- bug with order in args in mask conversion md code.
- flags can be both int and long on the alpha. kludge around that, by not
passing pointers to the conversion function.
- something is wrong with the toolchain look at the void * cast in copyout().
- sa_restorer is in a different place on the aplha. Deal with that, by not
initializing it as before. XXX: this should be changed in the MD signal
definition.
implement only the kernel version getting entries.
- make the signal conversion functions consistent; dest is arg 1.
- make linux_fakedev deal with block and char devices differently.
- add linux_sys_ugetrlimit, and flesh out the regular get/setrlimit so
that they work properly.
- add linux_mmap2 [untested].
- bump kernel to 2.4.18, and make the date be valentine's day :-)
- linux_sys_*stat64 family was totally busted. Fix it.
tested only on i386.
Status:
gdb, telnet, work
jdk-1.4.0 extractor works, jdk still coredumps.
christos
executable mappings. Stop overloading VTEXT for this purpose (VTEXT
also has another meaning).
- Rename vn_marktext() to vn_markexec(), and use it when executable
mappings of a vnode are established.
- In places where we want to set VTEXT, set it in v_flag directly, rather
than making a function call to do this (it no longer makes sense to
use a function call, since we no longer overload VTEXT with VEXECMAP's
meaning).
VEXECMAP suggested by Chuq Silvers.
format specific.
Struct emul has a e_setregs hook back, which points to emulation-specific
setregs function. es_setregs of struct execsw now only points to
optional executable-specific setup function (this is only used for
ECOFF).
pass socket address in our or out. This makes Linux programs using IPv6
working under emulation. This addresses kern/13279 by Jun-ichiro itojun Hagino.
Change sendmsg()/recvmsg() to expect the control information
to be passed the standard way (not 4.3BSD one, imlied by compat_43 code
or MSG_COMPAT). Linux uses different (arguably broken) CMSG_ALIGN() on some
architectures unfortunately, that would need to be handled eventually.
Also provide some other trivial stubs (like linux_sys_send()) to avoid
unneeded dependance on uipc_syscalls_43.c.
Async I/O OS specifities should now handled in OS specific code. Linux
has been done, but other emulation should be handled. See case LINUX_F_SETFL
in sys/compat/linux/common/linux_file.c:linux_sys_fcntl() for more details.
The data that has been collected yet:
Net Free Open Linux SunOS AIX OSF1 Darwin
send SIGIO to write end of pipe Y N N N N N Y Y
send SIGIO to read end of pipe Y Y N N N ? Y ?
send SIGIO to write end of socket Y Y Y N N Y Y Y
send SIGIO to read end of socket Y Y Y Y Y ? Y ?
neither it does send any SIGIO for pipes. If async I/O
was requested, we keep the SS_ASYNC in struct socket flag
set, but we clear SB_ASYNC flags on the sending buffer
(for socket), and on the sending and the receiving buffer
(for pipes).
Because we do not alter to SS_ASYNC in struct socket,
the Linux process keeps a consistent view of async I/O
status if it attemps to read the async flag (SS_ASYNC)
This async I/O problem does matters, since some Linux
a programs such as the JDK request async I/O on pipes,
but they fail if they happen to get a SIGIO to the write
end of the pipe.
between creation of a file descriptor and close(2) when using kernel
assisted threads. What we do is stick descriptors in the table, but
mark them as "larval". This causes essentially everything to treat
it as a non-existent descriptor, except for fdalloc(), which sees a
filled slot so that it won't (incorrectly) allocate it again. When
a descriptor is fully constructed, the code that has constructed it
marks it as "mature" (which actually clears the "larval" flag), and
things continue to work as normal.
While here, gather all the code that gets a descriptor from the table
into a fd_getfile() function, and call it, rather than having the
same (sometimes incorrect) code copied all over the place.
timezone in the kernel. We now keep track of it for emulated processes.
Tested for powerpc and i386. Untested on m68k, but it should be ok
Alpha has a particular implementation (osf1_sys_gettimeofday), and it has
therefore not been touched.
EMUL_BSD_ASYNCIO_PIPE notes that the emulated binaries expect the original
BSD pipe behavior for asynchronous I/O, which is to fire SIGIO on read() and
write(). OSes without this flag do not expect any SIGIO to be fired on
read() and write() for pipes, even when async I/O was requested. As far as
we know, the OSes that need EMUL_BSD_ASYNCIO_PIPE are NetBSD, OSF/1 and
Darwin.
