This is the procedure and patch to set up Ubuntu 12.04 64bit Linux Host PC and building i.MX28 L2.6.35_1.1.0_130130.  It has been tested to build GNOME profile and with FSL Standard MM codec. A) Basic Requirement: Set up the Linux Host PC using ubuntu-12.04.3-desktop-amd64.iso Make sure the previous LTIB installation and the /opt/freescale have been removed B) Installed the needed packages to the Linux Host PC $ sudo apt-get update $ sudo apt-get install gettext libgtk2.0-dev rpm bison m4 libfreetype6-dev $ sudo apt-get install libdbus-glib-1-dev liborbit2-dev intltool $ sudo apt-get install ccache ncurses-dev zlib1g zlib1g-dev gcc g++ libtool $ sudo apt-get install uuid-dev liblzo2-dev $ sudo apt-get install tcl dpkg $ sudo apt-get install asciidoc texlive-latex-base dblatex xutils-dev $ sudo apt-get install texlive texinfo $ sudo apt-get install ia32-libs libc6-dev-i386 lib32z1 $ sudo apt-get install uboot-mkimage $ sudo apt-get install scrollkeeper $ sudo apt-get install gparted $ sudo apt-get install nfs-common nfs-kernel-server $ sudo apt-get install git-core git-doc git-email git-gui gitk $ sudo apt-get install meld atftpd C) Unpack and install the LTIB source package and assume done on the home directory: $ cd ~ $ tar -zxvf L2.6.35_1.1.0_130130_source.tar.gz $ ./L2.6.35_1.1.0_130130_source/install After that, you will find ~/ltib directory created D) Apply the patch to make L2.6.35_1.1.0 could be installed and compiled on Ubuntu 12.04 64bit OS $ cd ~/ltib $ git apply 0001_make_L2.6.35_1.1.0_130130_compile_on_ubuntu_12.04_64bit_OS.patch a) The patch modifies the following files:    dist/lfs-5.1/base_libs/base_libs.spec    dist/lfs-5.1/lkc/lkc.spec    dist/lfs-5.1/mux_server/mux_server.spec    dist/lfs-5.1/ncurses/ncurses.spec b) Add the following files to the pkgs directory:    pkgs/lkc-1.4-lib.patch    pkgs/lkc-1.4-lib.patch.md5 E) Then, it is ready to proceed the rest of the LTIB env setup process: $ cd ~/ltib $ ./ltib -m config $ ./ltib Reference: L2.6.35_1.1.0_130130_docs/doc/mx28/Setting_Up_LTIB_Host_on_Ubuntu_9_04.pdf https://community.freescale.com/docs/DOC-93394 https://community.freescale.com/message/332385#332385 https://community.freescale.com/thread/271675 https://community.freescale.com/message/360556#360556 scrollkeeper is for the gnome-desktop compilation NOTE: When compiling gstreamer, this warning was pop up.  Just ignore it seems okay.

This is the procedure and patch to set up Ubuntu 12.04 64bit Linux Host PC and building i.MX6x L3.0.35_4.1.0.  It has been tested to build GNOME profile and with FSL Standard MM Codec for i.MX6Q SDB board. A) Basic Requirement: Set up the Linux Host PC using ubuntu-12.04.3-desktop-amd64.iso Make sure the previous LTIB installation and the /opt/freescale have been removed B) Installed the needed packages to the Linux Host PC $ sudo apt-get update $ sudo apt-get install gettext libgtk2.0-dev rpm bison m4 libfreetype6-dev $ sudo apt-get install libdbus-glib-1-dev liborbit2-dev intltool $ sudo apt-get install ccache ncurses-dev zlib1g zlib1g-dev gcc g++ libtool $ sudo apt-get install uuid-dev liblzo2-dev $ sudo apt-get install tcl dpkg $ sudo apt-get install asciidoc texlive-latex-base dblatex xutils-dev $ sudo apt-get install texlive texinfo $ sudo apt-get install ia32-libs libc6-dev-i386 lib32z1 $ sudo apt-get install uboot-mkimage $ sudo apt-get install scrollkeeper $ sudo apt-get install gparted $ sudo apt-get install nfs-common nfs-kernel-server $ sudo apt-get install git-core git-doc git-email git-gui gitk $ sudo apt-get install meld atftpd C) Unpack and install the LTIB source package and assume done on the home directory: $ cd ~ $ tar -zxvf L3.0.35_4.1.0_130816_source.tar.gz $ ./L3.0.35_4.1.0_130816_source/install After that, you will find ~/ltib directory created D) Apply the patch to make L3.0.35_4.1.0 could be installed and compiled on Ubuntu 12.04 64bit OS $ cd ~/ltib $ git apply 0001_make_L3.0.35_4.1.0_compile_on_Ubuntu_12.04_64bit_OS.patch The patch modifies the following files: dist/lfs-5.1/base_libs/base_libs.spec dist/lfs-5.1/ncurses/ncurses.spec E) Then, it is ready to proceed the rest of the LTIB env setup process: $ cd ~/ltib $ ./ltib -m config $ ./ltib Reference: L3.0.35_4.1.0_130816_docs/doc/mx6/Setting_Up_LTIB_host.pdf https://community.freescale.com/message/332385#332385 https://community.freescale.com/thread/271675 https://community.freescale.com/message/360556#360556 scrollkeeper is for the gnome-desktop compilation NOTE: When compiling gstreamer, this warning was pop up.  Just ignore it seems okay.

