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

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

Rescheduling in nested exception is not supported since nested exception
stack is per cpu, but without the assert such reschedule will cause
stack corruption which will be hard to debug.

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

AHCI has 32 cmd slots to issue cmd. Only first slot is used currently
which makes the queue-depth 1.

This patch uses all the cmd slots and makes the cmd completion async.
Now, the queue-depth is 32.

Test with "/tests/misc-bdev-write.so" on VBOX shows improvements:

   Before: ~10MB/s
   After: ~20MB/s

1000 round of "/tests/misc-bdev-rw.so" tests passed on VBOX and QEMU.

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

Assert to make sure the data buffer address satisfies the AHCI spec's data
alignment requirement.

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

AHCI spec '4.2.2 Command List Structure' says 'Command Table Descriptor
Base Address' must be aligned to 128-byte cache line, indicated by bits
06:00 being reserved.

All the 32 cmd_table are allocated in one linear space. The size of
cmd_table is 144 bytes for now which is larger than 128 bytes. So we pad
cmd_table to 256 byes.

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

One PRDT entry can contain 4MB buffer at most, we currently only use one
PRDT per AHCI cmd. Limit the size of a bio request to respect it. Larger
bio will be split into smaller bios by multiplex_strategy.

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

Add missing scandir() function from musl 1.0.0. Fixes #237.

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

We now print disk probe information like:

virtio-scsi: Add scsi device target=1, lun=0   as vblk0, devsize=10842275840

Disable it by default. The --verbose option will make this info
available which is useful when debug.

Reported-by: Pekka Enberg <penberg@cloudius-systems.com>
Signed-off-by: Asias He <asias@cloudius-systems.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>

It will create a file on-disk twice as large as memory, and then will map it entirely
in memory. The file is then read from using 3 different sequential patterns, and then
later on 2 threaded patterns.

This test does not handle writes.

It goes in misc because it takes a very long time to run (especially with a random pattern)

Example output:

Total Ram 586 Mb
Write done
Double Pass OK (13.6323 usec / page)
Recency OK (3.35954 usec / page)
Random Access OK (640.926 usec / page)
Threaded pass 1 address ended OK
Threaded pass many addresses ended OK
PASSED

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

Gleb have noticed a problem with the buffer mapping logic, that would make
unrelated mappings that would be adjacent to the file mapping be invalidated as
well. This is because we were invalidated an area as large as the buffer.  If
the mapping would, for instance, end in the middle of the buffer, we would keep
invalidated adjacent regions.

The solution is to change the code to handle per-page, not per-buffer mappings.
We would likely need something like that anyway for COW maps, so let's just
start handling it now.

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

This case is not supposed to be utterly common, but we have to handle it
anyway.  This implementation is slightly naive, but it should be okay if the
number of mappings is small.  We should revisit it this number start growing
too much

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

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

Historically, we only call vma::set() for non-fixed mappings. The reason is
that for fixed mappings, we already know and set which address we are putting
the map at. However, now that we have more operations tagging along vma::set(),
it is better to call it unconditionally.

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

The vma class has a method to update its range. Use it during split, instead
of naked coding it. It will be particularly important if we want to tag other
operations along the range update.

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

To better deal with shared buffers, we should expand the interface to
get_page/put_page.  Three changes are needed: first, the start of the virtual
address is passed to both. Together with the offset, it can be used to uniquely
identify the virtual address being mapped. That will be used to manage the
mapping list.

Regarding the offset itself, there is need to also pass the file offset
downwards, the one specified by the mapping. Reading that from the file pointer
is not enough, because not only this is not what the user has asked for, but it
can change during the map's lifetime. It also should be a separate parameter
from the in-mapping offset, not summed together with it. Doing otherwise would
make impossible for the mapping function to differentiate between those two
concepts

Also, put_page is called from the free operation, but does not receive the
actuall address from it. Knowing the physical address will be important to shut
down our mapping list.

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

Provide a map and unmap function that maps directly into the ARC cache.  Most
of the boilerplate code is inspired in the read function, but I have decided to
leave it separate to avoid complicating the original functions.

