linux/drivers/infiniband/core/uverbs_main.c

1253 lines
35 KiB
C
Raw Normal View History

/*
* Copyright (c) 2005 Topspin Communications. All rights reserved.
* Copyright (c) 2005, 2006 Cisco Systems. All rights reserved.
* Copyright (c) 2005 Mellanox Technologies. All rights reserved.
* Copyright (c) 2005 Voltaire, Inc. All rights reserved.
* Copyright (c) 2005 PathScale, Inc. All rights reserved.
*
* This software is available to you under a choice of one of two
* licenses. You may choose to be licensed under the terms of the GNU
* General Public License (GPL) Version 2, available from the file
* COPYING in the main directory of this source tree, or the
* OpenIB.org BSD license below:
*
* Redistribution and use in source and binary forms, with or
* without modification, are permitted provided that the following
* conditions are met:
*
* - Redistributions of source code must retain the above
* copyright notice, this list of conditions and the following
* disclaimer.
*
* - Redistributions in binary form must reproduce the above
* copyright notice, this list of conditions and the following
* disclaimer in the documentation and/or other materials
* provided with the distribution.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
* NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
* BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
* ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
* CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
* SOFTWARE.
*/
#include <linux/module.h>
#include <linux/init.h>
#include <linux/device.h>
#include <linux/err.h>
#include <linux/fs.h>
#include <linux/poll.h>
#include <linux/sched.h>
#include <linux/file.h>
#include <linux/cdev.h>
#include <linux/anon_inodes.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 08:04:11 +00:00
#include <linux/slab.h>
#include <linux/uaccess.h>
#include <rdma/ib.h>
#include <rdma/uverbs_std_types.h>
#include "uverbs.h"
#include "core_priv.h"
#include "rdma_core.h"
MODULE_AUTHOR("Roland Dreier");
MODULE_DESCRIPTION("InfiniBand userspace verbs access");
MODULE_LICENSE("Dual BSD/GPL");
enum {
IB_UVERBS_MAJOR = 231,
IB_UVERBS_BASE_MINOR = 192,
IB_UVERBS_MAX_DEVICES = RDMA_MAX_PORTS,
IB_UVERBS_NUM_FIXED_MINOR = 32,
IB_UVERBS_NUM_DYNAMIC_MINOR = IB_UVERBS_MAX_DEVICES - IB_UVERBS_NUM_FIXED_MINOR,
};
#define IB_UVERBS_BASE_DEV MKDEV(IB_UVERBS_MAJOR, IB_UVERBS_BASE_MINOR)
static dev_t dynamic_uverbs_dev;
static struct class *uverbs_class;
static DECLARE_BITMAP(dev_map, IB_UVERBS_MAX_DEVICES);
static ssize_t (*uverbs_cmd_table[])(struct ib_uverbs_file *file,
const char __user *buf, int in_len,
int out_len) = {
[IB_USER_VERBS_CMD_GET_CONTEXT] = ib_uverbs_get_context,
[IB_USER_VERBS_CMD_QUERY_DEVICE] = ib_uverbs_query_device,
[IB_USER_VERBS_CMD_QUERY_PORT] = ib_uverbs_query_port,
[IB_USER_VERBS_CMD_ALLOC_PD] = ib_uverbs_alloc_pd,
[IB_USER_VERBS_CMD_DEALLOC_PD] = ib_uverbs_dealloc_pd,
[IB_USER_VERBS_CMD_REG_MR] = ib_uverbs_reg_mr,
[IB_USER_VERBS_CMD_REREG_MR] = ib_uverbs_rereg_mr,
[IB_USER_VERBS_CMD_DEREG_MR] = ib_uverbs_dereg_mr,
[IB_USER_VERBS_CMD_ALLOC_MW] = ib_uverbs_alloc_mw,
[IB_USER_VERBS_CMD_DEALLOC_MW] = ib_uverbs_dealloc_mw,
[IB_USER_VERBS_CMD_CREATE_COMP_CHANNEL] = ib_uverbs_create_comp_channel,
[IB_USER_VERBS_CMD_CREATE_CQ] = ib_uverbs_create_cq,
[IB_USER_VERBS_CMD_RESIZE_CQ] = ib_uverbs_resize_cq,
[IB_USER_VERBS_CMD_POLL_CQ] = ib_uverbs_poll_cq,
[IB_USER_VERBS_CMD_REQ_NOTIFY_CQ] = ib_uverbs_req_notify_cq,
[IB_USER_VERBS_CMD_DESTROY_CQ] = ib_uverbs_destroy_cq,
[IB_USER_VERBS_CMD_CREATE_QP] = ib_uverbs_create_qp,
[IB_USER_VERBS_CMD_QUERY_QP] = ib_uverbs_query_qp,
[IB_USER_VERBS_CMD_MODIFY_QP] = ib_uverbs_modify_qp,
[IB_USER_VERBS_CMD_DESTROY_QP] = ib_uverbs_destroy_qp,
[IB_USER_VERBS_CMD_POST_SEND] = ib_uverbs_post_send,
[IB_USER_VERBS_CMD_POST_RECV] = ib_uverbs_post_recv,
[IB_USER_VERBS_CMD_POST_SRQ_RECV] = ib_uverbs_post_srq_recv,
[IB_USER_VERBS_CMD_CREATE_AH] = ib_uverbs_create_ah,
[IB_USER_VERBS_CMD_DESTROY_AH] = ib_uverbs_destroy_ah,
[IB_USER_VERBS_CMD_ATTACH_MCAST] = ib_uverbs_attach_mcast,
[IB_USER_VERBS_CMD_DETACH_MCAST] = ib_uverbs_detach_mcast,
[IB_USER_VERBS_CMD_CREATE_SRQ] = ib_uverbs_create_srq,
[IB_USER_VERBS_CMD_MODIFY_SRQ] = ib_uverbs_modify_srq,
[IB_USER_VERBS_CMD_QUERY_SRQ] = ib_uverbs_query_srq,
[IB_USER_VERBS_CMD_DESTROY_SRQ] = ib_uverbs_destroy_srq,
[IB_USER_VERBS_CMD_OPEN_XRCD] = ib_uverbs_open_xrcd,
[IB_USER_VERBS_CMD_CLOSE_XRCD] = ib_uverbs_close_xrcd,
[IB_USER_VERBS_CMD_CREATE_XSRQ] = ib_uverbs_create_xsrq,
[IB_USER_VERBS_CMD_OPEN_QP] = ib_uverbs_open_qp,
IB/core: extended command: an improved infrastructure for uverbs commands Commit 400dbc96583f ("IB/core: Infrastructure for extensible uverbs commands") added an infrastructure for extensible uverbs commands while later commit 436f2ad05a0b ("IB/core: Export ib_create/destroy_flow through uverbs") exported ib_create_flow()/ib_destroy_flow() functions using this new infrastructure. According to the commit 400dbc96583f, the purpose of this infrastructure is to support passing around provider (eg. hardware) specific buffers when userspace issue commands to the kernel, so that it would be possible to extend uverbs (eg. core) buffers independently from the provider buffers. But the new kernel command function prototypes were not modified to take advantage of this extension. This issue was exposed by Roland Dreier in a previous review[1]. So the following patch is an attempt to a revised extensible command infrastructure. This improved extensible command infrastructure distinguish between core (eg. legacy)'s command/response buffers from provider (eg. hardware)'s command/response buffers: each extended command implementing function is given a struct ib_udata to hold core (eg. uverbs) input and output buffers, and another struct ib_udata to hold the hw (eg. provider) input and output buffers. Having those buffers identified separately make it easier to increase one buffer to support extension without having to add some code to guess the exact size of each command/response parts: This should make the extended functions more reliable. Additionally, instead of relying on command identifier being greater than IB_USER_VERBS_CMD_THRESHOLD, the proposed infrastructure rely on unused bits in command field: on the 32 bits provided by command field, only 6 bits are really needed to encode the identifier of commands currently supported by the kernel. (Even using only 6 bits leaves room for about 23 new commands). So this patch makes use of some high order bits in command field to store flags, leaving enough room for more command identifiers than one will ever need (eg. 256). The new flags are used to specify if the command should be processed as an extended one or a legacy one. While designing the new command format, care was taken to make usage of flags itself extensible. Using high order bits of the commands field ensure that newer libibverbs on older kernel will properly fail when trying to call extended commands. On the other hand, older libibverbs on newer kernel will never be able to issue calls to extended commands. The extended command header includes the optional response pointer so that output buffer length and output buffer pointer are located together in the command, allowing proper parameters checking. This should make implementing functions easier and safer. Additionally the extended header ensure 64bits alignment, while making all sizes multiple of 8 bytes, extending the maximum buffer size: legacy extended Maximum command buffer: 256KBytes 1024KBytes (512KBytes + 512KBytes) Maximum response buffer: 256KBytes 1024KBytes (512KBytes + 512KBytes) For the purpose of doing proper buffer size accounting, the headers size are no more taken in account in "in_words". One of the odds of the current extensible infrastructure, reading twice the "legacy" command header, is fixed by removing the "legacy" command header from the extended command header: they are processed as two different parts of the command: memory is read once and information are not duplicated: it's making clear that's an extended command scheme and not a different command scheme. The proposed scheme will format input (command) and output (response) buffers this way: - command: legacy header + extended header + command data (core + hw): +----------------------------------------+ | flags | 00 00 | command | | in_words | out_words | +----------------------------------------+ | response | | response | | provider_in_words | provider_out_words | | padding | +----------------------------------------+ | | . <uverbs input> . . (in_words * 8) . | | +----------------------------------------+ | | . <provider input> . . (provider_in_words * 8) . | | +----------------------------------------+ - response, if present: +----------------------------------------+ | | . <uverbs output space> . . (out_words * 8) . | | +----------------------------------------+ | | . <provider output space> . . (provider_out_words * 8) . | | +----------------------------------------+ The overall design is to ensure that the extensible infrastructure is itself extensible while begin more reliable with more input and bound checking. Note: The unused field in the extended header would be perfect candidate to hold the command "comp_mask" (eg. bit field used to handle compatibility). This was suggested by Roland Dreier in a previous review[2]. But "comp_mask" field is likely to be present in the uverb input and/or provider input, likewise for the response, as noted by Matan Barak[3], so it doesn't make sense to put "comp_mask" in the header. [1]: http://marc.info/?i=CAL1RGDWxmM17W2o_era24A-TTDeKyoL6u3NRu_=t_dhV_ZA9MA@mail.gmail.com [2]: http://marc.info/?i=CAL1RGDXJtrc849M6_XNZT5xO1+ybKtLWGq6yg6LhoSsKpsmkYA@mail.gmail.com [3]: http://marc.info/?i=525C1149.6000701@mellanox.com Signed-off-by: Yann Droneaud <ydroneaud@opteya.com> Link: http://marc.info/?i=cover.1383773832.git.ydroneaud@opteya.com [ Convert "ret ? ret : 0" to the equivalent "ret". - Roland ] Signed-off-by: Roland Dreier <roland@purestorage.com>