Notes: + Run the pipelines in the presented order + The above example streams H263 video. + the gl command is equal to 'gst-launch' (two instead of 'gst-launch'.size() chars ) + Pending work: H264 test cases and other scenarios. Scenario Shell variables and pipelines # Export always these variables on the i.MX export VSALPHA=1 export WIDTH=320 export HEIGHT=240 export SEP=20 # decoded and displayed Uni-directional: from PC to i.MX. PC is streaming 4 H.263 streams and i.MX displays all in the screen. # On i.MX (Target) gl udpsrc caps='application/x-rtp, media=(string)video, clock-rate=(int)90000, encoding-name=(string)H263' port=8890 ! rtph263depay ! vpudec ! mfw_isink sync=false axis-top=0 axis-left=0 disp-width=$WIDTH disp-height=$HEIGHT & gl udpsrc caps='application/x-rtp, media=(string)video, clock-rate=(int)90000, encoding-name=(string)H263' port=8891 ! rtph263depay ! vpudec ! mfw_isink sync=false axis-top=0 axis-left=`expr $WIDTH + $SEP` disp-width=$WIDTH disp-height=$HEIGHT & gl udpsrc caps='application/x-rtp, media=(string)video, clock-rate=(int)90000, encoding-name=(string)H263' port=8892 ! rtph263depay ! vpudec ! mfw_isink sync=false axis-top=`expr $HEIGHT + $SEP` axis-left=0   disp-width=$WIDTH disp-height=$HEIGHT & gl udpsrc caps='application/x-rtp, media=(string)video, clock-rate=(int)90000, encoding-name=(string)H263' port=8893 ! rtph263depay ! vpudec ! mfw_isink sync=false axis-top=`expr $HEIGHT + $SEP` axis-left=`expr $WIDTH + $SEP` disp-width=$WIDTH disp-height=$HEIGHT & # On PC (Source) export IP_iMX= # Place the IP address of the i.MX board gst-launch -v videotestsrc ! ffenc_h263 ! rtph263pay ! multiudpsink clients=IP_iMX:8890,IP_iMX:8891,IP_iMX:8892,$IP_iMX:8893 Uni-directional: from PC to i.MX. PC is streaming one H.264 stream and i.MX displays it on the screen # On i.MX (Target) # Make sure you set the caps correctly, specially the sprop-parameter-sets cap. The one show below is just an example and works with the source file sintel_trailer-1080p.mp4 export VSALPHA=1 GST_DEBUG=*:2 gst-launch -v udpsrc caps='application/x-rtp, media=(string)video, clock-rate=(int)90000, encoding-name=(string)H264, sprop-parameter-sets=(string)\"Z2QAMqw05gHgCJ+WEAAAAwAQAAADAwDxgxmg\\,aOl4TLIs\", payload=(int)96' port=8890 ! rtph264depay ! vpudec ! mfw_isink sync=false # On PC (Source) gst-launch -v filesrc location=sintel_trailer-1080p.mp4 typefind=true ! qtdemux ! rtph264pay ! multiudpsink clients=10.112.102.168:8890 Bi-directional: PC is streaming 4 H.263 streams to i.MX, iMX displays it and sends the four back to PC # On i.MX export IP_PC= # Place the IP address of the PC host machine gl -v udpsrc caps='application/x-rtp, media=(string)video, clock-rate=(int)90000, encoding-name=(string)H263' port=8890 ! rtph263depay ! vpudec ! tee name=t ! queue ! mfw_isink sync=false axis-top=0 axis-left=0 disp-width=$WIDTH disp-height=$HEIGHT t. ! queue ! vpuenc codec=5 ! rtph263pay ! udpsink host=$IP_PC port=9990 & gl -v udpsrc caps='application/x-rtp, media=(string)video, clock-rate=(int)90000, encoding-name=(string)H263' port=8891 ! rtph263depay ! vpudec ! tee name=t ! queue ! mfw_isink sync=false axis-top=0 axis-left=`expr $WIDTH + $SEP` disp-width=$WIDTH disp-height=$HEIGHT t. ! queue ! vpuenc codec=5 ! rtph263pay ! udpsink host=$IP_PC port=9991 & gl -v udpsrc caps='application/x-rtp, media=(string)video, clock-rate=(int)90000, encoding-name=(string)H263' port=8892 ! rtph263depay ! vpudec ! tee name=t ! queue ! mfw_isink sync=false axis-top=`expr $HEIGHT + $SEP` axis-left=0   disp-width=$WIDTH disp-height=$HEIGHT t. ! queue ! vpuenc codec=5 ! rtph263pay ! udpsink host=$IP_PC port=9992 & gl -v udpsrc caps='application/x-rtp, media=(string)video, clock-rate=(int)90000, encoding-name=(string)H263' port=8893 ! rtph263depay ! vpudec ! tee name=t ! queue ! mfw_isink sync=false axis-top=`expr $HEIGHT + $SEP` axis-left=`expr $WIDTH + $SEP` disp-width=$WIDTH disp-height=$HEIGHT t. ! queue ! vpuenc codec=5 ! rtph263pay ! udpsink host=$IP_PC port=9993 & # On PC ## Stream received from iMX export IP_iMX= # Place the IP address of the i.MX board gl -v udpsrc caps='application/x-rtp, media=(string)video, clock-rate=(int)90000, encoding-name=(string)H263' port=9990 ! rtph263depay ! ffdec_h263 ! xvimagesink & gl -v udpsrc caps='application/x-rtp, media=(string)video, clock-rate=(int)90000, encoding-name=(string)H263' port=9991 ! rtph263depay ! ffdec_h263 ! xvimagesink & gl -v udpsrc caps='application/x-rtp, media=(string)video, clock-rate=(int)90000, encoding-name=(string)H263' port=9992 ! rtph263depay ! ffdec_h263 ! xvimagesink & gl -v udpsrc caps='application/x-rtp, media=(string)video, clock-rate=(int)90000, encoding-name=(string)H263' port=9993 ! rtph263depay ! ffdec_h263 ! xvimagesink & ## Stream sent to iMX gl -v videotestsrc ! videoscale ! video/x-raw-yuv,width=\(int\)1408,height=\(int\)1152 !  ffenc_h263 ! rtph263pay ! udpsink host=$IP_iMX port=8890 & gl -v videotestsrc ! videoscale ! video/x-raw-yuv,width=\(int\)1408,height=\(int\)1152 ! ffenc_h263 ! rtph263pay ! udpsink host=$IP_iMX port=8891 & gl -v videotestsrc ! videoscale ! video/x-raw-yuv,width=\(int\)1408,height=\(int\)1152 ! ffenc_h263 ! rtph263pay ! udpsink host=$IP_iMX port=8892 & gl -v videotestsrc ! videoscale ! video/x-raw-yuv,width=\(int\)1408,height=\(int\)1152 ! ffenc_h263 ! rtph263pay ! udpsink host=$IP_iMX port=8893 &

Kernel provides mtdoops to dump kmsg to MTD device, but MMC card is not a MTD device. We can let user-space program to execute the write operation to dump kmsg into block storage. The sample code is below. kernel space -- #include <linux/kernel.h> #include <linux/module.h> #include <linux/console.h> #include <linux/vmalloc.h> #include <linux/seq_file.h> #include <linux/workqueue.h> #include <linux/sched.h> #include <linux/wait.h> #include <linux/delay.h> #include <linux/interrupt.h> #include <linux/kmsg_dump.h> #include <linux/proc_fs.h> static struct kmsg_dumper dump; static struct proc_dir_entry *my_proc; static int is_panic = 0; static int my_proc_show(struct seq_file *m, void *v) {     seq_printf(m, "%d", is_panic);     return 0; } static int my_proc_open(struct inode *inode, struct file *file) {     return single_open(file, my_proc_show, NULL); } static const struct file_operations my_proc_ops = {     .open        = my_proc_open,     .read        = seq_read,     .llseek        = seq_lseek,     .release    = single_release, }; static void oops_do_dump(struct kmsg_dumper *dumper,         enum kmsg_dump_reason reason, const char *s1, unsigned long l1,         const char *s2, unsigned long l2) {     int i;     printk("### [%s:%d] reason = %d\n", __func__, __LINE__, reason);     is_panic = 1;     for (i = 0;i < 10; i++)         msleep(1000);     printk("### [%s:%d] should be done\n", __func__, __LINE__); } static int __init my_oops_init(void) {     int err;     dump.dump = oops_do_dump;     err = kmsg_dump_register(&dump);     if (err) {         printk(KERN_ERR "oops: registering kmsg dumper failed, error %d\n", err);         return -EINVAL;     }     my_proc = proc_create("dump_tester", 0, NULL, &my_proc_ops);     return 0; } static void __exit my_oops_exit(void) {     printk("### [%s:%d]\n", __func__, __LINE__);     if (my_proc)         remove_proc_entry( "dump_tester", NULL);     kmsg_dump_unregister(&dump); } module_init(my_oops_init); module_exit(my_oops_exit); MODULE_LICENSE("GPL"); User space -- #include <stdio.h> #include <stdlib.h> #include <string.h> #include <sys/types.h> #include <sys/stat.h> #include <unistd.h> #include <fcntl.h> #include <poll.h> #define BUF_LEN 40960 void main(int argc, char **argv) {     char tmp = 'X';     char buf[BUF_LEN];     int fd_src, fd_trg;     int fd = open("/proc/dump_tester", O_RDONLY, 0);     while(1) {         lseek(fd, 0, SEEK_SET);         read(fd, &tmp, 1);         //printf("### [%s:%d] ==> '%c'\n", __FUNCTION__, __LINE__, tmp);         if (tmp == '1') {             fd_src = open("/proc/kmsg", O_RDONLY, 0);             fd_trg = open("/dev/block/mmcblk0p6",  O_RDWR, 0);             memset(buf, 0, BUF_LEN);             write(fd_trg, buf, BUF_LEN);             lseek(fd_trg, 0, SEEK_SET);             read(fd_src, buf, BUF_LEN);             write(fd_trg, buf, BUF_LEN);             close(fd_src);             close(fd_trg);             sleep(1);             printf("### dump panic log into %s\n", "/dev/block/mmcblk0p6");             break;         }         sleep(1);     }     close(fd); }

