On Thu, Feb 22, 2018 at 6:02 AM, geraldo netto
wrote:
> this will avoid people having such situation below:
> ...
> CXX core/commands.cc
> virtual memory exhausted: Cannot allocate memory
>
I'm not comfortable with saying just that "2 GB of memory is needed for the
compilation". What is this n
I'll commit this, as it can't hurt, but note that it won't help either -
curl is installed by default on every Fedora installation:
$ dnf groups info Core
Group: Core
Description: Smallest possible installation
Mandatory Packages:
...
curl
--
Nadav Har'El
n...@scylladb
On Fri, Feb 16, 2018 at 5:19 AM, geraldo netto
wrote:
> Signed-off-by: geraldo netto
> ---
> scripts/manifest_common.py | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/scripts/manifest_common.py b/scripts/manifest_common.py
> index 58b0794..bfe5837 100644
> --- a/
On Sun, Feb 25, 2018 at 8:11 AM, Geraldo Netto
wrote:
> Hello Nadav/Waldek/All,
>
> It turns out most math functions just need a minor tuning
> I'm still facing a lot of issues with wide chars and some stdio things
> But I hope to start submitting the patches to upgrade musl
>
> Also, I was plann
d, instead, a patch which put these files in OSv's libc, and then
modifies the Makefile to take it from there.
Thanks,
Nadav.
--
Nadav Har'El
n...@scylladb.com
On Sat, Feb 24, 2018 at 7:40 PM, geraldo netto
wrote:
> update crypt functions to musl 1.1.18
> this includes encry
uot;not" is already in the
commit message, it's it?
Anyway, changing an already-comitted version is a mess. It's possible, but
I'd rather not do it unless there is a compelling reason to do it.
>
>
> Kind Regards,
> Geraldo Netto
>
>
> Em dom, 25 de fev de
On Sun, Feb 25, 2018 at 8:45 PM, Waldek Kozaczuk
wrote:
> I would like to publish new release of OSv. The last official one was 0.24
> and published on Oct 30, 2015 (a little over 2 years). I have never
> released OSv so any input and suggestions are very welcome especially from
> those that did
On Tue, Feb 27, 2018 at 1:25 AM, Waldek Kozaczuk
wrote:
> Please see my responses below. Also one question I forgot to ask was the
> cadence - how often and when to release. It could be:
>
>1. when enough new features/bugs are added/fixed (this requires more
>planning)
>2. when certai
On Tue, Feb 27, 2018 at 7:04 AM, Waldek Kozaczuk
wrote:
> I am also trying to understand the difference in behavior between java and
> nodejs images. In case of node.js we need to build it from source to make
> it a relocatable ELF but in case of java we use JRE so files as is because
> they are
On Tue, Feb 27, 2018 at 9:19 PM, Waldek Kozaczuk
wrote:
>
> Also the quote by de Saint-Exupery that Nadav posted “Perfection is
> achieved, not when there is nothing more to add, but when there is nothing
> left to take away.” made me think it applies to OSv as well. For example
> in my opinion
Please note that the message you write in the "cover letter" of a patch set
doesn't
get committed into git. In particular, the "fixes #945" message will not be
noticed,
and the issue will not be automatically closed by github. It's no biggie -
we can
always close the issue manually - but it's bette
On Wed, Feb 28, 2018 at 4:00 PM, Waldek Kozaczuk
wrote:
> I am going to review and test this patches. But can you elaborate why we
> need this patch? Is it part of your effort to upgrade musl? Also why does
> new musl add this functionality? And why OSv need this?
>
I think it is solving https:/
On Thu, Mar 1, 2018 at 5:25 AM, Geraldo Netto
wrote:
> Dear Friends,
>
> I know I should be RTFM'ing the code
> But just a quick doubt, do we always use cloud-init
> even for deployments that are not related to EC2?
