Recent changes to move ifconfig.so out of the main OSv makefile broke the
"mgmt" module build. This patch makes "scripts/build image=mgmt" work again.
Signed-off-by: Nadav Har'El <n...@scylladb.com>
---
module.py| 1 +
usr.manifest | 4 ++--
2 files changed, 3
On Sun, Jul 24, 2016 at 1:28 PM, Benoît Canet
wrote:
>
> But, Benoit, is it possible that ZFS *is* trying to coalesce writes, just
>> somehow misconfigured to aim at the wrong size? What I mean is, if you do
>> write() operations of 10 bytes, will you see 10-byte
On Tue, Aug 2, 2016 at 12:51 PM, wrote:
> Hi Team,
> I'm trying to build an OSv application. I have successfully build the same
> using capstan. But when I ran the application it prints the following logs
> and exits.
>
> ~/bin/capstan run
> Created instance: test
>
migration
lock - so let's start using it.
Refs #776.
Signed-off-by: Nadav Har'El <n...@scylladb.com>
---
include/osv/sched.hh | 1 +
libc/pthread.cc | 13 +
2 files changed, 6 insertions(+), 8 deletions(-)
diff --git a/include/osv/sched.hh b/include/osv/sched.hh
index 2
On Mon, Jul 4, 2016 at 10:27 AM, Justin Cinkelj
wrote:
> Signed-off-by: Justin Cinkelj
> ---
> libc/pthread.cc | 53 -
> 1 file changed, 28 insertions(+), 25 deletions(-)
>
> diff --git
this code :-)
This patch also adds a test for this bug, which passes on Linux, failed
on OSv before this patch, and now passes on OSv too.
Fixes #775
Signed-off-by: Nadav Har'El <n...@scylladb.com>
---
bsd/sys/kern/uipc_syscalls_wrap.cc | 8 +++
tests/tst-uio.cc
Signed-off-by: Nadav Har'El <n...@scylladb.com>
---
core/mempool.cc | 1 -
1 file changed, 1 deletion(-)
diff --git a/core/mempool.cc b/core/mempool.cc
index 0fbd857..50c938f 100644
--- a/core/mempool.cc
+++ b/core/mempool.cc
@@ -11,7 +11,6 @@
#include
#include
#include
-#i
enum, which do have conversion to
int. Put this enum inside a namespace so we can still use names like
init_prio::sched and don't need to change any of the callers.
Fixes #769
Signed-off-by: Nadav Har'El <n...@scylladb.com>
---
include/osv/prio.hh | 4 +++-
1 file changed, 3 insertions
Needed on gcc 6 for one of the files
Signed-off-by: Nadav Har'El <n...@scylladb.com>
---
Makefile | 2 ++
1 file changed, 2 insertions(+)
diff --git a/Makefile b/Makefile
index 63427bf..530e9a5 100644
--- a/Makefile
+++ b/Makefile
@@ -502,6 +502,8 @@ $(out)/bsd/sys/netinet/in_rmx.o:
be enough for applications
which want to use libquadmath (for example) for calculations but do not
intend to use its printf extensions.
Refs issue #765 (this issue should not be closed until we have a full
implementation of these functions, or make a decision that we never will).
Signed-off-by: Nadav
Allow running export_runtime() twice without ill-effect.
A following patch will allow migrating sleeping threads, and migrating it again
before it got to run is possible.
Signed-off-by: Nadav Har'El <n...@scylladb.com>
---
core/sched.cc | 6 --
1 file changed, 4 insertions(+), 2 del
with a wake() of the same thread, resulting in the thread remaining on
the old cpu's wakeup queues, and we better ignore it there.
Signed-off-by: Nadav Har'El <n...@scylladb.com>
---
core/sched.cc | 3 +++
1 file changed, 3 insertions(+)
diff --git a/core/sched.cc b/core/sched.cc
index 7
On Tue, Jul 5, 2016 at 9:56 AM, Justin Cinkelj
wrote:
> Signed-off-by: Justin Cinkelj
> ---
> Makefile | 1 +
> core/osv_c_wrappers.cc | 33 +
> include/osv/osv_c_wrappers.h | 27
glibc functions in
OSv - they all are supposed to have C linkage, but the the "extern C" is
almost always in the header file,
not the .c file. Maybe the prototype is not exactly identical? I'll take a
look at your patch.
