- l4-hackers - example.com Site

question about read clock API in ARM
by 박찬주 07 Aug '11

07 Aug '11

Hello, I'm newbie in L4 and now i'm trying to find out how to get system clock info. My test environment is QEMU, ARM realview EB. In fiasco kernel, there is a function to read system's clock Clock::read_count(){ return Io::read<Mword>(SYS_24MHZ); } then how can I use this in l4re application? Is there any related API in l4re? Thanks in advance, david

2 1

Sigma0 raised exception on boot
by Anna Fischer 04 Aug '11

04 Aug '11

I'm running Fiasco on the PandaBoard. On platform boot I'm getting an error that Sigma0 raised an exception and then Fiasco panics and the boot process stops. Any idea what the problem could be? I can't find any more information to debug this properly. I have compile Fiasco.OC and L4Re for the PandaBoard platform. The board has 1GB memory. L4 Bootstrapper Build: #162 Wed Aug 3 12:52:02 BST 2011, 4.5.2 Scanning up to 1024 MB RAM Memory size is 1024MB (80000000 - c0000000) RAM: 0000000080000000 - 00000000bfffffff: 1048576kB Total RAM: 1024MB [...] Relocated mbi to [0x8100f000-0x8100f1b1] Loading fiasco Loading sigma0 Loading moe find kernel info page... found kernel info page at 0x80002000 Regions of list regions [ 80001000, 800019ff] { a00} Kern fiasco [ 80002000, 8005cfff] { 5b000} Kern fiasco [ 80090000, 8009e17b] { e17c} Sigma0 sigma0 [ 80140000, 8017160b] { 3160c} Root moe [ 80178000, 8018eec7] { 16ec8} Root moe [ 81000000, 810143eb] { 143ec} Boot bootstrap [ 8100f000, 8100f2ae] { 2af} Root Multiboot info [ 81190000, 8194dfff] { 7be000} Root Modules Memory API Version: (87) experimental Sigma0 config ip:800900e0 sp:81013044 Roottask config ip:80140178 sp:00000000 Starting kernel fiasco at 80001000 Hello from Startup::stage2 Initialize page table Number of IRQs available at this GIC: 160 Vmem_alloc::init() Cache config: ON ID_PFR0: 00001231 ID_PFR1: 00000011 ID_DFR0: 00010444 ID_AFR0: 00000000 ID_MMFR0: 00100103 ID_MMFR1: 20000000 ID_MMFR2: 01230000 ID_MMFR3: 00102111 SERIAL ESC: allocated IRQ 106 for serial uart Not using serial hack in slow timer handler. Welcome to Fiasco.OC (arm)! L4/Fiasco.OC arm microkernel (C) 1998-2011 TU Dresden Rev: r36 compiled with gcc 4.5.2 for Pandaboard [] Build: #4 Wed Aug 3 12:46:10 BST 2011 Calibrating timer loop... done. KERNEL: Warning: Sigma0 raised an exception --> HALT Panic: ... Return reboots, "k" enters L4 kernel debugger...

2 2

Dependencies
by Bryce CR 03 Aug '11

03 Aug '11

Hi Hackers, I'm attempting to build the Fiasco.OC kernel including several additional files. When I attempt to make, I get endless complaints about multiple definitions of functions. The catch is, every single file I have added has a unique include guard... A few things I want to check: > My files are written in standard c++ format, not Preprocess compatible format. E.g. I make use of namespaces. I haven't added the files to any modules lists / Makerules files, so my thinking is that I'm just trying to link with these files but not use any of their features (for testing purposes). My impression was that namespace usage would be tolerated for files used thus: Preprocess doesn't complain. > The c++ preprocessor is still run, correct? I don't see include guards in kernel source files: are these generated by Preprocess, or simply not needed? > Specifically, the errors start appearing at "==> Archiving into jdb_compound.o," and they complain about finding a definition of a function in my source file which was already defined in an object file, e.g.: jdb-ia32-amd64.o: In function `prio_queue_rel::query_iterator_priority__element::next()': /home/bryce/parsnip/fiasco_modified_src/recovery/src/kernel/fiasco/src/kern/R_test11impl.cpp:499: multiple definition of `prio_queue_rel::query_iterator_priority__element::next()' jdb.o:/home/bryce/parsnip/fiasco_modified_src/recovery/src/kernel/fiasco/src/kern/R_test11impl.cpp:499: first defined here I know that's not a lot to go on, but I'm not sure what else would be useful. Please let me know what I do to help you help me... Thank you! --Bryce

