From: Bobby Eshleman
Make NS changes not break hyperv. Guest vsocks still remain in the
global namespace always, so the behavior is unchanged.
Signed-off-by: Bobby Eshleman
---
net/vmw_vsock/hyperv_transport.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/vmw_vsock
From: Bobby Eshleman
Add NS support to vsock loopback. In theory, loopback can be viewed as
a given CID, and so should collide with other vsocks when the namespaces
are in global mode, but should not collide if the namespace is in local
mode. This has not been tested yet, but will be by the next
From: Bobby Eshleman
Add hooks for new internal NS calls to avoid breaking vmci. Guest vsocks
remain in global mode namespaces, so behavior is unchanged.
Signed-off-by: Bobby Eshleman
---
net/vmw_vsock/vmci_transport.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a
From: Bobby Eshleman
Add the ability to isolate vsock flows using namespaces.
The namespace for a VM is inherited from the PID that opened the
vhost-vsock device.
Signed-off-by: Bobby Eshleman
---
drivers/vhost/vsock.c | 48 ++--
1 file changed, 38
From: Bobby Eshleman
This changes virtio to not be broken by new internal API changes
required for NS support on the host side. virtio-vsock namespaces are
always global mode, so behavior is unchanged for them.
Signed-off-by: Bobby Eshleman
---
net/vmw_vsock/virtio_transport.c | 5 +++--
1
From: Bobby Eshleman
Add support to the virtio common code for passing around net namespace
pointers (tx and rx). The series still requires non-common transport
support to be added by future patches.
Signed-off-by: Bobby Eshleman
---
include/linux/virtio_vsock.h| 1 +
net
From: Bobby Eshleman
Add NS functionality (initialization, passing to transports, procfs,
etc...) to the vsock socket layer. Later patches that add NS support to
transports will depend on this patch.
Signed-off-by: Bobby Eshleman
---
include/net/af_vsock.h | 7 +-
net/vmw_vsock/af_vsock.c
bridge with socat (perhaps ip tables would be better, because it can
reduce an out-of-tree tool dependency). Another option is to not use
vsock_test for the NS tests, but it seems more robust to test all of
vsock instead of just (for example) connectibility...
Signed-off-by: Bobby Eshleman
From: Bobby Eshleman
Add a net pointer to the vsock skb and helpers for getting/setting it.
This is in preparation for adding vsock NS support.
Signed-off-by: Bobby Eshleman
---
include/linux/virtio_vsock.h | 11 +++
1 file changed, 11 insertions(+)
diff --git a/include/linux
From: Bobby Eshleman
Add helper functions for setting/getting vsock NS modes. This commit is
in preparation for adding NS support to vsock.
Signed-off-by: Bobby Eshleman
---
include/net/af_vsock.h | 46 ++
1 file changed, 46 insertions(+)
diff
From: Bobby Eshleman
Add the per-net vsock NS mode state. This only adds the structure for
holding the mode necessary and some of the definitions, but does not
integrate the functionality yet.
Signed-off-by: Bobby Eshleman
---
MAINTAINERS | 1 +
include/net/net_namespace.h
e already
all done work on this together.
Thanks again for everyone's help and reviews!
Signed-off-by: Bobby Eshleman
---
Changes in v3:
- add notion of "modes"
- add procfs /proc/net/vsock_ns_mode
- local and global modes only
- no /dev/vhost-vsock-netns
- vmtest.sh already merge
a kselftest itself. This is different than other
systems that have a "vmtest.sh", where it is used as a utility script to
spin up a VM to run the selftests as a guest (but isn't hooked into
kselftest).
Signed-off-by: Bobby Eshleman
---
Changes in v10:
- remove dupes in tools/te
a kselftest itself. This is different than other
systems that have a "vmtest.sh", where it is used as a utility script to
spin up a VM to run the selftests as a guest (but isn't hooked into
kselftest).