Question: Using Linux SDK 4.1.0, with CAAM drivers enabled, there is little noticeable difference in the performance of openssd compared to a kernel without the CAAM drivers. Tests were done using openssd. Test image AES-128 8192 byte block (M Bytes/sec) “openssl speed –evp aes-128-cbc” AES-128 8192 byte block (M Bytes/sec) With /dev/crypto “openssl speed –evp aes-128-cbc -engine cryptodev”  Ubuntu 11.04 Image 19.010 N/A Timesys 20.518 N/A SDK 4.1.0 LTIB 22.013 21.984 (errors reported) One can see that with SDK 4.1.0, performance is worse with crypto enabled.  This is probably due to the overhead of a faulty driver or incorrect implementation. The lowest number is for Ubuntu which could be attributed to the Unity GUI. Conclusion:  CAAM driver is not functional or I am using an improper testing procedure. Test Procedure: Board used is iMX6Q Sabre SDP Openssl was used for testing. Two command line commands were used, with and without the cryptodev engine. openssl speed –evp aes-128-cbc openssl speed –evp aes-128-cbc -engine cryptodev Openssl versions used in each build are slightly different: Ubuntu:              openssl 1.0.0e Timesys:              openssl  1.0.1e SDK 4.1.0:            openssl  1.0.1c Three versions of Linux were tested. Default kernel  4.0.0 with Ubuntu rootfs form image tarballs. Timesys kernel and root file system Kernel built with SDK 4.1.0 using LTIB with hardware crypto enabled Both 1 and 2 above did not have CRYPTODEV set in .config which contains the line “# CONFIG_CRYPTO_CRYPTODEV is not set” Option 3 had the line in .config as, “CONFIG_CRYPTO_CRYPTODEV=y” All three builds generate “/proc/crypto”  whose contents are attached.  A partial listing of /proc/crypto lists “caam” as a driver for all encryption methods supported.  Example printout for aes shown below: ame         : cbc(aes) driver       : cbc-aes-caam module       : kernel priority     : 3000 refcnt       : 1 selftest     : passed type         : ablkcipher async        : yes blocksize    : 16 min keysize  : 16 max keysize  : 32 ivsize       : 16 geniv        : eseqiv All three builds have “caam” and “enable_wait_mode=off” in the kernel command line in u-boot. Only option #3 contains both device file in “/dev/crypto” and an entry in “/proc/crypto” root@freescale ~$ cd / root@freescale /$ ls /proc/cr* /proc/crypto root@freescale /$ ls /dev/cr* /dev/crypto root@freescale /$ Test #1—Kernel build 4.1.0 openssl speed test without caam engine root@freescale ~$ openssl speed -evp aes-128-cbc                    Doing aes-128-cbc for 3s on 16 size blocks: 3471184 aes-128-cbc's in 2.94s Doing aes-128-cbc for 3s on 64 size blocks: 986286 aes-128-cbc's in 3.00s Doing aes-128-cbc for 3s on 256 size blocks: 249743 aes-128-cbc's in 2.93s Doing aes-128-cbc for 3s on 1024 size blocks: 64343 aes-128-cbc's in 3.00s Doing aes-128-cbc for 3s on 8192 size blocks: 7954 aes-128-cbc's in 2.96s OpenSSL 1.0.1c 10 May 2012 built on: Sat Sep 7 18:47:34 PDT 2013 options:bn(64,32) rc4(ptr,char) des(idx,cisc,16,long) aes(partial) idea(int) blowfish(ptr) compiler: gcc -fPIC -DOPENSSL_PIC -DOPENSSL_THREADS -D_REENTRANT -DDSO_DLFCN -DHAVE_DLFCN_H -DL_ENDIAN -DTERMIO -O3 -fomit-frame-pointer -Wall The 'numbers' are in 1000s of bytes per second processed. type 16 bytes     64 bytes    256 bytes 1024 bytes   8192 bytes aes-128-cbc 18890.80k    21040.77k    21820.55k 21962.41k    22013.23k root@freescale ~$ Test #2—Timesys kernel build of openssd without /dev/crypto # openssl speed -evp aes-128-cbc Doing aes-128-cbc for 3s on 16 size blocks: 3361305 aes-128-cbc's in 3.00s Doing aes-128-cbc for 3s on 64 size blocks: 924423 aes-128-cbc's in 3.00s Doing aes-128-cbc for 3s on 256 size blocks: 236623 aes-128-cbc's in 3.00s Doing aes-128-cbc for 3s on 1024 size blocks: 59967 aes-128-cbc's in 3.00s Doing aes-128-cbc for 3s on 8192 size blocks: 7514 aes-128-cbc's in 3.00s OpenSSL 1.0.1e 11 Feb 2013 built on: Thu Sep 5 21:54:37 EDT 2013 options:bn(64,32) rc4(ptr,char) des(idx,cisc,16,long) aes(partial) blowfish(ptr) compiler: armv7l-timesys-linux-gnueabi-gcc -fPIC -DOPENSSL_PIC -DOPENSSL_THREADS -D_REENTRANT -DDSO_DLFCN -DHAVE_DLFCN_H -I/here/workdir/factory/build_armv7l-times ys-linux-gnueabi/toolchain/usr/include -DL_ENDIAN -DTERMIO -DOPENSSL_NO_KRB5 -DOPENSSL_NO_IDEA -DOPENSSL_NO_MDC2 -DOPENSSL_NO_RC5 -Os -pipe -Wa,--noexecstack -Wall The 'numbers' are in 1000s of bytes per second processed. type 16 bytes     64 bytes    256 bytes 1024 bytes   8192 bytes aes-128-cbc 17926.96k    19721.02k    20191.83k 20468.74k    20518.23k #  Test #3—Ubuntu rootfs and kernel image root@linaro-ubuntu-desktop:/# openssl speed -evp aes-128-cbc Doing aes-128-cbc for 3s on 16 size blocks: 3030128 aes-128-cbc's in 2.98s Doing aes-128-cbc for 3s on 64 size blocks: 852897 aes-128-cbc's in 3.00s Doing aes-128-cbc for 3s on 256 size blocks: 220572 aes-128-cbc's in 3.00s Doing aes-128-cbc for 3s on 1024 size blocks: 55534 aes-128-cbc's in 3.00s Doing aes-128-cbc for 3s on 8192 size blocks: 6846 aes-128-cbc's in 2.95s OpenSSL 1.0.0e 6 Sep 2011 built on: Wed Oct 5 01:45:02 UTC 2011 options:bn(64,32) rc4(ptr,char) des(idx,cisc,16,long) aes(partial) blowfish(ptr) compiler: cc -fPIC -DOPENSSL_PIC -DZLIB -DOPENSSL_THREADS -D_REENTRANT -DDSO_DLFCN -DHAVE_DLFCN_H -DL_ENDIAN -DTERMIO -O2 -Wa,--noexecstack -g -Wall The 'numbers' are in 1000s of bytes per second processed. type             16 bytes     64 bytes 256 bytes   1024 bytes   8192 bytes aes-128-cbc 16269.14k    18195.14k    18822.14k 18955.61k    19010.99k root@linaro-ubuntu-desktop:/# Test #4—SDK 4.1.0 openssl speed test with “/dev/crypto” .  Note errors. root@freescale ~$ openssl speed -evp aes-128-cbc -engine cryptodev  invalid engine "cryptodev" 716715216:error:25066067:DSO support routines:DLFCN_LOAD:could not load the shared library:dso_dlfcn.c:187:filename(/usr/lib/engines/libcryptodev.so): /usr/lib/eng ines/libcryptodev.so: cannot open shared object file: No such file or directory 716715216:error:25070067:DSO support routines:DSO_load:could not load the shared library:dso_lib.c:244: 716715216:error:260B6084:engine routines:DYNAMIC_LOAD:dso not found:eng_dyn.c:450: 716715216:error:2606A074:engine routines:ENGINE_by_id:no such engine:eng_list.c:417:id=cryptodev 716715216:error:25066067:DSO support routines:DLFCN_LOAD:could not load the shared library:dso_dlfcn.c:187:filename(libcryptodev.so): libcryptodev.so: cannot open shared object file: No such file or directory 716715216:error:25070067:DSO support routines:DSO_load:could not load the shared library:dso_lib.c:244: 716715216:error:260B6084:engine routines:DYNAMIC_LOAD:dso not found:eng_dyn.c:450: Doing aes-128-cbc for 3s on 16 size blocks: 3572980 aes-128-cbc's in 3.00s Doing aes-128-cbc for 3s on 64 size blocks: 966002 aes-128-cbc's in 2.94s Doing aes-128-cbc for 3s on 256 size blocks: 255307 aes-128-cbc's in 3.00s Doing aes-128-cbc for 3s on 1024 size blocks: 62967 aes-128-cbc's in 2.93s Doing aes-128-cbc for 3s on 8192 size blocks: 7890 aes-128-cbc's in 2.94s OpenSSL 1.0.1c 10 May 2012 built on: Sat Sep 7 18:47:34 PDT 2013 options:bn(64,32) rc4(ptr,char) des(idx,cisc,16,long) aes(partial) idea(int) blowfish(ptr) compiler: gcc -fPIC -DOPENSSL_PIC -DOPENSSL_THREADS -D_REENTRANT -DDSO_DLFCN -DHAVE_DLFCN_H -DL_ENDIAN -DTERMIO -O3 -fomit-frame-pointer -Wall The 'numbers' are in 1000s of bytes per second processed. type 16 bytes     64 bytes    256 bytes 1024 bytes   8192 bytes aes-128-cbc 19055.89k    21028.61k    21786.20k 22006.21k    21984.65k root@freescale ~$ Answer: I do not know what is recent state of official Freescale BSP regarding CAAM, but to get OpenSSL working under CAAM support with reasonable acceleration  : https://community.freescale.com/message/318188#318188 The patches was used below : http://git.freescale.com/git/cgit.cgi/imx/linux-2.6-imx.git/log/?h=imx_3.0.35_4.0.0 Direct link to the patches: http://git.freescale.com/git/cgit.cgi/imx/linux-2.6-imx.git/commit/?h=imx_3.0.35_4.0.0&id=6068d7a77b2101c172fc2f003f90b1febbf99505 http://git.freescale.com/git/cgit.cgi/imx/linux-2.6-imx.git/commit/?h=imx_3.0.35_4.0.0&id=b30237c79003223c6e8035d5be183cd4f0b469f9

