use less RAM and less stack space (some more RAM because I upped the number of cache buffers for large memory). There's very little memory lying fallow in rarely-used areas (see 'Key core changes' below). Also, all targets may open the same number of directory streams whereas before those with less than 8MB RAM were limited to 8, not 12 implying those targets will save slightly less. 4) Performance: The test_disk plugin shows markedly improved performance, particularly in the area of (uncached) directory scanning, due partly to more optimal directory reading and to a better sector cache algorithm. Uncached times tend to be better while there is a bit of a slowdown in dircache due to it being a bit heavier of an implementation. It's not noticeable by a human as far as I can say. Key core changes: 1) Files and directories share core code and data structures. 2) The filesystem code knows which descriptors refer to same file. This ensures that changes from one stream are appropriately reflected in every open descriptor for that file (fileobj_mgr.c). 3) File and directory cache buffers are borrowed from the main sector cache. This means that when they are not in use by a file, they are not wasted, but used for the cache. Most of the time, only a few of them are needed. It also means that adding more file and directory handles is less expensive. All one must do in ensure a large enough cache to borrow from. 4) Relative path components are supported and the namespace is unified. It does not support full relative paths to an implied current directory; what is does support is use of "." and "..". Adding the former would not be very difficult. The namespace is unified in the sense that volumes may be specified several times along with relative parts, e.g.: "/<0>/foo/../../<1>/bar" :<=> "/<1>/bar". 5) Stack usage is down due to sharing of data, static allocation and less duplication of strings on the stack. This requires more serialization than I would like but since the number of threads is limited to a low number, the tradoff in favor of the stack seems reasonable. 6) Separates and heirarchicalizes (sic) the SIM and APP filesystem code. SIM path and volume handling is just like the target. Some aspects of the APP file code get more straightforward (e.g. no path hashing is needed). Dircache: Deserves its own section. Dircache is new but pays homage to the old. The old one was not compatible and so it, since it got redone, does all the stuff it always should have done such as: 1) It may be update and used at any time during the build process. No longer has one to wait for it to finish building to do basic file management (create, remove, rename, etc.). 2) It does not need to be either fully scanned or completely disabled; it can be incomplete (i.e. overfilled, missing paths), still be of benefit and be correct. 3) Handles mounting and dismounting of individual volumes which means a full rebuild is not needed just because you pop a new SD card in the slot. Now, because it reuses its freed entry data, may rebuild only that volume. 4) Much more fundamental to the file code. When it is built, it is the keeper of the master file list whether enabled or not ("disabled" is just a state of the cache). Its must always to ready to be started and bind all streams opened prior to being enabled. 5) Maintains any short filenames in OEM format which means that it does not need to be rebuilt when changing the default codepage. Miscellaneous Compatibility: 1) Update any other code that would otherwise not work such as the hotswap mounting code in various card drivers. 2) File management: Clipboard needed updating because of the behavioral changes. Still needs a little more work on some finer points. 3) Remove now-obsolete functionality such as the mutex's "no preempt" flag (which was only for the prior FAT driver). 4) struct dirinfo uses time_t rather than raw FAT directory entry time fields. I plan to follow up on genericizing everything there (i.e. no FAT attributes). 5) unicode.c needed some redoing so that the file code does not try try to load codepages during a scan, which is actually a problem with the current code. The default codepage, if any is required, is now kept in RAM separarately (bufalloced) from codepages specified to iso_decode() (which must not be bufalloced because the conversion may be done by playback threads). Brings with it some additional reusable core code: 1) Revised file functions: Reusable code that does things such as safe path concatenation and parsing without buffer limitations or data duplication. Variants that copy or alter the input path may be based off these. To do: 1) Put dircache functionality back in the sim. Treating it internally as a different kind of file system seems the best approach at this time. 2) Restore use of dircache indexes in the playlist and database or something effectively the same. Since the cache doesn't have to be complete in order to be used, not getting a hit on the cache doesn't unambiguously say if the path exists or not. Change-Id: Ia30f3082a136253e3a0eae0784e3091d138915c8 Reviewed-on: http://gerrit.rockbox.org/566 Reviewed-by: Michael Sevakis <jethead71@rockbox.org> Tested: Michael Sevakis <jethead71@rockbox.org>
