Commit 762a9f2f authored by Linus Torvalds's avatar Linus Torvalds

Merge tag 'for-linus-5.7-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rw/uml

Pull UML updates from Richard Weinberger:

 - New mode for time travel, external via virtio

 - Fixes for ubd to make sure no requests can get lost

 - Fixes for vector networking

 - Allow CONFIG_STATIC_LINK only when possible

 - Minor cleanups and fixes

* tag 'for-linus-5.7-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rw/uml:
  um: Remove some unnecessary NULL checks in vector_user.c
  um: vector: Avoid NULL ptr deference if transport is unset
  um: Make CONFIG_STATIC_LINK actually static
  um: Implement cpu_relax() as ndelay(1) for time-travel
  um: Implement ndelay/udelay in time-travel mode
  um: Implement time-travel=ext
  um: virtio: Implement VHOST_USER_PROTOCOL_F_INBAND_NOTIFICATIONS
  um: time-travel: Rewrite as an event scheduler
  um: Move timer-internal.h to non-shared
  hostfs: Use kasprintf() instead of fixed buffer formatting
  um: falloc.h needs to be directly included for older libc
  um: ubd: Retry buffer read on any kind of error
  um: ubd: Prevent buffer overrun on command completion
  um: Fix overlapping ELF segments when statically linked
  um: Delete never executed timer
  um: Don't overwrite ethtool driver version
  um: Fix len of file in create_pid_file
  um: Don't use console_drivers directly
  um: Cleanup CONFIG_IOSCHED_CFQ
parents d5d24766 4a7c4624
......@@ -62,9 +62,12 @@ config NR_CPUS
source "arch/$(HEADER_ARCH)/um/Kconfig"
config FORBID_STATIC_LINK
bool
config STATIC_LINK
bool "Force a static link"
default n
depends on !FORBID_STATIC_LINK
help
This option gives you the ability to force a static link of UML.
Normally, UML is linked as a shared binary. This is inconvenient for
......@@ -73,6 +76,9 @@ config STATIC_LINK
Additionally, this option enables using higher memory spaces (up to
2.75G) for UML.
NOTE: This option is incompatible with some networking features which
depend on features that require being dynamically loaded (like NSS).
config LD_SCRIPT_STATIC
bool
default y
......@@ -191,6 +197,7 @@ config UML_TIME_TRAVEL_SUPPORT
prompt "Support time-travel mode (e.g. for test execution)"
# inf-cpu mode is incompatible with the benchmarking
depends on !RAID6_PQ_BENCHMARK
depends on !SMP
help
Enable this option to support time travel inside the UML instance.
......
......@@ -26,7 +26,7 @@ CONFIG_SLAB=y
CONFIG_MODULES=y
CONFIG_MODULE_UNLOAD=y
# CONFIG_BLK_DEV_BSG is not set
CONFIG_IOSCHED_CFQ=m
CONFIG_IOSCHED_BFQ=m
CONFIG_SSL=y
CONFIG_NULL_CHAN=y
CONFIG_PORT_CHAN=y
......
......@@ -24,7 +24,7 @@ CONFIG_SLAB=y
CONFIG_MODULES=y
CONFIG_MODULE_UNLOAD=y
# CONFIG_BLK_DEV_BSG is not set
CONFIG_IOSCHED_CFQ=m
CONFIG_IOSCHED_BFQ=m
CONFIG_SSL=y
CONFIG_NULL_CHAN=y
CONFIG_PORT_CHAN=y
......
......@@ -234,6 +234,7 @@ config UML_NET_DAEMON
config UML_NET_VECTOR
bool "Vector I/O high performance network devices"
depends on UML_NET
select FORBID_STATIC_LINK
help
This User-Mode Linux network driver uses multi-message send
and receive functions. The host running the UML guest must have
......@@ -245,6 +246,7 @@ config UML_NET_VECTOR
config UML_NET_VDE
bool "VDE transport (obsolete)"
depends on UML_NET
select FORBID_STATIC_LINK
help
This User-Mode Linux network transport allows one or more running
UMLs on a single host to communicate with each other and also
......@@ -292,6 +294,7 @@ config UML_NET_MCAST
config UML_NET_PCAP
bool "pcap transport (obsolete)"
depends on UML_NET
select FORBID_STATIC_LINK
help
The pcap transport makes a pcap packet stream on the host look
like an ethernet device inside UML. This is useful for making
......
