- l4-hackers - example.com Site

Re: libfs/ FUSE on L4Re
by Leslie Zhai 01 Jun '17

01 Jun '17

diff --git a/libfs/lib/libc_backend/l4fs.cc b/libfs/lib/libc_backend/l4fs.cc index 7a229ff..31ad120 100644 --- a/libfs/lib/libc_backend/l4fs.cc +++ b/libfs/lib/libc_backend/l4fs.cc @@ -219,10 +219,10 @@ int L4fs_dir::get_entry(const char *name, int flags, mode_t mode, //printf("mode=%x,flags=%x\n", mode, flags); if (S_ISDIR(mode)) { //printf("open '%s' as directory\n", name); - *file = new L4fs_dir(_svr, fh); + *file = cxx::ref_ptr(new L4fs_dir(_svr, fh)); } else { //printf("open '%s' as file\n", name); - *file = new L4fs_file(_svr, fh); + *file = cxx::ref_ptr(new L4fs_file(_svr, fh)); } return (*file) ? 0 : -ENOMEM; @@ -325,7 +325,7 @@ int L4fs_file_system::mount(char const *source, unsigned long mountflags, if (fh < 0) return fh; - *dir = new L4fs_dir(svr, fh); + *dir = cxx::ref_ptr(new L4fs_dir(svr, fh)); if (!dir) return -ENOMEM; 在 2017年06月01日 15:36, Xiang Zhai 写道: > Hi Florian, > > > /home/zhaixiang/project/l4re/l4/pkg/libfs/lib/libc_backend/l4fs.cc: In > member function ‘virtual int L4fs::L4fs_dir::get_entry(const char*, > int, mode_t, cxx::Ref_ptr<L4Re::Vfs::File>*)’: > /home/zhaixiang/project/l4re/l4/pkg/libfs/lib/libc_backend/l4fs.cc:222:38: > error: no match for ‘operator=’ (operand types are > ‘cxx::Ref_ptr<L4Re::Vfs::File>’ and ‘L4fs::L4fs_dir*’) > *file = new L4fs_dir(_svr, fh); > ^ > In file included from > /home/zhaixiang/project/l4re/l4/build/include/l4/l4re_vfs/vfs.h:41:0, > from > /home/zhaixiang/project/l4re/l4/build/include/l4/l4re_vfs/backend:22, > from > /home/zhaixiang/project/l4re/l4/pkg/libfs/lib/libc_backend/l4fs.cc:15: > /home/zhaixiang/project/l4re/l4/build/include/l4/cxx/ref_ptr:186:8: > note: candidate: template<class OT> void cxx::Ref_ptr<T, > CNT>::operator=(const cxx::Ref_ptr<OT>&) [with OT = OT; T = > L4Re::Vfs::File; CNT = cxx::Default_ref_counter] > void operator = (Ref_ptr<OT> const &o) throw() > ^~~~~~~~ > /home/zhaixiang/project/l4re/l4/build/include/l4/cxx/ref_ptr:186:8: > note: template argument deduction/substitution failed: > /home/zhaixiang/project/l4re/l4/pkg/libfs/lib/libc_backend/l4fs.cc:222:38: > note: mismatched types ‘const cxx::Ref_ptr<T>’ and ‘L4fs::L4fs_dir*’ > *file = new L4fs_dir(_svr, fh); > ^ > In file included from > /home/zhaixiang/project/l4re/l4/build/include/l4/l4re_vfs/vfs.h:41:0, > from > /home/zhaixiang/project/l4re/l4/build/include/l4/l4re_vfs/backend:22, > from > /home/zhaixiang/project/l4re/l4/pkg/libfs/lib/libc_backend/l4fs.cc:15: > /home/zhaixiang/project/l4re/l4/build/include/l4/cxx/ref_ptr:193:8: > note: candidate: void cxx::Ref_ptr<T, CNT>::operator=(const > cxx::Ref_ptr<T>&) [with T = L4Re::Vfs::File; CNT = > cxx::Default_ref_counter] > void operator = (Ref_ptr<T> const &o) throw() > ^~~~~~~~ > /home/zhaixiang/project/l4re/l4/build/include/l4/cxx/ref_ptr:193:8: > note: no known conversion for argument 1 from ‘L4fs::L4fs_dir*’ to > ‘const