2013-08-06 02:02:45 +00:00
}
return link;
}
#endif
#ifdef HAVE_MULTIDRIVE
/* we keep an open descriptor of the home directory to detect when it has been
opened by opendir() so that its "symlinks" may be enumerated */
static void cleanup_rbhome(void)
{
os_close(rbhome_fildes);
rbhome_fildes = -1;
}
#endif /* HAVE_MULTIDRIVE */
void paths_init(void)
{
#ifdef HAVE_SPECIAL_DIRS
/* make sure $HOME/.config/rockbox.org exists, it's needed for config.cfg */
#if (CONFIG_PLATFORM & PLATFORM_ANDROID)
os_mkdir("/sdcard/rockbox" __MKDIR_MODE_ARG);
os_mkdir("/sdcard/rockbox/rocks.data" __MKDIR_MODE_ARG);
Introducing Targets iBasso DX50 & iBasso DX90 The port to for this two targets has been entirely developped by Ilia Sergachev (alias Il or xzcc). His source can be found at https://bitbucket.org/isergachev/rockbox . The few necesary modifications for the DX90 port was done by headwhacker form head-fi.org. Unfortunately i could not try out the final state of the DX90 port. The port is hosted on android (without java) as standalone app. The official Firmware is required to run this port. Ilia did modify the source files for the "android" target in the rockbox source to make the DX port work. The work I did was to separate the code for DX50 (&DX90) from the android target. On this Target Ilia used source from tinyalsa from AOSP. I did not touch that part of the code because I do not understand it. What else I changed from Ilias sources besides the separation from the target "android": * removed a dirty hack to keep backlight off * changed value battery meter to voltage battery meter * made all plugins compile (named target as "standalone") and added keymaps * i added the graphics for the manual but did not do anything else for the manual yet * minor optimizations known bugs: * timers are slowed donw when playback is active (tinyalsa related?) * some minor bugs Things to do: * The main prolem will be how to install the app correctly. A guy called DOC2008 added a CWM (by androtab.info) to the official firmware and Ilia made a CWM installation script and a dualboot selector (rbutils/ibassoboot, build with ndk-build). We will have to find a way to install rockbox in a proper way without breaking any copyrights. Maybe ADB is an option but it is not enable with OF by default. Patching the OF is probably the way to go. * All the wiki and manual to build: needed: android ndk installed, android sdk installed with additional build-tools 19.1.0 installed ./tools/configure select iBasso DX50 or iBasso DX90 make -j apk the content of rockbox.zip/.rockbox needs to be copied to /system/rockbox/app_rockbox/rockbox/ (rockbox app not needed) the content of libs/armeabi to /system/rockbox/lib/ (rockbox app needed) The boot selector is needed as /system/bin/MangoPlayer and the iBasso app as /system/bin/MangoPlayer_original. There is also the "vold" file. The one from OF does not work with DX50 rockbox (DX90 works!?), the one from Ilia is necessary. Until we have found a proper way to install it, it can only be installed following the instructions of Ilia on his bitbucket page, using the CWM-OF and his installation script package. Change-Id: Ic4faaf84824c162aabcc08e492cee6e0068719d0 Reviewed-on: http://gerrit.rockbox.org/941 Tested: Chiwen Chang <rock1104.tw@yahoo.com.tw> Reviewed-by: Michael Giacomelli <giac2000@hotmail.com>
2014-08-30 11:15:53 +00:00
os_mkdir("/sdcard/rockbox/eqs" __MKDIR_MODE_ARG);
Rewrite filesystem code (WIP) This patch redoes the filesystem code from the FAT driver up to the clipboard code in onplay.c. Not every aspect of this is finished therefore it is still "WIP". I don't wish to do too much at once (haha!). What is left to do is get dircache back in the sim and find an implementation for the dircache indicies in the tagcache and playlist code or do something else that has the same benefit. Leaving these out for now does not make anything unusable. All the basics are done. Phone app code should probably get vetted (and app path handling just plain rewritten as environment expansions); the SDL app and Android run well. Main things addressed: 1) Thread safety: There is none right now in the trunk code. Most of what currently works is luck when multiple threads are involved or multiple descriptors to the same file are open. 2) POSIX