From: "Liam R. Howlett" <Liam.Howlett@Oracle.com>
To: Randy Dunlap <rdunlap@infradead.org>
Cc: maple-tree@lists.infradead.org, linux-mm@kvack.org,
linux-kernel@vger.kernel.org, Andrew Morton <akpm@google.com>,
Song Liu <songliubraving@fb.com>,
Davidlohr Bueso <dave@stgolabs.net>,
"Paul E . McKenney" <paulmck@kernel.org>,
Rik van Riel <riel@surriel.com>,
Peter Zijlstra <peterz@infradead.org>,
Matthew Wilcox <willy@infradead.org>,
Jerome Glisse <jglisse@redhat.com>,
David Rientjes <rientjes@google.com>,
Axel Rasmussen <axelrasmussen@google.com>,
Suren Baghdasaryan <surenb@google.com>,
Vlastimil Babka <vbabka@suse.cz>
Subject: Re: [PATCH v2 14/70] mm/mmap: Change do_brk_flags() to expand existing VMA and add do_brk_munmap()
Date: Wed, 13 Jan 2021 09:20:33 -0500 [thread overview]
Message-ID: <20210113142022.rbxbb77saxioednq@revolver> (raw)
In-Reply-To: <c02c31eb-7063-510f-2c2d-8d18a6bb5eee@infradead.org>
* Randy Dunlap <rdunlap@infradead.org> [210112 16:23]:
> Hi--
>
> On 1/12/21 8:11 AM, Liam R. Howlett wrote:
> > Avoid allocating a new VMA when it is not necessary. Expand or contract
> > the existing VMA instead. This avoids unnecessary tree manipulations
> > and allocations.
> >
> > Once the VMA is known, use it directly when populating to avoid
> > unnecessary lookup work.
> >
> > Signed-off-by: Liam R. Howlett <Liam.Howlett@Oracle.com>
> > ---
> > mm/mmap.c | 299 +++++++++++++++++++++++++++++++++++++++++++-----------
> > 1 file changed, 237 insertions(+), 62 deletions(-)
> >
> > diff --git a/mm/mmap.c b/mm/mmap.c
> > index a2b32202191d6..f500d5e490f1c 100644
> > --- a/mm/mmap.c
> > +++ b/mm/mmap.c
>
>
>
> > @@ -2022,8 +2068,7 @@ get_unmapped_area(struct file *file, unsigned long addr, unsigned long len,
> > EXPORT_SYMBOL(get_unmapped_area);
> >
> > /**
> > - * find_vma() - Find the VMA for a given address, or the next vma. May return
> > - * NULL in the case of no vma at addr or above
> > + * find_vma() - Find the VMA for a given address, or the next vma.
> > * @mm The mm_struct to check
>
> * @mm: ...
Ack
>
> > * @addr: The address
> > *
> > @@ -2777,16 +2825,102 @@ SYSCALL_DEFINE5(remap_file_pages, unsigned long, start, unsigned long, size,
> > }
> >
> > /*
> > - * this is really a simplified "do_mmap". it only handles
> > - * anonymous maps. eventually we may be able to do some
> > - * brk-specific accounting here.
> > + * bkr_munmap() - Unmap a parital vma.
> > + * @mas: The maple tree state.
> > + * @vma: The vma to be modified
> > + * @newbrk: the start of the address to unmap
> > + * @oldbrk: The end of the address to unmap
>
> missing:
> * @uf: ...
Thanks, yes. I will add the user fault.
>
> > + *
> > + * Returns: 0 on success.
> > + * unmaps a partial VMA mapping. Does not handle alignment, downgrades lock if
> > + * possible.
> > */
> > -static int do_brk_flags(unsigned long addr, unsigned long len,
> > - unsigned long flags, struct list_head *uf)
> > +static int do_brk_munmap(struct ma_state *mas, struct vm_area_struct *vma,
> > + unsigned long newbrk, unsigned long oldbrk,
> > + struct list_head *uf)
> > +{
> > + struct mm_struct *mm = vma->vm_mm;
> > + struct vm_area_struct unmap;
> > + unsigned long unmap_pages;
> > + int ret = 1;
> > +
> > + arch_unmap(mm, newbrk, oldbrk);
> > +
> > + if (likely(vma->vm_start >= newbrk)) { // remove entire mapping(s)
> > + mas_set(mas, newbrk);
> > + if (vma->vm_start != newbrk)
> > + mas_reset(mas); // cause a re-walk for the first overlap.