2 2

L4Linux and mmap
by Vasily Sartakov 03 Aug '11

03 Aug '11

Hallo. I try to use a mmap for communicate userspace application and other servers via kernel character driver. and its doesn't work. but, of course its work in native linux. my mmap call return no error, but when i try acces to memmory via returned pointer, i got seg fault Here is driver source and example program, based on mmap example: any of vmalloc or kalloc got seg fault: mmap ok, vaddr = 40034000 Segmentation fault //#include <linux/config.h> #include <linux/version.h> #include <linux/init.h> #include <linux/module.h> #include <linux/fs.h> #include <linux/cdev.h> #include <linux/slab.h> #include <linux/vmalloc.h> #include <linux/mm.h> #ifdef MODVERSIONS # include <linux/modversions.h> #endif #include <asm/io.h> /* character device structures */ static dev_t vhsm_dev; static struct cdev vhsm_cdev; /* methods of the character device */ static int vhsm_open(struct inode *inode, struct file *filp); static int vhsm_release(struct inode *inode, struct file *filp); static int vhsm_mmap(struct file *filp, struct vm_area_struct *vma); /* the file operations, i.e. all character device methods */ static struct file_operations vhsm_fops = { .open = vhsm_open, .release = vhsm_release, .mmap = vhsm_mmap, .owner = THIS_MODULE, }; // internal data // length of the two memory areas #define NPAGES 16 // pointer to the vmalloc'd area - alway page aligned static int *vmalloc_area; // pointer to the kmalloc'd area, rounded up to a page boundary static int *kmalloc_area; // original pointer for kmalloc'd area as returned by kmalloc static void *kmalloc_ptr; /* character device open method */ static int vhsm_open(struct inode *inode, struct file *filp) { return 0; } /* character device last close method */ static int vhsm_release(struct inode *inode, struct file *filp) { return 0; } // helper function, mmap's the kmalloc'd area which is physically contiguous int vhsm_kmem(struct file *filp, struct vm_area_struct *vma) { int ret; long length = vma->vm_end - vma->vm_start; /* check length - do not allow larger mappings than the number of pages allocated */ if (length > NPAGES * PAGE_SIZE) return -EIO; /* map the whole physically contiguous area in one piece */ if ((ret = remap_pfn_range(vma, vma->vm_start, virt_to_phys((void *)kmalloc_area) >> PAGE_SHIFT, length, vma->vm_page_prot)) < 0) { return ret; } return 0; } /* character device mmap method */ static int vhsm_mmap(struct file *filp, struct vm_area_struct *vma) { /* at offset 0 we map the vmalloc'd area */ if (vma->vm_pgoff == 0) { return vhsm_vmem(filp, vma); } /* at offset NPAGES we map the kmalloc'd area */ if (vma->vm_pgoff == NPAGES) { return vhsm_kmem(filp, vma); } /* at any other offset we return an error */ return -EIO; } /* module initialization - called at module load time */ static int __init vhsm_init(void) { int ret = 0; int i; /* allocate a memory area with kmalloc. Will be rounded up to a page boundary */ if ((kmalloc_ptr = kmalloc((NPAGES + 2) * PAGE_SIZE, GFP_KERNEL)) == NULL) { ret = -ENOMEM; goto out; } /* round it up to the page bondary */ kmalloc_area = (int *)((((unsigned long)kmalloc_ptr) + PAGE_SIZE - 1) & PAGE_MASK); /* allocate a memory area with vmalloc. */ if ((vmalloc_area = (int *)vmalloc(NPAGES * PAGE_SIZE)) == NULL) { ret = -ENOMEM; goto out_kfree; } /* get the major number of the character device */ if ((ret = alloc_chrdev_region(&vhsm_dev, 0, 1, "mmap")) < 0) { printk(KERN_ERR "could not allocate major number for mmap\n"); goto out_vfree; } /* initialize the device structure and register the device with the kernel */ cdev_init(&vhsm_cdev, &vhsm_fops); if ((ret = cdev_add(&vhsm_cdev, vhsm_dev, 1)) < 0) { printk(KERN_ERR "could not allocate chrdev for mmap\n"); goto out_unalloc_region; } /* mark the pages reserved */ for (i = 0; i < NPAGES * PAGE_SIZE; i+= PAGE_SIZE) { SetPageReserved(vmalloc_to_page((void *)(((unsigned long)vmalloc_area) + i))); SetPageReserved(virt_to_page(((unsigned long)kmalloc_area) + i)); } /* store a pattern in the memory - the test application will check for it */ for (i = 0; i < (NPAGES * PAGE_SIZE / sizeof(int)); i += 2) { vmalloc_area[i] = (0xaffe << 16) + i; vmalloc_area[i + 1] = (0xbeef << 16) + i; kmalloc_area[i] = (0xdead << 16) + i; kmalloc_area[i + 1] = (0xbeef << 16) + i; } return ret; out_unalloc_region: unregister_chrdev_region(vhsm_dev, 1); out_vfree: vfree(vmalloc_area); out_kfree: kfree(kmalloc_ptr); out: return ret; } /* module unload */ static void __exit vhsm_exit(void) { int i; /* remove the character deivce */ cdev_del(&vhsm_cdev); unregister_chrdev_region(vhsm_dev, 1); /* unreserve the pages */ for (i = 0; i < NPAGES * PAGE_SIZE; i+= PAGE_SIZE) { SetPageReserved(vmalloc_to_page((void *)(((unsigned long)vmalloc_area) + i))); SetPageReserved(virt_to_page(((unsigned long)kmalloc_area) + i)); } /* free the memory areas */ vfree(vmalloc_area); kfree(kmalloc_ptr); } module_init(vhsm_init); module_exit(vhsm_exit); test program: #include <stdio.h> #include <unistd.h> #include <sys/mman.h> #include <sys/types.h> #include <sys/stat.h> #include <fcntl.h> #include <stdlib.h> #define NPAGES 16 /* this is a test program that opens the mmap_drv. It reads out values of the kmalloc() and vmalloc() allocated areas and checks for correctness. You need a device special file to access the driver. The device special file is called 'node' and searched in the current directory. To create it - load the driver 'insmod mmap_mod.o' - find the major number assigned to the driver 'grep mmapdrv /proc/devices' - and create the special file (assuming major number 254) 'mknod node c 254 0' */ int main(void) { int fd; unsigned int *vadr; unsigned int *kadr; int len = NPAGES * getpagesize(); if ((fd=open("node", O_RDWR|O_SYNC))<0) { perror("open"); exit(-1); } vadr = mmap(0, len, PROT_READ, MAP_SHARED, fd, 0); if (vadr == MAP_FAILED) { perror("mmap"); exit(-1); } if ((vadr[0]!=0xaffe0000) || (vadr[1]!=0xbeef0000) || (vadr[len/sizeof(int)-2]!=(0xaffe0000+len/sizeof(int)-2)) || (vadr[len/sizeof(int)-1]!=(0xbeef0000+len/sizeof(int)-2))) { printf("0x%x 0x%x\n", vadr[0], vadr[1]); printf("0x%x 0x%x\n", vadr[len/sizeof(int)-2], vadr[len/sizeof(int)-1]); } kadr = mmap(0, len, PROT_READ|PROT_WRITE, MAP_SHARED| MAP_LOCKED, fd, len); if (kadr == MAP_FAILED) { perror("mmap"); exit(-1); } if ((kadr[0]!=0xdead0000) || (kadr[1]!=0xbeef0000) || (kadr[len / sizeof(int) - 2] != (0xdead0000 + len / sizeof(int) - 2)) || (kadr[len / sizeof(int) - 1] != (0xbeef0000 + len / sizeof(int) - 2))) { printf("0x%x 0x%x\n", kadr[0], kadr[1]); printf("0x%x 0x%x\n", kadr[len / sizeof(int) - 2], kadr[len / sizeof(int) - 1]); } close(fd); return(0); } -- Sartakov A. Vasily