2013-11-06 22:21:49 +00:00
};
static int (*uverbs_ex_cmd_table[])(struct ib_uverbs_file *file,
struct ib_udata *ucore,
struct ib_udata *uhw) = {
[IB_USER_VERBS_EX_CMD_CREATE_FLOW] = ib_uverbs_ex_create_flow,
[IB_USER_VERBS_EX_CMD_DESTROY_FLOW] = ib_uverbs_ex_destroy_flow,
IB/core: Add support for extended query device caps Add extensible query device capabilities verb to allow adding new features. ib_uverbs_ex_query_device is added and copy_query_dev_fields is used to copy capability fields to be used by both ib_uverbs_query_device and ib_uverbs_ex_query_device. Following the discussion about this patch [1], the code now validates the command's comp_mask is zero, returning -EINVAL for unknown values, in order to allow extending the verb in the future. The verb also checks the user-space provided response buffer size and only fills in capabilities that will fit in the buffer. In attempt to follow the spirit of presentation [2] by Tzahi Oved that was presented during OpenFabrics Alliance International Developer Workshop 2013, the comp_mask bits will only describe which fields are valid. Furthermore, fields that can simply be cleared when they are not supported, do not require a comp_mask bit at all. The verb returns a response_length field containing the actual number of bytes written by the kernel, so that a newer version running on an older kernel can tell which fields were actually returned. [1] [PATCH v1 0/5] IB/core: extended query device caps cleanup for v3.19 http://thread.gmane.org/gmane.linux.kernel.api/7889/ [2] https://www.openfabrics.org/images/docs/2013_Dev_Workshop/Tues_0423/2013_Workshop_Tues_0830_Tzahi_Oved-verbs_extensions_ofa_2013-tzahio.pdf Signed-off-by: Eli Cohen <eli@mellanox.com> Signed-off-by: Haggai Eran <haggaie@mellanox.com> Reviewed-by: Yann Droneaud <ydroneaud@opteya.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2015-02-08 11:28:50 +00:00
[IB_USER_VERBS_EX_CMD_QUERY_DEVICE] = ib_uverbs_ex_query_device,
[IB_USER_VERBS_EX_CMD_CREATE_CQ] = ib_uverbs_ex_create_cq,
[IB_USER_VERBS_EX_CMD_CREATE_QP] = ib_uverbs_ex_create_qp,
[IB_USER_VERBS_EX_CMD_CREATE_WQ] = ib_uverbs_ex_create_wq,
[IB_USER_VERBS_EX_CMD_MODIFY_WQ] = ib_uverbs_ex_modify_wq,
[IB_USER_VERBS_EX_CMD_DESTROY_WQ] = ib_uverbs_ex_destroy_wq,
[IB_USER_VERBS_EX_CMD_CREATE_RWQ_IND_TBL] = ib_uverbs_ex_create_rwq_ind_table,
[IB_USER_VERBS_EX_CMD_DESTROY_RWQ_IND_TBL] = ib_uverbs_ex_destroy_rwq_ind_table,
[IB_USER_VERBS_EX_CMD_MODIFY_QP] = ib_uverbs_ex_modify_qp,
[IB_USER_VERBS_EX_CMD_MODIFY_CQ] = ib_uverbs_ex_modify_cq,
};
static void ib_uverbs_add_one(struct ib_device *device);
static void ib_uverbs_remove_one(struct ib_device *device, void *client_data);
/*
* Must be called with the ufile->device->disassociate_srcu held, and the lock
* must be held until use of the ucontext is finished.
*/
struct ib_ucontext *ib_uverbs_get_ucontext(struct ib_uverbs_file *ufile)
{
/*
* We do not hold the hw_destroy_rwsem lock for this flow, instead
* srcu is used. It does not matter if someone races this with
* get_context, we get NULL or valid ucontext.
*/
struct ib_ucontext *ucontext = smp_load_acquire(&ufile->ucontext);
if (!srcu_dereference(ufile->device->ib_dev,
&ufile->device->disassociate_srcu))
return ERR_PTR(-EIO);
if (!ucontext)
return ERR_PTR(-EINVAL);
return ucontext;
}
EXPORT_SYMBOL(ib_uverbs_get_ucontext);
int uverbs_dealloc_mw(struct ib_mw *mw)
{
struct ib_pd *pd = mw->pd;
int ret;
ret = mw->device->dealloc_mw(mw);
if (!ret)
atomic_dec(&pd->usecnt);
return ret;
}
static void ib_uverbs_release_dev(struct device *device)
{
struct ib_uverbs_device *dev =
container_of(device, struct ib_uverbs_device, dev);
uverbs_destroy_api(dev->uapi);
cleanup_srcu_struct(&dev->disassociate_srcu);
kfree(dev);
}
static void ib_uverbs_release_async_event_file(struct kref *ref)
{
struct ib_uverbs_async_event_file *file =
container_of(ref, struct ib_uverbs_async_event_file, ref);
kfree(file);
}
void ib_uverbs_release_ucq(struct ib_uverbs_file *file,
struct ib_uverbs_completion_event_file *ev_file,
struct ib_ucq_object *uobj)
{
struct ib_uverbs_event *evt, *tmp;
if (ev_file) {
spin_lock_irq(&ev_file->ev_queue.lock);
list_for_each_entry_safe(evt, tmp, &uobj->comp_list, obj_list) {
list_del(&evt->list);
kfree(evt);
}
spin_unlock_irq(&ev_file->ev_queue.lock);
uverbs_uobject_put(&ev_file->uobj);
}
spin_lock_irq(&file->async_file->ev_queue.lock);
list_for_each_entry_safe(evt, tmp, &uobj->async_list, obj_list) {
list_del(&evt->list);
kfree(evt);
}
spin_unlock_irq(&file->async_file->ev_queue.lock);
}
void ib_uverbs_release_uevent(struct ib_uverbs_file *file,
struct ib_uevent_object *uobj)
{
struct ib_uverbs_event *evt, *tmp;
spin_lock_irq(&file->async_file->ev_queue.lock);
list_for_each_entry_safe(evt, tmp, &uobj->event_list, obj_list) {
list_del(&evt->list);
kfree(evt);
}
spin_unlock_irq(&file->async_file->ev_queue.lock);
}
void ib_uverbs_detach_umcast(struct ib_qp *qp,
struct ib_uqp_object *uobj)
{
struct ib_uverbs_mcast_entry *mcast, *tmp;
list_for_each_entry_safe(mcast, tmp, &uobj->mcast_list, list) {
ib_detach_mcast(qp, &mcast->gid, mcast->lid);
list_del(&mcast->list);
kfree(mcast);
}
}
static void ib_uverbs_comp_dev(struct ib_uverbs_device *dev)
{
complete(&dev->comp);
}
void ib_uverbs_release_file(struct kref *ref)
{
struct ib_uverbs_file *file =
container_of(ref, struct ib_uverbs_file, ref);
struct ib_device *ib_dev;
int srcu_key;
release_ufile_idr_uobject(file);
srcu_key = srcu_read_lock(&file->device->disassociate_srcu);
ib_dev = srcu_dereference(file->device->ib_dev,
&file->device->disassociate_srcu);
if (ib_dev && !ib_dev->disassociate_ucontext)
module_put(ib_dev->owner);
srcu_read_unlock(&file->device->disassociate_srcu, srcu_key);
if (atomic_dec_and_test(&file->device->refcount))
ib_uverbs_comp_dev(file->device);
put_device(&file->device->dev);
kfree(file);
}
static ssize_t ib_uverbs_event_read(struct ib_uverbs_event_queue *ev_queue,
struct ib_uverbs_file *uverbs_file,
struct file *filp, char __user *buf,
size_t count, loff_t *pos,
size_t eventsz)
{
struct ib_uverbs_event *event;
int ret = 0;
spin_lock_irq(&ev_queue->lock);
while (list_empty(&ev_queue->event_list)) {
spin_unlock_irq(&ev_queue->lock);
if (filp->f_flags & O_NONBLOCK)
return -EAGAIN;
if (wait_event_interruptible(ev_queue->poll_wait,
(!list_empty(&ev_queue->event_list) ||
/* The barriers built into wait_event_interruptible()
* and wake_up() guarentee this will see the null set
* without using RCU
*/
!uverbs_file->device->ib_dev)))
return -ERESTARTSYS;
/* If device was disassociated and no event exists set an error */
if (list_empty(&ev_queue->event_list) &&
!uverbs_file->device->ib_dev)
return -EIO;
spin_lock_irq(&ev_queue->lock);
}
event = list_entry(ev_queue->event_list.next, struct ib_uverbs_event, list);
if (eventsz > count) {
ret = -EINVAL;
event = NULL;
} else {
list_del(ev_queue->event_list.next);
if (event->counter) {
++(*event->counter);
list_del(&event->obj_list);
}
}
spin_unlock_irq(&ev_queue->lock);
if (event) {
if (copy_to_user(buf, event, eventsz))
ret = -EFAULT;
else
ret = eventsz;
}
kfree(event);
return ret;
}
static ssize_t ib_uverbs_async_event_read(struct file *filp, char __user *buf,
size_t count, loff_t *pos)
{
struct ib_uverbs_async_event_file *file = filp->private_data;
return ib_uverbs_event_read(&file->ev_queue, file->uverbs_file, filp,
buf, count, pos,
sizeof(struct ib_uverbs_async_event_desc));
}
static ssize_t ib_uverbs_comp_event_read(struct file *filp, char __user *buf,
size_t count, loff_t *pos)
{
struct ib_uverbs_completion_event_file *comp_ev_file =
filp->private_data;
return ib_uverbs_event_read(&comp_ev_file->ev_queue,
comp_ev_file->uobj.ufile, filp,
buf, count, pos,
sizeof(struct ib_uverbs_comp_event_desc));
}
static __poll_t ib_uverbs_event_poll(struct ib_uverbs_event_queue *ev_queue,
struct file *filp,
struct poll_table_struct *wait)
{
__poll_t pollflags = 0;
poll_wait(filp, &ev_queue->poll_wait, wait);
spin_lock_irq(&ev_queue->lock);
if (!list_empty(&ev_queue->event_list))
pollflags = EPOLLIN | EPOLLRDNORM;
spin_unlock_irq(&ev_queue->lock);
return pollflags;
}
static __poll_t ib_uverbs_async_event_poll(struct file *filp,
struct poll_table_struct *wait)
{
return ib_uverbs_event_poll(filp->private_data, filp, wait);
}
static __poll_t ib_uverbs_comp_event_poll(struct file *filp,
struct poll_table_struct *wait)
{
struct ib_uverbs_completion_event_file *comp_ev_file =
filp->private_data;
return ib_uverbs_event_poll(&comp_ev_file->ev_queue, filp, wait);
}
static int ib_uverbs_async_event_fasync(int fd, struct file *filp, int on)
{
struct ib_uverbs_event_queue *ev_queue = filp->private_data;
return fasync_helper(fd, filp, on, &ev_queue->async_queue);
}
static int ib_uverbs_comp_event_fasync(int fd, struct file *filp, int on)
