aboutsummaryrefslogtreecommitdiff
path: root/ubi-utils/README
diff options
context:
space:
mode:
authorFrank Haverkamp <haver@vnet.ibm.com>2006-06-14 11:53:59 +0200
committerFrank Haverkamp <haver@vnet.ibm.com>2006-10-31 15:06:06 +0100
commitf175083413f0f94de88def865eeb65e465ded389 (patch)
treef50ded679736272988ccce2a15d17fdeac2e09a5 /ubi-utils/README
parent37f40f5574e04ae050507133ade8fe0e6bae2f0d (diff)
UBI - Unsorted Block Images
UBI (Latin: "where?") manages multiple logical volumes on a single flash device, specifically supporting NAND flash devices. UBI provides a flexible partitioning concept which still allows for wear-levelling across the whole flash device. In a sense, UBI may be compared to the Logical Volume Manager (LVM). Whereas LVM maps logical sector numbers to physical HDD sector numbers, UBI maps logical eraseblocks to physical eraseblocks. More information may be found in the UBI design documentation: ubidesign.pdf. Which can be found here: http://www.linux-mtd.infradead.org/doc/ubi.html Partitioning/Re-partitioning An UBI volume occupies a certain number of erase blocks. This is limited by a configured maximum volume size, which could also be viewed as the partition size. Each individual UBI volume's size can be changed independently of the other UBI volumes, provided that the sum of all volume sizes doesn't exceed a certain limit. UBI supports dynamic volumes and static volumes. Static volumes are read-only and their contents are protected by CRC check sums. Bad eraseblocks handling UBI transparently handles bad eraseblocks. When a physical eraseblock becomes bad, it is substituted by a good physical eraseblock, and the user does not even notice this. Scrubbing On a NAND flash bit flips can occur on any write operation, sometimes also on read. If bit flips persist on the device, at first they can still be corrected by ECC, but once they accumulate, correction will become impossible. Thus it is best to actively scrub the affected eraseblock, by first copying it to a free eraseblock and then erasing the original. The UBI layer performs this type of scrubbing under the covers, transparently to the UBI volume users. Erase Counts UBI maintains an erase count header per eraseblock. This frees higher-level layers (like file systems) from doing this and allows for centralized erase count management instead. The erase counts are used by the wear-levelling algorithm in the UBI layer. The algorithm itself is exchangeable. Booting from NAND For booting directly from NAND flash the hardware must at least be capable of fetching and executing a small portion of the NAND flash. Some NAND flash controllers have this kind of support. They usually limit the window to a few kilobytes in erase block 0. This "initial program loader" (IPL) must then contain sufficient logic to load and execute the next boot phase. Due to bad eraseblocks, which may be randomly scattered over the flash device, it is problematic to store the "secondary program loader" (SPL) statically. Also, due to bit-flips it may become corrupted over time. UBI allows to solve this problem gracefully by storing the SPL in a small static UBI volume. UBI volumes vs. static partitions UBI volumes are still very similar to static MTD partitions: * both consist of eraseblocks (logical eraseblocks in case of UBI volumes, and physical eraseblocks in case of static partitions; * both support three basic operations - read, write, erase. But UBI volumes have the following advantages over traditional static MTD partitions: * there are no eraseblock wear-leveling constraints in case of UBI volumes, so the user should not care about this; * there are no bit-flips and bad eraseblocks in case of UBI volumes. So, UBI volumes may be considered as flash devices with relaxed restrictions. Where can it be found? Documentation, kernel code and applications can be found in the MTD gits. What are the applications for? The applications help to create binary flash images for two purposes: pfi files (partial flash images) for in-system update of UBI volumes, and plain binary images, with or without OOB data in case of NAND, for a manufacturing step. Furthermore some tools are/and will be created that allow flash content analysis after a system has crashed. Who did UBI? The original ideas, where UBI is based on, were developed by Andreas Arnez, Frank Haverkamp and Thomas Gleixner. Josh W. Boyer and some others were involved too. The implementation of the kernel layer was done by Artem B. Bityutskiy. The user-space applications and tools were written by Oliver Lohmann with contributions from Frank Haverkamp, Andreas Arnez, and Artem. Joern Engel contributed a patch which modifies JFFS2 so that it can be run on a UBI volume. Thomas Gleixner did modifications to the NAND layer and also some to JFFS2 to make it work. Signed-off-by: Frank Haverkamp <haver@vnet.ibm.com>
Diffstat (limited to 'ubi-utils/README')
-rw-r--r--ubi-utils/README236
1 files changed, 236 insertions, 0 deletions
diff --git a/ubi-utils/README b/ubi-utils/README
new file mode 100644
index 0000000..d976a76
--- /dev/null
+++ b/ubi-utils/README
@@ -0,0 +1,236 @@
+README
+======
+
+The programs and libraries in this directory provide a tool-chain to
+generate binary data for embedded systems which can be flashed either
+by a hardware flash programmer, e.g. JTAG debugger, or on the target
+system directly using pfiflash, or ubimkvol, ubirmvol, ubiwritevol.
