* Re: [PATCH net-next v11 08/13] selftests/vsock: add vm_dmesg_{warn,oops}_count() helpers
From: Stefano Garzarella @ 2025-11-21 14:42 UTC (permalink / raw)
To: Bobby Eshleman
Cc: David S. Miller, Eric Dumazet, Jakub Kicinski, Paolo Abeni,
Simon Horman, Stefan Hajnoczi, Michael S. Tsirkin, Jason Wang,
Eugenio Pérez, Xuan Zhuo, K. Y. Srinivasan, Haiyang Zhang,
Wei Liu, Dexuan Cui, Bryan Tan, Vishnu Dasa,
Broadcom internal kernel review list, Shuah Khan, linux-kernel,
virtualization, netdev, kvm, linux-hyperv, linux-kselftest,
berrange, Sargun Dhillon, Bobby Eshleman
In-Reply-To: <20251120-vsock-vmtest-v11-8-55cbc80249a7@meta.com>
On Thu, Nov 20, 2025 at 09:44:40PM -0800, Bobby Eshleman wrote:
>From: Bobby Eshleman <bobbyeshleman@meta.com>
>
>These functions are reused by the VM tests to collect and compare dmesg
>warnings and oops counts. The future VM-specific tests use them heavily.
>This patches relies on vm_ssh() already supporting namespaces.
>
>Signed-off-by: Bobby Eshleman <bobbyeshleman@meta.com>
>---
>Changes in v11:
>- break these out into an earlier patch so that they can be used
> directly in new patches (instead of causing churn by adding this
> later)
>---
> tools/testing/selftests/vsock/vmtest.sh | 19 +++++++++++++++----
> 1 file changed, 15 insertions(+), 4 deletions(-)
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
>
>diff --git a/tools/testing/selftests/vsock/vmtest.sh b/tools/testing/selftests/vsock/vmtest.sh
>index 4da91828a6a0..1623e4da15e2 100755
>--- a/tools/testing/selftests/vsock/vmtest.sh
>+++ b/tools/testing/selftests/vsock/vmtest.sh
>@@ -389,6 +389,17 @@ host_wait_for_listener() {
> fi
> }
>
>+vm_dmesg_oops_count() {
>+ local ns=$1
>+
>+ vm_ssh "${ns}" -- dmesg 2>/dev/null | grep -c -i 'Oops'
>+}
>+
>+vm_dmesg_warn_count() {
>+ local ns=$1
>+
>+ vm_ssh "${ns}" -- dmesg --level=warn 2>/dev/null | grep -c -i 'vsock'
>+}
>
> vm_vsock_test() {
> local ns=$1
>@@ -596,8 +607,8 @@ run_shared_vm_test() {
>
> host_oops_cnt_before=$(dmesg | grep -c -i 'Oops')
> host_warn_cnt_before=$(dmesg --level=warn | grep -c -i 'vsock')
>- vm_oops_cnt_before=$(vm_ssh -- dmesg | grep -c -i 'Oops')
>- vm_warn_cnt_before=$(vm_ssh -- dmesg --level=warn | grep -c -i 'vsock')
>+ vm_oops_cnt_before=$(vm_dmesg_oops_count "init_ns")
>+ vm_warn_cnt_before=$(vm_dmesg_warn_count "init_ns")
>
> name=$(echo "${1}" | awk '{ print $1 }')
> eval test_"${name}"
>@@ -615,13 +626,13 @@ run_shared_vm_test() {
> rc=$KSFT_FAIL
> fi
>
>- vm_oops_cnt_after=$(vm_ssh -- dmesg | grep -i 'Oops' | wc -l)
>+ vm_oops_cnt_after=$(vm_dmesg_oops_count "init_ns")
> if [[ ${vm_oops_cnt_after} -gt ${vm_oops_cnt_before} ]]; then
> echo "FAIL: kernel oops detected on vm" | log_host
> rc=$KSFT_FAIL
> fi
>
>- vm_warn_cnt_after=$(vm_ssh -- dmesg --level=warn | grep -c -i 'vsock')
>+ vm_warn_cnt_after=$(vm_dmesg_warn_count "init_ns")
> if [[ ${vm_warn_cnt_after} -gt ${vm_warn_cnt_before} ]]; then
> echo "FAIL: kernel warning detected on vm" | log_host
> rc=$KSFT_FAIL
>
>--
>2.47.3
>
^ permalink raw reply
* Re: [PATCH net-next v11 09/13] selftests/vsock: use ss to wait for listeners instead of /proc/net
From: Stefano Garzarella @ 2025-11-21 14:43 UTC (permalink / raw)
To: Bobby Eshleman
Cc: David S. Miller, Eric Dumazet, Jakub Kicinski, Paolo Abeni,
Simon Horman, Stefan Hajnoczi, Michael S. Tsirkin, Jason Wang,
Eugenio Pérez, Xuan Zhuo, K. Y. Srinivasan, Haiyang Zhang,
Wei Liu, Dexuan Cui, Bryan Tan, Vishnu Dasa,
Broadcom internal kernel review list, Shuah Khan, linux-kernel,
virtualization, netdev, kvm, linux-hyperv, linux-kselftest,
berrange, Sargun Dhillon, Bobby Eshleman
In-Reply-To: <20251120-vsock-vmtest-v11-9-55cbc80249a7@meta.com>
On Thu, Nov 20, 2025 at 09:44:41PM -0800, Bobby Eshleman wrote:
>From: Bobby Eshleman <bobbyeshleman@meta.com>
>
>Replace /proc/net parsing with ss(8) for detecting listening sockets in
>wait_for_listener() functions and add support for TCP, VSOCK, and Unix
>socket protocols.
>
>The previous implementation parsed /proc/net/tcp using awk to detect
>listening sockets, but this approach could not support vsock because
>vsock does not export socket information to /proc/net/.
>
>Instead, use ss so that we can detect listeners on tcp, vsock, and unix.
>
>The protocol parameter is now required for all wait_for_listener family
>functions (wait_for_listener, vm_wait_for_listener,
>host_wait_for_listener) to explicitly specify which socket type to wait
>for.
>
>ss is added to the dependency check in check_deps().
>
>Signed-off-by: Bobby Eshleman <bobbyeshleman@meta.com>
>---
> tools/testing/selftests/vsock/vmtest.sh | 47 +++++++++++++++++++++------------
> 1 file changed, 30 insertions(+), 17 deletions(-)
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
>
>diff --git a/tools/testing/selftests/vsock/vmtest.sh b/tools/testing/selftests/vsock/vmtest.sh
>index 1623e4da15e2..e32997db322d 100755
>--- a/tools/testing/selftests/vsock/vmtest.sh
>+++ b/tools/testing/selftests/vsock/vmtest.sh
>@@ -191,7 +191,7 @@ check_args() {
> }
>
> check_deps() {
>- for dep in vng ${QEMU} busybox pkill ssh; do
>+ for dep in vng ${QEMU} busybox pkill ssh ss; do
> if [[ ! -x $(command -v "${dep}") ]]; then
> echo -e "skip: dependency ${dep} not found!\n"
> exit "${KSFT_SKIP}"
>@@ -346,21 +346,32 @@ wait_for_listener()
> local port=$1
> local interval=$2
> local max_intervals=$3
>- local protocol=tcp
>- local pattern
>+ local protocol=$4
> local i
>
>- pattern=":$(printf "%04X" "${port}") "
>-
>- # for tcp protocol additionally check the socket state
>- [ "${protocol}" = "tcp" ] && pattern="${pattern}0A"
>-
> for i in $(seq "${max_intervals}"); do
>- if awk -v pattern="${pattern}" \
>- 'BEGIN {rc=1} $2" "$4 ~ pattern {rc=0} END {exit rc}' \
>- /proc/net/"${protocol}"*; then
>+ case "${protocol}" in
>+ tcp)
>+ if ss --listening --tcp --numeric | grep -q ":${port} "; then
>+ break
>+ fi
>+ ;;
>+ vsock)
>+ if ss --listening --vsock --numeric | grep -q ":${port} "; then
>+ break
>+ fi
>+ ;;
>+ unix)
>+ # For unix sockets, port is actually the socket path
>+ if ss --listening --unix | grep -q "${port}"; then
>+ break
>+ fi
>+ ;;
>+ *)
>+ echo "Unknown protocol: ${protocol}" >&2
> break
>- fi
>+ ;;
>+ esac
> sleep "${interval}"
> done
> }
>@@ -368,23 +379,25 @@ wait_for_listener()
> vm_wait_for_listener() {
> local ns=$1
> local port=$2
>+ local protocol=$3
>
> vm_ssh "${ns}" <<EOF
> $(declare -f wait_for_listener)
>-wait_for_listener ${port} ${WAIT_PERIOD} ${WAIT_PERIOD_MAX}
>+wait_for_listener ${port} ${WAIT_PERIOD} ${WAIT_PERIOD_MAX} ${protocol}
> EOF
> }
>
> host_wait_for_listener() {
> local ns=$1
> local port=$2
>+ local protocol=$3
>
> if [[ "${ns}" == "init_ns" ]]; then
>- wait_for_listener "${port}" "${WAIT_PERIOD}" "${WAIT_PERIOD_MAX}"
>+ wait_for_listener "${port}" "${WAIT_PERIOD}" "${WAIT_PERIOD_MAX}" "${protocol}"
> else
> ip netns exec "${ns}" bash <<-EOF
> $(declare -f wait_for_listener)
>- wait_for_listener ${port} ${WAIT_PERIOD} ${WAIT_PERIOD_MAX}
>+ wait_for_listener ${port} ${WAIT_PERIOD} ${WAIT_PERIOD_MAX} ${protocol}
> EOF
> fi
> }
>@@ -431,7 +444,7 @@ vm_vsock_test() {
> return $rc
> fi
>
>- vm_wait_for_listener "${ns}" "${port}"
>+ vm_wait_for_listener "${ns}" "${port}" "tcp"
> rc=$?
> fi
> set +o pipefail
>@@ -472,7 +485,7 @@ host_vsock_test() {
> return $rc
> fi
>
>- host_wait_for_listener "${ns}" "${port}"
>+ host_wait_for_listener "${ns}" "${port}" "tcp"
> rc=$?
> fi
> set +o pipefail
>
>--
>2.47.3
>
^ permalink raw reply
* Re: [PATCH net-next v11 10/13] selftests/vsock: add tests for proc sys vsock ns_mode
From: Stefano Garzarella @ 2025-11-21 14:44 UTC (permalink / raw)
To: Bobby Eshleman
Cc: David S. Miller, Eric Dumazet, Jakub Kicinski, Paolo Abeni,
Simon Horman, Stefan Hajnoczi, Michael S. Tsirkin, Jason Wang,
Eugenio Pérez, Xuan Zhuo, K. Y. Srinivasan, Haiyang Zhang,
Wei Liu, Dexuan Cui, Bryan Tan, Vishnu Dasa,
Broadcom internal kernel review list, Shuah Khan, linux-kernel,
virtualization, netdev, kvm, linux-hyperv, linux-kselftest,
berrange, Sargun Dhillon, Bobby Eshleman
In-Reply-To: <20251120-vsock-vmtest-v11-10-55cbc80249a7@meta.com>
On Thu, Nov 20, 2025 at 09:44:42PM -0800, Bobby Eshleman wrote:
>From: Bobby Eshleman <bobbyeshleman@meta.com>
>
>Add tests for the /proc/sys/net/vsock/ns_mode interface. Namely,
>that it accepts "global" and "local" strings and enforces a write-once
>policy.
>
>Start a convention of commenting the test name over the test
>description. Add test name comments over test descriptions that existed
>before this convention.
>
>Add a check_netns() function that checks if the test requires namespaces
>and if the current kernel supports namespaces. Skip tests that require
>namespaces if the system does not have namespace support.
>
>Add a test to verify that guest VMs with an active G2H transport
>(virtio-vsock) cannot set namespace mode to 'local'. This validates
>the mutual exclusion between G2H transports and LOCAL mode.
>
>This patch is the first to add tests that do *not* re-use the same
>shared VM. For that reason, it adds a run_tests() function to run these
>tests and filter out the shared VM tests.
>
>Signed-off-by: Bobby Eshleman <bobbyeshleman@meta.com>
>---
>Changes in v11:
>- Document ns_ prefix above TEST_NAMES (Stefano)
>
>Changes in v10:
>- Remove extraneous add_namespaces/del_namespaces calls.
>- Rename run_tests() to run_ns_tests() since it is designed to only
> run ns tests.
>
>Changes in v9:
>- add test ns_vm_local_mode_rejected to check that guests cannot use
> local mode
>---
> tools/testing/selftests/vsock/vmtest.sh | 143 +++++++++++++++++++++++++++++++-
> 1 file changed, 141 insertions(+), 2 deletions(-)
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
>
>diff --git a/tools/testing/selftests/vsock/vmtest.sh b/tools/testing/selftests/vsock/vmtest.sh
>index e32997db322d..2e077e8a1777 100755
>--- a/tools/testing/selftests/vsock/vmtest.sh
>+++ b/tools/testing/selftests/vsock/vmtest.sh
>@@ -41,14 +41,43 @@ readonly KERNEL_CMDLINE="\
> virtme.ssh virtme_ssh_channel=tcp virtme_ssh_user=$USER \
> "
> readonly LOG=$(mktemp /tmp/vsock_vmtest_XXXX.log)
>-readonly TEST_NAMES=(vm_server_host_client vm_client_host_server vm_loopback)
>+
>+# Namespace tests must use the ns_ prefix. This is checked in check_netns() and
>+# is used to determine if a test needs namespace setup before test execution.
>+readonly TEST_NAMES=(
>+ vm_server_host_client
>+ vm_client_host_server
>+ vm_loopback
>+ ns_host_vsock_ns_mode_ok
>+ ns_host_vsock_ns_mode_write_once_ok
>+ ns_vm_local_mode_rejected
>+)
> readonly TEST_DESCS=(
>+ # vm_server_host_client
> "Run vsock_test in server mode on the VM and in client mode on the host."
>+
>+ # vm_client_host_server
> "Run vsock_test in client mode on the VM and in server mode on the host."
>+
>+ # vm_loopback
> "Run vsock_test using the loopback transport in the VM."
>+
>+ # ns_host_vsock_ns_mode_ok
>+ "Check /proc/sys/net/vsock/ns_mode strings on the host."
>+
>+ # ns_host_vsock_ns_mode_write_once_ok
>+ "Check /proc/sys/net/vsock/ns_mode is write-once on the host."
>+
>+ # ns_vm_local_mode_rejected
>+ "Test that guest VM with G2H transport cannot set namespace mode to 'local'"
> )
>
>-readonly USE_SHARED_VM=(vm_server_host_client vm_client_host_server vm_loopback)
>+readonly USE_SHARED_VM=(
>+ vm_server_host_client
>+ vm_client_host_server
>+ vm_loopback
>+ ns_vm_local_mode_rejected
>+)
> readonly NS_MODES=("local" "global")
>
> VERBOSE=0
>@@ -205,6 +234,20 @@ check_deps() {
> fi
> }
>
>+check_netns() {
>+ local tname=$1
>+
>+ # If the test requires NS support, check if NS support exists
>+ # using /proc/self/ns
>+ if [[ "${tname}" =~ ^ns_ ]] &&
>+ [[ ! -e /proc/self/ns ]]; then
>+ log_host "No NS support detected for test ${tname}"
>+ return 1
>+ fi
>+
>+ return 0
>+}
>+
> check_vng() {
> local tested_versions
> local version
>@@ -528,6 +571,32 @@ log_guest() {
> LOG_PREFIX=guest log "$@"
> }
>
>+test_ns_host_vsock_ns_mode_ok() {
>+ for mode in "${NS_MODES[@]}"; do
>+ if ! ns_set_mode "${mode}0" "${mode}"; then
>+ return "${KSFT_FAIL}"
>+ fi
>+ done
>+
>+ return "${KSFT_PASS}"
>+}
>+
>+test_ns_host_vsock_ns_mode_write_once_ok() {
>+ for mode in "${NS_MODES[@]}"; do
>+ local ns="${mode}0"
>+ if ! ns_set_mode "${ns}" "${mode}"; then
>+ return "${KSFT_FAIL}"
>+ fi
>+
>+ # try writing again and expect failure
>+ if ns_set_mode "${ns}" "${mode}"; then
>+ return "${KSFT_FAIL}"
>+ fi
>+ done
>+
>+ return "${KSFT_PASS}"
>+}
>+
> test_vm_server_host_client() {
> if ! vm_vsock_test "init_ns" "server" 2 "${TEST_GUEST_PORT}"; then
> return "${KSFT_FAIL}"
>@@ -569,6 +638,26 @@ test_vm_loopback() {
> return "${KSFT_PASS}"
> }
>
>+test_ns_vm_local_mode_rejected() {
>+ # Guest VMs have a G2H transport (virtio-vsock) active, so they
>+ # should not be able to set namespace mode to 'local'.
>+ # This test verifies that the sysctl write fails as expected.
>+
>+ # Try to set local mode in the guest's init_ns
>+ if vm_ssh init_ns "echo local | tee /proc/sys/net/vsock/ns_mode &>/dev/null"; then
>+ return "${KSFT_FAIL}"
>+ fi
>+
>+ # Verify mode is still 'global'
>+ local mode
>+ mode=$(vm_ssh init_ns "cat /proc/sys/net/vsock/ns_mode")
>+ if [[ "${mode}" != "global" ]]; then
>+ return "${KSFT_FAIL}"
>+ fi
>+
>+ return "${KSFT_PASS}"
>+}
>+
> shared_vm_test() {
> local tname
>
>@@ -601,6 +690,11 @@ run_shared_vm_tests() {
> continue
> fi
>
>+ if ! check_netns "${arg}"; then
>+ check_result "${KSFT_SKIP}" "${arg}"
>+ continue
>+ fi
>+
> run_shared_vm_test "${arg}"
> check_result "$?" "${arg}"
> done
>@@ -654,6 +748,49 @@ run_shared_vm_test() {
> return "${rc}"
> }
>
>+run_ns_tests() {
>+ for arg in "${ARGS[@]}"; do
>+ if shared_vm_test "${arg}"; then
>+ continue
>+ fi
>+
>+ if ! check_netns "${arg}"; then
>+ check_result "${KSFT_SKIP}" "${arg}"
>+ continue
>+ fi
>+
>+ add_namespaces
>+
>+ name=$(echo "${arg}" | awk '{ print $1 }')
>+ log_host "Executing test_${name}"
>+
>+ host_oops_before=$(dmesg 2>/dev/null | grep -c -i 'Oops')
>+ host_warn_before=$(dmesg --level=warn 2>/dev/null | grep -c -i 'vsock')
>+ eval test_"${name}"
>+ rc=$?
