summaryrefslogtreecommitdiff
path: root/drivers/infiniband/hw/hfi1/file_ops.c
diff options
context:
space:
mode:
authorJames Morris <james.l.morris@oracle.com>2017-11-29 12:47:41 +1100
committerJames Morris <james.l.morris@oracle.com>2017-11-29 12:47:41 +1100
commitcf40a76e7d5874bb25f4404eecc58a2e033af885 (patch)
tree8fd81cbea03c87b3d41d7ae5b1d11eadd35d6ef5 /drivers/infiniband/hw/hfi1/file_ops.c
parentab5348c9c23cd253f5902980d2d8fe067dc24c82 (diff)
parent4fbd8d194f06c8a3fd2af1ce560ddb31f7ec8323 (diff)
Merge tag 'v4.15-rc1' into next-seccomp
Linux 4.15-rc1
Diffstat (limited to 'drivers/infiniband/hw/hfi1/file_ops.c')
-rw-r--r--drivers/infiniband/hw/hfi1/file_ops.c922
1 files changed, 529 insertions, 393 deletions
diff --git a/drivers/infiniband/hw/hfi1/file_ops.c b/drivers/infiniband/hw/hfi1/file_ops.c
index 3158128d57e8..7750a9c38b06 100644
--- a/drivers/infiniband/hw/hfi1/file_ops.c
+++ b/drivers/infiniband/hw/hfi1/file_ops.c
@@ -58,10 +58,10 @@
#include "device.h"
#include "common.h"
#include "trace.h"
+#include "mmu_rb.h"
#include "user_sdma.h"
#include "user_exp_rcv.h"
#include "aspm.h"
-#include "mmu_rb.h"
#undef pr_fmt
#define pr_fmt(fmt) DRIVER_NAME ": " fmt
@@ -78,29 +78,38 @@ static unsigned int hfi1_poll(struct file *fp, struct poll_table_struct *pt);
static int hfi1_file_mmap(struct file *fp, struct vm_area_struct *vma);
static u64 kvirt_to_phys(void *addr);
-static int assign_ctxt(struct hfi1_filedata *fd, struct hfi1_user_info *uinfo);
-static int init_subctxts(struct hfi1_ctxtdata *uctxt,
- const struct hfi1_user_info *uinfo);
-static int init_user_ctxt(struct hfi1_filedata *fd);
+static int assign_ctxt(struct hfi1_filedata *fd, unsigned long arg, u32 len);
+static void init_subctxts(struct hfi1_ctxtdata *uctxt,
+ const struct hfi1_user_info *uinfo);
+static int init_user_ctxt(struct hfi1_filedata *fd,
+ struct hfi1_ctxtdata *uctxt);
static void user_init(struct hfi1_ctxtdata *uctxt);
-static int get_ctxt_info(struct hfi1_filedata *fd, void __user *ubase,
- __u32 len);
-static int get_base_info(struct hfi1_filedata *fd, void __user *ubase,
- __u32 len);
-static int setup_base_ctxt(struct hfi1_filedata *fd);
+static int get_ctxt_info(struct hfi1_filedata *fd, unsigned long arg, u32 len);
+static int get_base_info(struct hfi1_filedata *fd, unsigned long arg, u32 len);
+static int user_exp_rcv_setup(struct hfi1_filedata *fd, unsigned long arg,
+ u32 len);
+static int user_exp_rcv_clear(struct hfi1_filedata *fd, unsigned long arg,
+ u32 len);
+static int user_exp_rcv_invalid(struct hfi1_filedata *fd, unsigned long arg,
+ u32 len);
+static int setup_base_ctxt(struct hfi1_filedata *fd,
+ struct hfi1_ctxtdata *uctxt);
static int setup_subctxt(struct hfi1_ctxtdata *uctxt);
static int find_sub_ctxt(struct hfi1_filedata *fd,
const struct hfi1_user_info *uinfo);
static int allocate_ctxt(struct hfi1_filedata *fd, struct hfi1_devdata *dd,
- struct hfi1_user_info *uinfo);
+ struct hfi1_user_info *uinfo,
+ struct hfi1_ctxtdata **cd);
+static void deallocate_ctxt(struct hfi1_ctxtdata *uctxt);
static unsigned int poll_urgent(struct file *fp, struct poll_table_struct *pt);
static unsigned int poll_next(struct file *fp, struct poll_table_struct *pt);
static int user_event_ack(struct hfi1_ctxtdata *uctxt, u16 subctxt,
- unsigned long events);
-static int set_ctxt_pkey(struct hfi1_ctxtdata *uctxt, u16 subctxt, u16 pkey);
+ unsigned long arg);
+static int set_ctxt_pkey(struct hfi1_ctxtdata *uctxt, unsigned long arg);
+static int ctxt_reset(struct hfi1_ctxtdata *uctxt);
static int manage_rcvq(struct hfi1_ctxtdata *uctxt, u16 subctxt,
- int start_stop);
+ unsigned long arg);
static int vma_fault(struct vm_fault *vmf);
static long hfi1_file_ioctl(struct file *fp, unsigned int cmd,
unsigned long arg);
@@ -116,7 +125,7 @@ static const struct file_operations hfi1_file_ops = {
.llseek = noop_llseek,
};
-static struct vm_operations_struct vm_ops = {
+static const struct vm_operations_struct vm_ops = {
.fault = vma_fault,
};
@@ -181,7 +190,7 @@ static int hfi1_file_open(struct inode *inode, struct file *fp)
struct hfi1_devdata,
user_cdev);
- if (!((dd->flags & HFI1_PRESENT) && dd->kregbase))
+ if (!((dd->flags & HFI1_PRESENT) && dd->kregbase1))
return -EINVAL;
if (!atomic_inc_not_zero(&dd->user_refcount))
@@ -217,13 +226,8 @@ static long hfi1_file_ioctl(struct file *fp, unsigned int cmd,
{
struct hfi1_filedata *fd = fp->private_data;
struct hfi1_ctxtdata *uctxt = fd->uctxt;
- struct hfi1_user_info uinfo;
- struct hfi1_tid_info tinfo;
int ret = 0;
- unsigned long addr;
int uval = 0;
- unsigned long ul_uval = 0;
- u16 uval16 = 0;
hfi1_cdbg(IOCTL, "IOCTL recv: 0x%x", cmd);
if (cmd != HFI1_IOCTL_ASSIGN_CTXT &&
@@ -233,170 +237,55 @@ static long hfi1_file_ioctl(struct file *fp, unsigned int cmd,
switch (cmd) {
case HFI1_IOCTL_ASSIGN_CTXT:
- if (uctxt)
- return -EINVAL;
-
- if (copy_from_user(&uinfo,
- (struct hfi1_user_info __user *)arg,
- sizeof(uinfo)))
- return -EFAULT;
-
- ret = assign_ctxt(fd, &uinfo);
+ ret = assign_ctxt(fd, arg, _IOC_SIZE(cmd));
break;
+
case HFI1_IOCTL_CTXT_INFO:
- ret = get_ctxt_info(fd, (void __user *)(unsigned long)arg,
- sizeof(struct hfi1_ctxt_info));
+ ret = get_ctxt_info(fd, arg, _IOC_SIZE(cmd));
break;
+
case HFI1_IOCTL_USER_INFO:
- ret = get_base_info(fd, (void __user *)(unsigned long)arg,
- sizeof(struct hfi1_base_info));
+ ret = get_base_info(fd, arg, _IOC_SIZE(cmd));
break;
+
case HFI1_IOCTL_CREDIT_UPD:
if (uctxt)
sc_return_credits(uctxt->sc);
break;
case HFI1_IOCTL_TID_UPDATE:
- if (copy_from_user(&tinfo,
- (struct hfi11_tid_info __user *)arg,
- sizeof(tinfo)))
- return -EFAULT;
-
- ret = hfi1_user_exp_rcv_setup(fd, &tinfo);
- if (!ret) {
- /*
- * Copy the number of tidlist entries we used
- * and the length of the buffer we registered.