>
No - cloud-init is a separate module that can be added, or not added, by
scrip
On Thu, Mar 1, 2018 at 5:20 AM, Geraldo Netto
wrote:
> Hello Waldek/Friends,
>
> It's interesting, for some reason, I could further strip the dummy-shlib
> and I couldn't find the stripped version for dummy-shlib
>
dummy-shlib is NOT included in the image, which is why it is not
automatically st
On Wed, Feb 28, 2018 at 1:28 PM, Geraldo Netto
wrote:
> Hello Nadav/All,
>
> Thanks for your feedback
> To be honest, I didn't know that
>
By the way, in another thread I also raised the question, do you really
need to update all the crypt() source files?
If all you wanted to add was encrypt(),
On Mon, Feb 5, 2018 at 7:28 AM, Waldemar Kozaczuk
wrote:
> This patch implements separate syscall call stack needed
> when runtimes like Golang use SYSCALL instruction to make
> system call.
>
> More specifically each application thread pre-alllocates
> "tiny" (1024 bytes deep) syscall call stack
On Mon, Feb 5, 2018 at 7:58 AM, Waldek Kozaczuk
wrote:
> I am expecting this patch will need some tweaking but I think it
> fundamentally implements SYSCALL call stack switch. I have tested in using
> my golang working branch.
>
Excellent.
>
> First of all I am assembly novice so I am not sure
On Wed, Mar 7, 2018 at 6:01 AM, Waldek Kozaczuk
wrote:
> Nadav,
> gic with malloc()?
>
>>
>> Hmm, I just now realized you were using mprotect() to guard one page on
>> the end of the stack (i.e., its beginning) and not the entire thing :-)
>> Makes more sense now...
>> I can't say I understand wh
On Tue, Mar 6, 2018 at 3:46 PM, Waldemar Kozaczuk
wrote:
> This patch adds logic to make sure that waitqueue object actually
> exists in queues map before it tries to dereference it.
>
> The content of this patch was authored by Benoit Canet.
>
Hi, I already reviewed this patch with Benoit, but
dified description. Thanks for bringing it back!
--
Nadav Har'El
n...@scylladb.com
On Tue, Mar 6, 2018 at 3:46 PM, Waldemar Kozaczuk
wrote:
> This patch adds logic to make sure that waitqueue object actually
> exists in queues map before it tries to dereference it.
>
>
you reproduce the bug before this patch? If
you can, maybe I can propose a simpler (and slightly more efficient) patch.
--
Nadav Har'El
n...@scylladb.com
On Wed, Mar 7, 2018 at 7:03 PM, Nadav Har'El wrote:
> Ok, so I gave the futex code a second look.
>
> First of all - I
On Fri, Mar 9, 2018 at 3:29 PM, Geraldo Netto
wrote:
> Dear Friends,
>
> I was doing more tests and for some reason, libz and libedit are not been
> included for groonga
>
So this should have worked (of course the better way is to add the
"requires" in module.py) -
scripts/build image=groonga,l
On Fri, Mar 9, 2018 at 11:15 PM, Geraldo Netto
wrote:
> Dear Friends,
>
> I was trying to reproduce the same issue on fedora 27
> So, I installed fedora 27 from scratch with all standard configuration
> (eg: next-next-finish)
> and I end up finding something else...
> It seems, for some reason, d
On Thu, Mar 15, 2018 at 5:24 AM, Waldek Kozaczuk
wrote:
> This is related to this series of patches by Benoit and the discussion
> mostly focused in the patch titled "Delay elf object initialization" -
> https://groups.google.com/forum/#!searchin/osv-dev/Delay$20elf$20object$
> 20initialization%7
On Sat, Mar 17, 2018 at 12:00 AM, Waldek Kozaczuk
wrote:
> Nadav,
> Thanks for the info.
>
> I have been studying the Drepper paper as well as OSv code (core/sched.*,
> core/elf.*, arch/x64/arch-*.*) and I must stay I am still far from
> understanding all bits and how exactly they play together -
On Sun, Mar 18, 2018 at 6:31 AM, Waldek Kozaczuk
wrote:
> So I have been thinking about it more and trying to answer my own
> question: "when accessing thread_local variables, what condition is used
> to determine whether to access *by offset* from the static TLS memory
> area that was allocated
On Sun, Mar 18, 2018 at 7:24 PM, Waldek Kozaczuk
wrote:
>
> Anyway you asked about the bug related to when the object becomes visible
> after/during initialization. I think the Benoit's branch (go4), I copied
> all this code from, has this corrected
>
Yes, I looked and it seems it does.