> BR Justin
>
> On 07/04/2016 06:54 PM, Nadav Har'El wrote:
Thanks.
Nitpick (but I'll commit anyway as-is): I think you could have used a const
reference to the thread th1 instead of a regular reference.
--
Nadav Har'El
n...@scylladb.com
On Mon, Jul 4, 2016 at 7:26 PM, Justin Cinkelj <justin.cink...@xlab.si>
wrote:
> Application is identifi
Hi, thanks. Looks mostly good but some comments below.
On Mon, Jul 4, 2016 at 7:26 PM, Justin Cinkelj
wrote:
> Signed-off-by: Justin Cinkelj
> ---
> Makefile | 1 +
> core/osv_c_wrappers.cc | 39
FYI, these tests have been failing for a long time on Jenkins, it's just
that a "scripts/build" bug hid this failure (see
https://github.com/cloudius-systems/osv/issues/779)
On Sun, Aug 7, 2016 at 10:46 PM, wrote:
> + exec ./scripts/test.py
> TEST java
Makefile:372: recipe for target 'build/release.x64/loader.o' failed
> make: *** [build/release.x64/loader.o] Error 1
> [benoit@scylla osv]$ ^C
> [benoit@scylla osv]$
>
>
> 2016-07-22 20:32 GMT+02:00 Nadav Har'El <n...@scylladb.com>:
>
>> Starting with gcc 6, som
On Thu, Jun 30, 2016 at 8:57 PM, Justin Cinkelj
wrote:
> Signed-off-by: Justin Cinkelj
> ---
> core/sched.cc| 45 +
> include/osv/sched.hh | 1 +
> 2 files changed, 46 insertions(+)
>
> diff
On Thu, Feb 2, 2017 at 12:12 AM, Nadav Har'El <n...@scylladb.com> wrote:
> +sched::with_all_threads([&](sched::thread ) {
>
>> +if (t.id() == tid && t.get_status() !=
>> sched::thread::status::terminated) {
>
On Tue, Jan 31, 2017 at 11:36 PM, Justin Cinkelj
wrote:
> Fixes #845.
>
> Signed-off-by: Justin Cinkelj
> ---
> modules/httpserver/api-doc/listings/app.json | 26
> ++
> modules/httpserver/api/app.cc| 18
On Wed, Feb 1, 2017 at 7:02 PM, Justin Cinkelj
wrote:
>
> >
> > Amnon, if we're changing the return value of the app/command REST API in
> > OSv from always returning a null string to returning a string with
> numbers
> > in it, are we supposed to modify some API
::with_thread_by_id() which runs
a given function on the thread with the given id - while holding a mutex
which prevents threads (including this one) from exiting. Because it holds
this mutex, the function should typically be very short - e.g., test
something about this thread.
Signed-off-by: Nadav
On Mon, Feb 6, 2017 at 3:45 PM, 'Sergiy Kibrik' via OSv Development <
osv-dev@googlegroups.com> wrote:
> Move x64-specific implementation out of common event channel code
> to platform-specific header, where other asm snippets reside.
> This will allow aarch64 to reuse common code.
