Java uses accesses to PROT_NONE region to stop threads sometimes, so it
worthwhile to be able to catch this as fast as possible without taking
vma_list_mutex. The patch does it by setting reserved bit on all ptes in
PROT_NONE VMA which causes RSVD bit to be set in a page fault error code.
It is enough to check it to know that access is to a valid VMA, but
permission is lacking.

Signed-off-by: Gleb Natapov <gleb@cloudius-systems.com>
Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

Currently address we get from pte includes reserved bits and setting
zeroes available and reserved bits.

Signed-off-by: Gleb Natapov <gleb@cloudius-systems.com>
Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

Later patches will use reserved bit, so we want to be sure that it is
available.

Signed-off-by: Gleb Natapov <gleb@cloudius-systems.com>
Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

At the moment, defining DEBUG_VFS fails OSv compilation. The patch
ensures OSv compiles and correct debug logs are emitted.

Reviewed-by: Glauber Costa <glommer@cloudius-systems.com>
Signed-off-by: Prasad Joshi <prasadjoshi.linux@gmail.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>

Currently covering three cases: 1 thread allocation, n threads allocation,
and n threads allocations with frees in a different cpu. We should improve
with more cases.

This test was designed to run on Linux as well. Let's keep it this way.

Signed-off-by: Glauber Costa <glommer@cloudius-systems.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>

According to http://gcc.gnu.org/onlinedocs/libstdc++/manual/abi.html

,
the versions of standard libraries have a strong matchup with versions
of compilers, so linkage errors occur when using libstdc++ in *external*
submodule when using GCC 4.9.0.

As there is such a mixup in the build system, the environments of
standard libraries in link time should be switchable to support GCC
whose version doesn't match up with external's.

*_env variables are introduced. To build a image with standard libraries
in the host, run `make build_env=host'. For fine-grained settings, use
gcc_lib_env and cxx_lib_env.

Reviewed-by: Nadav Har'El <nyh@cloudius-systems.com>
Signed-off-by: Boqun Feng <boqun.feng@linux.vnet.ibm.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>

Linked to wiki page.

Signed-off-by: Takuya ASADA <syuu@cloudius-systems.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>

As Nadav pointed out during review, this macro could use a bit more work, to
use a single parameter instead of one. That is what is done in this patch.
Unfortunately just pasting __COUNTER__ doesn't work because of preprocessor
rules, and we need some indirection to get it working. Also, visibility
"hidden" can go because that is already implied by "static". The problem then
becomes the fact that gcc does not really like unreferenced static variables,
which is solved by the "used" attribute. From gcc docs about "used":

   "This attribute, attached to a variable with the static storage, means that
    the variable must be emitted even if it appears that the variable is not
    referenced."

Signed-off-by: Glauber Costa <glommer@cloudius-systems.com>
Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

Signed-off-by: Claudio Fontana <claudio.fontana@huawei.com>
Cc: Avi Kivity <avi@cloudius-systems.com>
Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

Fixes https://github.com/cloudius-systems/mgmt/issues/33



If a user runs 'java xyz', instead of throwing the stacktrace, a simple
message informing the user about the missing class, should suffice.

Signed-off-by: Jaspal Singh Dhillon <jaspal.iiith@gmail.com>
Reviewed-by: Tomasz Grabiec <tgrabiec@gmail.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>

If a thread which invokes flush_tlb_all() is migrated between the
call to flush_tlb_local() and send_allbutself(), the CPU onto which it
was migrated would not get its TLB flushed.

Spotted during code inspection.

Signed-off-by: Tomasz Grabiec <tgrabiec@cloudius-systems.com>
Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

memory_order_acquire does not prevent previous stores from moving past
the barrier so if the _migration_lock_counter incrementation is split
into two accesses, this is eligible:

tmp = _migration_lock_counter;
atomic_signal_fence(std::memory_order_acquire); // load-load, load-store
<critical instructions here>
_migration_lock_counter = tmp + 1; // was moved past the barrier

To prevent this, we need to order previous stores with future
loads and stores, which is given only by std::memory_order_seq_cst.

Signed-off-by: Tomasz Grabiec <tgrabiec@cloudius-systems.com>
Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

Signed-off-by: Vlad Zolotarov <vladz@cloudius-systems.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>

Take the migration lock for pinned threads instead of a separate
check whether they are pinned or not.

Reviewed-by: Nadav Har'El <nyh@cloudius-systems.com>
Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

Instead of forcing a reload (and a flush) of all variables in memory,
use the minimum required barrier via std::atomic_signal_fence().

Reviewed-by: Tomasz Grabiec <tgrabiec@gmail.com>
Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

Proper memory ordering should be implied to loads and stores of _begin
field. Otherwise they may be reordered with the appropriate stores and
loads to/from the _ring array and in a corner case when the ring is full
it may lead to ring data corruption.

Signed-off-by: Vlad Zolotarov <vladz@cloudius-systems.com>
Reported-by: Nadav Har'el <nyh@cloudius-systems.com>
Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

Signed-off-by: Jani Kokkonen <jani.kokkonen@huawei.com>
Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

These functions are used to demark a critical section and should
follow a contract which says that no operation inside the critical
section should be moved before migrate_disable() and after
migrate_enable(). These functions are declared inline and the compiler
could theoretically move instructions across these.

