Add cwd (current directory) and env (environment variable) option to the
loader. Can be useful for certain applications that expect to be in a
certain directory, or certain environment variables to exist.

Example usage:

   run.py -e "--cwd=/tmp /usr/bin/something.so"

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

Test case for dladdr(), to make sure it returns correct symbol in these cases:
 - addr is less than 'vfprintf'. Should returns another symbol.
 - addr is equals to 'vfprintf'. Should returns 'vfprintf' as the result.
 - addr is bigger than 'vfprintf', and also inside of it. Should returns 'vfprintf' as the result.

Reviewed-by: Nadav Har'El <nyh@cloudius-systems.com>
Signed-off-by: Takuya ASADA <syuu@cloudius-systems.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>

Instead of acquiring sockbuf::sb_mtx inside sblock() and sbunlock(), rely
on the caller to take the lock for us.  Expand existing lock hold regions
in callers to make it so.  This reduces acquisitions of sb_mtx.

As a side effect, copies to and from userspace are done under the lock.
This can affect MSG_NOWAIT with demand paging major faults, but these are
screwed anyway.

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

sb_rwlock is used to serialize concurrent writers (or readers) to the same
socket buffer, but is quite expensive as it requires 4 atomic operations
per transaction, even if there is no contention.

Replace it with a waitqueue, and use the sockbuf::sb_mtx for serialization.
This still has exactly the same cost, but we can later move sblock() and
sbunlock() into contexts where the sockbuf::sb_mtx is already acquired.

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

sblock() takes sb_rwlock for reading, and is the only such locker, so it
clearly has no effect.

Change it to acquire the lock for writing, so it serializes access to the
socket buffer as intended.

Bug introduced in 6296cbab.

Reviewed-by: Glauber Costa <glommer@cloudius-systems.com>
Reviewed-by: Nadav Har'El <nyh@cloudius-systems.com>
Signed-off-by: Avi Kivity <avi@cloudius-systems.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>

This allows placing C++ objects in ifnet.

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

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

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

Include the system header and remove duplicate definitions.

Change some solaris imports to use _GNU_SOURCE to make rlim64_t defined
consistently.

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

Allows making ifnet a C++ class.

Reviewed-by: Glauber Costa <glommer@cloudius-systems.com>
Reviewed-by: Nadav Har'El <nyh@cloudius-systems.com>
Signed-off-by: Avi Kivity <avi@cloudius-systems.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>

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

Reviewed-by: Glauber Costa <glommer@cloudius-systems.com>
Reviewed-by: Nadav Har'El <nyh@cloudius-systems.com>
Signed-off-by: Avi Kivity <avi@cloudius-systems.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>

It is a bad practice to have DHCP discovery without timeout
and retries. In case discovery packet gets lost boot stucks.

Beside this there is an interesting phenomena on some systems.
A few first DHCP discovery packets sent on boot get lost in some cases.

This started to happen from time to time on my KVM system and almost
every time on my Xen system after installing recent Fedora Core updates.
Packet leaves VM's interface but never arrives to bridge interface.
The packet itself built properly and arrives to DHCP server just fine
after a few retransmissions.

Most probably this phenomena is a bug (or limitation) in the current
Linux bridge version so this patch is actually a work-around, but
since in general case it is a good idea to have DHCP timeouts/retries
it worth to have it anyway.

Signed-off-by: Dmitry Fleytman <dmitry@daynix.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>

Force version.h generation to make sure the version number matches the
git version.  I noticed the problem after tagging v0.05-rc1 and noticing
that OSv happily reported the same version even after applying patches.

Use the special ".PHONY" target suggested by Tomek to avoid breakage
from previous attempt that is documented in commit 996fdfde ('Revert
"build: Force version.h generation"').

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

I don't know how chdir() ever worked - apparently it didn't!

It took an argument "pathname", and then declared a local "path" and used
that, not pathname, as the path :-) Obviously, a call to task_conv, which
converts a relative "pathname" to an absolute "path", was missing...

chdir() is still a mess and incompatible in the error cases with Linux's
chdir(). I'll fix that, and add a test, in a follow-up patch.

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

Lightweight waitqueue and wait_for() primitive.

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

Dor Laor writes:

  - Ballooning causes to issues and we like to disable it:

    1. Parallel GC
        2 GC threads are trying to move memory around and they
        do it optimistically without locking. This fools our memmov logic.
        Gleb and Glauber have a solution (to use GCmonitor) but will take
        time to implement.

    2. Logger clash
        The existing GCmonitor bean access the logger early and we don't
        have the tomcat logger handy. Tomek is working on a solution but
        it will take few days to complete.

Therefore, disable the JVM balloon for v0.05.

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

Reviewed-by: Tomasz Grabiec <tgrabiec@gmail.com>
Signed-off-by: Gleb Natapov <gleb@cloudius-systems.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>

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

Allow arbitrary predicates to be waited on.

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

Add sockbuf::sb_cc_wq for using instead of msleep(&sb->sb_cc).  This paves
the way for lockless wakeups for net channels, as the wait primitive is now
thread::wait_for() instead of msleep().  In addition this reduces lock
acquisition and improves netperf bandwidth by about 10%.

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

This was used to ensure all socket-using code was converted to C++, but not
cleaned up later.