>
I think we
After this patch, I can't compile on aarch64:
$ make arch=aarch64
...
bsd/sys/xen/evtchn.cc:33:32: fatal error: machine/xen/xen-os.h: No such
file or directory
#include
--
Nadav Har'El
n...@scylladb.com
On Mon, Feb 6, 2017 at 3:45 PM, 'Sergiy Kibrik' via OSv Development <
osv-
On Thu, Feb 2, 2017 at 5:19 PM, Justin Cinkelj
wrote:
>
>
>> I think we can do something similar (at least to 1-3). Instead of my
>> previous patch, do something like this:
>>
>> 1. httpserver uses osv::run_background() to run the background operation,
>> and receives
On Mon, Feb 6, 2017 at 2:35 AM, 'Quentin Barnes' via OSv Development <
osv-dev@googlegroups.com> wrote:
> I ran acrossed a problem when building the jetty demo on Linux (RHEL
> 7.3). I'm not sure why no one else has hit this problem. That is,
> if no one has built this demo on Linux or if
On Sun, Feb 5, 2017 at 7:42 PM, Vasily A. Sartakov
wrote:
> Greetings
>
> I need a tool for tracing call stack. Something very simple: on demand, in
> some place, I want to obtain list of previous calls in form of addressed,
> maybe some more data,
You want to do a
On Mon, Jan 23, 2017 at 11:46 AM, Justin Cinkelj
wrote:
> Code to remove \0 at end of string tried to copy strlen()-1 data bytes.
> It didn't expect the .note.osv-mlock section to contains only 0x00.
> This caused attempt to allocate and copy -1 bytes, converted to about
On Mon, Jan 23, 2017 at 12:51 PM, 'Sergiy Kibrik' via OSv Development <
osv-dev@googlegroups.com> wrote:
> hi Roman,
>
> On 01/20/2017 06:57 PM, Roman Shaposhnik wrote:
>
>> this is super cool -- I've been very interested in seeing resurrected
>> support for OSv on Xen/ARM
>> for a different
Thanks. Comments inline below:
--
Nadav Har'El
n...@scylladb.com
On Tue, Jan 24, 2017 at 12:22 AM, Justin Cinkelj <justin.cink...@xlab.si>
wrote:
> MAC address of first "real" ethernet interface is used as uniqeu
>
typo: unique
> host identifier.
>
> Fixes
r changing application::start() massively.
Fixes #843.
Signed-off-by: Nadav Har'El <n...@scylladb.com>
---
include/osv/app.hh | 4
core/app.cc| 22 ++
2 files changed, 26 insertions(+)
diff --git a/include/osv/app.hh b/include/osv/app.hh
index 3237ebc
to osv::run_background().
Signed-off-by: Nadav Har'El <n...@scylladb.com>
---
include/osv/app.hh| 2 +-
core/app.cc | 2 +-
modules/httpserver/api/app.cc | 2 +-
3 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/include/osv/app.hh b/include/osv/app.hh
e inclusion of
from . This was added in 741b5563f9045ed07c387bbad63f6c8a9ee75209
and is not actually needed (the caller can include on its own,
it doesn't need to do it - see syscall(2)).
Signed-off-by: Nadav Har'El <n...@scylladb.com>
---
include/api/syscall.h | 1 -
include/osv/file.h
On Mon, Jan 30, 2017 at 4:19 PM, Justin Cinkelj
wrote:
>
>
> Maybe there is one significant difference between osv_execve() befor/after
> Nadav's fixed it, and the code above.
> osv_execve() tried to set new thread id in the code executed inside that
> new thread. So new
On Fri, Jan 20, 2017 at 1:04 PM, 'Sergiy Kibrik' via OSv Development <
osv-dev@googlegroups.com> wrote:
> Add Linux-resembling header at the begining of image,
> so that it can be recognized and loaded by Xen ARM64 image loader,
> and probably some other loaders too.
>
Thanks. Looks good. I have
Reviewed-by: Nadav Har'El <n...@scylladb.com>
--
Nadav Har'El
n...@scylladb.com
On Fri, Jan 20, 2017 at 1:04 PM, 'Sergiy Kibrik' via OSv Development <
osv-dev@googlegroups.com> wrote:
> Start device range from 0x300 so that aarch64 Xen guest can map gic.
