diff options
Diffstat (limited to 'Documentation/core-api/pin_user_pages.rst')
-rw-r--r-- | Documentation/core-api/pin_user_pages.rst | 143 |
1 files changed, 95 insertions, 48 deletions
diff --git a/Documentation/core-api/pin_user_pages.rst b/Documentation/core-api/pin_user_pages.rst index 1d490155ecd7..b18416f4500f 100644 --- a/Documentation/core-api/pin_user_pages.rst +++ b/Documentation/core-api/pin_user_pages.rst @@ -33,7 +33,7 @@ all combinations of get*(), pin*(), FOLL_LONGTERM, and more. Also, the pin_user_pages*() APIs are clearly distinct from the get_user_pages*() APIs, so that's a natural dividing line, and a good point to make separate wrapper calls. In other words, use pin_user_pages*() for DMA-pinned pages, and -get_user_pages*() for other cases. There are four cases described later on in +get_user_pages*() for other cases. There are five cases described later on in this document, to further clarify that concept. FOLL_PIN and FOLL_GET are mutually exclusive for a given gup call. However, @@ -52,8 +52,22 @@ Which flags are set by each wrapper For these pin_user_pages*() functions, FOLL_PIN is OR'd in with whatever gup flags the caller provides. The caller is required to pass in a non-null struct -pages* array, and the function then pin pages by incrementing each by a special -value. For now, that value is +1, just like get_user_pages*().:: +pages* array, and the function then pins pages by incrementing each by a special +value: GUP_PIN_COUNTING_BIAS. + +For compound pages, the GUP_PIN_COUNTING_BIAS scheme is not used. Instead, +an exact form of pin counting is achieved, by using the 2nd struct page +in the compound page. A new struct page field, compound_pincount, has +been added in order to support this. + +This approach for compound pages avoids the counting upper limit problems that +are discussed below. Those limitations would have been aggravated severely by +huge pages, because each tail page adds a refcount to the head page. And in +fact, testing revealed that, without a separate compound_pincount field, +page overflows were seen in some huge page stress tests. + +This also means that huge pages and compound pages do not suffer +from the false positives problem that is mentioned below.:: Function -------- @@ -99,27 +113,6 @@ pages: This also leads to limitations: there are only 31-10==21 bits available for a counter that increments 10 bits at a time. -TODO: for 1GB and larger huge pages, this is cutting it close. That's because -when pin_user_pages() follows such pages, it increments the head page by "1" -(where "1" used to mean "+1" for get_user_pages(), but now means "+1024" for -pin_user_pages()) for each tail page. So if you have a 1GB huge page: - -* There are 256K (18 bits) worth of 4 KB tail pages. -* There are 21 bits available to count up via GUP_PIN_COUNTING_BIAS (that is, - 10 bits at a time) -* There are 21 - 18 == 3 bits available to count. Except that there aren't, - because you need to allow for a few normal get_page() calls on the head page, - as well. Fortunately, the approach of using addition, rather than "hard" - bitfields, within page->_refcount, allows for sharing these bits gracefully. - But we're still looking at about 8 references. - -This, however, is a missing feature more than anything else, because it's easily -solved by addressing an obvious inefficiency in the original get_user_pages() -approach of retrieving pages: stop treating all the pages as if they were -PAGE_SIZE. Retrieve huge pages as huge pages. The callers need to be aware of -this, so some work is required. Once that's in place, this limitation mostly -disappears from view, because there will be ample refcounting range available. - * Callers must specifically request "dma-pinned tracking of pages". In other words, just calling get_user_pages() will not suffice; a new set of functions, pin_user_page() and related, must be used. @@ -155,26 +148,49 @@ NOTE: Some pages, such as DAX pages, cannot be pinned with longterm pins. That's because DAX pages do not have a separate page cache, and so "pinning" implies locking down file system blocks, which is not (yet) supported in that way. -CASE 3: Hardware with page faulting support -------------------------------------------- -Here, a well-written driver doesn't normally need to pin pages at all. However, -if the driver does choose to do so, it can register MMU notifiers for the range, -and will be called back upon invalidation. Either way (avoiding page pinning, or -using MMU notifiers to unpin upon request), there is proper synchronization with -both filesystem and mm (page_mkclean(), munmap(), etc). +CASE 3: MMU notifier registration, with or without page faulting hardware +------------------------------------------------------------------------- +Device drivers can pin pages via get_user_pages*(), and register for mmu +notifier callbacks for the memory range. Then, upon receiving a notifier +"invalidate range" callback , stop the device from using the range, and unpin +the pages. There may be other possible schemes, such as for example explicitly +synchronizing against pending IO, that accomplish approximately the same thing. -Therefore, neither flag needs to be set. +Or, if the hardware supports replayable page faults, then