compliance: Many of the functions behave nothing like their counterparts on a host system. This leads to inconsistent code or very different behavior from native to hosted. One huge offender was rename(). Going point by point would fill a book. 3) Actual running RAM usage: Many targets will use less RAM and less stack space (some more RAM because I upped the number of cache buffers for large memory). There's very little memory lying fallow in rarely-used areas (see 'Key core changes' below). Also, all targets may open the same number of directory streams whereas before those with less than 8MB RAM were limited to 8, not 12 implying those targets will save slightly less. 4) Performance: The test_disk plugin shows markedly improved performance, particularly in the area of (uncached) directory scanning, due partly to more optimal directory reading and to a better sector cache algorithm. Uncached times tend to be better while there is a bit of a slowdown in dircache due to it being a bit heavier of an implementation. It's not noticeable by a human as far as I can say. Key core changes: 1) Files and directories share core code and data structures. 2) The filesystem code knows which descriptors refer to same file. This ensures that changes from one stream are appropriately reflected in every open descriptor for that file (fileobj_mgr.c). 3) File and directory cache buffers are borrowed from the main sector cache. This means that when they are not in use by a file, they are not wasted, but used for the cache. Most of the time, only a few of them are needed. It also means that adding more file and directory handles is less expensive. All one must do in ensure a large enough cache to borrow from. 4) Relative path components are supported and the namespace is unified. It does not support full relative paths to an implied current directory; what is does support is use of "." and "..". Adding the former would not be very difficult. The namespace is unified in the sense that volumes may be specified several times along with relative parts, e.g.: "/<0>/foo/../../<1>/bar" :<=> "/<1>/bar". 5) Stack usage is down due to sharing of data, static allocation and less duplication of strings on the stack. This requires more serialization than I would like but since the number of threads is limited to a low number, the tradoff in favor of the stack seems reasonable. 6) Separates and heirarchicalizes (sic) the SIM and APP filesystem code. SIM path and volume handling is just like the target. Some aspects of the APP file code get more straightforward (e.g. no path hashing is needed). Dircache: Deserves its own section. Dircache is new but pays homage to the old. The old one was not compatible and so it, since it got redone, does all the stuff it always should have done such as: 1) It may be update and used at any time during the build process. No longer has one to wait for it to finish building to do basic file management (create, remove, rename, etc.). 2) It does not need to be either fully scanned or completely disabled; it can be incomplete (i.e. overfilled, missing paths), still be of benefit and be correct. 3) Handles mounting and dismounting of individual volumes which means a full rebuild is not needed just because you pop a new SD card in the slot. Now, because it reuses its freed entry data, may rebuild only that volume. 4) Much more fundamental to the file code. When it is built, it is the keeper of the master file list whether enabled or not ("disabled" is just a state of the cache). Its must always to ready to be started and bind all streams opened prior to being enabled. 5) Maintains any short filenames in OEM format which means that it does not need to be rebuilt when changing the default codepage. Miscellaneous Compatibility: 1) Update any other code that would otherwise not work such as the hotswap mounting code in various card drivers. 2) File management: Clipboard needed updating because of the behavioral changes. Still needs a little more work on some finer points. 3) Remove now-obsolete functionality such as the mutex's "no preempt" flag (which was only for the prior FAT driver). 4) struct dirinfo uses time_t rather than raw FAT directory entry time fields. I plan to follow up on genericizing everything there (i.e. no FAT attributes). 