{
struct ib_uverbs_completion_event_file *comp_ev_file =
filp->private_data;
return fasync_helper(fd, filp, on, &comp_ev_file->ev_queue.async_queue);
}
static int ib_uverbs_async_event_close(struct inode *inode, struct file *filp)
{
struct ib_uverbs_async_event_file *file = filp->private_data;
struct ib_uverbs_file *uverbs_file = file->uverbs_file;
struct ib_uverbs_event *entry, *tmp;
int closed_already = 0;
mutex_lock(&uverbs_file->device->lists_mutex);
spin_lock_irq(&file->ev_queue.lock);
closed_already = file->ev_queue.is_closed;
file->ev_queue.is_closed = 1;
list_for_each_entry_safe(entry, tmp, &file->ev_queue.event_list, list) {
if (entry->counter)
list_del(&entry->obj_list);
kfree(entry);
}
spin_unlock_irq(&file->ev_queue.lock);
if (!closed_already) {
list_del(&file->list);
ib_unregister_event_handler(&uverbs_file->event_handler);
}
mutex_unlock(&uverbs_file->device->lists_mutex);
kref_put(&uverbs_file->ref, ib_uverbs_release_file);
kref_put(&file->ref, ib_uverbs_release_async_event_file);
return 0;
}
static int ib_uverbs_comp_event_close(struct inode *inode, struct file *filp)
{
struct ib_uobject *uobj = filp->private_data;
struct ib_uverbs_completion_event_file *file = container_of(
uobj, struct ib_uverbs_completion_event_file, uobj);
struct ib_uverbs_event *entry, *tmp;
spin_lock_irq(&file->ev_queue.lock);
list_for_each_entry_safe(entry, tmp, &file->ev_queue.event_list, list) {
if (entry->counter)
list_del(&entry->obj_list);
kfree(entry);
}
spin_unlock_irq(&file->ev_queue.lock);
uverbs_close_fd(filp);
return 0;
}
const struct file_operations uverbs_event_fops = {
.owner = THIS_MODULE,
.read = ib_uverbs_comp_event_read,
.poll = ib_uverbs_comp_event_poll,
.release = ib_uverbs_comp_event_close,
.fasync = ib_uverbs_comp_event_fasync,
.llseek = no_llseek,
};
static const struct file_operations uverbs_async_event_fops = {
.owner = THIS_MODULE,
.read = ib_uverbs_async_event_read,
.poll = ib_uverbs_async_event_poll,
.release = ib_uverbs_async_event_close,
.fasync = ib_uverbs_async_event_fasync,
.llseek = no_llseek,
};
void ib_uverbs_comp_handler(struct ib_cq *cq, void *cq_context)
{
struct ib_uverbs_event_queue *ev_queue = cq_context;
struct ib_ucq_object *uobj;
struct ib_uverbs_event *entry;
unsigned long flags;
if (!ev_queue)
return;
spin_lock_irqsave(&ev_queue->lock, flags);
if (ev_queue->is_closed) {
spin_unlock_irqrestore(&ev_queue->lock, flags);
return;
}
entry = kmalloc(sizeof(*entry), GFP_ATOMIC);
if (!entry) {
spin_unlock_irqrestore(&ev_queue->lock, flags);
return;
}
uobj = container_of(cq->uobject, struct ib_ucq_object, uobject);
entry->desc.comp.cq_handle = cq->uobject->user_handle;
entry->counter = &uobj->comp_events_reported;
list_add_tail(&entry->list, &ev_queue->event_list);
list_add_tail(&entry->obj_list, &uobj->comp_list);
spin_unlock_irqrestore(&ev_queue->lock, flags);
wake_up_interruptible(&ev_queue->poll_wait);
kill_fasync(&ev_queue->async_queue, SIGIO, POLL_IN);
}
static void ib_uverbs_async_handler(struct ib_uverbs_file *file,
__u64 element, __u64 event,
struct list_head *obj_list,
u32 *counter)
{
struct ib_uverbs_event *entry;
unsigned long flags;
spin_lock_irqsave(&file->async_file->ev_queue.lock, flags);
if (file->async_file->ev_queue.is_closed) {
spin_unlock_irqrestore(&file->async_file->ev_queue.lock, flags);
return;
}
entry = kmalloc(sizeof(*entry), GFP_ATOMIC);
if (!entry) {
spin_unlock_irqrestore(&file->async_file->ev_queue.lock, flags);
return;
}
entry->desc.async.element = element;
entry->desc.async.event_type = event;
entry->desc.async.reserved = 0;
entry->counter = counter;
list_add_tail(&entry->list, &file->async_file->ev_queue.event_list);
if (obj_list)
list_add_tail(&entry->obj_list, obj_list);
spin_unlock_irqrestore(&file->async_file->ev_queue.lock, flags);
wake_up_interruptible(&file->async_file->ev_queue.poll_wait);
kill_fasync(&file->async_file->ev_queue.async_queue, SIGIO, POLL_IN);
}
void ib_uverbs_cq_event_handler(struct ib_event *event, void *context_ptr)
{
struct ib_ucq_object *uobj = container_of(event->element.cq->uobject,
struct ib_ucq_object, uobject);
ib_uverbs_async_handler(uobj->uobject.ufile, uobj->uobject.user_handle,
event->event, &uobj->async_list,
&uobj->async_events_reported);
}
void ib_uverbs_qp_event_handler(struct ib_event *event, void *context_ptr)
{
struct ib_uevent_object *uobj;
/* for XRC target qp's, check that qp is live */
if (!event->element.qp->uobject)
return;
uobj = container_of(event->element.qp->uobject,
struct ib_uevent_object, uobject);
ib_uverbs_async_handler(context_ptr, uobj->uobject.user_handle,
event->event, &uobj->event_list,
&uobj->events_reported);
}
void ib_uverbs_wq_event_handler(struct ib_event *event, void *context_ptr)
{
struct ib_uevent_object *uobj = container_of(event->element.wq->uobject,
struct ib_uevent_object, uobject);
ib_uverbs_async_handler(context_ptr, uobj->uobject.user_handle,
event->event, &uobj->event_list,
&uobj->events_reported);
}
void ib_uverbs_srq_event_handler(struct ib_event *event, void *context_ptr)
{
struct ib_uevent_object *uobj;
uobj = container_of(event->element.srq->uobject,
struct ib_uevent_object, uobject);
ib_uverbs_async_handler(context_ptr, uobj->uobject.user_handle,
event->event, &uobj->event_list,
&uobj->events_reported);
}
void ib_uverbs_event_handler(struct ib_event_handler *handler,
struct ib_event *event)
{
struct ib_uverbs_file *file =
container_of(handler, struct ib_uverbs_file, event_handler);
ib_uverbs_async_handler(file, event->element.port_num, event->event,
NULL, NULL);
}
void ib_uverbs_free_async_event_file(struct ib_uverbs_file *file)
{
kref_put(&file->async_file->ref, ib_uverbs_release_async_event_file);
file->async_file = NULL;
}
void ib_uverbs_init_event_queue(struct ib_uverbs_event_queue *ev_queue)
{
spin_lock_init(&ev_queue->lock);
INIT_LIST_HEAD(&ev_queue->event_list);
init_waitqueue_head(&ev_queue->poll_wait);
ev_queue->is_closed = 0;
ev_queue->async_queue = NULL;
}
struct file *ib_uverbs_alloc_async_event_file(struct ib_uverbs_file *uverbs_file,
struct ib_device *ib_dev)
{
struct ib_uverbs_async_event_file *ev_file;
struct file *filp;
ev_file = kzalloc(sizeof(*ev_file), GFP_KERNEL);
if (!ev_file)
return ERR_PTR(-ENOMEM);
ib_uverbs_init_event_queue(&ev_file->ev_queue);
ev_file->uverbs_file = uverbs_file;
kref_get(&ev_file->uverbs_file->ref);
kref_init(&ev_file->ref);
filp = anon_inode_getfile("[infinibandevent]", &uverbs_async_event_fops,
ev_file, O_RDONLY);
if (IS_ERR(filp))
goto err_put_refs;
mutex_lock(&uverbs_file->device->lists_mutex);
list_add_tail(&ev_file->list,
&uverbs_file->device->uverbs_events_file_list);
mutex_unlock(&uverbs_file->device->lists_mutex);
WARN_ON(uverbs_file->async_file);
uverbs_file->async_file = ev_file;
kref_get(&uverbs_file->async_file->ref);
INIT_IB_EVENT_HANDLER(&uverbs_file->event_handler,
ib_dev,
ib_uverbs_event_handler);
ib_register_event_handler(&uverbs_file->event_handler);
/* At that point async file stuff was fully set */
return filp;
err_put_refs:
kref_put(&ev_file->uverbs_file->ref, ib_uverbs_release_file);
kref_put(&ev_file->ref, ib_uverbs_release_async_event_file);
return filp;
}
static bool verify_command_mask(struct ib_uverbs_file *ufile, u32 command,
bool extended)
{
if (!extended)
return ufile->uverbs_cmd_mask & BIT_ULL(command);
return ufile->uverbs_ex_cmd_mask & BIT_ULL(command);
}
RDMA/uverbs: Protect from command mask overflow The command number is not bounds checked against the command mask before it is shifted, resulting in an ubsan hit. This does not cause malfunction since the command number is eventually bounds checked, but we can make this ubsan clean by moving the bounds check to before the mask check. ================================================================================ UBSAN: Undefined behaviour in drivers/infiniband/core/uverbs_main.c:647:21 shift exponent 207 is too large for 64-bit type 'long long unsigned int' CPU: 0 PID: 446 Comm: syz-executor3 Not tainted 4.15.0-rc2+ #61 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.7.5-0-ge51488c-20140602_164612-nilsson.home.kraxel.org 04/01/2014 Call Trace: dump_stack+0xde/0x164 ? dma_virt_map_sg+0x22c/0x22c ubsan_epilogue+0xe/0x81 __ubsan_handle_shift_out_of_bounds+0x293/0x2f7 ? debug_check_no_locks_freed+0x340/0x340 ? __ubsan_handle_load_invalid_value+0x19b/0x19b ? lock_acquire+0x440/0x440 ? lock_acquire+0x19d/0x440 ? __might_fault+0xf4/0x240 ? ib_uverbs_write+0x68d/0xe20 ib_uverbs_write+0x68d/0xe20 ? __lock_acquire+0xcf7/0x3940 ? uverbs_devnode+0x110/0x110 ? cyc2ns_read_end+0x10/0x10 ? sched_clock_cpu+0x18/0x200 ? sched_clock_cpu+0x18/0x200 __vfs_write+0x10d/0x700 ? uverbs_devnode+0x110/0x110 ? kernel_read+0x170/0x170 ? __fget+0x35b/0x5d0 ? security_file_permission+0x93/0x260 vfs_write+0x1b0/0x550 SyS_write+0xc7/0x1a0 ? SyS_read+0x1a0/0x1a0 ? trace_hardirqs_on_thunk+0x1a/0x1c entry_SYSCALL_64_fastpath+0x18/0x85 RIP: 0033:0x448e29 RSP: 002b:00007f033f567c58 EFLAGS: 00000246 ORIG_RAX: 0000000000000001 RAX: ffffffffffffffda RBX: 00007f033f5686bc RCX: 0000000000448e29 RDX: 0000000000000060 RSI: 0000000020001000 RDI: 0000000000000012 RBP: 000000000070bea0 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000246 R12: 00000000ffffffff R13: 00000000000056a0 R14: 00000000006e8740 R15: 0000000000000000 ================================================================================ Cc: syzkaller <syzkaller@googlegroups.com> Cc: <stable@vger.kernel.org> # 4.5 Fixes: 2dbd5186a39c ("IB/core: IB/core: Allow legacy verbs through extended interfaces") Reported-by: Noa Osherovich <noaos@mellanox.com> Reviewed-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-02-13 10:18:41 +00:00