+
+The latter is the case when there is already Linux running which has
+build in UBI support.
+
+Authors: Oliver Lohmann
+ Frank Haverkamp
+ Andreas Arnez
+
+mkpfi - tool for flash content generation in PFI
+ format
+pfi2bin - conversion tool to transfer a PFI file into a
+ binary image
+pfiflash - tool to update the embedded systems flash using
+ pfi files created by mkpfi
+libbootenv - library for boot-parameter processing
+libpfi - library for partial flash image (PFI) creation
+ and handling
+ubigen - tool to create binary UBI images e.g. for a
+ jtag flashing tool
+nandimg - tool to add OOB data to binary images intended
+ for NAND flash systems
+ubilib - UBI library
+
+!!! NOTICE !!!
+If you execute ./configure in the top_level directory the helper Makefile
+gets overwritten. Thats actually no problem, but be aware of that.
+
+1. Build Process
+
+1.1 Build, install and forget
+ o Build all and everything
+ $make all (takes a while, builds ppc and x86 binaries/libs)
+ o Installation:
+ $make install
+ o Uninstallation:
+ $make uninstall
+
+ o x86 only would be:
+ $make x86 && make install_x86
+
+1.2 Usage for a developer
+
+ 1.2.1 The build process in detail
+
+ o If you've checked out the sources from the CVS repository you'll find a
+ directory setup like this:
+
+ flashutils/
+ -rw-r--r-- 1 olli olli 1.3K Mar 14 11:53 Makefile
+ -rw-r--r-- 1 olli olli 1.9K Mar 14 10:50 Makefile.am
+ -rwxr-xr-x 1 olli olli 265 Mar 9 00:47 bootstrap
+ -rw-r--r-- 1 olli olli 1.1K Mar 9 16:55 configure.ac
+ drwxr-xr-x 2 olli olli 4.0K Mar 9 00:28 doc
+ drwxr-xr-x 2 olli olli 4.0K Mar 14 11:56 inc
+ drwxr-xr-x 2 olli olli 4.0K Mar 14 11:56 lib
+ drwxr-xr-x 17 olli olli 4.0K Mar 13 16:50 src
+
+ o To generate the initial build templates you have to call the bootstrap
+ script:
+ $ ./bootstrap
+ o Create a directory for the target platform
+ $ mkdir build_x86
+ o Descend into the directory and call the top-level configure script
+ with the desired options.
+ $ cd build_x86
+ $ ../configure --prefix=/usr/local [...]
+ o Now you'll find a directory structure like this:
+
+ flashutils/build_x86/
+ -rw-r--r-- 1 olli olli 47K Mar 14 13:33 Makefile
+ -rw-r--r-- 1 olli olli 33K Mar 14 13:33 config.log
+ -rwxr-xr-x 1 olli olli 38K Mar 14 13:33 config.status
+ drwxr-xr-x 2 olli olli 4.0K Mar 14 13:33 inc
+ drwxr-xr-x 3 olli olli 4.0K Mar 14 13:33 lib
+ -rwxr-xr-x 1 olli olli 202K Mar 14 13:33 libtool
+
+ o The config.guess script can be used to update the Makefiles in the
+ target directory after a change of the top-level template files
+ (i.e. the Makefile.in files).
+ $ ./config.guess
+ o To compile everything for this platform just invoke make in
+ flashutils/build_x86:
+ $ make
+ or from toplevel:
+ $ make -C ./build_x86
+ o The build process creates a new directory "bin":
+ flashutils/build_x86/
+ [...]
+ drwxr-xr-x 3 olli olli 4.0K Mar 14 13:41 bin
+ [...]
+
+ This directory contains all binary files which will be installed
+ by make install, e.g.:
+
+ flashutils/build_x86/bin/
+ -rwxr-xr-x 1 olli olli 7.2K Mar 14 13:41 bin2nand
+ -rwxr-xr-x 1 olli olli 15K Mar 14 13:41 mkbootenv
+ -rwxr-xr-x 1 olli olli 16K Mar 14 13:41 pddcustomize
+ -rwxr-xr-x 1 olli olli 36K Mar 14 13:41 pfi2bin
+ -rwxr-xr-x 1 olli olli 6.8K Mar 14 13:41 pfiflash
+ -rwxr-xr-x 1 olli olli 5.0K Mar 14 13:41 ubicrc32
+ -rwxr-xr-x 1 olli olli 13K Mar 14 13:41 ubigen
+ -rwxr-xr-x 1 olli olli 6.3K Mar 14 13:41 ubimirror
+
+
+ 1.2.2 Modifying and Adding Sources
+
+ o There is a dedicated directory which contains all source code
+ of the flashutils package, e.g.:
+
+ flashutils/src/
+ drwxr-xr-x 2 olli olli 4.0K Mar 13 11:42 libbootenv
+ drwxr-xr-x 2 olli olli 4.0K Mar 13 11:42 liberror
+ drwxr-xr-x 2 olli olli 4.0K Mar 13 16:48 mkpfi
+ drwxr-xr-x 2 olli olli 4.0K Mar 13 16:12 pddcustomize
+
+
+
+ The prefix "lib" is used to mark directories as part of a convenience
+ library. Binaries have no special prefix.
