| Message ID | 20221201005931.3877608-2-jiaqiyan@google.com (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | Memory poison recovery in khugepaged collapsing | expand |
On Wed, 30 Nov 2022 16:59:30 -0800 Jiaqi Yan <jiaqiyan@google.com> wrote: > Make __collapse_huge_page_copy return whether copying anonymous pages > succeeded, and make collapse_huge_page handle the return status. > > Break existing PTE scan loop into two for-loops. The first loop copies > source pages into target huge page, and can fail gracefully when running > into memory errors in source pages. If copying all pages succeeds, the > second loop releases and clears up these normal pages. Otherwise, the > second loop rolls back the page table and page states by: > - re-establishing the original PTEs-to-PMD connection. > - releasing source pages back to their LRU list. > > Tested manually: > 0. Enable khugepaged on system under test. > 1. Start a two-thread application. Each thread allocates a chunk of > non-huge anonymous memory buffer. > 2. Pick 4 random buffer locations (2 in each thread) and inject > uncorrectable memory errors at corresponding physical addresses. > 3. Signal both threads to make their memory buffer collapsible, i.e. > calling madvise(MADV_HUGEPAGE). > 4. Wait and check kernel log: khugepaged is able to recover from poisoned > pages and skips collapsing them. > 5. Signal both threads to inspect their buffer contents and make sure no > data corruption. Looks like a nice patchset. I'd like to give it a run in linux-next but we're at -rc7 and we have no review/ack tags. So it should be a post-6.2-rc1 thing. I have a quibble. > --- a/include/linux/highmem.h > +++ b/include/linux/highmem.h > @@ -361,6 +361,27 @@ static inline void copy_highpage(struct page *to, struct page *from) > > #endif > > +/* > + * Machine check exception handled version of copy_highpage. Return number > + * of bytes not copied if there was an exception; otherwise 0 for success. > + * Note handling #MC requires arch opt-in. > + */ > +static inline int copy_mc_highpage(struct page *to, struct page *from) > +{ > + char *vfrom, *vto; > + unsigned long ret; > + > + vfrom = kmap_local_page(from); > + vto = kmap_local_page(to); > + ret = copy_mc_to_kernel(vto, vfrom, PAGE_SIZE); > + if (ret == 0) > + kmsan_copy_page_meta(to, from); > + kunmap_local(vto); > + kunmap_local(vfrom); > + > + return ret; > +} Why inlined? It's large, it's slow, it's called only from khugepaged.c. A regular out-of-line function which is static to khugepaged.c seems more appropriate?
On 2022/12/2 7:09, Andrew Morton wrote: > On Wed, 30 Nov 2022 16:59:30 -0800 Jiaqi Yan <jiaqiyan@google.com> wrote: > >> Make __collapse_huge_page_copy return whether copying anonymous pages >> succeeded, and make collapse_huge_page handle the return status. >> >> Break existing PTE scan loop into two for-loops. The first loop copies >> source pages into target huge page, and can fail gracefully when running >> into memory errors in source pages. If copying all pages succeeds, the >> second loop releases and clears up these normal pages. Otherwise, the >> second loop rolls back the page table and page states by: >> - re-establishing the original PTEs-to-PMD connection. >> - releasing source pages back to their LRU list. >> >> Tested manually: >> 0. Enable khugepaged on system under test. >> 1. Start a two-thread application. Each thread allocates a chunk of >> non-huge anonymous memory buffer. >> 2. Pick 4 random buffer locations (2 in each thread) and inject >> uncorrectable