- * These fields are adjacent in the structure so
- * we can copy them at the same time.
- */
- addr = arg + offsetof(struct hfi1_tid_info, tidcnt);
- if (copy_to_user((void __user *)addr, &tinfo.tidcnt,
- sizeof(tinfo.tidcnt) +
- sizeof(tinfo.length)))
- ret = -EFAULT;
- }
+ ret = user_exp_rcv_setup(fd, arg, _IOC_SIZE(cmd));
break;
case HFI1_IOCTL_TID_FREE:
- if (copy_from_user(&tinfo,
- (struct hfi11_tid_info __user *)arg,
- sizeof(tinfo)))
- return -EFAULT;
-
- ret = hfi1_user_exp_rcv_clear(fd, &tinfo);
- if (ret)
- break;
- addr = arg + offsetof(struct hfi1_tid_info, tidcnt);
- if (copy_to_user((void __user *)addr, &tinfo.tidcnt,
- sizeof(tinfo.tidcnt)))
- ret = -EFAULT;
+ ret = user_exp_rcv_clear(fd, arg, _IOC_SIZE(cmd));
break;
case HFI1_IOCTL_TID_INVAL_READ:
- if (copy_from_user(&tinfo,
- (struct hfi11_tid_info __user *)arg,
- sizeof(tinfo)))
- return -EFAULT;
-
- ret = hfi1_user_exp_rcv_invalid(fd, &tinfo);
- if (ret)
- break;
- addr = arg + offsetof(struct hfi1_tid_info, tidcnt);
- if (copy_to_user((void __user *)addr, &tinfo.tidcnt,
- sizeof(tinfo.tidcnt)))
- ret = -EFAULT;
+ ret = user_exp_rcv_invalid(fd, arg, _IOC_SIZE(cmd));
break;
case HFI1_IOCTL_RECV_CTRL:
- ret = get_user(uval, (int __user *)arg);
- if (ret != 0)
- return -EFAULT;
- ret = manage_rcvq(uctxt, fd->subctxt, uval);
+ ret = manage_rcvq(uctxt, fd->subctxt, arg);
break;
case HFI1_IOCTL_POLL_TYPE:
- ret = get_user(uval, (int __user *)arg);
- if (ret != 0)
+ if (get_user(uval, (int __user *)arg))
return -EFAULT;
uctxt->poll_type = (typeof(uctxt->poll_type))uval;
break;
case HFI1_IOCTL_ACK_EVENT:
- ret = get_user(ul_uval, (unsigned long __user *)arg);
- if (ret != 0)
- return -EFAULT;
- ret = user_event_ack(uctxt, fd->subctxt, ul_uval);
+ ret = user_event_ack(uctxt, fd->subctxt, arg);
break;
case HFI1_IOCTL_SET_PKEY:
- ret = get_user(uval16, (u16 __user *)arg);
- if (ret != 0)
- return -EFAULT;
- if (HFI1_CAP_IS_USET(PKEY_CHECK))
- ret = set_ctxt_pkey(uctxt, fd->subctxt, uval16);
- else
- return -EPERM;
+ ret = set_ctxt_pkey(uctxt, arg);
break;
- case HFI1_IOCTL_CTXT_RESET: {
- struct send_context *sc;
- struct hfi1_devdata *dd;
-
- if (!uctxt || !uctxt->dd || !uctxt->sc)
- return -EINVAL;
-
- /*
- * There is no protection here. User level has to
- * guarantee that no one will be writing to the send
- * context while it is being re-initialized.
- * If user level breaks that guarantee, it will break
- * it's own context and no one else's.
- */
- dd = uctxt->dd;
- sc = uctxt->sc;
- /*
- * Wait until the interrupt handler has marked the
- * context as halted or frozen. Report error if we time
- * out.
- */
- wait_event_interruptible_timeout(
- sc->halt_wait, (sc->flags & SCF_HALTED),
- msecs_to_jiffies(SEND_CTXT_HALT_TIMEOUT));
- if (!(sc->flags & SCF_HALTED))
- return -ENOLCK;
-
- /*
- * If the send context was halted due to a Freeze,
- * wait until the device has been "unfrozen" before
- * resetting the context.
- */
- if (sc->flags & SCF_FROZEN) {
- wait_event_interruptible_timeout(
- dd->event_queue,
- !(ACCESS_ONCE(dd->flags) & HFI1_FROZEN),
- msecs_to_jiffies(SEND_CTXT_HALT_TIMEOUT));
- if (dd->flags & HFI1_FROZEN)
- return -ENOLCK;
-
- if (dd->flags & HFI1_FORCED_FREEZE)
- /*
- * Don't allow context reset if we are into
- * forced freeze
- */
- return -ENODEV;
-
- sc_disable(sc);
- ret = sc_enable(sc);
- hfi1_rcvctrl(dd, HFI1_RCVCTRL_CTXT_ENB,
- uctxt->ctxt);
- } else {
- ret = sc_restart(sc);
- }
- if (!ret)
- sc_return_credits(sc);
+ case HFI1_IOCTL_CTXT_RESET:
+ ret = ctxt_reset(uctxt);
break;
- }
case HFI1_IOCTL_GET_VERS:
uval = HFI1_USER_SWVERSION;
@@ -425,8 +314,7 @@ static ssize_t hfi1_write_iter(struct kiocb *kiocb, struct iov_iter *from)
if (!iter_is_iovec(from) || !dim)
return -EINVAL;
- hfi1_cdbg(SDMA, "SDMA request from %u:%u (%lu)",
- fd->uctxt->ctxt, fd->subctxt, dim);
+ trace_hfi1_sdma_request(fd->dd, fd->uctxt->ctxt, fd->subctxt, dim);
if (atomic_read(&pq->n_reqs) == pq->n_max_reqs)
return -ENOSPC;
@@ -591,9 +479,8 @@ static int hfi1_file_mmap(struct file *fp, struct vm_area_struct *vma)
* Use the page where this context's flags are. User level
* knows where it's own bitmap is within the page.