>+
>+ host_oops_after=$(dmesg 2>/dev/null | grep -c -i 'Oops')
>+ if [[ "${host_oops_after}" -gt "${host_oops_before}" ]]; then
>+ echo "FAIL: kernel oops detected on host" | log_host
>+ check_result "${KSFT_FAIL}" "${name}"
>+ del_namespaces
>+ continue
>+ fi
>+
>+ host_warn_after=$(dmesg --level=warn 2>/dev/null | grep -c -i 'vsock')
>+ if [[ "${host_warn_after}" -gt "${host_warn_before}" ]]; then
>+ echo "FAIL: kernel warning detected on host" | log_host
>+ check_result "${KSFT_FAIL}" "${name}"
>+ del_namespaces
>+ continue
>+ fi
>+
>+ check_result "${rc}" "${name}"
>+
>+ del_namespaces
>+ done
>+}
>+
> BUILD=0
> QEMU="qemu-system-$(uname -m)"
>
>@@ -699,6 +836,8 @@ if shared_vm_tests_requested "${ARGS[@]}"; then
> terminate_pidfiles "${pidfile}"
> fi
>
>+run_ns_tests "${ARGS[@]}"
>+
> echo "SUMMARY: PASS=${cnt_pass} SKIP=${cnt_skip} FAIL=${cnt_fail}"
> echo "Log: ${LOG}"
>
>
>--
>2.47.3
>
^ permalink raw reply
* Re: [PATCH net-next v11 11/13] selftests/vsock: add namespace tests for CID collisions
From: Stefano Garzarella @ 2025-11-21 14:48 UTC (permalink / raw)
To: Bobby Eshleman
Cc: David S. Miller, Eric Dumazet, Jakub Kicinski, Paolo Abeni,
Simon Horman, Stefan Hajnoczi, Michael S. Tsirkin, Jason Wang,
Eugenio Pérez, Xuan Zhuo, K. Y. Srinivasan, Haiyang Zhang,
Wei Liu, Dexuan Cui, Bryan Tan, Vishnu Dasa,
Broadcom internal kernel review list, Shuah Khan, linux-kernel,
virtualization, netdev, kvm, linux-hyperv, linux-kselftest,
berrange, Sargun Dhillon, Bobby Eshleman
In-Reply-To: <20251120-vsock-vmtest-v11-11-55cbc80249a7@meta.com>
On Thu, Nov 20, 2025 at 09:44:43PM -0800, Bobby Eshleman wrote:
>From: Bobby Eshleman <bobbyeshleman@meta.com>
>
>Add tests to verify CID collision rules across different vsock namespace
>modes.
>
>1. Two VMs with the same CID cannot start in different global namespaces
> (ns_global_same_cid_fails)
>2. Two VMs with the same CID can start in different local namespaces
> (ns_local_same_cid_ok)
>3. VMs with the same CID can coexist when one is in a global namespace
> and another is in a local namespace (ns_global_local_same_cid_ok and
> ns_local_global_same_cid_ok)
>
>The tests ns_global_local_same_cid_ok and ns_local_global_same_cid_ok
>make sure that ordering does not matter.
>
>The tests use a shared helper function namespaces_can_boot_same_cid()
>that attempts to start two VMs with identical CIDs in the specified
>namespaces and verifies whether VM initialization failed or succeeded.
>
>Signed-off-by: Bobby Eshleman <bobbyeshleman@meta.com>
>---
>Changes in v11:
>- check vm_start() rc in namespaces_can_boot_same_cid() (Stefano)
>- fix ns_local_same_cid_ok() to use local0 and local1 instead of reusing
> local0 twice. This check should pass, ensuring local namespaces do not
> collide (Stefano)
>---
> tools/testing/selftests/vsock/vmtest.sh | 78 +++++++++++++++++++++++++++++++++
> 1 file changed, 78 insertions(+)
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
>
>diff --git a/tools/testing/selftests/vsock/vmtest.sh b/tools/testing/selftests/vsock/vmtest.sh
>index 2e077e8a1777..f84da1e8ad14 100755
>--- a/tools/testing/selftests/vsock/vmtest.sh
>+++ b/tools/testing/selftests/vsock/vmtest.sh
>@@ -51,6 +51,10 @@ readonly TEST_NAMES=(
> ns_host_vsock_ns_mode_ok
> ns_host_vsock_ns_mode_write_once_ok
> ns_vm_local_mode_rejected
>+ ns_global_same_cid_fails
>+ ns_local_same_cid_ok
>+ ns_global_local_same_cid_ok
>+ ns_local_global_same_cid_ok
> )
> readonly TEST_DESCS=(
> # vm_server_host_client
>@@ -70,6 +74,18 @@ readonly TEST_DESCS=(
>
> # ns_vm_local_mode_rejected
> "Test that guest VM with G2H transport cannot set namespace mode to 'local'"
>+
>+ # ns_global_same_cid_fails
>+ "Check QEMU fails to start two VMs with same CID in two different global namespaces."
>+
>+ # ns_local_same_cid_ok
>+ "Check QEMU successfully starts two VMs with same CID in two different local namespaces."
>+
>+ # ns_global_local_same_cid_ok
>+ "Check QEMU successfully starts one VM in a global ns and then another VM in a local ns with the same CID."
>+
>+ # ns_local_global_same_cid_ok
>+ "Check QEMU successfully starts one VM in a local ns and then another VM in a global ns with the same CID."
> )
>
> readonly USE_SHARED_VM=(
>@@ -581,6 +597,68 @@ test_ns_host_vsock_ns_mode_ok() {
> return "${KSFT_PASS}"
> }
>
>+namespaces_can_boot_same_cid() {
>+ local ns0=$1
>+ local ns1=$2
>+ local pidfile1 pidfile2
>+ local rc
>+
>+ pidfile1="$(create_pidfile)"
>+
>+ # The first VM should be able to start. If it can't then we have
>+ # problems and need to return non-zero.
>+ if ! vm_start "${pidfile1}" "${ns0}"; then
>+ return 1
>+ fi
>+
>+ pidfile2="$(create_pidfile)"
>+ vm_start "${pidfile2}" "${ns1}"
>+ rc=$?
>+ terminate_pidfiles "${pidfile1}" "${pidfile2}"
>+
>+ return "${rc}"
>+}
>+
>+test_ns_global_same_cid_fails() {
>+ init_namespaces
>+
>+ if namespaces_can_boot_same_cid "global0" "global1"; then
>+ return "${KSFT_FAIL}"
>+ fi
>+
>+ return "${KSFT_PASS}"
>+}
>+
>+test_ns_local_global_same_cid_ok() {
>+ init_namespaces
>+
>+ if namespaces_can_boot_same_cid "local0" "global0"; then
>+ return "${KSFT_PASS}"
>+ fi
>+
>+ return "${KSFT_FAIL}"
>+}
>+
>+test_ns_global_local_same_cid_ok() {
>+ init_namespaces
>+
>+ if namespaces_can_boot_same_cid "global0" "local0"; then
>+ return "${KSFT_PASS}"
>+ fi
>+
>+ return "${KSFT_FAIL}"
>+}
>+
>+test_ns_local_same_cid_ok() {
>+ init_namespaces
>+
>+ if namespaces_can_boot_same_cid "local0" "local1"; then
>+ return "${KSFT_PASS}"
>+ fi
>+
>+ return "${KSFT_FAIL}"
>+}
>+
> test_ns_host_vsock_ns_mode_write_once_ok() {
> for mode in "${NS_MODES[@]}"; do
> local ns="${mode}0"
>
>--
>2.47.3
>
^ permalink raw reply
* Re: [PATCH net-next v11 12/13] selftests/vsock: add tests for host <-> vm connectivity with namespaces
From: Stefano Garzarella @ 2025-11-21 14:52 UTC (permalink / raw)
To: Bobby Eshleman
Cc: David S. Miller, Eric Dumazet, Jakub Kicinski, Paolo Abeni,
Simon Horman, Stefan Hajnoczi, Michael S. Tsirkin, Jason Wang,
Eugenio Pérez, Xuan Zhuo, K. Y. Srinivasan, Haiyang Zhang,
Wei Liu, Dexuan Cui, Bryan Tan, Vishnu Dasa,
Broadcom internal kernel review list, Shuah Khan, linux-kernel,
virtualization, netdev, kvm, linux-hyperv, linux-kselftest,
berrange, Sargun Dhillon, Bobby Eshleman
In-Reply-To: <20251120-vsock-vmtest-v11-12-55cbc80249a7@meta.com>
On Thu, Nov 20, 2025 at 09:44:44PM -0800, Bobby Eshleman wrote:
>From: Bobby Eshleman <bobbyeshleman@meta.com>
>
>Add tests to validate namespace correctness using vsock_test and socat.
>The vsock_test tool is used to validate expected success tests, but
>socat is used for expected failure tests. socat is used to ensure that
>connections are rejected outright instead of failing due to some other
>socket behavior (as tested in vsock_test). Additionally, socat is
>already required for tunneling TCP traffic from vsock_test. Using only
>one of the vsock_test tests like 'test_stream_client_close_client' would
>have yielded a similar result, but doing so wouldn't remove the socat
>dependency.
>
>Additionally, check for the dependency socat. socat needs special
>handling beyond just checking if it is on the path because it must be
>compiled with support for both vsock and unix. The function
>check_socat() checks that this support exists.
>
>Add more padding to test name printf strings because the tests added in
>this patch would otherwise overflow.
>
>Add vm_dmesg_start() and vm_dmesg_check() to encapsulate checking dmesg
>for oops and warnings.
>
>Signed-off-by: Bobby Eshleman <bobbyeshleman@meta.com>
>---
>Changes in v11:
>- add 'sleep "${WAIT_PERIOD}"' after any non-TCP socat LISTEN cmd
> (Stefano)
I guess you meant wait_for_listener, right?
>- add host_wait_for_listener() after any socat TCP-LISTEN (Stefano)
>- reuse vm_dmesg_{oops,warn}_count() inside vm_dmesg_check()
>- fix copy-paste in test_ns_same_local_vm_connect_to_local_host_ok()
> (Stefano)
>
>Changes in v10:
>- add vm_dmesg_start() and vm_dmesg_check()
>
>Changes in v9:
>- consistent variable quoting
>---
> tools/testing/selftests/vsock/vmtest.sh | 557 +++++++++++++++++++++++++++++++-
> 1 file changed, 555 insertions(+), 2 deletions(-)
LGTM:
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
>
>diff --git a/tools/testing/selftests/vsock/vmtest.sh b/tools/testing/selftests/vsock/vmtest.sh
>index f84da1e8ad14..dfa895abfc7f 100755
>--- a/tools/testing/selftests/vsock/vmtest.sh
>+++ b/tools/testing/selftests/vsock/vmtest.sh
>@@ -7,6 +7,7 @@
> # * virtme-ng
> # * busybox-static (used by virtme-ng)
> # * qemu (used by virtme-ng)
>+# * socat
> #
> # shellcheck disable=SC2317,SC2119
>
>@@ -55,6 +56,19 @@ readonly TEST_NAMES=(
> ns_local_same_cid_ok
> ns_global_local_same_cid_ok
> ns_local_global_same_cid_ok
>+ ns_diff_global_host_connect_to_global_vm_ok
>+ ns_diff_global_host_connect_to_local_vm_fails
>+ ns_diff_global_vm_connect_to_global_host_ok
>+ ns_diff_global_vm_connect_to_local_host_fails
>+ ns_diff_local_host_connect_to_local_vm_fails
>+ ns_diff_local_vm_connect_to_local_host_fails
>+ ns_diff_global_to_local_loopback_local_fails
>+ ns_diff_local_to_global_loopback_fails
>+ ns_diff_local_to_local_loopback_fails
>+ ns_diff_global_to_global_loopback_ok
>+ ns_same_local_loopback_ok
>+ ns_same_local_host_connect_to_local_vm_ok
>+ ns_same_local_vm_connect_to_local_host_ok
> )
> readonly TEST_DESCS=(
> # vm_server_host_client
>@@ -86,6 +100,45 @@ readonly TEST_DESCS=(
>
> # ns_local_global_same_cid_ok
> "Check QEMU successfully starts one VM in a local ns and then another VM in a global ns with the same CID."
>+
>+ # ns_diff_global_host_connect_to_global_vm_ok
>+ "Run vsock_test client in global ns with server in VM in another global ns."
>+
>+ # ns_diff_global_host_connect_to_local_vm_fails
>+ "Run socat to test a process in a global ns fails to connect to a VM in a local ns."
>+
>+ # ns_diff_global_vm_connect_to_global_host_ok
>+ "Run vsock_test client in VM in a global ns with server in another global ns."
>+
>+ # ns_diff_global_vm_connect_to_local_host_fails
>+ "Run socat to test a VM in a global ns fails to connect to a host process in a local ns."
>+
>+ # ns_diff_local_host_connect_to_local_vm_fails
>+ "Run socat to test a host process in a local ns fails to connect to a VM in another local ns."
>+
>+ # ns_diff_local_vm_connect_to_local_host_fails
>+ "Run socat to test a VM in a local ns fails to connect to a host process in another local ns."
>+
>+ # ns_diff_global_to_local_loopback_local_fails
>+ "Run socat to test a loopback vsock in a global ns fails to connect to a vsock in a local ns."
>+
>+ # ns_diff_local_to_global_loopback_fails
>+ "Run socat to test a loopback vsock in a local ns fails to connect to a vsock in a global ns."
>+
>+ # ns_diff_local_to_local_loopback_fails
>+ "Run socat to test a loopback vsock in a local ns fails to connect to a vsock in another local ns."
>+
>+ # ns_diff_global_to_global_loopback_ok
>+ "Run socat to test a loopback vsock in a global ns successfully connects to a vsock in another global ns."
>+
>+ # ns_same_local_loopback_ok
>+ "Run socat to test a loopback vsock in a local ns successfully connects to a vsock in the same ns."
>+
>+ # ns_same_local_host_connect_to_local_vm_ok
>+ "Run vsock_test client in a local ns with server in VM in same ns."
>+
>+ # ns_same_local_vm_connect_to_local_host_ok
>+ "Run vsock_test client in VM in a local ns with server in same ns."
> )
>
> readonly USE_SHARED_VM=(
>@@ -117,7 +170,7 @@ usage() {
> for ((i = 0; i < ${#TEST_NAMES[@]}; i++)); do
> name=${TEST_NAMES[${i}]}
> desc=${TEST_DESCS[${i}]}
>- printf "\t%-35s%-35s\n" "${name}" "${desc}"
>+ printf "\t%-55s%-35s\n" "${name}" "${desc}"
> done
> echo
>
>@@ -236,7 +289,7 @@ check_args() {
> }
>
> check_deps() {
>- for dep in vng ${QEMU} busybox pkill ssh ss; do
>+ for dep in vng ${QEMU} busybox pkill ssh ss socat; do
> if [[ ! -x $(command -v "${dep}") ]]; then
> echo -e "skip: dependency ${dep} not found!\n"
> exit "${KSFT_SKIP}"
>@@ -287,6 +340,20 @@ check_vng() {
> fi
> }
>
>+check_socat() {
>+ local support_string
>+
>+ support_string="$(socat -V)"
>+
>+ if [[ "${support_string}" != *"WITH_VSOCK 1"* ]]; then
>+ die "err: socat is missing vsock support"
>+ fi
>+
>+ if [[ "${support_string}" != *"WITH_UNIX 1"* ]]; then
>+ die "err: socat is missing unix support"
>+ fi
>+}
>+
> handle_build() {
> if [[ ! "${BUILD}" -eq 1 ]]; then
> return
>@@ -335,6 +402,14 @@ terminate_pidfiles() {
> done
> }
>
>+terminate_pids() {
>+ local pid
>+
>+ for pid in "$@"; do
>+ kill -SIGTERM "${pid}" &>/dev/null || :
>+ done
>+}
>+
> vm_start() {
> local pidfile=$1
> local ns=$2
>@@ -473,6 +548,28 @@ vm_dmesg_warn_count() {
> vm_ssh "${ns}" -- dmesg --level=warn 2>/dev/null | grep -c -i 'vsock'
> }
>
>+vm_dmesg_check() {
>+ local pidfile=$1
>+ local ns=$2
>+ local oops_before=$3
>+ local warn_before=$4
>+ local oops_after warn_after
>+
>+ oops_after=$(vm_dmesg_oops_count "${ns}")
>+ if [[ "${oops_after}" -gt "${oops_before}" ]]; then
>+ echo "FAIL: kernel oops detected on vm in ns ${ns}" | log_host
>+ return 1
>+ fi
>+
>+ warn_after=$(vm_dmesg_warn_count "${ns}")
>+ if [[ "${warn_after}" -gt "${warn_before}" ]]; then
>+ echo "FAIL: kernel warning detected on vm in ns ${ns}" | log_host
>+ return 1
>+ fi
>+
>+ return 0
>+}
>+
> vm_vsock_test() {
> local ns=$1
> local host=$2
>@@ -597,6 +694,461 @@ test_ns_host_vsock_ns_mode_ok() {
> return "${KSFT_PASS}"
> }
>
>+test_ns_diff_global_host_connect_to_global_vm_ok() {
>+ local oops_before warn_before
>+ local pids pid pidfile
>+ local ns0 ns1 port
>+ declare -a pids
>+ local unixfile
>+ ns0="global0"
>+ ns1="global1"
>+ port=1234
>+ local rc
>+
>+ init_namespaces
>+
>+ pidfile="$(create_pidfile)"
>+
>+ if ! vm_start "${pidfile}" "${ns0}"; then
>+ return "${KSFT_FAIL}"
>+ fi
>+
>+ vm_wait_for_ssh "${ns0}"
>+ oops_before=$(vm_dmesg_oops_count "${ns0}")
>+ warn_before=$(vm_dmesg_warn_count "${ns0}")
>+
>+ unixfile=$(mktemp -u /tmp/XXXX.sock)
>+ ip netns exec "${ns1}" \
>+ socat TCP-LISTEN:"${TEST_HOST_PORT}",fork \
>+ UNIX-CONNECT:"${unixfile}" &
>+ pids+=($!)
>+ host_wait_for_listener "${ns1}" "${TEST_HOST_PORT}" "tcp"
>+
>+ ip netns exec "${ns0}" socat UNIX-LISTEN:"${unixfile}",fork \
>+ TCP-CONNECT:localhost:"${TEST_HOST_PORT}" &
>+ pids+=($!)
>+ host_wait_for_listener "${ns0}" "${unixfile}" "unix"
>+
>+ vm_vsock_test "${ns0}" "server" 2 "${TEST_GUEST_PORT}"
>+ vm_wait_for_listener "${ns0}" "${TEST_GUEST_PORT}" "tcp"
>+ host_vsock_test "${ns1}" "127.0.0.1" "${VSOCK_CID}" "${TEST_HOST_PORT}"
>+ rc=$?
>+
>+ vm_dmesg_check "${pidfile}" "${ns0}" "${oops_before}" "${warn_before}"
>+ dmesg_rc=$?