static bool verify_command_idx(u32 command, bool extended)
{
if (extended)
return command < ARRAY_SIZE(uverbs_ex_cmd_table) &&
uverbs_ex_cmd_table[command];
RDMA/uverbs: Protect from command mask overflow The command number is not bounds checked against the command mask before it is shifted, resulting in an ubsan hit. This does not cause malfunction since the command number is eventually bounds checked, but we can make this ubsan clean by moving the bounds check to before the mask check. ================================================================================ UBSAN: Undefined behaviour in drivers/infiniband/core/uverbs_main.c:647:21 shift exponent 207 is too large for 64-bit type 'long long unsigned int' CPU: 0 PID: 446 Comm: syz-executor3 Not tainted 4.15.0-rc2+ #61 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.7.5-0-ge51488c-20140602_164612-nilsson.home.kraxel.org 04/01/2014 Call Trace: dump_stack+0xde/0x164 ? dma_virt_map_sg+0x22c/0x22c ubsan_epilogue+0xe/0x81 __ubsan_handle_shift_out_of_bounds+0x293/0x2f7 ? debug_check_no_locks_freed+0x340/0x340 ? __ubsan_handle_load_invalid_value+0x19b/0x19b ? lock_acquire+0x440/0x440 ? lock_acquire+0x19d/0x440 ? __might_fault+0xf4/0x240 ? ib_uverbs_write+0x68d/0xe20 ib_uverbs_write+0x68d/0xe20 ? __lock_acquire+0xcf7/0x3940 ? uverbs_devnode+0x110/0x110 ? cyc2ns_read_end+0x10/0x10 ? sched_clock_cpu+0x18/0x200 ? sched_clock_cpu+0x18/0x200 __vfs_write+0x10d/0x700 ? uverbs_devnode+0x110/0x110 ? kernel_read+0x170/0x170 ? __fget+0x35b/0x5d0 ? security_file_permission+0x93/0x260 vfs_write+0x1b0/0x550 SyS_write+0xc7/0x1a0 ? SyS_read+0x1a0/0x1a0 ? trace_hardirqs_on_thunk+0x1a/0x1c entry_SYSCALL_64_fastpath+0x18/0x85 RIP: 0033:0x448e29 RSP: 002b:00007f033f567c58 EFLAGS: 00000246 ORIG_RAX: 0000000000000001 RAX: ffffffffffffffda RBX: 00007f033f5686bc RCX: 0000000000448e29 RDX: 0000000000000060 RSI: 0000000020001000 RDI: 0000000000000012 RBP: 000000000070bea0 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000246 R12: 00000000ffffffff R13: 00000000000056a0 R14: 00000000006e8740 R15: 0000000000000000 ================================================================================ Cc: syzkaller <syzkaller@googlegroups.com> Cc: <stable@vger.kernel.org> # 4.5 Fixes: 2dbd5186a39c ("IB/core: IB/core: Allow legacy verbs through extended interfaces") Reported-by: Noa Osherovich <noaos@mellanox.com> Reviewed-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-02-13 10:18:41 +00:00
return command < ARRAY_SIZE(uverbs_cmd_table) &&
uverbs_cmd_table[command];
RDMA/uverbs: Protect from command mask overflow The command number is not bounds checked against the command mask before it is shifted, resulting in an ubsan hit. This does not cause malfunction since the command number is eventually bounds checked, but we can make this ubsan clean by moving the bounds check to before the mask check. ================================================================================ UBSAN: Undefined behaviour in drivers/infiniband/core/uverbs_main.c:647:21 shift exponent 207 is too large for 64-bit type 'long long unsigned int' CPU: 0 PID: 446 Comm: syz-executor3 Not tainted 4.15.0-rc2+ #61 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.7.5-0-ge51488c-20140602_164612-nilsson.home.kraxel.org 04/01/2014 Call Trace: dump_stack+0xde/0x164 ? dma_virt_map_sg+0x22c/0x22c ubsan_epilogue+0xe/0x81 __ubsan_handle_shift_out_of_bounds+0x293/0x2f7 ? debug_check_no_locks_freed+0x340/0x340 ? __ubsan_handle_load_invalid_value+0x19b/0x19b ? lock_acquire+0x440/0x440 ? lock_acquire+0x19d/0x440 ? __might_fault+0xf4/0x240 ? ib_uverbs_write+0x68d/0xe20 ib_uverbs_write+0x68d/0xe20 ? __lock_acquire+0xcf7/0x3940 ? uverbs_devnode+0x110/0x110 ? cyc2ns_read_end+0x10/0x10 ? sched_clock_cpu+0x18/0x200 ? sched_clock_cpu+0x18/0x200 __vfs_write+0x10d/0x700 ? uverbs_devnode+0x110/0x110 ? kernel_read+0x170/0x170 ? __fget+0x35b/0x5d0 ? security_file_permission+0x93/0x260 vfs_write+0x1b0/0x550 SyS_write+0xc7/0x1a0 ? SyS_read+0x1a0/0x1a0 ? trace_hardirqs_on_thunk+0x1a/0x1c entry_SYSCALL_64_fastpath+0x18/0x85 RIP: 0033:0x448e29 RSP: 002b:00007f033f567c58 EFLAGS: 00000246 ORIG_RAX: 0000000000000001 RAX: ffffffffffffffda RBX: 00007f033f5686bc RCX: 0000000000448e29 RDX: 0000000000000060 RSI: 0000000020001000 RDI: 0000000000000012 RBP: 000000000070bea0 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000246 R12: 00000000ffffffff R13: 00000000000056a0 R14: 00000000006e8740 R15: 0000000000000000 ================================================================================ Cc: syzkaller <syzkaller@googlegroups.com> Cc: <stable@vger.kernel.org> # 4.5 Fixes: 2dbd5186a39c ("IB/core: IB/core: Allow legacy verbs through extended interfaces") Reported-by: Noa Osherovich <noaos@mellanox.com> Reviewed-by: Matan Barak <matanb@mellanox.com> Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
2018-02-13 10:18:41 +00:00
}
static ssize_t process_hdr(struct ib_uverbs_cmd_hdr *hdr,
u32 *command, bool *extended)
{
if (hdr->command & ~(u32)(IB_USER_VERBS_CMD_FLAG_EXTENDED |
IB_USER_VERBS_CMD_COMMAND_MASK))
return -EINVAL;
*command = hdr->command & IB_USER_VERBS_CMD_COMMAND_MASK;
*extended = hdr->command & IB_USER_VERBS_CMD_FLAG_EXTENDED;
if (!verify_command_idx(*command, *extended))
return -EOPNOTSUPP;
return 0;
}
static ssize_t verify_hdr(struct ib_uverbs_cmd_hdr *hdr,
struct ib_uverbs_ex_cmd_hdr *ex_hdr,
size_t count, bool extended)
{
if (extended) {
count -= sizeof(*hdr) + sizeof(*ex_hdr);
if ((hdr->in_words + ex_hdr->provider_in_words) * 8 != count)
return -EINVAL;
if (ex_hdr->cmd_hdr_reserved)
return -EINVAL;
if (ex_hdr->response) {
if (!hdr->out_words && !ex_hdr->provider_out_words)
return -EINVAL;
if (!access_ok(VERIFY_WRITE,
u64_to_user_ptr(ex_hdr->response),
(hdr->out_words + ex_hdr->provider_out_words) * 8))
return -EFAULT;
} else {
if (hdr->out_words || ex_hdr->provider_out_words)
return -EINVAL;
}
return 0;
}
/* not extended command */
if (hdr->in_words * 4 != count)
return -EINVAL;
return 0;
}
static ssize_t ib_uverbs_write(struct file *filp, const char __user *buf,
size_t count, loff_t *pos)
{
struct ib_uverbs_file *file = filp->private_data;
struct ib_uverbs_ex_cmd_hdr ex_hdr;
struct ib_uverbs_cmd_hdr hdr;
bool extended;
int srcu_key;
u32 command;
ssize_t ret;
if (!ib_safe_file_access(filp)) {
pr_err_once("uverbs_write: process %d (%s) changed security contexts after opening file descriptor, this is not allowed.\n",
task_tgid_vnr(current), current->comm);
return -EACCES;
}
if (count < sizeof(hdr))
return -EINVAL;
if (copy_from_user(&hdr, buf, sizeof(hdr)))
return -EFAULT;
ret = process_hdr(&hdr, &command, &extended);
if (ret)
return ret;
if (extended) {
if (count < (sizeof(hdr) + sizeof(ex_hdr)))
return -EINVAL;
if (copy_from_user(&ex_hdr, buf + sizeof(hdr), sizeof(ex_hdr)))
return -EFAULT;
}
ret = verify_hdr(&hdr, &ex_hdr, count, extended);
if (ret)
return ret;
srcu_key = srcu_read_lock(&file->device->disassociate_srcu);
if (!verify_command_mask(file, command, extended)) {
ret = -EOPNOTSUPP;
goto out;
}
buf += sizeof(hdr);
IB/core: Infrastructure for extensible uverbs commands Add infrastructure to support extended uverbs capabilities in a forward/backward manner. Uverbs command opcodes which are based on the verbs extensions approach should be greater or equal to IB_USER_VERBS_CMD_THRESHOLD. They have new header format and processed a bit differently. Whenever a specific IB_USER_VERBS_CMD_XXX is extended, which practically means it needs to have additional arguments, we will be able to add them without creating a completely new IB_USER_VERBS_CMD_YYY command or bumping the uverbs ABI version. This patch for itself doesn't provide the whole scheme which is also dependent on adding a comp_mask field to each extended uverbs command struct. The new header framework allows for future extension of the CMD arguments (ib_uverbs_cmd_hdr.in_words, ib_uverbs_cmd_hdr.out_words) for an existing new command (that is a command that supports the new uverbs command header format suggested in this patch) w/o bumping ABI version and with maintaining backward and formward compatibility to new and old libibverbs versions. In the uverbs command we are passing both uverbs arguments and the provider arguments. We split the ib_uverbs_cmd_hdr.in_words to ib_uverbs_cmd_hdr.in_words which will now carry only uverbs input argument struct size and ib_uverbs_cmd_hdr.provider_in_words that will carry the provider input argument size. Same goes for the response (the uverbs CMD output argument). For example take the create_cq call and the mlx4_ib provider: The uverbs layer gets libibverb's struct ibv_create_cq (named struct ib_uverbs_create_cq in the kernel), mlx4_ib gets libmlx4's struct mlx4_create_cq (which includes struct ibv_create_cq and is named struct mlx4_ib_create_cq in the kernel) and in_words = sizeof(mlx4_create_cq)/4 . Thus ib_uverbs_cmd_hdr.in_words carry both uverbs plus mlx4_ib input argument sizes, where uverbs assumes it knows the size of its input argument - struct ibv_create_cq. Now, if we wish to add a variable to struct ibv_create_cq, we can add a comp_mask field to the struct which is basically bit field indicating which fields exists in the struct (as done for the libibverbs API extension), but we need a way to tell what is the total size of the struct and not assume the struct size is predefined (since we may get different struct sizes from different user libibverbs versions). So we know at which point the provider input argument (struct mlx4_create_cq) begins. Same goes for extending the provider struct mlx4_create_cq. Thus we split the ib_uverbs_cmd_hdr.in_words to ib_uverbs_cmd_hdr.in_words which will now carry only uverbs input argument struct size and ib_uverbs_cmd_hdr.provider_in_words that will carry the provider (mlx4_ib) input argument size. Signed-off-by: Igor Ivanov <Igor.Ivanov@itseez.com> Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2013-08-14 10:58:29 +00:00
