1.. SPDX-License-Identifier: GPL-2.0
2
3====================================================
4pin_user_pages() and related calls
5====================================================
6
7.. contents:: :local:
8
9Overview
10========
11
12This document describes the following functions::
13
14 pin_user_pages()
15 pin_user_pages_fast()
16 pin_user_pages_remote()
17
18Basic description of FOLL_PIN
19=============================
20
21FOLL_PIN and FOLL_LONGTERM are flags that can be passed to the get_user_pages*()
22("gup") family of functions. FOLL_PIN has significant interactions and
23interdependencies with FOLL_LONGTERM, so both are covered here.
24
25FOLL_PIN is internal to gup, meaning that it should not appear at the gup call
26sites. This allows the associated wrapper functions  (pin_user_pages*() and
27others) to set the correct combination of these flags, and to check for problems
28as well.
29
30FOLL_LONGTERM, on the other hand, *is* allowed to be set at the gup call sites.
31This is in order to avoid creating a large number of wrapper functions to cover
32all combinations of get*(), pin*(), FOLL_LONGTERM, and more. Also, the
33pin_user_pages*() APIs are clearly distinct from the get_user_pages*() APIs, so
34that's a natural dividing line, and a good point to make separate wrapper calls.
35In other words, use pin_user_pages*() for DMA-pinned pages, and
36get_user_pages*() for other cases. There are five cases described later on in
37this document, to further clarify that concept.
38
39FOLL_PIN and FOLL_GET are mutually exclusive for a given gup call. However,
40multiple threads and call sites are free to pin the same struct pages, via both
41FOLL_PIN and FOLL_GET. It's just the call site that needs to choose one or the
42other, not the struct page(s).
43
44The FOLL_PIN implementation is nearly the same as FOLL_GET, except that FOLL_PIN
45uses a different reference counting technique.
46
47FOLL_PIN is a prerequisite to FOLL_LONGTERM. Another way of saying that is,
48FOLL_LONGTERM is a specific case, more restrictive case of FOLL_PIN.
49
50Which flags are set by each wrapper
51===================================
52
53For these pin_user_pages*() functions, FOLL_PIN is OR'd in with whatever gup
54flags the caller provides. The caller is required to pass in a non-null struct
55pages* array, and the function then pins pages by incrementing each by a special
56value: GUP_PIN_COUNTING_BIAS.
57
58For large folios, the GUP_PIN_COUNTING_BIAS scheme is not used. Instead,
59the extra space available in the struct folio is used to store the
60pincount directly.
61
62This approach for large folios avoids the counting upper limit problems
63that are discussed below. Those limitations would have been aggravated
64severely by huge pages, because each tail page adds a refcount to the
65head page. And in fact, testing revealed that, without a separate pincount
66field, refcount overflows were seen in some huge page stress tests.
67
68This also means that huge pages and large folios do not suffer
69from the false positives problem that is mentioned below.::
70
71 Function
72 --------
73 pin_user_pages          FOLL_PIN is always set internally by this function.
74 pin_user_pages_fast     FOLL_PIN is always set internally by this function.
75 pin_user_pages_remote   FOLL_PIN is always set internally by this function.
76
77For these get_user_pages*() functions, FOLL_GET might not even be specified.
78Behavior is a little more complex than above. If FOLL_GET was *not* specified,
79but the caller passed in a non-null struct pages* array, then the function
80sets FOLL_GET for you, and proceeds to pin pages by incrementing the refcount
81of each page by +1.::
82
83 Function
84 --------
85 get_user_pages           FOLL_GET is sometimes set internally by this function.
86 get_user_pages_fast      FOLL_GET is sometimes set internally by this function.
87 get_user_pages_remote    FOLL_GET is sometimes set internally by this function.
88
89Tracking dma-pinned pages
90=========================
91
92Some of the key design constraints, and solutions, for tracking dma-pinned
93pages:
94
95* An actual reference count, per struct page, is required. This is because
96  multiple processes may pin and unpin a page.