2 1

CfP: Workshop on Isolation and Integration in Embedded Systems (IIES)
by Michael Engel 01 Aug '11

01 Aug '11

Dear colleagues, the fourth Workshop on Isolation and Integration in Embedded Systems, organized by Olaf Spinczyk (TU Dortmund), Jörg Nolte (BTU Cottbus) and myself (TU Dortmund), takes place on October 9th, 2011 at ESweek in Taipei, Taiwan. Embedded systems developers today are confronted with a large variety of challenges in the design and implementation of a system. Issues such as co-location of different applications, platform integration of different embedded operating environments as well as safety and protection of embedded software systems require novel approaches to embedded systems development as well as the adaptation of technologies from mainframe and desktop computing, like virtualization and emulation. This, in turn, will change traditional embedded design processes to include hardware and software of growing complexity. The IIES workshop is a platform to discuss recent developments and novel concepts in these areas. We would be happy if you submitted a research or position paper. Submissions may be up to six pages long (ACM SIGOPS Format). Submission deadline: August 15th, 2011 Notification of acceptance: September 12th, 2011 Workshop: October 9th, 2011 Best regards, Michael Engel (michael.engel(a)tu-dortmund.de) -- Dr. Michael Engel michael.engel(a)tu-dortmund.de - engel(a)multicores.org TU Dortmund Phone: +49 231 755 6141 Faculty of Computer Science XII Fax: +49 231 755 6116 Embedded Systems Group - Otto-Hahn-Str. 16 - D-44221 Dortmund, Germany

