fix the sigaltstack hack by lowering the stack size to the original value
if the request was for less than MINSIGSTKSZ (in which case we "round up").
discussed with: marcel

Some linux apps, such as IBM's JDK 1.3, will attempt to mmap thread
stacks near the top of their address space. If their TOS is greater
than vm_maxsaddr, vm_map_growstack() will confuse the thread s

Some linux apps, such as IBM's JDK 1.3, will attempt to mmap thread
stacks near the top of their address space. If their TOS is greater
than vm_maxsaddr, vm_map_growstack() will confuse the thread stack
with the process stack and deliver a SEGV if they attempt to grow the
thread stack past their current stacksize rlimit. To avoid this,
adjust vm_maxsaddr upwards to reflect the current stacksize rlimit
rather than the maximum possible stacksize. It would be better to
adjust the mmap'ed region, but some apps (again, IBM's JDK 1.3) do not
check mmap's return value..

This commit (in conjunction with setting MINSIGSTKSZ to 2048 &
rebuilding your kernel and modules) will get IBM's JDK 1.3 working
with FreeBSD at least well enough to run many of the example applets.

Reviewed by: marcel
Tested by: sto@stat.duke.edu, many others on freebsd-java@

show more ...

This is the first of 3 commits that will get IBM's JDK 1.3 working
with FreeBSD (not including the MINSIGSTKSZ issue, which belongs to
Marcel). Due to time constraints, I'm going to space them out o

This is the first of 3 commits that will get IBM's JDK 1.3 working
with FreeBSD (not including the MINSIGSTKSZ issue, which belongs to
Marcel). Due to time constraints, I'm going to space them out over a
few days.

This fixes two problems with linux_sigaltstack()

o ss == 0 is perfectly valid use, so do not fail in this case.

o Fix flag handling:
- Our SS_DISABLE is 4, linux's is 2, so we need conversion routines.
These conversion routines will be needed by linux_rt_sendsig()
and linux_rt_sigreturn (forthcoming), so they are not static.
- Linux's flag 0 historically meant SS_ONSTACK according to a comment
in their linux/kernel/signal.c file.

Among other things, this fixes a warning from Sun's JDK 1.3:
"Java HotSpot(TM) Client VM warning: cannot uninstall alt signal stack"

Reviewed by: marcel
Tested by: sto@stat.duke.edu, many others on freebsd-java@

show more ...

When sigaltstack is called with a stack size that's not smaller
than LINUX_MINSIGSTKSZ but smaller than MINSIGSTKSZ, cheat and
pass MINSIGSTKSZ to the kernel. This is a workaround.

Submitted through

When sigaltstack is called with a stack size that's not smaller
than LINUX_MINSIGSTKSZ but smaller than MINSIGSTKSZ, cheat and
pass MINSIGSTKSZ to the kernel. This is a workaround.

Submitted through: nate

show more ...

Fix typo in license.

Collect the MD syscalls from /sys/compat/linux here. Since this
is a new file, fix most of the style bugs at the same time.

MFC @199204

Fix some problems with effective mmap() offsets > 32 bits. This was
partially fixed on amd64 earlier. Rather than forcing linux_mmap_common()
to use a 32-bit offset, have it accept a 64-bit file of

Fix some problems with effective mmap() offsets > 32 bits. This was
partially fixed on amd64 earlier. Rather than forcing linux_mmap_common()
to use a 32-bit offset, have it accept a 64-bit file offset. This offset
is then passed to the real mmap() call. Rather than inventing a structure
to hold the normal linux_mmap args that has a 64-bit offset, just pass
each of the arguments individually to linux_mmap_common() since that more
closes matches the existing style of various kern_foo() functions.

Submitted by: Christian Zander @ Nvidia
MFC after: 1 week

show more ...

- MFC

Return ENOSYS instead of EINVAL for invalid function codes to match the
behavior of Linux.

Reported by: Alexander Best alexbestms of math.uni-muenster.de
Approved by: re (kib)

Update the projects tree to a newer FreeBSD current.

Adapt linux emulation to use cv for vfork wait.

Submitted by: Takahiro Kurosawa <takahiro.kurosawa gmail com>
PR: kern/131506

Several cleanups related to pipe(2).

- Use `fildes[2]' instead of `*fildes' to make more clear that pipe(2)
fills an array with two descriptors.

- Remove EFAULT from the manual page. Because of t

Several cleanups related to pipe(2).

- Use `fildes[2]' instead of `*fildes' to make more clear that pipe(2)
fills an array with two descriptors.

- Remove EFAULT from the manual page. Because of the current calling
convention, pipe(2) raises a segmentation fault when an invalid
address is passed.