......@@ -266,7 +266,6 @@ static void uml_net_get_drvinfo(struct net_device *dev,
struct ethtool_drvinfo *info)
{
strlcpy(info->driver, DRIVER_NAME, sizeof(info->driver));
strlcpy(info->version, "42", sizeof(info->version));
}
static const struct ethtool_ops uml_net_ethtool_ops = {
......@@ -275,17 +274,6 @@ static const struct ethtool_ops uml_net_ethtool_ops = {
.get_ts_info = ethtool_op_get_ts_info,
};
static void uml_net_user_timer_expire(struct timer_list *t)
{
#ifdef undef
struct uml_net_private *lp = from_timer(lp, t, tl);
struct connection *conn = &lp->user;
dprintk(KERN_INFO "uml_net_user_timer_expire [%p]\n", conn);
do_connect(conn);
#endif
}
void uml_net_setup_etheraddr(struct net_device *dev, char *str)
{
unsigned char *addr = dev->dev_addr;
......@@ -456,7 +444,6 @@ static void eth_configure(int n, void *init, char *mac,
.add_address = transport->user->add_address,
.delete_address = transport->user->delete_address });
timer_setup(&lp->tl, uml_net_user_timer_expire, 0);
spin_lock_init(&lp->lock);
memcpy(lp->mac, dev->dev_addr, sizeof(lp->mac));
......
......@@ -1592,11 +1592,11 @@ int io_thread(void *arg)
&io_remainder_size,
UBD_REQ_BUFFER_SIZE
);
if (n < 0) {
if (n == -EAGAIN) {
if (n <= 0) {
if (n == -EAGAIN)
ubd_read_poll(-1);
continue;
}
continue;
}
for (count = 0; count < n/sizeof(struct io_thread_req *); count++) {
......@@ -1607,7 +1607,9 @@ int io_thread(void *arg)
written = 0;
do {
res = os_write_file(kernel_fd, ((char *) io_req_buffer) + written, n);
res = os_write_file(kernel_fd,
((char *) io_req_buffer) + written,
n - written);
if (res >= 0) {
written += res;
}
......
......@@ -46,7 +46,6 @@
#define DRIVER_NAME "uml-vector"
#define DRIVER_VERSION "01"
struct vector_cmd_line_arg {
struct list_head list;
int unit;
......@@ -198,6 +197,9 @@ static int get_transport_options(struct arglist *def)
long parsed;
int result = 0;
if (transport == NULL)
return -EINVAL;
if (vector != NULL) {
if (kstrtoul(vector, 10, &parsed) == 0) {
if (parsed == 0) {
......@@ -1378,7 +1380,6 @@ static void vector_net_get_drvinfo(struct net_device *dev,
struct ethtool_drvinfo *info)
{
strlcpy(info->driver, DRIVER_NAME, sizeof(info->driver));
strlcpy(info->version, DRIVER_VERSION, sizeof(info->version));
}
static int vector_net_load_bpf_flash(struct net_device *dev,
......
......@@ -221,8 +221,7 @@ static struct vector_fds *user_init_tap_fds(struct arglist *ifspec)
return result;
tap_cleanup:
printk(UM_KERN_ERR "user_init_tap: init failed, error %d", fd);
if (result != NULL)
kfree(result);
kfree(result);
return NULL;
}
......@@ -266,8 +265,7 @@ static struct vector_fds *user_init_hybrid_fds(struct arglist *ifspec)
return result;
hybrid_cleanup:
printk(UM_KERN_ERR "user_init_hybrid: init failed");
if (result != NULL)
kfree(result);
kfree(result);
return NULL;
}
......@@ -344,10 +342,8 @@ static struct vector_fds *user_init_unix_fds(struct arglist *ifspec, int id)
unix_cleanup:
if (fd >= 0)
os_close_file(fd);
if (remote_addr != NULL)
kfree(remote_addr);
if (result != NULL)
kfree(result);
kfree(remote_addr);
kfree(result);
return NULL;
}
......@@ -382,8 +378,7 @@ static struct vector_fds *user_init_raw_fds(struct arglist *ifspec)
return result;
raw_cleanup:
printk(UM_KERN_ERR "user_init_raw: init failed, error %d", err);
if (result != NULL)
kfree(result);
kfree(result);
return NULL;
}
......