*/
- memaddr = (unsigned long)(dd->events +
- ((uctxt->ctxt - dd->first_dyn_alloc_ctxt) *
- HFI1_MAX_SHARED_CTXTS)) & PAGE_MASK;
+ memaddr = (unsigned long)
+ (dd->events + uctxt_offset(uctxt)) & PAGE_MASK;
memlen = PAGE_SIZE;
/*
* v3.7 removes VM_RESERVED but the effect is kept by
@@ -752,12 +639,11 @@ static int hfi1_file_close(struct inode *inode, struct file *fp)
if (!uctxt)
goto done;
- hfi1_cdbg(PROC, "freeing ctxt %u:%u", uctxt->ctxt, fdata->subctxt);
- mutex_lock(&hfi1_mutex);
+ hfi1_cdbg(PROC, "closing ctxt %u:%u", uctxt->ctxt, fdata->subctxt);
flush_wc();
/* drain user sdma queue */
- hfi1_user_sdma_free_queues(fdata);
+ hfi1_user_sdma_free_queues(fdata, uctxt);
/* release the cpu */
hfi1_put_proc_affinity(fdata->rec_cpu_num);
@@ -766,20 +652,27 @@ static int hfi1_file_close(struct inode *inode, struct file *fp)
hfi1_user_exp_rcv_free(fdata);
/*
+ * fdata->uctxt is used in the above cleanup. It is not ready to be
+ * removed until here.
+ */
+ fdata->uctxt = NULL;
+ hfi1_rcd_put(uctxt);
+
+ /*
* Clear any left over, unhandled events so the next process that
* gets this context doesn't get confused.
*/
- ev = dd->events + ((uctxt->ctxt - dd->first_dyn_alloc_ctxt) *
- HFI1_MAX_SHARED_CTXTS) + fdata->subctxt;
+ ev = dd->events + uctxt_offset(uctxt) + fdata->subctxt;
*ev = 0;
+ spin_lock_irqsave(&dd->uctxt_lock, flags);
__clear_bit(fdata->subctxt, uctxt->in_use_ctxts);
if (!bitmap_empty(uctxt->in_use_ctxts, HFI1_MAX_SHARED_CTXTS)) {
- mutex_unlock(&hfi1_mutex);
+ spin_unlock_irqrestore(&dd->uctxt_lock, flags);
goto done;
}
+ spin_unlock_irqrestore(&dd->uctxt_lock, flags);
- spin_lock_irqsave(&dd->uctxt_lock, flags);
/*
* Disable receive context and interrupt available, reset all
* RcvCtxtCtrl bits to default values.
@@ -790,34 +683,24 @@ static int hfi1_file_close(struct inode *inode, struct file *fp)
HFI1_RCVCTRL_TAILUPD_DIS |
HFI1_RCVCTRL_ONE_PKT_EGR_DIS |
HFI1_RCVCTRL_NO_RHQ_DROP_DIS |
- HFI1_RCVCTRL_NO_EGR_DROP_DIS, uctxt->ctxt);
+ HFI1_RCVCTRL_NO_EGR_DROP_DIS, uctxt);
/* Clear the context's J_KEY */
- hfi1_clear_ctxt_jkey(dd, uctxt->ctxt);
+ hfi1_clear_ctxt_jkey(dd, uctxt);
/*
- * Reset context integrity checks to default.
- * (writes to CSRs probably belong in chip.c)
+ * If a send context is allocated, reset context integrity
+ * checks to default and disable the send context.
*/
- write_kctxt_csr(dd, uctxt->sc->hw_context, SEND_CTXT_CHECK_ENABLE,
- hfi1_pkt_default_send_ctxt_mask(dd, uctxt->sc->type));
- sc_disable(uctxt->sc);
- spin_unlock_irqrestore(&dd->uctxt_lock, flags);
-
- dd->rcd[uctxt->ctxt] = NULL;
+ if (uctxt->sc) {
+ set_pio_integrity(uctxt->sc);
+ sc_disable(uctxt->sc);
+ }
- hfi1_user_exp_rcv_grp_free(uctxt);
+ hfi1_free_ctxt_rcv_groups(uctxt);
hfi1_clear_ctxt_pkey(dd, uctxt);
- uctxt->rcvwait_to = 0;
- uctxt->piowait_to = 0;
- uctxt->rcvnowait = 0;
- uctxt->pionowait = 0;
uctxt->event_flags = 0;
- hfi1_stats.sps_ctxts--;
- if (++dd->freectxts == dd->num_user_contexts)
- aspm_enable_all(dd);
- mutex_unlock(&hfi1_mutex);
- hfi1_free_ctxtdata(dd, uctxt);
+ deallocate_ctxt(uctxt);
done:
mmdrop(fdata->mm);
kobject_put(&dd->kobj);
@@ -845,135 +728,212 @@ static u64 kvirt_to_phys(void *addr)
return paddr;
}
-static int assign_ctxt(struct hfi1_filedata *fd, struct hfi1_user_info *uinfo)
+/**
+ * complete_subctxt
+ * @fd: valid filedata pointer
+ *
+ * Sub-context info can only be set up after the base context
+ * has been completed. This is indicated by the clearing of the
+ * HFI1_CTXT_BASE_UINIT bit.
+ *
+ * Wait for the bit to be cleared, and then complete the subcontext
+ * initialization.
+ *
+ */
+static int complete_subctxt(struct hfi1_filedata *fd)
+{
+ int ret;
+ unsigned long flags;
+
+ /*
+ * sub-context info can only be set up after the base context
+ * has been completed.
+ */
+ ret = wait_event_interruptible(
+ fd->uctxt->wait,
+ !test_bit(HFI1_CTXT_BASE_UNINIT, &fd->uctxt->event_flags));
+
+ if (test_bit(HFI1_CTXT_BASE_FAILED, &fd->uctxt->event_flags))
+ ret = -ENOMEM;
+
+ /* Finish the sub-context init */
+ if (!ret) {
+ fd->rec_cpu_num = hfi1_get_proc_affinity(fd->uctxt->numa_id);
+ ret = init_user_ctxt(fd, fd->uctxt);
+ }
+
+ if (ret) {
+ hfi1_rcd_put(fd->uctxt);
+ fd->uctxt = NULL;
+ spin_lock_irqsave(&fd->dd->uctxt_lock, flags);
+ __clear_bit(fd->subctxt, fd->uctxt->in_use_ctxts);
+ spin_unlock_irqrestore(&fd->dd->uctxt_lock, flags);
+ }
+
+ return ret;
+}
+
+static int assign_ctxt(struct hfi1_filedata *fd, unsigned long arg, u32 len)
{
int ret;
- unsigned int swmajor, swminor;
+ unsigned int swmajor;
+ struct hfi1_ctxtdata *uctxt = NULL;
+ struct hfi1_user_info uinfo;
- swmajor = uinfo->userversion >> 16;
+ if (fd->uctxt)
+ return -EINVAL;
+
+ if (sizeof(uinfo) != len)
+ return -EINVAL;
+
+ if (copy_from_user(&uinfo, (void __user *)arg, sizeof(uinfo)))
+ return -EFAULT;
+
+ swmajor = uinfo.userversion >> 16;
if (swmajor != HFI1_USER_SWMAJOR)
return -ENODEV;
- swminor = uinfo->userversion & 0xffff;
+ if (uinfo.subctxt_cnt > HFI1_MAX_SHARED_CTXTS)
+ return -EINVAL;
+ /*
+ * Acquire the mutex to protect against multiple creations of what
+ * could be a shared base context.