> and
On Thu, Mar 22, 2018 at 7:10 AM, Waldemar Kozaczuk
wrote:
> All these 4 patches are identical as in the set from March 2. However
> they are in different order with the 4-th one being most controversial.
>
Thanks! I pushed the first two patches (those actually adding rofs), and
will look at the
On Thu, Mar 22, 2018 at 11:44 AM, Nadav Har'El wrote:
>
> On Thu, Mar 22, 2018 at 7:10 AM, Waldemar Kozaczuk
> wrote:
>
>> All these 4 patches are identical as in the set from March 2. However
>> they are in different order with the 4-th one being most controversial
Looks good, I just have a few suggestions below.
On Thu, Mar 22, 2018 at 7:10 AM, Waldemar Kozaczuk
wrote:
> This patch enhances ramfs implementation to handle time and
> mode node attributes. Additionally it adds extra checks
> in relevant places to enforce maximum name length.
>
> Signed-off-b
On Fri, Mar 23, 2018 at 3:38 PM, Waldek Kozaczuk
wrote:
> I managed to build and run simple Rust hello example on OSv.
>
> Initially OSv complained about missing __res_init symbol. Based on what I
> read it is some old libc function kept for backwards compatibility and it
> is not clear why rust
On Fri, Mar 23, 2018 at 3:21 PM, Waldek Kozaczuk
wrote:
> I am trying to avoid this code ("GoMain" part) in application constructor:
>
> _main = _lib->lookup("main");
> *if (!_main) {*
> *_entry_point = _lib->lookup("GoMain");*
> *}*
> if (!_main && !_entry_point) {
>
On Fri, Apr 13, 2018 at 7:36 AM, Commit Bot
wrote:
> From: Wang Yu
> Committer: Waldemar Kozaczuk
> Branch: master
>
> add boot_time event of uncompress lzloader.elf
>
> sometimes i found uncompres cost serival times,
> so add boot_time event of uncompress lzloader.elf
>
Excellent. The importa
On Tue, Apr 10, 2018 at 10:29 PM, Waldek Kozaczuk
wrote:
> Last week I have been trying to hack OSv to run on hyperkit and finally I
> managed to execute native hello world example with ROFS.
>
Excellent :-)
>
> Here is a timing on hyperkit/OSX (the bootchart does not work on hyperkit
> due to
On Sun, Apr 8, 2018 at 7:02 AM, Waldemar Kozaczuk
wrote:
> This patch adds new golang module in order to bootstrap Golang
> applications on OSv. Its role is to pass golang shared object
> specific main function name - GoMain - and terminate all lingering
> Golang application threads once main thr
On Sun, Apr 15, 2018 at 8:12 PM, Waldek Kozaczuk
wrote:
> Is it host setup dependent? I just pulled latest master and reran the
> tests and all of them passed. Is it because my host has different version
> of the libraries that my laptop has?
>
I saw the same problem on my build machine a few da
not familiar with rust, but is the file "Cargo.lock" a file you
wrote, or some sort of temporary file which shouldn't have been committed?
--
Nadav Har'El
n...@scylladb.com
On Sun, Apr 22, 2018 at 4:37 PM, Commit Bot
wrote:
> From: Waldemar Kozaczuk
> Committer: Wal
On Fri, Apr 20, 2018 at 4:07 AM, Waldek Kozaczuk
wrote:
> To make SMP working I had to hack OSv to pass fee00900 when
> enabling APIC for first CPU and fee00800 for all other CPUs. It
> looks like (based on source code of hyperkit) it requires that the APIC
> registers memory area
osed lock exact version of
> dependencies - https://doc.rust-lang.org/cargo/guide/cargo-toml-vs-
> cargo-lock.html. Kind of like package-lock.json in Node.JS land.