......@@ -10,9 +10,10 @@
/* Feature bits */
#define VHOST_USER_F_PROTOCOL_FEATURES 30
/* Protocol feature bits */
#define VHOST_USER_PROTOCOL_F_REPLY_ACK 3
#define VHOST_USER_PROTOCOL_F_SLAVE_REQ 5
#define VHOST_USER_PROTOCOL_F_CONFIG 9
#define VHOST_USER_PROTOCOL_F_REPLY_ACK 3
#define VHOST_USER_PROTOCOL_F_SLAVE_REQ 5
#define VHOST_USER_PROTOCOL_F_CONFIG 9
#define VHOST_USER_PROTOCOL_F_INBAND_NOTIFICATIONS 14
/* Vring state index masks */
#define VHOST_USER_VRING_INDEX_MASK 0xff
#define VHOST_USER_VRING_POLL_MASK BIT(8)
......@@ -24,7 +25,8 @@
/* Supported protocol features */
#define VHOST_USER_SUPPORTED_PROTOCOL_F (BIT_ULL(VHOST_USER_PROTOCOL_F_REPLY_ACK) | \
BIT_ULL(VHOST_USER_PROTOCOL_F_SLAVE_REQ) | \
BIT_ULL(VHOST_USER_PROTOCOL_F_CONFIG))
BIT_ULL(VHOST_USER_PROTOCOL_F_CONFIG) | \
BIT_ULL(VHOST_USER_PROTOCOL_F_INBAND_NOTIFICATIONS))
enum vhost_user_request {
VHOST_USER_GET_FEATURES = 1,
......@@ -52,12 +54,14 @@ enum vhost_user_request {
VHOST_USER_SET_VRING_ENDIAN = 23,
VHOST_USER_GET_CONFIG = 24,
VHOST_USER_SET_CONFIG = 25,
VHOST_USER_VRING_KICK = 35,
};
enum vhost_user_slave_request {
VHOST_USER_SLAVE_IOTLB_MSG = 1,
VHOST_USER_SLAVE_CONFIG_CHANGE_MSG = 2,
VHOST_USER_SLAVE_VRING_HOST_NOTIFIER_MSG = 3,
VHOST_USER_SLAVE_VRING_CALL = 4,
};
struct vhost_user_header {
......
......@@ -26,6 +26,7 @@
#include <linux/virtio.h>
#include <linux/virtio_config.h>
#include <linux/virtio_ring.h>
#include <linux/time-internal.h>
#include <shared/as-layout.h>
#include <irq_kern.h>
#include <init.h>
......@@ -53,6 +54,7 @@ struct virtio_uml_device {
struct virtio_device vdev;
struct platform_device *pdev;
spinlock_t sock_lock;
int sock, req_fd;
u64 features;
u64 protocol_features;
......@@ -63,6 +65,11 @@ struct virtio_uml_device {
struct virtio_uml_vq_info {
int kick_fd, call_fd;
char name[32];
#ifdef CONFIG_UML_TIME_TRAVEL_SUPPORT
struct virtqueue *vq;
vq_callback_t *callback;
struct time_travel_event defer;
#endif
};
extern unsigned long long physmem_size, highmem;
......@@ -117,10 +124,27 @@ static int vhost_user_recv_header(int fd, struct vhost_user_msg *msg)
static int vhost_user_recv(struct virtio_uml_device *vu_dev,
int fd, struct vhost_user_msg *msg,
size_t max_payload_size)
size_t max_payload_size, bool wait)
{
size_t size;
int rc = vhost_user_recv_header(fd, msg);
int rc;
/*
* In virtio time-travel mode, we're handling all the vhost-user
* FDs by polling them whenever appropriate. However, we may get
* into a situation where we're sending out an interrupt message
* to a device (e.g. a net device) and need to handle a simulation
* time message while doing so, e.g. one that tells us to update
* our idea of how long we can run without scheduling.