memory errors at corresponding physical addresses. >> 3. Signal both threads to make their memory buffer collapsible, i.e. >> calling madvise(MADV_HUGEPAGE). >> 4. Wait and check kernel log: khugepaged is able to recover from poisoned >> pages and skips collapsing them. >> 5. Signal both threads to inspect their buffer contents and make sure no >> data corruption. > Looks like a nice patchset. I'd like to give it a run in linux-next > but we're at -rc7 and we have no review/ack tags. So it should be a > post-6.2-rc1 thing. > > I have a quibble. > >> --- a/include/linux/highmem.h >> +++ b/include/linux/highmem.h >> @@ -361,6 +361,27 @@ static inline void copy_highpage(struct page *to, struct page *from) >> >> #endif >> >> +/* >> + * Machine check exception handled version of copy_highpage. Return number >> + * of bytes not copied if there was an exception; otherwise 0 for success. >> + * Note handling #MC requires arch opt-in. >> + */ >> +static inline int copy_mc_highpage(struct page *to, struct page *from) >> +{ >> + char *vfrom, *vto; >> + unsigned long ret; >> + >> + vfrom = kmap_local_page(from); >> + vto = kmap_local_page(to); >> + ret = copy_mc_to_kernel(vto, vfrom, PAGE_SIZE); >> + if (ret == 0) >> + kmsan_copy_page_meta(to, from); >> + kunmap_local(vto); >> + kunmap_local(vfrom); >> + >> + return ret; >> +} > Why inlined? It's large, it's slow, it's called only from > khugepaged.c. A regular out-of-line function which is static to > khugepaged.c seems more appropriate? There is a similar function copy_mc_user_highpage(), could we reuse it , see commit a873dfe1032a mm, hwpoison: try to recover from copy-on write faults > > .
On Thu, Dec 1, 2022 at 6:25 PM Kefeng Wang <wangkefeng.wang@huawei.com> wrote: > > > On 2022/12/2 7:09, Andrew Morton wrote: > > On Wed, 30 Nov 2022 16:59:30 -0800 Jiaqi Yan <jiaqiyan@google.com> wrote: > > > >> Make __collapse_huge_page_copy return whether copying anonymous pages > >> succeeded, and make collapse_huge_page handle the return status. > >> > >> Break existing PTE scan loop into two for-loops. The first loop copies > >> source pages into target huge page, and can fail gracefully when running > >> into memory errors in source pages. If copying all pages succeeds, the > >> second loop releases and clears up these normal pages. Otherwise, the > >> second loop rolls back the page table and page states by: > >> - re-establishing the original PTEs-to-PMD connection. > >> - releasing source pages back to their LRU list. > >> > >> Tested manually: > >> 0. Enable khugepaged on system under test. > >> 1. Start a two-thread application. Each thread allocates a chunk of > >> non-huge anonymous memory buffer. > >> 2. Pick 4 random buffer locations (2 in each thread) and inject > >> uncorrectable memory errors at corresponding physical addresses. > >> 3. Signal both threads to make their memory buffer collapsible, i.e. > >> calling madvise(MADV_HUGEPAGE). > >> 4. Wait and check kernel log: khugepaged is able to recover from poisoned > >> pages and skips collapsing them. > >> 5. Signal both threads to inspect their buffer contents and make sure no > >> data corruption. > > Looks like a nice patchset. I'd like to give it a run in linux-next > > but we're at -rc7 and we have no review/ack tags. So it should be a > > post-6.2-rc1 thing. > > > > I have a quibble. > > > >> --- a/include/linux/highmem.h > >> +++ b/include/linux/highmem.h > >> @@ -361,6 +361,27 @@ static inline void copy_highpage(struct page *to, struct page *from) > >> > >> #endif > >> > >> +/* > >> + * Machine