> This seems to
Reviewed-by: Nadav Har'El <n...@scylladb.com>
--
Nadav Har'El
n...@scylladb.com
On Fri, Jan 20, 2017 at 1:04 PM, 'Sergiy Kibrik' via OSv Development <
osv-dev@googlegroups.com> wrote:
> Signed-off-by: Sergiy Kibrik <sergiy.kib...@globallogic.com>
> ---
> bsd/aarc
Reviewed-by: Nadav Har'El <n...@scylladb.com>
We could have also changed the interface of HYPERVISOR_console_io, but I
guess you preferred not to modify that unmodified BSD header?
--
Nadav Har'El
n...@scylladb.com
On Fri, Jan 20, 2017 at 1:04 PM, 'Sergiy Kibrik' via OSv Development <
On Fri, Jan 20, 2017 at 1:04 PM, 'Sergiy Kibrik' via OSv Development <
osv-dev@googlegroups.com> wrote:
> hi OSv developers,
> Here let me present you with some little work to allow OSv kernel to be
> loaded
> under Xen hypervisor on AARCH64 platform.
>
> This is basic stuff, which allows to
Reviewed-by: Nadav Har'El <n...@scylladb.com>
Just a typo in the commit message - that's __ASSEMBLY__, not __ASSEMBLER__.
--
Nadav Har'El
n...@scylladb.com
On Fri, Jan 20, 2017 at 1:04 PM, 'Sergiy Kibrik' via OSv Development <
osv-dev@googlegroups.com> wrote:
> __ASSEMBLER__ def
Reviewed-by: Nadav Har'El <n...@scylladb.com>
--
Nadav Har'El
n...@scylladb.com
On Fri, Jan 20, 2017 at 1:04 PM, 'Sergiy Kibrik' via OSv Development <
osv-dev@googlegroups.com> wrote:
> Signed-off-by: Sergiy Kibrik <sergiy.kib...@globallogic.com>
> ---
> ar
g., in x64, for the "early console" we
use console::isa_serial_console::early_init()).
--
Nadav Har'El
n...@scylladb.com
On Fri, Jan 20, 2017 at 1:04 PM, 'Sergiy Kibrik' via OSv Development <
osv-dev@googlegroups.com> wrote:
> Primitive driver that uses HYPERVISOR_console_io hyperca
Hi,
Here I have doubts on whether this is the right approach -
If we have the pl011_console class and the xen_console class, why can't
early_console simply being set to one of them, instead of needing to choose
between them on every write()?
On Fri, Jan 20, 2017 at 1:04 PM, 'Sergiy Kibrik' via
On Sat, Jan 28, 2017 at 1:22 AM, Justin Cinkelj
wrote:
> One thing confuses me - how/why the original commit worked for Nadav?
> Again some compiler difference (gcc 4.8...)
>
To be honest, I was surprised that the "magic number" was 3, I'm not even
sure who holds all
On Mon, Feb 20, 2017 at 3:32 PM, 'Sergiy Kibrik' via OSv Development <
osv-dev@googlegroups.com> wrote:
> evtchn and gnttab drivers use these operations extensively.
>
> Signed-off-by: Sergiy Kibrik
> ---
> bsd/aarch64/machine/xen/synch_bitops.h | 59
Hi, I don't have access to any version of VMware so I can't test this
myself, but I can ask you guys questions:
On Fri, Feb 10, 2017 at 1:49 PM, wrote:
>
> Here is what worked for us:
>
> 1. Download the VMware workstation image from http://downloads.osv.io.s3.
>
On Fri, Feb 17, 2017 at 1:15 PM, Justin Cinkelj
wrote:
> Running multiple instances of same program at the same time (or even
> same program multiple times) can fail as ELF object and requied libraries
> are already loaded. When program is run for the second time, global
aiting for the mutex, on a different wr), but when it
does wake up, and poll()s all the wr, it will see the woken wr as being
woken, as expected.
Fixes #824
Signed-off-by: Nadav Har'El <n...@scylladb.com>
---
core/sched.cc | 1 +
1 file changed, 1 insertion(+)
diff --git a/core/sched.cc b/core/sc
Hmm, looks reasonable to me, but I am not familar at all with the context,
so I have no idea if this change has any downsides. Glauber, maybe you can
take a look at this patch?