*
* Thus, we need to not just read() from the given fd, but need
* to also handle messages for the simulation time - this function
* does that for us while waiting for the given fd to be readable.
*/
if (wait)
time_travel_wait_readable(fd);
rc = vhost_user_recv_header(fd, msg);
if (rc == -ECONNRESET && vu_dev->registered) {
struct virtio_uml_platform_data *pdata;
......@@ -142,7 +166,8 @@ static int vhost_user_recv_resp(struct virtio_uml_device *vu_dev,
struct vhost_user_msg *msg,
size_t max_payload_size)
{
int rc = vhost_user_recv(vu_dev, vu_dev->sock, msg, max_payload_size);
int rc = vhost_user_recv(vu_dev, vu_dev->sock, msg,
max_payload_size, true);
if (rc)
return rc;
......@@ -172,7 +197,8 @@ static int vhost_user_recv_req(struct virtio_uml_device *vu_dev,
struct vhost_user_msg *msg,
size_t max_payload_size)
{
int rc = vhost_user_recv(vu_dev, vu_dev->req_fd, msg, max_payload_size);
int rc = vhost_user_recv(vu_dev, vu_dev->req_fd, msg,
max_payload_size, false);
if (rc)
return rc;
......@@ -189,6 +215,7 @@ static int vhost_user_send(struct virtio_uml_device *vu_dev,
int *fds, size_t num_fds)
{
size_t size = sizeof(msg->header) + msg->header.size;
unsigned long flags;
bool request_ack;
int rc;
......@@ -207,24 +234,28 @@ static int vhost_user_send(struct virtio_uml_device *vu_dev,
if (request_ack)
msg->header.flags |= VHOST_USER_FLAG_NEED_REPLY;
spin_lock_irqsave(&vu_dev->sock_lock, flags);
rc = full_sendmsg_fds(vu_dev->sock, msg, size, fds, num_fds);
if (rc < 0)
return rc;
goto out;
if (request_ack) {
uint64_t status;
rc = vhost_user_recv_u64(vu_dev, &status);
if (rc)
return rc;
goto out;
if (status) {
vu_err(vu_dev, "slave reports error: %llu\n", status);
return -EIO;
rc = -EIO;
goto out;
}
}
return 0;
out:
spin_unlock_irqrestore(&vu_dev->sock_lock, flags);
return rc;
}
static int vhost_user_send_no_payload(struct virtio_uml_device *vu_dev,
......@@ -324,6 +355,7 @@ static void vhost_user_reply(struct virtio_uml_device *vu_dev,
static irqreturn_t vu_req_interrupt(int irq, void *data)
{
struct virtio_uml_device *vu_dev = data;
struct virtqueue *vq;
int response = 1;
struct {
struct vhost_user_msg msg;
......@@ -343,6 +375,15 @@ static irqreturn_t vu_req_interrupt(int irq, void *data)
virtio_config_changed(&vu_dev->vdev);
response = 0;
break;
case VHOST_USER_SLAVE_VRING_CALL:
virtio_device_for_each_vq((&vu_dev->vdev), vq) {
if (vq->index == msg.msg.payload.vring_state.index) {
response = 0;
vring_interrupt(0 /* ignored */, vq);
break;
}
}
break;
case VHOST_USER_SLAVE_IOTLB_MSG:
/* not supported - VIRTIO_F_IOMMU_PLATFORM */
case VHOST_USER_SLAVE_VRING_HOST_NOTIFIER_MSG:
......@@ -684,6 +725,17 @@ static bool vu_notify(struct virtqueue *vq)
const uint64_t n = 1;
int rc;
time_travel_propagate_time();
if (info->kick_fd < 0) {
struct virtio_uml_device *vu_dev;
vu_dev = to_virtio_uml_device(vq->vdev);
return vhost_user_set_vring_state(vu_dev, VHOST_USER_VRING_KICK,
vq->index, 0) == 0;
}
do {
rc = os_write_file(info->kick_fd, &n, sizeof(n));
} while (rc == -EINTR);
......@@ -749,10 +801,13 @@ static void vu_del_vq(struct virtqueue *vq)
{
struct virtio_uml_vq_info *info = vq->priv;
um_free_irq(VIRTIO_IRQ, vq);
if (info->call_fd >= 0) {
um_free_irq(VIRTIO_IRQ, vq);
os_close_file(info->call_fd);
}
os_close_file(info->call_fd);
os_close_file(info->kick_fd);
if (info->kick_fd >= 0)
os_close_file(info->kick_fd);
vring_del_virtqueue(vq);
kfree(info);