if (!extended) {
ret = uverbs_cmd_table[command](file, buf,
hdr.in_words * 4,
hdr.out_words * 4);
} else {
IB/core: extended command: an improved infrastructure for uverbs commands Commit 400dbc96583f ("IB/core: Infrastructure for extensible uverbs commands") added an infrastructure for extensible uverbs commands while later commit 436f2ad05a0b ("IB/core: Export ib_create/destroy_flow through uverbs") exported ib_create_flow()/ib_destroy_flow() functions using this new infrastructure. According to the commit 400dbc96583f, the purpose of this infrastructure is to support passing around provider (eg. hardware) specific buffers when userspace issue commands to the kernel, so that it would be possible to extend uverbs (eg. core) buffers independently from the provider buffers. But the new kernel command function prototypes were not modified to take advantage of this extension. This issue was exposed by Roland Dreier in a previous review[1]. So the following patch is an attempt to a revised extensible command infrastructure. This improved extensible command infrastructure distinguish between core (eg. legacy)'s command/response buffers from provider (eg. hardware)'s command/response buffers: each extended command implementing function is given a struct ib_udata to hold core (eg. uverbs) input and output buffers, and another struct ib_udata to hold the hw (eg. provider) input and output buffers. Having those buffers identified separately make it easier to increase one buffer to support extension without having to add some code to guess the exact size of each command/response parts: This should make the extended functions more reliable. Additionally, instead of relying on command identifier being greater than IB_USER_VERBS_CMD_THRESHOLD, the proposed infrastructure rely on unused bits in command field: on the 32 bits provided by command field, only 6 bits are really needed to encode the identifier of commands currently supported by the kernel. (Even using only 6 bits leaves room for about 23 new commands). So this patch makes use of some high order bits in command field to store flags, leaving enough room for more command identifiers than one will ever need (eg. 256). The new flags are used to specify if the command should be processed as an extended one or a legacy one. While designing the new command format, care was taken to make usage of flags itself extensible. Using high order bits of the commands field ensure that newer libibverbs on older kernel will properly fail when trying to call extended commands. On the other hand, older libibverbs on newer kernel will never be able to issue calls to extended commands. The extended command header includes the optional response pointer so that output buffer length and output buffer pointer are located together in the command, allowing proper parameters checking. This should make implementing functions easier and safer. Additionally the extended header ensure 64bits alignment, while making all sizes multiple of 8 bytes, extending the maximum buffer size: legacy extended Maximum command buffer: 256KBytes 1024KBytes (512KBytes + 512KBytes) Maximum response buffer: 256KBytes 1024KBytes (512KBytes + 512KBytes) For the purpose of doing proper buffer size accounting, the headers size are no more taken in account in "in_words". One of the odds of the current extensible infrastructure, reading twice the "legacy" command header, is fixed by removing the "legacy" command header from the extended command header: they are processed as two different parts of the command: memory is read once and information are not duplicated: it's making clear that's an extended command scheme and not a different command scheme. The proposed scheme will format input (command) and output (response) buffers this way: - command: legacy header + extended header + command data (core + hw): +----------------------------------------+ | flags | 00 00 | command | | in_words | out_words | +----------------------------------------+ | response | | response | | provider_in_words | provider_out_words | | padding | +----------------------------------------+ | | . <uverbs input> . . (in_words * 8) . | | +----------------------------------------+ | | . <provider input> . . (provider_in_words * 8) . | | +----------------------------------------+ - response, if present: +----------------------------------------+ | | . <uverbs output space> . . (out_words * 8) . | | +----------------------------------------+ | | . <provider output space> . . (provider_out_words * 8) . | | +----------------------------------------+ The overall design is to ensure that the extensible infrastructure is itself extensible while begin more reliable with more input and bound checking. Note: The unused field in the extended header would be perfect candidate to hold the command "comp_mask" (eg. bit field used to handle compatibility). This was suggested by Roland Dreier in a previous review[2]. But "comp_mask" field is likely to be present in the uverb input and/or provider input, likewise for the response, as noted by Matan Barak[3], so it doesn't make sense to put "comp_mask" in the header. [1]: http://marc.info/?i=CAL1RGDWxmM17W2o_era24A-TTDeKyoL6u3NRu_=t_dhV_ZA9MA@mail.gmail.com [2]: http://marc.info/?i=CAL1RGDXJtrc849M6_XNZT5xO1+ybKtLWGq6yg6LhoSsKpsmkYA@mail.gmail.com [3]: http://marc.info/?i=525C1149.6000701@mellanox.com Signed-off-by: Yann Droneaud <ydroneaud@opteya.com> Link: http://marc.info/?i=cover.1383773832.git.ydroneaud@opteya.com [ Convert "ret ? ret : 0" to the equivalent "ret". - Roland ] Signed-off-by: Roland Dreier <roland@purestorage.com>
2013-11-06 22:21:49 +00:00
struct ib_udata ucore;
struct ib_udata uhw;
buf += sizeof(ex_hdr);
IB/core: extended command: an improved infrastructure for uverbs commands Commit 400dbc96583f ("IB/core: Infrastructure for extensible uverbs commands") added an infrastructure for extensible uverbs commands while later commit 436f2ad05a0b ("IB/core: Export ib_create/destroy_flow through uverbs") exported ib_create_flow()/ib_destroy_flow() functions using this new infrastructure. According to the commit 400dbc96583f, the purpose of this infrastructure is to support passing around provider (eg. hardware) specific buffers when userspace issue commands to the kernel, so that it would be possible to extend uverbs (eg. core) buffers independently from the provider buffers. But the new kernel command function prototypes were not modified to take advantage of this extension. This issue was exposed by Roland Dreier in a previous review[1]. So the following patch is an attempt to a revised extensible command infrastructure. This improved extensible command infrastructure distinguish between core (eg. legacy)'s command/response buffers from provider (eg. hardware)'s command/response buffers: each extended command implementing function is given a struct ib_udata to hold core (eg. uverbs) input and output buffers, and another struct ib_udata to hold the hw (eg. provider) input and output buffers. Having those buffers identified separately make it easier to increase one buffer to support extension without having to add some code to guess the exact size of each command/response parts: This should make the extended functions more reliable. Additionally, instead of relying on command identifier being greater than IB_USER_VERBS_CMD_THRESHOLD, the proposed infrastructure rely on unused bits in command field: on the 32 bits provided by command field, only 6 bits are really needed to encode the identifier of commands currently supported by the kernel. (Even using only 6 bits leaves room for about 23 new commands). So this patch makes use of some high order bits in command field to store flags, leaving enough room for more command identifiers than one will ever need (eg. 256). The new flags are used to specify if the command should be processed as an extended one or a legacy one. While designing the new command format, care was taken to make usage of flags itself extensible. Using high order bits of the commands field ensure that newer libibverbs on older kernel will properly fail when trying to call extended commands. On the other hand, older libibverbs on newer kernel will never be able to issue calls to extended commands. The extended command header includes the optional response pointer so that output buffer length and output buffer pointer are located together in the command, allowing proper parameters checking. This should make implementing functions easier and safer. Additionally the extended header ensure 