Signed-off-by: Bobby Eshleman
---
Changes in v9:
- make kselftest build target
On Tue, May 27, 2025 at 05:55:13PM +0200, Stefano Garzarella wrote:
> On Tue, May 27, 2025 at 07:30:29AM -0700, Bobby Eshleman wrote:
> > On Mon, May 26, 2025 at 01:18:18PM +0200, Stefano Garzarella wrote:
> > > On Thu, May 22, 2025 at 09:59:07PM -0700, Bobby Eshleman wrote:
&g
On Mon, May 26, 2025 at 01:18:18PM +0200, Stefano Garzarella wrote:
> On Thu, May 22, 2025 at 09:59:07PM -0700, Bobby Eshleman wrote:
> > This commit introduces a new vmtest.sh runner for vsock.
> >
> > It uses virtme-ng/qemu to run tests in a VM. The tests validate G2H,
a kselftest itself. This is different than other
systems that have a "vmtest.sh", where it is used as a utility script to
spin up a VM to run the selftests as a guest (but isn't hooked into
kselftest).
Signed-off-by: Bobby Eshleman
---
Changes in v8:
- remove NIPA comment from comm
On Tue, May 20, 2025 at 01:09:25PM +0200, Stefano Garzarella wrote:
> On Tue, May 20, 2025 at 12:58:18PM +0200, Paolo Abeni wrote:
> > On 5/20/25 10:24 AM, Stefano Garzarella wrote:
> >
> > Still it could be worthy to re-introduce (behind a command line option)
> > the ability to build the kernel
On Tue, May 20, 2025 at 12:58:18PM +0200, Paolo Abeni wrote:
> On 5/20/25 10:24 AM, Stefano Garzarella wrote:
>
> https://github.com/linux-netdev/nipa/wiki/How-to-run-netdev-selftests-CI-style
>
> @Bobby: AFAICS this now has all the ingredients to fit NIPA integration
> am I correct? the last com
On Tue, May 20, 2025 at 01:24:43PM +0200, Paolo Abeni wrote:
> On 5/16/25 12:00 AM, Bobby Eshleman wrote:
> > +tap_prefix() {
> > + sed -e "s/^/${TAP_PREFIX}/"
> > +}
>
> I think there is no need to the tap prefix to the output you intend to
> 'com
as a kselftest itself. This is different than other
systems that have a "vmtest.sh", where it is used as a utility script to
spin up a VM to run the selftests as a guest (but isn't hooked into
kselftest).
Testing in NIPA is still WIP.
Signed-off-by: Bobby Eshleman
---
Changes in v7:
- fi
On Thu, May 15, 2025 at 11:02:00AM +0200, Stefano Garzarella wrote:
> On Tue, May 13, 2025 at 11:31:13PM -0700, Bobby Eshleman wrote:
> > This commit introduces a new vmtest.sh runner for vsock.
> >
> > It uses virtme-ng/qemu to run tests in a VM. The tests validate G2H,
WIP.
Signed-off-by: Bobby Eshleman
---
Changes in v6:
- add make cmd in commit message in vmtest.sh example (Stefano)
- check nonzero size of QEMU_PIDFILE using -s conditional (Stefano)
- display log file path after tests so it is easier to find amongst other
random names
- cleanup qemu pidfile
st.sh as a kselftest itself. This is different than other
systems that have a "vmtest.sh", where it is used as a utility script to
spin up a VM to run the selftests as a guest (but isn't hooked into
kselftest).
Testing in NIPA is still WIP.
Signed-off-by: Bobby Eshleman
---
Chang
On Tue, May 13, 2025 at 12:08:48PM +0200, Paolo Abeni wrote:
> On 5/8/25 3:23 AM, Bobby Eshleman wrote:
> > This commit introduces a new vmtest.sh runner for vsock.
> >
> > It uses virtme-ng/qemu to run tests in a VM. The tests validate G2H,
> > H2G, and loopback. T
p_output
+echo "1..${#ARGS[@]}" | tap_output
+
+rm -f "${LOG}"
+log_setup "Booting up VM"
+vm_setup
+vm_wait_for_ssh
+log_setup "VM booted up"
+
+cnt_pass=0
+cnt_fail=0
+cnt_skip=0
+cnt_total=0
+exitcode=0
+for arg in "${ARGS[@]}"; do
+ run_test "${arg}"
+ rc=$?