......@@ -782,6 +837,15 @@ static int vu_setup_vq_call_fd(struct virtio_uml_device *vu_dev,
int call_fds[2];
int rc;
/* no call FD needed/desired in this case */
if (vu_dev->protocol_features &
BIT_ULL(VHOST_USER_PROTOCOL_F_INBAND_NOTIFICATIONS) &&
vu_dev->protocol_features &
BIT_ULL(VHOST_USER_PROTOCOL_F_SLAVE_REQ)) {
info->call_fd = -1;
return 0;
}
/* Use a pipe for call fd, since SIGIO is not supported for eventfd */
rc = os_pipe(call_fds, true, true);
if (rc < 0)
......@@ -810,6 +874,23 @@ static int vu_setup_vq_call_fd(struct virtio_uml_device *vu_dev,
return rc;
}
#ifdef CONFIG_UML_TIME_TRAVEL_SUPPORT
static void vu_defer_irq_handle(struct time_travel_event *d)
{
struct virtio_uml_vq_info *info;
info = container_of(d, struct virtio_uml_vq_info, defer);
info->callback(info->vq);
}
static void vu_defer_irq_callback(struct virtqueue *vq)
{
struct virtio_uml_vq_info *info = vq->priv;
time_travel_add_irq_event(&info->defer);
}
#endif
static struct virtqueue *vu_setup_vq(struct virtio_device *vdev,
unsigned index, vq_callback_t *callback,
const char *name, bool ctx)
......@@ -829,6 +910,19 @@ static struct virtqueue *vu_setup_vq(struct virtio_device *vdev,
snprintf(info->name, sizeof(info->name), "%s.%d-%s", pdev->name,
pdev->id, name);
#ifdef CONFIG_UML_TIME_TRAVEL_SUPPORT
/*
* When we get an interrupt, we must bounce it through the simulation
* calendar (the simtime device), except for the simtime device itself
* since that's part of the simulation control.
*/
if (time_travel_mode == TT_MODE_EXTERNAL && callback) {
info->callback = callback;
callback = vu_defer_irq_callback;
time_travel_set_event_fn(&info->defer, vu_defer_irq_handle);
}
#endif
vq = vring_create_virtqueue(index, num, PAGE_SIZE, vdev, true, true,
ctx, vu_notify, callback, info->name);
if (!vq) {
......@@ -837,11 +931,19 @@ static struct virtqueue *vu_setup_vq(struct virtio_device *vdev,
}
vq->priv = info;
num = virtqueue_get_vring_size(vq);
#ifdef CONFIG_UML_TIME_TRAVEL_SUPPORT
info->vq = vq;
#endif
rc = os_eventfd(0, 0);
if (rc < 0)
goto error_kick;
info->kick_fd = rc;
if (vu_dev->protocol_features &
BIT_ULL(VHOST_USER_PROTOCOL_F_INBAND_NOTIFICATIONS)) {
info->kick_fd = -1;
} else {
rc = os_eventfd(0, 0);
if (rc < 0)
goto error_kick;
info->kick_fd = rc;
}
rc = vu_setup_vq_call_fd(vu_dev, vq);
if (rc)
......@@ -866,10 +968,13 @@ static struct virtqueue *vu_setup_vq(struct virtio_device *vdev,
return vq;
error_setup:
um_free_irq(VIRTIO_IRQ, vq);
os_close_file(info->call_fd);
if (info->call_fd >= 0) {
um_free_irq(VIRTIO_IRQ, vq);
os_close_file(info->call_fd);
}
error_call:
os_close_file(info->kick_fd);
if (info->kick_fd >= 0)
os_close_file(info->kick_fd);
error_kick:
vring_del_virtqueue(vq);
error_create:
......@@ -908,10 +1013,12 @@ static int vu_find_vqs(struct virtio_device *vdev, unsigned nvqs,
list_for_each_entry(vq, &vdev->vqs, list) {
struct virtio_uml_vq_info *info = vq->priv;
rc = vhost_user_set_vring_kick(vu_dev, vq->index,
info->kick_fd);
if (rc)
goto error_setup;
if (info->kick_fd >= 0) {
rc = vhost_user_set_vring_kick(vu_dev, vq->index,
info->kick_fd);
if (rc)
goto error_setup;
}
rc = vhost_user_set_vring_enable(vu_dev, vq->index, true);
if (rc)
......@@ -1008,6 +1115,8 @@ static int virtio_uml_probe(struct platform_device *pdev)
return rc;
vu_dev->sock = rc;
spin_lock_init(&vu_dev->sock_lock);
rc = vhost_user_init(vu_dev);
if (rc)
goto error_init;
......