+ */
mutex_lock(&hfi1_mutex);
/*
- * Get a sub context if necessary.
+ * Get a sub context if available (fd->uctxt will be set).
* ret < 0 error, 0 no context, 1 sub-context found
*/
- ret = 0;
- if (uinfo->subctxt_cnt) {
- ret = find_sub_ctxt(fd, uinfo);
- if (ret > 0)
- fd->rec_cpu_num =
- hfi1_get_proc_affinity(fd->uctxt->numa_id);
- }
+ ret = find_sub_ctxt(fd, &uinfo);
/*
- * Allocate a base context if context sharing is not required or we
- * couldn't find a sub context.
+ * Allocate a base context if context sharing is not required or a
+ * sub context wasn't found.
*/
if (!ret)
- ret = allocate_ctxt(fd, fd->dd, uinfo);
+ ret = allocate_ctxt(fd, fd->dd, &uinfo, &uctxt);
mutex_unlock(&hfi1_mutex);
- /* Depending on the context type, do the appropriate init */
- if (ret > 0) {
- /*
- * sub-context info can only be set up after the base
- * context has been completed.
- */
- ret = wait_event_interruptible(fd->uctxt->wait, !test_bit(
- HFI1_CTXT_BASE_UNINIT,
- &fd->uctxt->event_flags));
- if (test_bit(HFI1_CTXT_BASE_FAILED, &fd->uctxt->event_flags)) {
- clear_bit(fd->subctxt, fd->uctxt->in_use_ctxts);
- return -ENOMEM;
- }
- /* The only thing a sub context needs is the user_xxx stuff */
- if (!ret)
- ret = init_user_ctxt(fd);
-
+ /* Depending on the context type, finish the appropriate init */
+ switch (ret) {
+ case 0:
+ ret = setup_base_ctxt(fd, uctxt);
if (ret)
- clear_bit(fd->subctxt, fd->uctxt->in_use_ctxts);
- } else if (!ret) {
- ret = setup_base_ctxt(fd);
- if (fd->uctxt->subctxt_cnt) {
- /* If there is an error, set the failed bit. */
- if (ret)
- set_bit(HFI1_CTXT_BASE_FAILED,
- &fd->uctxt->event_flags);
- /*
- * Base context is done, notify anybody using a
- * sub-context that is waiting for this completion
- */
- clear_bit(HFI1_CTXT_BASE_UNINIT,
- &fd->uctxt->event_flags);
- wake_up(&fd->uctxt->wait);
- }
+ deallocate_ctxt(uctxt);
+ break;
+ case 1:
+ ret = complete_subctxt(fd);
+ break;
+ default:
+ break;
}
return ret;
}
-/*
- * The hfi1_mutex must be held when this function is called. It is
- * necessary to ensure serialized access to the bitmask in_use_ctxts.
+/**
+ * match_ctxt
+ * @fd: valid filedata pointer
+ * @uinfo: user info to compare base context with
+ * @uctxt: context to compare uinfo to.
+ *
+ * Compare the given context with the given information to see if it
+ * can be used for a sub context.
*/
-static int find_sub_ctxt(struct hfi1_filedata *fd,
- const struct hfi1_user_info *uinfo)
+static int match_ctxt(struct hfi1_filedata *fd,
+ const struct hfi1_user_info *uinfo,
+ struct hfi1_ctxtdata *uctxt)
{
- int i;
struct hfi1_devdata *dd = fd->dd;
+ unsigned long flags;
u16 subctxt;
- for (i = dd->first_dyn_alloc_ctxt; i < dd->num_rcv_contexts; i++) {
- struct hfi1_ctxtdata *uctxt = dd->rcd[i];
+ /* Skip dynamically allocated kernel contexts */
+ if (uctxt->sc && (uctxt->sc->type == SC_KERNEL))
+ return 0;
- /* Skip ctxts which are not yet open */
- if (!uctxt ||
- bitmap_empty(uctxt->in_use_ctxts,
- HFI1_MAX_SHARED_CTXTS))
- continue;
+ /* Skip ctxt if it doesn't match the requested one */
+ if (memcmp(uctxt->uuid, uinfo->uuid, sizeof(uctxt->uuid)) ||
+ uctxt->jkey != generate_jkey(current_uid()) ||
+ uctxt->subctxt_id != uinfo->subctxt_id ||
+ uctxt->subctxt_cnt != uinfo->subctxt_cnt)
+ return 0;
- /* Skip dynamically allocted kernel contexts */
- if (uctxt->sc && (uctxt->sc->type == SC_KERNEL))
- continue;
+ /* Verify the sharing process matches the base */
+ if (uctxt->userversion != uinfo->userversion)
+ return -EINVAL;
- /* Skip ctxt if it doesn't match the requested one */
- if (memcmp(uctxt->uuid, uinfo->uuid,
- sizeof(uctxt->uuid)) ||
- uctxt->jkey != generate_jkey(current_uid()) ||
- uctxt->subctxt_id != uinfo->subctxt_id ||
- uctxt->subctxt_cnt != uinfo->subctxt_cnt)
- continue;
+ /* Find an unused sub context */
+ spin_lock_irqsave(&dd->uctxt_lock, flags);
+ if (bitmap_empty(uctxt->in_use_ctxts, HFI1_MAX_SHARED_CTXTS)) {
+ /* context is being closed, do not use */
+ spin_unlock_irqrestore(&dd->uctxt_lock, flags);
+ return 0;
+ }
- /* Verify the sharing process matches the master */
- if (uctxt->userversion != uinfo->userversion)
- return -EINVAL;
+ subctxt = find_first_zero_bit(uctxt->in_use_ctxts,
+ HFI1_MAX_SHARED_CTXTS);
+ if (subctxt >= uctxt->subctxt_cnt) {
+ spin_unlock_irqrestore(&dd->uctxt_lock, flags);
+ return -EBUSY;
+ }
+
+ fd->subctxt = subctxt;
+ __set_bit(fd->subctxt, uctxt->in_use_ctxts);
+ spin_unlock_irqrestore(&dd->uctxt_lock, flags);
- /* Find an unused context */
- subctxt = find_first_zero_bit(uctxt->in_use_ctxts,
- HFI1_MAX_SHARED_CTXTS);
- if (subctxt >= uctxt->subctxt_cnt)
- return -EBUSY;
+ fd->uctxt = uctxt;
+ hfi1_rcd_get(uctxt);
- fd->uctxt = uctxt;
- fd->subctxt = subctxt;
- __set_bit(fd->subctxt, uctxt->in_use_ctxts);
+ return 1;
+}
- return 1;
+/**
+ * find_sub_ctxt
+ * @fd: valid filedata pointer
+ * @uinfo: matching info to use to find a possible context to share.