>+
>+ terminate_pids "${pids[@]}"
>+ terminate_pidfiles "${pidfile}"
>+
>+ if [[ "${rc}" -ne 0 ]] || [[ "${dmesg_rc}" -ne 0 ]]; then
>+ return "${KSFT_FAIL}"
>+ fi
>+
>+ return "${KSFT_PASS}"
>+}
>+
>+test_ns_diff_global_host_connect_to_local_vm_fails() {
>+ local oops_before warn_before
>+ local ns0="global0"
>+ local ns1="local0"
>+ local port=12345
>+ local dmesg_rc
>+ local pidfile
>+ local result
>+ local pid
>+
>+ init_namespaces
>+
>+ outfile=$(mktemp)
>+
>+ pidfile="$(create_pidfile)"
>+ if ! vm_start "${pidfile}" "${ns1}"; then
>+ log_host "failed to start vm (cid=${VSOCK_CID}, ns=${ns0})"
>+ return "${KSFT_FAIL}"
>+ fi
>+
>+ vm_wait_for_ssh "${ns1}"
>+ oops_before=$(vm_dmesg_oops_count "${ns1}")
>+ warn_before=$(vm_dmesg_warn_count "${ns1}")
>+
>+ vm_ssh "${ns1}" -- socat VSOCK-LISTEN:"${port}" STDOUT > "${outfile}" &
>+ vm_wait_for_listener "${ns1}" "${port}" "vsock"
>+ echo TEST | ip netns exec "${ns0}" \
>+ socat STDIN VSOCK-CONNECT:"${VSOCK_CID}":"${port}" 2>/dev/null
>+
>+ vm_dmesg_check "${pidfile}" "${ns1}" "${oops_before}" "${warn_before}"
>+ dmesg_rc=$?
>+
>+ terminate_pidfiles "${pidfile}"
>+ result=$(cat "${outfile}")
>+ rm -f "${outfile}"
>+
>+ if [[ "${result}" == "TEST" ]] || [[ "${dmesg_rc}" -ne 0 ]]; then
>+ return "${KSFT_FAIL}"
>+ fi
>+
>+ return "${KSFT_PASS}"
>+}
>+
>+test_ns_diff_global_vm_connect_to_global_host_ok() {
>+ local oops_before warn_before
>+ local ns0="global0"
>+ local ns1="global1"
>+ local port=12345
>+ local unixfile
>+ local dmesg_rc
>+ local pidfile
>+ local pids
>+ local rc
>+
>+ init_namespaces
>+
>+ declare -a pids
>+
>+ log_host "Setup socat bridge from ns ${ns0} to ns ${ns1} over port ${port}"
>+
>+ unixfile=$(mktemp -u /tmp/XXXX.sock)
>+
>+ ip netns exec "${ns0}" \
>+ socat TCP-LISTEN:"${port}" UNIX-CONNECT:"${unixfile}" &
>+ pids+=($!)
>+ host_wait_for_listener "${ns0}" "${port}" "tcp"
>+
>+ ip netns exec "${ns1}" \
>+ socat UNIX-LISTEN:"${unixfile}" TCP-CONNECT:127.0.0.1:"${port}" &
>+ pids+=($!)
>+ host_wait_for_listener "${ns1}" "${unixfile}" "unix"
>+
>+ log_host "Launching ${VSOCK_TEST} in ns ${ns1}"
>+ host_vsock_test "${ns1}" "server" "${VSOCK_CID}" "${port}"
>+
>+ pidfile="$(create_pidfile)"
>+ if ! vm_start "${pidfile}" "${ns0}"; then
>+ log_host "failed to start vm (cid=${cid}, ns=${ns0})"
>+ terminate_pids "${pids[@]}"
>+ rm -f "${unixfile}"
>+ return "${KSFT_FAIL}"
>+ fi
>+
>+ vm_wait_for_ssh "${ns0}"
>+
>+ oops_before=$(vm_dmesg_oops_count "${ns0}")
>+ warn_before=$(vm_dmesg_warn_count "${ns0}")
>+
>+ vm_vsock_test "${ns0}" "10.0.2.2" 2 "${port}"
>+ rc=$?
>+
>+ vm_dmesg_check "${pidfile}" "${ns0}" "${oops_before}" "${warn_before}"
>+ dmesg_rc=$?
>+
>+ terminate_pidfiles "${pidfile}"
>+ terminate_pids "${pids[@]}"
>+ rm -f "${unixfile}"
>+
>+ if [[ "${rc}" -ne 0 ]] || [[ "${dmesg_rc}" -ne 0 ]]; then
>+ return "${KSFT_FAIL}"
>+ fi
>+
>+ return "${KSFT_PASS}"
>+
>+}
>+
>+test_ns_diff_global_vm_connect_to_local_host_fails() {
>+ local ns0="global0"
>+ local ns1="local0"
>+ local port=12345
>+ local oops_before warn_before
>+ local dmesg_rc
>+ local pidfile
>+ local result
>+ local pid
>+
>+ init_namespaces
>+
>+ log_host "Launching socat in ns ${ns1}"
>+ outfile=$(mktemp)
>+
>+ ip netns exec "${ns1}" socat VSOCK-LISTEN:"${port}" STDOUT &> "${outfile}" &
>+ pid=$!
>+ host_wait_for_listener "${ns1}" "${port}" "vsock"
>+
>+ pidfile="$(create_pidfile)"
>+ if ! vm_start "${pidfile}" "${ns0}"; then
>+ log_host "failed to start vm (cid=${cid}, ns=${ns0})"
>+ terminate_pids "${pid}"
>+ rm -f "${outfile}"
>+ return "${KSFT_FAIL}"
>+ fi
>+
>+ vm_wait_for_ssh "${ns0}"
>+
>+ oops_before=$(vm_dmesg_oops_count "${ns0}")
>+ warn_before=$(vm_dmesg_warn_count "${ns0}")
>+
>+ vm_ssh "${ns0}" -- \
>+ bash -c "echo TEST | socat STDIN VSOCK-CONNECT:2:${port}" 2>&1 | log_guest
>+
>+ vm_dmesg_check "${pidfile}" "${ns0}" "${oops_before}" "${warn_before}"
>+ dmesg_rc=$?
>+
>+ terminate_pidfiles "${pidfile}"
>+ terminate_pids "${pid}"
>+
>+ result=$(cat "${outfile}")
>+ rm -f "${outfile}"
>+
>+ if [[ "${result}" != TEST ]] && [[ "${dmesg_rc}" -eq 0 ]]; then
>+ return "${KSFT_PASS}"
>+ fi
>+
>+ return "${KSFT_FAIL}"
>+}
>+
>+test_ns_diff_local_host_connect_to_local_vm_fails() {
>+ local ns0="local0"
>+ local ns1="local1"
>+ local port=12345
>+ local oops_before warn_before
>+ local dmesg_rc
>+ local pidfile
>+ local result
>+ local pid
>+
>+ init_namespaces
>+
>+ outfile=$(mktemp)
>+
>+ pidfile="$(create_pidfile)"
>+ if ! vm_start "${pidfile}" "${ns1}"; then
>+ log_host "failed to start vm (cid=${cid}, ns=${ns0})"
>+ return "${KSFT_FAIL}"
>+ fi
>+
>+ vm_wait_for_ssh "${ns1}"
>+ oops_before=$(vm_dmesg_oops_count "${ns1}")
>+ warn_before=$(vm_dmesg_warn_count "${ns1}")
>+
>+ vm_ssh "${ns1}" -- socat VSOCK-LISTEN:"${port}" STDOUT > "${outfile}" &
>+ vm_wait_for_listener "${ns1}" "${port}" "vsock"
>+
>+ echo TEST | ip netns exec "${ns0}" \
>+ socat STDIN VSOCK-CONNECT:"${VSOCK_CID}":"${port}" 2>/dev/null
>+
>+ vm_dmesg_check "${pidfile}" "${ns1}" "${oops_before}" "${warn_before}"
>+ dmesg_rc=$?
>+
>+ terminate_pidfiles "${pidfile}"
>+
>+ result=$(cat "${outfile}")
>+ rm -f "${outfile}"
>+
>+ if [[ "${result}" != TEST ]] && [[ "${dmesg_rc}" -eq 0 ]]; then
>+ return "${KSFT_PASS}"
>+ fi
>+
>+ return "${KSFT_FAIL}"
>+}
>+
>+test_ns_diff_local_vm_connect_to_local_host_fails() {
>+ local oops_before warn_before
>+ local ns0="local0"
>+ local ns1="local1"
>+ local port=12345
>+ local dmesg_rc
>+ local pidfile
>+ local result
>+ local pid
>+
>+ init_namespaces
>+
>+ log_host "Launching socat in ns ${ns1}"
>+ outfile=$(mktemp)
>+ ip netns exec "${ns1}" socat VSOCK-LISTEN:"${port}" STDOUT &> "${outfile}" &
>+ pid=$!
>+ host_wait_for_listener "${ns1}" "${port}" "vsock"
>+
>+ pidfile="$(create_pidfile)"
>+ if ! vm_start "${pidfile}" "${ns0}"; then
>+ log_host "failed to start vm (cid=${cid}, ns=${ns0})"
>+ rm -f "${outfile}"
>+ return "${KSFT_FAIL}"
>+ fi
>+
>+ vm_wait_for_ssh "${ns0}"
>+ oops_before=$(vm_dmesg_oops_count "${ns0}")
>+ warn_before=$(vm_dmesg_warn_count "${ns0}")
>+
>+ vm_ssh "${ns0}" -- \
>+ bash -c "echo TEST | socat STDIN VSOCK-CONNECT:2:${port}" 2>&1 | log_guest
>+
>+ vm_dmesg_check "${pidfile}" "${ns0}" "${oops_before}" "${warn_before}"
>+ dmesg_rc=$?
>+
>+ terminate_pidfiles "${pidfile}"
>+ terminate_pids "${pid}"
>+
>+ result=$(cat "${outfile}")
>+ rm -f "${outfile}"
>+
>+ if [[ "${result}" != TEST ]] && [[ "${dmesg_rc}" -eq 0 ]]; then
>+ return "${KSFT_PASS}"
>+ fi
>+
>+ return "${KSFT_FAIL}"
>+}
>+
>+__test_loopback_two_netns() {
>+ local ns0=$1
>+ local ns1=$2
>+ local port=12345
>+ local result
>+ local pid
>+
>+ modprobe vsock_loopback &> /dev/null || :
>+
>+ log_host "Launching socat in ns ${ns1}"
>+ outfile=$(mktemp)
>+
>+ ip netns exec "${ns1}" socat VSOCK-LISTEN:"${port}" STDOUT > "${outfile}" 2>/dev/null &
>+ pid=$!
>+ host_wait_for_listener "${ns1}" "${port}" "vsock"
>+
>+ log_host "Launching socat in ns ${ns0}"
>+ echo TEST | ip netns exec "${ns0}" socat STDIN VSOCK-CONNECT:1:"${port}" 2>/dev/null
>+ terminate_pids "${pid}"
>+
>+ result=$(cat "${outfile}")
>+ rm -f "${outfile}"
>+
>+ if [[ "${result}" == TEST ]]; then
>+ return 0
>+ fi
>+
>+ return 1
>+}
>+
>+test_ns_diff_global_to_local_loopback_local_fails() {
>+ init_namespaces
>+
>+ if ! __test_loopback_two_netns "global0" "local0"; then
>+ return "${KSFT_PASS}"
>+ fi
>+
>+ return "${KSFT_FAIL}"
>+}
>+
>+test_ns_diff_local_to_global_loopback_fails() {
>+ init_namespaces
>+
>+ if ! __test_loopback_two_netns "local0" "global0"; then
>+ return "${KSFT_PASS}"
>+ fi
>+
>+ return "${KSFT_FAIL}"
>+}
>+
>+test_ns_diff_local_to_local_loopback_fails() {
>+ init_namespaces
>+
>+ if ! __test_loopback_two_netns "local0" "local1"; then
>+ return "${KSFT_PASS}"
>+ fi
>+
>+ return "${KSFT_FAIL}"
>+}
>+
>+test_ns_diff_global_to_global_loopback_ok() {
>+ init_namespaces
>+
>+ if __test_loopback_two_netns "global0" "global1"; then
>+ return "${KSFT_PASS}"
>+ fi
>+
>+ return "${KSFT_FAIL}"
>+}
>+
>+test_ns_same_local_loopback_ok() {
>+ init_namespaces
>+
>+ if __test_loopback_two_netns "local0" "local0"; then
>+ return "${KSFT_PASS}"
>+ fi
>+
>+ return "${KSFT_FAIL}"
>+}
>+
>+test_ns_same_local_host_connect_to_local_vm_ok() {
>+ local oops_before warn_before
>+ local ns="local0"
>+ local port=1234
>+ local dmesg_rc
>+ local pidfile
>+ local rc
>+
>+ init_namespaces
>+
>+ pidfile="$(create_pidfile)"
>+
>+ if ! vm_start "${pidfile}" "${ns}"; then
>+ return "${KSFT_FAIL}"
>+ fi
>+
>+ vm_wait_for_ssh "${ns}"
>+ oops_before=$(vm_dmesg_oops_count "${ns}")
>+ warn_before=$(vm_dmesg_warn_count "${ns}")
>+
>+ vm_vsock_test "${ns}" "server" 2 "${TEST_GUEST_PORT}"
>+ host_vsock_test "${ns}" "127.0.0.1" "${VSOCK_CID}" "${TEST_HOST_PORT}"
>+ rc=$?
>+
>+ vm_dmesg_check "${pidfile}" "${ns}" "${oops_before}" "${warn_before}"
>+ dmesg_rc=$?
>+
>+ terminate_pidfiles "${pidfile}"
>+
>+ if [[ "${rc}" -ne 0 ]] || [[ "${dmesg_rc}" -ne 0 ]]; then
>+ return "${KSFT_FAIL}"
>+ fi
>+
>+ return "${KSFT_PASS}"
>+}
>+
>+test_ns_same_local_vm_connect_to_local_host_ok() {
>+ local oops_before warn_before
>+ local ns="local0"
>+ local port=1234
>+ local dmesg_rc
>+ local pidfile
>+ local rc
>+
>+ init_namespaces
>+
>+ pidfile="$(create_pidfile)"
>+
>+ if ! vm_start "${pidfile}" "${ns}"; then
>+ return "${KSFT_FAIL}"
>+ fi
>+
>+ vm_wait_for_ssh "${ns}"
>+ oops_before=$(vm_dmesg_oops_count "${ns}")
>+ warn_before=$(vm_dmesg_warn_count "${ns}")
>+
>+ host_vsock_test "${ns}" "server" "${VSOCK_CID}" "${port}"
>+ vm_vsock_test "${ns}" "10.0.2.2" 2 "${port}"
>+ rc=$?
>+
>+ vm_dmesg_check "${pidfile}" "${ns}" "${oops_before}" "${warn_before}"
>+ dmesg_rc=$?
>+
>+ terminate_pidfiles "${pidfile}"
>+
>+ if [[ "${rc}" -ne 0 ]] || [[ "${dmesg_rc}" -ne 0 ]]; then
>+ return "${KSFT_FAIL}"
>+ fi
>+
>+ return "${KSFT_PASS}"
>+}
>+
> namespaces_can_boot_same_cid() {
> local ns0=$1
> local ns1=$2
>@@ -894,6 +1446,7 @@ fi
> check_args "${ARGS[@]}"
> check_deps
> check_vng
>+check_socat
> handle_build
>
> echo "1..${#ARGS[@]}"
>
>--
>2.47.3
>
^ permalink raw reply
* Re: [PATCH net-next v11 13/13] selftests/vsock: add tests for namespace deletion and mode changes
From: Stefano Garzarella @ 2025-11-21 14:53 UTC (permalink / raw)
To: Bobby Eshleman
Cc: David S. Miller, Eric Dumazet, Jakub Kicinski, Paolo Abeni,
Simon Horman, Stefan Hajnoczi, Michael S. Tsirkin, Jason Wang,
Eugenio Pérez, Xuan Zhuo, K. Y. Srinivasan, Haiyang Zhang,
Wei Liu, Dexuan Cui, Bryan Tan, Vishnu Dasa,
Broadcom internal kernel review list, Shuah Khan, linux-kernel,
virtualization, netdev, kvm, linux-hyperv, linux-kselftest,
berrange, Sargun Dhillon, Bobby Eshleman
In-Reply-To: <20251120-vsock-vmtest-v11-13-55cbc80249a7@meta.com>
On Thu, Nov 20, 2025 at 09:44:45PM -0800, Bobby Eshleman wrote:
>From: Bobby Eshleman <bobbyeshleman@meta.com>
>
>Add tests that validate vsock sockets are resilient to deleting
>namespaces or changing namespace modes from global to local. The vsock
>sockets should still function normally.
>
>The function check_ns_changes_dont_break_connection() is added to re-use
>the step-by-step logic of 1) setup connections, 2) do something that
>would maybe break the connections, 3) check that the connections are
>still ok.
>
>Signed-off-by: Bobby Eshleman <bobbyeshleman@meta.com>
>---
>Changes in v11:
>- remove pipefile (Stefano)
>
>Changes in v9:
>- more consistent shell style
>- clarify -u usage comment for pipefile
>---
> tools/testing/selftests/vsock/vmtest.sh | 119 ++++++++++++++++++++++++++++++++
> 1 file changed, 119 insertions(+)
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
>
>diff --git a/tools/testing/selftests/vsock/vmtest.sh b/tools/testing/selftests/vsock/vmtest.sh
>index dfa895abfc7f..5f0b24845fad 100755
>--- a/tools/testing/selftests/vsock/vmtest.sh
>+++ b/tools/testing/selftests/vsock/vmtest.sh
>@@ -69,6 +69,12 @@ readonly TEST_NAMES=(
> ns_same_local_loopback_ok
> ns_same_local_host_connect_to_local_vm_ok
> ns_same_local_vm_connect_to_local_host_ok
>+ ns_mode_change_connection_continue_vm_ok
>+ ns_mode_change_connection_continue_host_ok
>+ ns_mode_change_connection_continue_both_ok
>+ ns_delete_vm_ok
>+ ns_delete_host_ok
>+ ns_delete_both_ok
> )
> readonly TEST_DESCS=(
> # vm_server_host_client
>@@ -139,6 +145,24 @@ readonly TEST_DESCS=(
>
> # ns_same_local_vm_connect_to_local_host_ok
> "Run vsock_test client in VM in a local ns with server in same ns."
>+
>+ # ns_mode_change_connection_continue_vm_ok
>+ "Check that changing NS mode of VM namespace from global to local after a connection is established doesn't break the connection"
>+
>+ # ns_mode_change_connection_continue_host_ok
>+ "Check that changing NS mode of host namespace from global to local after a connection is established doesn't break the connection"
>+
>+ # ns_mode_change_connection_continue_both_ok
>+ "Check that changing NS mode of host and VM namespaces from global to local after a connection is established doesn't break the connection"
>+
>+ # ns_delete_vm_ok
>+ "Check that deleting the VM's namespace does not break the socket connection"
>+
>+ # ns_delete_host_ok
>+ "Check that deleting the host's namespace does not break the socket connection"
>+
>+ # ns_delete_both_ok
>+ "Check that deleting the VM and host's namespaces does not break the socket connection"
> )
>
> readonly USE_SHARED_VM=(
>@@ -1288,6 +1312,101 @@ test_ns_vm_local_mode_rejected() {
> return "${KSFT_PASS}"
> }
>
>+check_ns_changes_dont_break_connection() {
>+ local pipefile pidfile outfile
>+ local ns0="global0"
>+ local ns1="global1"
>+ local port=12345
>+ local pids=()
>+ local rc=0
>+
>+ init_namespaces
>+
>+ pidfile="$(create_pidfile)"
>+ if ! vm_start "${pidfile}" "${ns0}"; then
>+ return "${KSFT_FAIL}"
>+ fi
>+ vm_wait_for_ssh "${ns0}"
>+
>+ outfile=$(mktemp)
>+ vm_ssh "${ns0}" -- \
>+ socat VSOCK-LISTEN:"${port}",fork STDOUT > "${outfile}" 2>/dev/null &
>+ pids+=($!)