64bits alignment, while making all sizes multiple of 8 bytes, extending the maximum buffer size: legacy extended Maximum command buffer: 256KBytes 1024KBytes (512KBytes + 512KBytes) Maximum response buffer: 256KBytes 1024KBytes (512KBytes + 512KBytes) For the purpose of doing proper buffer size accounting, the headers size are no more taken in account in "in_words". One of the odds of the current extensible infrastructure, reading twice the "legacy" command header, is fixed by removing the "legacy" command header from the extended command header: they are processed as two different parts of the command: memory is read once and information are not duplicated: it's making clear that's an extended command scheme and not a different command scheme. The proposed scheme will format input (command) and output (response) buffers this way: - command: legacy header + extended header + command data (core + hw): +----------------------------------------+ | flags | 00 00 | command | | in_words | out_words | +----------------------------------------+ | response | | response | | provider_in_words | provider_out_words | | padding | +----------------------------------------+ | | . <uverbs input> . . (in_words * 8) . | | +----------------------------------------+ | | . <provider input> . . (provider_in_words * 8) . | | +----------------------------------------+ - response, if present: +----------------------------------------+ | | . <uverbs output space> . . (out_words * 8) . | | +----------------------------------------+ | | . <provider output space> . . (provider_out_words * 8) . | | +----------------------------------------+ The overall design is to ensure that the extensible infrastructure is itself extensible while begin more reliable with more input and bound checking. Note: The unused field in the extended header would be perfect candidate to hold the command "comp_mask" (eg. bit field used to handle compatibility). This was suggested by Roland Dreier in a previous review[2]. But "comp_mask" field is likely to be present in the uverb input and/or provider input, likewise for the response, as noted by Matan Barak[3], so it doesn't make sense to put "comp_mask" in the header. [1]: http://marc.info/?i=CAL1RGDWxmM17W2o_era24A-TTDeKyoL6u3NRu_=t_dhV_ZA9MA@mail.gmail.com [2]: http://marc.info/?i=CAL1RGDXJtrc849M6_XNZT5xO1+ybKtLWGq6yg6LhoSsKpsmkYA@mail.gmail.com [3]: http://marc.info/?i=525C1149.6000701@mellanox.com Signed-off-by: Yann Droneaud <ydroneaud@opteya.com> Link: http://marc.info/?i=cover.1383773832.git.ydroneaud@opteya.com [ Convert "ret ? ret : 0" to the equivalent "ret". - Roland ] Signed-off-by: Roland Dreier <roland@purestorage.com>
2013-11-06 22:21:49 +00:00
ib_uverbs_init_udata_buf_or_null(&ucore, buf,
u64_to_user_ptr(ex_hdr.response),
hdr.in_words * 8, hdr.out_words * 8);
ib_uverbs_init_udata_buf_or_null(&uhw,
buf + ucore.inlen,
u64_to_user_ptr(ex_hdr.response) + ucore.outlen,
ex_hdr.provider_in_words * 8,
ex_hdr.provider_out_words * 8);
IB/core: extended command: an improved infrastructure for uverbs commands Commit 400dbc96583f ("IB/core: Infrastructure for extensible uverbs commands") added an infrastructure for extensible uverbs commands while later commit 436f2ad05a0b ("IB/core: Export ib_create/destroy_flow through uverbs") exported ib_create_flow()/ib_destroy_flow() functions using this new infrastructure. According to the commit 400dbc96583f, the purpose of this infrastructure is to support passing around provider (eg. hardware) specific buffers when userspace issue commands to the kernel, so that it would be possible to extend uverbs (eg. core) buffers independently from the provider buffers. But the new kernel command function prototypes were not modified to take advantage of this extension. This issue was exposed by Roland Dreier in a previous review[1]. So the following patch is an attempt to a revised extensible command infrastructure. This improved extensible command infrastructure distinguish between core (eg. legacy)'s command/response buffers from provider (eg. hardware)'s command/response buffers: each extended command implementing function is given a struct ib_udata to hold core (eg. uverbs) input and output buffers, and another struct ib_udata to hold the hw (eg. provider) input and output buffers. Having those buffers identified separately make it easier to increase one buffer to support extension without having to add some code to guess the exact size of each command/response parts: This should make the extended functions more reliable. Additionally, instead of relying on command identifier being greater than IB_USER_VERBS_CMD_THRESHOLD, the proposed infrastructure rely on unused bits in command field: on the 32 bits provided by command field, only 6 bits are really needed to encode the identifier of commands currently supported by the kernel. (Even using only 6 bits leaves room for about 23 new commands). So this patch makes use of some high order bits in command field to store flags, leaving enough room for more command identifiers than one will ever need (eg. 256). The new flags are used to specify if the command should be processed as an extended one or a legacy one. While designing the new command format, care was taken to make usage of flags itself extensible. Using high order bits of the commands field ensure that newer libibverbs on older kernel will properly fail when trying to call extended commands. On the other hand, older libibverbs on newer kernel will never be able to issue calls to extended commands. The extended command header includes the optional response pointer so that output buffer length and output buffer pointer are located together in the command, allowing proper parameters checking. This should make implementing functions easier and safer. Additionally the extended header ensure 64bits alignment, while making all sizes multiple of 8 bytes, extending the maximum buffer size: legacy extended Maximum command buffer: 256KBytes 1024KBytes (512KBytes + 512KBytes) Maximum response buffer: 256KBytes 1024KBytes (512KBytes + 512KBytes) For the purpose of doing proper buffer size accounting, the headers size are no more taken in account in "in_words". One of the odds of the current extensible infrastructure, reading twice the "legacy" command header, is fixed by removing the "legacy" command header from the extended command header: they are processed as two different parts of the command: memory is read once and information are not duplicated: it's making clear that's an extended command scheme and not a different command scheme. The proposed scheme will format input (command) and output (response) buffers this way: - command: legacy header + extended header + command data (core + hw): +----------------------------------------+ | flags | 00 00 | command | | in_words | out_words | +----------------------------------------+ | response | | response | | provider_in_words | provider_out_words | | padding | +----------------------------------------+ | | . <uverbs input> . . (in_words * 8) . | | +----------------------------------------+ | | . <provider input> . . (provider_in_words * 8) . | | +----------------------------------------+ - response, if present: +----------------------------------------+ | | . <uverbs output space> . . (out_words * 8) . | | +----------------------------------------+ | | . <provider output space> . . (provider_out_words * 8) . | | +----------------------------------------+ The overall design is to ensure that the extensible infrastructure is itself extensible while begin more reliable with more input and bound checking. Note: The unused field in the extended header would be perfect candidate to hold the command "comp_mask" (eg. bit field used to handle compatibility). This was suggested by Roland Dreier in a previous review[2]. But "comp_mask" field is likely to be present in the uverb input and/or provider input, likewise for the response, as noted by Matan Barak[3], so it doesn't make sense to put "comp_mask" in the header. [1]: http://marc.info/?i=CAL1RGDWxmM17W2o_era24A-TTDeKyoL6u3NRu_=t_dhV_ZA9MA@mail.gmail.com [2]: http://marc.info/?i=CAL1RGDXJtrc849M6_XNZT5xO1+ybKtLWGq6yg6LhoSsKpsmkYA@mail.gmail.com [3]: http://marc.info/?i=525C1149.6000701@mellanox.com Signed-off-by: Yann Droneaud <ydroneaud@opteya.com> Link: http://marc.info/?i=cover.1383773832.git.ydroneaud@opteya.com [ Convert "ret ? ret : 0" to the equivalent "ret". - Roland ] Signed-off-by: Roland Dreier <roland@purestorage.com>
2013-11-06 22:21:49 +00:00
ret = uverbs_ex_cmd_table[command](file, &ucore, &uhw);
ret = (ret) ? : count;