check exception handled version of copy_highpage. Return number > >> + * of bytes not copied if there was an exception; otherwise 0 for success. > >> + * Note handling #MC requires arch opt-in. > >> + */ > >> +static inline int copy_mc_highpage(struct page *to, struct page *from) > >> +{ > >> + char *vfrom, *vto; > >> + unsigned long ret; > >> + > >> + vfrom = kmap_local_page(from); > >> + vto = kmap_local_page(to); > >> + ret = copy_mc_to_kernel(vto, vfrom, PAGE_SIZE); > >> + if (ret == 0) > >> + kmsan_copy_page_meta(to, from); > >> + kunmap_local(vto); > >> + kunmap_local(vfrom); > >> + > >> + return ret; > >> +} > > Why inlined? It's large, it's slow, it's called only from > > khugepaged.c. A regular out-of-line function which is static to > > khugepaged.c seems more appropriate? > > There is a similar function copy_mc_user_highpage(), could we reuse > it , see commit a873dfe1032a mm, hwpoison: try to recover from copy-on > write faults > > To Kefeng: As I explained in v7, besides `to` and `from` pages, copy_mc_user_highpage takes `struct vm_area_struct *vma` and `vaddr`. While it fits __collapse_huge_page_copy, it doesn't really fit well for collapse_file (needed for the 2nd commit). When Shi Yang reviewed my patches, we agreed that we should borrow this opportunity to unify the copying routines in khugepaged.c (for both file-backed and anon memory), and copy_highpage fits both (alternatively we can use copy_user_highpage and passing vaddr=null and vma=null, but I don't like that). So I choose to make copy_highpage to be MC recoverable. Does this make sense to you? To Andrew: I think it is a reasonable "quibble". I will prepare the update in v9 while waiting for more reviews on v8 if there is.
On 2022/12/3 1:29, Jiaqi Yan wrote: > On Thu, Dec 1, 2022 at 6:25 PM Kefeng Wang <wangkefeng.wang@huawei.com> wrote: >> >> On 2022/12/2 7:09, Andrew Morton wrote: >>> On Wed, 30 Nov 2022 16:59:30 -0800 Jiaqi Yan <jiaqiyan@google.com> wrote: >>> >>>> Make __collapse_huge_page_copy return whether copying anonymous pages >>>> succeeded, and make collapse_huge_page handle the return status. >>>> >>>> Break existing PTE scan loop into two for-loops. The first loop copies >>>> source pages into target huge page, and can fail gracefully when running >>>> into memory errors in source pages. If copying all pages succeeds, the >>>> second loop releases and clears up these normal pages. Otherwise, the >>>> second loop rolls back the page table and page states by: >>>> - re-establishing the original PTEs-to-PMD connection. >>>> - releasing source pages back to their LRU list. >>>> >>>> Tested manually: >>>> 0. Enable khugepaged on system under test. >>>> 1. Start a two-thread application. Each thread allocates a chunk of >>>> non-huge anonymous memory buffer. >>>> 2. Pick 4 random buffer locations (2 in each thread) and inject >>>> uncorrectable memory errors at corresponding physical addresses. >>>> 3. Signal both threads to make their memory buffer collapsible, i.e. >>>> calling madvise(MADV_HUGEPAGE). >>>> 4. Wait and check kernel log: khugepaged is able to recover from poisoned >>>> pages and skips collapsing them. >>>> 5. Signal both threads to inspect their buffer contents and make sure no >>>> data corruption. >>> Looks like a nice patchset. I'd like to give it a run in linux-next >>> but we're at -rc7 and we have no review/ack tags. So it should be a >>> post-6.2-rc1 thing. >>> >>> I have a quibble. >>> >>>> --- a/include/linux/highmem.h >>>> +++ b/include/linux/highmem.h >>>> @@ -361,6 +361,27 @@ static inline void copy_highpage(struct page *to, struct page *from) >>>> >>>> #endif >>>> >>>> +/* >>>> + * Machine check exception handled version of copy_highpage. Return number >>>> + * of bytes not copied if there was an exception; otherwise 0 for success. >>>> + * Note handling #MC requires arch opt-in. >>>> + */ >>>> +static inline int copy_mc_highpage(struct page *to, struct page *from) >>>> +{ >>>> + char *vfrom, *vto; >>>> + unsigned long ret; >>>> + >>>> + vfrom = kmap_local_page(from); >>>> + vto = kmap_local_page(to); >>>> + ret = copy_mc_to_kernel(vto, vfrom, PAGE_SIZE); >>>> + if (ret == 0) >>>> + kmsan_copy_page_meta(to, from); >>>> + kunmap_local(vto); >>>> + kunmap_local(vfrom); >>>> + >>>> + return ret; >>>> +} >>> Why inlined? It's large, it's slow, it's called only from >>> khugepaged.c. A regular out-of-line function which is static to >>> khugepaged.c seems more appropriate? >> There is a similar function copy_mc_user_highpage(), could we reuse >> it , see commit a873dfe1032a mm, hwpoison: try to recover from copy-on >> write faults >> >> > To Kefeng: As I explained in v7, besides `to` and `from` pages, > copy_mc_user_highpage takes `struct vm_area_struct *vma` and `vaddr`. > While it fits __collapse_huge_page_copy, it doesn't really fit well > for collapse_file (needed for the 2nd commit). When Shi Yang reviewed > my patches, we agreed that we should borrow this opportunity to unify > the copying routines in khugepaged.c (for both file-backed and anon > memory), and copy_highpage fits both (alternatively we can use > copy_user_highpage and passing vaddr=null and vma=null, but I don't > like that). So I choose to make copy_highpage to be MC recoverable. > Does this make sense to you? Yes，thanks for your explanation. > > To Andrew: I think it is a reasonable "quibble". I will prepare the > update in v9 while waiting for more reviews on v8 if there is. > .
diff --git a/include/linux/highmem.h b/include/linux/highmem.h index 44242268f53bd..f45fc5ceff524 100644 --- a/include/linux/highmem.h +++ b/include/linux/highmem.h @@ -361,6 +361,27 @@ static inline void copy_highpage(struct page *to, struct page *from) #endif +/* + * Machine check exception handled version of copy_highpage. Return number + * of bytes not copied if there was an exception; otherwise 0 for success. + * Note handling #MC requires arch opt-in. + */ +static inline int copy_mc_highpage(struct page *to, struct page *from) +{ + char *vfrom, *vto; + unsigned long ret; + + vfrom = kmap_local_page(from); + vto = kmap_local_page(to); + ret = copy_mc_to_kernel(vto, vfrom, PAGE_SIZE); + if (ret == 0) + kmsan_copy_page_meta(to, from); + kunmap_local(vto); + kunmap_local(vfrom); + + return ret; +} + static inline void memcpy_page(struct page *dst_page, size_t dst_off, struct page *src_page, size_t src_off, size_t len) diff --git a/include/trace/events/huge_memory.h b/include/trace/events/huge_memory.h index 35d759d3b0104..5743ae970af31 100644 --- a/include/trace/events/huge_memory.h +++ b/include/trace/events/huge_memory.h @@ -36,7 +36,8 @@ EM( SCAN_ALLOC_HUGE_PAGE_FAIL, "alloc_huge_page_failed") \ EM( SCAN_CGROUP_CHARGE_FAIL, "ccgroup_charge_failed") \ EM( SCAN_TRUNCATED, "truncated") \ - EMe(SCAN_PAGE_HAS_PRIVATE, "page_has_private") \ + EM( SCAN_PAGE_HAS_PRIVATE, "page_has_private") \ + EMe(SCAN_COPY_MC, "copy_poisoned_page") \ #undef EM #undef EMe diff --git a/mm/khugepaged.c b/mm/khugepaged.c index 78ec2771cc652..d9ff99980daea 100644 --- a/mm/khugepaged.c +++ b/mm/khugepaged.c @@ -55,6 +55,7 @@ enum scan_result { SCAN_CGROUP_CHARGE_FAIL, SCAN_TRUNCATED, SCAN_PAGE_HAS_PRIVATE, + SCAN_COPY_MC, }; #define CREATE_TRACE_POINTS @@ -670,56 +671,124 @@ static int __collapse_huge_page_isolate(struct vm_area_struct *vma, return result; } -static void __collapse_huge_page_copy(pte_t *pte, struct page *page, - struct vm_area_struct *vma, - unsigned long address, - spinlock_t *ptl, - struct list_head *compound_pagelist) +/* + * __collapse_huge_page_copy - attempts to copy memory contents from normal + * pages to a hugepage. Cleans up the normal pages if copying succeeds; + * otherwise restores the original page table and releases isolated normal pages. + * Returns SCAN_SUCCEED if copying succeeds, otherwise returns SCAN_COPY_MC. + * + * @pte: starting of the PTEs to copy from + * @page: the new hugepage to copy contents to + * @pmd: pointer to the new hugepage's PMD + * @rollback: the original normal pages' PMD + * @vma: the original normal pages' virtual memory area + * @address: starting address to copy + * @pte_ptl: lock on normal pages' PTEs + * @compound_pagelist: list that stores compound pages + */ +static int __collapse_huge_page_copy(pte_t *pte, + struct page *page, + pmd_t *pmd, + pmd_t rollback, + struct vm_area_struct *vma, + unsigned long address, + spinlock_t *pte_ptl, + struct list_head *compound_pagelist) { struct page *src_page, *tmp; pte_t *_pte; - for (_pte = pte; _pte < pte + HPAGE_PMD_NR; - _pte++, page++, address += PAGE_SIZE) { - pte_t pteval = *_pte; + pte_t pteval; + unsigned long _address; + spinlock_t *pmd_ptl; + int result = SCAN_SUCCEED; - if (pte_none(pteval) || is_zero_pfn(pte_pfn(pteval))) { - clear_user_highpage(page, address); - add_mm_counter(vma->vm_mm, MM_ANONPAGES, 1); - if (is_zero_pfn(pte_pfn(pteval))) { + /* + * Copying pages' contents is subject to memory poison at any iteration. + */ + for (_pte = pte, _address = address; _pte < pte + HPAGE_PMD_NR; + _pte++, page++, _address += PAGE_SIZE) { + pteval = *_pte; + + if (pte_none(pteval) || is_zero_pfn(pte_pfn(pteval))) + clear_user_highpage(page, _address); + else { + src_page = pte_page(pteval); + if (copy_mc_highpage(page, src_page) > 0) { + result = SCAN_COPY_MC; + break; + } + } + } + + if (likely(result == SCAN_SUCCEED)) { + for (_pte = pte, _address = address; _pte < pte + HPAGE_PMD_NR; + _pte++, _address += PAGE_SIZE) { + pteval = *_pte; + if (pte_none(pteval) || is_zero_pfn(pte_pfn(pteval))) { + add_mm_counter(vma->vm_mm, MM_ANONPAGES, 1); + if (is_zero_pfn(pte_pfn(pteval))) { + /* + * pte_ptl mostly unnecessary. + */ + spin_lock(pte_ptl); + pte_clear(vma->vm_mm, _address, _pte); + spin_unlock(pte_ptl); + } + } else { + src_page = pte_page(pteval); + if (!PageCompound(src_page)) + release_pte_page(src_page); /* - * ptl mostly unnecessary. + * pte_ptl mostly unnecessary, but preempt has + * to be disabled to update the per-cpu stats + * inside page_remove_rmap(). */ - spin_lock(ptl); - ptep_clear(vma->vm_mm, address, _pte); - spin_unlock(ptl); + spin_lock(pte_ptl); + ptep_clear(vma->vm_mm, _address, _pte); + page_remove_rmap(src_page, vma, false); + spin_unlock(pte_ptl); + free_page_and_swap_cache(src_page); } - } else { - src_page = pte_page(pteval); - copy_user_highpage(page, src_page, address, vma); - if (!PageCompound(src_page)) - release_pte_page(src_page); - /* - * ptl mostly unnecessary, but preempt has to - * be disabled to update the per-cpu stats - * inside page_remove_rmap(). - */ - spin_lock(ptl); - ptep_clear(vma->vm_mm, address, _pte); - page_remove_rmap(src_page, vma, false); - spin_unlock(ptl); - free_page_and_swap_cache(src_page); + } + list_for_each_entry_safe(src_page, tmp, compound_pagelist, lru) { + list_del(&src_page->lru); + mod_node_page_state(page_pgdat(src_page), + NR_ISOLATED_ANON + page_is_file_lru(src_page), + -compound_nr(src_page)); + unlock_page(src_page); + free_swap_cache(src_page); + putback_lru_page(src_page); + } + } else { + /* + * Re-establish the regular PMD that points to the regular + * page table. Restoring PMD needs to be done prior to + * releasing pages. Since pages are still isolated and + * locked here, acquiring anon_vma_lock_write is unnecessary. + */ + pmd_ptl = pmd_lock(vma->vm_mm, pmd); + pmd_populate(vma->vm_mm, pmd, pmd_pgtable(rollback)); + spin_unlock(pmd_ptl); + /* + * Release both raw and compound pages isolated + * in __collapse_huge_page_isolate. + */ + for (_pte = pte, _address = address; _pte < pte + HPAGE_PMD_NR; + _pte++, _address += PAGE_SIZE) { + pteval = *_pte; + if (!pte_none(pteval) && !is_zero_pfn(pte_pfn(pteval))) { + src_page = pte_page(pteval); + if (!PageCompound(src_page)) + release_pte_page(src_page); + } + } + list_for_each_entry_safe(src_page, tmp, compound_pagelist, lru) { + list_del(&src_page->lru); + release_pte_page(src_page); } } - list_for_each_entry_safe(src_page, tmp, compound_pagelist, lru) { - list_del(&src_page->lru); - mod_node_page_state(page_pgdat(src_page), - NR_ISOLATED_ANON + page_is_file_lru(src_page), - -compound_nr(src_page)); - unlock_page(src_page); - free_swap_cache(src_page); - putback_lru_page(src_page); - } + return result; } static void khugepaged_alloc_sleep(void) @@ -1079,9 +1148,13 @@ static int collapse_huge_page(struct mm_struct *mm, unsigned long address, */ anon_vma_unlock_write(vma->anon_vma); - __collapse_huge_page_copy(pte, hpage, vma, address, pte_ptl, - &compound_pagelist); + result = __collapse_huge_page_copy(pte, hpage, pmd, _pmd, + vma, address, pte_ptl, + &compound_pagelist); pte_unmap(pte); + if (unlikely(result != SCAN_SUCCEED)) + goto out_up_write; + /* * spin_lock() below is not the equivalent of smp_wmb(), but * the smp_wmb() inside __SetPageUptodate() can be reused to
Make __collapse_huge_page_copy return whether copying anonymous pages succeeded, and make collapse_huge_page handle the return status. Break existing PTE scan loop into two for-loops. The first loop copies source pages into target huge page, and can fail gracefully when running into memory errors in source pages. If copying all pages succeeds, the second loop releases and clears up these normal pages. Otherwise, the second loop rolls back the page table and page states by: - re-establishing the original PTEs-to-PMD connection. - releasing source pages back to their LRU list. Tested manually: 0. Enable khugepaged on system under test. 1. Start a two-thread application. Each thread allocates a chunk of non-huge anonymous memory buffer. 2. Pick 4 random buffer locations (2 in each thread) and inject uncorrectable memory errors at corresponding physical addresses. 3. Signal both threads to make their memory buffer collapsible, i.e. calling madvise(MADV_HUGEPAGE). 4. Wait and check kernel log: khugepaged is able to recover from poisoned pages and skips collapsing them. 5. Signal both threads to inspect their buffer contents and make sure no data corruption. Signed-off-by: Jiaqi Yan <jiaqiyan@google.com> --- include/linux/highmem.h | 21 ++++ include/trace/events/huge_memory.h | 3 +- mm/khugepaged.c | 157 +++++++++++++++++++++-------- 3 files changed, 138 insertions(+), 43 deletions(-)