* introduce fsetown(), fgetown(), fownsignal() - this sets/retrieves/signals
the owner of descriptor, according to appropriate sematics
of TIOCSPGRP/FIOSETOWN/SIOCSPGRP/TIOCGPGRP/FIOGETOWN/SIOCGPGRP ioctl; use
these routines instead of custom code where appropriate
* make every place handling TIOCSPGRP/TIOCGPGRP handle also FIOSETOWN/FIOGETOWN
properly, and remove the translation of FIO[SG]OWN to TIOC[SG]PGRP
in sys_ioctl() & sys_fcntl()
* also remove the socket-specific hack in sys_ioctl()/sys_fcntl() and
pass the ioctls down to soo_ioctl() as any other ioctl
change discussed on tech-kern@
Some initial support for IOHIDSystem. This gives us some pointer moves in X11
(not really related the the actual mouse movement, but this will come).
The darwin_iohidsystem_thread reads events from wscons, translates them
into IOHIDSystem events and wakes up the userland client with a notification.
To do this, I had to improve the void implementation of
io_connect_set_notification_port() to actually register something (I assumed
a single notification port, which makes some sense since only one process
seems to be able to open the driver)
Missing bits:
- we do not take event masks given by the process into account.
- the notification message has not been checked against Darwin
- events are badly translated
(not really related the the actual mouse movement, but this will come).
The darwin_iohidsystem_thread reads events from wscons, translates them
into IOHIDSystem events and wakes up the userland client with a notification.
To do this, I had to improve the void implementation of
io_connect_set_notification_port() to actually register something (I assumed
a single notification port, which makes some sense since only one process
seems to be able to open the driver)
Missing bits:
- we do not take event masks given by the process into account.
- the notification message has not been checked against Darwin
- events are badly translated
and make the stack and heap non-executable by default. the changes
fall into two basic catagories:
- pmap and trap-handler changes. these are all MD:
= alpha: we already track per-page execute permission with the (software)
PG_EXEC bit, so just have the trap handler pay attention to it.
= i386: use a new GDT segment for %cs for processes that have no
executable mappings above a certain threshold (currently the
bottom of the stack). track per-page execute permission with
the last unused PTE bit.
= powerpc/ibm4xx: just use the hardware exec bit.
= powerpc/oea: we already track per-page exec bits, but the hardware only
implements non-exec mappings at the segment level. so track the
number of executable mappings in each segment and turn on the no-exec
segment bit iff the count is 0. adjust the trap handler to deal.
= sparc (sun4m): fix our use of the hardware protection bits.
fix the trap handler to recognize text faults.
= sparc64: split the existing unified TSB into data and instruction TSBs,
and only load TTEs into the appropriate TSB(s) for the permissions.
fix the trap handler to check for execute permission.
= not yet implemented: amd64, hppa, sh5
- changes in all the emulations that put a signal trampoline on the stack.
instead, we now put the trampoline into a uvm_aobj and map that into
the process separately.
originally from openbsd, adapted for netbsd by me.
truncate64() wrapper to translate args structure
NetBSD truncate() and ftrucate() have hidden 'pad' argument, so we have
to do the argument translation
Problem found and patch supplied in PR kern/22360 by Ales Krenek
This is the last of syscalls with hidden 'pad' arg we didn't have
wrapper for; all the others (lseek, mmap, pwrite, pread) already had
wrapper before.
address. We do that through the irix_load_addr function, which is
not IRIX specific at all. If another emulation needs it, it can
easily move to compat_util.c
With this change, IRIX dynamic binaries are able to link and run again
(top down UVM broke them a few weeks ago)
* translate MSG_* flags
* rewrite CMSG level/type to appropriate NetBSD value on input, and to Linux
value on output
* handle different CMSG_DATA alignment for some archs
This fixes SCM_RIGHTS passing. Other SCM_* types are not supported -
the set is different on NetBSD and Linux. SCM_TIMESTAMP doesn't seem
to be actually implemented in Linux 2.5.15, so it's not supported
for Linux binaries either (for now).
PR: 21577 by Todd Vierling
fix value for LINUX_SO_DETACH_FILTER on couple places (unused anyway)
g/c LINUX_SCM_TIMESTAMP definition for some archs, it will be defined
in generic linux_socket.h
be inserted into ktrace records. The general change has been to replace
"struct proc *" with "struct lwp *" in various function prototypes, pass
the lwp through and use l_proc to get the process pointer when needed.
Bump the kernel rev up to 1.6V
We try to map the console framebuffer through wsdisplay. It seems to work,
but we still need to _see_ something.
XXX Framebuffer size discovery is borken.