Question: What’s the best way to rotate a MX6 image 90 degrees, thought the IPU correct? IPU is limited to 1024x1024. Apparently we don’t support frame buffer rotation in the IPU, so we have to use some middleware. I know that Android’s surface flinger uses the GPU but do you know what we can use in Linux that uses H/W acceleration also? It looks look like X-server can rotate only when the Vivante driver is not  loaded, which means the hardware is not implementing rotations. Answer: it should be possible to split the picture into two halves and rotate them separately. Well, two halves if you can reduce the line count to 1024 … otherwise it would be 4 rotates. X11 Xrandr will be implemented on GPU sometime this year. It's in the R&D queue but as low priority. They could use GC320 low level API to rotate (if they use linux frame buffer). It implies a blit but it would be done by GC320 they will probably need to use virtualFB too. The API documentation is the BSP documentation (iMX6.2D.API.pdf) Attached a simple source using the 2D low level API. VirtualFB: https://community.freescale.com/message/289198

Hi All, The new i.MX 6 Q/D/DL/S/SL L3.0.35_4.1.0 GA release is now available on the http://www.freescale.com/site. ·         Files available                                   # Name Description 1 L3.0.35_4.1.0_LINUX_DOCS i.MX   6Quad, i.MX 6Dual, i.MX 6DualLite, i.MX 6Solo and i.MX 6Sololite Linux BSP   Documentation. Includes Release Notes, Reference Manual, User guide. API   Documentation 2 L3.0.35_4.1.0_LINUX_MMDOCS i.MX   6Quad, i.MX 6Dual, i.MX 6DualLite, i.MX 6Solo and i.MX 6Sololite  Linux Multimedia Codecs Documentation.   Includes CODECs Release Notes and User's Guide 3 L3.0.35_4.1.0_SOURCE_BSP i.MX   6Quad, i.MX 6Dual, i.MX 6DualLite, i.MX 6Solo and i.MX 6Sololite Linux BSP   Source Code Files 4 L3.0.35_4.1.0_MM_CODECS i.MX   6Quad, i.MX 6Dual, i.MX 6DualLite, i.MX 6Solo and i.MX 6Sololite  Linux Multimedia Codecs Sources 5 L3.0.35_4.1.0_AACP_CODECS i.MX   6Quad, i.MX 6Dual, i.MX 6DualLite, i.MX 6Solo and i.MX 6Sololite  Linux AAC Plus Codec 6 L3.0.35_4.1.0_DEMO_IMAGE_BSP i.MX   6Quad, i.MX 6Dual, i.MX 6DualLite, i.MX 6Solo and i.MX 6Sololite  Linux Binary Demo Files 7 L3.0.35_4.1.0_UBUNTU_RFS_BSP i.MX   6Quad, i.MX 6Dual, i.MX 6DualLite, i.MX 6Solo and i.MX 6Sololite  Linux File System for the Ubuntu Images 8 i.MX_6D/Q_Vivante_VDK_146_Tools Set   of applications for the Linux L3.0.35_4.1.0 BSP, designed to be used by   graphics application developers to rapidly develop and port graphics   applications. Includes applications, GPU Driver with vprofiler enabled and   documentation. 9 IMX_6DL_6S_MFG_TOOL Tool   and documentation for downloading OS images to the i.MX 6DualLite and i.MX   6Solo. 10 IMX_6DQ_MFG_TOOL Tool   and documentation for downloading OS images to the i.MX 6Quad and i.MX 6Dual. 11 IMX_6SL_MFG_TOOL Tool   and documentation for downloading OS images to the i.MX 6Sololite. ·         Target HW boards o   i.MX 6Quad SABRE-SDP o   i.MX 6Quad SABRE-SDB o   i.MX 6Quad SABRE-AI o   i.MX 6DualLite SABRE-SDP o   i.MX 6DualLite SABRE-AI o   i.MX 6SL EVK ·         New features o   BSP New Features on i.MX 6D/Q, i.MX 6DL/S and MX 6SL: §  HDCP §  CEC §  GPU4.6.9p12 §  Audio playback IRAM/SDMA §  V4L capture resize on MX6SL §  MX6DQ disable the double line fill feature of PL310 ·         Known issues o   For known issues and limitations please consult the release notes.

This document provide an overall guide how to get started with i.MX6 development. There are several chapters: 1. how to get necessary docs from freescale website; 2. how to setup environment and build your own images;3. Hardware design consideration;4. How to get help. I hope the doc will bring you in i.MX world more easily, and hope you all have a fun in it.

Q: What is latency figures for the VPU to decode Device: i.MX6Q OS: Linux Resolution:                         1920x1080(HD) Frame rate:                        30 FPS Function:                             Overlay messages. Input/Output:                   8 bit / YUV 4:2:0 / NAL stream Profile level:                      4.1. Constrained Baseline. I and P frames support. A: It depend on the syntax in H.264, includes num_reorder_frames,max_dec_frame_buffering,num_ref_frames,MaxDpbSize,etc. for start latency: it cover vpu driver loading, allocate buffers, init, decoding the first frame less than 100ms on iMX6/Linux. This document was generated from the following discussion: VPU Latency i.MX6