IB/core: Infrastructure for extensible uverbs commands Add infrastructure to support extended uverbs capabilities in a forward/backward manner. Uverbs command opcodes which are based on the verbs extensions approach should be greater or equal to IB_USER_VERBS_CMD_THRESHOLD. They have new header format and processed a bit differently. Whenever a specific IB_USER_VERBS_CMD_XXX is extended, which practically means it needs to have additional arguments, we will be able to add them without creating a completely new IB_USER_VERBS_CMD_YYY command or bumping the uverbs ABI version. This patch for itself doesn't provide the whole scheme which is also dependent on adding a comp_mask field to each extended uverbs command struct. The new header framework allows for future extension of the CMD arguments (ib_uverbs_cmd_hdr.in_words, ib_uverbs_cmd_hdr.out_words) for an existing new command (that is a command that supports the new uverbs command header format suggested in this patch) w/o bumping ABI version and with maintaining backward and formward compatibility to new and old libibverbs versions. In the uverbs command we are passing both uverbs arguments and the provider arguments. We split the ib_uverbs_cmd_hdr.in_words to ib_uverbs_cmd_hdr.in_words which will now carry only uverbs input argument struct size and ib_uverbs_cmd_hdr.provider_in_words that will carry the provider input argument size. Same goes for the response (the uverbs CMD output argument). For example take the create_cq call and the mlx4_ib provider: The uverbs layer gets libibverb's struct ibv_create_cq (named struct ib_uverbs_create_cq in the kernel), mlx4_ib gets libmlx4's struct mlx4_create_cq (which includes struct ibv_create_cq and is named struct mlx4_ib_create_cq in the kernel) and in_words = sizeof(mlx4_create_cq)/4 . Thus ib_uverbs_cmd_hdr.in_words carry both uverbs plus mlx4_ib input argument sizes, where uverbs assumes it knows the size of its input argument - struct ibv_create_cq. Now, if we wish to add a variable to struct ibv_create_cq, we can add a comp_mask field to the struct which is basically bit field indicating which fields exists in the struct (as done for the libibverbs API extension), but we need a way to tell what is the total size of the struct and not assume the struct size is predefined (since we may get different struct sizes from different user libibverbs versions). So we know at which point the provider input argument (struct mlx4_create_cq) begins. Same goes for extending the provider struct mlx4_create_cq. Thus we split the ib_uverbs_cmd_hdr.in_words to ib_uverbs_cmd_hdr.in_words which will now carry only uverbs input argument struct size and ib_uverbs_cmd_hdr.provider_in_words that will carry the provider (mlx4_ib) input argument size. Signed-off-by: Igor Ivanov <Igor.Ivanov@itseez.com> Signed-off-by: Hadar Hen Zion <hadarh@mellanox.com> Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com> Signed-off-by: Roland Dreier <roland@purestorage.com>
2013-08-14 10:58:29 +00:00
}
IB/core: extended command: an improved infrastructure for uverbs commands Commit 400dbc96583f ("IB/core: Infrastructure for extensible uverbs commands") added an infrastructure for extensible uverbs commands while later commit 436f2ad05a0b ("IB/core: Export ib_create/destroy_flow through uverbs") exported ib_create_flow()/ib_destroy_flow() functions using this new infrastructure. According to the commit 400dbc96583f, the purpose of this infrastructure is to support passing around provider (eg. hardware) specific buffers when userspace issue commands to the kernel, so that it would be possible to extend uverbs (eg. core) buffers independently from the provider buffers. But the new kernel command function prototypes were not modified to take advantage of this extension. This issue was exposed by Roland Dreier in a previous review[1]. So the following patch is an attempt to a revised extensible command infrastructure. This improved extensible command infrastructure distinguish between core (eg. legacy)'s command/response buffers from provider (eg. hardware)'s command/response buffers: each extended command implementing function is given a struct ib_udata to hold core (eg. uverbs) input and output buffers, and another struct ib_udata to hold the hw (eg. provider) input and output buffers. Having those buffers identified separately make it easier to increase one buffer to support extension without having to add some code to guess the exact size of each command/response parts: This should make the extended functions more reliable. Additionally, instead of relying on command identifier being greater than IB_USER_VERBS_CMD_THRESHOLD, the proposed infrastructure rely on unused bits in command field: on the 32 bits provided by command field, only 6 bits are really needed to encode the identifier of commands currently supported by the kernel. (Even using only 6 bits leaves room for about 23 new commands). So this patch makes use of some high order bits in command field to store flags, leaving enough room for more command identifiers than one will ever need (eg. 256). The new flags are used to specify if the command should be processed as an extended one or a legacy one. While designing the new command format, care was taken to make usage of flags itself extensible. Using high order bits of the commands field ensure that newer libibverbs on older kernel will properly fail when trying to call extended commands. On the other hand, older libibverbs on newer kernel will never be able to issue calls to extended commands. The extended command header includes the optional response pointer so that output buffer length and output buffer pointer are located together in the command, allowing proper parameters checking. This should make implementing functions easier and safer. Additionally the extended header ensure 64bits alignment, while making all sizes multiple of 8 bytes, extending the maximum buffer size: legacy extended Maximum command buffer: 256KBytes 1024KBytes (512KBytes + 512KBytes) Maximum response buffer: 256KBytes 1024KBytes (512KBytes + 512KBytes) For the purpose of doing proper buffer size accounting, the headers size are no more taken in account in "in_words". One of the odds of the current extensible infrastructure, reading twice the "legacy" command header, is fixed by removing the "legacy" command header from the extended command header: they are processed as two different parts of the command: memory is read once and information are not duplicated: it's making clear that's an extended command scheme and not a different command scheme. The proposed scheme will format input (command) and output (response) buffers this way: - command: legacy header + extended header + command data (core + hw): +----------------------------------------+ | flags | 00 00 | command | | in_words | out_words | +----------------------------------------+ | response | | response | | provider_in_words | provider_out_words | | padding | +----------------------------------------+ | | . <uverbs input> . . (in_words * 8) . | | +----------------------------------------+ | | . <provider input> . . (provider_in_words * 8) . | | +----------------------------------------+ - response, if present: +----------------------------------------+ | | . <uverbs output space> . . (out_words * 8) . | | +----------------------------------------+ | | . <provider output space> . . (provider_out_words * 8) . | | +----------------------------------------+ The overall design is to ensure that the extensible infrastructure is itself extensible while begin more reliable with more input and bound checking. Note: The unused field in the extended header would be perfect candidate to hold the command "comp_mask" (eg. bit field used to handle compatibility). This was suggested by Roland Dreier in a previous review[2]. But "comp_mask" field is likely to be present in the uverb input and/or provider input, likewise for the response, as noted by Matan Barak[3], so it doesn't make sense to put "comp_mask" in the header. [1]: http://marc.info/?i=CAL1RGDWxmM17W2o_era24A-TTDeKyoL6u3NRu_=t_dhV_ZA9MA@mail.gmail.com [2]: http://marc.info/?i=CAL1RGDXJtrc849M6_XNZT5xO1+ybKtLWGq6yg6LhoSsKpsmkYA@mail.gmail.com [3]: http://marc.info/?i=525C1149.6000701@mellanox.com Signed-off-by: Yann Droneaud <ydroneaud@opteya.com> Link: http://marc.info/?i=cover.1383773832.git.ydroneaud@opteya.com [ Convert "ret ? ret : 0" to the equivalent "ret". - Roland ] Signed-off-by: Roland Dreier <roland@purestorage.com>
2013-11-06 22:21:49 +00:00
out:
srcu_read_unlock(&file->device->disassociate_srcu, srcu_key);
return ret;
}
static int ib_uverbs_mmap(struct file *filp, struct vm_area_struct *vma)
{
struct ib_uverbs_file *file = filp->private_data;
struct ib_ucontext *ucontext;
int ret = 0;
int srcu_key;
srcu_key = srcu_read_lock(&file->device->disassociate_srcu);
ucontext = ib_uverbs_get_ucontext(file);
if (IS_ERR(ucontext)) {
ret = PTR_ERR(ucontext);
goto out;
}
ret = ucontext->device->mmap(ucontext, vma);
out:
srcu_read_unlock(&file->device->disassociate_srcu, srcu_key);
return ret;
}
/*
* ib_uverbs_open() does not need the BKL:
*
* - the ib_uverbs_device structures are properly reference counted and
* everything else is purely local to the file being created, so
* races against other open calls are not a problem;
* - there is no ioctl method to race against;
* - the open method will either immediately run -ENXIO, or all
* required initialization will be done.
*/
static int ib_uverbs_open(struct inode *inode, struct file *filp)
{
struct ib_uverbs_device *dev;
struct ib_uverbs_file *file;
struct ib_device *ib_dev;
int ret;
int module_dependent;
int srcu_key;
dev = container_of(inode->i_cdev, struct ib_uverbs_device, cdev);
if (!atomic_inc_not_zero(&dev->refcount))
return -ENXIO;
get_device(&dev->dev);
srcu_key = srcu_read_lock(&dev->disassociate_srcu);
mutex_lock(&dev->lists_mutex);
ib_dev = srcu_dereference(dev->ib_dev,
&dev->disassociate_srcu);
if (!ib_dev) {
ret = -EIO;
goto err;
}
/* In case IB device supports disassociate ucontext, there is no hard
* dependency between uverbs device and its low level device.