5) unicode.c needed some redoing so that the file code does not try try to load codepages during a scan, which is actually a problem with the current code. The default codepage, if any is required, is now kept in RAM separarately (bufalloced) from codepages specified to iso_decode() (which must not be bufalloced because the conversion may be done by playback threads). Brings with it some additional reusable core code: 1) Revised file functions: Reusable code that does things such as safe path concatenation and parsing without buffer limitations or data duplication. Variants that copy or alter the input path may be based off these. To do: 1) Put dircache functionality back in the sim. Treating it internally as a different kind of file system seems the best approach at this time. 2) Restore use of dircache indexes in the playlist and database or something effectively the same. Since the cache doesn't have to be complete in order to be used, not getting a hit on the cache doesn't unambiguously say if the path exists or not. Change-Id: Ia30f3082a136253e3a0eae0784e3091d138915c8 Reviewed-on: http://gerrit.rockbox.org/566 Reviewed-by: Michael Sevakis <jethead71@rockbox.org> Tested: Michael Sevakis <jethead71@rockbox.org>
2013-08-06 02:02:45 +00:00
#else
char config_dir[MAX_PATH];
const char *home = getenv("RBROOT");
if (!home)
{
home = getenv("HOME");
}
if (!home)
{
logf("HOME environment var not set. Can't write config");
return;
}
rbhome = home;
snprintf(config_dir, sizeof(config_dir), "%s/.config", home);
os_mkdir(config_dir __MKDIR_MODE_ARG);
snprintf(config_dir, sizeof(config_dir), "%s/.config/rockbox.org", home);
os_mkdir(config_dir __MKDIR_MODE_ARG);
/* Plugin data directory */
snprintf(config_dir, sizeof(config_dir), "%s/.config/rockbox.org/rocks.data", home);
os_mkdir(config_dir __MKDIR_MODE_ARG);
#endif
#endif /* HAVE_SPECIAL_DIRS */
#ifdef HAVE_MULTIDRIVE
/* if this fails then alternate volumes will not work, but this function
cannot return that fact */
rbhome_fildes = os_opendirfd(rbhome);
if (rbhome_fildes >= 0)
atexit(cleanup_rbhome);
#endif /* HAVE_MULTIDRIVE */
}
#ifdef HAVE_SPECIAL_DIRS
static const char* _get_user_file_path(const char *path,
unsigned flags,
char* buf,
const size_t bufsize)
{
const char *ret = path;
const char *pos = path;
/* replace ROCKBOX_DIR in path with $HOME/.config/rockbox.org */
pos += ROCKBOX_DIR_LEN;
if (*pos == '/') pos += 1;
#if (CONFIG_PLATFORM & PLATFORM_ANDROID)
if (path_append(buf, "/sdcard/rockbox", pos, bufsize) >= bufsize)
return NULL;
#else
if (path_append(buf, rbhome, ".config/rockbox.org", bufsize) >= bufsize ||
path_append(buf, PA_SEP_SOFT, pos, bufsize) >= bufsize)
return NULL;
#endif
/* always return the replacement buffer (pointing to $HOME) if
* write access is needed */
if (flags & NEED_WRITE)
ret = buf;
else if (os_file_exists(buf))
ret = buf;
if (ret != buf) /* not found in $HOME, try ROCKBOX_BASE_DIR, !NEED_WRITE only */
{
if (path_append(buf, ROCKBOX_SHARE_PATH, pos, bufsize) >= bufsize)
return NULL;
if (os_file_exists(buf))
ret = buf;
}
return ret;
}
#endif
const char * handle_special_dirs(const char *dir, unsigned flags,
char *buf, const size_t bufsize)
{
(void) flags; (void) buf; (void) bufsize;
#ifdef HAVE_SPECIAL_DIRS
if (!strncmp(HOME_DIR, dir, HOME_DIR_LEN))
{
const char *p = dir + HOME_DIR_LEN;
while (*p == '/') p++;
snprintf(buf, bufsize, "%s/%s", rbhome, p);
dir = buf;
}
else if (!strncmp(ROCKBOX_DIR, dir, ROCKBOX_DIR_LEN))
dir = _get_user_file_path(dir, flags, buf, bufsize);
#endif
#ifdef HAVE_MULTIDRIVE
dir = handle_special_links(dir, flags, buf, bufsize);
#endif
return dir;
}
int app_open(const char *path, int oflag, ...)
{
int flags = IS_FILE;
if (oflag & O_ACCMODE)
flags |= NEED_WRITE;
char realpath[MAX_PATH];
const char *fpath = handle_special_dirs(path, flags, realpath,
sizeof (realpath));
if (!fpath)
FILE_ERROR_RETURN(ENAMETOOLONG, -1);
return os_open(fpath, oflag __OPEN_MODE_ARG);
}
int app_creat(const char *path, mode_t mode)
{
return app_open(path, O_CREAT|O_WRONLY|O_TRUNC, mode);
}
int app_remove(const char *path)
{
char realpath[MAX_PATH];
const char *fpath = handle_special_dirs(path, NEED_WRITE, realpath,
sizeof (realpath));
if (!fpath)
FILE_ERROR_RETURN(ENAMETOOLONG, -1);
return os_remove(fpath);
}
int app_rename(const char *old, const char *new)
{
char realpath_old[MAX_PATH], realpath_new[MAX_PATH];
const char *fold = handle_special_dirs(old, NEED_WRITE, realpath_old,
sizeof (realpath_old));
const char *fnew = handle_special_dirs(new, NEED_WRITE, realpath_new,
sizeof (realpath_new));
if (!fold || !fnew)
FILE_ERROR_RETURN(ENAMETOOLONG, -1);
return os_rename(fold, fnew);
}
#ifdef HAVE_SDL_THREADS
ssize_t app_read(int fd, void *buf, size_t nbyte)
{
return os_read(fd, buf, nbyte);
}
ssize_t app_write(int fd, const void *buf, size_t nbyte)
{
return os_write(fd, buf, nbyte);
}
#endif /* HAVE_SDL_THREADS */
int app_relate(const char *path1, const char *path2)
{
char realpath_1[MAX_PATH], realpath_2[MAX_PATH];
const char *fpath1 = handle_special_dirs(path1, 0, realpath_1,
sizeof (realpath_1));
const char *fpath2 = handle_special_dirs(path2, 0, realpath_2,
sizeof (realpath_2));
if (!fpath1 || !fpath2)
FILE_ERROR_RETURN(ENAMETOOLONG, -1);
return os_relate(fpath1, fpath2);
}
