April 2017 - Devel - lists.sel4.systems

Questions about Memory Management in seL4
by Daniel Wang 19 May '23

19 May '23

Hi all, Sorry to bother, I have some basic questions about seL4 memory management. Can someone help me? 1. where is the kernel’s address space? I know there are basically three methods: a) separate virtual address space, e.g. through changing page tables on entry into privileged mode to access b) physical space, e.g. through disable automatic hardware translation of virtual addresses on entry into privilege mode c) privilege region in each process’s virtual address space, e.g. through page table to map kernel address into use-mode process. My question is which of this is being used by seL4 microkernel? 2. Objects like CSpace, VSpace, CNode, TCB, Endpoint, etc. are called kernel objects, but it seems all dynamically allocated in thread running/creating time. Are those objects (e.g. Figure 3.3 in seL4 manual) in kernel space or in user space? 3. I read that after the kernel boot up, it passes all resources, untyped memory to the first thread. Are those untyped memory physical memory or virtual memory in a single page? The reason I ask is I’m learning the code of the UNSW AOS project (two of the assignment projects are implementing frame and papers). I’m confused that before the implementation of frames and pages what kind of memory is passed to the the first thread? I saw int AOS project, it uses ut_table_init(), ut_steal_mem(), I wonder what kind of memory it operates on? Thanks -Dan

5 4

seL4 Upgrading Issues
by Chris Guikema 04 May '17

04 May '17

Hello, I have developed an Ethernet application using CAmkES 2.2 and seL4 3.2. Because we are going to open-source the design and other documentation, I thought it would be a good idea to upgrade to the latest version of seL4 and CAmkES. I can pull the latest kernel and merge the compatible libraries with my changes and get everything to compile; however, when I load it on the board it fails... Failed to find any untyped capable of creating an object at address 0x20c4000 Failed to find a way to map address 0x20c4000 Failed to find any untyped capable of creating an object at address 0x20c8000 Failed to find a way to map address 0x20c8000 Failed to find any untyped capable of creating an object at address 0x20e0000 Failed to find a way to map address 0x20e0000 Failed to find any untyped capable of creating an object at address 0x21bc000 Failed to find a way to map address 0x21bc000 Here is my CAmkES configuration which sets up the Ethdrivers cnode and untypes. Keep in mind this works with seL4 3.2 ethdriver.simple = true; ethdriver.cnode_size_bits = 12; ethdriver.simple_untyped20_pool = 4; ethdriver.heap_size = 0; ethdriver.mmios = "0x02188000:0x4000:12, 0x021BC000:0x4000:12, 0x020E0000:0x4000:12, 0x020A4000:0x4000:12, 0x020B0000:0x4000:12, 0x020C4000:0x4000:12, 0x020C8000:0x1000:12"; Can anyone see the problem? Thanks, Chris Guikema

2 2

questions about TK1 VM
by PX 01 May '17

01 May '17

Hi, all I have some questions about TK1 VM. I want to save the execution status of the TK1 VM. First, I stop the VM, then I use the function, seL4_TCB_ReadRegisters() to read the VM's registers into a buffer. Here are my questions: 1. Before I call the function, seL4_TCB_ReadRegisters(), do I need to flush TLB of CPU? If so, where is the library I need to check for this function? 2. The register values in the structure, seL4_UserContext, are VMM's physical addresses or virtual addresses? Best wishes Peng

2 1

sel4test failing TEST_SCHED0004 on iMX6
by Robert Kaiser 20 Apr '17

20 Apr '17

Hi all, I just tried to run the test suite on iMX6. Using the current version from github, I got this result: ---------------------------- ..... Test suite failed. 200/202 tests passed. 200/202 tests passed. *** FAILURES DETECTED *** ---------------------------- Further checking into the log shows this: ---------------------------- ..... Starting test 51: TEST_SCHED0004 TEST_SCHED0004 Running test SCHED0004 (Test threads at all priorities) _allocman_cspace_alloc@allocman.c:252 Regular cspace alloc failed, and failed from watermark vka_alloc_object_at_maybe_dev@object.h:46 Failed to allocate cslot: error 1 new_pages_at_vaddr@vspace.c:331 Failed to allocate page sel4utils_configure_thread_config@thread.c:77 ipc buffer allocation failed Assertion failed: error == 0 ([.....]/seL4test/apps/sel4test-tests/src/helpers.c: create_helper_thread: 346) Failure: result == SUCCESS at line 343 of file [.....]/seL4test/apps/sel4test-driver/src/main.c Failure: Test check failure (see log). at line 0 of file ..... ---------------------------- The same test is run a second time further down below the log, giving the same result. I tested this on two different IMX6 platforms (SabreLite and Wandboard). Both showed identical results. Is this a known problem? Cheers Robert -- Robert Kaiser Technische Informatik Hochschule RheinMain Wiesbaden Rüsselsheim Computer Engineering RheinMain University of Applied Sciences robert.kaiser(a)hs-rm.de http://www.cs.hs-rm.de/~kaiser tel:(+49)611-9495-1292 fax:(+49)611-9495-1210 Postanschrift/Postal Address: Robert Kaiser, Hochschule RheinMain, FB DCSM/Informatik Unter den Eichen 5, 65195 Wiesbaden, Germany

2 2

smaccm/sel4arm-vmm-manifest complie problem
by 高文博 19 Apr '17

19 Apr '17

hi everyone! i am doing research about sel4.i used https://github.com/smaccm/sel4arm-vmm-manifest code. and complie it.than it build out kernel.elf only. but where is linux? expecting your answers. thanks very much! gao. -- 高文博数据通信科学技术研究所

2 1

smaccm/sel4arm-vmm-manifest compile problem
by 高文博 15 Apr '17

15 Apr '17

hi everyone! i am doing research about sel4.i used https://github.com/smaccm/sel4arm-vmm-manifest code. and compile it.than it build out kernel.elf only. but where is linux? expecting your answers. thanks very much! gao.