Q: When trying to mount his SDIO WiFi module (an Azurewave module containing a Marvel 88W8790)  in 4 bit mode, and got error wifi module on iMX6 Smart SD dev board using the alpha kernel (3.5.7+3285970). mwifiex_sdio mmc0:0001:1: WLAN FW is active mwifiex_sdio mmc0:0001:1: mwifiex_cmd_timeout_func: Timeout cmd id (2004.239824) = 0xa9, act = 0x0 mwifiex_sdio mmc0:0001:1: num_data_h2c_failure = 0 mwifiex_sdio mmc0:0001:1: num_cmd_h2c_failure = 0 mwifiex_sdio mmc0:0001:1: num_cmd_timeout = 1 mwifiex_sdio mmc0:0001:1: num_tx_timeout = 0 mwifiex_sdio mmc0:0001:1: last_cmd_index = 1 mwifiex_sdio mmc0:0001:1: last_cmd_resp_index = 0 mwifiex_sdio mmc0:0001:1: last_event_index = 0 mwifiex_sdio mmc0:0001:1: data_sent=1 cmd_sent=1 mwifiex_sdio mmc0:0001:1: ps_mode=1 ps_state=0 +++++++++++++++++++++++ There are only two known issues with the SDHC driver as noted in the release notes and these don't seem to match.  The Linux Reference Manual states the operation of the SDIO was veriifed usign the AR6003.  My assumption is that this is the Silex module we have standardized on.  Was 4 bit mode verified using this module? Does anyone have any idea what could be happening here? A: L3.5.7 is an alpha release, and WiFi function is not stable. The test report shows that open WiFi issue exists. Wifi module sometime can't work on mx6q_smd. After insmod ath, ath6kl_core and ath6kl_sdio. Insert the wifi card, sometime(50%) will display -------------------------- ath6kl: unable to read RX_LOOKAHEAD_VALID                                      ath6kl: Unable to recv target info: -84                                        ath6kl: Failed to init ath6kl core                                             ath6kl_sdio: probe of mmc0:0001:1 failed with error -84    -------------------------- reinsert the wifi card may solve the problem. when execute "udhcpc -i wlan0", sometime (30%) will cause program exception. Sometime can't get the wlan0 ip(program hang). Environment(OS,Platform,Driver, etc): HW: MX6Q_SMD Num014 and Num017 MX6Q_ARD don't have this problem SW: Kernel 3.5.7-1.0.0 GNU/Linux Case ID: TGE-LV-WIFI-0043 Reproduce Steps: #modprobe ath #modprobe ath6kl_core #modprobe ath6kl_sdio insert the wifi card # iwconfig wlan0 mode managed # iwlist wlan0 scanning | grep MAD-wifi #iwconfig wlan0 key 00112233445566778899123456 #iwconfig wlan0 essid MAD-wifi #udhcpc -i wlan0 Attached MX6 ARD WiFi issue also. [Kernel3.5.7_MX6QARD]Wifi:wifi card can't work during suspend and resume. 100% -- Bug detailed description: With wifi card inserted in the board. doing system suspend and resume test. wifi card can't work well after the system suspend and resume. Always report : ath6kl: Unable to decrement the command credit count register: -84             ath6kl: Unable to write to the device: -84                                     ath6kl: bmi_write_memory for uart debug failed                                 ath6kl: Failed to boot hw in resume: -5          Environment(OS,Platform,Driver, etc): HW: MX6QARD -023 Only tried on this platform SW: root@imx6qsabreauto:~# uname -a                                                Linux imx6qsabreauto 3.5.7-1.0.0+3285970 #1 SMP PREEMPT Sat Jun 29 10:20:45 CDT 2013 armv7l GNU/Linux Case ID:  TGE-LV-WIFI-1060 and TGE-LV-WIFI-1062 Reproduce Steps: 1. boot the kernel with wif card inserted 2. doing wifi stress test 3. doing suspend and resume

The Yocto Project is open-source, so anyone can contribute. No matter what your contribution is (bug fixing or new metadata), contributions are sent through patches to a community list. Many eyes will look into your patch and at some point it is either rejected or accepted. Follow these steps to contribute: Make sure you have previously configured your personal info $ git config --global user.name "Your Name Here" $ git config --global user.email "your_email@example.com" Subscribed to the Freescale Yocto Project Mailing List Download `master` branches fsl-community-bsp $ repo init \   -u https://github.com/Freescale/fsl-community-bsp-platform \   -b master Update fsl-community-bsp $ repo sync Create local branches so your work is *not* done on master fsl-community-bsp $ repo start <branch name> --all Where `<branch name>` is any name you want to give to your local branch (e.g. `fix_uboot_recipe`, `new_gstreamer_recipe`, etc.) Make your changes in any Freescale related folder (e.g. sources/meta-fsl-arm). In case you modified a recipe (.bb) or include (.inc) file, do not forget to *bump* (increase the value by one) either the `PR` or `INC_PR` value Commit your changes using `git`. In this example we assume your change is on `meta-fsl-arm` folder sources/meta-fsl-arm $ git add <file 1> <file 2> sources/meta-fsl-arm $ git commit On the commit's log, the title must start with the filename change or introduced, then a brief description of the patch's goal, following with a long description. Make sure you follow the standards (type ` git log --pretty=oneline` to see previous commits) Create a patch sources/meta-fsl-arm $ git format-patch -s  --subject-prefix='<meta-fsl-arm][PATCH' -1 Where the last parameter (`-1`) indicate to patch last commit. In case you want to create patches for older commits, just indicate the correct index. If your patch is done in other folder, just make sure you change the `--subject-prefix` value. Send your patch or patches with git send-email --to meta-freescale@yoctoproject.org <patch> where `<patch>` is the file created by `git format-patch`. Keep track of patch's responses on the mailing list. In case you need to rework your patch, repeat the steps but this time the patch's subject changes to `--subject-prefix='<meta-fsl-*][PATCH v2'` Once your patch has been approved, you can delete your working branches fsl-community-bsp $ repo abandon <branch name>

1. To setup the Yocto environment, from the BASE folder run fsl-community-bsp $ . setup-environment build 2. Build the toolchain build $ bitbake meta-toolchain # Other toolchains: # Qt Embedded toolchain build: bitbake meta-toolchain-qte # Qt X11 toolchain build: bitbake meta-toolchain-qt 3. Install it on your PC build $ sudo sh \   tmp/deploy/sdk/poky-eglibc-x86_64-arm-toolchain-<version>.sh 4. Setup the toolchain environment build $ source \   /opt/poky/<version>/environment-setup-armv7a-vfp-neon-poky-linux-gnueabi 5. Get the Linux Kernel's source code. $ git clone git://git.freescale.com/imx/linux-2.6-imx.git linux-imx $ cd linux-imx 6. Create a local branch linux-imx $ BRANCH=imx_3.0.35_4.0.0 # Change to any branch you want,   # Use 'git branch -a' to list all linux-imx $ git checkout -b ${BRANCH} origin/${BRANCH} 7. Export ARCH and CROSS_COMPILE linux-imx $ export ARCH=arm  linux-imx $ export CROSS_COMPILE=arm-poky-linux-gnueabi- linux-imx $ unset LDFLAGS 8. Choose configuration and compile linux-imx $ make imx6_defconfig  linux-imx $ make uImage  9. To Test your changes, copy the `uImage` into your SD Card linux-imx $ sudo cp arch/arm/boot/uImage /media/boot 10. If case you want your changes to be reflected on your Yocto Framework, create the patches following the document i.MX Yocto Project: How can I patch the kernel?

The Linux Kernel is just another recipe for Yocto, so learning to patch it you learn to patch any other package. In the other hand, Yocto **should not** be used for package development, but in those rare cases follow the below steps. It is assumed that you have already build the package you want to patch. 1. Create the patch or patches. In this example we are patching the Linux kernel for [wandboard-dual](http://www.wandboard.org/) machine; in other words, the value of MACHINE on the `build/conf/local.conf` is `MACHINE ??= 'wandboard-dual'`. In case you already have the patches, make sure these can be nicely applied with the commands `git apply --check <PATCH_NAME>`, and jump this step build $ cd tmp/work/wandboard_dual-poky-linux-gnueabi/linux-wandboard/3.0.35-r0/git build $ # Edit any files you want to change build $ git add <modified file 1> <modified file 2> .. # Select the files you want to commit build $ git commit -s -m '<your commit's title>' # Create the commit build $ git format-patch -1 # Create the patch 2. Create a new layer (see document i.MX Yocto Proyect: How can I create a new Layer?) 3. On the new layer (e.g `meta-fsl-custom`) , create the corresponding subfolders and the `.bbfile` sources $ mkdir -p meta-fsl-custom/recipes-kernel/linux/linux-wandboard-3.0.35/ sources $ cat > meta-fsl-custom/recipes-kernel/linux/linux-wandboard_3.0.35.bbappend FILESEXTRAPATHS_prepend := "${THISDIR}/${PN}-${PV}:" SRC_URI += "file://0001-calibrate-Add-printk-example.patch" PRINC := "${@int(PRINC) + 1}" # SEE NOTE BELLOW ^d (The PRINC variable is not needed starting at Yocto 1.6 ([RFC] base.bbclass: Deprecate the PRINC logic - Patchwork)) 4. Move the patch to the new layer sources $ cp \ ../build/tmp/work/wandboard_dual-poky-linux-gnueabi/linux-wandboard/3.0.35-r0/git/0001-calibrate-Add-printk-example.patch \ meta-fsl-custom/recipes-kernel/linux/linux-wandboard-3.0.35 5. Setup the enviroment and clean previous package's build data (sstate) fsl-community-bsp $ . setup-environment build build $ bitbake -c cleansstate linux-wandboard 6. Compile and Deploy build $ bitbake -f -c compile linux-wandboard build $ bitbake -c deploy linux-wandboard 7. Insert the SD into your Host and copy the `uImage` into the first partition. Do not forget to unmount the partition before removing the card! build $ sudo cp tmp/deploy/images/uImage /media/Boot\ wandbo/ 8. Insert the SD into your board and test your change.