>
> Waldek
>
> On Sunday, April 22, 2018 at 4:22:27 PM UTC-4, Nadav Har'El wrote:
>>
>> Hi, I think some f
On Mon, Apr 23, 2018 at 7:44 AM, wrote:
> Hi All.
>
> Our purpose is to run python-applications as unikernels on xen on arm.
> Till now, we have evaluated rumprun and unikraft. However, rumprun does not
> have support for arm on xen, while unikraft is still in very nascent stages.
>
> Does OSv ha
On Mon, Apr 23, 2018 at 10:52 AM, wang Yu wrote:
> 1、scripts/build image=mysql
> 2、scripts/run.py -nv to start mysql server on osv
> OSv v0.24-511-ge60339d eth0: 192.168.122.76
> 3、on host i use mysqlslap to test the benchmark
> mysqlslap -h 192.168.122.76 -P 3306 --concurrency=100--iteratio
On Thu, Apr 19, 2018 at 2:50 PM, Waldek Kozaczuk
wrote:
> Nadav,
>
> Thanks for reviewing the patch. Please see my responses below.
>
> On Sunday, April 15, 2018 at 12:36:31 PM UTC-4, Nadav Har'El wrote:
>>
>> I'm not sure I understood why post_main() is n
Thanks. I'm committing this, but it would be preferable if the test also
prints a failure message (as in the other tests) but I think it should work
this way too (the test will print a message about exiting with a non-zero
exit code).
--
Nadav Har'El
n...@scylladb.com
On Fri, Apr 6,
On Thu, Apr 5, 2018 at 12:40 AM, Waldek Kozaczuk
wrote:
> Which version of Java is being used by build process? I have a feeling it
> is higher than java 8.
>
If require Java 7 or 8 for the build, I wonder if we shouldn't make an
effort to find the appropriate
executable on the system - or maybe
On Mon, Apr 23, 2018 at 7:36 PM, Nadav Har'El wrote:
>
> On Thu, Apr 5, 2018 at 12:40 AM, Waldek Kozaczuk
> wrote:
>
>> Which version of Java is being used by build process? I have a feeling it
>> is higher than java 8.
>>
>
> If require Java 7 or 8 for
On Fri, Apr 27, 2018 at 5:41 PM, Waldek Kozaczuk
wrote:
> This time tst-unordered-ring-mpsc failed (line 60).
>
What do you mean "this time"? In the message you quote below, the error is,
"as usual", in tst-tcp-nbwrite.so.
Or do you mean that you ran tests on your own machine, and saw a failure
On Fri, Apr 27, 2018 at 6:19 PM, Waldek Kozaczuk
wrote:
> My theory might be right.
>
> I ran simple experiment by adding this before first assert
> in test_when_ring_gets_full test:
> sched::thread::pin(sched::cpus[0]);
>
> and then before assert(!ring.push(5));
> sched::thread::pin(sched::cpus[
On Thu, Apr 26, 2018 at 8:43 AM, Waldemar Kozaczuk
wrote:
> This patch fixes an intermittent issue happening in TCP stack code
> during session 3-way handshake protocol due to inconsistent state
> of transmission control block (TCB) data structure. Specifically
> right after server sends SYN+ACK
On Mon, Apr 30, 2018 at 1:56 AM, Waldek Kozaczuk
wrote:
> See my comments below.
>
> On Sunday, April 29, 2018 at 6:41:01 PM UTC-4, Nadav Har'El wrote:
>
>>
>> On Thu, Apr 26, 2018 at 8:43 AM, Waldemar Kozaczuk
>> wrote:
>>
>>> This patch fixes
> all these missing problems would go away.
>
>
> But again there may be value in making OSv work fine with Fedora distribution
> of open jdk8.
>
>
Yes, indeed, that's what I think too.
>
> Waldek
>
>
> On Wed, May 2, 2018 at 11:39 AM, wrote:
>
>
Ok, so after the barrage of patches I committed today, OSv finally builds
and works, mostly, on Fedora 28.
Most of the things I tested appear to work now.