......@@ -3,7 +3,6 @@ generic-y += bpf_perf_event.h
generic-y += bug.h
generic-y += compat.h
generic-y += current.h
generic-y += delay.h
generic-y += device.h
generic-y += emergency-restart.h
generic-y += exec.h
......
/* SPDX-License-Identifier: GPL-2.0 */
#ifndef __UM_DELAY_H
#define __UM_DELAY_H
#include <asm-generic/delay.h>
#include <linux/time-internal.h>
static inline void um_ndelay(unsigned long nsecs)
{
if (time_travel_mode == TT_MODE_INFCPU ||
time_travel_mode == TT_MODE_EXTERNAL) {
time_travel_ndelay(nsecs);
return;
}
ndelay(nsecs);
}
#undef ndelay
#define ndelay um_ndelay
static inline void um_udelay(unsigned long usecs)
{
if (time_travel_mode == TT_MODE_INFCPU ||
time_travel_mode == TT_MODE_EXTERNAL) {
time_travel_ndelay(1000 * usecs);
return;
}
udelay(usecs);
}
#undef udelay
#define udelay um_udelay
#endif /* __UM_DELAY_H */
/* SPDX-License-Identifier: GPL-2.0 */
/*
* Copyright (C) 2012 - 2014 Cisco Systems
* Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com)
*/
#ifndef __TIMER_INTERNAL_H__
#define __TIMER_INTERNAL_H__
#include <linux/list.h>
#define TIMER_MULTIPLIER 256
#define TIMER_MIN_DELTA 500
enum time_travel_mode {
TT_MODE_OFF,
TT_MODE_BASIC,
TT_MODE_INFCPU,
TT_MODE_EXTERNAL,
};
#ifdef CONFIG_UML_TIME_TRAVEL_SUPPORT
struct time_travel_event {
unsigned long long time;
void (*fn)(struct time_travel_event *d);
struct list_head list;
bool pending, onstack;
};
extern enum time_travel_mode time_travel_mode;
void time_travel_sleep(unsigned long long duration);
static inline void
time_travel_set_event_fn(struct time_travel_event *e,
void (*fn)(struct time_travel_event *d))
{
e->fn = fn;
}
void __time_travel_propagate_time(void);
static inline void time_travel_propagate_time(void)
{
if (time_travel_mode == TT_MODE_EXTERNAL)
__time_travel_propagate_time();
}
void __time_travel_wait_readable(int fd);
static inline void time_travel_wait_readable(int fd)
{
if (time_travel_mode == TT_MODE_EXTERNAL)
__time_travel_wait_readable(fd);
}
void time_travel_add_irq_event(struct time_travel_event *e);
#else
struct time_travel_event {
};
#define time_travel_mode TT_MODE_OFF
static inline void time_travel_sleep(unsigned long long duration)
{
}
/* this is a macro so the event/function need not exist */
#define time_travel_set_event_fn(e, fn) do {} while (0)
static inline void time_travel_propagate_time(void)
{
}
static inline void time_travel_wait_readable(int fd)
{
}
#endif /* CONFIG_UML_TIME_TRAVEL_SUPPORT */
/*
* Without CONFIG_UML_TIME_TRAVEL_SUPPORT this is a linker error if used,