Thanks,
Nadav.
--
Nadav Har'El
n...@scylladb.com
On Mon, Feb 20, 2017 at 3:32 PM, 'Sergiy Kibrik' via OSv Development
On Mon, Feb 20, 2017 at 3:32 PM, 'Sergiy Kibrik' via OSv Development <
osv-dev@googlegroups.com> wrote:
> hi OSv developers,
> Let me present patches that allow OSv to work with Xen event channels on
> aarch64 platform.
> This series contains some clean up work -- to remove unused (and probably
>
Reviewed-by: Nadav Har'El <n...@scylladb.com>
One nitpick below.
On Fri, Jan 20, 2017 at 1:04 PM, 'Sergiy Kibrik' via OSv Development <
osv-dev@googlegroups.com> wrote:
> Xen crafts and provides a DTB with special node by which we can know
> which hypervisor we're using and
Allow unpinning a thread with pthread_setaffinity_np() and
sched_setaffinity() by setting the CPU mask to all CPUs.
We still do not support setting the CPU mask to an arbitrary subset of
CPUs - just to either one CPU, or all of them.
Fixes #778
Signed-off-by: Nadav Har'El <n...@scylladb.
Thanks, looks good - also according to https://www.ietf.org/rfc/rfc4627.txt
section 2.5.
I'll commit it now.
--
Nadav Har'El
n...@scylladb.com
On Thu, Jan 19, 2017 at 6:52 AM, Waldemar Kozaczuk <jwkozac...@gmail.com>
wrote:
> Changed formatter::to_json methods that format strin
On Thu, Jan 19, 2017 at 3:02 PM, wrote:
> I ran OSv on VMware player with the following modules
> (image=httpserver.fg_ssl,java,cli,certs). When I ran cli with almost
> commands, it returned the following message: "Unknown response code
> closed". But when I run the image
On Thu, Jan 19, 2017 at 3:32 PM, <mld.priv...@gmail.com> wrote:
> Vào 20:11:30 UTC+7 Thứ Năm, ngày 19 tháng 1 năm 2017, Nadav Har'El đã viết:
> > On Thu, Jan 19, 2017 at 3:02 PM, <mld.p...@gmail.com> wrote:
> > I ran OSv on VMware player with the following modules
>
not yet change sched_setaffinity() and friends to call
the new unpin() method. This will come in a later patch.
Refs #778.
Signed-off-by: Nadav Har'El <n...@scylladb.com>
---
include/osv/sched.hh | 7 +++
core/sched.cc| 36
tests/tst-
On Wed, Aug 17, 2016 at 9:54 PM, Benoit Canet <
benoit.canet.cont...@gmail.com> wrote:
> Signed-of-by: BenoÃt Canet
> ---
> bsd/porting/networking.cc | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/bsd/porting/networking.cc
Hi Justin,
On Tue, Aug 16, 2016 at 9:50 AM, Justin Cinkelj
wrote:
> pthread_getid is only a convience function to convert pthread_t to
> corresponding thread ID. It is added here to avoid manually casting
> pthread_t to osv::pthread at other places.
>
> Signed-off-by:
Hi Benoit,
On Wed, Aug 17, 2016 at 9:54 PM, Benoit Canet <
benoit.canet.cont...@gmail.com> wrote:
> This fix a bug in OSv musl that has been fixed in upstream musl
> by using netlink. Since OSv does not implement netlink add
> a function in if.cc in order to copy and interface mac address
>
Thanks Justin.
I've been away from civilization for a week (a vacation plus a
malfunctioning laptop) so hopefully I'll be able to commit your patch
tomorrow.
The patch looks good and correct..
--
Nadav Har'El
n...@scylladb.com
On Thu, Aug 18, 2016 at 3:30 PM, Justin Cinkelj <justin.c
On Thu, Aug 18, 2016 at 2:41 PM, Benoît Canet
wrote:
>
> For Makefile they are a well know UNIX mess.