cxx::Ref_ptr<L4Re::Vfs::File>&’ > /home/zhaixiang/project/l4re/l4/build/include/l4/cxx/ref_ptr:203:8: > note: candidate: void cxx::Ref_ptr<T, CNT>::operator=(cxx::Ref_ptr<T, > CNT>::Null_type) [with T = L4Re::Vfs::File; CNT = > cxx::Default_ref_counter; cxx::Ref_ptr<T, CNT>::Null_type = > std::nullptr_t] > void operator = (Null_type) throw() > ^~~~~~~~ > /home/zhaixiang/project/l4re/l4/build/include/l4/cxx/ref_ptr:203:8: > note: no known conversion for argument 1 from ‘L4fs::L4fs_dir*’ to > ‘cxx::Ref_ptr<L4Re::Vfs::File>::Null_type {aka std::nullptr_t}’ > /home/zhaixiang/project/l4re/l4/build/include/l4/cxx/ref_ptr:218:8: > note: candidate: template<class OT> void cxx::Ref_ptr<T, > CNT>::operator=(cxx::Ref_ptr<OT>&&) [with OT = OT; T = > L4Re::Vfs::File; CNT = cxx::Default_ref_counter] > void operator = (Ref_ptr<OT> &&o) throw() > ^~~~~~~~ > /home/zhaixiang/project/l4re/l4/build/include/l4/cxx/ref_ptr:218:8: > note: template argument deduction/substitution failed: > /home/zhaixiang/project/l4re/l4/pkg/libfs/lib/libc_backend/l4fs.cc:222:38: > note: mismatched types ‘cxx::Ref_ptr<T>’ and ‘L4fs::L4fs_dir*’ > *file = new L4fs_dir(_svr, fh); > ^ > In file included from > /home/zhaixiang/project/l4re/l4/build/include/l4/l4re_vfs/vfs.h:41:0, > from > /home/zhaixiang/project/l4re/l4/build/include/l4/l4re_vfs/backend:22, > from > /home/zhaixiang/project/l4re/l4/pkg/libfs/lib/libc_backend/l4fs.cc:15: > /home/zhaixiang/project/l4re/l4/build/include/l4/cxx/ref_ptr:224:8: > note: candidate: void cxx::Ref_ptr<T, > CNT>::operator=(cxx::Ref_ptr<T>&&) [with T = L4Re::Vfs::File; CNT = > cxx::Default_ref_counter] > void operator = (Ref_ptr<T> &&o) throw() > ^~~~~~~~ > /home/zhaixiang/project/l4re/l4/build/include/l4/cxx/ref_ptr:224:8: > note: no known conversion for argument 1 from ‘L4fs::L4fs_dir*’ to > ‘cxx::Ref_ptr<L4Re::Vfs::File>&&’ > > > So I simply added cxx::ref_ptr() for reference counting pointer with > automatic cleanup: > > > diff --git a/libfs/lib/libc_backend/l4fs.cc > b/libfs/lib/libc_backend/l4fs.cc > index 7a229ff..31ad120 100644 > --- a/libfs/lib/libc_backend/l4fs.cc > +++ b/libfs/lib/libc_backend/l4fs.cc > @@ -219,10 +219,10 @@ int L4fs_dir::get_entry(const char *name, int > flags, mode_t mode, > //printf("mode=%x,flags=%x\n", mode, flags); > if (S_ISDIR(mode)) { > //printf("open '%s' as directory\n", name); > - *file = new L4fs_dir(_svr, fh); > + *file = cxx::ref_ptr(new L4fs_dir(_svr, fh)); > } else { > //printf("open '%s' as file\n", name); > - *file = new L4fs_file(_svr, fh); > + *file = cxx::ref_ptr(new L4fs_file(_svr, fh)); > } > > return (*file) ? 0 : -ENOMEM; > @@ -325,7 +325,7 @@ int L4fs_file_system::mount(char const *source, > unsigned long mountflags, > if (fh < 0) > return fh; > > - *dir = new L4fs_dir(svr, fh); > + *dir = cxx::ref_ptr(new L4fs_dir(svr, fh)); > if (!dir) > return -ENOMEM; > > > please review my patch, thanks a lot! > -- Regards, Leslie Zhai - a LLVM hacker https://reviews.llvm.org/p/xiangzhai/