Clean it up now.

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

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

A waitqueue is an object on which multiple threads can wait; other threads
can wake up either one or all waiting threads.  A waitqueue is associated
with an external mutex which the user must supply for both wait and wake
operations.

Waitqueues differ from condition variables in three respects:
- waitqueues do not contain an internal mutex.  This makes them smaller, and
  reduces lock acquisitions.  On the other hand the waker must hold the
  associated mutex, whereas this is not required with condition variables.
- waitqueues support sched::thread::wait_for()

waitqueues support wait morphing and do not cause excess lock contention,
even with wake_all().

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

This adds a facility to wake a thread, but with the intention that it will
acquire a certain lock after waking, and while the waker holds the lock.
This is implemented using the regular wait morphing code (send_lock() and
receive_lock()), but with additional mutual exclusion to allow regular
wake()s in parallel.

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

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

Add a signal catching class (signal_catcher).

Use signal_catcher::wait() to wait for a signal.
Use sched::thread::wait_for(..., signal_catcher&, ...) to wait for a signal
in addition to other events.

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

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

Waiting is complicated; you can wait for a condition variable, a timer,
wait queues and signals (when we have them), semaphores, and more.

When you need to wait for multiple events, things become even more
complicated as some sort of abstraction is needed.  poll() for example
supports waiting for multiple objects that conform to the fd interface,
or for a timeout.

To allow more flexible waiting, this patch adds a poll-like interface
to the scheduler, except that the interfaces are all compile time.

We add a new function wait_for() that accepts a variable number of waitable
objects.  A waitable object is an object that supports the three methods:

 - poll(): check if waiting is complete
 - arm(): prepare for waiting
 - disarm(): finish waiting

to make a waitable object usable with wait_for, these functions have to be
supplied.

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

Reviewed-by: Nadav Har'El <nyh@cloudius-systems.com>
Signed-off-by: Vlad Zolotarov <vladz@cloudius-systems.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>

Reviewed-by: Nadav Har'El <nyh@cloudius-systems.com>
Signed-off-by: Vlad Zolotarov <vladz@cloudius-systems.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>

Reviewed-by: Nadav Har'El <nyh@cloudius-systems.com>
Signed-off-by: Vlad Zolotarov <vladz@cloudius-systems.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>

Remove the extra check on size just like the remark above implies.

Reviewed-by: Nadav Har'El <nyh@cloudius-systems.com>
Signed-off-by: Vlad Zolotarov <vladz@cloudius-systems.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>

The check is moved to bdev_read and bdev_write.

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

setup_features() is called twice, once in virtio_driver constructor and
once in virtio device driver constructor. Doing it in driver constructor
is enough.

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

This would be useful for device with mmio bar, e.g. AHCI controller.

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

The OSv loader has a "--verbose" option to enable debug() output to be
printed to the console and not just to a memory buffer. However, needing to
pass this option in the "-e" string is annoying, and even worse when you want
to use the default command line and didn't even want an explict "-e" option.

So this patch adds a --verbose to run.py, which can be used whether or not
a "-e" option is given. For example:

    run.py --verbose
or
    run.py --verbose tests/tst-timerfd.so

As a shorthand for --verbose, I used "-V". Unfortunately, lowercase "-v"
was already taken (by the "vhost" option).

Fixes #173.

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

As issue #128 explains, it is annoying that once the "-e" option was used
in run.py, it overrides the default command line, so that running run.py
again without -e reruns the same modified command line, and not the default
one as usual.

This simple patch changes this behavior: On every run of run.py, the command
line is set - if "-e" is given it is set to the given command line, but if
"-e" is absent, the command line is reset to the default one. This is
basically the behavior that people always wanted from run.py.

This patch, and the behavior it performs, are much simpler than any of
the alternatives proposed in issue #128.

Please don't complain that this can only be used for development, and not
for EC2, for creating images to distribute, and so on. After all, this is a
patch to *run.py*, and it is obviously not relevant to anyone who is running
an OSv image not through run.py. This patch only makes run.py more convenient
to use.

Fixes #128.

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

sysinfo() until now was a stub. This patches fills some of the information
we can already provide, namely the uptime in seconds (through the recently
added osv::clock::uptime()), the total and free memory (through the recently
added memory::stats::free()/total() functions), and number of processes
(always 1 on OSv).

This patch also changes a couple of types from unsized names (like "unsigned")
to fixed-sized ones (like "u32"), to match the definition of this structure
on Linux. It won't make any difference on x86_64, but perhaps in some future
architecture it will.

Tested-by: Amnon Heiman <amnon@cloudius-systems.com>
Signed-off-by: Nadav Har'El <nyh@cloudius-systems.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>

sysinfo() used sizeof(struct sysinfo) to decide how many zeros to write.
The problem is that our struct was much larger than Linux's (we had a much
larger padding at the end), so if we run code compiled on Linux, that code
would allocate a small structure on the stack and then zero a much larger
structure, causing the stack to be corrupted.

Drop the "reserved" field making our structure larger. It's not needed.

Fixes #174.

Tested-by: Amnon Heiman <amnon@cloudius-systems.com>
Signed-off-by: Nadav Har'El <nyh@cloudius-systems.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>