> One of the official documentation is here: https://www.gnu.org/
> software/make/manual/make.html but it's not easy to grok.
> You will probably end up doing like
this new application::run_and_join().
The osv_execve() suffered from the unnecessary thread in that the actual
application code does not run in the thread id returned by it, but rather
in a child thread. After this patch, it should return the right thread id.
Signed-off-by: Nadav Har'El &l
On Sun, Feb 26, 2017 at 1:45 PM, Nadav Har'El <n...@scylladb.com> wrote:
>
> The "-s" option runs all the tests (I assume only the simple C ones, not
> more complex tests involving host-side processing, etc.?) on one OSv guest.
> We haven't done this in a long while,
Thanks. I can reproduce this on any machine by running on pure qemu without
KVM. I opened an issue: https://github.com/cloudius-systems/osv/issues/855
--
Nadav Har'El
n...@scylladb.com
On Fri, Feb 24, 2017 at 2:31 PM, Waldek Kozaczuk <jwkozac...@gmail.com>
wrote:
> This test fails con
I opened https://github.com/cloudius-systems/osv/issues/859 on this test
failure.
I can't reproduce it on my own machine - I tried 1,000 times and it never
failed once.
--
Nadav Har'El
n...@scylladb.com
On Tue, Feb 28, 2017 at 3:47 PM, <jenk...@scylladb.com> wrote:
>
On Tue, Aug 23, 2016 at 12:32 AM, Nadav Har'El <n...@scylladb.com> wrote:
>
> On Mon, Aug 22, 2016 at 6:51 PM, Justin Cinkelj <justin.cink...@xlab.si>
> wrote:
>
>> Seems that here and there, osv_execve returns with thread_id still at 0,
>> instead of containin
r only know use case.
Signed-off-by: Nadav Har'El <n...@scylladb.com>
---
include/osv/app.hh | 7 +--
core/app.cc| 11 ---
core/osv_execve.cc | 23 ---
3 files changed, 25 insertions(+), 16 deletions(-)
diff --git a/include/osv/app.hh b/includ
Hi Benoit (and Pekka), thanks. Here are some comments and questions:
--
Nadav Har'El
n...@scylladb.com
On Thu, Aug 25, 2016 at 1:07 PM, Benoit Canet <
benoit.canet.cont...@gmail.com> wrote:
> Enable "fast system calls" via the 'syscall' instruction on OSv. The
> ins
I just sent another RFC patch which should cause osv_execve to wait until
app_runtime() is the new application's.
I haven't really tested it much beyond the usual "make check" - can you
please try it out?
Thanks,
Nadav.
--
Nadav Har'El
n...@scylladb.com
On Wed, Aug 24, 2016 at 7:21
/.gitorderfile
(the "renames" thing is unrelated, but also useful)
And then in ~/.gitorderfile something like:
*.hh
*.cc
*
--
Nadav Har'El
n...@scylladb.com
On Fri, Sep 2, 2016 at 3:24 PM, Benoit Canet <benoit.canet.cont...@gmail.com
> wrote:
> The go c-shared object need th
On Fri, Sep 2, 2016 at 5:35 PM, Benoit Canet wrote:
> This will allow go program to get the vdso library address.
>
> Signed-off-by: Benoît Canet
> ---
> core/app.cc| 29 ++---
> include/osv/app.hh | 8
does *something* similar to what I want - it seems it has
a single TLS variable (in the "initial exec" variant of the ELF TLS
standard). On the other hand, I also read it still uses prctl()...
So *should* c-shared do what I am looking for? Does it in fact do this
currently?
On Fri, Sep 2, 2016 at 12:58 AM, Paweł Dziepak <pdzie...@scylladb.com>
wrote:
> On 1 September 2016 at 12:05, Nadav Har'El <n...@scylladb.com> wrote:
>
> BUT, from the code we've inspected, it seems that golang-compiled code
>> doesn't use this technique. Instead, it
On Tue, Aug 30, 2016 at 4:45 PM, Benoit Canet <
benoit.canet.cont...@gmail.com> wrote:
> The last signal of the list didn't have enough room for
> him because signal numbers start at 1.