bool app_file_exists(const char *path)
{
char realpath[MAX_PATH];
const char *fpath = handle_special_dirs(path, NEED_WRITE, realpath,
sizeof (realpath));
if (!fpath)
FILE_ERROR_RETURN(ENAMETOOLONG, false);
return os_file_exists(fpath);
}
/* need to wrap around DIR* because we need to save the parent's directory
* path in order to determine dirinfo for volumes or convert the path to UTF-8;
* also is required to implement get_dir_info() */
struct __dir
{
OS_DIR_T *osdirp;
#ifdef HAVE_MULTIDRIVE
int volumes_returned;
#endif
int osfd;
bool osfd_is_opened;
#if defined(OS_DIRENT_CONVERT) || defined (HAVE_MULTIDRIVE)
#define USE_DIRENTP
struct dirent *direntp;
size_t d_name_size;
#endif
char path[];
};
static void __dir_free(struct __dir *this)
{
if (!this)
return;
#ifdef USE_DIRENTP
free(this->direntp);
#endif
if (this->osfd_is_opened)
os_close(this->osfd);
free(this);
}
DIR * app_opendir(const char *dirname)
{
int rc;
char realpath[MAX_PATH];
const char *fname = handle_special_dirs(dirname, 0, realpath,
sizeof (realpath));
if (!fname)
FILE_ERROR_RETURN(ENAMETOOLONG, NULL);
size_t name_len = path_strip_trailing_separators(fname, &fname);
struct __dir *this = calloc(1, sizeof (*this) + name_len + 1);
if (!this)
FILE_ERROR(ENOMEM, RC);
#ifdef USE_DIRENTP
/* allocate what we're really going to return to callers, making certain
it has at least the d_name size we want */
this->d_name_size = MAX(MAX_PATH, sizeof (this->direntp->d_name));
this->direntp = calloc(1, offsetof(typeof (*this->direntp), d_name) +
this->d_name_size);
if (!this->direntp)
FILE_ERROR(ENOMEM, RC);
/* only the d_name field will be valid but that is all that anyone may
truely count on portably existing */
#endif /* USE_DIRENTP */
strmemcpy(this->path, fname, name_len);
rc = os_opendir_and_fd(this->path, &this->osdirp, &this->osfd);
if (rc < 0)
FILE_ERROR(ERRNO, RC);
this->osfd_is_opened = rc > 0;
#ifdef HAVE_MULTIDRIVE
this->volumes_returned = INT_MAX; /* assume NOT $HOME */
if (rbhome_fildes >= 0 && os_fsamefile(rbhome_fildes, this->osfd) > 0)
Rewrite filesystem code (WIP) This patch redoes the filesystem code from the FAT driver up to the clipboard code in onplay.c. Not every aspect of this is finished therefore it is still "WIP". I don't wish to do too much at once (haha!). What is left to do is get dircache back in the sim and find an implementation for the dircache indicies in the tagcache and playlist code or do something else that has the same benefit. Leaving these out for now does not make anything unusable. All the basics are done. Phone app code should probably get vetted (and app path handling just plain rewritten as environment expansions); the SDL app and Android run well. Main things addressed: 1) Thread safety: There is none right now in the trunk code. Most of what currently works is luck when multiple threads are involved or multiple descriptors to the same file are open. 2) POSIX compliance: Many of the functions behave nothing like their counterparts on a host system. This leads to inconsistent code or very different behavior from native to hosted. One huge offender was rename(). Going point by point would fill a book. 3) Actual running RAM usage: Many targets will use less RAM and less stack space (some more RAM because I upped the number of cache buffers for large memory). There's very little memory lying fallow in rarely-used areas (see 'Key core changes' below). Also, all targets may open the same number of directory streams whereas before those with less than 8MB RAM were limited to 8, not 12 implying those targets will save slightly less. 4) Performance: The test_disk plugin shows markedly improved performance, particularly in the area of (uncached) directory scanning, due partly to more optimal directory reading and to a better sector cache algorithm. Uncached times tend to be better while there is a bit of a slowdown in dircache due to it being a bit heavier of an implementation. It's not noticeable by a human as far as I can say. Key core changes: 1) Files and directories share core code and data structures. 2) The filesystem code knows which descriptors refer to same file. This ensures that changes from one stream are appropriately reflected in every open descriptor for that file (fileobj_mgr.c). 3) File and directory cache buffers are borrowed from the main sector cache. This means that when they are not in use by a file, they are not wasted, but used for the cache. Most of the time, only a few of them are needed. It also means that adding more file and directory handles is less expensive. All one must do in ensure a large enough cache to borrow from. 4) Relative path components are supported and the namespace is unified. It does not support full relative paths to an implied current directory; what is does support is use of "." and "..". Adding the former would not be very difficult. The namespace is unified in the sense that volumes may be specified several times along with relative parts, e.g.: "/<0>/foo/../../<1>/bar" :<=> "/<1>/bar". 