- Introduce kern_pipe() to make it easier for binary emulations to
implement pipe(2).

- Make Linux binary emulation use kern_pipe(), which means we don't have
to recover td_retval after calling the FreeBSD system call.

Approved by: rdivacky
Discussed on: arch

show more ...

Fix Linux mmap with MAP_GROWSDOWN flag.

Reported by: Andriy Gapon (avg at icyb dot net dot ua)
Tested by: Andriy Gapon (avg at icyb dot net dot ua)
Pointyhat: me
MFC after: 3 days

Implement read_default_ldt in linux_modify_ldt(). It copies out zeroed
descriptor, like real Linux does.

Tested by: Yuriy Tsibizov <yuriy.tsibizov at gmail com>
Submitted by: rdivacky
MFC after: 1 w

Implement read_default_ldt in linux_modify_ldt(). It copies out zeroed
descriptor, like real Linux does.

Tested by: Yuriy Tsibizov <yuriy.tsibizov at gmail com>
Submitted by: rdivacky
MFC after: 1 week

show more ...

i386_set_ioperm, i386_get_ldt and i386_set_ldt are now MPSAFE
(Giant/sched_lock free) so remove unuseful Giant cruft.

Approved by: jeff
Approved by: re
Sponsorized by: NGX Italy (http://www.ngx.it)

Don't add the 'pad' argument to the mmap/truncate/etc syscalls.

Submitted by: kensmith
Approved by: re (kensmith)

Commit 14/14 of sched_lock decomposition.
- Use thread_lock() rather than sched_lock for per-thread scheduling
sychronization.
- Use the per-process spinlock rather than the sched_lock for per-p

Commit 14/14 of sched_lock decomposition.
- Use thread_lock() rather than sched_lock for per-thread scheduling
sychronization.
- Use the per-process spinlock rather than the sched_lock for per-process
scheduling synchronization.

Tested by: kris, current@
Tested on: i386, amd64, ULE, 4BSD, libthr, libkse, PREEMPTION, etc.
Discussed with: kris, attilio, kmacy, jhb, julian, bde (small parts each)

show more ...

Do not dereference linux_to_bsd_signal[-1] if userland has
passed zero as exit signal.

GCC 4.2 changes the kernel data segment layout not to have 0
in that memory location. This code ran by luck bef

Do not dereference linux_to_bsd_signal[-1] if userland has
passed zero as exit signal.

GCC 4.2 changes the kernel data segment layout not to have 0
in that memory location. This code ran by luck before and now
the luck has run out.

show more ...

MFP4: 115220, 115222

- Fix style(9) and reduce diff between amd64 and i386.
- Prefix Linuxulator macros with LINUX_ to prevent future collision.

MFP4: 115094

Linux does not check file descriptor when MAP_ANONYMOUS is set.
This should fix recent LTP test regressions.

Reported by: Scot Hetzel (swhetzel at gmail dot com)
netchild

Partial MFp4 of 114977:
Whitespace commit: Fix grammar, spelling and punctuation.

Submitted by: "Scot Hetzel" <swhetzel@gmail.com>

MFp4 (114193 (i386 part), 114194, 114195, 114200):
- Dont "return" in linux_clone() after we forked the new process in a case
of problems.
- Move the copyout of p2->p_pid outside the emul_lock c

MFp4 (114193 (i386 part), 114194, 114195, 114200):
- Dont "return" in linux_clone() after we forked the new process in a case
of problems.
- Move the copyout of p2->p_pid outside the emul_lock coverage in
linux_clone().
- Cache the em->pdeath_signal in a local variable and move the copyout
out of the emul_lock coverage.
- Move the free() out of the emul_shared_lock coverage in a preparation
to switch emul_lock to non-sleepable lock (mutex).

Submitted by: rdivacky

show more ...

MFP4: 113025, 113146, 113177, 113203, 113500, 113546, 113570

- PROT_READ, PROT_WRITE, or PROT_EXEC implies PROT_READ and PROT_EXEC.
Linux/ia64's i386 emulation layer does this and it complies with L

MFP4: 113025, 113146, 113177, 113203, 113500, 113546, 113570

- PROT_READ, PROT_WRITE, or PROT_EXEC implies PROT_READ and PROT_EXEC.
Linux/ia64's i386 emulation layer does this and it complies with Linux
header files. This fixes mmap05 LTP test case on amd64.
- Do not adjust stack size when failure has occurred.
- Synchronize i386 mmap/mprotect with amd64.

show more ...

Fix LOR that occurs because proctree_lock was acquired while holding
emuldata lock by moving the code upwards outside the emul_lock coverage.

Submitted by: rdivacky