97
98* False positives (reporting that a page is dma-pinned, when in fact it is not)
99  are acceptable, but false negatives are not.
100
101* struct page may not be increased in size for this, and all fields are already
102  used.
103
104* Given the above, we can overload the page->_refcount field by using, sort of,
105  the upper bits in that field for a dma-pinned count. "Sort of", means that,
106  rather than dividing page->_refcount into bit fields, we simple add a medium-
107  large value (GUP_PIN_COUNTING_BIAS, initially chosen to be 1024: 10 bits) to
108  page->_refcount. This provides fuzzy behavior: if a page has get_page() called
109  on it 1024 times, then it will appear to have a single dma-pinned count.
110  And again, that's acceptable.
111
112This also leads to limitations: there are only 31-10==21 bits available for a
113counter that increments 10 bits at a time.
114
115* Callers must specifically request "dma-pinned tracking of pages". In other
116  words, just calling get_user_pages() will not suffice; a new set of functions,
117  pin_user_page() and related, must be used.
118
119FOLL_PIN, FOLL_GET, FOLL_LONGTERM: when to use which flags
120==========================================================
121
122Thanks to Jan Kara, Vlastimil Babka and several other -mm people, for describing
123these categories:
124
125CASE 1: Direct IO (DIO)
126-----------------------
127There are GUP references to pages that are serving
128as DIO buffers. These buffers are needed for a relatively short time (so they
129are not "long term"). No special synchronization with page_mkclean() or
130munmap() is provided. Therefore, flags to set at the call site are: ::
131
132    FOLL_PIN
133
134...but rather than setting FOLL_PIN directly, call sites should use one of
135the pin_user_pages*() routines that set FOLL_PIN.
136
137CASE 2: RDMA
138------------
139There are GUP references to pages that are serving as DMA
140buffers. These buffers are needed for a long time ("long term"). No special
141synchronization with page_mkclean() or munmap() is provided. Therefore, flags
142to set at the call site are: ::
143
144    FOLL_PIN | FOLL_LONGTERM
145
146NOTE: Some pages, such as DAX pages, cannot be pinned with longterm pins. That's
147because DAX pages do not have a separate page cache, and so "pinning" implies
148locking down file system blocks, which is not (yet) supported in that way.
149
150CASE 3: MMU notifier registration, with or without page faulting hardware
151-------------------------------------------------------------------------
152Device drivers can pin pages via get_user_pages*(), and register for mmu
153notifier callbacks for the memory range. Then, upon receiving a notifier
154"invalidate range" callback , stop the device from using the range, and unpin
155the pages. There may be other possible schemes, such as for example explicitly
156synchronizing against pending IO, that accomplish approximately the same thing.
157
158Or, if the hardware supports replayable page faults, then the device driver can
159avoid pinning entirely (this is ideal), as follows: register for mmu notifier
160callbacks as above, but instead of stopping the device and unpinning in the
161callback, simply remove the range from the device's page tables.
162
163Either way, as long as the driver unpins the pages upon mmu notifier callback,
164then there is proper synchronization with both filesystem and mm
165(page_mkclean(), munmap(), etc). Therefore, neither flag needs to be set.
166
167CASE 4: Pinning for struct page manipulation only
168-------------------------------------------------
169If only struct page data (as opposed to the actual memory contents that a page
170is tracking) is affected, then normal GUP calls are sufficient, and neither flag
171needs to be set.
172
173CASE 5: Pinning in order to write to the data within the page
174-------------------------------------------------------------
175Even though neither DMA nor Direct IO is involved, just a simple case of "pin,
176write to a page's data, unpin" can cause a problem. Case 5 may be considered a
177superset of Case 1, plus Case 2, plus anything that invokes that pattern. In
178other words, if the code is neither Case 1 nor Case 2, it may still require
179FOLL_PIN, for patterns like this:
180
181Correct (uses FOLL_PIN calls):
182    pin_user_pages()
183    write to the data within the pages
184    unpin_user_pages()
185
186INCORRECT (uses FOLL_GET calls):
187    get_user_pages()
188    write to the data within the pages
189    put_page()
190
191page_maybe_dma_pinned(): the whole point of pinning
192===================================================
193
194The whole point of marking pages as "DMA-pinned" or "gup-pinned" is to be able
195to query, "is this page DMA-pinned?" That allows code such as page_mkclean()
196(and file system writeback code in general) to make informed decisions about
197what to do when a page cannot be unmapped due to such pins.