5) Stack usage is down due to sharing of data, static allocation and less duplication of strings on the stack. This requires more serialization than I would like but since the number of threads is limited to a low number, the tradoff in favor of the stack seems reasonable. 6) Separates and heirarchicalizes (sic) the SIM and APP filesystem code. SIM path and volume handling is just like the target. Some aspects of the APP file code get more straightforward (e.g. no path hashing is needed). Dircache: Deserves its own section. Dircache is new but pays homage to the old. The old one was not compatible and so it, since it got redone, does all the stuff it always should have done such as: 1) It may be update and used at any time during the build process. No longer has one to wait for it to finish building to do basic file management (create, remove, rename, etc.). 2) It does not need to be either fully scanned or completely disabled; it can be incomplete (i.e. overfilled, missing paths), still be of benefit and be correct. 3) Handles mounting and dismounting of individual volumes which means a full rebuild is not needed just because you pop a new SD card in the slot. Now, because it reuses its freed entry data, may rebuild only that volume. 4) Much more fundamental to the file code. When it is built, it is the keeper of the master file list whether enabled or not ("disabled" is just a state of the cache). Its must always to ready to be started and bind all streams opened prior to being enabled. 5) Maintains any short filenames in OEM format which means that it does not need to be rebuilt when changing the default codepage. Miscellaneous Compatibility: 1) Update any other code that would otherwise not work such as the hotswap mounting code in various card drivers. 2) File management: Clipboard needed updating because of the behavioral changes. Still needs a little more work on some finer points. 3) Remove now-obsolete functionality such as the mutex's "no preempt" flag (which was only for the prior FAT driver). 4) struct dirinfo uses time_t rather than raw FAT directory entry time fields. I plan to follow up on genericizing everything there (i.e. no FAT attributes). 5) unicode.c needed some redoing so that the file code does not try try to load codepages during a scan, which is actually a problem with the current code. The default codepage, if any is required, is now kept in RAM separarately (bufalloced) from codepages specified to iso_decode() (which must not be bufalloced because the conversion may be done by playback threads). Brings with it some additional reusable core code: 1) Revised file functions: Reusable code that does things such as safe path concatenation and parsing without buffer limitations or data duplication. Variants that copy or alter the input path may be based off these. To do: 1) Put dircache functionality back in the sim. Treating it internally as a different kind of file system seems the best approach at this time. 2) Restore use of dircache indexes in the playlist and database or something effectively the same. Since the cache doesn't have to be complete in order to be used, not getting a hit on the cache doesn't unambiguously say if the path exists or not. Change-Id: Ia30f3082a136253e3a0eae0784e3091d138915c8 Reviewed-on: http://gerrit.rockbox.org/566 Reviewed-by: Michael Sevakis <jethead71@rockbox.org> Tested: Michael Sevakis <jethead71@rockbox.org>
2013-08-06 02:02:45 +00:00
this->volumes_returned = 0; /* there's no place like $HOME */
#endif /* HAVE_MULTIDRIVE */
return (DIR *)this;
file_error:
__dir_free(this);
return NULL;
}
int app_closedir(DIR *dirp)
{
struct __dir *this = (struct __dir *)dirp;
if (!this)
FILE_ERROR_RETURN(EBADF, -1);
OS_DIR_T *osdirp = this->osdirp;
__dir_free(this);
return os_closedir(osdirp);
}
struct dirent * app_readdir(DIR *dirp)
{
struct __dir *this = (struct __dir *)dirp;
if (!this)
FILE_ERROR_RETURN(EBADF, NULL);
#ifdef HAVE_MULTIDRIVE
if (this->volumes_returned < NUM_VOLUMES)
{
while (++this->volumes_returned < NUM_VOLUMES)
{
if (!volume_present(this->volumes_returned))
continue;
get_volume_name(this->volumes_returned, this->direntp->d_name);
return this->direntp;
}
}
/* do normal directory reads */
#endif /* HAVE_MULTIDRIVE */
OS_DIRENT_T *osdirent = os_readdir(this->osdirp);
#ifdef OS_DIRENT_CONVERT
if (strlcpy_from_os(this->direntp->d_name, osdirent->d_name,
this->d_name_size) >= this->d_name_size)
{
this->direntp->d_name[0] = '\0';
errno = EOVERFLOW;
return NULL;
}
osdirent = (OS_DIRENT_T *)this->direntp;