But "make check" still fails on the weird "tracing_smoke_test" - see issue
964, maybe because of some gdb interaction.
On Sat, May 5, 2018 at 7:48 PM, Waldek Kozaczuk
wrote:
> What if there was a way early into the booting process to load specific
> features of OSv, that right now are linked in as part of kernel, as
> libraries? For example if we mounted ROFS as early and possible we could
> load other elements o
On Sun, May 6, 2018 at 7:52 AM, Dor Laor wrote:
> On Sat, May 5, 2018 at 9:48 AM, Waldek Kozaczuk
> wrote:
>
>> What if there was a way early into the booting process to load specific
>> features of OSv, that right now are linked in as part of kernel, as
>> libraries? For example if we mounted R
object, it is treated as bind-now.
Fixes #967.
Signed-off-by: Nadav Har'El
---
include/osv/elf.hh | 9 +
core/elf.cc| 4 +++-
2 files changed, 12 insertions(+), 1 deletion(-)
diff --git a/include/osv/elf.hh b/include/osv/elf.hh
index e0966c40..304021af 100644
--- a/include/o
On Wed, May 9, 2018 at 7:55 PM, Geraldo Netto
wrote:
> Dear Friends,
>
> Just a quick thing, I was syncing my local OSv repos
> and I noticed something with osv-apps:
>
> netto@besta01 ~/Desktop/osv/osv-clean $ git pull
> remote: Counting objects: 7, done.
> remote: Compressing objects: 100% (1/1
On Sun, May 13, 2018 at 7:02 AM, Commit Bot
wrote:
> From: Waldemar Kozaczuk
> Committer: Waldemar Kozaczuk
> Branch: master
>
> Gracefully handle the case of missing libvsdo.so
>
> Current implementation of application::prepare_argv() aborts execution
> if it fails to load libvdso library. So
Hi, thanks for being so insistent on this important patch.
I really wanted to commit this version, but unfortunately after another
reading,
I'm really worried if what you're doing with rbx is safe. Please see my
comments below.
Sorry about dragging this on for so long :-(
On Fri, Mar 16, 2018 a
ain a ZFS build, even
though the only thing changed is the command line?
If this doesn't work, then Jenkins won't work properly (I believe it
doesn't bother with a "make clean" every time).
>
> Waldek
>
> On Sunday, May 13, 2018 at 5:44:29 PM UTC-4, Bot Droid
Thanks, and sorry for all the round-trips... I have one last (I hope) issue
below (two comments, actually):
--
Nadav Har'El
n...@scylladb.com
On Tue, May 15, 2018 at 8:01 AM, Waldemar Kozaczuk
wrote:
> This patch implements separate syscall call stack needed
> when runtimes like
--
Nadav Har'El
n...@scylladb.com
On Tue, May 15, 2018 at 2:59 PM, Waldemar Kozaczuk
wrote:
> This patch implements separate syscall call stack needed
> when runtimes like Golang use SYSCALL instruction to execute
> system calls. More specifically in case of Golang tiny st
On Tue, May 15, 2018 at 10:20 PM, Waldemar Kozaczuk
wrote:
> This patch implements separate syscall call stack needed
> when runtimes like Golang use SYSCALL instruction to execute
> system calls. More specifically in case of Golang tiny stacks
> used by coroutines are not deep enough to execute
fore this patch, adding an extra bit not listed, like EPOLLRDHUP,
caused POLLNVAL to be returned. Worse, when using epoll, it did not
expect to ever get back POLLNVAL from poll(), and caused an assertion
failure.
Refs #971.
Signed-off-by: Nadav Har'El
---
core/poll.cc | 11 +--
1 file
nfuse poll() inplementations, and in particular poll_no_poll()
which caused an assertion failure.
Refs #971.
Signed-off-by: Nadav Har'El
---
core/epoll.cc | 9 -
1 file changed, 4 insertions(+), 5 deletions(-)
diff --git a/core/epoll.cc b/core/epoll.cc
index f425c4ad..422a8170 100644
appened in #971).
The test can now be run on both Linux and OSv, and succeds on both.