1 0

libfs/ FUSE on L4Re
by Stark, Josef 31 May '17

31 May '17

Hello, yesterday I came across a very interesting paper titled "Porting FUSE to L4Re", accessible here: http://os.inf.tu-dresden.de/papers_ps/pester-beleg.pdf For porting FUSE, the author uses a library called "libfs" which provides filesystem request forwarding and processing, if I understood correctly. It reads as if libfs is already included in L4Re. However, I couldn't find neither FUSE nor libfs in the current L4Re snapshot. Are they by any chance publicly available somewhere else, or are both not to be published? Cheers, Josef

4 7

Microkernel + Alumni Meetup
by Bjoern Doebel 30 May '17

30 May '17

Hi all, it has been a while since we last met for a beer. Water went down the Elbe river, companies were closed, companies were started, research papers got accepted, technical advancements were made. Let's catch up on all of that! To find a suitable time, please visit https://dudle.inf.tu-dresden.de/ukm-01-2017/ Bjoern

2 3

clang-5.0 is not supported. Please use one of the following versions
by Leslie Zhai 28 May '17

28 May '17

Hi L4 hackers, I am using clang (svn trunk 303321) to build fiasco, it thrown such error: kernel/fiasco/src/Makeconf:264: *** clang-5.0 is not supported. Please use one of the following versions: 3.5 3.6 3.7 3.8 so I simply add other versions for clang, please review my patch, thanks a lot! Index: kernel/fiasco/src/Makeconf =================================================================== --- kernel/fiasco/src/Makeconf (revision 72) +++ kernel/fiasco/src/Makeconf (working copy) @@ -245,7 +245,7 @@ CC_WHITELIST-gcc := 4.5 4.6 4.7 4.8 4.9 5 6 -CC_WHITELIST-clang := 3.5 3.6 3.7 3.8 +CC_WHITELIST-clang := 3.5 3.6 3.7 3.8 3.9 4.0 5.0 CC_WHITELIST := $(CC_WHITELIST-$(CC_TYPE)) CC_BLACKLIST-arm-gcc := 4.8.0 4.8.1 4.8.2 -- Regards, Leslie Zhai https://reviews.llvm.org/p/xiangzhai/

2 1

Share exist memory over IPC with L4Re
by Alexander Fratzer 15 May '17

15 May '17

Dear L4-Hackers, I have a very simple problem but I can't find the solution in the L4Re-Doc and examples so far. I wrote a very simple Unix-Like Shell with a Filesystem-Server and a Loader-Server and I implemented some Posix-Call as an abstraction for the communication with the Filesystem-Server (virtual). Now I have a problem of double copies for a read-Operation. I want a complete abstraction from the L4Re mechanisms and the underlying Client-Server architecture for the usage of this interface. The (bad) mechanism is as follow: Function (client side): ssize_t read(int fd, void *buf, size_t count) Algorithm: 1. Allocate Dataspace with size of count. 2. Send Cap of Dataspace to the Filesystem-Server. 3. Copy content of Dataspace to buf. 4. Cleanups. Function (server side): long VFSSrv::op_file_read(VFS::Rights, TaskRegistry::TaskId task, TaskRegistry::FileDescriptor fd, size_t size, L4::Ipc::Snd_fpage& flexp, size_t& rsize) Algorithm: 1. Get Dataspace-Cap from Ipc-fpage. 2. Reserve area with Region-Manager. 3. Attach Dataspace to the area. 4. Copy content of the file to the attached area. 5. Cleanups. This was my first working solution without focus on runtime but the creation of a new task from an ELF-File of 1.4 MB size takes 20 seconds on Cortex A8. This suxx so I have to optimize. Therefor I need a way to share the content of my file (virtual, heap allocated memory) without copying to the client but i don't find a way to share exist memory over IPC. I think I'm using the Dataspace-mechanism in a wrong way but maybe a Dataspace isn't the right choice in my case. A simple solution would be a new Filesystem which is based on Dataspaces but I don't want to do this. Have anybody a better idea? Thanks for help. Best regards Alex.

2 1

Fiasco.OC: null-pointer dereference?
by Stefan Kalkowski 09 May '17

09 May '17

Dear L4-Hackers, recently, I started to upgrade the Fiasco.OC kernel version that is used by the Genode OS framework to the lastly released version (r72). I took the opportunity to upgrade, because the upcoming Genode release uses a fresh compiler toolchain that refused to build the very old Fiasco.OC kernel version that was used until now (r56). Everything went quite smoothly, and I'm glad to see how the kernel develops further. Thanks to all developers at this point! Unfortunately, I stumbled across an issue when it comes to thread destruction. In our system all threads are constructed and destructed by the roottask that is called 'core'. In some cases, not always but quite often, the Ram_quota pointer of the thread object is zero during the call of the Thread_oject's delete operator, which leads to a page-fault within the kernel-code. A simple check[1] before dereferencing the pointer solves the problem, but I wonder whether we will leak quota or memory then, or in general cover some more serious problem. Obviously, we have different usage patterns of syscalls, e.g.: the order of destructing IPC-gates, threads, IRQs, and tasks. Moreover, we still have some very few patches[2] so that the kernel meets our requirements. But none of them explains the thread's Ram_quota pointer getting zero. The page-fault triggers across all x86 and arm platforms that we use. Any hint would be very much appreciated, all the best! Stefan [1] https://github.com/skalk/foc/commit/2b01c9d16fd8e29e6af18fe750be2c8a312b4762 [2] https://github.com/skalk/foc/commits/r72 -- Stefan Kalkowski Genode Labs https://github.com/skalk · http://genode.org/