#endif /* OS_DIRENT_CONVERT */
return (struct dirent *)osdirent;
}
int app_mkdir(const char *path)
{
char realpath[MAX_PATH];
const char *fname = handle_special_dirs(path, NEED_WRITE, realpath,
sizeof (realpath));
if (!fname)
FILE_ERROR_RETURN(ENAMETOOLONG, -1);
return os_mkdir(fname __MKDIR_MODE_ARG);
}
int app_rmdir(const char *path)
{
char realpath[MAX_PATH];
const char *fname = handle_special_dirs(path, NEED_WRITE, realpath,
sizeof (realpath));
if (!fname)
FILE_ERROR_RETURN(ENAMETOOLONG, -1);
return os_rmdir(fname);
}
int app_samedir(DIR *dirp1, DIR *dirp2)
{
struct __dir *this1 = (struct __dir *)dirp1;
struct __dir *this2 = (struct __dir *)dirp2;
if (!this1 || !this2)
{
errno = EBADF;
return -1;
}
return os_fsamefile(this1->osfd, this2->osfd);
}
bool app_dir_exists(const char *dirname)
{
char realpath[MAX_PATH];
const char *fname = handle_special_dirs(dirname, 0, realpath,
sizeof (realpath));
if (!fname)
FILE_ERROR_RETURN(ENAMETOOLONG, false);
OS_DIR_T *osdirp = os_opendir(fname);
if (!osdirp)
return false;
os_closedir(osdirp);
return true;
}
struct dirinfo dir_get_info(DIR *dirp, struct dirent *entry)
{
struct __dir *this = (struct __dir *)dirp;
struct dirinfo ret = { .attribute = 0,
.mtime = 0 };
Rewrite filesystem code (WIP) This patch redoes the filesystem code from the FAT driver up to the clipboard code in onplay.c. Not every aspect of this is finished therefore it is still "WIP". I don't wish to do too much at once (haha!). What is left to do is get dircache back in the sim and find an implementation for the dircache indicies in the tagcache and playlist code or do something else that has the same benefit. Leaving these out for now does not make anything unusable. All the basics are done. Phone app code should probably get vetted (and app path handling just plain rewritten as environment expansions); the SDL app and Android run well. Main things addressed: 1) Thread safety: There is none right now in the trunk code. Most of what currently works is luck when multiple threads are involved or multiple descriptors to the same file are open. 2) POSIX compliance: Many of the functions behave nothing like their counterparts on a host system. This leads to inconsistent code or very different behavior from native to hosted. One huge offender was rename(). Going point by point would fill a book. 3) Actual running RAM usage: Many targets will use less RAM and less stack space (some more RAM because I upped the number of cache buffers for large memory). There's very little memory lying fallow in rarely-used areas (see 'Key core changes' below). Also, all targets may open the same number of directory streams whereas before those with less than 8MB RAM were limited to 8, not 12 implying those targets will save slightly less. 4) Performance: The test_disk plugin shows markedly improved performance, particularly in the area of (uncached) directory scanning, due partly to more optimal directory reading and to a better sector cache algorithm. Uncached times tend to be better while there is a bit of a slowdown in dircache due to it being a bit heavier of an implementation. It's not noticeable by a human as far as I can say. Key core changes: 1) Files and directories share core code and data structures. 2) The filesystem code knows which descriptors refer to same file. This ensures that changes from one stream are appropriately reflected in every open descriptor for that file (fileobj_mgr.c). 3) File and directory cache buffers are borrowed from the main sector cache. This means that when they are not in use by a file, they are not wasted, but used for the cache. Most of the time, only a few of them are needed. It also means that adding more file and directory handles is less expensive. All one must do in ensure a large enough cache to borrow from. 4) Relative path components are supported and the namespace is unified. It does not support full relative paths to an implied current directory; what is does support is use of "." and "..". Adding the former would not be very difficult. The namespace is unified in the sense that volumes may be specified several times along with relative parts, e.g.: "/<0>/foo/../../<1>/bar" :<=> "/<1>/bar". 5) Stack usage is down due to sharing of data, static allocation and less duplication of strings on the stack. This requires more serialization than I would like but since the number of threads is limited to a low number, the tradoff in favor of the stack seems reasonable. 