>+ vm_wait_for_listener "${ns0}" "${port}" "vsock"
>+
>+ # We use a pipe here so that we can echo into the pipe instead of using
>+ # socat and a unix socket file. We just need a name for the pipe (not a
>+ # regular file) so use -u.
>+ pipefile=$(mktemp -u /tmp/vmtest_pipe_XXXX)
>+ ip netns exec "${ns1}" \
>+ socat PIPE:"${pipefile}" VSOCK-CONNECT:"${VSOCK_CID}":"${port}" &
>+ pids+=($!)
>+
>+ timeout "${WAIT_PERIOD}" \
>+ bash -c 'while [[ ! -e '"${pipefile}"' ]]; do sleep 1; done; exit 0'
>+
>+ if [[ $2 == "delete" ]]; then
>+ if [[ "$1" == "vm" ]]; then
>+ ip netns del "${ns0}"
>+ elif [[ "$1" == "host" ]]; then
>+ ip netns del "${ns1}"
>+ elif [[ "$1" == "both" ]]; then
>+ ip netns del "${ns0}"
>+ ip netns del "${ns1}"
>+ fi
>+ elif [[ $2 == "change_mode" ]]; then
>+ if [[ "$1" == "vm" ]]; then
>+ ns_set_mode "${ns0}" "local"
>+ elif [[ "$1" == "host" ]]; then
>+ ns_set_mode "${ns1}" "local"
>+ elif [[ "$1" == "both" ]]; then
>+ ns_set_mode "${ns0}" "local"
>+ ns_set_mode "${ns1}" "local"
>+ fi
>+ fi
>+
>+ echo "TEST" > "${pipefile}"
>+
>+ timeout "${WAIT_PERIOD}" \
>+ bash -c 'while [[ ! -s '"${outfile}"' ]]; do sleep 1; done; exit 0'
>+
>+ if grep -q "TEST" "${outfile}"; then
>+ rc="${KSFT_PASS}"
>+ else
>+ rc="${KSFT_FAIL}"
>+ fi
>+
>+ terminate_pidfiles "${pidfile}"
>+ terminate_pids "${pids[@]}"
>+ rm -f "${outfile}" "${pipefile}"
>+
>+ return "${rc}"
>+}
>+
>+test_ns_mode_change_connection_continue_vm_ok() {
>+ check_ns_changes_dont_break_connection "vm" "change_mode"
>+}
>+
>+test_ns_mode_change_connection_continue_host_ok() {
>+ check_ns_changes_dont_break_connection "host" "change_mode"
>+}
>+
>+test_ns_mode_change_connection_continue_both_ok() {
>+ check_ns_changes_dont_break_connection "both" "change_mode"
>+}
>+
>+test_ns_delete_vm_ok() {
>+ check_ns_changes_dont_break_connection "vm" "delete"
>+}
>+
>+test_ns_delete_host_ok() {
>+ check_ns_changes_dont_break_connection "host" "delete"
>+}
>+
>+test_ns_delete_both_ok() {
>+ check_ns_changes_dont_break_connection "both" "delete"
>+}
>+
> shared_vm_test() {
> local tname
>
>
>--
>2.47.3
>
^ permalink raw reply
* Re: [PATCH 0/6] arch,sysfb: Move screen and edid info into single place
From: Arnd Bergmann @ 2025-11-21 15:09 UTC (permalink / raw)
To: Thomas Zimmermann, Ard Biesheuvel, Javier Martinez Canillas
Cc: x86, linux-arm-kernel, linux-kernel, linux-efi, loongarch,
linux-riscv, dri-devel, linux-hyperv, linux-pci, linux-fbdev
In-Reply-To: <20251121135624.494768-1-tzimmermann@suse.de>
On Fri, Nov 21, 2025, at 14:36, Thomas Zimmermann wrote:
> Replace screen_info and edid_info with sysfb_primary_device of type
> struct sysfb_display_info. Update all users.
>
> Sysfb DRM drivers currently fetch the global edid_info directly, when
> they should get that information together with the screen_info from their
> device. Wrapping screen_info and edid_info in sysfb_primary_display and
> passing this to drivers enables this.
>
> Replacing both with sysfb_primary_display has been motivate by the EFI
> stub. EFI wants to transfer EDID via config table in a single entry.
> Using struct sysfb_display_info this will become easily possible. Hence
> accept some churn in architecture code for the long-term improvements.
This all looks good to me,
Acked-by: Arnd Bergmann <arnd@arndb.de>
It should also bring us one step closer to eventually
disconnecting the x86 boot ABI from the kernel-internal
sysfb_primary_display.
Arnd
^ permalink raw reply
* Re: [PATCH 0/6] arch,sysfb: Move screen and edid info into single place
From: Ard Biesheuvel @ 2025-11-21 15:16 UTC (permalink / raw)
To: Arnd Bergmann
Cc: Thomas Zimmermann, Javier Martinez Canillas, x86,
linux-arm-kernel, linux-kernel, linux-efi, loongarch, linux-riscv,
dri-devel, linux-hyperv, linux-pci, linux-fbdev
In-Reply-To: <96a8d591-29d5-4764-94f9-6042252e53ff@app.fastmail.com>
On Fri, 21 Nov 2025 at 16:10, Arnd Bergmann <arnd@arndb.de> wrote:
>
> On Fri, Nov 21, 2025, at 14:36, Thomas Zimmermann wrote:
> > Replace screen_info and edid_info with sysfb_primary_device of type
> > struct sysfb_display_info. Update all users.
> >
> > Sysfb DRM drivers currently fetch the global edid_info directly, when
> > they should get that information together with the screen_info from their
> > device. Wrapping screen_info and edid_info in sysfb_primary_display and
> > passing this to drivers enables this.
> >
> > Replacing both with sysfb_primary_display has been motivate by the EFI
> > stub. EFI wants to transfer EDID via config table in a single entry.
> > Using struct sysfb_display_info this will become easily possible. Hence
> > accept some churn in architecture code for the long-term improvements.
>
> This all looks good to me,
>
> Acked-by: Arnd Bergmann <arnd@arndb.de>
>
> It should also bring us one step closer to eventually
> disconnecting the x86 boot ABI from the kernel-internal
> sysfb_primary_display.
>
Agreed
Acked-by: Ard Biesheuvel <ardb@kernel.org>
I can take patches 1-2 right away, if that helps during the next cycle.
^ permalink raw reply
* Re: [PATCH 0/6] arch, sysfb: Move screen and edid info into single place
From: Thomas Zimmermann @ 2025-11-21 15:53 UTC (permalink / raw)
To: Ard Biesheuvel, Arnd Bergmann
Cc: Javier Martinez Canillas, x86, linux-arm-kernel, linux-kernel,
linux-efi, loongarch, linux-riscv, dri-devel, linux-hyperv,
linux-pci, linux-fbdev
In-Reply-To: <CAMj1kXF1Dh0RbuqYc0fhAPf-CM0mdYh8BhenM8-ugKVHfwnhBg@mail.gmail.com>
Hi
Am 21.11.25 um 16:16 schrieb Ard Biesheuvel:
> On Fri, 21 Nov 2025 at 16:10, Arnd Bergmann <arnd@arndb.de> wrote:
>> On Fri, Nov 21, 2025, at 14:36, Thomas Zimmermann wrote:
>>> Replace screen_info and edid_info with sysfb_primary_device of type
>>> struct sysfb_display_info. Update all users.
>>>
>>> Sysfb DRM drivers currently fetch the global edid_info directly, when
>>> they should get that information together with the screen_info from their
>>> device. Wrapping screen_info and edid_info in sysfb_primary_display and
>>> passing this to drivers enables this.
>>>
>>> Replacing both with sysfb_primary_display has been motivate by the EFI
>>> stub. EFI wants to transfer EDID via config table in a single entry.
>>> Using struct sysfb_display_info this will become easily possible. Hence
>>> accept some churn in architecture code for the long-term improvements.
>> This all looks good to me,
>>
>> Acked-by: Arnd Bergmann <arnd@arndb.de>
Thanks
>>
>> It should also bring us one step closer to eventually
>> disconnecting the x86 boot ABI from the kernel-internal
>> sysfb_primary_display.
>>
> Agreed
>
> Acked-by: Ard Biesheuvel <ardb@kernel.org>
Thanks
>
> I can take patches 1-2 right away, if that helps during the next cycle.
From my sysfb-focused POV, these patches would ideally all go through
the same tree, say efi or generic arch, or whatever fits best. Most of
the other code is only renames anyway.
Best regards
Thomas
--
--
Thomas Zimmermann
Graphics Driver Developer
SUSE Software Solutions Germany GmbH
Frankenstr. 146, 90461 Nürnberg, Germany, www.suse.com
GF: Jochen Jaser, Andrew McDonald, Werner Knoblich, (HRB 36809, AG Nürnberg)
^ permalink raw reply
* Re: [PATCH 6/6] sysfb: Move edid_info into sysfb_primary_display
From: Thomas Zimmermann @ 2025-11-21 15:56 UTC (permalink / raw)
To: ardb, javierm, arnd
Cc: x86, linux-arm-kernel, linux-kernel, linux-efi, loongarch,
linux-riscv, dri-devel, linux-hyperv, linux-pci, linux-fbdev
In-Reply-To: <20251121135624.494768-7-tzimmermann@suse.de>
Am 21.11.25 um 14:36 schrieb Thomas Zimmermann:
> Move x86's edid_info into sysfb_primary_display as a new field named
> edid. Adapt all users.
>
> An instance of edid_info has only been defined on x86. With the move
> into sysfb_primary_display, it becomes available on all architectures.
> Therefore remove this contraint from CONFIG_FIRMWARE_EDID.
s/contraint/constraint/
>
> x86 fills the EDID data from boot_params.edid_info. DRM drivers pick
> up the raw data and make it available to DRM clients. Replace the
> drivers' references to edid_info and instead use the sysfb_display_info
> as passed from sysfb.
>
> Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
> ---
> arch/x86/kernel/setup.c | 6 +-----
> drivers/gpu/drm/sysfb/efidrm.c | 5 ++---
> drivers/gpu/drm/sysfb/vesadrm.c | 5 ++---
> drivers/video/Kconfig | 1 -
> drivers/video/fbdev/core/fbmon.c | 8 +++++---
> include/linux/sysfb.h | 6 ++++++
> include/video/edid.h | 4 ----
> 7 files changed, 16 insertions(+), 19 deletions(-)
>
> diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
> index 675e4b9deb1f..d9bfe2032cd9 100644
> --- a/arch/x86/kernel/setup.c
> +++ b/arch/x86/kernel/setup.c
> @@ -215,10 +215,6 @@ arch_initcall(init_x86_sysctl);
>
> struct sysfb_display_info sysfb_primary_display;
> EXPORT_SYMBOL(sysfb_primary_display);
> -#if defined(CONFIG_FIRMWARE_EDID)
> -struct edid_info edid_info;
> -EXPORT_SYMBOL_GPL(edid_info);
> -#endif
>
> extern int root_mountflags;
>
> @@ -530,7 +526,7 @@ static void __init parse_boot_params(void)
> ROOT_DEV = old_decode_dev(boot_params.hdr.root_dev);
> sysfb_primary_display.screen = boot_params.screen_info;
> #if defined(CONFIG_FIRMWARE_EDID)
> - edid_info = boot_params.edid_info;
> + sysfb_primary_display.edid = boot_params.edid_info;
> #endif
> #ifdef CONFIG_X86_32
> apm_info.bios = boot_params.apm_bios_info;
> diff --git a/drivers/gpu/drm/sysfb/efidrm.c b/drivers/gpu/drm/sysfb/efidrm.c
> index 29533ae8fbbf..50e0aeef709c 100644
> --- a/drivers/gpu/drm/sysfb/efidrm.c
> +++ b/drivers/gpu/drm/sysfb/efidrm.c
> @@ -24,7 +24,6 @@
> #include <drm/drm_print.h>
> #include <drm/drm_probe_helper.h>
>
> -#include <video/edid.h>
> #include <video/pixel_format.h>
>
> #include "drm_sysfb_helper.h"
> @@ -207,8 +206,8 @@ static struct efidrm_device *efidrm_device_create(struct drm_driver *drv,
> &format->format, width, height, stride);
>
> #if defined(CONFIG_FIRMWARE_EDID)
> - if (drm_edid_header_is_valid(edid_info.dummy) == 8)
> - sysfb->edid = edid_info.dummy;
> + if (drm_edid_header_is_valid(dpy->edid.dummy) == 8)
> + sysfb->edid = dpy->edid.dummy;
> #endif
> sysfb->fb_mode = drm_sysfb_mode(width, height, 0, 0);
> sysfb->fb_format = format;
> diff --git a/drivers/gpu/drm/sysfb/vesadrm.c b/drivers/gpu/drm/sysfb/vesadrm.c
> index 16fc223f8c5b..0680638b8131 100644
> --- a/drivers/gpu/drm/sysfb/vesadrm.c
> +++ b/drivers/gpu/drm/sysfb/vesadrm.c
> @@ -25,7 +25,6 @@
> #include <drm/drm_print.h>
> #include <drm/drm_probe_helper.h>
>
> -#include <video/edid.h>
> #include <video/pixel_format.h>
> #include <video/vga.h>
>
> @@ -474,8 +473,8 @@ static struct vesadrm_device *vesadrm_device_create(struct drm_driver *drv,
> }
>
> #if defined(CONFIG_FIRMWARE_EDID)
> - if (drm_edid_header_is_valid(edid_info.dummy) == 8)
> - sysfb->edid = edid_info.dummy;
> + if (drm_edid_header_is_valid(dpy->edid.dummy) == 8)
> + sysfb->edid = dpy->edid.dummy;
> #endif
> sysfb->fb_mode = drm_sysfb_mode(width, height, 0, 0);
> sysfb->fb_format = format;
> diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig
> index d51777df12d1..ad55e7d62159 100644
> --- a/drivers/video/Kconfig
> +++ b/drivers/video/Kconfig
> @@ -63,7 +63,6 @@ endif # HAS_IOMEM
>
> config FIRMWARE_EDID
> bool "Enable firmware EDID"
> - depends on X86
> help
> This enables access to the EDID transferred from the firmware.
> On x86, this is from the VESA BIOS. DRM display drivers will
> diff --git a/drivers/video/fbdev/core/fbmon.c b/drivers/video/fbdev/core/fbmon.c
> index 0a65bef01e3c..07df7e98f8a3 100644
> --- a/drivers/video/fbdev/core/fbmon.c
> +++ b/drivers/video/fbdev/core/fbmon.c
> @@ -32,11 +32,13 @@
> #include <linux/module.h>
> #include <linux/pci.h>
> #include <linux/slab.h>
> -#include <video/edid.h>
> +#include <linux/string_choices.h>
> +#include <linux/sysfb.h>
> +
> #include <video/of_videomode.h>
> #include <video/videomode.h>
> +
> #include "../edid.h"
> -#include <linux/string_choices.h>
>
> /*
> * EDID parser
> @@ -1504,7 +1506,7 @@ const unsigned char *fb_firmware_edid(struct device *device)
> res = &dev->resource[PCI_ROM_RESOURCE];
>
> if (res && res->flags & IORESOURCE_ROM_SHADOW)
> - edid = edid_info.dummy;
> + edid = sysfb_primary_display.edid.dummy;
>
> return edid;
> }
> diff --git a/include/linux/sysfb.h b/include/linux/sysfb.h
> index e8bde392c690..5226efde9ad4 100644
> --- a/include/linux/sysfb.h
> +++ b/include/linux/sysfb.h
> @@ -12,6 +12,8 @@
> #include <linux/screen_info.h>
> #include <linux/types.h>
>
> +#include <video/edid.h>
> +
> struct device;
> struct platform_device;
> struct screen_info;
> @@ -62,6 +64,10 @@ struct efifb_dmi_info {
>
> struct sysfb_display_info {
> struct screen_info screen;
> +
> +#if defined(CONFIG_FIRMWARE_EDID)
> + struct edid_info edid;
> +#endif
> };
>
> extern struct sysfb_display_info sysfb_primary_display;
> diff --git a/include/video/edid.h b/include/video/edid.h
> index c2b186b1933a..52aabb706032 100644
> --- a/include/video/edid.h
> +++ b/include/video/edid.h
> @@ -4,8 +4,4 @@
>
> #include <uapi/video/edid.h>
>
> -#if defined(CONFIG_FIRMWARE_EDID)
> -extern struct edid_info edid_info;
> -#endif
> -
> #endif /* __linux_video_edid_h__ */
--
--
Thomas Zimmermann
Graphics Driver Developer
SUSE Software Solutions Germany GmbH
Frankenstr. 146, 90461 Nürnberg, Germany, www.suse.com
GF: Jochen Jaser, Andrew McDonald, Werner Knoblich, (HRB 36809, AG Nürnberg)
^ permalink raw reply
* Re: [PATCH 0/6] arch, sysfb: Move screen and edid info into single place
From: Ard Biesheuvel @ 2025-11-21 15:56 UTC (permalink / raw)
To: Thomas Zimmermann
Cc: Arnd Bergmann, Javier Martinez Canillas, x86, linux-arm-kernel,
linux-kernel, linux-efi, loongarch, linux-riscv, dri-devel,
linux-hyperv, linux-pci, linux-fbdev
In-Reply-To: <199e7538-5b4a-483b-8976-84e4a8a0f2fd@suse.de>
On Fri, 21 Nov 2025 at 16:53, Thomas Zimmermann <tzimmermann@suse.de> wrote:
>
> Hi
>
> Am 21.11.25 um 16:16 schrieb Ard Biesheuvel:
> > On Fri, 21 Nov 2025 at 16:10, Arnd Bergmann <arnd@arndb.de> wrote:
> >> On Fri, Nov 21, 2025, at 14:36, Thomas Zimmermann wrote:
> >>> Replace screen_info and edid_info with sysfb_primary_device of type
> >>> struct sysfb_display_info. Update all users.
> >>>
> >>> Sysfb DRM drivers currently fetch the global edid_info directly, when
> >>> they should get that information together with the screen_info from their
> >>> device. Wrapping screen_info and edid_info in sysfb_primary_display and
> >>> passing this to drivers enables this.
> >>>
> >>> Replacing both with sysfb_primary_display has been motivate by the EFI
> >>> stub. EFI wants to transfer EDID via config table in a single entry.