1 0

how to debug libsel4allocman
by Vasily A. Sartakov 14 Apr '17

14 Apr '17

Greetings I am using a 4.0 compatible seL4 snapshot derived from mainline several months ago as a base for seL4/MIPS port. Currently, I am debugging sel4tests. My goal — make everything working before moving to hardware. By «everything» I mean different configurations of the system (optimization flags, syscall attributes, fast/slow path, etc). For example, most stable build uses «CONFIG_LIB_SEL4_INLINE_INVOCATIONS», less stable build, which I am debugging now, uses «CONFIG_LIB_SEL4_PUBLIC_SYMBOLS». This configuration crashes at the start when I am trying to allocate memory for the first test: Starting test suite sel4test Starting test 0: TEST_TRIVIAL0001 TEST_TRIVIAL0001 b24e68 b0cc4c b0cc2c b0cc2c sel4test-tests 0 0 * Loading segment 00400000-->005748a4 * Loading segment 00575000-->00689d84 alloc->num_cspace_slots = 30 alloc->num_cspace_slots = 29 alloc->num_cspace_slots = 28 alloc->num_cspace_slots = 27 alloc->num_cspace_slots = 26 alloc->num_cspace_slots = 25 alloc->num_cspace_slots = 24 alloc->num_cspace_slots = 23 alloc->num_cspace_slots = 22 alloc->num_cspace_slots = 21 alloc->num_cspace_slots = 20 alloc->num_cspace_slots = 19 alloc->num_cspace_slots = 18 alloc->num_cspace_slots = 17 alloc->num_cspace_slots = 16 alloc->num_cspace_slots = 15 alloc->num_cspace_slots = 14 alloc->num_cspace_slots = 13 alloc->num_cspace_slots = 12 alloc->num_cspace_slots = 11 alloc->num_cspace_slots = 10 alloc->num_cspace_slots = 9 alloc->num_cspace_slots = 8 alloc->num_cspace_slots = 7 alloc->num_cspace_slots = 6 alloc->num_cspace_slots = 5 alloc->num_cspace_slots = 4 alloc->num_cspace_slots = 3 alloc->num_cspace_slots = 2 alloc->num_cspace_slots = 1 _allocman_cspace_alloc@allocman.c:254 Regular cspace alloc failed, and failed from watermark alloc->num_cspace_slots == 0_refill_watermark:374 slota = 0 alloc->num_cspace_slots = 1 _allocman_cspace_alloc@allocman.c:254 Regular cspace alloc failed, and failed from watermark alloc->num_cspace_slots == 0_refill_watermark:374 slota = 0 alloc->num_cspace_slots = 1 _allocman_cspace_alloc@allocman.c:254 Regular cspace alloc failed, and failed from watermark I see, that: 243: error = alloc->cspace.alloc(alloc, alloc->cspace.cspace, slot); returns the error, and then we allocating memory from the watermark. In other builds (for example, with CONFIG_LIB_SEL4_INLINE_INVOCATIONS attributes), I do not see this issue - there are no errors on return. So, I need to debug function to be sure that I am receiving in the kernel proper arguments and so one, but I do not understand how this part of the code works. Do you have any documentation, or slides or just simple diagrams how this allocator work? What functions are responsible for generating error return, where exactly allocation is performed and how to debug this. Thank you.

1 0

Questions about userspace memory allocation
by Daniel (Xiaolong) Wang 11 Apr '17

11 Apr '17

Hi all, I have a couple questions hope someone can help me clear it out. I understand that after booting up, the kernel keep tracking of all physical memory as Untyped and hand it over to the init thread. It seems like the kernel create both idlethread and init thread and, but it only create a TCB object. 1). I wonder when the init thread's frame and PD are created, and how the thread's stack and heap is allocated? 2). I wonder is there a limit or default value for how much physical memory init thread uses at booting time? Thanks -Dan

2 1

Questions about Userspace memory allocation
by Daniel Wang 09 Apr '17

09 Apr '17

Hi all, I have a couple questions hope someone can help me clear it out. I understand that after booting up, the kernel keep tracking of all physical memory as Untyped and hand it over to the init thread. It seems like the kernel create both idlethread and init thread and, but it only create a TCB object. 1). I wonder when the init thread's frame and PD are created, and how the thread's stack and heap is allocated? 2). I wonder is there a limit or default value for how much physical memory init thread uses at booting time? Thanks -Dan

1 0

Creating process from not root process
by rad021993＠yandex.ru 02 Apr '17

02 Apr '17

Hi all, I try to create process (process2) from not root process (process1). Process1 has own cspace and vspace (fields 'create_cspace' and 'create_vspace' are set to true in process1' config). I try to create process2 like in the root process (4-th example in sel4 tutorials), but I can't obtain bootinfo to create allocman. As I understood it, this info available only from root process. Also I found function to create allocman via bootstrap function (like in example_bootstrap.c), but I also need to get process1' vspace and cspace to attach them to allocman. I do not understand, how I can get the needed info. Could you tell me how I can do it? Best regards, Rinat Dobrokhotov

2 1