EMUL_NO_SIGIO_ON_READ notes that the emulated binaries that requested
asynchrnous I/O expect the reader process to be notified by a SIGIO, but
not the writer process. OSes without this flag expect the reader and the
writer to be notified when some data has arrived or when some data have been
read. As far as we know, the OSes that need EMUL_NO_SIGIO_ON_READ are Linux
and SunOS.
unimplemented syscalls. Difference is that linux_sys_nosys() doesn't
post SIGSYS to the process. Linux doesn't do that and thus glibc (and
probably other Linux code) doesn't deal with SIGSYS, even through it
generally contains fallback code for ENOSYS case.
This also addresses port-i386/12503, though the issue was already discussed
earlier.
have vmspace setup yet. This fixes lossage discussed on current-users,
thread "hard lockups with -current and Netscape", and should fix
kern/12433 by <kawamoto@tenjin.org>, maybe also kern/12455.
XXX Threads forked via linux_sys_clone() should share the brk value
XXX if they share vmspace. This needs to be implemented.
only signal handler array sharable between threads
move other random signal stuff from struct proc to struct sigctx
This addresses kern/10981 by Matthew Orgass.
* __HAVE_SYSCALL_INTERN. If this is defined, e_syscall is replaced by
e_syscall_intern, which is called at key places in the kernel. This can be
used to set a MD syscall handler pointer. This obsoletes and replaces the
*_HAS_SEPARATED_SYSCALL flags.
* __HAVE_MINIMAL_EMUL. If this is defined, certain (deprecated) elements in
struct emul are omitted.
*_emul_path variables
change macros CHECK_ALT_{CREAT|EXIST} to use that, 'root' doesn't need
to be passed explicitly any more and *_CHECK_ALT_{CREAT|EXIST} are removed
change explicit emul_find() calls in probe functions to get the emulation
path from the checked exec switch entry's emulation
remove no longer needed header files
add e_flags and e_syscall to struct emul; these are unsed and empty for now
* move all exec-type specific information from struct emul to execsw[] and
provide single struct emul per emulation
* elf:
- kern/exec_elf32.c:probe_funcs[] is gone, execsw[] how has one entry
per emulation and contains pointer to respective probe function
- interp is allocated via MALLOC() rather than on stack
- elf_args structure is allocated via MALLOC() rather than malloc()
* ecoff: the per-emulation hooks moved from alpha and mips specific code
to OSF1 and Ultrix compat code as appropriate, execsw[] has one entry per
emulation supporting ecoff with appropriate probe function
* the makecmds/probe functions don't set emulation, pointer to emulation is
part of appropriate execsw[] entry
* constify couple of structures
Two main changes:
Create a linux_elf64_copyargs that uses the linux specific LinuxAuxInfo
structure. This is only used on the alpha. i386 and m68k use the
standard elf copyargs function.
Since linux's approach to binary compatibilty is to look as much
like osf1 as possible, add all the osf1 syscalls that we have
implemented to the linux syscall table. This includes get/setsysinfo,
ported from FreeBSD.
In order for linux compat to work you must have COMPAT_OSF1, COMPAT_43,
COMPAT_09, COMPAT_12 and COMPAT_13 on also.
semantics as NetBSD's PT_STEP, so support it on archs which have it
under Linux, ptrace(PTRACE_CONT, pid, 0, 0) is used to continue the
traced process where it left off; adjust addr to (caddr_t) 1 so that
the NetBSD syscall code DTRT
some sligh const poisoning
FreeBSD-SA00:42.
Change stackgap_alloc to bounds-check the allocation vs. the stack
gap, returning NULL if there isn't room for the allocation.
Change emul_find() to check for a NULL return from stackgap_alloc()
and convert that into ENAMETOOLONG.
Reorder various emulation syscalls so that all *_CHECK_ALT_{EXIST,CREAT}
calls (which turn into emul_find() under the covers come *after* small,
fixed-size stackgap_alloc() calls.
Clean up ibcs2 {get,set}groups.
zeroes) if the mapped signal number comes up zero. Previously, sigaction1()
would return an error, confusing some linux apps trying to set handlers for
"all" signals.
doing a cpu_set_kpc(), just pass the entry point and argument all
the way down the fork path starting with fork1(). In order to
avoid special-casing the normal fork in every cpu_fork(), MI code
passes down child_return() and the child process pointer explicitly.
This fixes a race condition on multiprocessor systems; a CPU could
grab the newly created processes (which has been placed on a run queue)
before cpu_set_kpc() would be performed.
timeout()/untimeout() API:
- Clients supply callout handle storage, thus eliminating problems of
resource allocation.
- Insertion and removal of callouts is constant time, important as
this facility is used quite a lot in the kernel.
The old timeout()/untimeout() API has been removed from the kernel.