the device driver can +avoid pinning entirely (this is ideal), as follows: register for mmu notifier +callbacks as above, but instead of stopping the device and unpinning in the +callback, simply remove the range from the device's page tables. -In this case, ideally, neither get_user_pages() nor pin_user_pages() should be -called. Instead, the software should be written so that it does not pin pages. -This allows mm and filesystems to operate more efficiently and reliably. +Either way, as long as the driver unpins the pages upon mmu notifier callback, +then there is proper synchronization with both filesystem and mm +(page_mkclean(), munmap(), etc). Therefore, neither flag needs to be set. CASE 4: Pinning for struct page manipulation only ------------------------------------------------- -Here, normal GUP calls are sufficient, so neither flag needs to be set. - -page_dma_pinned(): the whole point of pinning -============================================= +If only struct page data (as opposed to the actual memory contents that a page +is tracking) is affected, then normal GUP calls are sufficient, and neither flag +needs to be set. + +CASE 5: Pinning in order to write to the data within the page +------------------------------------------------------------- +Even though neither DMA nor Direct IO is involved, just a simple case of "pin, +write to a page's data, unpin" can cause a problem. Case 5 may be considered a +superset of Case 1, plus Case 2, plus anything that invokes that pattern. In +other words, if the code is neither Case 1 nor Case 2, it may still require +FOLL_PIN, for patterns like this: + +Correct (uses FOLL_PIN calls): + pin_user_pages() + write to the data within the pages + unpin_user_pages() + +INCORRECT (uses FOLL_GET calls): + get_user_pages() + write to the data within the pages + put_page() + +page_maybe_dma_pinned(): the whole point of pinning +=================================================== The whole point of marking pages as "DMA-pinned" or "gup-pinned" is to be able to query, "is this page DMA-pinned?" That allows code such as page_mkclean() @@ -186,7 +202,7 @@ and debates (see the References at the end of this document). It's a TODO item here: fill in the details once that's worked out. Meanwhile, it's safe to say that having this available: :: - static inline bool page_dma_pinned(struct page *page) + static inline bool page_maybe_dma_pinned(struct page *page) ...is a prerequisite to solving the long-running gup+DMA problem. @@ -205,22 +221,52 @@ Unit testing ============ This file:: - tools/testing/selftests/vm/gup_benchmark.c + tools/testing/selftests/vm/gup_test.c has the following new calls to exercise the new pin*() wrapper functions: -* PIN_FAST_BENCHMARK (./gup_benchmark -a) -* PIN_BENCHMARK (./gup_benchmark -b) +* PIN_FAST_BENCHMARK (./gup_test -a) +* PIN_BASIC_TEST (./gup_test -b) You can monitor how many total dma-pinned pages have been acquired and released since the system was booted, via two new /proc/vmstat entries: :: - /proc/vmstat/nr_foll_pin_requested - /proc/vmstat/nr_foll_pin_requested + /proc/vmstat/nr_foll_pin_acquired + /proc/vmstat/nr_foll_pin_released + +Under normal conditions, these two values will be equal unless there are any +long-term [R]DMA pins in place, or during pin/unpin transitions. + +* nr_foll_pin_acquired: This is the number of logical pins that have been + acquired since the system was powered on. For huge pages, the head page is + pinned once for each page (head page and each tail page) within the huge page. + This follows the same sort of behavior that get_user_pages() uses for huge + pages: the head page is refcounted once for each tail or head page in the huge + page, when get_user_pages() is applied to a huge page. + +* nr_foll_pin_released: The number of logical pins that have been released since + the system was powered on. Note that pages are released (unpinned) on a + PAGE_SIZE granularity, even if the original pin was applied to a huge page. + Becaused of the pin count behavior described above in "nr_foll_pin_acquired", + the accounting balances out, so that after doing this:: + + pin_user_pages(huge_page); + for (each page in huge_page) + unpin_user_page(page); + +...the following is expected:: + + nr_foll_pin_released == nr_foll_pin_acquired + +(...unless it was already out of balance due to a long-term RDMA pin being in +place.) + +Other diagnostics +================= -Those are both going to show zero, unless CONFIG_DEBUG_VM is set. This is -because there is a noticeable performance drop in unpin_user_page(), when they -are activated. +dump_page() has been enhanced slightly, to handle these new counting +fields, and to better report on compound pages in general. Specifically, +for compound pages, the exact (compound_pincount) pincount is reported. References ========== @@ -228,5 +274,6 @@ References * `Some slow progress on get_user_pages() (Apr 2, 2019) <https://lwn.net/Articles/784574/>`_ * `DMA and get_user_pages() (LPC: Dec 12, 2018) <https://lwn.net/Articles/774411/>`_ * `The trouble with get_user_pages() (Apr 30, 2018) <https://lwn.net/Articles/753027/>`_ +* `LWN kernel index: get_user_pages() <https://lwn.net/Kernel/Index/#Memory_management-get_user_pages>`_ John Hubbard, October, 2019 |