+ if [[ ${rc} == $KSFT_PASS ]]; then
+ cnt_pass=$(( cnt_pass + 1 ))
+ echo "[PASS]" | tap_prefix
+ echo "ok ${cnt_total} ${arg}" | tap_output
+ elif [[ ${rc} == $KSFT_SKIP ]]; then
+ cnt_skip=$(( cnt_skip + 1 ))
+ echo "[SKIP]" | tap_prefix
+ echo "ok ${cnt_total} ${arg} # SKIP" | tap_output
+ exitcode=$KSFT_SKIP
+ elif [[ ${rc} == $KSFT_FAIL ]]; then
+ cnt_fail=$(( cnt_fail + 1 ))
+ echo "[FAIL]" | tap_prefix
+ echo "not ok ${cnt_fail} ${arg} # exit=$rc" | tap_output
+ exitcode=$KSFT_FAIL
+ fi
+ cnt_total=$(( cnt_total + 1 ))
+done
+
+echo "SUMMARY: PASS=${cnt_pass} SKIP=${cnt_skip} FAIL=${cnt_fail}" | tap_prefix
+echo "1..${cnt_total}" | tap_output
+
+exit ${rc}
---
base-commit: 8066e388be48f1ad62b0449dc1d31a25489fa12a
change-id: 20250325-vsock-vmtest-b3a21d2102c2
Best regards,
--
Bobby Eshleman
On Fri, May 02, 2025 at 12:22:46PM +0200, Paolo Abeni wrote:
> On 4/29/25 1:48 AM, Bobby Eshleman wrote:
> > This commit introduces a new vmtest.sh runner for vsock.
> >
> > It uses virtme-ng/qemu to run tests in a VM. The tests validate G2H,
> > H2G, and loopback. T
PS. sorry for weird formatting, writing from the gmail web UI instead of mutt.
Tried to limit line lengths manually. Looks weird after hitting "send"... lol.
Best,
Bobby
On Wed, Apr 30, 2025 at 7:37 PM Bobby Eshleman wrote:
>
> On Wed, Apr 30, 2025 at 6:06 AM Stefano Garz
On Wed, Apr 30, 2025 at 6:06 AM Stefano Garzarella wrote:
>
> On Mon, Apr 28, 2025 at 04:48:11PM -0700, Bobby Eshleman wrote:
> >This commit introduces a new vmtest.sh runner for vsock.
> >
> >It uses virtme-ng/qemu to run tests in a VM. The tests validate G2H,
> >
(but isn't hooked into
kselftest). This aspect is worth review, as I'm not aware of all of the
enviroments where this would run.
Signed-off-by: Bobby Eshleman
---
Changes in v3:
- use common conditional syntax for checking variables
- use return value instead of global rc
- fix typo TES
used.
Only tested on x86.
To run:
$ tools/testing/selftests/vsock/vmtest.sh
Future work can include vsock_diag_test.
Signed-off-by: Bobby Eshleman
---
Changes in v2:
- add kernel oops and warnings checker
- change testname variable to use FUNCNAME
- fix spacing in test_vm_server_host_client
On Fri, Apr 11, 2025 at 10:32:45AM +0200, Stefano Garzarella wrote:
> On Thu, Apr 10, 2025 at 06:07:59PM -0700, Bobby Eshleman wrote:
> > This commit introduces a new vmtest.sh runner for vsock.
> >
> > It uses virtme-ng/qemu to run tests in a VM. The tests are designed to
:
$ tools/testing/selftests/vsock/vmtest.sh
Signed-off-by: Bobby Eshleman
---
MAINTAINERS| 1 +
tools/testing/selftests/vsock/.gitignore | 1 +
tools/testing/selftests/vsock/config.vsock | 6 +
tools/testing/selftests/vsock/vmtest.sh| 247
33 matches
Mail list logo