Signed-off-by: Nadav Har'El
---
tests/tst-epoll.cc | 40
1 file changed, 40 insertions(+)
diff --git a/tests/tst-epoll.cc b/tests/tst-epoll.cc
index 6a489ccc..826589
edb837c05be for more information). So although
the EPOLLET bit is *not* traditionally supported by the poll() method, we
in OSv we do need to pass it on to poll(). Other bits, namely EPOLLONESHOT,
we still don't need to pass.
Fixes #974.
Signed-off-by: Nadav Har'El
---
core/epoll.cc |
On Sat, Aug 4, 2018 at 6:03 AM, Charles Myers
wrote:
> Used in FreeBSD IPv6 code sys/netinet6/frag6.c IP6Q_TRYLOCK().
>
Thanks, I'll commit these two trivial patches as-is, but the next patches,
can you please add a "Signed-off-by:" line to your patch?
This is explained in the CONTRIBUTING file.
On Sat, Aug 4, 2018 at 6:00 AM, Charles Myers
wrote:
> It took me a while to split out the patches according to your guidelines,
> since they were not organized that way.
> I will be submitting them probably starting tomorrow or maybe early next
> week. The largest patch will be around 1 MB for
On Fri, Aug 3, 2018 at 9:35 PM, Waldek Kozaczuk
wrote:
> There is an outstanding issue https://github.com/
> cloudius-systems/osv/issues/979 that describes how OSv creates extra
> copies of file data when mmap() is called for RAMFS or ROFS node. Possible
> improvement to eliminate this extra copy
On Wed, Aug 1, 2018 at 7:48 AM, Waldemar Kozaczuk
wrote:
> This patch optimizes RAM utilization of bootfs by
> eliminating unnecessary copy of data. It does so by
> pointing created file nodes to existing data offset in memory
> which is part of area where kernel is copied after decompression.
>
On Tue, Jul 31, 2018 at 9:08 PM, Waldek Kozaczuk
wrote:
> When possible, it's a worthy goal to have a kernel and usr.img which can
>> work for all platforms (xen, qemu, qemu-with-kernel, etc.) and we don't
>> need
>> to recompile for each different cloud platform. I didn't fully understand
>> if
On Mon, Aug 6, 2018 at 6:17 PM, Nadav Har'El wrote:
>
> Oh, I found the problem. It is OSv's bug, on the way we define uint64_t
> (we define it ourselves, we don't use gcc's headers):
> In:
>
> include/api/x64/bits/alltypes.h.sh:TYPEDEF unsigned long ui
Makes sense :-) How did this function ever previously work? I guess it
didn't, because we never even implemented this ioctl, I think.
I guess we don't have a unit test for it?
--
Nadav Har'El
n...@scylladb.com
On Tue, Aug 7, 2018 at 5:49 AM, Charles Myers
wrote:
> Use
Looks good in general. Some trivial questions inline below.
On Tue, Aug 7, 2018 at 5:49 AM, Charles Myers
wrote:
> Signed-off-by: Charles Myers
> ---
> bsd/sys/netinet/icmp6.h | 2 +
> bsd/sys/netinet/ip6.h | 93 +--
> bsd/sys/netinet6/in6.h| 125 +
The patch looks reasonable to me, but there are some things that I didn't
understand how they fit in.
For example, how do the changes bsd/sys/netinet/tcp_reass.cc related to
some function for iterating link-level sockets?
--
Nadav Har'El
n...@scylladb.com
On Tue, Aug 7, 2018 at 5:49 A
On Tue, Aug 7, 2018 at 5:49 AM, Charles Myers
wrote:
> Signed-off-by: Charles Myers
> ---
> bsd/porting/networking.cc | 178 ++
> +++-
> bsd/porting/networking.hh | 8 +++
> bsd/porting/route.cc | 86 ++
> bsd/sys/netinet/in.cc
h uses
these netlink sockets directly?
Some more minor questions inline below.