Spotted during code contemplation.

Signed-off-by: Tomasz Grabiec <tgrabiec@cloudius-systems.com>
Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

The lock has been pushed down to a helper class and cannot be accessed here,
so remove the assert.

Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

revert to a simple implementation which is just atomic,
adding a barrier for the compiler.
The store release is used without load acquire.

Signed-off-by: Claudio Fontana <claudio.fontana@huawei.com>

we start with only a tst-hello.so.

Signed-off-by: Claudio Fontana <claudio.fontana@huawei.com>

move driver setup and console creation to arch-setup,
and ioapic init for x64 to smp_launch, so that we
can remove ifdefs and increase the amount of common code.

Signed-off-by: Claudio Fontana <claudio.fontana@huawei.com>
Reviewed-by: Nadav Har'El <nyh@cloudius-systems.com>

allow execution to flow until main_cont
so we can reach the backtrace.

Signed-off-by: Claudio Fontana <claudio.fontana@huawei.com>

we need to move one instruction back from the return address
to get the address we want to show. Also format the pointers
as long.

Signed-off-by: Claudio Fontana <claudio.fontana@huawei.com>

Signed-off-by: Jani Kokkonen <jani.kokkonen@huawei.com>

[claudio: renamed fixup.hh to fault-fixup.hh]
Signed-off-by: Claudio Fontana <claudio.fontana@huawei.com>

implement fixup fault and the backtrace functionality which is
its first simple user.

Signed-off-by: Jani Kokkonen <jani.kokkonen@huawei.com>

[claudio: added elf changes to allow lookup and demangling to work]

Signed-off-by: Claudio Fontana <claudio.fontana@huawei.com>

Signed-off-by: Jani Kokkonen <jani.kokkonen@huawei.com>
Signed-off-by: Claudio Fontana <claudio.fontana@huawei.com>

Signed-off-by: Claudio Fontana <claudio.fontana@huawei.com>

generate exception frame

Signed-off-by: Jani Kokkonen <jani.kokkonen@huawei.com>
Signed-off-by: Claudio Fontana <claudio.fontana@huawei.com>

Signed-off-by: Claudio Fontana <claudio.fontana@huawei.com>

Signed-off-by: Claudio Fontana <claudio.fontana@huawei.com>

arch::init_on_cpu should initialize the gic cpu interface,
but not for the boot CPU, for which this has already been
done.

Signed-off-by: Claudio Fontana <claudio.fontana@huawei.com>

we need to do that because load/store instructions
behave differently with Device-nGnRnE memory.

Signed-off-by: Claudio Fontana <claudio.fontana@huawei.com>

Signed-off-by: Claudio Fontana <claudio.fontana@huawei.com>

Signed-off-by: Claudio Fontana <claudio.fontana@huawei.com>

Add qemu command path argument to test different versions of qemu.

Signed-off-by: Takuya ASADA <syuu@cloudius-systems.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>

thread::current()->thread_clock() returns the CPU time consumed by this
thread. A thread that wishes to measure the amount of CPU time consumed
by some short section of code will want this clock to have high resolution,
but in the existing code it was only updated on context switches, so shorter
durations could not be measured with it.

This patch fixes thread_clock() to also add the time that passed since
the the time slice started.

When running thread_clock() on *another* thread (not thread::current()),
we still return a cpu time snapshot from the last context switch - even
if the thread happens to be running now (on another CPU). Fixing that case
is quite difficult (and will probably require additional memory-ordering
guarantees), and anyway not very important: Usually we don't need a
high-resolution estimate of a different thread's cpu time.

Fixes #302.

Reviewed-by: Gleb Natapov <gleb@cloudius-systems.com>
Signed-off-by: Nadav Har'El <nyh@cloudius-systems.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>

Again, we are currently calling a function everytime we disable/enable preemption
(actually a pair of functions), where simple mov instructions would do.

Signed-off-by: Glauber Costa <glommer@cloudius-systems.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>

We are heavily using this function to grab the address of the current thread.
That means a function call will be issued every time that is done, where a
simple mov instruction would do.

For objects outside the main ELF, we don't want that to be inlined, since that
would mean the resolution would have to go through an expensive __tls_get_addr.
So what we do is that we don't present the symbol as inline for them, and make
sure the symbol is always generated.

Signed-off-by: Glauber Costa <glommer@cloudius-systems.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>

It is sometimes useful to programatically learn if something will end up
in the main ELF or in a shared library. This is the function usually
performed by _KERNEL. However, using that proved quite difficult when
Asias was trying to conditionally compile some zfs tools.  This is
because a lot of our headers expect _KERNEL to be defined and they are
also included in the tools build. Too messy, unfortunately.

Because of that, I am defining a new constant that will be available for
every object that will end up in the main ELF, but won't be defined
otherwise.

Signed-off-by: Glauber Costa <glommer@cloudius-systems.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>