+ In ioframebuffer and iohidsystem:
More methods and more debug for io_connect_method_*
+ A few new IOKit mach traps.
Peeking at Darwin's header files gives some ideas of what we are doing here.
We know understand most of the stuff called by XDarwin and WindowServer before
mapping the framebuffer.
return packet was wrong (port descriptor instead of VM area descriptor),
thus leading to SIGSEGV when calling IOKit's IOFBCreateSharedCursor().
Now this is fixed we can move forward again on graphic mode.
for the IOHIDSystem driver in Darwin. IOHIDSystem provides access
to user interface devices.
Communication between userspace and kernel is done through a shared
memory page. Our plan is to have a kernel thread waiting on wscons
events (for now it just does nothing), and have it feeding the
shared page.
the off_t argument, Darwin does not. In order to get the off_t argument
going through our syscall machinery, we declare it as two long arguments,
and we reassemble it in darwin_sys_lseek.
request a mapping at fixed address, overlaying previously existing mappings.
The overlay is done by unmapping (sys_munmap) before mapping at fixed
location. This makes XDarwin able to link and run.
Avoids a lot of casting and removes the need for some line breaks.
Removed a load of (caddr_t) casts from calls to copyin/copyout as well.
(approved by christos - he has a plan to remove caddr_t...)
lookup and allocation, and any dependency on NPROC or MAXUSERS.
NO_PID changed to -1 (and renamed NO_PGID) to remove artificial limit
on PID_MAX.
As discussed on tech-kern.
of cycles off the syscall overhead.
Since all COMPAT_LINUX platforms now support __HAVE_SYSCALL_INTERN,
garbage-collect the LINUX_SYSCALL_FUNCTION stuff.
Darwin specific files. The handler does nothing yet.
This needs some improvement. Darwin attaches an IOHIDSystem device to
each human-interface device in the tree: mouse, keyboard, and a few others.
For now we only emulate one IOHIDSystem device.
- introduce mach_port_destroy (badly emulated for now)
- on fork/exec, don't set l_private, as the process might be not mature
enough to survive a proc_representative_lwp call (-> kernel panic)
Once we give WindowServer the io_master_port, it sends a message to it:
305 WindowServer MMSG id 2804 [0x1310009 -> 0x131000e] -12016 bytes, flags 0
x1513
0x0000 0x00001513 0xffffd110 0x0131000e 0x01310009 .........1...1..
0x0010 0xffffcfd0 0x00000af4 0x00000000 0x00000000 ................
0x0020 0xffffcf60 0x00000052 0x3c646963 0x74204944 ...`...R<dict ID
0x0030 0x3d223022 0x3e3c6b65 0x793e494f 0x50726f76 ="0"><key>IOProv
0x0040 0x69646572 0x436c6173 0x733c2f6b 0x65793e3c iderClass</key><
0x0050 0x73747269 0x6e672049 0x443d2231 0x223e494f string ID="1">IO
0x0060 0x48494453 0x79737465 0x6d3c2f73 0x7472696e HIDSystem</strin
0x0070 0x673e3c2f 0x64696374 0x3e0047cc g></dict>.G.
Did you ever dreamt about system call arguments in XML?
are using lwp.
Enable kill -9 to kill parent thread waiting for its child.
Use upcallret instead of child_return for the newly created lwp.
Add debug messages for thread creation.
Thread creation are still broken...
There are two new mach traps:
mach_vm_make_memory_entry, mach_host_get_io_master. Now we need to find
the documentation to understand what they are supposed to do.
send, send once, and dead names, not for port sets and receive rights.
This make vi and telnet able to work again.
Also removed the all process right list and its lock, which got useless. The
all process lock is replaced by a per process lock, located in struct
mach_emuldata.
Also one bug fix: we did not correctly called Mach hooks for struct emuldata
initialization and release for Darwin processes.
the right in the destination process.
This is a small step backward for functionnality: vi does not work anymore
because our right checks cause some spurious errors, but this will be fixed
later.
mach-O header. This is a guess based on what we have on Darwin, we need
to check other Mach systems to see if this value is Darwin specific or not.
Also updated copyright and removed useless debug
and a body.
- If mach_init is not availabkle for boostrap requests, try to handle them
in the kernel (we don't really handle them, we just try to avoid hanging there)
- minor tweaks.
1) rights should be shared by the threads within a process. While it would
be easier to handle this with the struct proc/struct lwp split, we attempt to
do this now by sharing the right lists. Because each right holds a reference
to struct proc, this might cause some problems later.
2) in pthread_exit, really exit the thread. Also reintialize the righ tlist to
make sure we will not destroy the parent's right list
3) rights can hold multiple permissions on a port (ie: send and receive). Fix th
is.