> >>> Using struct sysfb_display_info this will become easily possible. Hence
> >>> accept some churn in architecture code for the long-term improvements.
> >> This all looks good to me,
> >>
> >> Acked-by: Arnd Bergmann <arnd@arndb.de>
>
> Thanks
>
> >>
> >> It should also bring us one step closer to eventually
> >> disconnecting the x86 boot ABI from the kernel-internal
> >> sysfb_primary_display.
> >>
> > Agreed
> >
> > Acked-by: Ard Biesheuvel <ardb@kernel.org>
>
> Thanks
>
> >
> > I can take patches 1-2 right away, if that helps during the next cycle.
>
> From my sysfb-focused POV, these patches would ideally all go through
> the same tree, say efi or generic arch, or whatever fits best. Most of
> the other code is only renames anyway.
>
I don't mind queueing all of it, but I did get a conflict on
drivers/pci/vgaarb.c
^ permalink raw reply
* Re: [PATCH 0/6] arch, sysfb: Move screen and edid info into single place
From: Thomas Zimmermann @ 2025-11-21 16:08 UTC (permalink / raw)
To: Ard Biesheuvel
Cc: Arnd Bergmann, Javier Martinez Canillas, x86, linux-arm-kernel,
linux-kernel, linux-efi, loongarch, linux-riscv, dri-devel,
linux-hyperv, linux-pci, linux-fbdev
In-Reply-To: <CAMj1kXE+mS1Sm5GaROU0P97J2w1pew0P_To4sKiw8h1iOMuLcg@mail.gmail.com>
Hi
Am 21.11.25 um 16:56 schrieb Ard Biesheuvel:
> On Fri, 21 Nov 2025 at 16:53, Thomas Zimmermann <tzimmermann@suse.de> wrote:
>> Hi
>>
>> Am 21.11.25 um 16:16 schrieb Ard Biesheuvel:
>>> On Fri, 21 Nov 2025 at 16:10, Arnd Bergmann <arnd@arndb.de> wrote:
>>>> On Fri, Nov 21, 2025, at 14:36, Thomas Zimmermann wrote:
>>>>> Replace screen_info and edid_info with sysfb_primary_device of type
>>>>> struct sysfb_display_info. Update all users.
>>>>>
>>>>> Sysfb DRM drivers currently fetch the global edid_info directly, when
>>>>> they should get that information together with the screen_info from their
>>>>> device. Wrapping screen_info and edid_info in sysfb_primary_display and
>>>>> passing this to drivers enables this.
>>>>>
>>>>> Replacing both with sysfb_primary_display has been motivate by the EFI
>>>>> stub. EFI wants to transfer EDID via config table in a single entry.
>>>>> Using struct sysfb_display_info this will become easily possible. Hence
>>>>> accept some churn in architecture code for the long-term improvements.
>>>> This all looks good to me,
>>>>
>>>> Acked-by: Arnd Bergmann <arnd@arndb.de>
>> Thanks
>>
>>>> It should also bring us one step closer to eventually
>>>> disconnecting the x86 boot ABI from the kernel-internal
>>>> sysfb_primary_display.
>>>>
>>> Agreed
>>>
>>> Acked-by: Ard Biesheuvel <ardb@kernel.org>
>> Thanks
>>
>>> I can take patches 1-2 right away, if that helps during the next cycle.
>> From my sysfb-focused POV, these patches would ideally all go through
>> the same tree, say efi or generic arch, or whatever fits best. Most of
>> the other code is only renames anyway.
>>
> I don't mind queueing all of it, but I did get a conflict on
> drivers/pci/vgaarb.c
Probably from a78835b86a44 ("PCI/VGA: Select SCREEN_INFO on X86") that I
have in my tree. You can fix it by replacing screen_info with
sysfb_primary_display.screen. The later merge conflict seems manageable.
Best regards
Thomas
--
--
Thomas Zimmermann
Graphics Driver Developer
SUSE Software Solutions Germany GmbH
Frankenstr. 146, 90461 Nürnberg, Germany, www.suse.com
GF: Jochen Jaser, Andrew McDonald, Werner Knoblich, (HRB 36809, AG Nürnberg)
^ permalink raw reply
* Re: [PATCH 0/6] arch, sysfb: Move screen and edid info into single place
From: Thomas Zimmermann @ 2025-11-21 16:09 UTC (permalink / raw)
To: Ard Biesheuvel
Cc: Arnd Bergmann, Javier Martinez Canillas, x86, linux-arm-kernel,
linux-kernel, linux-efi, loongarch, linux-riscv, dri-devel,
linux-hyperv, linux-pci, linux-fbdev
In-Reply-To: <d080729c-6586-4b9c-b234-470977849d3d@suse.de>
Am 21.11.25 um 17:08 schrieb Thomas Zimmermann:
> Hi
>
> Am 21.11.25 um 16:56 schrieb Ard Biesheuvel:
>> On Fri, 21 Nov 2025 at 16:53, Thomas Zimmermann <tzimmermann@suse.de>
>> wrote:
>>> Hi
>>>
>>> Am 21.11.25 um 16:16 schrieb Ard Biesheuvel:
>>>> On Fri, 21 Nov 2025 at 16:10, Arnd Bergmann <arnd@arndb.de> wrote:
>>>>> On Fri, Nov 21, 2025, at 14:36, Thomas Zimmermann wrote:
>>>>>> Replace screen_info and edid_info with sysfb_primary_device of type
>>>>>> struct sysfb_display_info. Update all users.
>>>>>>
>>>>>> Sysfb DRM drivers currently fetch the global edid_info directly,
>>>>>> when
>>>>>> they should get that information together with the screen_info
>>>>>> from their
>>>>>> device. Wrapping screen_info and edid_info in
>>>>>> sysfb_primary_display and
>>>>>> passing this to drivers enables this.
>>>>>>
>>>>>> Replacing both with sysfb_primary_display has been motivate by
>>>>>> the EFI
>>>>>> stub. EFI wants to transfer EDID via config table in a single entry.
>>>>>> Using struct sysfb_display_info this will become easily possible.
>>>>>> Hence
>>>>>> accept some churn in architecture code for the long-term
>>>>>> improvements.
>>>>> This all looks good to me,
>>>>>
>>>>> Acked-by: Arnd Bergmann <arnd@arndb.de>
>>> Thanks
>>>
>>>>> It should also bring us one step closer to eventually
>>>>> disconnecting the x86 boot ABI from the kernel-internal
>>>>> sysfb_primary_display.
>>>>>
>>>> Agreed
>>>>
>>>> Acked-by: Ard Biesheuvel <ardb@kernel.org>
>>> Thanks
>>>
>>>> I can take patches 1-2 right away, if that helps during the next
>>>> cycle.
>>> From my sysfb-focused POV, these patches would ideally all go through
>>> the same tree, say efi or generic arch, or whatever fits best. Most of
>>> the other code is only renames anyway.
>>>
>> I don't mind queueing all of it, but I did get a conflict on
>> drivers/pci/vgaarb.c
>
> Probably from a78835b86a44 ("PCI/VGA: Select SCREEN_INFO on X86")
https://lore.kernel.org/all/20251013220829.1536292-1-superm1@kernel.org/
> that I have in my tree. You can fix it by replacing screen_info with
> sysfb_primary_display.screen. The later merge conflict seems manageable.
>
> Best regards
> Thomas
>
>
--
--
Thomas Zimmermann
Graphics Driver Developer
SUSE Software Solutions Germany GmbH
Frankenstr. 146, 90461 Nürnberg, Germany, www.suse.com
GF: Jochen Jaser, Andrew McDonald, Werner Knoblich, (HRB 36809, AG Nürnberg)
^ permalink raw reply
* Re: [PATCH 0/6] arch, sysfb: Move screen and edid info into single place
From: Ard Biesheuvel @ 2025-11-21 16:19 UTC (permalink / raw)
To: Thomas Zimmermann
Cc: Arnd Bergmann, Javier Martinez Canillas, x86, linux-arm-kernel,
linux-kernel, linux-efi, loongarch, linux-riscv, dri-devel,
linux-hyperv, linux-pci, linux-fbdev
In-Reply-To: <6dff8e7e-c99b-443d-a1d8-22650ca0b595@suse.de>
On Fri, 21 Nov 2025 at 17:09, Thomas Zimmermann <tzimmermann@suse.de> wrote:
>
>
>
> Am 21.11.25 um 17:08 schrieb Thomas Zimmermann:
> > Hi
> >
> > Am 21.11.25 um 16:56 schrieb Ard Biesheuvel:
> >> On Fri, 21 Nov 2025 at 16:53, Thomas Zimmermann <tzimmermann@suse.de>
> >> wrote:
> >>> Hi
> >>>
> >>> Am 21.11.25 um 16:16 schrieb Ard Biesheuvel:
> >>>> On Fri, 21 Nov 2025 at 16:10, Arnd Bergmann <arnd@arndb.de> wrote:
> >>>>> On Fri, Nov 21, 2025, at 14:36, Thomas Zimmermann wrote:
> >>>>>> Replace screen_info and edid_info with sysfb_primary_device of type
> >>>>>> struct sysfb_display_info. Update all users.
> >>>>>>
> >>>>>> Sysfb DRM drivers currently fetch the global edid_info directly,
> >>>>>> when
> >>>>>> they should get that information together with the screen_info
> >>>>>> from their
> >>>>>> device. Wrapping screen_info and edid_info in
> >>>>>> sysfb_primary_display and
> >>>>>> passing this to drivers enables this.
> >>>>>>
> >>>>>> Replacing both with sysfb_primary_display has been motivate by
> >>>>>> the EFI
> >>>>>> stub. EFI wants to transfer EDID via config table in a single entry.
> >>>>>> Using struct sysfb_display_info this will become easily possible.
> >>>>>> Hence
> >>>>>> accept some churn in architecture code for the long-term
> >>>>>> improvements.
> >>>>> This all looks good to me,
> >>>>>
> >>>>> Acked-by: Arnd Bergmann <arnd@arndb.de>
> >>> Thanks
> >>>
> >>>>> It should also bring us one step closer to eventually
> >>>>> disconnecting the x86 boot ABI from the kernel-internal
> >>>>> sysfb_primary_display.
> >>>>>
> >>>> Agreed
> >>>>
> >>>> Acked-by: Ard Biesheuvel <ardb@kernel.org>
> >>> Thanks
> >>>
> >>>> I can take patches 1-2 right away, if that helps during the next
> >>>> cycle.
> >>> From my sysfb-focused POV, these patches would ideally all go through
> >>> the same tree, say efi or generic arch, or whatever fits best. Most of
> >>> the other code is only renames anyway.
> >>>
> >> I don't mind queueing all of it, but I did get a conflict on
> >> drivers/pci/vgaarb.c
> >
> > Probably from a78835b86a44 ("PCI/VGA: Select SCREEN_INFO on X86")
>
> https://lore.kernel.org/all/20251013220829.1536292-1-superm1@kernel.org/
>
Yes, if I merge back -rc2 first, I can apply patches 1-5 onto my
efi/next tree. But then I hit
Applying: sysfb: Move edid_info into sysfb_primary_display
error: sha1 information is lacking or useless (drivers/gpu/drm/sysfb/efidrm.c).
error: could not build fake ancestor
Patch failed at 0006 sysfb: Move edid_info into sysfb_primary_display
If you prefer, you can take the whole lot via the sysfb tree instead,
assuming it does not depend on the EDID changes I already queued up?
^ permalink raw reply
* Re: [PATCH 0/6] arch, sysfb: Move screen and edid info into single place
From: Thomas Zimmermann @ 2025-11-21 16:26 UTC (permalink / raw)
To: Ard Biesheuvel
Cc: Arnd Bergmann, Javier Martinez Canillas, x86, linux-arm-kernel,
linux-kernel, linux-efi, loongarch, linux-riscv, dri-devel,
linux-hyperv, linux-pci, linux-fbdev
In-Reply-To: <CAMj1kXGpC_162bFL65kQw=7qVP7ezYw77Q76y217dDs8pqHogw@mail.gmail.com>
Hi
Am 21.11.25 um 17:19 schrieb Ard Biesheuvel:
> On Fri, 21 Nov 2025 at 17:09, Thomas Zimmermann <tzimmermann@suse.de> wrote:
>>
>>
>> Am 21.11.25 um 17:08 schrieb Thomas Zimmermann:
>>> Hi
>>>
>>> Am 21.11.25 um 16:56 schrieb Ard Biesheuvel:
>>>> On Fri, 21 Nov 2025 at 16:53, Thomas Zimmermann <tzimmermann@suse.de>
>>>> wrote:
>>>>> Hi
>>>>>
>>>>> Am 21.11.25 um 16:16 schrieb Ard Biesheuvel:
>>>>>> On Fri, 21 Nov 2025 at 16:10, Arnd Bergmann <arnd@arndb.de> wrote:
>>>>>>> On Fri, Nov 21, 2025, at 14:36, Thomas Zimmermann wrote:
>>>>>>>> Replace screen_info and edid_info with sysfb_primary_device of type
>>>>>>>> struct sysfb_display_info. Update all users.
>>>>>>>>
>>>>>>>> Sysfb DRM drivers currently fetch the global edid_info directly,
>>>>>>>> when
>>>>>>>> they should get that information together with the screen_info
>>>>>>>> from their
>>>>>>>> device. Wrapping screen_info and edid_info in
>>>>>>>> sysfb_primary_display and
>>>>>>>> passing this to drivers enables this.
>>>>>>>>
>>>>>>>> Replacing both with sysfb_primary_display has been motivate by
>>>>>>>> the EFI
>>>>>>>> stub. EFI wants to transfer EDID via config table in a single entry.
>>>>>>>> Using struct sysfb_display_info this will become easily possible.
>>>>>>>> Hence
>>>>>>>> accept some churn in architecture code for the long-term
>>>>>>>> improvements.
>>>>>>> This all looks good to me,
>>>>>>>
>>>>>>> Acked-by: Arnd Bergmann <arnd@arndb.de>
>>>>> Thanks
>>>>>
>>>>>>> It should also bring us one step closer to eventually
>>>>>>> disconnecting the x86 boot ABI from the kernel-internal
>>>>>>> sysfb_primary_display.
>>>>>>>
>>>>>> Agreed
>>>>>>
>>>>>> Acked-by: Ard Biesheuvel <ardb@kernel.org>
>>>>> Thanks
>>>>>
>>>>>> I can take patches 1-2 right away, if that helps during the next
>>>>>> cycle.
>>>>> From my sysfb-focused POV, these patches would ideally all go through
>>>>> the same tree, say efi or generic arch, or whatever fits best. Most of
>>>>> the other code is only renames anyway.
>>>>>
>>>> I don't mind queueing all of it, but I did get a conflict on
>>>> drivers/pci/vgaarb.c
>>> Probably from a78835b86a44 ("PCI/VGA: Select SCREEN_INFO on X86")
>> https://lore.kernel.org/all/20251013220829.1536292-1-superm1@kernel.org/
>>
> Yes, if I merge back -rc2 first, I can apply patches 1-5 onto my
> efi/next tree. But then I hit
>
> Applying: sysfb: Move edid_info into sysfb_primary_display
> error: sha1 information is lacking or useless (drivers/gpu/drm/sysfb/efidrm.c).
> error: could not build fake ancestor
> Patch failed at 0006 sysfb: Move edid_info into sysfb_primary_display
>
> If you prefer, you can take the whole lot via the sysfb tree instead,
> assuming it does not depend on the EDID changes I already queued up?
Sure, I can also add it to the drm-misc tree. ETA in upstream would be
v6.20-rc1.
Best regards
Thomas
--
--
Thomas Zimmermann
Graphics Driver Developer
SUSE Software Solutions Germany GmbH
Frankenstr. 146, 90461 Nürnberg, Germany, www.suse.com
GF: Jochen Jaser, Andrew McDonald, Werner Knoblich, (HRB 36809, AG Nürnberg)
^ permalink raw reply
* Re: [PATCH 0/6] arch, sysfb: Move screen and edid info into single place
From: Ard Biesheuvel @ 2025-11-21 16:31 UTC (permalink / raw)
To: Thomas Zimmermann
Cc: Arnd Bergmann, Javier Martinez Canillas, x86, linux-arm-kernel,
linux-kernel, linux-efi, loongarch, linux-riscv, dri-devel,
linux-hyperv, linux-pci, linux-fbdev
In-Reply-To: <8d0bc096-e346-462b-a274-f0cc1456eea3@suse.de>
On Fri, 21 Nov 2025 at 17:26, Thomas Zimmermann <tzimmermann@suse.de> wrote:
>
> Hi
>
> Am 21.11.25 um 17:19 schrieb Ard Biesheuvel:
> > On Fri, 21 Nov 2025 at 17:09, Thomas Zimmermann <tzimmermann@suse.de> wrote:
> >>
> >>
> >> Am 21.11.25 um 17:08 schrieb Thomas Zimmermann:
> >>> Hi
> >>>
> >>> Am 21.11.25 um 16:56 schrieb Ard Biesheuvel:
> >>>> On Fri, 21 Nov 2025 at 16:53, Thomas Zimmermann <tzimmermann@suse.de>
> >>>> wrote:
> >>>>> Hi
> >>>>>
> >>>>> Am 21.11.25 um 16:16 schrieb Ard Biesheuvel:
> >>>>>> On Fri, 21 Nov 2025 at 16:10, Arnd Bergmann <arnd@arndb.de> wrote:
> >>>>>>> On Fri, Nov 21, 2025, at 14:36, Thomas Zimmermann wrote:
> >>>>>>>> Replace screen_info and edid_info with sysfb_primary_device of type
> >>>>>>>> struct sysfb_display_info. Update all users.
> >>>>>>>>
> >>>>>>>> Sysfb DRM drivers currently fetch the global edid_info directly,
> >>>>>>>> when
> >>>>>>>> they should get that information together with the screen_info
> >>>>>>>> from their
> >>>>>>>> device. Wrapping screen_info and edid_info in
> >>>>>>>> sysfb_primary_display and
> >>>>>>>> passing this to drivers enables this.
> >>>>>>>>
> >>>>>>>> Replacing both with sysfb_primary_display has been motivate by
> >>>>>>>> the EFI
> >>>>>>>> stub. EFI wants to transfer EDID via config table in a single entry.
> >>>>>>>> Using struct sysfb_display_info this will become easily possible.
> >>>>>>>> Hence
> >>>>>>>> accept some churn in architecture code for the long-term
> >>>>>>>> improvements.
> >>>>>>> This all looks good to me,
> >>>>>>>
> >>>>>>> Acked-by: Arnd Bergmann <arnd@arndb.de>
> >>>>> Thanks
> >>>>>
> >>>>>>> It should also bring us one step closer to eventually
> >>>>>>> disconnecting the x86 boot ABI from the kernel-internal
> >>>>>>> sysfb_primary_display.
> >>>>>>>
> >>>>>> Agreed
> >>>>>>
> >>>>>> Acked-by: Ard Biesheuvel <ardb@kernel.org>
> >>>>> Thanks
> >>>>>
> >>>>>> I can take patches 1-2 right away, if that helps during the next
> >>>>>> cycle.
