Hi,
On 10/06/2017 06:09 AM, Yevgeny Lavrov wrote:
Hi
When trying various test scripts, I ran into the following problem while executing "run/init" script on Sabre Lite .... [init -> test-init] step 32 (init_config) [init -> test-init] step 33 (expect_log) [init -> test-init] [init -> log] config 3: became unavailable (ignored) no RM attachment (READ pf_addr=0x5f75c pf_ip=0x5f75c from pager_object: pd='init -> init -> log' thread='ep')
Are there any suggestions on what might be causing it?
well, it is not strange that non-cooperatively killed programs are triggering last page-faults, because during their destruction several resources are hardly withdrawn. The init.run test script repeatedly spawns components and removes them again. Therefore, above log messages are nothing worrying but normal. The question is, whether you get "Run script execution successful." in the end.
And, is it possible to reproduce the USB Armory's "run/tz_vmm" scenario on i.MX6 platform?
This cannot be done out of the box. The most easy part would be to add support to the hw kernel to execute i.MX6 in the secure world, provide the TrustZone specific VM interface to the user-land, and to partition the memory into secure and normal one. The next probably more time-consuming step is the para-virtualization of the i.MX6 specific Linux Kernel that is run as "guest" in the tz_vmm.run scenario.
Regards Stefan
Thanks
On Fri, Sep 29, 2017 at 2:07 AM, Yevgeny Lavrov <ylavrov13@...9... mailto:ylavrov13@...9...> wrote:
Hi Now that I have Genode running on i.MX6 under both kernels, hw and seL4, I want to move to my next objective, which is to bring up Linux on top of Genode/hw on ARM. Is there any already existing run script that I may use (as a reference) to achieve that on i.MX6 platform? Do I need virtualization for this? Cheers On Wed, Sep 13, 2017 at 4:45 PM, Yevgeny Lavrov <ylavrov13@...9... <mailto:ylavrov13@...9...>> wrote: Thank you for quick reply. It was very helpful, and pointed me in the right direction.The answer was hiding within repos/base-sel4/patches/wand_quad.config, I somehow overlooked this directory while looking for answers. It turned out to be a one line fix in <genode_dir>/contrib/sel4-<SHA>/src/kernel/sel4/configs/imx6/sabre_lite/autoconf.h file. Changing#define CONFIG_PLAT_WANDQ 1 to #define CONFIG_PLAT_SABRE 1 did the trick. Now I can observe outputs from elf-loader tool, seL4 and Genode. Cheers On Wed, Sep 13, 2017 at 8:30 AM, Christian Helmuth <christian.helmuth@...1... <mailto:christian.helmuth@...1...>> wrote: On Wed, Sep 13, 2017 at 12:44:11PM +0200, Alexander Boettcher wrote: > On 13.09.2017 08:48, Alexander Boettcher wrote: > > The UART for Sabre Lite and Wand Quad are not the same, as far as I know. > > Sorry, my assumption seems to be wrong, since the messages above are > from the seL4 elfloader already. > > So I presume you have just to adjust the seL4 kernel autoconf.h to match > the Sabre lite board. The kernel differentiates Sabre Lite and Wandboard regarding UART config and available RAM. $ grep -rnA2 'CONFIG_PLAT_SABRE\|CONFIG_PLAT_WANDQ' . ./src/kernel/sel4/configs/imx6/wand_quad/autoconf.h:75:#define CONFIG_PLAT_WANDQ 1 ./src/kernel/sel4/configs/imx6/wand_quad/autoconf.h-76-#define CONFIG_HAVE_LIB_PLATSUPPORT 1 ./src/kernel/sel4/configs/imx6/wand_quad/autoconf.h-77-#define CONFIG_NUM_DOMAINS 16 -- ./src/kernel/sel4/include/plat/imx6/plat/machine/hardware.h:55:#if defined(CONFIG_PLAT_SABRE) ./src/kernel/sel4/include/plat/imx6/plat/machine/hardware.h-56- /* Sabre has 1 GiB */ ./src/kernel/sel4/include/plat/imx6/plat/machine/hardware.h-57- { /* .start = */ 0x10000000, /* .end = */ 0x50000000 } ./src/kernel/sel4/include/plat/imx6/plat/machine/hardware.h:58:#elif defined(CONFIG_PLAT_WANDQ) ./src/kernel/sel4/include/plat/imx6/plat/machine/hardware.h-59- /* Wandboard Quad: 2 GiB */ ./src/kernel/sel4/include/plat/imx6/plat/machine/hardware.h-60- { /* .start = */ 0x10000000, /* .end = */ 0x90000000 } -- ./src/kernel/sel4/include/plat/imx6/plat/machine/devices.h:15:#ifdef CONFIG_PLAT_SABRE ./src/kernel/sel4/include/plat/imx6/plat/machine/devices.h-16-#define UART_PADDR UART2_PADDR ./src/kernel/sel4/include/plat/imx6/plat/machine/devices.h:17:#elif defined(CONFIG_PLAT_WANDQ) ./src/kernel/sel4/include/plat/imx6/plat/machine/devices.h-18-#define UART_PADDR UART1_PADDR ./src/kernel/sel4/include/plat/imx6/plat/machine/devices.h-19-#else Greets -- Christian Helmuth Genode Labs https://www.genode-labs.com/ · https://genode.org/ https://twitter.com/GenodeLabs · /ˈdʒiː.nəʊd/ Genode Labs GmbH · Amtsgericht Dresden · HRB 28424 · Sitz Dresden Geschäftsführer: Dr.-Ing. Norman Feske, Christian Helmuth ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot _______________________________________________ genode-main mailing list genode-main@lists.sourceforge.net <mailto:genode-main@lists.sourceforge.net> https://lists.sourceforge.net/lists/listinfo/genode-main <https://lists.sourceforge.net/lists/listinfo/genode-main>
Check out the vibrant tech community on one of the world's most engaging tech sites, Slashdot.org! http://sdm.link/slashdot
genode-main mailing list genode-main@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/genode-main