+ *
+ * The hfi1_mutex must be held when this function is called. It is
+ * necessary to ensure serialized creation of shared contexts.
+ *
+ * Return:
+ * 0 No sub-context found
+ * 1 Subcontext found and allocated
+ * errno EINVAL (incorrect parameters)
+ * EBUSY (all sub contexts in use)
+ */
+static int find_sub_ctxt(struct hfi1_filedata *fd,
+ const struct hfi1_user_info *uinfo)
+{
+ struct hfi1_ctxtdata *uctxt;
+ struct hfi1_devdata *dd = fd->dd;
+ u16 i;
+ int ret;
+
+ if (!uinfo->subctxt_cnt)
+ return 0;
+
+ for (i = dd->first_dyn_alloc_ctxt; i < dd->num_rcv_contexts; i++) {
+ uctxt = hfi1_rcd_get_by_index(dd, i);
+ if (uctxt) {
+ ret = match_ctxt(fd, uinfo, uctxt);
+ hfi1_rcd_put(uctxt);
+ /* value of != 0 will return */
+ if (ret)
+ return ret;
+ }
}
return 0;
}
static int allocate_ctxt(struct hfi1_filedata *fd, struct hfi1_devdata *dd,
- struct hfi1_user_info *uinfo)
+ struct hfi1_user_info *uinfo,
+ struct hfi1_ctxtdata **rcd)
{
struct hfi1_ctxtdata *uctxt;
- unsigned int ctxt;
int ret, numa;
if (dd->flags & HFI1_FROZEN) {
@@ -987,22 +947,9 @@ static int allocate_ctxt(struct hfi1_filedata *fd, struct hfi1_devdata *dd,
return -EIO;
}
- /*
- * This check is sort of redundant to the next EBUSY error. It would
- * also indicate an inconsistancy in the driver if this value was
- * zero, but there were still contexts available.
- */
if (!dd->freectxts)
return -EBUSY;
- for (ctxt = dd->first_dyn_alloc_ctxt;
- ctxt < dd->num_rcv_contexts; ctxt++)
- if (!dd->rcd[ctxt])
- break;
-
- if (ctxt == dd->num_rcv_contexts)
- return -EBUSY;
-
/*
* If we don't have a NUMA node requested, preference is towards
* device NUMA node.
@@ -1012,11 +959,10 @@ static int allocate_ctxt(struct hfi1_filedata *fd, struct hfi1_devdata *dd,
numa = cpu_to_node(fd->rec_cpu_num);
else
numa = numa_node_id();
- uctxt = hfi1_create_ctxtdata(dd->pport, ctxt, numa);
- if (!uctxt) {
- dd_dev_err(dd,
- "Unable to allocate ctxtdata memory, failing open\n");
- return -ENOMEM;
+ ret = hfi1_create_ctxtdata(dd->pport, numa, &uctxt);
+ if (ret < 0) {
+ dd_dev_err(dd, "user ctxtdata allocation failed\n");
+ return ret;
}
hfi1_cdbg(PROC, "[%u:%u] pid %u assigned to CPU %d (NUMA %u)",
uctxt->ctxt, fd->subctxt, current->pid, fd->rec_cpu_num,
@@ -1025,8 +971,7 @@ static int allocate_ctxt(struct hfi1_filedata *fd, struct hfi1_devdata *dd,
/*
* Allocate and enable a PIO send context.
*/
- uctxt->sc = sc_alloc(dd, SC_USER, uctxt->rcvhdrqentsize,
- uctxt->dd->node);
+ uctxt->sc = sc_alloc(dd, SC_USER, uctxt->rcvhdrqentsize, dd->node);
if (!uctxt->sc) {
ret = -ENOMEM;
goto ctxdata_free;
@@ -1038,28 +983,19 @@ static int allocate_ctxt(struct hfi1_filedata *fd, struct hfi1_devdata *dd,
goto ctxdata_free;
/*
- * Setup sub context resources if the user-level has requested
+ * Setup sub context information if the user-level has requested
* sub contexts.
* This has to be done here so the rest of the sub-contexts find the
- * proper master.
+ * proper base context.
*/
- if (uinfo->subctxt_cnt) {
- ret = init_subctxts(uctxt, uinfo);
- /*
- * On error, we don't need to disable and de-allocate the
- * send context because it will be done during file close
- */
- if (ret)
- goto ctxdata_free;
- }
+ if (uinfo->subctxt_cnt)
+ init_subctxts(uctxt, uinfo);
uctxt->userversion = uinfo->userversion;
uctxt->flags = hfi1_cap_mask; /* save current flag state */
init_waitqueue_head(&uctxt->wait);
strlcpy(uctxt->comm, current->comm, sizeof(uctxt->comm));
memcpy(uctxt->uuid, uinfo->uuid, sizeof(uctxt->uuid));
uctxt->jkey = generate_jkey(current_uid());
- INIT_LIST_HEAD(&uctxt->sdma_queues);
- spin_lock_init(&uctxt->sdma_qlock);
hfi1_stats.sps_ctxts++;
/*
* Disable ASPM when there are open user/PSM contexts to avoid
@@ -1067,31 +1003,33 @@ static int allocate_ctxt(struct hfi1_filedata *fd, struct hfi1_devdata *dd,
*/
if (dd->freectxts-- == dd->num_user_contexts)
aspm_disable_all(dd);
- fd->uctxt = uctxt;
+
+ *rcd = uctxt;
return 0;
ctxdata_free:
- dd->rcd[ctxt] = NULL;
- hfi1_free_ctxtdata(dd, uctxt);
+ hfi1_free_ctxt(uctxt);
return ret;
}
-static int init_subctxts(struct hfi1_ctxtdata *uctxt,
- const struct hfi1_user_info *uinfo)
+static void deallocate_ctxt(struct hfi1_ctxtdata *uctxt)
{
- u16 num_subctxts;
+ mutex_lock(&hfi1_mutex);
+ hfi1_stats.sps_ctxts--;
+ if (++uctxt->dd->freectxts == uctxt->dd->num_user_contexts)
+ aspm_enable_all(uctxt->dd);
+ mutex_unlock(&hfi1_mutex);
- num_subctxts = uinfo->subctxt_cnt;
- if (num_subctxts > HFI1_MAX_SHARED_CTXTS)
- return -EINVAL;
+ hfi1_free_ctxt(uctxt);
+}
+static void init_subctxts(struct hfi1_ctxtdata *uctxt,
+ const struct hfi1_user_info *uinfo)
+{
uctxt->subctxt_cnt = uinfo->subctxt_cnt;
uctxt->subctxt_id = uinfo->subctxt_id;
- uctxt->redirect_seq_cnt = 1;
set_bit(HFI1_CTXT_BASE_UNINIT, &uctxt->event_flags);
-
- return 0;
}
static int setup_subctxt(struct hfi1_ctxtdata *uctxt)
@@ -1153,7 +1091,7 @@ static void user_init(struct hfi1_ctxtdata *uctxt)
clear_rcvhdrtail(uctxt);
/* Setup J_KEY before enabling the context */
- hfi1_set_ctxt_jkey(uctxt->dd, uctxt->ctxt, uctxt->jkey);
+ hfi1_set_ctxt_jkey(uctxt->dd, uctxt, uctxt->jkey);
rcvctrl_ops = HFI1_RCVCTRL_CTXT_ENB;
if (HFI1_CAP_UGET_MASK(uctxt->flags, HDRSUPP))
@@ -1179,15 +1117,16 @@ static void user_init(struct hfi1_ctxtdata *uctxt)
rcvctrl_ops |= HFI1_RCVCTRL_TAILUPD_ENB;
else
rcvctrl_ops |= HFI1_RCVCTRL_TAILUPD_DIS;
- hfi1_rcvctrl(uctxt->dd, rcvctrl_ops, uctxt->ctxt);
+ hfi1_rcvctrl(uctxt->dd, rcvctrl_ops, uctxt);
}
-static int get_ctxt_info(struct hfi1_filedata *fd, void __user *ubase,
- __u32 len)
+static int get_ctxt_info(struct hfi1_filedata *fd, unsigned long arg, u32 len)
{
struct hfi1_ctxt_info cinfo;
struct hfi1_ctxtdata *uctxt = fd->uctxt;
- int ret = 0;
+
+ if (sizeof(cinfo) != len)
+ return -EINVAL;
memset(&cinfo, 0, sizeof(cinfo));
cinfo.runtime_flags = (((uctxt->flags >> HFI1_CAP_MISC_SHIFT) &
@@ -1217,29 +1156,31 @@ static int get_ctxt_info(struct hfi1_filedata *fd, void __user *ubase,
cinfo.rcvegr_size = uctxt->egrbufs.rcvtid_size;
trace_hfi1_ctxt_info(uctxt->dd, uctxt->ctxt, fd->subctxt, cinfo);
- if (copy_to_user(ubase, &cinfo, sizeof(cinfo)))
- ret = -EFAULT;
+ if (copy_to_user((void __user *)arg, &cinfo, len))
+ return -EFAULT;
- return ret;
+ return 0;
}
-static int init_user_ctxt(struct hfi1_filedata *fd)
+static int init_user_ctxt(struct hfi1_filedata *fd,
+ struct hfi1_ctxtdata *uctxt)
{
- struct hfi1_ctxtdata *uctxt = fd->uctxt;
int ret;
ret = hfi1_user_sdma_alloc_queues(uctxt, fd);
if (ret)
return ret;
- ret = hfi1_user_exp_rcv_init(fd);
+ ret = hfi1_user_exp_rcv_init(fd, uctxt);
+ if (ret)
+ hfi1_user_sdma_free_queues(fd, uctxt);
return ret;
}
-static int setup_base_ctxt(struct hfi1_filedata *fd)
+static int setup_base_ctxt(struct hfi1_filedata *fd,
+ struct hfi1_ctxtdata *uctxt)
{
- struct hfi1_ctxtdata *uctxt = fd->uctxt;
struct hfi1_devdata *dd = uctxt->dd;
int ret = 0;
@@ -1248,47 +1189,64 @@ static int setup_base_ctxt(struct hfi1_filedata *fd)
/* Now allocate the RcvHdr queue and eager buffers. */
ret = hfi1_create_rcvhdrq(dd, uctxt);
if (ret)
- return ret;
+ goto done;
ret = hfi1_setup_eagerbufs(uctxt);
if (ret)
- goto setup_failed;
+ goto done;
/* If sub-contexts are enabled, do the appropriate setup */
if (uctxt->subctxt_cnt)
ret = setup_subctxt(uctxt);
if (ret)
- goto setup_failed;
+ goto done;
- ret = hfi1_user_exp_rcv_grp_init(fd);
+ ret = hfi1_alloc_ctxt_rcv_groups(uctxt);
if (ret)
- goto setup_failed;
+ goto done;
- ret = init_user_ctxt(fd);
+ ret = init_user_ctxt(fd, uctxt);
if (ret)
- goto setup_failed;
+ goto done;
user_init(uctxt);
- return 0;
+ /* Now that the context is set up, the fd can get a reference. */
+ fd->uctxt = uctxt;
+ hfi1_rcd_get(uctxt);
+
+done:
+ if (uctxt->subctxt_cnt) {
+ /*
+ * On error, set the failed bit so sub-contexts will clean up
+ * correctly.
+ */
+ if (ret)
+ set_bit(HFI1_CTXT_BASE_FAILED, &uctxt->event_flags);
+
+ /*
+ * Base context is done (successfully or not), notify anybody
+ * using a sub-context that is waiting for this completion.
+ */
+ clear_bit(HFI1_CTXT_BASE_UNINIT, &uctxt->event_flags);
+ wake_up(&uctxt->wait);
+ }
-setup_failed:
- hfi1_free_ctxtdata(dd, uctxt);
return ret;
}
-static int get_base_info(struct hfi1_filedata *fd, void __user *ubase,
- __u32 len)
+static int get_base_info(struct hfi1_filedata *fd, unsigned long arg, u32 len)
{
struct hfi1_base_info binfo;
struct hfi1_ctxtdata *uctxt = fd->uctxt;
struct hfi1_devdata *dd = uctxt->dd;
- ssize_t sz;
unsigned offset;
- int ret = 0;
trace_hfi1_uctxtdata(uctxt->dd, uctxt, fd->subctxt);
+ if (sizeof(binfo) != len)
+ return -EINVAL;
+
memset(&binfo, 0, sizeof(binfo));
binfo.hw_version = dd->revision;
binfo.sw_version = HFI1_KERN_SWVERSION;
@@ -1318,39 +1276,152 @@ static int get_base_info(struct hfi1_filedata *fd, void __user *ubase,
fd->subctxt,
uctxt->egrbufs.rcvtids[0].dma);
binfo.sdma_comp_bufbase = HFI1_MMAP_TOKEN(SDMA_COMP, uctxt->ctxt,
- fd->subctxt, 0);
+ fd->subctxt, 0);
/*
* user regs are at
* (RXE_PER_CONTEXT_USER + (ctxt * RXE_PER_CONTEXT_SIZE))
*/
binfo.user_regbase = HFI1_MMAP_TOKEN(UREGS, uctxt->ctxt,
- fd->subctxt, 0);
- offset = offset_in_page((((uctxt->ctxt - dd->first_dyn_alloc_ctxt) *
- HFI1_MAX_SHARED_CTXTS) + fd->subctxt) *
- sizeof(*dd->events));
+ fd->subctxt, 0);
+ offset = offset_in_page((uctxt_offset(uctxt) + fd->subctxt) *
+ sizeof(*dd->events));
binfo.events_bufbase = HFI1_MMAP_TOKEN(EVENTS, uctxt->ctxt,
- fd->subctxt,
- offset);
+ fd->subctxt,
+ offset);
binfo.status_bufbase = HFI1_MMAP_TOKEN(STATUS, uctxt->ctxt,
- fd->subctxt,
- dd->status);
+ fd->subctxt,
+ dd->status);
if (HFI1_CAP_IS_USET(DMA_RTAIL))
binfo.rcvhdrtail_base = HFI1_MMAP_TOKEN(RTAIL, uctxt->ctxt,
- fd->subctxt, 0);
+ fd->subctxt, 0);
if (uctxt->subctxt_cnt) {
binfo.subctxt_uregbase = HFI1_MMAP_TOKEN(SUBCTXT_UREGS,
- uctxt->ctxt,
- fd->subctxt, 0);
- binfo.subctxt_rcvhdrbuf = HFI1_MMAP_TOKEN(SUBCTXT_RCV_HDRQ,
uctxt->ctxt,
fd->subctxt, 0);
+ binfo.subctxt_rcvhdrbuf = HFI1_MMAP_TOKEN(SUBCTXT_RCV_HDRQ,
+ uctxt->ctxt,
+ fd->subctxt, 0);
binfo.subctxt_rcvegrbuf = HFI1_MMAP_TOKEN(SUBCTXT_EGRBUF,
- uctxt->ctxt,
- fd->subctxt, 0);
+ uctxt->ctxt,
+ fd->subctxt, 0);
}
- sz = (len < sizeof(binfo)) ? len : sizeof(binfo);
- if (copy_to_user(ubase, &binfo, sz))
+
+ if (copy_to_user((void __user *)arg, &binfo, len))
+ return -EFAULT;
+
+ return 0;
+}
+
+/**
+ * user_exp_rcv_setup - Set up the given tid rcv list
+ * @fd: file data of the current driver instance
+ * @arg: ioctl argumnent for user space information
+ * @len: length of data structure associated with ioctl command
+ *
+ * Wrapper to validate ioctl information before doing _rcv_setup.
+ *
+ */
+static int user_exp_rcv_setup(struct hfi1_filedata *fd, unsigned long arg,
+ u32 len)
+{
+ int ret;
+ unsigned long addr;
+ struct hfi1_tid_info tinfo;
+
+ if (sizeof(tinfo) != len)
+ return -EINVAL;
+
+ if (copy_from_user(&tinfo, (void __user *)arg, (sizeof(tinfo))))
+ return -EFAULT;
+
+ ret = hfi1_user_exp_rcv_setup(fd, &tinfo);
+ if (!ret) {
+ /*
+ * Copy the number of tidlist entries we used
+ * and the length of the buffer we registered.
+ */
+ addr = arg + offsetof(struct hfi1_tid_info, tidcnt);
+ if (copy_to_user((void __user *)addr, &tinfo.tidcnt,
+ sizeof(tinfo.tidcnt)))
+ return -EFAULT;
+
+ addr = arg + offsetof(struct hfi1_tid_info, length);
+ if (copy_to_user((void __user *)addr, &tinfo.length,
+ sizeof(tinfo.length)))
+ ret = -EFAULT;
+ }
+
+ return ret;
+}
+
+/**
+ * user_exp_rcv_clear - Clear the given tid rcv list
+ * @fd: file data of the current driver instance
+ * @arg: ioctl argumnent for user space information
+ * @len: length of data structure associated with ioctl command
+ *
+ * The hfi1_user_exp_rcv_clear() can be called from the error path. Because
+ * of this, we need to use this wrapper to copy the user space information
+ * before doing the clear.
+ */
+static int user_exp_rcv_clear(struct hfi1_filedata *fd, unsigned long arg,
+ u32 len)
+{
+ int ret;
+ unsigned long addr;
+ struct hfi1_tid_info tinfo;
+
+ if (sizeof(tinfo) != len)
+ return -EINVAL;
+
+ if (copy_from_user(&tinfo, (void __user *)arg, (sizeof(tinfo))))
+ return -EFAULT;
+
+ ret = hfi1_user_exp_rcv_clear(fd, &tinfo);
+ if (!ret) {
+ addr = arg + offsetof(struct hfi1_tid_info, tidcnt);
+ if (copy_to_user((void __user *)addr, &tinfo.tidcnt,
+ sizeof(tinfo.tidcnt)))
+ return -EFAULT;
+ }
+
+ return ret;
+}
+
+/**
+ * user_exp_rcv_invalid - Invalidate the given tid rcv list
+ * @fd: file data of the current driver instance
+ * @arg: ioctl argumnent for user space information
+ * @len: length of data structure associated with ioctl command
+ *
+ * Wrapper to validate ioctl information before doing _rcv_invalid.
+ *
+ */
+static int user_exp_rcv_invalid(struct hfi1_filedata *fd, unsigned long arg,
+ u32 len)
+{
+ int ret;
+ unsigned long addr;
+ struct hfi1_tid_info tinfo;
+
+ if (sizeof(tinfo) != len)
+ return -EINVAL;
+
+ if (!fd->invalid_tids)
+ return -EINVAL;
+
+ if (copy_from_user(&tinfo, (void __user *)arg, (sizeof(tinfo))))
+ return -EFAULT;
+
+ ret = hfi1_user_exp_rcv_invalid(fd, &tinfo);
+ if (ret)
+ return ret;
+
+ addr = arg + offsetof(struct hfi1_tid_info, tidcnt);
+ if (copy_to_user((void __user *)addr, &tinfo.tidcnt,
+ sizeof(tinfo.tidcnt)))
ret = -EFAULT;
+
return ret;
}
@@ -1390,7 +1461,7 @@ static unsigned int poll_next(struct file *fp,
spin_lock_irq(&dd->uctxt_lock);
if (hdrqempty(uctxt)) {
set_bit(HFI1_CTXT_WAITING_RCV, &uctxt->event_flags);
- hfi1_rcvctrl(dd, HFI1_RCVCTRL_INTRAVAIL_ENB, uctxt->ctxt);
+ hfi1_rcvctrl(dd, HFI1_RCVCTRL_INTRAVAIL_ENB, uctxt);
pollflag = 0;
} else {
pollflag = POLLIN | POLLRDNORM;
@@ -1409,36 +1480,30 @@ int hfi1_set_uevent_bits(struct hfi1_pportdata *ppd, const int evtbit)
{
struct hfi1_ctxtdata *uctxt;
struct hfi1_devdata *dd = ppd->dd;
- unsigned ctxt;
- int ret = 0;
- unsigned long flags;
+ u16 ctxt;
- if (!dd->events) {
- ret = -EINVAL;
- goto done;
- }
+ if (!dd->events)
+ return -EINVAL;
- spin_lock_irqsave(&dd->uctxt_lock, flags);
for (ctxt = dd->first_dyn_alloc_ctxt; ctxt < dd->num_rcv_contexts;
ctxt++) {
- uctxt = dd->rcd[ctxt];
+ uctxt = hfi1_rcd_get_by_index(dd, ctxt);
if (uctxt) {
- unsigned long *evs = dd->events +
- (uctxt->ctxt - dd->first_dyn_alloc_ctxt) *
- HFI1_MAX_SHARED_CTXTS;
+ unsigned long *evs;
int i;
/*
* subctxt_cnt is 0 if not shared, so do base
* separately, first, then remaining subctxt, if any
*/
+ evs = dd->events + uctxt_offset(uctxt);
set_bit(evtbit, evs);
for (i = 1; i < uctxt->subctxt_cnt; i++)
set_bit(evtbit, evs + i);
+ hfi1_rcd_put(uctxt);
}
}
- spin_unlock_irqrestore(&dd->uctxt_lock, flags);
-done:
- return ret;
+
+ return 0;
}
/**
@@ -1452,13 +1517,18 @@ done:
* re-init the software copy of the head register
*/
static int manage_rcvq(struct hfi1_ctxtdata *uctxt, u16 subctxt,
- int start_stop)
+ unsigned long arg)
{
struct hfi1_devdata *dd = uctxt->dd;
unsigned int rcvctrl_op;
+ int start_stop;
if (subctxt)
- goto bail;
+ return 0;
+
+ if (get_user(start_stop, (int __user *)arg))
+ return -EFAULT;
+
/* atomically clear receive enable ctxt. */
if (start_stop) {
/*
@@ -1475,9 +1545,9 @@ static int manage_rcvq(struct hfi1_ctxtdata *uctxt, u16 subctxt,
} else {
rcvctrl_op = HFI1_RCVCTRL_CTXT_DIS;
}
- hfi1_rcvctrl(dd, rcvctrl_op, uctxt->ctxt);
+ hfi1_rcvctrl(dd, rcvctrl_op, uctxt);
/* always; new head should be equal to new tail; see above */
-bail:
+
return 0;
}
@@ -1487,17 +1557,20 @@ bail:
* set, if desired, and checks again in future.
*/
static int user_event_ack(struct hfi1_ctxtdata *uctxt, u16 subctxt,
- unsigned long events)
+ unsigned long arg)
{
int i;
struct hfi1_devdata *dd = uctxt->dd;
unsigned long *evs;
+ unsigned long events;
if (!dd->events)
return 0;
- evs = dd->events + ((uctxt->ctxt - dd->first_dyn_alloc_ctxt) *
- HFI1_MAX_SHARED_CTXTS) + subctxt;
+ if (get_user(events, (unsigned long __user *)arg))
+ return -EFAULT;
+
+ evs = dd->events + uctxt_offset(uctxt) + subctxt;
for (i = 0; i <= _HFI1_MAX_EVENT_BIT; i++) {
if (!test_bit(i, &events))
@@ -1507,26 +1580,89 @@ static int user_event_ack(struct hfi1_ctxtdata *uctxt, u16 subctxt,
return 0;
}
-static int set_ctxt_pkey(struct hfi1_ctxtdata *uctxt, u16 subctxt, u16 pkey)
+static int set_ctxt_pkey(struct hfi1_ctxtdata *uctxt, unsigned long arg)
{
- int ret = -ENOENT, i, intable = 0;
+ int i;
struct hfi1_pportdata *ppd = uctxt->ppd;
struct hfi1_devdata *dd = uctxt->dd;
+ u16 pkey;
- if (pkey == LIM_MGMT_P_KEY || pkey == FULL_MGMT_P_KEY) {
- ret = -EINVAL;
- goto done;
- }
+ if (!HFI1_CAP_IS_USET(PKEY_CHECK))
+ return -EPERM;
+
+ if (get_user(pkey, (u16 __user *)arg))
+ return -EFAULT;
+
+ if (pkey == LIM_MGMT_P_KEY || pkey == FULL_MGMT_P_KEY)
+ return -EINVAL;
for (i = 0; i < ARRAY_SIZE(ppd->pkeys); i++)
- if (pkey == ppd->pkeys[i]) {
- intable = 1;
- break;
- }
+ if (pkey == ppd->pkeys[i])
+ return hfi1_set_ctxt_pkey(dd, uctxt, pkey);
+
+ return -ENOENT;
+}
+
+/**
+ * ctxt_reset - Reset the user context
+ * @uctxt: valid user context
+ */
+static int ctxt_reset(struct hfi1_ctxtdata *uctxt)
+{
+ struct send_context *sc;
+ struct hfi1_devdata *dd;
+ int ret = 0;
+
+ if (!uctxt || !uctxt->dd || !uctxt->sc)
+ return -EINVAL;
+
+ /*
+ * There is no protection here. User level has to guarantee that
+ * no one will be writing to the send context while it is being
+ * re-initialized. If user level breaks that guarantee, it will
+ * break it's own context and no one else's.
+ */
+ dd = uctxt->dd;
+ sc = uctxt->sc;
+
+ /*
+ * Wait until the interrupt handler has marked the context as
+ * halted or frozen. Report error if we time out.
+ */
+ wait_event_interruptible_timeout(
+ sc->halt_wait, (sc->flags & SCF_HALTED),
+ msecs_to_jiffies(SEND_CTXT_HALT_TIMEOUT));
+ if (!(sc->flags & SCF_HALTED))
+ return -ENOLCK;
+
+ /*
+ * If the send context was halted due to a Freeze, wait until the
+ * device has been "unfrozen" before resetting the context.
+ */
+ if (sc->flags & SCF_FROZEN) {
+ wait_event_interruptible_timeout(
+ dd->event_queue,
+ !(READ_ONCE(dd->flags) & HFI1_FROZEN),
+ msecs_to_jiffies(SEND_CTXT_HALT_TIMEOUT));
+ if (dd->flags & HFI1_FROZEN)
+ return -ENOLCK;
+
+ if (dd->flags & HFI1_FORCED_FREEZE)
+ /*
+ * Don't allow context reset if we are into
+ * forced freeze
+ */
+ return -ENODEV;
+
+ sc_disable(sc);
+ ret = sc_enable(sc);
+ hfi1_rcvctrl(dd, HFI1_RCVCTRL_CTXT_ENB, uctxt);
+ } else {
+ ret = sc_restart(sc);
+ }
+ if (!ret)
+ sc_return_credits(sc);
- if (intable)
- ret = hfi1_set_ctxt_pkey(dd, uctxt->ctxt, pkey);
-done:
return ret;
}