> >>>>> From my sysfb-focused POV, these patches would ideally all go through
> >>>>> the same tree, say efi or generic arch, or whatever fits best. Most of
> >>>>> the other code is only renames anyway.
> >>>>>
> >>>> I don't mind queueing all of it, but I did get a conflict on
> >>>> drivers/pci/vgaarb.c
> >>> Probably from a78835b86a44 ("PCI/VGA: Select SCREEN_INFO on X86")
> >> https://lore.kernel.org/all/20251013220829.1536292-1-superm1@kernel.org/
> >>
> > Yes, if I merge back -rc2 first, I can apply patches 1-5 onto my
> > efi/next tree. But then I hit
> >
> > Applying: sysfb: Move edid_info into sysfb_primary_display
> > error: sha1 information is lacking or useless (drivers/gpu/drm/sysfb/efidrm.c).
> > error: could not build fake ancestor
> > Patch failed at 0006 sysfb: Move edid_info into sysfb_primary_display
> >
> > If you prefer, you can take the whole lot via the sysfb tree instead,
> > assuming it does not depend on the EDID changes I already queued up?
>
> Sure, I can also add it to the drm-misc tree. ETA in upstream would be
> v6.20-rc1.
>
But does that mean the EDID firmware on non-x86 will have to wait for
6.21? I was trying to avoid making this a 6 month effort.
^ permalink raw reply
* Re: [PATCH net-next v11 03/13] vsock: reject bad VSOCK_NET_MODE_LOCAL configuration for G2H
From: Bobby Eshleman @ 2025-11-21 19:01 UTC (permalink / raw)
To: Stefano Garzarella
Cc: David S. Miller, Eric Dumazet, Jakub Kicinski, Paolo Abeni,
Simon Horman, Stefan Hajnoczi, Michael S. Tsirkin, Jason Wang,
Eugenio Pérez, Xuan Zhuo, K. Y. Srinivasan, Haiyang Zhang,
Wei Liu, Dexuan Cui, Bryan Tan, Vishnu Dasa,
Broadcom internal kernel review list, Shuah Khan, linux-kernel,
virtualization, netdev, kvm, linux-hyperv, linux-kselftest,
berrange, Sargun Dhillon, Bobby Eshleman
In-Reply-To: <swa5xpovczqucynffqgfotyx34lziccwpqomnm5a7iwmeyixfv@uehtzbdj53b4>
On Fri, Nov 21, 2025 at 03:24:25PM +0100, Stefano Garzarella wrote:
> On Thu, Nov 20, 2025 at 09:44:35PM -0800, Bobby Eshleman wrote:
> > From: Bobby Eshleman <bobbyeshleman@meta.com>
> >
> > Reject setting VSOCK_NET_MODE_LOCAL with -EOPNOTSUPP if a G2H transport
> > is operational. Additionally, reject G2H transport registration if there
> > already exists a namespace in local mode.
> >
> > G2H sockets break in local mode because the G2H transports don't support
> > namespacing yet. The current approach is to coerce packets coming out of
> > G2H transports into VSOCK_NET_MODE_GLOBAL mode, but it is not possible
> > to coerce sockets in the same way because it cannot be deduced which
> > transport will be used by the socket. Specifically, when bound to
> > VMADDR_CID_ANY in a nested VM (both G2H and H2G available), it is not
> > until a packet is received and matched to the bound socket that we
> > assign the transport. This presents a chicken-and-egg problem, because
> > we need the namespace to lookup the socket and resolve the transport,
> > but we need the transport to know how to use the namespace during
> > lookup.
> >
> > For that reason, this patch prevents VSOCK_NET_MODE_LOCAL from being
> > used on systems that support G2H, even nested systems that also have H2G
> > transports.
> >
> > Local mode is blocked based on detecting the presence of G2H devices
> > (when possible, as hyperv is special). This means that a host kernel
> > with G2H support compiled in (or has the module loaded), will still
> > support local mode if there is no G2H (e.g., virtio-vsock) device
> > detected. This enables using the same kernel in the host and in the
> > guest, as we do in kselftest.
> >
> > Systems with only namespace-aware transports (vhost-vsock, loopback) can
> > still use both VSOCK_NET_MODE_GLOBAL and VSOCK_NET_MODE_LOCAL modes as
> > intended.
> >
> > Add supports_local_mode() transport callback to indicate
> > transport-specific local mode support.
> >
> > These restrictions can be lifted in a future patch series when G2H
> > transports gain namespace support.
> >
> > Signed-off-by: Bobby Eshleman <bobbyeshleman@meta.com>
> > ---
> > Changes in v11:
> > - vhost_transport_supports_local_mode() returns false to keep things
> > disabled until support comes online (Stefano)
> > - add comment above supports_local_mode() cb to clarify (Stefano)
> > - Remove redundant `ret = 0` initialization in vsock_net_mode_string()
> > (Stefano)
> > - Refactor vsock_net_mode_string() to separate parsing from validation
> > (Stefano)
> > - vmci returns false for supports_local_mode(), with comment
> >
> > Changes in v10:
> > - move this patch before any transports bring online namespacing (Stefano)
> > - move vsock_net_mode_string into critical section (Stefano)
> > - add ->supports_local_mode() callback to transports (Stefano)
> > ---
> > drivers/vhost/vsock.c | 6 ++++++
> > include/net/af_vsock.h | 11 +++++++++++
> > net/vmw_vsock/af_vsock.c | 32 ++++++++++++++++++++++++++++++++
> > net/vmw_vsock/hyperv_transport.c | 6 ++++++
> > net/vmw_vsock/virtio_transport.c | 13 +++++++++++++
> > net/vmw_vsock/vmci_transport.c | 12 ++++++++++++
> > net/vmw_vsock/vsock_loopback.c | 6 ++++++
> > 7 files changed, 86 insertions(+)
> >
> > diff --git a/drivers/vhost/vsock.c b/drivers/vhost/vsock.c
> > index 69074656263d..4e3856aa2479 100644
> > --- a/drivers/vhost/vsock.c
> > +++ b/drivers/vhost/vsock.c
> > @@ -64,6 +64,11 @@ static u32 vhost_transport_get_local_cid(void)
> > return VHOST_VSOCK_DEFAULT_HOST_CID;
> > }
> >
> > +static bool vhost_transport_supports_local_mode(void)
> > +{
> > + return false;
> > +}
> > +
> > /* Callers that dereference the return value must hold vhost_vsock_mutex or the
> > * RCU read lock.
> > */
> > @@ -412,6 +417,7 @@ static struct virtio_transport vhost_transport = {
> > .module = THIS_MODULE,
> >
> > .get_local_cid = vhost_transport_get_local_cid,
> > + .supports_local_mode = vhost_transport_supports_local_mode,
> >
> > .init = virtio_transport_do_socket_init,
> > .destruct = virtio_transport_destruct,
> > diff --git a/include/net/af_vsock.h b/include/net/af_vsock.h
> > index 59d97a143204..e24ef1d9fe02 100644
> > --- a/include/net/af_vsock.h
> > +++ b/include/net/af_vsock.h
> > @@ -180,6 +180,17 @@ struct vsock_transport {
> > /* Addressing. */
> > u32 (*get_local_cid)(void);
> >
> > + /* Return true if the transport is compatible with
> > + * VSOCK_NET_MODE_LOCAL. Otherwise, return false.
> > + *
> > + * Transports should return false if they lack local-mode namespace
> > + * support (e.g., G2H transports like hyperv-vsock and vmci-vsock).
> > + * virtio-vsock returns true only if no device is present in order to
> > + * enable local mode in nested scenarios in which virtio-vsock is
> > + * loaded or built-in, but nonetheless unusable by sockets.
> > + */
> > + bool (*supports_local_mode)(void);
> > +
> > /* Read a single skb */
> > int (*read_skb)(struct vsock_sock *, skb_read_actor_t);
> >
> > diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c
> > index 243c0d588682..120adb9dad9f 100644
> > --- a/net/vmw_vsock/af_vsock.c
> > +++ b/net/vmw_vsock/af_vsock.c
> > @@ -91,6 +91,12 @@
> > * and locked down by a namespace manager. The default is "global". The mode
> > * is set per-namespace.
> > *
> > + * Note: LOCAL mode is only supported when using namespace-aware transports
> > + * (vhost-vsock, loopback). If a guest-to-host transport (virtio-vsock,
> > + * hyperv-vsock, vmci-vsock) is operational, attempts to set LOCAL mode will
> > + * fail with EOPNOTSUPP, as these transports do not support per-namespace
> > + * isolation.
> > + *
> > * The modes affect the allocation and accessibility of CIDs as follows:
> > *
> > * - global - access and allocation are all system-wide
> > @@ -2794,6 +2800,15 @@ static int vsock_net_mode_string(const struct ctl_table *table, int write,
> > else
> > return -EINVAL;
> >
> > + mutex_lock(&vsock_register_mutex);
> > + if (mode == VSOCK_NET_MODE_LOCAL &&
> > + transport_g2h && transport_g2h->supports_local_mode &&
> > + !transport_g2h->supports_local_mode()) {
> > + mutex_unlock(&vsock_register_mutex);
> > + return -EOPNOTSUPP;
> > + }
> > + mutex_unlock(&vsock_register_mutex);
>
> Wait, I think we already discussed about this, vsock_net_write_mode() must
> be called with the lock held.
>
> See
> https://lore.kernel.org/netdev/aRTTwuuXSz5CvNjt@devvm11784.nha0.facebook.com/
>
Ah right, oversight on my part.
> Since I guess we need another version of this patch, can you check the
> commit description to see if it reflects what we are doing now
> (e.g vhost is not enabled)?
>
> Also I don't understand why for vhost we will enable it later, but for
> virtio_transport and vsock_loopback we are enabling it now, also if this
> patch is before the support on that transports. I'm a bit confused.
>
> If something is unclear, let's discuss it before sending a new version.
>
>
> What I had in mind was, add this patch and explain why we need this new
> callback (like you did), but enable the support in the patches that
> really enable it for any transport. But maybe what is not clear to me is
> that we need this only for G2H. But now I'm confused about the discussion
> around vmci H2G. We decided to discard also that one, but here we are not
> checking that?
> I mean here we are calling supports_local_mode() only on G2H IIUC.
Ah right, VMCI broke my original mental model of only needing this check
for G2H (originally I didn't realize VMCI was H2G too).
I think now, we actually need to do this check for all of the transports
no? Including h2g, g2h, local, and dgram?
Additionally, the commit description needs to be updated to reflect that.
With this, we then end up with two commits:
commit 1) This commit which adds the callbacks and gives each
transport stubs to return false. Checks all transports (not just
G2H). Update the commit. Fix vsock_net_write_mode() race above.
commit 2) change the virtio-vsock/vhost-vsock/vsock-loopback to
add the real implementations (vhost + loopback return true,
virtio detects device). The other transports keep their return
false stubs so no changes.
Does that seem about right?
>
> > +
> > if (!vsock_net_write_mode(net, mode))
> > return -EPERM;
> >
> > @@ -2938,6 +2953,7 @@ int vsock_core_register(const struct vsock_transport *t, int features)
> > {
> > const struct vsock_transport *t_h2g, *t_g2h, *t_dgram, *t_local;
> > int err = mutex_lock_interruptible(&vsock_register_mutex);
> > + struct net *net;
> >
> > if (err)
> > return err;
> > @@ -2960,6 +2976,22 @@ int vsock_core_register(const struct vsock_transport *t, int features)
> > err = -EBUSY;
> > goto err_busy;
> > }
> > +
> > + /* G2H sockets break in LOCAL mode namespaces because G2H
>
> And also here we are talking about only of G2H, so what happen if vmci is
> loaded as H2G?
>
> IMO we should discuss this a bit more and make it a bit more generic, like
> check all the transports.
>
> Thanks,
> Stefano
Agreed.
Best,
Bobby
^ permalink raw reply
* Re: [PATCH net-next v11 05/13] vsock: add netns support to virtio transports
From: Bobby Eshleman @ 2025-11-21 19:03 UTC (permalink / raw)
To: Stefano Garzarella
Cc: David S. Miller, Eric Dumazet, Jakub Kicinski, Paolo Abeni,
Simon Horman, Stefan Hajnoczi, Michael S. Tsirkin, Jason Wang,
Eugenio Pérez, Xuan Zhuo, K. Y. Srinivasan, Haiyang Zhang,
Wei Liu, Dexuan Cui, Bryan Tan, Vishnu Dasa,
Broadcom internal kernel review list, Shuah Khan, linux-kernel,
virtualization, netdev, kvm, linux-hyperv, linux-kselftest,
berrange, Sargun Dhillon, Bobby Eshleman
In-Reply-To: <v6dpp4j4pjnrsa5amw7uubbqtpnxb4odpjhyjksr4mqes2qbzg@3bsjx5ofbwl4>
On Fri, Nov 21, 2025 at 03:39:25PM +0100, Stefano Garzarella wrote:
> On Thu, Nov 20, 2025 at 09:44:37PM -0800, Bobby Eshleman wrote:
> > From: Bobby Eshleman <bobbyeshleman@meta.com>
> >
> > Add netns support to loopback and vhost. Keep netns disabled for
> > virtio-vsock, but add necessary changes to comply with common API
> > updates.
> >
> > This is the patch in the series when vhost-vsock namespaces actually
> > come online. Hence, vhost_transport_supports_local_mode() is switched
> > to return true.
> >
> > Signed-off-by: Bobby Eshleman <bobbyeshleman@meta.com>
> > ---
> > Changes in v11:
> > - reorder with the skb ownership patch for loopback (Stefano)
> > - toggle vhost_transport_supports_local_mode() to true
> >
> > Changes in v10:
> > - Splitting patches complicates the series with meaningless placeholder
> > values that eventually get replaced anyway, so to avoid that this
> > patch combines into one. Links to previous patches here:
> > - Link: https://lore.kernel.org/all/20251111-vsock-vmtest-v9-3-852787a37bed@meta.com/
> > - Link: https://lore.kernel.org/all/20251111-vsock-vmtest-v9-6-852787a37bed@meta.com/
> > - Link: https://lore.kernel.org/all/20251111-vsock-vmtest-v9-7-852787a37bed@meta.com/
> > - remove placeholder values (Stefano)
> > - update comment describe net/net_mode for
> > virtio_transport_reset_no_sock()
> > ---
> > drivers/vhost/vsock.c | 47 ++++++++++++++++++------
> > include/linux/virtio_vsock.h | 8 +++--
> > net/vmw_vsock/virtio_transport.c | 10 ++++--
> > net/vmw_vsock/virtio_transport_common.c | 63 ++++++++++++++++++++++++---------
> > net/vmw_vsock/vsock_loopback.c | 8 +++--
> > 5 files changed, 103 insertions(+), 33 deletions(-)
>
> Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
If we move the supports_local_mode() changes into this patch (for virtio
and loopback, as I bring up in other discussion), should I drop this
trailer or carry it forward?
Thanks,
Bobby
^ permalink raw reply
* Re: [PATCH v2 3/3] x86/hyperv: Remove ASM_CALL_CONSTRAINT with VMMCALL insn
From: H. Peter Anvin @ 2025-11-22 0:06 UTC (permalink / raw)
To: Uros Bizjak, linux-hyperv, x86, linux-kernel
Cc: Michael Kelley, K. Y. Srinivasan, Haiyang Zhang, Wei Liu,
Dexuan Cui, Thomas Gleixner, Ingo Molnar, Borislav Petkov,
Dave Hansen
In-Reply-To: <20251121141437.205481-3-ubizjak@gmail.com>
On November 21, 2025 6:14:11 AM PST, Uros Bizjak <ubizjak@gmail.com> wrote:
>Unlike CALL instruction, VMMCALL does not push to the stack, so it's
>OK to allow the compiler to insert it before the frame pointer gets
>set up by the containing function. ASM_CALL_CONSTRAINT is for CALLs
>that must be inserted after the frame pointer is set up, so it is
>over-constraining here and can be removed.
>
>Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
>Tested-by: Michael Kelley <mhklinux@outlook.com>
>Cc: "K. Y. Srinivasan" <kys@microsoft.com>
>Cc: Haiyang Zhang <haiyangz@microsoft.com>
>Cc: Wei Liu <wei.liu@kernel.org>
>Cc: Dexuan Cui <decui@microsoft.com>
>Cc: Thomas Gleixner <tglx@linutronix.de>
>Cc: Ingo Molnar <mingo@redhat.com>
>Cc: Borislav Petkov <bp@alien8.de>
>Cc: Dave Hansen <dave.hansen@linux.intel.com>
>Cc: "H. Peter Anvin" <hpa@zytor.com>
>---
>v2: Expand commit message and include ASM_CALL_CONSTRAINT explanation
>---
> arch/x86/hyperv/ivm.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
>diff --git a/arch/x86/hyperv/ivm.c b/arch/x86/hyperv/ivm.c
>index 7365d8f43181..be7fad43a88d 100644
>--- a/arch/x86/hyperv/ivm.c
>+++ b/arch/x86/hyperv/ivm.c
>@@ -392,7 +392,7 @@ u64 hv_snp_hypercall(u64 control, u64 param1, u64 param2)
>
> register u64 __r8 asm("r8") = param2;
> asm volatile("vmmcall"
>- : "=a" (hv_status), ASM_CALL_CONSTRAINT,
>+ : "=a" (hv_status),
> "+c" (control), "+d" (param1), "+r" (__r8)
> : : "cc", "memory", "r9", "r10", "r11");
>
I think it would be good to have a comment at the point where ASM_CALL_CONSTRAINT is defined explaining its proper use.
Specifically, instructions like syscall, vmcall, vmfunc, vmmcall, int xx and VM-specific escape instructions are not "calls" because they either don't modify the stack or create an exception frame (kernel) or signal frame (user space) which is completely special.
^ permalink raw reply
* Re: [PATCH net-next, v3] net: mana: Implement ndo_tx_timeout and serialize queue resets per port.
From: Dipayaan Roy @ 2025-11-22 7:03 UTC (permalink / raw)
To: Simon Horman
Cc: kys, haiyangz, wei.liu, decui, andrew+netdev, davem, edumazet,
kuba, pabeni, longli, kotaranov, shradhagupta, ssengar, ernis,
shirazsaleem, linux-hyperv, netdev, linux-kernel, linux-rdma,
dipayanroy
In-Reply-To: <aRNsHUjW3PybGXCK@horms.kernel.org>
On Tue, Nov 11, 2025 at 05:02:21PM +0000, Simon Horman wrote:
> On Mon, Nov 10, 2025 at 02:35:41AM -0800, Dipayaan Roy wrote:
> > Implement .ndo_tx_timeout for MANA so any stalled TX queue can be detected
> > and a device-controlled port reset for all queues can be scheduled to a
> > ordered workqueue. The reset for all queues on stall detection is
> > recomended by hardware team.
> >
> > The change introduces a single ordered workqueue
> > ("mana_per_port_queue_reset_wq") with WQ_UNBOUND | WQ_MEM_RECLAIM and
> > queues exactly one work_struct per port onto it.