1 0

L4Re Library Usage
by Wesley Miaw 30 Jul '11

30 Jul '11

I have been able to get my module to compile as an ext-pkg. However I'm a little confused by the description of L4Re listed at http://os.inf.tu-dresden.de/L4Re/overview.html What exactly do I need to replace in my code to use the L4Re C or C++ libraries, or libstdc++ etc.? I didn't get any compilation errors with code that directly calls malloc() and I saw some other modules call malloc() too. But I would have thought all memory allocation/deallocation needs to be changed to use an L4 function. When I looked inside packages like cxx or libstdc++ it doesn't actually look like these are replacements for the GNU headers or libraries. Is the only code I need to have use L4Re functions and types the code that is going to handle message receiving and sending? Everything else could be as if it was not part of L4? Thanks, Wes

2 1

[ANNOUNCE] NOVA Microhypervisor 0.4 prerelease
by Udo A. Steinberg 28 Jul '11

28 Jul '11

Hi, The NOVA project is happy to announce that there is a new prerelease of the NOVA microhypervisor available for download at http://www.hypervisor.org/ under the terms of the GNU General Public License version 2. NOVA is based on a modern microhypervisor written in C++ and assembler. The current implementation runs on multi-core x86 machines that support ACPI, under QEMU (including VM support), and as a microkernel in a virtual machine on top of itself. NOVA makes use of hardware virtualization features such as Intel VT-x/VT-d or AMD-V, and facilitates running unmodified guest operating systems in virtual machines with near-native performance. Among other things, the new prerelease improves the security of device- and interrupt assignment by using capabilities for these operations. There is also a new hypercall for interacting with the scheduler. At the same time, we've released a new version of the NOVA Userland (NUL). While there are few user-visible changes, a lot has changed under the hood. We have integrated a central admission server that is going to implement global policies for CPU time distribution in a future release. Another notable development is rudimentary libvirt support, which will eventually replace the old configuration syntax. Cheers, - Udo

1 0

Re: l4-hackers Digest, Vol 99, Issue 18
by Wesley Miaw 27 Jul '11

27 Jul '11

> On Mon Jul 25, 2011 at 16:43:09 -0700, Wesley Miaw wrote: >> I'm trying to add a new lib/server module to L4 but I'm having some >> trouble. I followed the snapshot top-level README suggestion to >> place the new code in an external directory, so I placed it into >> src/mymodule/ next to src/l4re and src/l4linux. Then I created a >> Makefile under src/mymodule/src/ based on the example with include >> $(L4DIR)/mk/lib.mk. >> >> However, this doesn't appear to be working. $(L4DIR)/mk/Makeconf is >> complaining that files existing inside the obj/l4/... directory are >> not to be found in the obj/mymodule/... directory. (I tried to copy >> the obj/... logic found in the top-level Makefile.) >> >> Is the README's suggestion accurate? > > I think so. Could you show me some of the failed output + surroundings > to better judge what's going on? My bad. I didn't realize the build directory (O=) had to be the l4 obj directory for my own packages. (Missed that part of the README.) It's working for me now. Thanks! -- Wesley Miaw wesley(a)wesman.net

1 0

No configuration file found in build directory
by Wesley Miaw 26 Jul '11

26 Jul '11

I'm trying to add a new lib/server module to L4 but I'm having some trouble. I followed the snapshot top-level README suggestion to place the new code in an external directory, so I placed it into src/ mymodule/ next to src/l4re and src/l4linux. Then I created a Makefile under src/mymodule/src/ based on the example with include $(L4DIR)/mk/ lib.mk. However, this doesn't appear to be working. $(L4DIR)/mk/Makeconf is complaining that files existing inside the obj/l4/... directory are not to be found in the obj/mymodule/... directory. (I tried to copy the obj/... logic found in the top-level Makefile.) Is the README's suggestion accurate? Thanks, -- Wesley Miaw wesley(a)wesman.net

2 1

transfer_msg_items
by Chen Tian 21 Jul '11

21 Jul '11

Hello, Why was a sender's send item (e.g. mr[0] and mr[1] ) merged and stored into the receiver's mr[0]? Basically I have trouble understanding the following code. *rcv_words=(item->b.raw() & ~0x0ff7) | (sfp.raw())&0x0ff0); rcv_word +=2 ; Why didn't we use receiver's mr[1] as well? In the case of page fault handling IPC, the pf address stored in sender's mr[1] is not used ? Thanks. Chen

1 1