*/
module_dependent = !(ib_dev->disassociate_ucontext);
if (module_dependent) {
if (!try_module_get(ib_dev->owner)) {
ret = -ENODEV;
goto err;
}
}
file = kzalloc(sizeof(*file), GFP_KERNEL);
if (!file) {
ret = -ENOMEM;
if (module_dependent)
goto err_module;
goto err;
}
file->device = dev;
kref_init(&file->ref);
mutex_init(&file->ucontext_lock);
spin_lock_init(&file->uobjects_lock);
INIT_LIST_HEAD(&file->uobjects);
init_rwsem(&file->hw_destroy_rwsem);
filp->private_data = file;
list_add_tail(&file->list, &dev->uverbs_file_list);
mutex_unlock(&dev->lists_mutex);
srcu_read_unlock(&dev->disassociate_srcu, srcu_key);
file->uverbs_cmd_mask = ib_dev->uverbs_cmd_mask;
file->uverbs_ex_cmd_mask = ib_dev->uverbs_ex_cmd_mask;
setup_ufile_idr_uobject(file);
return nonseekable_open(inode, filp);
err_module:
module_put(ib_dev->owner);
err:
mutex_unlock(&dev->lists_mutex);
srcu_read_unlock(&dev->disassociate_srcu, srcu_key);
if (atomic_dec_and_test(&dev->refcount))
ib_uverbs_comp_dev(dev);
put_device(&dev->dev);
return ret;
}
static int ib_uverbs_close(struct inode *inode, struct file *filp)
{
struct ib_uverbs_file *file = filp->private_data;
uverbs_destroy_ufile_hw(file, RDMA_REMOVE_CLOSE);
mutex_lock(&file->device->lists_mutex);
list_del_init(&file->list);
mutex_unlock(&file->device->lists_mutex);
if (file->async_file)
kref_put(&file->async_file->ref,
ib_uverbs_release_async_event_file);
kref_put(&file->ref, ib_uverbs_release_file);
return 0;
}
static const struct file_operations uverbs_fops = {
.owner = THIS_MODULE,
.write = ib_uverbs_write,
.open = ib_uverbs_open,
.release = ib_uverbs_close,
.llseek = no_llseek,
.unlocked_ioctl = ib_uverbs_ioctl,
.compat_ioctl = ib_uverbs_ioctl,
};
static const struct file_operations uverbs_mmap_fops = {
.owner = THIS_MODULE,
.write = ib_uverbs_write,
.mmap = ib_uverbs_mmap,
.open = ib_uverbs_open,
.release = ib_uverbs_close,
.llseek = no_llseek,
.unlocked_ioctl = ib_uverbs_ioctl,
.compat_ioctl = ib_uverbs_ioctl,
};
static struct ib_client uverbs_client = {
.name = "uverbs",
.add = ib_uverbs_add_one,
.remove = ib_uverbs_remove_one
};
static ssize_t ibdev_show(struct device *device, struct device_attribute *attr,
char *buf)
{
struct ib_uverbs_device *dev =
container_of(device, struct ib_uverbs_device, dev);
int ret = -ENODEV;
int srcu_key;
struct ib_device *ib_dev;
srcu_key = srcu_read_lock(&dev->disassociate_srcu);
ib_dev = srcu_dereference(dev->ib_dev, &dev->disassociate_srcu);
if (ib_dev)
ret = sprintf(buf, "%s\n", ib_dev->name);
srcu_read_unlock(&dev->disassociate_srcu, srcu_key);
return ret;
}
static DEVICE_ATTR_RO(ibdev);
static ssize_t abi_version_show(struct device *device,
struct device_attribute *attr, char *buf)
{
struct ib_uverbs_device *dev =
container_of(device, struct ib_uverbs_device, dev);
int ret = -ENODEV;
int srcu_key;
struct ib_device *ib_dev;
srcu_key = srcu_read_lock(&dev->disassociate_srcu);
ib_dev = srcu_dereference(dev->ib_dev, &dev->disassociate_srcu);
if (ib_dev)
ret = sprintf(buf, "%d\n", ib_dev->uverbs_abi_ver);
srcu_read_unlock(&dev->disassociate_srcu, srcu_key);
return ret;
}
static DEVICE_ATTR_RO(abi_version);
static struct attribute *ib_dev_attrs[] = {
&dev_attr_abi_version.attr,
&dev_attr_ibdev.attr,
NULL,
};
static const struct attribute_group dev_attr_group = {
.attrs = ib_dev_attrs,
};
static CLASS_ATTR_STRING(abi_version, S_IRUGO,
__stringify(IB_USER_VERBS_ABI_VERSION));
static int ib_uverbs_create_uapi(struct ib_device *device,
struct ib_uverbs_device *uverbs_dev)
{
struct uverbs_api *uapi;
uapi = uverbs_alloc_api(device->driver_specs, device->driver_id);
if (IS_ERR(uapi))
return PTR_ERR(uapi);
uverbs_dev->uapi = uapi;
return 0;
}
static void ib_uverbs_add_one(struct ib_device *device)
{
int devnum;
dev_t base;
struct ib_uverbs_device *uverbs_dev;
int ret;
if (!device->alloc_ucontext)
return;
uverbs_dev = kzalloc(sizeof(*uverbs_dev), GFP_KERNEL);
if (!uverbs_dev)
return;
ret = init_srcu_struct(&uverbs_dev->disassociate_srcu);
if (ret) {
kfree(uverbs_dev);
return;
}
device_initialize(&uverbs_dev->dev);
uverbs_dev->dev.class = uverbs_class;
uverbs_dev->dev.parent = device->dev.parent;
uverbs_dev->dev.release = ib_uverbs_release_dev;
uverbs_dev->groups[0] = &dev_attr_group;
uverbs_dev->dev.groups = uverbs_dev->groups;
atomic_set(&uverbs_dev->refcount, 1);
init_completion(&uverbs_dev->comp);
uverbs_dev->xrcd_tree = RB_ROOT;
mutex_init(&uverbs_dev->xrcd_tree_mutex);
mutex_init(&uverbs_dev->lists_mutex);
INIT_LIST_HEAD(&uverbs_dev->uverbs_file_list);
INIT_LIST_HEAD(&uverbs_dev->uverbs_events_file_list);
rcu_assign_pointer(uverbs_dev->ib_dev, device);
uverbs_dev->num_comp_vectors = device->num_comp_vectors;
devnum = find_first_zero_bit(dev_map, IB_UVERBS_MAX_DEVICES);
if (devnum >= IB_UVERBS_MAX_DEVICES)
goto err;
uverbs_dev->devnum = devnum;
set_bit(devnum, dev_map);
if (devnum >= IB_UVERBS_NUM_FIXED_MINOR)
base = dynamic_uverbs_dev + devnum - IB_UVERBS_NUM_FIXED_MINOR;
else
base = IB_UVERBS_BASE_DEV + devnum;
if (ib_uverbs_create_uapi(device, uverbs_dev))
goto err_uapi;
uverbs_dev->dev.devt = base;
dev_set_name(&uverbs_dev->dev, "uverbs%d", uverbs_dev->devnum);
cdev_init(&uverbs_dev->cdev,
device->mmap ? &uverbs_mmap_fops : &uverbs_fops);
uverbs_dev->cdev.owner = THIS_MODULE;
ret = cdev_device_add(&uverbs_dev->cdev, &uverbs_dev->dev);
if (ret)
goto err_uapi;
ib_set_client_data(device, &uverbs_client, uverbs_dev);
return;
err_uapi:
clear_bit(devnum, dev_map);
err:
if (atomic_dec_and_test(&uverbs_dev->refcount))
ib_uverbs_comp_dev(uverbs_dev);
wait_for_completion(&uverbs_dev->comp);
put_device(&uverbs_dev->dev);
return;
}
static void ib_uverbs_free_hw_resources(struct ib_uverbs_device *uverbs_dev,
struct ib_device *ib_dev)
{
struct ib_uverbs_file *file;
struct ib_uverbs_async_event_file *event_file;
struct ib_event event;
/* Pending running commands to terminate */
uverbs_disassociate_api_pre(uverbs_dev);
event.event = IB_EVENT_DEVICE_FATAL;
event.element.port_num = 0;
event.device = ib_dev;
mutex_lock(&uverbs_dev->lists_mutex);
while (!list_empty(&uverbs_dev->uverbs_file_list)) {
file = list_first_entry(&uverbs_dev->uverbs_file_list,
struct ib_uverbs_file, list);
list_del_init(&file->list);
kref_get(&file->ref);
/* We must release the mutex before going ahead and calling
* uverbs_cleanup_ufile, as it might end up indirectly calling
* uverbs_close, for example due to freeing the resources (e.g
* mmput).
*/
mutex_unlock(&uverbs_dev->lists_mutex);
ib_uverbs_event_handler(&file->event_handler, &event);
uverbs_destroy_ufile_hw(file, RDMA_REMOVE_DRIVER_REMOVE);
kref_put(&file->ref, ib_uverbs_release_file);
mutex_lock(&uverbs_dev->lists_mutex);
}
while (!list_empty(&uverbs_dev->uverbs_events_file_list)) {
event_file = list_first_entry(&uverbs_dev->
uverbs_events_file_list,
struct ib_uverbs_async_event_file,
list);
spin_lock_irq(&event_file->ev_queue.lock);
event_file->ev_queue.is_closed = 1;
spin_unlock_irq(&event_file->ev_queue.lock);
list_del(&event_file->list);
ib_unregister_event_handler(
&event_file->uverbs_file->event_handler);
event_file->uverbs_file->event_handler.device =
NULL;
wake_up_interruptible(&event_file->ev_queue.poll_wait);
kill_fasync(&event_file->ev_queue.async_queue, SIGIO, POLL_IN);
}
mutex_unlock(&uverbs_dev->lists_mutex);
uverbs_disassociate_api(uverbs_dev->uapi);
}
static void ib_uverbs_remove_one(struct ib_device *device, void *client_data)
{
struct ib_uverbs_device *uverbs_dev = client_data;
int wait_clients = 1;
if (!uverbs_dev)
return;
cdev_device_del(&uverbs_dev->cdev, &uverbs_dev->dev);
clear_bit(uverbs_dev->devnum, dev_map);
if (device->disassociate_ucontext) {
/* We disassociate HW resources and immediately return.
* Userspace will see a EIO errno for all future access.
* Upon returning, ib_device may be freed internally and is not
* valid any more.
* uverbs_device is still available until all clients close
* their files, then the uverbs device ref count will be zero
* and its resources will be freed.
* Note: At this point no more files can be opened since the
* cdev was deleted, however active clients can still issue
* commands and close their open files.
*/
ib_uverbs_free_hw_resources(uverbs_dev, device);
wait_clients = 0;
}
if (atomic_dec_and_test(&uverbs_dev->refcount))
ib_uverbs_comp_dev(uverbs_dev);
if (wait_clients)
wait_for_completion(&uverbs_dev->comp);
put_device(&uverbs_dev->dev);
}
static char *uverbs_devnode(struct device *dev, umode_t *mode)
{
if (mode)
*mode = 0666;
return kasprintf(GFP_KERNEL, "infiniband/%s", dev_name(dev));
}
static int __init ib_uverbs_init(void)
{
int ret;
ret = register_chrdev_region(IB_UVERBS_BASE_DEV,
IB_UVERBS_NUM_FIXED_MINOR,
"infiniband_verbs");
if (ret) {
pr_err("user_verbs: couldn't register device number\n");
goto out;
}
ret = alloc_chrdev_region(&dynamic_uverbs_dev, 0,
IB_UVERBS_NUM_DYNAMIC_MINOR,
"infiniband_verbs");
if (ret) {
pr_err("couldn't register dynamic device number\n");
goto out_alloc;
}
uverbs_class = class_create(THIS_MODULE, "infiniband_verbs");
if (IS_ERR(uverbs_class)) {
ret = PTR_ERR(uverbs_class);
pr_err("user_verbs: couldn't create class infiniband_verbs\n");
goto out_chrdev;
}
uverbs_class->devnode = uverbs_devnode;
ret = class_create_file(uverbs_class, &class_attr_abi_version.attr);
if (ret) {
pr_err("user_verbs: couldn't create abi_version attribute\n");
goto out_class;
}
ret = ib_register_client(&uverbs_client);
if (ret) {
pr_err("user_verbs: couldn't register client\n");
goto out_class;
}
return 0;
out_class:
class_destroy(uverbs_class);
out_chrdev:
unregister_chrdev_region(dynamic_uverbs_dev,
IB_UVERBS_NUM_DYNAMIC_MINOR);
out_alloc:
unregister_chrdev_region(IB_UVERBS_BASE_DEV,
IB_UVERBS_NUM_FIXED_MINOR);
out:
return ret;
}
static void __exit ib_uverbs_cleanup(void)
{
ib_unregister_client(&uverbs_client);
class_destroy(uverbs_class);
unregister_chrdev_region(IB_UVERBS_BASE_DEV,
IB_UVERBS_NUM_FIXED_MINOR);
unregister_chrdev_region(dynamic_uverbs_dev,
IB_UVERBS_NUM_DYNAMIC_MINOR);
}
module_init(ib_uverbs_init);
module_exit(ib_uverbs_cleanup);