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

We won't require all filesystems to implement it, those who don't should return
ENOSYS.

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

That will allow us to read a dmu arc-backed buffer and acquire a pointer to it,
instead of copying it somewhere else. This function will both do mapping and
unmapping, depending on its last parameter.

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

I am using the same name as solaris code expect. The first field is an uio,
and the rest is something that we will use in specialized functions.

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

The underlying filesystem won't be able to back it with huge pages, so force
small.

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

To avoid overloading "change_perm"'s meaning, and taking advantage of our nice
template based page walker, this patch creates another walker that will unmap
a range of pages, marking them all as non present.

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

This comes together with the placeholder in the mmu-side to get this
warning. The actual functionality is not yet implemented

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

A new (private) flag is added that marks a buffer as shared between
ZFS and the page cache. It indicates that before evicting this buffer,
ZFS will have to take some action. Right now no action is taken besides
clearing the flag.

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

If the flags is set on vma it is backed by small pages only.

Signed-off-by: Gleb Natapov <gleb@cloudius-systems.com>
Reviewed-by: Glauber Costa <glommer@cloudius-systems.com>

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

Add support for SA_RESETHAND signal handler flag, which means that the signal
handler is reset to the default one after handling the signal once.

I admit it's not a very useful feature (our default handler is powering off
the system...) but no reason not to support it.

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

This script list up missing libc function called from a shared library.
Usage:
./scripts/check-libcfunc-avail.sh release apps/memcached/memcached.so

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

In a case of nested submodules, like swagger-ui inside mgmt inside osv,
only the --recursive update all the submodules, all the way down.

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

Thread pointed to by tlb_flush_waiter may exit before tlb_flush_ipi()
calls wake() on it which will cause access to freed memory. Fix that by
using thread_handle which was introduces to handle exactly those kind of
races.

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

After commit 5c652796, our thread id
is 32 bit, not 64 bit. This patch fixes the return type of the id()
function, and replaces an incorrect comment (which claimed that the
ids are unique during the lifetime of the system) with a more accurate
description on this function, using a doxygen comment (so it can be
seen in our doxygen API documentation).

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

This variable is for keeping value of uint32_t shiftcode[n], so it
should be uint32_t as well.

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>

Fixes #219.

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>

drivers/libtsm includes a bunch of Doxygen comments, and worst of all, a
"@mainpage" tag controlling Doxygen's main page. Since libtsm isn't really
part of the API we intend to advertise, let's exclude this directory from
being scanned by doxygen.

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

Don Martin writes:

  I'm experimenting with Capstan, trying to rule out some things for
  getting a bigger application working. I have a simple program to
  recursively list directory contents, here...

    https://github.com/dmarti/mini-ls



  It works on the host, but when I build the .so, the "capstan build" step
  works, and then when I do "capstan run"...

  $ capstan run
  OSv v0.05-501-gdbdf4f7
  Failed looking up symbol alphasort

This is missing libc functionality in OSv. To fix that, add the
alphasort function from musl.

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

When we discard a bad buffer, we don't advance to the next one, so we'll free
it again on the next pass, resulting in a double-free.

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

The API is listening to port 8000, not 8080.

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

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

Add similar function of the following route command
in Freebsd to OSv.

   $ route add -host $host_ip -interface $interface_name
   e.g.
   $ route add -host 10.240.0.1 -interface eth0

This is useful when the gateway is not reachable within the network
specified by the interface. For instance, GCE's DHCP server provides:

   ip=10.240.10.15, netmask=255.255.255.255, gw=10.240.10.1

A host route entry is needed before we can add the default route entry.

With this patch, the route table looks like:

   Destination         Gateway             Flags  Netif
   default             10.240.0.1          UGS    eth0
   10.240.0.1          42:01:0a:f0:e0:c5   UHS    eth0
   10.240.10.15        link#2              U      eth0
   10.240.10.15        link#2              UHS    lo0
   127.0.0.1           link#1              UH     lo0

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

Fixes #198.

Signed-off-by: Sanidhya Kashyap <sanidhya.k@gmail.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>