6) Separates and heirarchicalizes (sic) the SIM and APP filesystem code. SIM path and volume handling is just like the target. Some aspects of the APP file code get more straightforward (e.g. no path hashing is needed). Dircache: Deserves its own section. Dircache is new but pays homage to the old. The old one was not compatible and so it, since it got redone, does all the stuff it always should have done such as: 1) It may be update and used at any time during the build process. No longer has one to wait for it to finish building to do basic file management (create, remove, rename, etc.). 2) It does not need to be either fully scanned or completely disabled; it can be incomplete (i.e. overfilled, missing paths), still be of benefit and be correct. 3) Handles mounting and dismounting of individual volumes which means a full rebuild is not needed just because you pop a new SD card in the slot. Now, because it reuses its freed entry data, may rebuild only that volume. 4) Much more fundamental to the file code. When it is built, it is the keeper of the master file list whether enabled or not ("disabled" is just a state of the cache). Its must always to ready to be started and bind all streams opened prior to being enabled. 5) Maintains any short filenames in OEM format which means that it does not need to be rebuilt when changing the default codepage. Miscellaneous Compatibility: 1) Update any other code that would otherwise not work such as the hotswap mounting code in various card drivers. 2) File management: Clipboard needed updating because of the behavioral changes. Still needs a little more work on some finer points. 3) Remove now-obsolete functionality such as the mutex's "no preempt" flag (which was only for the prior FAT driver). 4) struct dirinfo uses time_t rather than raw FAT directory entry time fields. I plan to follow up on genericizing everything there (i.e. no FAT attributes). 5) unicode.c needed some redoing so that the file code does not try try to load codepages during a scan, which is actually a problem with the current code. The default codepage, if any is required, is now kept in RAM separarately (bufalloced) from codepages specified to iso_decode() (which must not be bufalloced because the conversion may be done by playback threads). Brings with it some additional reusable core code: 1) Revised file functions: Reusable code that does things such as safe path concatenation and parsing without buffer limitations or data duplication. Variants that copy or alter the input path may be based off these. To do: 1) Put dircache functionality back in the sim. Treating it internally as a different kind of file system seems the best approach at this time. 2) Restore use of dircache indexes in the playlist and database or something effectively the same. Since the cache doesn't have to be complete in order to be used, not getting a hit on the cache doesn't unambiguously say if the path exists or not. Change-Id: Ia30f3082a136253e3a0eae0784e3091d138915c8 Reviewed-on: http://gerrit.rockbox.org/566 Reviewed-by: Michael Sevakis <jethead71@rockbox.org> Tested: Michael Sevakis <jethead71@rockbox.org>
2013-08-06 02:02:45 +00:00
if (!this)
FILE_ERROR_RETURN(EBADF, ret);
if (!entry || entry->d_name[0] == '\0')
FILE_ERROR_RETURN(ENOENT, ret);
char path[MAX_PATH];
#ifdef HAVE_MULTIDRIVE
if (this->volumes_returned < NUM_VOLUMES)
{
/* last thing read was a "symlink" */
ret.attribute = ATTR_LINK;
strcpy(path, MULTIDRIVE_DIR);
}
else
#endif
if (path_append(path, this->path, entry->d_name, sizeof (path))
>= sizeof (path))
{
FILE_ERROR_RETURN(ENAMETOOLONG, ret);
}
struct stat s;
if (os_lstat(path, &s) < 0)
FILE_ERROR_RETURN(ERRNO, ret);
int err = 0;
if (S_ISLNK(s.st_mode))
{
ret.attribute |= ATTR_LINK;
err = os_stat(path, &s);
}
if (err < 0)
FILE_ERROR_RETURN(ERRNO, ret);
if (S_ISDIR(s.st_mode))
ret.attribute |= ATTR_DIRECTORY;
ret.size = s.st_size;
struct tm tm;
if (!localtime_r(&s.st_mtime, &tm))
FILE_ERROR_RETURN(ERRNO, ret);
ret.mtime = mktime(&tm);
return ret;
}
/* On MD we create a virtual symlink for the external drive,
* for this we need to override readlink(). */
ssize_t app_readlink(const char *path, char *buf, size_t bufsiz)
{
char _buf[MAX_PATH];
path = handle_special_dirs(path, 0, _buf, sizeof(_buf));
#ifdef HAVE_MULTIDRIVE
/* if path == _buf then we can be sure handle_special_dir() did something
* and path is not an ordinary directory */
if (path == _buf && !strncmp(path, MULTIDRIVE_DIR, sizeof(MULTIDRIVE_DIR)-1))
{
/* copying NUL is not required as per readlink specification */
ssize_t len = strlen(path);
memcpy(buf, path, len);
return len;
}
#endif
/* does not append NUL !! */
return os_readlink(path, buf, bufsiz);
(void) path; (void) buf; (void) bufsiz;
}
int os_volume_path(IF_MV(int volume, ) char *buffer, size_t bufsize)
{
#ifdef HAVE_MULTIVOLUME
char volname[VOL_MAX_LEN + 1];
get_volume_name(volume, volname);
#else
const char *volname = "/";
#endif
if (!handle_special_dirs(volname, NEED_WRITE, buffer, bufsize))
{
errno = ENAMETOOLONG;
return -1;
}
return 0;
}