--
Nadav Har'El
n...@scylladb.com
On Tue, Aug 7, 2018 at 5:49 AM, Charles Myers
wrote:
> Signed-off-by: Charles Myers
> ---
> Makefile | 1 +
> bsd/net.cc
On Tue, Aug 7, 2018 at 5:49 AM, Charles Myers
wrote:
> libc code is from musl library
Which version? It's not the version we have in musl/, right? So which
version?
> with minor change to prevent partial reads if faster than sender.
>
Is this fixing a bug in musl (in which case you should pr
On Tue, Aug 7, 2018 at 5:49 AM, Charles Myers
wrote:
> Signed-off-by: Charles Myers
> ---
> bsd/sys/compat/linux/linux_ioctl.cc | 37 ++
> +++
> 1 file changed, 33 insertions(+), 4 deletions(-)
>
> diff --git a/bsd/sys/compat/linux/linux_ioctl.cc
> b/bsd/sys/comp
All looks good. Thanks.
--
Nadav Har'El
n...@scylladb.com
On Tue, Aug 7, 2018 at 5:49 AM, Charles Myers
wrote:
> https://cloudinit.readthedocs.io/en/latest/topics/network-
> config-format-v1.html
>
> Currently only interface naming, static IP, routes and DNS are supported.
Thanks. I am very happy you added these tests :-)
Only some minor comments below.
On Tue, Aug 7, 2018 at 5:49 AM, Charles Myers
wrote:
> Signed-off-by: Charles Myers
> ---
> modules/tests/Makefile| 16 ++-
> tests/tst-pktinfo.cc | 245 ++
> ++
On Tue, Aug 7, 2018 at 5:49 AM, Charles Myers
wrote:
> Signed-off-by: Charles Myers
> ---
> Makefile | 27 ++-
> bsd/net.cc | 23 +++-
> bsd/porting/netport.h| 4 +++
> bsd/sys/dev/xen/netfront/ne
On Tue, Aug 7, 2018 at 5:49 AM, Charles Myers
wrote:
> This patch series adds IPv6 support to OSv.
>
> The first 5 parts of the series should have no impact on existing
> OSv code. The rest of the patches starting from the net channel
> changes do impact existing code.
>
Thanks. Very impressive
On Tue, Aug 14, 2018 at 12:34 AM, Waldek Kozaczuk
wrote:
> I managed to get Python 3.5 working on OSv (I only tested simple 'Hello
> world!' and httpserver example). However to get it working I had to add
> missing __wcscpy_chk to libc.
>
> According to http://refspecs.linux-foundation.org/LSB_4.
On Tue, Aug 14, 2018 at 7:16 AM, Waldek Kozaczuk
wrote:
> I managed to get Python 3.5 working on OSv but reporting missing 318
> syscall. With version 3.6 and above Python is trying to invoke getrandom
> which obviously fails due to missing symbol.
>
Yes, both are getrandom() and it is indeed mi
On Tue, Aug 14, 2018 at 6:42 AM, Waldemar Kozaczuk
wrote:
> This patch renames python27 to python2x and improves it
> by discovering Python home directory by reading sys.* variables.
>
I committed this, but it doesn't seem to actually work for me, on Fedora 28:
$ scripts/run.py -e /python
OSv v
ifru_addr;
>
> @@ -144,7 +142,6 @@ struct l_ifreq {
>
> } ifr_ifru;
>
> } __packed;
>
>
>
> -#defineifr_nameifr_ifrn.ifrn_name /* Interface name
> */
>
> #defineifr_hwaddr ifr_ifru.ifru_hwaddr/* MAC address
> Sent from my iPhone
>
> On Aug 14, 2018, at 03:31, Nadav Har'El wrote:
>
>
> On Tue, Aug 14, 2018 at 6:42 AM, Waldemar Kozaczuk
> wrote:
>
>> This patch renames python27 to python2x and improves it
>> by discovering Python home directory by reading sys.*
ter this patch, I can build the image on Fedora 28 but it fails on
missing mkfifoat:
/usr/lib/libpython3.6m.so.1.0: failed looking up symbol mkfifoat
We are indeed missing this function.