198
199What to do in those cases is the subject of a years-long series of discussions
200and debates (see the References at the end of this document). It's a TODO item
201here: fill in the details once that's worked out. Meanwhile, it's safe to say
202that having this available: ::
203
204        static inline bool page_maybe_dma_pinned(struct page *page)
205
206...is a prerequisite to solving the long-running gup+DMA problem.
207
208Another way of thinking about FOLL_GET, FOLL_PIN, and FOLL_LONGTERM
209===================================================================
210
211Another way of thinking about these flags is as a progression of restrictions:
212FOLL_GET is for struct page manipulation, without affecting the data that the
213struct page refers to. FOLL_PIN is a *replacement* for FOLL_GET, and is for
214short term pins on pages whose data *will* get accessed. As such, FOLL_PIN is
215a "more severe" form of pinning. And finally, FOLL_LONGTERM is an even more
216restrictive case that has FOLL_PIN as a prerequisite: this is for pages that
217will be pinned longterm, and whose data will be accessed.
218
219Unit testing
220============
221This file::
222
223 tools/testing/selftests/mm/gup_test.c
224
225has the following new calls to exercise the new pin*() wrapper functions:
226
227* PIN_FAST_BENCHMARK (./gup_test -a)
228* PIN_BASIC_TEST (./gup_test -b)
229
230You can monitor how many total dma-pinned pages have been acquired and released
231since the system was booted, via two new /proc/vmstat entries: ::
232
233    /proc/vmstat/nr_foll_pin_acquired
234    /proc/vmstat/nr_foll_pin_released
235
236Under normal conditions, these two values will be equal unless there are any
237long-term [R]DMA pins in place, or during pin/unpin transitions.
238
239* nr_foll_pin_acquired: This is the number of logical pins that have been
240  acquired since the system was powered on. For huge pages, the head page is
241  pinned once for each page (head page and each tail page) within the huge page.
242  This follows the same sort of behavior that get_user_pages() uses for huge
243  pages: the head page is refcounted once for each tail or head page in the huge
244  page, when get_user_pages() is applied to a huge page.
245
246* nr_foll_pin_released: The number of logical pins that have been released since
247  the system was powered on. Note that pages are released (unpinned) on a
248  PAGE_SIZE granularity, even if the original pin was applied to a huge page.
249  Becaused of the pin count behavior described above in "nr_foll_pin_acquired",
250  the accounting balances out, so that after doing this::
251
252    pin_user_pages(huge_page);
253    for (each page in huge_page)
254        unpin_user_page(page);
255
256...the following is expected::
257
258    nr_foll_pin_released == nr_foll_pin_acquired
259
260(...unless it was already out of balance due to a long-term RDMA pin being in
261place.)
262
263Other diagnostics
264=================
265
266dump_page() has been enhanced slightly to handle these new counting
267fields, and to better report on large folios in general.  Specifically,
268for large folios, the exact pincount is reported.
269
270References
271==========
272
273* `Some slow progress on get_user_pages() (Apr 2, 2019) <https://lwn.net/Articles/784574/>`_
274* `DMA and get_user_pages() (LPC: Dec 12, 2018) <https://lwn.net/Articles/774411/>`_
275* `The trouble with get_user_pages() (Apr 30, 2018) <https://lwn.net/Articles/753027/>`_
276* `LWN kernel index: get_user_pages() <https://lwn.net/Kernel/Index/#Memory_management-get_user_pages>`_
277
278John Hubbard, October, 2019
279