>
> I see that this goes some way to addressing Jakub's feedback
> on the commit message in his review of v2. But I this paragraph
> isn't adding much in it's current form. It seems to me some
> explanation of why why WQ_UNBOUND and WQ_MEM_RECLAIM are used is
> appropriate.
>
> [1] https://lore.kernel.org/all/20251029182233.59aea2d3@kernel.org/
>
Sure, I can add short explanation on the flag usage.
> > Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
> > Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com>
> > Signed-off-by: Dipayaan Roy <dipayanroy@linux.microsoft.com>
> > ---
> > Changes in v3:
> > -Fixed commit meesage, removed rtnl_trylock and added
> > disable_work_sync, fixed mana_queue_reset_work, and few
> > cosmetics.
> > Changes in v2:
> > -Fixed cosmetic changes.
> > ---
> > ---
> > drivers/net/ethernet/microsoft/mana/mana_en.c | 78 ++++++++++++++++++-
> > include/net/mana/gdma.h | 7 +-
> > include/net/mana/mana.h | 7 ++
> > 3 files changed, 90 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c
> > index cccd5b63cee6..636df3b066c5 100644
> > --- a/drivers/net/ethernet/microsoft/mana/mana_en.c
> > +++ b/drivers/net/ethernet/microsoft/mana/mana_en.c
> > @@ -298,6 +298,42 @@ static int mana_get_gso_hs(struct sk_buff *skb)
> > return gso_hs;
> > }
> >
> > +static void mana_per_port_queue_reset_work_handler(struct work_struct *work)
> > +{
> > + struct mana_queue_reset_work *reset_queue_work =
> > + container_of(work, struct mana_queue_reset_work, work);
> > +
> > + struct mana_port_context *apc = container_of(reset_queue_work,
> > + struct mana_port_context,
> > + queue_reset_work);
> > + struct net_device *ndev = apc->ndev;
> > + int err;
> > +
> > + rtnl_lock();
> > +
> > + /* Pre-allocate buffers to prevent failure in mana_attach later */
> > + err = mana_pre_alloc_rxbufs(apc, ndev->mtu, apc->num_queues);
> > + if (err) {
> > + netdev_err(ndev, "Insufficient memory for reset post tx stall detection\n");
> > + goto out;
> > + }
> > +
> > + err = mana_detach(ndev, false);
> > + if (err) {
> > + netdev_err(ndev, "mana_detach failed: %d\n", err);
> > + goto dealloc_pre_rxbufs;
> > + }
> > +
> > + err = mana_attach(ndev);
> > + if (err)
> > + netdev_err(ndev, "mana_attach failed: %d\n", err);
> > +
> > +dealloc_pre_rxbufs:
> > + mana_pre_dealloc_rxbufs(apc);
> > +out:
> > + rtnl_unlock();
> > +}
> > +
> > netdev_tx_t mana_start_xmit(struct sk_buff *skb, struct net_device *ndev)
> > {
> > enum mana_tx_pkt_format pkt_fmt = MANA_SHORT_PKT_FMT;
> > @@ -802,6 +838,23 @@ static int mana_change_mtu(struct net_device *ndev, int new_mtu)
> > return err;
> > }
> >
> > +static void mana_tx_timeout(struct net_device *netdev, unsigned int txqueue)
> > +{
> > + struct mana_port_context *apc = netdev_priv(netdev);
> > + struct mana_context *ac = apc->ac;
> > + struct gdma_context *gc = ac->gdma_dev->gdma_context;
> > +
> > + /* Already in service, hence tx queue reset is not required.*/
> > + if (gc->in_service)
> > + return;
> > +
> > + /* Note: If there are pending queue reset work for this port(apc),
> > + * subsequent request queued up from here are ignored. This is because
> > + * we are using the same work instance per port(apc).
> > + */
> > + queue_work(ac->per_port_queue_reset_wq, &apc->queue_reset_work.work);
> > +}
> > +
> > static int mana_shaper_set(struct net_shaper_binding *binding,
> > const struct net_shaper *shaper,
> > struct netlink_ext_ack *extack)
> > @@ -884,7 +937,9 @@ static const struct net_device_ops mana_devops = {
> > .ndo_bpf = mana_bpf,
> > .ndo_xdp_xmit = mana_xdp_xmit,
> > .ndo_change_mtu = mana_change_mtu,
> > - .net_shaper_ops = &mana_shaper_ops,
> > + .ndo_tx_timeout = mana_tx_timeout,
> > + .net_shaper_ops = &mana_shaper_ops,
> > +
> > };
> >
> > static void mana_cleanup_port_context(struct mana_port_context *apc)
> > @@ -3244,6 +3299,7 @@ static int mana_probe_port(struct mana_context *ac, int port_idx,
> > ndev->min_mtu = ETH_MIN_MTU;
> > ndev->needed_headroom = MANA_HEADROOM;
> > ndev->dev_port = port_idx;
> > + ndev->watchdog_timeo = 15 * HZ;
> > SET_NETDEV_DEV(ndev, gc->dev);
> >
> > netif_set_tso_max_size(ndev, GSO_MAX_SIZE);
> > @@ -3283,6 +3339,10 @@ static int mana_probe_port(struct mana_context *ac, int port_idx,
> >
> > debugfs_create_u32("current_speed", 0400, apc->mana_port_debugfs, &apc->speed);
> >
> > + /* Initialize the per port queue reset work.*/
> > + INIT_WORK(&apc->queue_reset_work.work,
> > + mana_per_port_queue_reset_work_handler);
> > +
>
> I think it would make more sense to move this to before the call to
> register_netdev(), which is a few lines above this hunk.
>
> I suppose that because a watchdog timeout is involved, it won't happen in
> practice, but in theory could fire ndo_tx_timeout before INIT_WORK is
> called, resulting in access to the work queue before it is initialised.
>
Sure, will address this in next version.
> > return 0;
> >
> > free_indir:
> > @@ -3488,6 +3548,15 @@ int mana_probe(struct gdma_dev *gd, bool resuming)
> > if (ac->num_ports > MAX_PORTS_IN_MANA_DEV)
> > ac->num_ports = MAX_PORTS_IN_MANA_DEV;
> >
> > + ac->per_port_queue_reset_wq =
> > + alloc_ordered_workqueue("mana_per_port_queue_reset_wq",
> > + WQ_UNBOUND | WQ_MEM_RECLAIM);
> > + if (!ac->per_port_queue_reset_wq) {
> > + dev_err(dev, "Failed to allocate per port queue reset workqueue\n");
> > + err = -ENOMEM;
> > + goto out;
> > + }
> > +
> > if (!resuming) {
> > for (i = 0; i < ac->num_ports; i++) {
> > err = mana_probe_port(ac, i, &ac->ports[i]);
>
> It is not strictly related to this patch, but the lines above the hunk
> below look like this:
>
> apc = netdev_priv(ndev);
> if (!ndev) {
> if (i == 0)
> dev_err(dev, "No net device to remove\n");
>
> If ndev is null then the call to netdev_priv() will result in a
> NULL pointer dereference. So I think it should be moved
> to after the check for !ndev.
>
Thanks for pointing this, I will fix this along with my next version.
> > @@ -3557,6 +3626,8 @@ void mana_remove(struct gdma_dev *gd, bool suspending)
> > goto out;
> > }
> >
> > + disable_work_sync(&apc->queue_reset_work.work);
> > +
> > /* All cleanup actions should stay after rtnl_lock(), otherwise
> > * other functions may access partially cleaned up data.
> > */
>
> Comments on code flagged by Claude Code with
> https://github.com/masoncl/review-prompts/
Thanks Simon for the review. I will send the rebased next version with all the
comments addressed.
Regards
^ permalink raw reply
* Re: [PATCH v2 3/3] x86/hyperv: Remove ASM_CALL_CONSTRAINT with VMMCALL insn
From: Uros Bizjak @ 2025-11-22 9:33 UTC (permalink / raw)
To: H. Peter Anvin
Cc: linux-hyperv, x86, linux-kernel, Michael Kelley, K. Y. Srinivasan,
Haiyang Zhang, Wei Liu, Dexuan Cui, Thomas Gleixner, Ingo Molnar,
Borislav Petkov, Dave Hansen
In-Reply-To: <8F5147DF-E0E2-4942-99D9-4242F3013635@zytor.com>
On Sat, Nov 22, 2025 at 1:06 AM H. Peter Anvin <hpa@zytor.com> wrote:
>
> On November 21, 2025 6:14:11 AM PST, Uros Bizjak <ubizjak@gmail.com> wrote:
> >Unlike CALL instruction, VMMCALL does not push to the stack, so it's
> >OK to allow the compiler to insert it before the frame pointer gets
> >set up by the containing function. ASM_CALL_CONSTRAINT is for CALLs
> >that must be inserted after the frame pointer is set up, so it is
> >over-constraining here and can be removed.
> >
> >Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
> >Tested-by: Michael Kelley <mhklinux@outlook.com>
> >Cc: "K. Y. Srinivasan" <kys@microsoft.com>
> >Cc: Haiyang Zhang <haiyangz@microsoft.com>
> >Cc: Wei Liu <wei.liu@kernel.org>
> >Cc: Dexuan Cui <decui@microsoft.com>
> >Cc: Thomas Gleixner <tglx@linutronix.de>
> >Cc: Ingo Molnar <mingo@redhat.com>
> >Cc: Borislav Petkov <bp@alien8.de>
> >Cc: Dave Hansen <dave.hansen@linux.intel.com>
> >Cc: "H. Peter Anvin" <hpa@zytor.com>
> >---
> >v2: Expand commit message and include ASM_CALL_CONSTRAINT explanation
> >---
> > arch/x86/hyperv/ivm.c | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> >diff --git a/arch/x86/hyperv/ivm.c b/arch/x86/hyperv/ivm.c
> >index 7365d8f43181..be7fad43a88d 100644
> >--- a/arch/x86/hyperv/ivm.c
> >+++ b/arch/x86/hyperv/ivm.c
> >@@ -392,7 +392,7 @@ u64 hv_snp_hypercall(u64 control, u64 param1, u64 param2)
> >
> > register u64 __r8 asm("r8") = param2;
> > asm volatile("vmmcall"
> >- : "=a" (hv_status), ASM_CALL_CONSTRAINT,
> >+ : "=a" (hv_status),
> > "+c" (control), "+d" (param1), "+r" (__r8)
> > : : "cc", "memory", "r9", "r10", "r11");
> >
>
> I think it would be good to have a comment at the point where ASM_CALL_CONSTRAINT is defined explaining its proper use.
>
> Specifically, instructions like syscall, vmcall, vmfunc, vmmcall, int xx and VM-specific escape instructions are not "calls" because they either don't modify the stack or create an exception frame (kernel) or signal frame (user space) which is completely special.
The existing comment already mentions CALL instruction only:
/*
* This output constraint should be used for any inline asm which has a "call"
* instruction. Otherwise the asm may be inserted before the frame pointer
* gets set up by the containing function. If you forget to do this, objtool
* may print a "call without frame pointer save/setup" warning.
*/
Uros.
^ permalink raw reply
* Re: [PATCH 0/6] arch, sysfb: Move screen and edid info into single place
From: Thomas Zimmermann @ 2025-11-22 10:52 UTC (permalink / raw)
To: Ard Biesheuvel
Cc: Arnd Bergmann, Javier Martinez Canillas, x86, linux-arm-kernel,
linux-kernel, linux-efi, loongarch, linux-riscv, dri-devel,
linux-hyperv, linux-pci, linux-fbdev
In-Reply-To: <CAMj1kXFdethf2sb1tm1V4wRW1SyPt-OnCmaAXc5cHNKuLJMXWA@mail.gmail.com>
Hi
Am 21.11.25 um 17:31 schrieb Ard Biesheuvel:
> On Fri, 21 Nov 2025 at 17:26, Thomas Zimmermann <tzimmermann@suse.de> wrote:
>> Hi
>>
>> Am 21.11.25 um 17:19 schrieb Ard Biesheuvel:
>>> On Fri, 21 Nov 2025 at 17:09, Thomas Zimmermann <tzimmermann@suse.de> wrote:
>>>>
>>>> Am 21.11.25 um 17:08 schrieb Thomas Zimmermann:
>>>>> Hi
>>>>>
>>>>> Am 21.11.25 um 16:56 schrieb Ard Biesheuvel:
>>>>>> On Fri, 21 Nov 2025 at 16:53, Thomas Zimmermann <tzimmermann@suse.de>
>>>>>> wrote:
>>>>>>> Hi
>>>>>>>
>>>>>>> Am 21.11.25 um 16:16 schrieb Ard Biesheuvel:
>>>>>>>> On Fri, 21 Nov 2025 at 16:10, Arnd Bergmann <arnd@arndb.de> wrote:
>>>>>>>>> On Fri, Nov 21, 2025, at 14:36, Thomas Zimmermann wrote:
>>>>>>>>>> Replace screen_info and edid_info with sysfb_primary_device of type
>>>>>>>>>> struct sysfb_display_info. Update all users.
>>>>>>>>>>
>>>>>>>>>> Sysfb DRM drivers currently fetch the global edid_info directly,
>>>>>>>>>> when
>>>>>>>>>> they should get that information together with the screen_info
>>>>>>>>>> from their
>>>>>>>>>> device. Wrapping screen_info and edid_info in
>>>>>>>>>> sysfb_primary_display and
>>>>>>>>>> passing this to drivers enables this.
>>>>>>>>>>
>>>>>>>>>> Replacing both with sysfb_primary_display has been motivate by
>>>>>>>>>> the EFI
>>>>>>>>>> stub. EFI wants to transfer EDID via config table in a single entry.
>>>>>>>>>> Using struct sysfb_display_info this will become easily possible.
>>>>>>>>>> Hence
>>>>>>>>>> accept some churn in architecture code for the long-term
>>>>>>>>>> improvements.
>>>>>>>>> This all looks good to me,
>>>>>>>>>
>>>>>>>>> Acked-by: Arnd Bergmann <arnd@arndb.de>
>>>>>>> Thanks
>>>>>>>
>>>>>>>>> It should also bring us one step closer to eventually
>>>>>>>>> disconnecting the x86 boot ABI from the kernel-internal
>>>>>>>>> sysfb_primary_display.
>>>>>>>>>
>>>>>>>> Agreed
>>>>>>>>
>>>>>>>> Acked-by: Ard Biesheuvel <ardb@kernel.org>
>>>>>>> Thanks
>>>>>>>
>>>>>>>> I can take patches 1-2 right away, if that helps during the next
>>>>>>>> cycle.
>>>>>>> From my sysfb-focused POV, these patches would ideally all go through
>>>>>>> the same tree, say efi or generic arch, or whatever fits best. Most of
>>>>>>> the other code is only renames anyway.
>>>>>>>
>>>>>> I don't mind queueing all of it, but I did get a conflict on
>>>>>> drivers/pci/vgaarb.c
>>>>> Probably from a78835b86a44 ("PCI/VGA: Select SCREEN_INFO on X86")
>>>> https://lore.kernel.org/all/20251013220829.1536292-1-superm1@kernel.org/
>>>>
>>> Yes, if I merge back -rc2 first, I can apply patches 1-5 onto my
>>> efi/next tree. But then I hit
>>>
>>> Applying: sysfb: Move edid_info into sysfb_primary_display
>>> error: sha1 information is lacking or useless (drivers/gpu/drm/sysfb/efidrm.c).
>>> error: could not build fake ancestor
>>> Patch failed at 0006 sysfb: Move edid_info into sysfb_primary_display
>>>
>>> If you prefer, you can take the whole lot via the sysfb tree instead,
>>> assuming it does not depend on the EDID changes I already queued up?
>> Sure, I can also add it to the drm-misc tree. ETA in upstream would be
>> v6.20-rc1.
>>
> But does that mean the EDID firmware on non-x86 will have to wait for
> 6.21? I was trying to avoid making this a 6 month effort.
No problem. Then let me rebase onto linux-next and put the existing EDID
patches for EFI on top. It's mostly acked or reviewed already. Once we
have it in good shape we can merged it all at once via the linux-efi
tree. Does that work for you?
Best regards
Thomas
--
--
Thomas Zimmermann
Graphics Driver Developer
SUSE Software Solutions Germany GmbH
Frankenstr. 146, 90461 Nürnberg, Germany, www.suse.com
GF: Jochen Jaser, Andrew McDonald, Werner Knoblich, (HRB 36809, AG Nürnberg)
^ permalink raw reply
* Re: [PATCH 0/6] arch, sysfb: Move screen and edid info into single place
From: Ard Biesheuvel @ 2025-11-22 11:42 UTC (permalink / raw)
To: Thomas Zimmermann
Cc: Arnd Bergmann, Javier Martinez Canillas, x86, linux-arm-kernel,
linux-kernel, linux-efi, loongarch, linux-riscv, dri-devel,
linux-hyperv, linux-pci, linux-fbdev
In-Reply-To: <4c716fd1-c989-4e48-9878-e7312fefc302@suse.de>
On Sat, 22 Nov 2025 at 11:52, Thomas Zimmermann <tzimmermann@suse.de> wrote:
>
> Hi
>
> Am 21.11.25 um 17:31 schrieb Ard Biesheuvel:
> > On Fri, 21 Nov 2025 at 17:26, Thomas Zimmermann <tzimmermann@suse.de> wrote:
> >> Hi
> >>
> >> Am 21.11.25 um 17:19 schrieb Ard Biesheuvel:
> >>> On Fri, 21 Nov 2025 at 17:09, Thomas Zimmermann <tzimmermann@suse.de> wrote:
> >>>>
> >>>> Am 21.11.25 um 17:08 schrieb Thomas Zimmermann:
> >>>>> Hi
> >>>>>
> >>>>> Am 21.11.25 um 16:56 schrieb Ard Biesheuvel:
> >>>>>> On Fri, 21 Nov 2025 at 16:53, Thomas Zimmermann <tzimmermann@suse.de>
> >>>>>> wrote:
> >>>>>>> Hi
> >>>>>>>
> >>>>>>> Am 21.11.25 um 16:16 schrieb Ard Biesheuvel:
> >>>>>>>> On Fri, 21 Nov 2025 at 16:10, Arnd Bergmann <arnd@arndb.de> wrote:
> >>>>>>>>> On Fri, Nov 21, 2025, at 14:36, Thomas Zimmermann wrote:
> >>>>>>>>>> Replace screen_info and edid_info with sysfb_primary_device of type
> >>>>>>>>>> struct sysfb_display_info. Update all users.
> >>>>>>>>>>
> >>>>>>>>>> Sysfb DRM drivers currently fetch the global edid_info directly,
> >>>>>>>>>> when
> >>>>>>>>>> they should get that information together with the screen_info
> >>>>>>>>>> from their
> >>>>>>>>>> device. Wrapping screen_info and edid_info in
> >>>>>>>>>> sysfb_primary_display and
> >>>>>>>>>> passing this to drivers enables this.
> >>>>>>>>>>
> >>>>>>>>>> Replacing both with sysfb_primary_display has been motivate by
> >>>>>>>>>> the EFI
> >>>>>>>>>> stub. EFI wants to transfer EDID via config table in a single entry.