+
+ o How to add sources?
+
+ Just create a new directory at flashutils/src/, e.g.:
+
+ For a binary:
+ $ mkdir rider
+ $ cd rider
+ $ vi rider.c
+ /* do sth with that file... */
+
+ For a convenience library (as well as for "normal libs")
+ $ mkdir libworld
+ $ cd libworld
+ $ vi world.c
+ /* do sth with that file... */
+
+ o How to register sources in the build process (for binaries)?
+
+ You have to register your sources at the top-level automake Makefile:
+
+ In directory flashutils/
+ $ vi Makefile.am
+
+ Binaries have to be registered at "bin_PROGRAMS", e.g.:
+ bin_PROGRAMS = bin/pddcustomize \
+ bin/rider
+
+ Add the rule how the binary is assembled, e.g.:
+ bin_pddcustomize_SOURCES = \
+ $(top_srcdir)/src/pddcustomize/pddcustomize.c
+ bin_pddcustomize_LDADD = \
+ $(top_builddir)/lib/libbootenv.la \
+ $(top_builddir)/lib/liberror.la
+
+ bin_rider_SOURCES = \
+ $(top_srcdir)/src/rider/rider.c
+
+ This example reflects a simple build process for "rider". "rider"
+ is built without any other dependencies or convenience libraries.
+ The example for pddcustomize is a bit more complicated.
+ "_LDADD" adds some convenience libraris into the link process of
+ "pddcustomize". Imagine, that your "rider" has common code
+ with "dragon_bin" which is held in a library called "libworld".
+ The build rules would like like the following:
+
+ bin_rider_SOURCES = \
+ $(top_srcdir)/src/rider/rider.c
+ bin_rider_LDADD = \
+ $(top_builddir)/lib/libworld.la
+
+ bin_dragon_SOURCES = \
+ $(top_srcdir)/src/dragon_bin/dragon_bin.c
+ bin_dragon_LDADD = \
+ $(top_builddir)/lib/libworld.la
+
+ Don't forget to add "dragon" to "bin_PROGRAMS"!
+ Don't forget to set the build rule for the "libworld" itself!
+ This is documented in the next section.
+
+
+ o How to register sources in the build process (for libraries)?
+
+ Until now we didn't care about the build process of "libworld".
+ Libraries are handled special in this build process because
+ they are handled as "modules", i.e. they are able to be built
+ without building the binaries in the same step. Additionally,
+ it is possible to assemble complex libraries out of simple ones.
+ That especially makes sense if you want to export (install) a
+ library on a system which uses some common code and makes
+ some adoptions for usability and presents a comfortable interface to
+ the user (see libpfiflash in the sources for an example).
+
+ o Registering "libworld" as convenience library.
+
+ Instead of editing the "Makefile.am" in "flashtools/", we have to
+ edit now the "Makefile.am" in "flashtools/lib/":
+
+ noinst_LTLIBRARIES = libworld.la
+
+ libworld_la_SOURCES = $(top_srcdir)/src/libworld/world.c
+
+ o Registering "libworld" as library which gets installed.
+
+ lib_LTLIBRARIES = libworld.la
+ libworld_la_SOURCES = $(top_srcdir)/src/libworld/world.c
+ libworld_la_LDFLAGS = -no-undefined -version-info 0:0:0
+
+ o Header files
+
+ All header files are stored at "flashutils/inc", regardless
+ if convenience library or not.
+
+ If you want to export headers you have to specify this in the Makefile.am
+ located at "flashutils/inc", e.g. (this should not be done
+ for convenience libraries):
+
+ nobase_include_HEADERS = world.h
+
+
+
+Appendix
+
+A.1. FAQ
+
+ Q How to call configure to setup a cross-platform build?
+ A $ ./configure --build=i686-pc-linux-gnu --host=ppc-linux \
+ --prefix=/opt/.../ppcnf/crossroot/ \
+ --exec-prefix=/opt/..../ppcnf/crossroot/usr