> Make him some room so sigaction won't fail for him
> when go lib execute.
>
We have a bunch of other places
No functional changes.
Signed-off-by: Nadav Har'El <n...@scylladb.com>
---
arch/x64/entry.S | 166 ---
1 file changed, 86 insertions(+), 80 deletions(-)
diff --git a/arch/x64/entry.S b/arch/x64/entry.S
index 48a0a71..e265a30 100644
---
out of the registers saved on the stack, so I took
this opportunity to remove some of the useless things we saved (like rax,
which will be overwritten by a return value anyway) or saved multiple times
(rcx, rsp).
Signed-off-by: Nadav Har'El <n...@scylladb.com>
-
On Mon, Aug 29, 2016 at 8:12 PM, Avi Kivity wrote:
>
>
> On 08/29/2016 08:10 PM, Benoît Canet wrote:
>
>
>
> On Mon, Aug 29, 2016 at 7:08 PM, Avi Kivity wrote:
>
>> diff --git a/arch/x64/entry.S b/arch/x64/entry.S
>>
>>> index e265a30..e3be0bc 100644
>>>
On Tue, Aug 30, 2016 at 6:08 AM, wrote:
> Hello everyone,
> I am new to OSv,these days I have done some experiments on OSv,I
> found it was a great project,I am interested in it very much.But now I have
> some questions to ask,I will appreciate it very much if
should now be possible
(I hope) in cases which failed previously.
Signed-off-by: Nadav Har'El <n...@scylladb.com>
---
arch/x64/entry.S | 28
1 file changed, 20 insertions(+), 8 deletions(-)
diff --git a/arch/x64/entry.S b/arch/x64/entry.S
index 8fdc5a0..dda7358
e system
call.
3. It's even worse if after a thread does ARCH_SET_FS is calls a function
directly, not through a system call. e.g., malloc(). It is possible (???)
that Go is aware of this problem and restores the FS_BASE before calling C
functions, but if it doesn't, we're in big problems.
--
N
On Wed, Aug 31, 2016 at 4:26 PM, Benoît Canet
wrote:
>
>
> Go use it's own tls see: http://blog.altoros.com/golang-internals-part-5-
> runtime-bootstrap-process.html
>
Thanks. Interesting.
It is very sad, and surprising, that Go does this: If I understand
On Wed, Aug 31, 2016 at 4:26 PM, Benoît Canet
wrote:
>
> Buildmode shared alone crash.
>
>
>
the flags are |= between them.
>
This is not what I see in https://golang.org/src/cmd/go/build.go
But it seems these two options indeed have a funny interaction between
static_cast<u64*>(addr) = t->dyn_tabs.lookup(sym)->st_value -
> t->tls.size;
>break;
>case R_X86_64_IRELATIVE:
>
>
> On Wed, Aug 31, 2016 at 4:07 PM, Benoît Canet <ben...@cloudius-systems.com
> > wrote:
>
>> This is what I saw in the code managing th
On Wed, Aug 31, 2016 at 5:33 PM, Nadav Har'El <n...@scylladb.com> wrote:
> I'm more worried about the symbol(sym) part of this code... What if there
> is no symbol involved? In this case (again see readelf --relocs for your
> object), there is no object...
>
> I th
On Wed, Aug 31, 2016 at 4:04 PM, Benoit Canet <
benoit.canet.cont...@gmail.com> wrote:
> Needed by libraries produced by the Go compiler.
>
> With this patch and the golang app the runtime
> init crash later because of a vdso issue.
>
> It can be examinated by insertion an abort()
> before osv
Thanks, I committed these patches.
I think that the three separate example apps were not strictly necessary,
as one can do something like
scripts/build image=openjdk8-zulu-compact1,java-example
to build an image with this specific version of Java and the same Java
example.