4) first attempt on right carried by messages. We still have to do rights carrie
d in the message body (complex messages).
for forking the traditional UNIX init(8) and it does the Mach port naming
service. We need mach_init for the naming service, but unfortunately, it
will only act as such if its PID is 1. We introduce a sysctl
(emul.darwin.init_pid) to fool a given process into thinking its PID is 1.
That way we can get mach_init into behaving as the name server.
Typical use:
/sbin/sysctl -w emul.darwin.init_pid=$$ ; exec /emul/darwin/sbin/mach_init
Perhaps this file should be changed to use trace_enter() so that
systrace works for aliased syscalls? Leave that for someone familiar
with the systrace system.
keep track of the receive right for the port. There can be only one, and
the receive right references struct proc should we need it.
This makes easier to find the receive right associated with the port.
This does not buy us new functionnality for now, because we still have to
discover how mach_init (which acts as a name server, enabling processes to
discover each other's ports) is able to receive messages from other processes
(this is a bootstrap problem, and the bootstrap port might be the place to
search).
While we are there:
- removed a lot of debug which is now available using ktrace.
- reworked message handling to avoid mutliple copyin/copyout of the
same data. ktrace of Mach message now uses the in-kernel copy of the
message instead of copying it from userland.
- packed mach trap handlers arguments into a structure to avoid modifying
everything next time we have to add an argument.
mach_port_move_member, mach_port_set_attributes, mach_task_set_special_port,
(none do anything)
Added mach_thread_create_running, which creates a new Mach thread. It
provides the register context of the new thread. We use it in a child
function provided to fork1(). The child function is machine dependent and
is not yet implemented for i386.
The new thread crashes quickly, but at least it starts.
the passed device number to only contain lower 16 bits; older glibc
appears to pass some junk in upper bits sometimes, resulting in incorrect
device entries being created
Problem initially analyzed by Thor Lancelot Simon.
two ways:
- the child gets its pid as retval[0] (userland stub will turn it into a 0),
retval[1] is 1 and it is 0 in the parent.
- in the child, the fork syscall is successful, hence we must skip the next
instruction.
compat/common, so that they can be shared by several emulations, and use
them for Darwin.
This removes the ugly dependance on FreeBSD freebsd_file.c for COMPAT_DARWIN
used to get and set the thread user value, which is an opaque pointer to
a per thread structure stored in userland. cthread_self() is used by Darwin
as an implementation for pthread_self(), which return the thread id.
We use the p_emuldata field of struct proc in order to keep track of the
thread user value. For now the value is per-process, but we will make it
per-thread when we will take care of threading.
While we are there, do some KNF
merge the two emul_irix structures; the only difference was
setregs function, which can be handled by exec-specific setregs hook
rename setregs_n32() to irix_n32_setregs(), and make it suitable
as the exec-specific setregs hook
make irix_check_exec() a macro now that just single compare
it checks both the alternative/emul tree, and the non-emul tree.
This makes it possible to run chrooted emulated binaries without need
to setup shadow /emul tree within the chroot hierarchy.
XXX sligh code rearrangement was necessary, change not tested
it checks both the alternative/emul tree, and the non-emul tree.
This makes it possible to run chrooted emulated binaries without need
to setup shadow /emul tree within the chroot hierarchy.
Only tested for COMPAT_LINUX, changes to other compat modules were
mechanical.
Fixes kern/19161 by Christian Groessler.
in alternative emul tree first, and updates interpreter pathname
if found there; if not found in alternative emul tree, pathname
without the emul prefix is checked, and error returned if the file
doesn't exist
Check for target buffer length, and fail if it is too short
Move mach_msg_trap and mach_msg_overwrite_trap to their own file
Remove some useless debug messages now we have ktrace
Remove __P()
macho_hdr, argc, *argv, NULL, *envp, NULL, progname, NULL,
*progname, **argv, **envp
Where progname is a pointer to the program name as given in the first
argument to execve(), and macho_hdr a pointer to the Mach-O header at
the beginning of the executable file.
is already used, we want to remap to another place. The vmcmd must be
completely rebuild, else things does not work as expected, as the vmcmd has
been modified by the first mapping attempt.
this gives:
* linux sysconf(_SC_CLK_TCK) gives correct value for linux binaries (hz)
even if hz != 100
* glibc gets proper information on real/effective uid and enables
secure mode for suid binaries
g/c LINUX_COPYARGS_FUNCTION, replaced by linux ELF copyargs function
g/c alpha-specific linux ELF copyargs function and linux ELF defines
cl