2 3

Trying to build L4 on NixOS
by Mateusz Czaplinski 09 Apr '17

09 Apr '17

I'm trying to build L4 on NixOS in a VM (long term goal: try to learn how to replace NixOS's kernel with L4Linux, then maybe Genode). I'm getting an error like below: undefined reference to `__stack_chk_fail' during linking of bootstrap.elf - in a GCC's error-barf just after: [bootstrap] ==> Linking bootstrap.elf I've put up the precise Nix expression/script I'm using to build it on github - at: https://github.com/akavel/l4.nix/blob/0108dfcf8d84780e1ba884c41ece0d2dbf4dd… but what it generally boils down to is a sequence of commands like this: $ wget http://os.inf.tu-dresden.de/download/snapshots/l4re-core-2016082114.tar.xz $ tar xf l4re-core-2016082114.tar.xz $ cd l4re-core-2016082114 $ mkdir out $ cp src/l4/mk/defconfig/config.amd64 out/l4/.kconfig $ make -C src/l4 O=out/l4 olddefconfig $ make -C out/l4 I've tried various additional changes, based on some snippets found here and there on teh Internets, like: - adding V=0 to the last make (no idea what it does yet, just some cargo culting) - adding NIX_CFLAGS_COMPILE = "-fno-stack-protector"; to the Nix expression/script (also not really sure where it applies in the stack); but none of the above helped. Any idea what I could try doing to push it further? Thanks, /Mateusz.

2 7

Questions about Fiasco.OC scheduler and L4Linux
by Marc CHALAND 02 Apr '17

02 Apr '17

Hi, I'm doing a study to port old software on top of Fiasco.OC. This soft uses routines which blocks thread scheduling between EnterRegion and LeaveRegion call. Is there any equivalent on TUDOS ? Did somebody already get this challenge ? I would like to know if L4Linux uses Fiasco.OC scheduler or a user land one ? For our i.MX7d board, Freescale provides Linux drivers on 4.1. I don't know yet how many drivers we will need. Do you think it is better to port drivers from 4.1 to L4Linux 4.3 or later or get a L4Linux 4.1. I have no idea about the complexity on doing this. Regards Marc

4 13

L4linux __raw_writesl/readl undefined
by Manolis Ragkousis 01 Apr '17

01 Apr '17

Hello everyone, I am trying to enable sd card support on my zedboard running l4linux and I get this error: ERROR: "__raw_writesl" [drivers/mmc/host/mmci.ko] undefined! ERROR: "__raw_readsl" [drivers/mmc/host/mmci.ko] undefined! make[2]: *** [../scripts/Makefile.modpost:91: __modpost] Error 1 make[1]: *** [/home/manolis/repos/l4linux/Makefile:1173: modules] Error 2 make[1]: *** Waiting for unfinished jobs.... LD ldrImage GZIP vmlinuz LOCINST vmlinuz make[1]: Leaving directory '/home/manolis/repos/l4linux/build' make: *** [Makefile:150: sub-make] Error 2 In order to enable sd card support I understand that I need to enable MMC/SD/SDIO card support in the kernel + ARM AMBA Multimedia Card Interface support. I have followed the instructions from previous emails in this list [1][2]. What am I missing? Is there a config option which enables raw writes/reads on top of L4? Thank you for you time and help, Manolis [1] https://os.inf.tu-dresden.de/pipermail/l4-hackers/2014/006296.html [2] https://os.inf.tu-dresden.de/pipermail/l4-hackers/2014/006417.html

2 2

Can't find device node in /dev/ directory!
by li94575 20 Mar '17

20 Mar '17

Hi, I added some drivers in l4linux, of course, I assigned the corresponding peripheral resources (iomem, irq) to these drivers. From the driver's output log, it seems that they all work well, but when entering the command line, I could not find any relevant device node in /dev/ directory. To my confusion, I can see these devices info in /sys/ directory. The command line args is showing below: mem=64M console=ttyLv0 l4x_rd=rom/ramdisk-" .. L4.Info.arch() .. ".rd ramdisk_size=4000 Any idea?

5 6