We use PCIe to connect Intersil TW6865 chip for the surround view solution. This is the connection of PCIe to iMX6Q SabreSD board.   This is the block diagram of the connection: This is the 4 camera surround view:   Code base is L3.0.35_12.10.02 release. You can merge the patch file to the latest Freescale release. Please check the attach file for the patch code.   Note:  It is only a test version. The last code for L3.0.35 BSP: L3.0.35_GA4.1.0 Patches.7z The last code for L3.10.53 BSP: L3.10.53_TW686x_patch.7z Patch for L4.1.15 1.1.0 GA BSP: TW6865 driver for Linux L4.1.15_1.1.0-ga.7z

The solution works when I use mx53_loco bsp. Modify u-boot and kernel, keep the same. Then you may find you can't login into the system regardless of whatever you input after freescale login: It confused me for a long time. If you  also met this problem,try to check the iomux-mx53.h(linux/arch/arm/plat-mxc/include/mach/). #define _MX53_PAD_PATA_DIOW__UART1_TXD_MUX   IOMUX_PAD(the fourth argument 0x878 should be changed to 0x0) I think this is a small bug in header files. Haifeng

In this post we see how to setup a Debian server, to allow booting the i.MX6 sabre sd platform (mostly) from the network. Booting from the network instead of e.g. the SD card is very handy for day to day development and testing, as it eliminates almost all physical interactions with the board and saves much time. Also, fortunately for us, both u-boot and Linux for i.MX6 support network booting out of the box. Boot sequence principles Before we setup the server, here are some more details on the boot sequence we will obtain in the end: i.MX6 boots, loads u-boot from SD card. u-boot starts, loads its environment (boot commands) from SD card. u-boot obtains its network address by DHCP, loads a Linux kernel uImage and a dtb by TFTP. Linux boots; obtains its network address by DHCP (again), mounts its root filesystem on NFS. Setting up DHCP and TFTP One can easily setup a Debian server to act as DHCP and TFTP server with Dnsmasq; just install the dnsmasq package. The default configuration is mostly empty; so we need to enhance it a bit. For the following we will assume that your Debian server has IP address 192.168.111.1 on the network where it sees the i.MX6 sabre sd platform. You can add some options to a dnsmasq config file such as e.g. /etc/dnsmasq.d/my-custom-config-file:   dhcp-range=192.168.111.50,192.168.111.150,12h   enable-tftp   tftp-root=/var/ftpd This informs dnsmasq to act as a DHCP server for addresses range 192.168.111.50-150 and act as TFTP server, which serves files under /var/ftpd. That means you will need to copy a Linux uImage and an imx6q-sabresd.dtb under /var/ftpd/. See this post for more details about compiling Linux to obtain those two files. Setting up NFS If we want the root filesystem to be mounted on the network we will need to export some folders with NFS from the Debian server. We need to install the nfs-kernel-server package and setup /etc/exports with a line such as:   /tftpboot       192.168.111.*(rw,no_root_squash,subtree_check) This allows clients on the 192.168.111.0 network to access filesystems under the /tftpboot folder. So you will need to create a /tftpboot folder on the server, and install some "filesystem" under there. For this example we assume you will have a busybox installed under a /tftpboot/busybox/ folder. That means we want to have under there all folders such as bin, dev, etc... See this post for details on how to compile busybox to populate this folder. Do not forget to restart the NFS server after configuration, with:   # /etc/init.d/nfs-kernel-server restart We are now setup on the server side. Setting up u-boot At the time of this writing we need to help u-boot a bit when booting the i.MX6 sabre sd platform from the network. Stop at u-boot prompt and configure a few things:   env default -a   setenv netargs $netargs rw   setenv serverip 192.168.111.1   setenv nfsroot /tftpboot/busybox   setenv bootcmd run netboot   saveenv Reset your board; it should now boot from the network:   U-Boot 2013.07-rc1-00210-gc623eb0 (Jun 27 2013 - 21:10:47)   (..)   Hit any key to stop autoboot:  0   Booting from net ...   BOOTP broadcast 1   DHCP client bound to address 192.168.111.121   Using FEC device   TFTP from server 192.168.111.1; our IP address is 192.168.111.121   Filename 'uImage'.   Load address: 0x12000000   Loading: #################################################################            #################################################################            #################################################################            #################################################################            ##########################            4 MiB/s   done   Bytes transferred = 4185600 (3fde00 hex)   BOOTP broadcast 1   DHCP client bound to address 192.168.111.121   Using FEC device   TFTP from server 192.168.111.1; our IP address is 192.168.111.121   Filename 'imx6q-sabresd.dtb'.   Load address: 0x11000000   Loading: ##            2.7 MiB/s   done   Bytes transferred = 22818 (5922 hex)   ## Booting kernel from Legacy Image at 12000000 ...      Image Name:   Linux-3.10.0-rc7   (..)   Starting kernel ...   Booting Linux on physical CPU 0x0   Linux version 3.10.0-rc7 (jenkins@debian) (gcc version 4.7.2 (Debian 4.7.2-5) ) #1 SMP Tue Jun 25 08:28:31 CEST 2013   (..)   Kernel command line: console=ttymxc0,115200 root=/dev/nfs ip=dhcp nfsroot=192.168.111.1:/tftpboot/busybox,v3,tcp rw   (..)   fec 2188000.ethernet eth0: Freescale FEC PHY driver [Generic PHY] (mii_bus:phy_addr=2188000.ethernet:01, irq=-1)   IPv6: ADDRCONF(NETDEV_UP): eth0: link is not ready   libphy: 2188000.ethernet:01 - Link is Up - 1000/Full   IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link becomes ready   Sending DHCP requests ., OK   IP-Config: Got DHCP answer from 192.168.111.1, my address is 192.168.111.121   IP-Config: Complete:        device=eth0, hwaddr=00:04:9f:02:b7:fd, ipaddr=192.168.111.121, mask=255.255.255.0, gw=192.168.111.1        host=192.168.111.121, domain=, nis-domain=(none)        bootserver=192.168.111.1, rootserver=192.168.111.1, rootpath=        nameserver0=192.168.111.1   ALSA device list:     No soundcards found.   VFS: Mounted root (nfs filesystem) on device 0:11.   devtmpfs: mounted   Freeing unused kernel memory: 292K (806d5000 - 8071e000)   Please press Enter to activate this console. Enjoy! Bonus: updating u-boot by the network One last piece remains on the SD card: u-boot. If you do not want to move your SD card out of its slot any more, here is a method for you to update even u-boot from the network. You will need to copy u-boot.imx under /var/ftpd. See this post for details on how to compile u-boot and obtain u-boot.imx. Then, at u-boot prompt, do:   dhcp $loadaddr u-boot.imx   mmc dev 1   mmc write $loadaddr 2 600 This will download a new u-boot.imx from the network and flash it to your SD card; reboot your board and you are done. Note that we give 600 as the number of SD card blocks to write; this is a rough estimate of ~300KB, which should work in most of the cases as writing a bit "too much" blocks does not harm. If you are very picky, you can compute the exact number of blocks by dividing your u-boot.imx size by 512 and rounding it up. See also... Did you know that dnsmasq primary role is to be used to "relay" the DNS queries? A feature that come very handy when you want to let your i.MX6 platform "see" the internet.