> >>>>>>>>>> Using struct sysfb_display_info this will become easily possible.
> >>>>>>>>>> Hence
> >>>>>>>>>> accept some churn in architecture code for the long-term
> >>>>>>>>>> improvements.
> >>>>>>>>> This all looks good to me,
> >>>>>>>>>
> >>>>>>>>> Acked-by: Arnd Bergmann <arnd@arndb.de>
> >>>>>>> Thanks
> >>>>>>>
> >>>>>>>>> It should also bring us one step closer to eventually
> >>>>>>>>> disconnecting the x86 boot ABI from the kernel-internal
> >>>>>>>>> sysfb_primary_display.
> >>>>>>>>>
> >>>>>>>> Agreed
> >>>>>>>>
> >>>>>>>> Acked-by: Ard Biesheuvel <ardb@kernel.org>
> >>>>>>> Thanks
> >>>>>>>
> >>>>>>>> I can take patches 1-2 right away, if that helps during the next
> >>>>>>>> cycle.
> >>>>>>> From my sysfb-focused POV, these patches would ideally all go through
> >>>>>>> the same tree, say efi or generic arch, or whatever fits best. Most of
> >>>>>>> the other code is only renames anyway.
> >>>>>>>
> >>>>>> I don't mind queueing all of it, but I did get a conflict on
> >>>>>> drivers/pci/vgaarb.c
> >>>>> Probably from a78835b86a44 ("PCI/VGA: Select SCREEN_INFO on X86")
> >>>> https://lore.kernel.org/all/20251013220829.1536292-1-superm1@kernel.org/
> >>>>
> >>> Yes, if I merge back -rc2 first, I can apply patches 1-5 onto my
> >>> efi/next tree. But then I hit
> >>>
> >>> Applying: sysfb: Move edid_info into sysfb_primary_display
> >>> error: sha1 information is lacking or useless (drivers/gpu/drm/sysfb/efidrm.c).
> >>> error: could not build fake ancestor
> >>> Patch failed at 0006 sysfb: Move edid_info into sysfb_primary_display
> >>>
> >>> If you prefer, you can take the whole lot via the sysfb tree instead,
> >>> assuming it does not depend on the EDID changes I already queued up?
> >> Sure, I can also add it to the drm-misc tree. ETA in upstream would be
> >> v6.20-rc1.
> >>
> > But does that mean the EDID firmware on non-x86 will have to wait for
> > 6.21? I was trying to avoid making this a 6 month effort.
>
> No problem. Then let me rebase onto linux-next and put the existing EDID
> patches for EFI on top. It's mostly acked or reviewed already. Once we
> have it in good shape we can merged it all at once via the linux-efi
> tree. Does that work for you?
>
Sounds good.
^ permalink raw reply
* [PATCH net-next, v4] net: mana: Implement ndo_tx_timeout and serialize queue resets per port.
From: Dipayaan Roy @ 2025-11-23 18:08 UTC (permalink / raw)
To: kys, haiyangz, wei.liu, decui, andrew+netdev, davem, edumazet,
kuba, pabeni, longli, kotaranov, horms, shradhagupta, ssengar,
ernis, shirazsaleem, linux-hyperv, netdev, linux-kernel,
linux-rdma, dipayanroy
Implement .ndo_tx_timeout for MANA so any stalled TX queue can be detected
and a device-controlled port reset for all queues can be scheduled to a
ordered workqueue. The reset for all queues on stall detection is
recomended by hardware team.
The change introduces a single ordered workqueue
"mana_per_port_queue_reset_wq" queuing one work_struct per port,
using WQ_UNBOUND | WQ_MEM_RECLAIM so stalled queue reset work can
run on any CPU and still make forward progress under memory
pressure.
Reviewed-by: Pavan Chebbi <pavan.chebbi@broadcom.com>
Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Dipayaan Roy <dipayanroy@linux.microsoft.com>
---
Changes in v4:
-Fixed commit message, work initialization before registering netdev,
fixed potential null pointer de-reference bug.
Changes in v3:
-Fixed commit meesage, removed rtnl_trylock and added
disable_work_sync, fixed mana_queue_reset_work, and few
cosmetics.
Changes in v2:
-Fixed cosmetic changes.
---
---
drivers/net/ethernet/microsoft/mana/mana_en.c | 77 ++++++++++++++++++-
include/net/mana/gdma.h | 7 +-
include/net/mana/mana.h | 8 +-
3 files changed, 89 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/microsoft/mana/mana_en.c b/drivers/net/ethernet/microsoft/mana/mana_en.c
index 1ad154f9db1a..7c71257b43a4 100644
--- a/drivers/net/ethernet/microsoft/mana/mana_en.c
+++ b/drivers/net/ethernet/microsoft/mana/mana_en.c
@@ -299,6 +299,42 @@ static int mana_get_gso_hs(struct sk_buff *skb)
return gso_hs;
}
+static void mana_per_port_queue_reset_work_handler(struct work_struct *work)
+{
+ struct mana_queue_reset_work *reset_queue_work =
+ container_of(work, struct mana_queue_reset_work, work);
+
+ struct mana_port_context *apc = container_of(reset_queue_work,
+ struct mana_port_context,
+ queue_reset_work);
+ struct net_device *ndev = apc->ndev;
+ int err;
+
+ rtnl_lock();
+
+ /* Pre-allocate buffers to prevent failure in mana_attach later */
+ err = mana_pre_alloc_rxbufs(apc, ndev->mtu, apc->num_queues);
+ if (err) {
+ netdev_err(ndev, "Insufficient memory for reset post tx stall detection\n");
+ goto out;
+ }
+
+ err = mana_detach(ndev, false);
+ if (err) {
+ netdev_err(ndev, "mana_detach failed: %d\n", err);
+ goto dealloc_pre_rxbufs;
+ }
+
+ err = mana_attach(ndev);
+ if (err)
+ netdev_err(ndev, "mana_attach failed: %d\n", err);
+
+dealloc_pre_rxbufs:
+ mana_pre_dealloc_rxbufs(apc);
+out:
+ rtnl_unlock();
+}
+
netdev_tx_t mana_start_xmit(struct sk_buff *skb, struct net_device *ndev)
{
enum mana_tx_pkt_format pkt_fmt = MANA_SHORT_PKT_FMT;
@@ -839,6 +875,23 @@ static int mana_change_mtu(struct net_device *ndev, int new_mtu)
return err;
}
+static void mana_tx_timeout(struct net_device *netdev, unsigned int txqueue)
+{
+ struct mana_port_context *apc = netdev_priv(netdev);
+ struct mana_context *ac = apc->ac;
+ struct gdma_context *gc = ac->gdma_dev->gdma_context;
+
+ /* Already in service, hence tx queue reset is not required.*/
+ if (gc->in_service)
+ return;
+
+ /* Note: If there are pending queue reset work for this port(apc),
+ * subsequent request queued up from here are ignored. This is because
+ * we are using the same work instance per port(apc).
+ */
+ queue_work(ac->per_port_queue_reset_wq, &apc->queue_reset_work.work);
+}
+
static int mana_shaper_set(struct net_shaper_binding *binding,
const struct net_shaper *shaper,
struct netlink_ext_ack *extack)
@@ -924,6 +977,7 @@ static const struct net_device_ops mana_devops = {
.ndo_bpf = mana_bpf,
.ndo_xdp_xmit = mana_xdp_xmit,
.ndo_change_mtu = mana_change_mtu,
+ .ndo_tx_timeout = mana_tx_timeout,
.net_shaper_ops = &mana_shaper_ops,
};
@@ -3287,6 +3341,7 @@ static int mana_probe_port(struct mana_context *ac, int port_idx,
ndev->min_mtu = ETH_MIN_MTU;
ndev->needed_headroom = MANA_HEADROOM;
ndev->dev_port = port_idx;
+ ndev->watchdog_timeo = 15 * HZ;
SET_NETDEV_DEV(ndev, gc->dev);
netif_set_tso_max_size(ndev, GSO_MAX_SIZE);
@@ -3303,6 +3358,10 @@ static int mana_probe_port(struct mana_context *ac, int port_idx,
if (err)
goto reset_apc;
+ /* Initialize the per port queue reset work.*/
+ INIT_WORK(&apc->queue_reset_work.work,
+ mana_per_port_queue_reset_work_handler);
+
netdev_lockdep_set_classes(ndev);
ndev->hw_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_IPV6_CSUM;
@@ -3549,6 +3608,15 @@ int mana_probe(struct gdma_dev *gd, bool resuming)
if (ac->num_ports > MAX_PORTS_IN_MANA_DEV)
ac->num_ports = MAX_PORTS_IN_MANA_DEV;
+ ac->per_port_queue_reset_wq =
+ alloc_ordered_workqueue("mana_per_port_queue_reset_wq",
+ WQ_UNBOUND | WQ_MEM_RECLAIM);
+ if (!ac->per_port_queue_reset_wq) {
+ dev_err(dev, "Failed to allocate per port queue reset workqueue\n");
+ err = -ENOMEM;
+ goto out;
+ }
+
if (!resuming) {
for (i = 0; i < ac->num_ports; i++) {
err = mana_probe_port(ac, i, &ac->ports[i]);
@@ -3616,13 +3684,15 @@ void mana_remove(struct gdma_dev *gd, bool suspending)
for (i = 0; i < ac->num_ports; i++) {
ndev = ac->ports[i];
- apc = netdev_priv(ndev);
if (!ndev) {
if (i == 0)
dev_err(dev, "No net device to remove\n");
goto out;
}
+ apc = netdev_priv(ndev);
+ disable_work_sync(&apc->queue_reset_work.work);
+
/* All cleanup actions should stay after rtnl_lock(), otherwise
* other functions may access partially cleaned up data.
*/
@@ -3647,6 +3717,11 @@ void mana_remove(struct gdma_dev *gd, bool suspending)
free_netdev(ndev);
}
+ if (ac->per_port_queue_reset_wq) {
+ destroy_workqueue(ac->per_port_queue_reset_wq);
+ ac->per_port_queue_reset_wq = NULL;
+ }
+
mana_destroy_eq(ac);
out:
mana_gd_deregister_device(gd);
diff --git a/include/net/mana/gdma.h b/include/net/mana/gdma.h
index a4cf307859f8..808622ae5ccc 100644
--- a/include/net/mana/gdma.h
+++ b/include/net/mana/gdma.h
@@ -592,6 +592,10 @@ enum {
/* Driver can self reset on FPGA Reconfig EQE notification */
#define GDMA_DRV_CAP_FLAG_1_HANDLE_RECONFIG_EQE BIT(17)
+
+/* Driver detects stalled send queues and recovers them */
+#define GDMA_DRV_CAP_FLAG_1_HANDLE_STALL_SQ_RECOVERY BIT(18)
+
#define GDMA_DRV_CAP_FLAG_1_HW_VPORT_LINK_AWARE BIT(6)
/* Driver supports linearizing the skb when num_sge exceeds hardware limit */
@@ -611,7 +615,8 @@ enum {
GDMA_DRV_CAP_FLAG_1_HANDLE_RECONFIG_EQE | \
GDMA_DRV_CAP_FLAG_1_HW_VPORT_LINK_AWARE | \
GDMA_DRV_CAP_FLAG_1_PERIODIC_STATS_QUERY | \
- GDMA_DRV_CAP_FLAG_1_SKB_LINEARIZE)
+ GDMA_DRV_CAP_FLAG_1_SKB_LINEARIZE | \
+ GDMA_DRV_CAP_FLAG_1_HANDLE_STALL_SQ_RECOVERY)
#define GDMA_DRV_CAP_FLAGS2 0
diff --git a/include/net/mana/mana.h b/include/net/mana/mana.h
index d7e089c6b694..cef78a871c7c 100644
--- a/include/net/mana/mana.h
+++ b/include/net/mana/mana.h
@@ -480,7 +480,7 @@ struct mana_context {
struct mana_ethtool_hc_stats hc_stats;
struct mana_eq *eqs;
struct dentry *mana_eqs_debugfs;
-
+ struct workqueue_struct *per_port_queue_reset_wq;
/* Workqueue for querying hardware stats */
struct delayed_work gf_stats_work;
bool hwc_timeout_occurred;
@@ -492,9 +492,15 @@ struct mana_context {
u32 link_event;
};
+struct mana_queue_reset_work {
+ /* Work structure */
+ struct work_struct work;
+};
+
struct mana_port_context {
struct mana_context *ac;
struct net_device *ndev;
+ struct mana_queue_reset_work queue_reset_work;
u8 mac_addr[ETH_ALEN];
--
2.43.0
^ permalink raw reply related
* Re: [PATCH 1/6] efi: earlycon: Reduce number of references to global screen_info
From: Richard Lyu @ 2025-11-24 9:28 UTC (permalink / raw)
To: Thomas Zimmermann, ardb, javierm, arnd
Cc: x86, linux-arm-kernel, linux-kernel, linux-efi, loongarch,
linux-riscv, dri-devel, linux-hyperv, linux-pci, linux-fbdev
In-Reply-To: <20251121135624.494768-2-tzimmermann@suse.de>
Reviewed-by: Richard Lyu <richard.lyu@suse.com>
On 2025/11/21 14:36, Thomas Zimmermann wrote:
> Replace usage of global screen_info with local pointers. This will
> later reduce churn when screen_info is being moved.
>
> Signed-off-by: Thomas Zimmermann <tzimmermann@suse.de>
> ---
> drivers/firmware/efi/earlycon.c | 40 ++++++++++++++++-----------------
> 1 file changed, 20 insertions(+), 20 deletions(-)
>
> diff --git a/drivers/firmware/efi/earlycon.c b/drivers/firmware/efi/earlycon.c
> index d18a1a5de144..fac3a295c57f 100644
> --- a/drivers/firmware/efi/earlycon.c
> +++ b/drivers/firmware/efi/earlycon.c
> @@ -32,12 +32,13 @@ static void *efi_fb;
> */
> static int __init efi_earlycon_remap_fb(void)
> {
> + const struct screen_info *si = &screen_info;
> +
> /* bail if there is no bootconsole or it was unregistered already */
> if (!earlycon_console || !console_is_registered(earlycon_console))
> return 0;
>
> - efi_fb = memremap(fb_base, screen_info.lfb_size,
> - fb_wb ? MEMREMAP_WB : MEMREMAP_WC);
> + efi_fb = memremap(fb_base, si->lfb_size, fb_wb ? MEMREMAP_WB : MEMREMAP_WC);
>
> return efi_fb ? 0 : -ENOMEM;
> }
> @@ -71,12 +72,12 @@ static __ref void efi_earlycon_unmap(void *addr, unsigned long len)
> early_memunmap(addr, len);
> }
>
> -static void efi_earlycon_clear_scanline(unsigned int y)
> +static void efi_earlycon_clear_scanline(unsigned int y, const struct screen_info *si)
> {
> unsigned long *dst;
> u16 len;
>
> - len = screen_info.lfb_linelength;
> + len = si->lfb_linelength;
> dst = efi_earlycon_map(y*len, len);
> if (!dst)
> return;
> @@ -85,7 +86,7 @@ static void efi_earlycon_clear_scanline(unsigned int y)
> efi_earlycon_unmap(dst, len);
> }
>
> -static void efi_earlycon_scroll_up(void)
> +static void efi_earlycon_scroll_up(const struct screen_info *si)
> {
> unsigned long *dst, *src;
> u16 maxlen = 0;
> @@ -99,8 +100,8 @@ static void efi_earlycon_scroll_up(void)
> }
> maxlen *= 4;
>
> - len = screen_info.lfb_linelength;
> - height = screen_info.lfb_height;
> + len = si->lfb_linelength;
> + height = si->lfb_height;
>
> for (i = 0; i < height - font->height; i++) {
> dst = efi_earlycon_map(i*len, len);
> @@ -120,7 +121,8 @@ static void efi_earlycon_scroll_up(void)
> }
> }
>
> -static void efi_earlycon_write_char(u32 *dst, unsigned char c, unsigned int h)
> +static void efi_earlycon_write_char(u32 *dst, unsigned char c, unsigned int h,
> + const struct screen_info *si)
> {
> const u32 color_black = 0x00000000;
> const u32 color_white = 0x00ffffff;
> @@ -145,13 +147,12 @@ static void efi_earlycon_write_char(u32 *dst, unsigned char c, unsigned int h)
> static void
> efi_earlycon_write(struct console *con, const char *str, unsigned int num)
> {
> - struct screen_info *si;
> + const struct screen_info *si = &screen_info;
> u32 cur_efi_x = efi_x;
> unsigned int len;
> const char *s;
> void *dst;
>
> - si = &screen_info;
> len = si->lfb_linelength;
>
> while (num) {
> @@ -174,7 +175,7 @@ efi_earlycon_write(struct console *con, const char *str, unsigned int num)
> x = efi_x;
>
> while (n-- > 0) {
> - efi_earlycon_write_char(dst + x*4, *s, h);
> + efi_earlycon_write_char(dst + x*4, *s, h, si);
> x += font->width;
> s++;
> }
> @@ -207,10 +208,10 @@ efi_earlycon_write(struct console *con, const char *str, unsigned int num)
> cur_line_y = (cur_line_y + 1) % max_line_y;
>
> efi_y -= font->height;
> - efi_earlycon_scroll_up();
> + efi_earlycon_scroll_up(si);
>
> for (i = 0; i < font->height; i++)
> - efi_earlycon_clear_scanline(efi_y + i);
> + efi_earlycon_clear_scanline(efi_y + i, si);
> }
> }
> }
> @@ -226,22 +227,21 @@ void __init efi_earlycon_reprobe(void)
> static int __init efi_earlycon_setup(struct earlycon_device *device,
> const char *opt)
> {
> - struct screen_info *si;
> + const struct screen_info *si = &screen_info;
> u16 xres, yres;
> u32 i;
>
> fb_wb = opt && !strcmp(opt, "ram");
>
> - if (screen_info.orig_video_isVGA != VIDEO_TYPE_EFI) {
> + if (si->orig_video_isVGA != VIDEO_TYPE_EFI) {
> fb_probed = true;
> return -ENODEV;
> }
>
> - fb_base = screen_info.lfb_base;
> - if (screen_info.capabilities & VIDEO_CAPABILITY_64BIT_BASE)
> - fb_base |= (u64)screen_info.ext_lfb_base << 32;
> + fb_base = si->lfb_base;
> + if (si->capabilities & VIDEO_CAPABILITY_64BIT_BASE)
> + fb_base |= (u64)si->ext_lfb_base << 32;
>
> - si = &screen_info;
> xres = si->lfb_width;
> yres = si->lfb_height;
>
> @@ -266,7 +266,7 @@ static int __init efi_earlycon_setup(struct earlycon_device *device,
>
> efi_y -= font->height;
> for (i = 0; i < (yres - efi_y) / font->height; i++)
> - efi_earlycon_scroll_up();
> + efi_earlycon_scroll_up(si);
>
> device->con->write = efi_earlycon_write;
> earlycon_console = device->con;
> --
> 2.51.1
>
>
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox