2016-02-06 00:57:46 +08:00
|
|
|
/*
|
2018-06-05 02:43:54 +08:00
|
|
|
* Copyright(c) 2015-2018 Intel Corporation.
|
2016-02-06 00:57:46 +08:00
|
|
|
*
|
|
|
|
* This file is provided under a dual BSD/GPLv2 license. When using or
|
|
|
|
* redistributing this file, you may do so under either license.
|
|
|
|
*
|
|
|
|
* GPL LICENSE SUMMARY
|
|
|
|
*
|
|
|
|
* This program is free software; you can redistribute it and/or modify
|
|
|
|
* it under the terms of version 2 of the GNU General Public License as
|
|
|
|
* published by the Free Software Foundation.
|
|
|
|
*
|
|
|
|
* This program is distributed in the hope that it will be useful, but
|
|
|
|
* WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
|
|
|
|
* General Public License for more details.
|
|
|
|
*
|
|
|
|
* BSD LICENSE
|
|
|
|
*
|
|
|
|
* 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.
|
|
|
|
* - Neither the name of Intel Corporation nor the names of its
|
|
|
|
* contributors may be used to endorse or promote products derived
|
|
|
|
* from this software without specific prior written permission.
|
|
|
|
*
|
|
|
|
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
|
|
|
|
* "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
|
|
|
|
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
|
|
|
|
* A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
|
|
|
|
* OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
|
|
|
|
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
|
|
|
|
* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
|
|
|
|
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
|
|
|
|
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
|
|
|
|
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
|
|
|
|
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
|
|
|
*
|
|
|
|
*/
|
|
|
|
#include <asm/page.h>
|
2017-02-08 21:28:31 +08:00
|
|
|
#include <linux/string.h>
|
2016-02-06 00:57:46 +08:00
|
|
|
|
2017-08-22 09:27:23 +08:00
|
|
|
#include "mmu_rb.h"
|
2016-02-06 00:57:46 +08:00
|
|
|
#include "user_exp_rcv.h"
|
|
|
|
#include "trace.h"
|
2016-02-06 00:57:52 +08:00
|
|
|
|
2017-05-04 20:14:39 +08:00
|
|
|
static void unlock_exp_tids(struct hfi1_ctxtdata *uctxt,
|
|
|
|
struct exp_tid_set *set,
|
|
|
|
struct hfi1_filedata *fd);
|
2017-08-22 09:26:51 +08:00
|
|
|
static u32 find_phys_blocks(struct tid_user_buf *tidbuf, unsigned int npages);
|
|
|
|
static int set_rcvarray_entry(struct hfi1_filedata *fd,
|
|
|
|
struct tid_user_buf *tbuf,
|
2017-05-04 20:14:39 +08:00
|
|
|
u32 rcventry, struct tid_group *grp,
|
2017-08-22 09:26:51 +08:00
|
|
|
u16 pageidx, unsigned int npages);
|
2017-05-04 20:14:39 +08:00
|
|
|
static int tid_rb_insert(void *arg, struct mmu_rb_node *node);
|
2016-07-29 03:21:26 +08:00
|
|
|
static void cacheless_tid_rb_remove(struct hfi1_filedata *fdata,
|
|
|
|
struct tid_rb_node *tnode);
|
2017-05-04 20:14:39 +08:00
|
|
|
static void tid_rb_remove(void *arg, struct mmu_rb_node *node);
|
|
|
|
static int tid_rb_invalidate(void *arg, struct mmu_rb_node *mnode);
|
2017-08-22 09:26:51 +08:00
|
|
|
static int program_rcvarray(struct hfi1_filedata *fd, struct tid_user_buf *,
|
|
|
|
struct tid_group *grp,
|
|
|
|
unsigned int start, u16 count,
|
|
|
|
u32 *tidlist, unsigned int *tididx,
|
|
|
|
unsigned int *pmapped);
|
2017-05-04 20:14:45 +08:00
|
|
|
static int unprogram_rcvarray(struct hfi1_filedata *fd, u32 tidinfo,
|
2017-05-04 20:14:39 +08:00
|
|
|
struct tid_group **grp);
|
2016-07-29 03:21:26 +08:00
|
|
|
static void clear_tid_node(struct hfi1_filedata *fd, struct tid_rb_node *node);
|
2016-03-09 03:14:20 +08:00
|
|
|
|
|
|
|
static struct mmu_rb_ops tid_rb_ops = {
|
2016-07-29 00:27:37 +08:00
|
|
|
.insert = tid_rb_insert,
|
|
|
|
.remove = tid_rb_remove,
|
|
|
|
.invalidate = tid_rb_invalidate
|
2016-03-09 03:14:20 +08:00
|
|
|
};
|
2016-02-06 00:57:52 +08:00
|
|
|
|
2016-02-06 00:57:46 +08:00
|
|
|
/*
|
|
|
|
* Initialize context and file private data needed for Expected
|
|
|
|
* receive caching. This needs to be done after the context has
|
|
|
|
* been configured with the eager/expected RcvEntry counts.
|
|
|
|
*/
|
2017-07-29 23:43:32 +08:00
|
|
|
int hfi1_user_exp_rcv_init(struct hfi1_filedata *fd,
|
|
|
|
struct hfi1_ctxtdata *uctxt)
|
2016-02-06 00:57:46 +08:00
|
|
|
{
|
2016-02-06 00:57:54 +08:00
|
|
|
struct hfi1_devdata *dd = uctxt->dd;
|
2017-05-04 20:15:09 +08:00
|
|
|
int ret = 0;
|
2016-02-06 00:57:54 +08:00
|
|
|
|
|
|
|
spin_lock_init(&fd->tid_lock);
|
|
|
|
spin_lock_init(&fd->invalid_lock);
|
|
|
|
|
2016-02-04 06:34:41 +08:00
|
|
|
fd->entry_to_rb = kcalloc(uctxt->expected_count,
|
2017-05-04 20:15:09 +08:00
|
|
|
sizeof(struct rb_node *),
|
|
|
|
GFP_KERNEL);
|
2016-02-04 06:34:41 +08:00
|
|
|
if (!fd->entry_to_rb)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
2016-07-29 03:21:21 +08:00
|
|
|
if (!HFI1_CAP_UGET_MASK(uctxt->flags, TID_UNMAP)) {
|
2016-02-06 00:57:54 +08:00
|
|
|
fd->invalid_tid_idx = 0;
|
2017-02-09 22:30:53 +08:00
|
|
|
fd->invalid_tids = kcalloc(uctxt->expected_count,
|
|
|
|
sizeof(*fd->invalid_tids),
|
|
|
|
GFP_KERNEL);
|
2017-05-04 20:15:21 +08:00
|
|
|
if (!fd->invalid_tids) {
|
|
|
|
kfree(fd->entry_to_rb);
|
|
|
|
fd->entry_to_rb = NULL;
|
2017-05-04 20:15:09 +08:00
|
|
|
return -ENOMEM;
|
2017-05-04 20:15:21 +08:00
|
|
|
}
|
2016-02-04 06:34:41 +08:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Register MMU notifier callbacks. If the registration
|
2016-07-29 03:21:21 +08:00
|
|
|
* fails, continue without TID caching for this context.
|
2016-02-04 06:34:41 +08:00
|
|
|
*/
|
2016-07-29 03:21:24 +08:00
|
|
|
ret = hfi1_mmu_rb_register(fd, fd->mm, &tid_rb_ops,
|
|
|
|
dd->pport->hfi1_wq,
|
|
|
|
&fd->handler);
|
2016-02-04 06:34:41 +08:00
|
|
|
if (ret) {
|
|
|
|
dd_dev_info(dd,
|
|
|
|
"Failed MMU notifier registration %d\n",
|
|
|
|
ret);
|
|
|
|
ret = 0;
|
2016-02-06 00:57:54 +08:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* PSM does not have a good way to separate, count, and
|
|
|
|
* effectively enforce a limit on RcvArray entries used by
|
|
|
|
* subctxts (when context sharing is used) when TID caching
|
|
|
|
* is enabled. To help with that, we calculate a per-process
|
|
|
|
* RcvArray entry share and enforce that.
|
|
|
|
* If TID caching is not in use, PSM deals with usage on its
|
|
|
|
* own. In that case, we allow any subctxt to take all of the
|
|
|
|
* entries.
|
|
|
|
*
|
|
|
|
* Make sure that we set the tid counts only after successful
|
|
|
|
* init.
|
|
|
|
*/
|
2016-02-06 00:57:56 +08:00
|
|
|
spin_lock(&fd->tid_lock);
|
2016-07-29 03:21:21 +08:00
|
|
|
if (uctxt->subctxt_cnt && fd->handler) {
|
2016-02-06 00:57:54 +08:00
|
|
|
u16 remainder;
|
|
|
|
|
|
|
|
fd->tid_limit = uctxt->expected_count / uctxt->subctxt_cnt;
|
|
|
|
remainder = uctxt->expected_count % uctxt->subctxt_cnt;
|
|
|
|
if (remainder && fd->subctxt < remainder)
|
|
|
|
fd->tid_limit++;
|
|
|
|
} else {
|
|
|
|
fd->tid_limit = uctxt->expected_count;
|
|
|
|
}
|
2016-02-06 00:57:56 +08:00
|
|
|
spin_unlock(&fd->tid_lock);
|
2017-05-04 20:15:09 +08:00
|
|
|
|
2016-02-06 00:57:54 +08:00
|
|
|
return ret;
|
2016-02-06 00:57:46 +08:00
|
|
|
}
|
|
|
|
|
2017-05-04 20:15:09 +08:00
|
|
|
void hfi1_user_exp_rcv_free(struct hfi1_filedata *fd)
|
2016-02-06 00:57:46 +08:00
|
|
|
{
|
2016-02-06 00:57:54 +08:00
|
|
|
struct hfi1_ctxtdata *uctxt = fd->uctxt;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* The notifier would have been removed when the process'es mm
|
|
|
|
* was freed.
|
|
|
|
*/
|
2017-05-04 20:14:34 +08:00
|
|
|
if (fd->handler) {
|
2016-07-29 03:21:20 +08:00
|
|
|
hfi1_mmu_rb_unregister(fd->handler);
|
2017-05-04 20:14:34 +08:00
|
|
|
} else {
|
2016-02-06 00:57:54 +08:00
|
|
|
if (!EXP_TID_SET_EMPTY(uctxt->tid_full_list))
|
2016-07-29 03:21:20 +08:00
|
|
|
unlock_exp_tids(uctxt, &uctxt->tid_full_list, fd);
|
2016-02-06 00:57:54 +08:00
|
|
|
if (!EXP_TID_SET_EMPTY(uctxt->tid_used_list))
|
2016-07-29 03:21:20 +08:00
|
|
|
unlock_exp_tids(uctxt, &uctxt->tid_used_list, fd);
|
2016-02-06 00:57:54 +08:00
|
|
|
}
|
2016-02-04 06:34:41 +08:00
|
|
|
|
2017-05-04 20:14:34 +08:00
|
|
|
kfree(fd->invalid_tids);
|
|
|
|
fd->invalid_tids = NULL;
|
|
|
|
|
2016-02-04 06:34:41 +08:00
|
|
|
kfree(fd->entry_to_rb);
|
2017-05-04 20:14:34 +08:00
|
|
|
fd->entry_to_rb = NULL;
|
2016-02-06 00:57:46 +08:00
|
|
|
}
|
|
|
|
|
2017-08-22 09:26:51 +08:00
|
|
|
/**
|
|
|
|
* Release pinned receive buffer pages.
|
|
|
|
*
|
|
|
|
* @mapped - true if the pages have been DMA mapped. false otherwise.
|
|
|
|
* @idx - Index of the first page to unpin.
|
|
|
|
* @npages - No of pages to unpin.
|
|
|
|
*
|
|
|
|
* If the pages have been DMA mapped (indicated by mapped parameter), their
|
|
|
|
* info will be passed via a struct tid_rb_node. If they haven't been mapped,
|
|
|
|
* their info will be passed via a struct tid_user_buf.
|
|
|
|
*/
|
|
|
|
static void unpin_rcv_pages(struct hfi1_filedata *fd,
|
|
|
|
struct tid_user_buf *tidbuf,
|
|
|
|
struct tid_rb_node *node,
|
|
|
|
unsigned int idx,
|
|
|
|
unsigned int npages,
|
|
|
|
bool mapped)
|
|
|
|
{
|
|
|
|
struct page **pages;
|
|
|
|
struct hfi1_devdata *dd = fd->uctxt->dd;
|
|
|
|
|
|
|
|
if (mapped) {
|
|
|
|
pci_unmap_single(dd->pcidev, node->dma_addr,
|
|
|
|
node->mmu.len, PCI_DMA_FROMDEVICE);
|
|
|
|
pages = &node->pages[idx];
|
|
|
|
} else {
|
|
|
|
pages = &tidbuf->pages[idx];
|
|
|
|
}
|
|
|
|
hfi1_release_user_pages(fd->mm, pages, npages, mapped);
|
|
|
|
fd->tid_n_pinned -= npages;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Pin receive buffer pages.
|
|
|
|
*/
|
|
|
|
static int pin_rcv_pages(struct hfi1_filedata *fd, struct tid_user_buf *tidbuf)
|
|
|
|
{
|
|
|
|
int pinned;
|
|
|
|
unsigned int npages;
|
|
|
|
unsigned long vaddr = tidbuf->vaddr;
|
|
|
|
struct page **pages = NULL;
|
|
|
|
struct hfi1_devdata *dd = fd->uctxt->dd;
|
|
|
|
|
|
|
|
/* Get the number of pages the user buffer spans */
|
|
|
|
npages = num_user_pages(vaddr, tidbuf->length);
|
|
|
|
if (!npages)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
if (npages > fd->uctxt->expected_count) {
|
|
|
|
dd_dev_err(dd, "Expected buffer too big\n");
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Verify that access is OK for the user buffer */
|
Remove 'type' argument from access_ok() function
Nobody has actually used the type (VERIFY_READ vs VERIFY_WRITE) argument
of the user address range verification function since we got rid of the
old racy i386-only code to walk page tables by hand.
It existed because the original 80386 would not honor the write protect
bit when in kernel mode, so you had to do COW by hand before doing any
user access. But we haven't supported that in a long time, and these
days the 'type' argument is a purely historical artifact.
A discussion about extending 'user_access_begin()' to do the range
checking resulted this patch, because there is no way we're going to
move the old VERIFY_xyz interface to that model. And it's best done at
the end of the merge window when I've done most of my merges, so let's
just get this done once and for all.
This patch was mostly done with a sed-script, with manual fix-ups for
the cases that weren't of the trivial 'access_ok(VERIFY_xyz' form.
There were a couple of notable cases:
- csky still had the old "verify_area()" name as an alias.
- the iter_iov code had magical hardcoded knowledge of the actual
values of VERIFY_{READ,WRITE} (not that they mattered, since nothing
really used it)
- microblaze used the type argument for a debug printout
but other than those oddities this should be a total no-op patch.
I tried to fix up all architectures, did fairly extensive grepping for
access_ok() uses, and the changes are trivial, but I may have missed
something. Any missed conversion should be trivially fixable, though.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2019-01-04 10:57:57 +08:00
|
|
|
if (!access_ok((void __user *)vaddr,
|
2017-08-22 09:26:51 +08:00
|
|
|
npages * PAGE_SIZE)) {
|
|
|
|
dd_dev_err(dd, "Fail vaddr %p, %u pages, !access_ok\n",
|
|
|
|
(void *)vaddr, npages);
|
|
|
|
return -EFAULT;
|
|
|
|
}
|
|
|
|
/* Allocate the array of struct page pointers needed for pinning */
|
|
|
|
pages = kcalloc(npages, sizeof(*pages), GFP_KERNEL);
|
|
|
|
if (!pages)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Pin all the pages of the user buffer. If we can't pin all the
|
|
|
|
* pages, accept the amount pinned so far and program only that.
|
|
|
|
* User space knows how to deal with partially programmed buffers.
|
|
|
|
*/
|
|
|
|
if (!hfi1_can_pin_pages(dd, fd->mm, fd->tid_n_pinned, npages)) {
|
|
|
|
kfree(pages);
|
|
|
|
return -ENOMEM;
|
|
|
|
}
|
|
|
|
|
|
|
|
pinned = hfi1_acquire_user_pages(fd->mm, vaddr, npages, true, pages);
|
|
|
|
if (pinned <= 0) {
|
|
|
|
kfree(pages);
|
|
|
|
return pinned;
|
|
|
|
}
|
|
|
|
tidbuf->pages = pages;
|
|
|
|
tidbuf->npages = npages;
|
|
|
|
fd->tid_n_pinned += pinned;
|
|
|
|
return pinned;
|
|
|
|
}
|
|
|
|
|
2016-02-06 00:57:57 +08:00
|
|
|
/*
|
|
|
|
* RcvArray entry allocation for Expected Receives is done by the
|
|
|
|
* following algorithm:
|
|
|
|
*
|
|
|
|
* The context keeps 3 lists of groups of RcvArray entries:
|
|
|
|
* 1. List of empty groups - tid_group_list
|
|
|
|
* This list is created during user context creation and
|
|
|
|
* contains elements which describe sets (of 8) of empty
|
|
|
|
* RcvArray entries.
|
|
|
|
* 2. List of partially used groups - tid_used_list
|
|
|
|
* This list contains sets of RcvArray entries which are
|
|
|
|
* not completely used up. Another mapping request could
|
|
|
|
* use some of all of the remaining entries.
|
|
|
|
* 3. List of full groups - tid_full_list
|
|
|
|
* This is the list where sets that are completely used
|
|
|
|
* up go.
|
|
|
|
*
|
|
|
|
* An attempt to optimize the usage of RcvArray entries is
|
|
|
|
* made by finding all sets of physically contiguous pages in a
|
|
|
|
* user's buffer.
|
|
|
|
* These physically contiguous sets are further split into
|
|
|
|
* sizes supported by the receive engine of the HFI. The
|
|
|
|
* resulting sets of pages are stored in struct tid_pageset,
|
|
|
|
* which describes the sets as:
|
|
|
|
* * .count - number of pages in this set
|
|
|
|
* * .idx - starting index into struct page ** array
|
|
|
|
* of this set
|
|
|
|
*
|
|
|
|
* From this point on, the algorithm deals with the page sets
|
|
|
|
* described above. The number of pagesets is divided by the
|
|
|
|
* RcvArray group size to produce the number of full groups
|
|
|
|
* needed.
|
|
|
|
*
|
|
|
|
* Groups from the 3 lists are manipulated using the following
|
|
|
|
* rules:
|
|
|
|
* 1. For each set of 8 pagesets, a complete group from
|
|
|
|
* tid_group_list is taken, programmed, and moved to
|
|
|
|
* the tid_full_list list.
|
|
|
|
* 2. For all remaining pagesets:
|
|
|
|
* 2.1 If the tid_used_list is empty and the tid_group_list
|
|
|
|
* is empty, stop processing pageset and return only
|
|
|
|
* what has been programmed up to this point.
|
|
|
|
* 2.2 If the tid_used_list is empty and the tid_group_list
|
|
|
|
* is not empty, move a group from tid_group_list to
|
|
|
|
* tid_used_list.
|
|
|
|
* 2.3 For each group is tid_used_group, program as much as
|
|
|
|
* can fit into the group. If the group becomes fully
|
|
|
|
* used, move it to tid_full_list.
|
|
|
|
*/
|
2017-05-04 20:14:45 +08:00
|
|
|
int hfi1_user_exp_rcv_setup(struct hfi1_filedata *fd,
|
|
|
|
struct hfi1_tid_info *tinfo)
|
2016-02-06 00:57:46 +08:00
|
|
|
{
|
2016-02-06 00:57:57 +08:00
|
|
|
int ret = 0, need_group = 0, pinned;
|
|
|
|
struct hfi1_ctxtdata *uctxt = fd->uctxt;
|
|
|
|
struct hfi1_devdata *dd = uctxt->dd;
|
2017-08-22 09:26:51 +08:00
|
|
|
unsigned int ngroups, pageidx = 0, pageset_count,
|
2016-02-06 00:57:57 +08:00
|
|
|
tididx = 0, mapped, mapped_pages = 0;
|
|
|
|
u32 *tidlist = NULL;
|
2017-08-22 09:26:51 +08:00
|
|
|
struct tid_user_buf *tidbuf;
|
2016-02-06 00:57:57 +08:00
|
|
|
|
2019-05-24 23:45:04 +08:00
|
|
|
if (!PAGE_ALIGNED(tinfo->vaddr))
|
|
|
|
return -EINVAL;
|
|
|
|
|
2017-08-22 09:26:51 +08:00
|
|
|
tidbuf = kzalloc(sizeof(*tidbuf), GFP_KERNEL);
|
|
|
|
if (!tidbuf)
|
2016-02-06 00:57:57 +08:00
|
|
|
return -ENOMEM;
|
|
|
|
|
2017-08-22 09:26:51 +08:00
|
|
|
tidbuf->vaddr = tinfo->vaddr;
|
|
|
|
tidbuf->length = tinfo->length;
|
|
|
|
tidbuf->psets = kcalloc(uctxt->expected_count, sizeof(*tidbuf->psets),
|
|
|
|
GFP_KERNEL);
|
|
|
|
if (!tidbuf->psets) {
|
|
|
|
kfree(tidbuf);
|
|
|
|
return -ENOMEM;
|
2016-04-13 01:46:29 +08:00
|
|
|
}
|
|
|
|
|
2017-08-22 09:26:51 +08:00
|
|
|
pinned = pin_rcv_pages(fd, tidbuf);
|
2016-02-06 00:57:57 +08:00
|
|
|
if (pinned <= 0) {
|
2017-08-22 09:26:51 +08:00
|
|
|
kfree(tidbuf->psets);
|
|
|
|
kfree(tidbuf);
|
|
|
|
return pinned;
|
2016-02-06 00:57:57 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
/* Find sets of physically contiguous pages */
|
2017-08-22 09:26:51 +08:00
|
|
|
tidbuf->n_psets = find_phys_blocks(tidbuf, pinned);
|
2016-02-06 00:57:57 +08:00
|
|
|
|
|
|
|
/*
|
|
|
|
* We don't need to access this under a lock since tid_used is per
|
|
|
|
* process and the same process cannot be in hfi1_user_exp_rcv_clear()
|
|
|
|
* and hfi1_user_exp_rcv_setup() at the same time.
|
|
|
|
*/
|
|
|
|
spin_lock(&fd->tid_lock);
|
2017-08-22 09:26:51 +08:00
|
|
|
if (fd->tid_used + tidbuf->n_psets > fd->tid_limit)
|
2016-02-06 00:57:57 +08:00
|
|
|
pageset_count = fd->tid_limit - fd->tid_used;
|
|
|
|
else
|
2017-08-22 09:26:51 +08:00
|
|
|
pageset_count = tidbuf->n_psets;
|
2016-02-06 00:57:57 +08:00
|
|
|
spin_unlock(&fd->tid_lock);
|
|
|
|
|
|
|
|
if (!pageset_count)
|
|
|
|
goto bail;
|
|
|
|
|
|
|
|
ngroups = pageset_count / dd->rcv_entries.group_size;
|
|
|
|
tidlist = kcalloc(pageset_count, sizeof(*tidlist), GFP_KERNEL);
|
|
|
|
if (!tidlist) {
|
|
|
|
ret = -ENOMEM;
|
|
|
|
goto nomem;
|
|
|
|
}
|
|
|
|
|
|
|
|
tididx = 0;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* From this point on, we are going to be using shared (between master
|
|
|
|
* and subcontexts) context resources. We need to take the lock.
|
|
|
|
*/
|
2018-06-05 02:43:54 +08:00
|
|
|
mutex_lock(&uctxt->exp_mutex);
|
2016-02-06 00:57:57 +08:00
|
|
|
/*
|
|
|
|
* The first step is to program the RcvArray entries which are complete
|
|
|
|
* groups.
|
|
|
|
*/
|
|
|
|
while (ngroups && uctxt->tid_group_list.count) {
|
|
|
|
struct tid_group *grp =
|
|
|
|
tid_group_pop(&uctxt->tid_group_list);
|
|
|
|
|
2017-08-22 09:26:51 +08:00
|
|
|
ret = program_rcvarray(fd, tidbuf, grp,
|
2016-02-06 00:57:57 +08:00
|
|
|
pageidx, dd->rcv_entries.group_size,
|
2017-08-22 09:26:51 +08:00
|
|
|
tidlist, &tididx, &mapped);
|
2016-02-06 00:57:57 +08:00
|
|
|
/*
|
|
|
|
* If there was a failure to program the RcvArray
|
|
|
|
* entries for the entire group, reset the grp fields
|
|
|
|
* and add the grp back to the free group list.
|
|
|
|
*/
|
|
|
|
if (ret <= 0) {
|
|
|
|
tid_group_add_tail(grp, &uctxt->tid_group_list);
|
|
|
|
hfi1_cdbg(TID,
|
|
|
|
"Failed to program RcvArray group %d", ret);
|
|
|
|
goto unlock;
|
|
|
|
}
|
|
|
|
|
|
|
|
tid_group_add_tail(grp, &uctxt->tid_full_list);
|
|
|
|
ngroups--;
|
|
|
|
pageidx += ret;
|
|
|
|
mapped_pages += mapped;
|
|
|
|
}
|
|
|
|
|
|
|
|
while (pageidx < pageset_count) {
|
|
|
|
struct tid_group *grp, *ptr;
|
|
|
|
/*
|
|
|
|
* If we don't have any partially used tid groups, check
|
|
|
|
* if we have empty groups. If so, take one from there and
|
|
|
|
* put in the partially used list.
|
|
|
|
*/
|
|
|
|
if (!uctxt->tid_used_list.count || need_group) {
|
|
|
|
if (!uctxt->tid_group_list.count)
|
|
|
|
goto unlock;
|
|
|
|
|
|
|
|
grp = tid_group_pop(&uctxt->tid_group_list);
|
|
|
|
tid_group_add_tail(grp, &uctxt->tid_used_list);
|
|
|
|
need_group = 0;
|
|
|
|
}
|
|
|
|
/*
|
|
|
|
* There is an optimization opportunity here - instead of
|
|
|
|
* fitting as many page sets as we can, check for a group
|
|
|
|
* later on in the list that could fit all of them.
|
|
|
|
*/
|
|
|
|
list_for_each_entry_safe(grp, ptr, &uctxt->tid_used_list.list,
|
|
|
|
list) {
|
|
|
|
unsigned use = min_t(unsigned, pageset_count - pageidx,
|
|
|
|
grp->size - grp->used);
|
|
|
|
|
2017-08-22 09:26:51 +08:00
|
|
|
ret = program_rcvarray(fd, tidbuf, grp,
|
|
|
|
pageidx, use, tidlist,
|
2016-02-06 00:57:57 +08:00
|
|
|
&tididx, &mapped);
|
|
|
|
if (ret < 0) {
|
|
|
|
hfi1_cdbg(TID,
|
|
|
|
"Failed to program RcvArray entries %d",
|
|
|
|
ret);
|
|
|
|
goto unlock;
|
|
|
|
} else if (ret > 0) {
|
|
|
|
if (grp->used == grp->size)
|
|
|
|
tid_group_move(grp,
|
|
|
|
&uctxt->tid_used_list,
|
|
|
|
&uctxt->tid_full_list);
|
|
|
|
pageidx += ret;
|
|
|
|
mapped_pages += mapped;
|
|
|
|
need_group = 0;
|
|
|
|
/* Check if we are done so we break out early */
|
|
|
|
if (pageidx >= pageset_count)
|
|
|
|
break;
|
|
|
|
} else if (WARN_ON(ret == 0)) {
|
|
|
|
/*
|
|
|
|
* If ret is 0, we did not program any entries
|
|
|
|
* into this group, which can only happen if
|
|
|
|
* we've screwed up the accounting somewhere.
|
|
|
|
* Warn and try to continue.
|
|
|
|
*/
|
|
|
|
need_group = 1;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
unlock:
|
2018-06-05 02:43:54 +08:00
|
|
|
mutex_unlock(&uctxt->exp_mutex);
|
2016-02-06 00:57:57 +08:00
|
|
|
nomem:
|
|
|
|
hfi1_cdbg(TID, "total mapped: tidpairs:%u pages:%u (%d)", tididx,
|
|
|
|
mapped_pages, ret);
|
|
|
|
if (tididx) {
|
|
|
|
spin_lock(&fd->tid_lock);
|
|
|
|
fd->tid_used += tididx;
|
|
|
|
spin_unlock(&fd->tid_lock);
|
|
|
|
tinfo->tidcnt = tididx;
|
|
|
|
tinfo->length = mapped_pages * PAGE_SIZE;
|
|
|
|
|
2018-03-28 04:18:47 +08:00
|
|
|
if (copy_to_user(u64_to_user_ptr(tinfo->tidlist),
|
2016-02-06 00:57:57 +08:00
|
|
|
tidlist, sizeof(tidlist[0]) * tididx)) {
|
|
|
|
/*
|
|
|
|
* On failure to copy to the user level, we need to undo
|
|
|
|
* everything done so far so we don't leak resources.
|
|
|
|
*/
|
|
|
|
tinfo->tidlist = (unsigned long)&tidlist;
|
2017-05-04 20:14:45 +08:00
|
|
|
hfi1_user_exp_rcv_clear(fd, tinfo);
|
2016-02-06 00:57:57 +08:00
|
|
|
tinfo->tidlist = 0;
|
|
|
|
ret = -EFAULT;
|
|
|
|
goto bail;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If not everything was mapped (due to insufficient RcvArray entries,
|
|
|
|
* for example), unpin all unmapped pages so we can pin them nex time.
|
|
|
|
*/
|
2017-08-22 09:26:51 +08:00
|
|
|
if (mapped_pages != pinned)
|
|
|
|
unpin_rcv_pages(fd, tidbuf, NULL, mapped_pages,
|
|
|
|
(pinned - mapped_pages), false);
|
2016-02-06 00:57:57 +08:00
|
|
|
bail:
|
2017-08-22 09:26:51 +08:00
|
|
|
kfree(tidbuf->psets);
|
2016-02-06 00:57:57 +08:00
|
|
|
kfree(tidlist);
|
2017-08-22 09:26:51 +08:00
|
|
|
kfree(tidbuf->pages);
|
|
|
|
kfree(tidbuf);
|
2016-02-06 00:57:57 +08:00
|
|
|
return ret > 0 ? 0 : ret;
|
2016-02-06 00:57:46 +08:00
|
|
|
}
|
|
|
|
|
2017-05-04 20:14:45 +08:00
|
|
|
int hfi1_user_exp_rcv_clear(struct hfi1_filedata *fd,
|
|
|
|
struct hfi1_tid_info *tinfo)
|
2016-02-06 00:57:46 +08:00
|
|
|
{
|
2016-02-06 00:57:56 +08:00
|
|
|
int ret = 0;
|
|
|
|
struct hfi1_ctxtdata *uctxt = fd->uctxt;
|
|
|
|
u32 *tidinfo;
|
|
|
|
unsigned tididx;
|
|
|
|
|
2017-04-10 01:16:03 +08:00
|
|
|
if (unlikely(tinfo->tidcnt > fd->tid_used))
|
|
|
|
return -EINVAL;
|
|
|
|
|
2018-03-28 04:18:47 +08:00
|
|
|
tidinfo = memdup_user(u64_to_user_ptr(tinfo->tidlist),
|
2017-02-08 21:28:31 +08:00
|
|
|
sizeof(tidinfo[0]) * tinfo->tidcnt);
|
|
|
|
if (IS_ERR(tidinfo))
|
|
|
|
return PTR_ERR(tidinfo);
|
2016-02-06 00:57:56 +08:00
|
|
|
|
2018-06-05 02:43:54 +08:00
|
|
|
mutex_lock(&uctxt->exp_mutex);
|
2016-02-06 00:57:56 +08:00
|
|
|
for (tididx = 0; tididx < tinfo->tidcnt; tididx++) {
|
2017-05-04 20:14:45 +08:00
|
|
|
ret = unprogram_rcvarray(fd, tidinfo[tididx], NULL);
|
2016-02-06 00:57:56 +08:00
|
|
|
if (ret) {
|
|
|
|
hfi1_cdbg(TID, "Failed to unprogram rcv array %d",
|
|
|
|
ret);
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
spin_lock(&fd->tid_lock);
|
|
|
|
fd->tid_used -= tididx;
|
|
|
|
spin_unlock(&fd->tid_lock);
|
|
|
|
tinfo->tidcnt = tididx;
|
2018-06-05 02:43:54 +08:00
|
|
|
mutex_unlock(&uctxt->exp_mutex);
|
2017-02-08 21:28:31 +08:00
|
|
|
|
2016-02-06 00:57:56 +08:00
|
|
|
kfree(tidinfo);
|
|
|
|
return ret;
|
2016-02-06 00:57:46 +08:00
|
|
|
}
|
|
|
|
|
2017-05-04 20:14:45 +08:00
|
|
|
int hfi1_user_exp_rcv_invalid(struct hfi1_filedata *fd,
|
|
|
|
struct hfi1_tid_info *tinfo)
|
2016-02-06 00:57:46 +08:00
|
|
|
{
|
2016-02-06 00:57:56 +08:00
|
|
|
struct hfi1_ctxtdata *uctxt = fd->uctxt;
|
|
|
|
unsigned long *ev = uctxt->dd->events +
|
2017-09-26 22:00:56 +08:00
|
|
|
(uctxt_offset(uctxt) + fd->subctxt);
|
2016-02-06 00:57:56 +08:00
|
|
|
u32 *array;
|
|
|
|
int ret = 0;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* copy_to_user() can sleep, which will leave the invalid_lock
|
|
|
|
* locked and cause the MMU notifier to be blocked on the lock
|
|
|
|
* for a long time.
|
|
|
|
* Copy the data to a local buffer so we can release the lock.
|
|
|
|
*/
|
|
|
|
array = kcalloc(uctxt->expected_count, sizeof(*array), GFP_KERNEL);
|
|
|
|
if (!array)
|
|
|
|
return -EFAULT;
|
|
|
|
|
|
|
|
spin_lock(&fd->invalid_lock);
|
|
|
|
if (fd->invalid_tid_idx) {
|
|
|
|
memcpy(array, fd->invalid_tids, sizeof(*array) *
|
|
|
|
fd->invalid_tid_idx);
|
|
|
|
memset(fd->invalid_tids, 0, sizeof(*fd->invalid_tids) *
|
|
|
|
fd->invalid_tid_idx);
|
|
|
|
tinfo->tidcnt = fd->invalid_tid_idx;
|
|
|
|
fd->invalid_tid_idx = 0;
|
|
|
|
/*
|
|
|
|
* Reset the user flag while still holding the lock.
|
|
|
|
* Otherwise, PSM can miss events.
|
|
|
|
*/
|
|
|
|
clear_bit(_HFI1_EVENT_TID_MMU_NOTIFY_BIT, ev);
|
|
|
|
} else {
|
|
|
|
tinfo->tidcnt = 0;
|
|
|
|
}
|
|
|
|
spin_unlock(&fd->invalid_lock);
|
|
|
|
|
|
|
|
if (tinfo->tidcnt) {
|
|
|
|
if (copy_to_user((void __user *)tinfo->tidlist,
|
|
|
|
array, sizeof(*array) * tinfo->tidcnt))
|
|
|
|
ret = -EFAULT;
|
|
|
|
}
|
|
|
|
kfree(array);
|
|
|
|
|
|
|
|
return ret;
|
2016-02-06 00:57:46 +08:00
|
|
|
}
|
|
|
|
|
2017-08-22 09:26:51 +08:00
|
|
|
static u32 find_phys_blocks(struct tid_user_buf *tidbuf, unsigned int npages)
|
2016-02-06 00:57:52 +08:00
|
|
|
{
|
|
|
|
unsigned pagecount, pageidx, setcount = 0, i;
|
|
|
|
unsigned long pfn, this_pfn;
|
2017-08-22 09:26:51 +08:00
|
|
|
struct page **pages = tidbuf->pages;
|
|
|
|
struct tid_pageset *list = tidbuf->psets;
|
2016-02-06 00:57:52 +08:00
|
|
|
|
|
|
|
if (!npages)
|
|
|
|
return 0;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Look for sets of physically contiguous pages in the user buffer.
|
|
|
|
* This will allow us to optimize Expected RcvArray entry usage by
|
|
|
|
* using the bigger supported sizes.
|
|
|
|
*/
|
|
|
|
pfn = page_to_pfn(pages[0]);
|
|
|
|
for (pageidx = 0, pagecount = 1, i = 1; i <= npages; i++) {
|
|
|
|
this_pfn = i < npages ? page_to_pfn(pages[i]) : 0;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If the pfn's are not sequential, pages are not physically
|
|
|
|
* contiguous.
|
|
|
|
*/
|
|
|
|
if (this_pfn != ++pfn) {
|
|
|
|
/*
|
|
|
|
* At this point we have to loop over the set of
|
|
|
|
* physically contiguous pages and break them down it
|
|
|
|
* sizes supported by the HW.
|
|
|
|
* There are two main constraints:
|
|
|
|
* 1. The max buffer size is MAX_EXPECTED_BUFFER.
|
|
|
|
* If the total set size is bigger than that
|
|
|
|
* program only a MAX_EXPECTED_BUFFER chunk.
|
|
|
|
* 2. The buffer size has to be a power of two. If
|
|
|
|
* it is not, round down to the closes power of
|
|
|
|
* 2 and program that size.
|
|
|
|
*/
|
|
|
|
while (pagecount) {
|
|
|
|
int maxpages = pagecount;
|
|
|
|
u32 bufsize = pagecount * PAGE_SIZE;
|
|
|
|
|
|
|
|
if (bufsize > MAX_EXPECTED_BUFFER)
|
|
|
|
maxpages =
|
|
|
|
MAX_EXPECTED_BUFFER >>
|
|
|
|
PAGE_SHIFT;
|
|
|
|
else if (!is_power_of_2(bufsize))
|
|
|
|
maxpages =
|
|
|
|
rounddown_pow_of_two(bufsize) >>
|
|
|
|
PAGE_SHIFT;
|
|
|
|
|
|
|
|
list[setcount].idx = pageidx;
|
|
|
|
list[setcount].count = maxpages;
|
|
|
|
pagecount -= maxpages;
|
|
|
|
pageidx += maxpages;
|
|
|
|
setcount++;
|
|
|
|
}
|
|
|
|
pageidx = i;
|
|
|
|
pagecount = 1;
|
|
|
|
pfn = this_pfn;
|
|
|
|
} else {
|
|
|
|
pagecount++;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
return setcount;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* program_rcvarray() - program an RcvArray group with receive buffers
|
2017-05-04 20:14:45 +08:00
|
|
|
* @fd: filedata pointer
|
2017-08-22 09:26:51 +08:00
|
|
|
* @tbuf: pointer to struct tid_user_buf that has the user buffer starting
|
|
|
|
* virtual address, buffer length, page pointers, pagesets (array of
|
|
|
|
* struct tid_pageset holding information on physically contiguous
|
|
|
|
* chunks from the user buffer), and other fields.
|
2016-02-06 00:57:52 +08:00
|
|
|
* @grp: RcvArray group
|
|
|
|
* @start: starting index into sets array
|
|
|
|
* @count: number of struct tid_pageset's to program
|
|
|
|
* @tidlist: the array of u32 elements when the information about the
|
|
|
|
* programmed RcvArray entries is to be encoded.
|
|
|
|
* @tididx: starting offset into tidlist
|
|
|
|
* @pmapped: (output parameter) number of pages programmed into the RcvArray
|
|
|
|
* entries.
|
|
|
|
*
|
|
|
|
* This function will program up to 'count' number of RcvArray entries from the
|
|
|
|
* group 'grp'. To make best use of write-combining writes, the function will
|
|
|
|
* perform writes to the unused RcvArray entries which will be ignored by the
|
|
|
|
* HW. Each RcvArray entry will be programmed with a physically contiguous
|
|
|
|
* buffer chunk from the user's virtual buffer.
|
|
|
|
*
|
|
|
|
* Return:
|
|
|
|
* -EINVAL if the requested count is larger than the size of the group,
|
|
|
|
* -ENOMEM or -EFAULT on error from set_rcvarray_entry(), or
|
|
|
|
* number of RcvArray entries programmed.
|
|
|
|
*/
|
2017-08-22 09:26:51 +08:00
|
|
|
static int program_rcvarray(struct hfi1_filedata *fd, struct tid_user_buf *tbuf,
|
2016-02-06 00:57:52 +08:00
|
|
|
struct tid_group *grp,
|
2017-08-22 09:26:51 +08:00
|
|
|
unsigned int start, u16 count,
|
|
|
|
u32 *tidlist, unsigned int *tididx,
|
|
|
|
unsigned int *pmapped)
|
2016-02-06 00:57:52 +08:00
|
|
|
{
|
|
|
|
struct hfi1_ctxtdata *uctxt = fd->uctxt;
|
|
|
|
struct hfi1_devdata *dd = uctxt->dd;
|
|
|
|
u16 idx;
|
|
|
|
u32 tidinfo = 0, rcventry, useidx = 0;
|
|
|
|
int mapped = 0;
|
|
|
|
|
|
|
|
/* Count should never be larger than the group size */
|
|
|
|
if (count > grp->size)
|
|
|
|
return -EINVAL;
|
|
|
|
|
|
|
|
/* Find the first unused entry in the group */
|
|
|
|
for (idx = 0; idx < grp->size; idx++) {
|
|
|
|
if (!(grp->map & (1 << idx))) {
|
|
|
|
useidx = idx;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
rcv_array_wc_fill(dd, grp->base + idx);
|
|
|
|
}
|
|
|
|
|
|
|
|
idx = 0;
|
|
|
|
while (idx < count) {
|
|
|
|
u16 npages, pageidx, setidx = start + idx;
|
|
|
|
int ret = 0;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If this entry in the group is used, move to the next one.
|
|
|
|
* If we go past the end of the group, exit the loop.
|
|
|
|
*/
|
|
|
|
if (useidx >= grp->size) {
|
|
|
|
break;
|
|
|
|
} else if (grp->map & (1 << useidx)) {
|
|
|
|
rcv_array_wc_fill(dd, grp->base + useidx);
|
|
|
|
useidx++;
|
|
|
|
continue;
|
|
|
|
}
|
|
|
|
|
|
|
|
rcventry = grp->base + useidx;
|
2017-08-22 09:26:51 +08:00
|
|
|
npages = tbuf->psets[setidx].count;
|
|
|
|
pageidx = tbuf->psets[setidx].idx;
|
2016-02-06 00:57:52 +08:00
|
|
|
|
2017-08-22 09:26:51 +08:00
|
|
|
ret = set_rcvarray_entry(fd, tbuf,
|
|
|
|
rcventry, grp, pageidx,
|
2016-02-06 00:57:52 +08:00
|
|
|
npages);
|
|
|
|
if (ret)
|
|
|
|
return ret;
|
|
|
|
mapped += npages;
|
|
|
|
|
|
|
|
tidinfo = rcventry2tidinfo(rcventry - uctxt->expected_base) |
|
|
|
|
EXP_TID_SET(LEN, npages);
|
|
|
|
tidlist[(*tididx)++] = tidinfo;
|
|
|
|
grp->used++;
|
|
|
|
grp->map |= 1 << useidx++;
|
|
|
|
idx++;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Fill the rest of the group with "blank" writes */
|
|
|
|
for (; useidx < grp->size; useidx++)
|
|
|
|
rcv_array_wc_fill(dd, grp->base + useidx);
|
|
|
|
*pmapped = mapped;
|
|
|
|
return idx;
|
|
|
|
}
|
|
|
|
|
2017-08-22 09:26:51 +08:00
|
|
|
static int set_rcvarray_entry(struct hfi1_filedata *fd,
|
|
|
|
struct tid_user_buf *tbuf,
|
2016-02-06 00:57:52 +08:00
|
|
|
u32 rcventry, struct tid_group *grp,
|
2017-08-22 09:26:51 +08:00
|
|
|
u16 pageidx, unsigned int npages)
|
2016-02-06 00:57:52 +08:00
|
|
|
{
|
|
|
|
int ret;
|
|
|
|
struct hfi1_ctxtdata *uctxt = fd->uctxt;
|
2016-03-09 03:14:20 +08:00
|
|
|
struct tid_rb_node *node;
|
2016-02-06 00:57:52 +08:00
|
|
|
struct hfi1_devdata *dd = uctxt->dd;
|
|
|
|
dma_addr_t phys;
|
2017-08-22 09:26:51 +08:00
|
|
|
struct page **pages = tbuf->pages + pageidx;
|
2016-02-06 00:57:52 +08:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Allocate the node first so we can handle a potential
|
|
|
|
* failure before we've programmed anything.
|
|
|
|
*/
|
|
|
|
node = kzalloc(sizeof(*node) + (sizeof(struct page *) * npages),
|
|
|
|
GFP_KERNEL);
|
|
|
|
if (!node)
|
|
|
|
return -ENOMEM;
|
|
|
|
|
|
|
|
phys = pci_map_single(dd->pcidev,
|
|
|
|
__va(page_to_phys(pages[0])),
|
|
|
|
npages * PAGE_SIZE, PCI_DMA_FROMDEVICE);
|
|
|
|
if (dma_mapping_error(&dd->pcidev->dev, phys)) {
|
|
|
|
dd_dev_err(dd, "Failed to DMA map Exp Rcv pages 0x%llx\n",
|
|
|
|
phys);
|
|
|
|
kfree(node);
|
|
|
|
return -EFAULT;
|
|
|
|
}
|
|
|
|
|
2017-08-22 09:26:51 +08:00
|
|
|
node->mmu.addr = tbuf->vaddr + (pageidx * PAGE_SIZE);
|
2016-03-09 03:14:20 +08:00
|
|
|
node->mmu.len = npages * PAGE_SIZE;
|
2016-02-06 00:57:52 +08:00
|
|
|
node->phys = page_to_phys(pages[0]);
|
|
|
|
node->npages = npages;
|
|
|
|
node->rcventry = rcventry;
|
|
|
|
node->dma_addr = phys;
|
|
|
|
node->grp = grp;
|
|
|
|
node->freed = false;
|
|
|
|
memcpy(node->pages, pages, sizeof(struct page *) * npages);
|
|
|
|
|
2016-07-29 03:21:21 +08:00
|
|
|
if (!fd->handler)
|
2016-07-29 03:21:20 +08:00
|
|
|
ret = tid_rb_insert(fd, &node->mmu);
|
2016-03-09 03:14:42 +08:00
|
|
|
else
|
2016-07-29 03:21:20 +08:00
|
|
|
ret = hfi1_mmu_rb_insert(fd->handler, &node->mmu);
|
2016-02-06 00:57:52 +08:00
|
|
|
|
|
|
|
if (ret) {
|
|
|
|
hfi1_cdbg(TID, "Failed to insert RB node %u 0x%lx, 0x%lx %d",
|
2016-03-09 03:14:20 +08:00
|
|
|
node->rcventry, node->mmu.addr, node->phys, ret);
|
2016-02-06 00:57:52 +08:00
|
|
|
pci_unmap_single(dd->pcidev, phys, npages * PAGE_SIZE,
|
|
|
|
PCI_DMA_FROMDEVICE);
|
|
|
|
kfree(node);
|
|
|
|
return -EFAULT;
|
|
|
|
}
|
|
|
|
hfi1_put_tid(dd, rcventry, PT_EXPECTED, phys, ilog2(npages) + 1);
|
2016-03-09 03:14:20 +08:00
|
|
|
trace_hfi1_exp_tid_reg(uctxt->ctxt, fd->subctxt, rcventry, npages,
|
|
|
|
node->mmu.addr, node->phys, phys);
|
2016-02-06 00:57:52 +08:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2017-05-04 20:14:45 +08:00
|
|
|
static int unprogram_rcvarray(struct hfi1_filedata *fd, u32 tidinfo,
|
2016-02-06 00:57:52 +08:00
|
|
|
struct tid_group **grp)
|
|
|
|
{
|
|
|
|
struct hfi1_ctxtdata *uctxt = fd->uctxt;
|
|
|
|
struct hfi1_devdata *dd = uctxt->dd;
|
2016-03-09 03:14:20 +08:00
|
|
|
struct tid_rb_node *node;
|
2016-02-06 00:57:52 +08:00
|
|
|
u8 tidctrl = EXP_TID_GET(tidinfo, CTRL);
|
2016-02-04 06:34:41 +08:00
|
|
|
u32 tididx = EXP_TID_GET(tidinfo, IDX) << 1, rcventry;
|
2016-02-06 00:57:52 +08:00
|
|
|
|
|
|
|
if (tididx >= uctxt->expected_count) {
|
|
|
|
dd_dev_err(dd, "Invalid RcvArray entry (%u) index for ctxt %u\n",
|
|
|
|
tididx, uctxt->ctxt);
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (tidctrl == 0x3)
|
|
|
|
return -EINVAL;
|
|
|
|
|
2016-02-04 06:34:41 +08:00
|
|
|
rcventry = tididx + (tidctrl - 1);
|
2016-02-06 00:57:52 +08:00
|
|
|
|
2016-02-04 06:34:41 +08:00
|
|
|
node = fd->entry_to_rb[rcventry];
|
2016-03-09 03:14:20 +08:00
|
|
|
if (!node || node->rcventry != (uctxt->expected_base + rcventry))
|
2016-02-06 00:57:52 +08:00
|
|
|
return -EBADF;
|
2016-07-29 03:21:26 +08:00
|
|
|
|
|
|
|
if (grp)
|
|
|
|
*grp = node->grp;
|
|
|
|
|
2016-07-29 03:21:21 +08:00
|
|
|
if (!fd->handler)
|
2016-07-29 03:21:26 +08:00
|
|
|
cacheless_tid_rb_remove(fd, node);
|
2016-03-09 03:14:42 +08:00
|
|
|
else
|
2016-07-29 03:21:20 +08:00
|
|
|
hfi1_mmu_rb_remove(fd->handler, &node->mmu);
|
2016-03-09 03:14:20 +08:00
|
|
|
|
2016-02-06 00:57:52 +08:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2016-07-29 00:27:32 +08:00
|
|
|
static void clear_tid_node(struct hfi1_filedata *fd, struct tid_rb_node *node)
|
2016-02-06 00:57:52 +08:00
|
|
|
{
|
|
|
|
struct hfi1_ctxtdata *uctxt = fd->uctxt;
|
|
|
|
struct hfi1_devdata *dd = uctxt->dd;
|
|
|
|
|
2016-02-06 00:57:58 +08:00
|
|
|
trace_hfi1_exp_tid_unreg(uctxt->ctxt, fd->subctxt, node->rcventry,
|
2016-03-09 03:14:20 +08:00
|
|
|
node->npages, node->mmu.addr, node->phys,
|
2016-02-06 00:57:58 +08:00
|
|
|
node->dma_addr);
|
|
|
|
|
2016-02-06 00:57:52 +08:00
|
|
|
/*
|
|
|
|
* Make sure device has seen the write before we unpin the
|
|
|
|
* pages.
|
|
|
|
*/
|
IB/hfi1: Fix bar0 mapping to use write combining
When the debugpat kernel boot flag is turned on the following
traces are printed:
[ 1884.793168] x86/PAT: Overlap at 0x90000000-0x92000000
[ 1884.803510] x86/PAT: reserve_memtype added [mem 0x91200000-0x9127ffff],
track uncached-minus, req write-combining, ret uncached-minus
[ 1884.818167] hfi1 0000:05:00.0: hfi1_0: WC Remapped RcvArray:
ffffc9000a980000
The ioremap_wc() clearly is not returning a write combining mapping due
to an overlap where the RcvArray is mapped in a uncached mapping prior
to creating the proposed write combining mapping.
The patch replaces the single base register for uncached CSRs that
used to overlap the RcvArray with two mappings. One, kregbase1, from the
bar0 up to the RcvArray and another, kregbase2, from the end of the
RcvArray to the pio send buffer space. A new dd field, base2_start,
is used to convert the zero-based offset in the CSR routines to the
correct kregbase1/kregbase2 mapping. A single direct write of the
RcvArray CSRs is replaced with hfi1_put_tid() to insure correct access
using the new disjoint mapping.
Additionally, the kregend field is deleted since it is only ever written.
patdebug now shows the RcvArray as write combining:
[ 35.688990] x86/PAT: reserve_memtype added [mem 0x91200000-0x9127ffff],
track write-combining, req write-combining, ret write-combining
To insulate from any potential issues with write combining, all
writeq are now flushed in hfi1_put_tid() and rcv_array_wc_fill().
Reviewed-by: Mitko Haralanov <mitko.haralanov@intel.com>
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
2017-07-24 22:45:31 +08:00
|
|
|
hfi1_put_tid(dd, node->rcventry, PT_INVALID_FLUSH, 0, 0);
|
2016-02-06 00:57:52 +08:00
|
|
|
|
2017-08-22 09:26:51 +08:00
|
|
|
unpin_rcv_pages(fd, NULL, node, 0, node->npages, true);
|
2016-02-06 00:57:52 +08:00
|
|
|
|
|
|
|
node->grp->used--;
|
|
|
|
node->grp->map &= ~(1 << (node->rcventry - node->grp->base));
|
|
|
|
|
|
|
|
if (node->grp->used == node->grp->size - 1)
|
|
|
|
tid_group_move(node->grp, &uctxt->tid_full_list,
|
|
|
|
&uctxt->tid_used_list);
|
|
|
|
else if (!node->grp->used)
|
|
|
|
tid_group_move(node->grp, &uctxt->tid_used_list,
|
|
|
|
&uctxt->tid_group_list);
|
|
|
|
kfree(node);
|
|
|
|
}
|
|
|
|
|
2016-07-29 03:21:26 +08:00
|
|
|
/*
|
|
|
|
* As a simple helper for hfi1_user_exp_rcv_free, this function deals with
|
|
|
|
* clearing nodes in the non-cached case.
|
|
|
|
*/
|
2016-02-06 00:57:52 +08:00
|
|
|
static void unlock_exp_tids(struct hfi1_ctxtdata *uctxt,
|
2016-07-29 03:21:20 +08:00
|
|
|
struct exp_tid_set *set,
|
|
|
|
struct hfi1_filedata *fd)
|
2016-02-06 00:57:52 +08:00
|
|
|
{
|
|
|
|
struct tid_group *grp, *ptr;
|
|
|
|
int i;
|
|
|
|
|
|
|
|
list_for_each_entry_safe(grp, ptr, &set->list, list) {
|
|
|
|
list_del_init(&grp->list);
|
|
|
|
|
|
|
|
for (i = 0; i < grp->size; i++) {
|
|
|
|
if (grp->map & (1 << i)) {
|
|
|
|
u16 rcventry = grp->base + i;
|
2016-03-09 03:14:20 +08:00
|
|
|
struct tid_rb_node *node;
|
2016-02-06 00:57:52 +08:00
|
|
|
|
2016-02-04 06:34:41 +08:00
|
|
|
node = fd->entry_to_rb[rcventry -
|
|
|
|
uctxt->expected_base];
|
|
|
|
if (!node || node->rcventry != rcventry)
|
2016-02-06 00:57:52 +08:00
|
|
|
continue;
|
2016-07-29 03:21:26 +08:00
|
|
|
|
|
|
|
cacheless_tid_rb_remove(fd, node);
|
2016-02-06 00:57:52 +08:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2016-07-29 03:21:26 +08:00
|
|
|
/*
|
|
|
|
* Always return 0 from this function. A non-zero return indicates that the
|
|
|
|
* remove operation will be called and that memory should be unpinned.
|
|
|
|
* However, the driver cannot unpin out from under PSM. Instead, retain the
|
|
|
|
* memory (by returning 0) and inform PSM that the memory is going away. PSM
|
|
|
|
* will call back later when it has removed the memory from its list.
|
|
|
|
*/
|
2016-07-29 03:21:20 +08:00
|
|
|
static int tid_rb_invalidate(void *arg, struct mmu_rb_node *mnode)
|
2016-02-06 00:57:46 +08:00
|
|
|
{
|
2016-07-29 03:21:20 +08:00
|
|
|
struct hfi1_filedata *fdata = arg;
|
2016-03-09 03:14:20 +08:00
|
|
|
struct hfi1_ctxtdata *uctxt = fdata->uctxt;
|
|
|
|
struct tid_rb_node *node =
|
|
|
|
container_of(mnode, struct tid_rb_node, mmu);
|
2016-02-06 00:57:46 +08:00
|
|
|
|
2016-03-09 03:14:20 +08:00
|
|
|
if (node->freed)
|
|
|
|
return 0;
|
2016-02-06 00:57:55 +08:00
|
|
|
|
2016-03-09 03:14:20 +08:00
|
|
|
trace_hfi1_exp_tid_inval(uctxt->ctxt, fdata->subctxt, node->mmu.addr,
|
|
|
|
node->rcventry, node->npages, node->dma_addr);
|
|
|
|
node->freed = true;
|
2016-02-06 00:57:58 +08:00
|
|
|
|
2016-03-09 03:14:20 +08:00
|
|
|
spin_lock(&fdata->invalid_lock);
|
|
|
|
if (fdata->invalid_tid_idx < uctxt->expected_count) {
|
|
|
|
fdata->invalid_tids[fdata->invalid_tid_idx] =
|
|
|
|
rcventry2tidinfo(node->rcventry - uctxt->expected_base);
|
|
|
|
fdata->invalid_tids[fdata->invalid_tid_idx] |=
|
|
|
|
EXP_TID_SET(LEN, node->npages);
|
|
|
|
if (!fdata->invalid_tid_idx) {
|
|
|
|
unsigned long *ev;
|
2016-02-06 00:57:55 +08:00
|
|
|
|
|
|
|
/*
|
2016-03-09 03:14:20 +08:00
|
|
|
* hfi1_set_uevent_bits() sets a user event flag
|
|
|
|
* for all processes. Because calling into the
|
|
|
|
* driver to process TID cache invalidations is
|
|
|
|
* expensive and TID cache invalidations are
|
|
|
|
* handled on a per-process basis, we can
|
|
|
|
* optimize this to set the flag only for the
|
|
|
|
* process in question.
|
2016-02-06 00:57:55 +08:00
|
|
|
*/
|
2016-03-09 03:14:20 +08:00
|
|
|
ev = uctxt->dd->events +
|
2017-09-26 22:00:56 +08:00
|
|
|
(uctxt_offset(uctxt) + fdata->subctxt);
|
2016-03-09 03:14:20 +08:00
|
|
|
set_bit(_HFI1_EVENT_TID_MMU_NOTIFY_BIT, ev);
|
2016-02-06 00:57:55 +08:00
|
|
|
}
|
2016-03-09 03:14:20 +08:00
|
|
|
fdata->invalid_tid_idx++;
|
2016-02-06 00:57:55 +08:00
|
|
|
}
|
2016-03-09 03:14:20 +08:00
|
|
|
spin_unlock(&fdata->invalid_lock);
|
|
|
|
return 0;
|
2016-02-06 00:57:46 +08:00
|
|
|
}
|
|
|
|
|
2016-07-29 03:21:20 +08:00
|
|
|
static int tid_rb_insert(void *arg, struct mmu_rb_node *node)
|
2016-02-06 00:57:46 +08:00
|
|
|
{
|
2016-07-29 03:21:20 +08:00
|
|
|
struct hfi1_filedata *fdata = arg;
|
2016-03-09 03:14:20 +08:00
|
|
|
struct tid_rb_node *tnode =
|
|
|
|
container_of(node, struct tid_rb_node, mmu);
|
2016-02-04 06:34:41 +08:00
|
|
|
u32 base = fdata->uctxt->expected_base;
|
2016-02-06 00:57:46 +08:00
|
|
|
|
2016-03-09 03:14:20 +08:00
|
|
|
fdata->entry_to_rb[tnode->rcventry - base] = tnode;
|
2016-02-06 00:57:46 +08:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2016-07-29 03:21:26 +08:00
|
|
|
static void cacheless_tid_rb_remove(struct hfi1_filedata *fdata,
|
|
|
|
struct tid_rb_node *tnode)
|
|
|
|
{
|
|
|
|
u32 base = fdata->uctxt->expected_base;
|
|
|
|
|
|
|
|
fdata->entry_to_rb[tnode->rcventry - base] = NULL;
|
|
|
|
clear_tid_node(fdata, tnode);
|
|
|
|
}
|
|
|
|
|
2016-07-29 03:21:25 +08:00
|
|
|
static void tid_rb_remove(void *arg, struct mmu_rb_node *node)
|
2016-02-04 06:34:41 +08:00
|
|
|
{
|
2016-07-29 03:21:20 +08:00
|
|
|
struct hfi1_filedata *fdata = arg;
|
2016-03-09 03:14:20 +08:00
|
|
|
struct tid_rb_node *tnode =
|
|
|
|
container_of(node, struct tid_rb_node, mmu);
|
2016-02-04 06:34:41 +08:00
|
|
|
|
2016-07-29 03:21:26 +08:00
|
|
|
cacheless_tid_rb_remove(fdata, tnode);
|
2016-02-04 06:34:41 +08:00
|
|
|
}
|