Here we show how to generate a minimal root filesystem fairly quickly with BusyBox, for the i.MX6 sabre sd platform. This document assumes you are able to boot a Linux kernel on your platform already. See this post for details on how to do it. This implies you already have a "working" Linux development environment with some ARM cross-compilers at hand (e.g. Debian + Emdebian). busybox is so small that we will go for a ramdisk as our main root filesystem. Get busybox sources We will use git to fetch busybox sources:   $ git clone git://git.busybox.net/busybox This should create a busybox directory with all the latest sources. Note that for more stability you might want to checkout a release instead of the latest version; to do so, list the available release tags with e.g. git tag -l, and git checkout <the-desired-tag>. Compile Assuming your cross compiler is called e.g. arm-linux-gnueabihf-gcc, you can compile by doing:   $ cd busybox   $ export ARCH=arm   $ export CROSS_COMPILE=arm-linux-gnueabihf-   $ make defconfig   $ sed -i.orig 's/^#.*CONFIG_STATIC.*/CONFIG_STATIC=y/' .config   $ make   $ make install This should create an _install folder hierarchy containing binaries and links. Note that we force the build of a static binary with the sed command. Configure the root filesystem We need to add some more configuration into the _install folder before we can call it a minimal filesystem. Create some folders We need to create some mountpoints and folders:   $ mkdir _install/dev   $ mkdir _install/proc   $ mkdir _install/sys   $ mkdir -p _install/etc/init.d Add some configuration files and scripts We need to prepare the main init configuration file, _install/etc/inittab, with this contents:   ::sysinit:/etc/init.d/rcS   ::askfirst:/bin/sh   ::ctrlaltdel:/sbin/reboot   ::shutdown:/sbin/swapoff -a   ::shutdown:/bin/umount -a -r   ::restart:/sbin/init This is very close to the default behavior busybox init has with no inittab file. It just suppresses some warnings about missing tty. We need to add some more configuration to mount a few filesystems at boot for convenience. This is done with an _install/etc/fstab file containing:   proc     /proc proc     defaults 0 0   sysfs    /sys  sysfs    defaults 0 0   devtmpfs /dev  devtmpfs defaults 0 0 We also need to actually trigger the mount in the _install/etc/init.d/rcS script, which is called from the inittab. It should contain:   #!/bin/sh   mount -a And we need to make it executable:   $ chmod +x _install/etc/init.d/rcS Generate the ramdisk contents Now that we have adapted the root filesystem contents, we can generate a busybox ramdisk image for u-boot with the following commands:   $ (cd _install ; find |cpio -o -H newc |gzip -c > ../initramfs.cpio.gz)   $ mkimage -A arm -T ramdisk -d initramfs.cpio.gz uInitrd This results in a uInitrd file, suitable for u-boot. Prepare a boot script The default u-boot commands are not sufficient to boot our system, so we need to edit a boot.txt file with the following contents:   run loaduimage   run loadfdt   setenv rdaddr 0x13000000   fatload mmc ${mmcdev}:$mmcpart $rdaddr uInitrd   setenv bootargs console=${console},${baudrate} rdinit=/sbin/init   bootm $loadaddr $rdaddr $fdt_addr Then we generate a boot.scr script, which can be loaded by u-boot with:   $ mkimage -A arm -T script -d boot.txt boot.scr Put on SD card Assuming you have prepared your SD card with u-boot and Linux as explained in this post, you have a single FAT partition on your card with your kernel and dtb. Our boot script and ramdisk image should be copied alongside:   $ mount /dev/<your-sd-card-first-partition> /mnt   $ cp uInitrd boot.scr /mnt/   $ umount /mnt Your SD card first partition is typically something in /dev/sd<X>1 or /dev/mmcblk<X>p1. Note that you need write permissions on the SD card for the command to succeed, so you might need to su - as root, or use sudo, or do achmod a+w as root on the SD card device node to grant permissions to users. Boot! Your SD card is ready for booting. Insert it in the SD card slot of your i.MX6 sabre sd platform, connect to the USB to UART port with a serial terminal set to 115200 baud, no parity, 8bit data and power up the platform. Your busybox system should boot to a prompt:   ...   Freeing unused kernel memory: 292K (806d5000 - 8071e000)   Please press Enter to activate this console. After pressing enter you should have a functional busybox shell on the target. Enjoy! See also... For a more featured root filesystem you might want to try a Debian filesystem in a second SD card partition, as explained in this post, or generate your filesystem with Buildroot. If you plan to compile busybox often, you might want to use a C compiler cache; see this post.

Here we show how to bootstrap the Debian Linux distribution from a PC to the i.MX6 sabre sd platform. While bootstrapping Debian on any architecture "natively" is pretty straightforward, "cross-bootstrapping" requires some techniques that we will explain. This document assumes you are able to boot a Linux kernel on your platform already. See this post for details on how to do it. Also, this document assumes you are using a Debian PC for preparing your SD card. You will require the following packages to be installed: binfmt-support qemu-user-static debootstrap Note: all the commands found in the following steps need to be run as root. Formatting the SD card We need to format the SD card with two partitions; one small FAT partition to contain the Linux kernel and its dtb, and one large ext4 partition, which will contain the root filesystem with the Debian userspace. Also, we need to make sure we leave some space for u-boot starting from offset 1024B. Here is an example SD card layout:   +-----+------+--------+-----+---------------+-----------------   | MBR |  ... | u-boot | ... | FAT partition | Linux partition ...   +-----+------+--------+-----+---------------+-----------------   0     512    1024           1M              ~257M (offsets in bytes) Here is an example SD card layout, as displayed by fdisk:   Device    Boot      Start         End      Blocks   Id  System   /dev/sdc1            2048      526335      262144    c  W95 FAT32 (LBA)   /dev/sdc2          526336     8054783     3764224   83  Linux (units: 512B sectors) You can format and mount the Linux partition with:   # mkfs.ext4 /dev/<your-sd-card-second-partition>   # mount /dev/<your-sd-card-second-partition> /mnt Your SD card second partition is typically something in /dev/sd<X>2 or /dev/mmcblk<X>p2. Do not forget to install u-boot and a Linux kernel as explained in those posts. Bootstrapping Debian First stage The first stage of Debian bootstrapping is done with:   # debootstrap --foreign --arch=armhf testing /mnt This will retrieve the base Debian packages from the internet, and perform a first stage of installation:   I: Retrieving Release   I: Retrieving Release.gpg   I: Checking Release signature   I: Valid Release signature (key id A1BD8E9D78F7FE5C3E65D8AF8B48AD6246925553)   I: Validating Packages   I: Resolving dependencies of required packages...   I: Resolving dependencies of base packages...   I: Found additional required dependencies: insserv libbz2-1.0 libcap2 libdb5.1 libsemanage-common libsemanage1 libslang2 libustr-1.0-1   I: Found additional base dependencies: libee0 libept1.4.12 libestr0 libgcrypt11 libgnutls-openssl27 libgnutls26 libgpg-error0 libidn11 libjson-c2 liblognorm0 libmnl0 libnetfilter-acct1 libnfnetlink0 libp11-kit0 libsqlite3-0 libtasn1-3 libxapian22   I: Checking component main on http://ftp.us.debian.org/debian...   (...)   I: Extracting util-linux...   I: Extracting liblzma5...   I: Extracting zlib1g... At this point, the necessary tools for second stage of installation are under /mnt/debootstrap/. Second stage The second stage needs to run natively; on an arm platform, that is. But we can use the combination of two techniques to perform this stage on the PC anyway:   # cp /usr/bin/qemu-arm-static /mnt/usr/bin/   # chroot /mnt /debootstrap/debootstrap --second-stage Those commands copy an arm emulator on the target filesystem, and use the chroot command to execute the second stage of the installation into the SD card, on the PC, with transparent emulation:   I: Installing core packages...   I: Unpacking required packages...   I: Unpacking libacl1:armhf...   I: Unpacking libattr1:armhf...   I: Unpacking base-files...   (...)   I: Configuring tasksel...   I: Configuring tasksel-data...   I: Configuring libc-bin...   I: Base system installed successfully. You can now remove /mnt/usr/bin/qemu-arm-static, or keep it for later, subsequent chroot under emulation. Finetuning the root filesystem For development it is handy to remove the root password on the target by removing the '*' from /mnt/etc/shadow on the SD card:   root::15880:0:99999:7::: Also, we can add the following line in /mnt/etc/inittab to obtain a login prompt on the UART:   T0:23:respawn:/sbin/getty -L ttymxc0 115200 vt100 You can now unmount the filesystem with:   # umount /mnt Boot! Your SD card is ready for booting. Insert it in the SD card slot of your i.MX6 sabre sd platform, connect to the USB to UART port with a serial terminal set to 115200 baud, no parity, 8bit data and power up the platform. At the time of writing u-boot tells the kernel to boot from the wrong partition by default, so we need to interrupt by pressing enter at u-boot prompt for the first boot and setup u-boot environment to fix this:   U-Boot > setenv mmcroot /dev/mmcblk0p2 rootwait rw   U-Boot > saveenv   Saving Environment to MMC...   Writing to MMC(1)... done As this is saved in the SD card it need only to be done once at first boot. You can reboot your board or type boot; your Debian system should boot to a prompt:   (...)   [ ok ] Starting periodic command scheduler: cron.   [ ok ] Running local boot scripts (/etc/rc.local).   Debian GNU/Linux jessie/sid debian ttymxc0   debian login: From there you may login as root. It is recommended to setup the network connection and install an ssh server inside the target for further development. Enjoy! See also... With the amounts of memory we have today in the systems, it is even possible to boot Debian in a ramdisk. See this post about busybox for the ramdisk generation. Another way of generating a root filesystem is by building it with buildroot. See and this post for details.