--
Nadav Har'El
On Fri, Sep 23, 2016 at 4:08 PM, Waldek Kozaczuk
wrote:
> When I build it with java 8 (./scripts/build image=openjdk8-fedora,derby)
> and try to run it I get OSv to crash like this (missing PR_CallOnce system
> call?):
>
> OSv v0.24-193-gd64db4a
> eth0: 192.168.122.15
>
I'm having a hard time understanding this patch. Maybe Avi (CCed) remembers
some of the code involved (which he wrote) and might have better comments.
Some questions inline below:
--
Nadav Har'El
n...@scylladb.com
On Fri, Sep 9, 2016 at 9:08 PM, Timmons C. Player <
timmons.pla...@spirent.
d objects in
mmap()ed areas.
One of the places modified is rcu_flush(), whose on-stack sched::thread
objects are what caused issue #790.
Fixes #790.
Signed-off-by: Nadav Har'El <n...@scylladb.com>
---
drivers/virtio-net.hh | 6 ++--
drivers/virtio-rng.hh | 2 +-
driver
tack sched::thread
objects are what caused issue #790 to surface.
Signed-off-by: Nadav Har'El <n...@scylladb.com>
---
drivers/virtio-net.hh | 6 +--
drivers/virtio-rng.hh | 2 +-
drivers/vmxnet3.hh| 4 +-
include/osv/dhcp.hh | 2 +-
includ
On Fri, Sep 16, 2016 at 3:28 AM, Ian Eyberg wrote:
Hi All:
>
> I gave a presentation last night at the ACCU meetup on the 'Current
> State of C++ Unikernels'. Dor happened to be around and mentioned that this
> list might be interested in seeing the slides I was presenting
dler
could then zero it (if not already zero)?
Thanks,
Nadav.
--
Nadav Har'El
n...@scylladb.com
--
You received this message because you are subscribed to the Google Groups "OSv
Development" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to osv-de
By the way, I wonder if this solves
https://github.com/cloudius-systems/osv/issues/755 which features a crash
in a similar area of the code.
Justin (CC'ed) might be able to confirm if he can still reproduce issue 755.
--
Nadav Har'El
n...@scylladb.com
On Thu, Sep 15, 2016 at 2:18 PM, Player
I am guessing that this patch fixes
https://github.com/cloudius-systems/osv/issues/726
*Justin* - can you still reproduce this issue? If so, can you please check
if this patch fixes it?
--
Nadav Har'El
n...@scylladb.com
On Fri, Sep 9, 2016 at 8:08 PM, Timmons C. Player <
timmons.
On Mon, Sep 19, 2016 at 2:10 AM, Waldek Kozaczuk
wrote:
> OSv has already a concept of module but I wonder if its core can be made
> modular as well. For example there could be something like osv.base,
> osv.io (osv.io.read, osv.io.write, etc), osv.net, etc I do not know
>
On Mon, Sep 19, 2016 at 1:11 PM, Gleb Natapov wrote:
> > So if I understand correctly, both your and Gleb's sentiment is there is
> no
> > need not to make the scheduler more forgiving for page table changes, but
> > rather require that sched::threads objects live in a mapping
On Mon, Sep 19, 2016 at 4:37 PM, Player, Timmons wrote:
> Well, KASSERT only actually fails in debug builds, so you’d never see a
> problem with a release build
> (since NDEBUG is defined in conf/release.mk).
>
So this patch looks like it is doing reasonable things
Thanks for this patch. Looks serious, and I wonder if it explains
https://github.com/cloudius-systems/osv/issues/755
Some comments and questions below.
On Fri, Sep 9, 2016 at 8:12 PM, Timmons C. Player <
timmons.pla...@spirent.com> wrote:
> Under high levels of memory page churn, the array of
On Fri, Sep 9, 2016 at 9:26 PM, Timmons C. Player <
timmons.pla...@spirent.com> wrote:
> There is a potential lock order inversion issue here. Closing
> the file from the network side requires taking the file lock
> after taking the socket lock. Conversely, polling a socket from
> epoll takes
1 - 100 of 1121 matches
Mail list logo