--
Nadav Har'El
n...@scylladb.com
On Tue, Aug 14, 2018 at 2:47 PM, Waldemar Kozaczuk
wrote:
&g
On Tue, Aug 14, 2018 at 3:01 PM, Nadav Har'El wrote:
> I committed this, but it doesn't compile on Fedora 28... On Fedora, you
> can't compile with -lpython3.6m, you need just "-lpython3".
>
Turns out that the missing symbolic links are added when you install
August 14, 2018 at 8:15:38 AM UTC-4, Bot Droid wrote:
>>
>> From: Nadav Har'El
>> Committer: Nadav Har'El
>> Branch: master
>>
>> Stub implementation of mkfifoat()
>>
>> Implement a stub mkfifoat() which just prints a message about not
On Wed, Aug 15, 2018 at 5:49 AM, Waldemar Kozaczuk
wrote:
> Fixes #989
>
> Signed-off-by: Waldemar Kozaczuk
> ---
> Makefile | 1 +
> include/api/sys/random.h | 1 +
> include/api/x64/bits/syscall.h | 1 +
> libc/include/sys/random.h | 13 ++
> libc/r
;>> Python 3.6.6 (default, Jun 28 2018, 04:42:43)
>>> [GCC 5.4.0 20160609] on linux
>>> Type "help", "copyright", "credits" or "license" for more information.
>>> >>>
>>>
>>> On my regular machi
On Wed, Aug 15, 2018 at 3:19 PM, Nadav Har'El wrote:
>
> On Wed, Aug 15, 2018 at 2:18 PM, Waldek Kozaczuk
> wrote:
>
>> I think added random.h header in wrong placed messed things up. I forgot
>> to do full rebuild.
>>
>
> By __NR_getrandom comes from s
On Wed, Aug 15, 2018 at 3:24 PM, Nadav Har'El wrote:
>
> On Wed, Aug 15, 2018 at 3:19 PM, Nadav Har'El wrote:
>
>>
>> On Wed, Aug 15, 2018 at 2:18 PM, Waldek Kozaczuk
>> wrote:
>>
>>> I think added random.h header in wrong placed messed thin
On Wed, Aug 15, 2018 at 4:10 PM, Nadav Har'El wrote:
>
> On Wed, Aug 15, 2018 at 3:24 PM, Nadav Har'El wrote:
>
>>
>> On Wed, Aug 15, 2018 at 3:19 PM, Nadav Har'El wrote:
>>
>>>
>>> On Wed, Aug 15, 2018 at 2:18 PM, Waldek Kozaczuk
On Wed, Aug 15, 2018 at 9:41 PM, Waldek Kozaczuk
wrote:
> I hacked elf.cc to print missing symbols instead of aborting and here is a
> list I got:
>
Beyond providing us with a valuable list of functions we should,
eventually, implement (would be nice to enter them into a bug tracker issue
or iss
}
> if (!ret.symbol) {
> -abort("%s: failed looking up symbol %s\n",
> +debugf("%s: failed looking up symbol %s\n",
> pathname().c_str(), demangle(name).c_str());
> +return symbol_module(sym, this);
> }
>
On Fri, Aug 24, 2018 at 1:53 AM, Waldemar Kozaczuk
wrote:
> This patch optimizes RAM utilization of bootfs by
> eliminating unnecessary copy of data. It does so by
> pointing created file nodes to existing data offset in memory
> which is part of area where kernel is copied after decompression.
>
This is the old bug https://github.com/cloudius-systems/osv/issues/382
raising it's ugly head again. I updated the issue.
--
Nadav Har'El
n...@scylladb.com
On Fri, Aug 24, 2018 at 1:07 AM, wrote:
> See <http://jenkins.cloudius-systems.com:8080/job/osv-
> build-nightly/15
ed these libraries.
After all, OSv does *not* really have systemd. I wonder if we just delete
(blacklist) /lib/python3.6/site-packages/systemd/ or something, may python
will not look for it in the first place.
--
Nadav Har'El
n...@scylladb.com
On Fri, Aug 17, 2018 at 7:01 AM, Waldek Kozaczuk
601 - 700 of 1373 matches
Mail list logo