ccache is a C compiler cache. ccache can save a large amount of compilation time on recurring builds and builds restarted from a clean repository after make clean or git clean. It is well suited for e.g. u-boot and Linux compilation. Caching the host compiler Caching "native" builds is easily done by adding in the beginning of your $PATH a special directory, which contains links to ccache to override the usual compiler. On e.g. Debian this directory is readily available as /usr/lib/ccache, So you can do:   $ export PATH="/usr/lib/ccache:$PATH" Typical links found in this folder are:   c++ -> ../../bin/ccache   cc -> ../../bin/ccache   g++ -> ../../bin/ccache   gcc -> ../../bin/ccache etc... Caching the cross compiler Caching cross-compiled builds can be done in the same way as native builds, provided you create links of the form e.g. arm-linux-gnueabihf-gcc pointing to ccache. But there is an even more convenient way for those projects, which rely on a $CROSS_COMPILE environment variable (as is the case for e.g. u-boot and Linux). You can prefix the cross compiler with ccache there in e.g. the following way:   $ export CROSS_COMPILE="ccache arm-linux-gnueabihf-" Monitoring efficiency Now that your builds are cached, you might want to see how much is "spared" with this technique. ccache -s will tell you all sorts of statistics, such as:   cache directory                     /home/vstehle/.ccache   cache hit (direct)                 10852   cache hit (preprocessed)            3225   cache miss                         19000   called for link                    33267   called for preprocessing            9463   compile failed                         3   preprocessor error                     1   couldn't find the compiler           117   unsupported source language          921   unsupported compiler option         2167   no input file                      31681   files in cache                     51694   cache size                           1.3 Gbytes   max cache size                       4.0 Gbytes Here you see a somewhat typical 50%/50% hit/miss ratio. Enjoy! See Also ccache is usually supported natively by build systems, such as Buildroot or Yocto.

Here is a quick summary at booting Linux on the i.MX 6 sabre sd platform. This assumes you already have u-boot working on your platform as described here. This implies you already have a "working" Linux development environment with some ARM cross-compilers at hand (e.g. Debian + Emdebian). Get Linux sources We will use git to fetch Linux sources:   $ git clone git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git This should create a linux directory with all the latest sources (after a while). Note that for more stability you might want to checkout a release instead of the latest version; to do so, list the available release tags with e.g. git tag -l 'v*', and git checkout <the-desired-tag>. Compile Assuming your cross compiler is called e.g. arm-linux-gnueabihf-gcc, you can compile by doing:   $ cd linux   $ export ARCH=arm   $ export CROSS_COMPILE=arm-linux-gnueabihf-   $ make imx_v6_v7_defconfig   $ make You then need to supply a LOADADDR (as joowonkim pointed out); do:   $ make uImage LOADADDR=0x10008000 This should create a number of files, including arch/arm/boot/uImage and arch/arm/boot/dts/imx6q-sabresd.dtb. Put on SD We need a proper FAT partition on the SD card, from which u-boot will be able to load the kernel and dtb. Also, we need to make sure we leave some space for u-boot starting from offset 1024B. Here is an example SD card layout:   +-----+------+--------+-----+----------------   | MBR |  ... | u-boot | ... | FAT partition ...   +-----+------+--------+-----+----------------   0     512    1024           1M (offsets in bytes) Here is an example SD card layout, as displayed by fdisk:   Device    Boot      Start         End      Blocks   Id  System   /dev/sdc1            2048     8054783     4026368    c  W95 FAT32 (LBA) (units: 512B sectors) You can format the FAT partition, mount, copy and unmount with:   $ mkfs.vfat /dev/<your-sd-card-first-partition>   $ mount /dev/<your-sd-card-first-partition> /mnt   $ cp arch/arm/boot/uImage arch/arm/boot/dts/imx6q-sabresd.dtb /mnt/   $ umount /mnt Your SD card first partition is typically something in /dev/sd<X>1 or /dev/mmcblk<X>p1. Note that you need write permissions on the SD card for the command to succeed, so you might need to su - as root, or use sudo, or do a chmod a+w as root on the SD card device node to grant permissions to users. Also, be sure to have u-boot on the SD card as explained in this post. Boot! That's it; u-boot already knows how to deal with your kernel by default so you are good to go. Insert the SD card into the SD card slot of your i.MX6 sabre sd platform, connect to the USB to UART port with a serial terminal set to 115200 baud, no parity, 8bit data and power up the platform. You should see u-boot messages:   U-Boot 2013.07-rc1-00014-g74771f4 (Jun 21 2013 - 16:27:39) u-boot should load the uImage and dtb from SD card and boot the kernel:   (...)   reading uImage   4215344 bytes read in 449 ms (9 MiB/s)   Booting from mmc ...   reading imx6q-sabresd.dtb   22818 bytes read in 22 ms (1012.7 KiB/s)   ## Booting kernel from Legacy Image at 12000000 ...      Image Name:   Linux-3.10.0-rc6      Image Type:   ARM Linux Kernel Image (uncompressed)      Data Size:    4215280 Bytes = 4 MiB      Load Address: 10008000      Entry Point:  10008000      Verifying Checksum ... OK   ## Flattened Device Tree blob at 11000000      Booting using the fdt blob at 0x11000000      Loading Kernel Image ... OK   OK      Using Device Tree in place at 11000000, end 11008921   Starting kernel ... The kernel should boot:   Booting Linux on physical CPU 0x0   Linux version 3.10.0-rc6 (vstehle@debian) (gcc version 4.7.2 (Debian 4.7.2-5) ) #1 SMP Fri Jun 21 18:09:26 CEST 2013 By default, the kernel will try to mount a root filesystem from the SD card second partition, as can be read in the default kernel command line:   (...)   Kernel command line: console=ttymxc0,115200 root=/dev/mmcblk1p2 rootwait rw ...but we did not prepare a root filesystem partition, so after a number of boot messages the kernel will wait indefinitely:   (...)   mmc1: new SDHC card at address b368   (...)    mmcblk0: p1   (...)   Waiting for root device /dev/mmcblk1p2... We will see in another post how to prepare this root filesystem on the second SD card partition. Enjoy! See also... If you plan to compile Linux often, you might want to use a C compiler cache; see this post. Once you have Linux booting on your platform the next step is to give it a root filesystem. See this post for a Debian root filesystem, this post for a minimal busybox filesystem and this post for generating a root filesystem with buildroot.