> > + ret = do_mas_munmap(mas, mm, newbrk, oldbrk-newbrk, uf, true);
> > + goto munmap_full_vma;
> > + }
> > +
> > + vma_init(&unmap, mm);
> > + unmap.vm_start = newbrk;
> > + unmap.vm_end = oldbrk;
> > + ret = userfaultfd_unmap_prep(&unmap, newbrk, oldbrk, uf);
> > + if (ret)
> > + return ret;
> > + ret = 1;
> > +
> > + // Change the oldbrk of vma to the newbrk of the munmap area
> > + vma_adjust_trans_huge(vma, vma->vm_start, newbrk, 0);
> > + if (vma->anon_vma) {
> > + anon_vma_lock_write(vma->anon_vma);
> > + anon_vma_interval_tree_pre_update_vma(vma);
> > + }
> > +
> > + vma->vm_end = newbrk;
> > + if (vma_mas_remove(&unmap, mas))
> > + goto mas_store_fail;
> > +
> > + vmacache_invalidate(vma->vm_mm);
> > + if (vma->anon_vma) {
> > + anon_vma_interval_tree_post_update_vma(vma);
> > + anon_vma_unlock_write(vma->anon_vma);
> > + }
> > +
> > + unmap_pages = vma_pages(&unmap);
> > + if (unmap.vm_flags & VM_LOCKED) {
> > + mm->locked_vm -= unmap_pages;
> > + munlock_vma_pages_range(&unmap, newbrk, oldbrk);
> > + }
> > +
> > + mmap_write_downgrade(mm);
> > + unmap_region(mm, &unmap, vma, newbrk, oldbrk);
> > + /* Statistics */
> > + vm_stat_account(mm, unmap.vm_flags, -unmap_pages);
> > + if (unmap.vm_flags & VM_ACCOUNT)
> > + vm_unacct_memory(unmap_pages);
> > +
> > +munmap_full_vma:
> > + validate_mm_mt(mm);
> > + return ret;
> > +
> > +mas_store_fail:
> > + vma->vm_end = oldbrk;
> > + anon_vma_interval_tree_post_update_vma(vma);
> > + anon_vma_unlock_write(vma->anon_vma);
> > + return -ENOMEM;
> > +}
> > +
> > +/*
> > + * do_brk_flags() - Increase the brk vma if the flags match.
> > + * @mas: The maple tree state.
> > + * @addr: The start address
> > + * @len: The length of the increase
> > + * @vma: The vma,
>
> s/@vma/@brkvma/ ??
yes, sorry. I tried to clarify the name in a later revision.
>
> > + * @flags: The VMA Flags
> > + *
> > + * Extend the brk VMA from addr to addr + len. If the VMA is NULL or the flags
> > + * do not match then create a new anonymous VMA. Eventually we may be able to
> > + * do some brk-specific accounting here.
> > + */
> > +static int do_brk_flags(struct ma_state *mas, struct vm_area_struct **brkvma,
> > + unsigned long addr, unsigned long len,
> > + unsigned long flags)
> > {
> > struct mm_struct *mm = current->mm;
> > - struct vm_area_struct *vma, *prev;
> > - pgoff_t pgoff = addr >> PAGE_SHIFT;
> > + struct vm_area_struct *prev = NULL, *vma;
> > int error;
> > unsigned long mapped_addr;
> > validate_mm_mt(mm);
>
>
>
>
Thank you for looking at this. I will fix all the issues you have
pointed out across the three emails.
Thanks,
Liam
next prev parent reply other threads:[~2021-01-13 14:21 UTC|newest]
Thread overview: 76+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-01-12 16:11 [PATCH v2 00/70] RFC mm: Introducing the Maple Tree Liam R. Howlett
2021-01-12 16:11 ` [PATCH v2 01/70] radix tree test suite: Enhancements for " Liam R. Howlett
2021-01-12 16:11 ` [PATCH v2 02/70] radix tree test suite: Add support for fallthrough attribute Liam R. Howlett
2021-01-12 16:11 ` [PATCH v2 03/70] radix tree test suite: Add support for kmem_cache_free_bulk Liam R. Howlett
2021-01-12 16:11 ` [PATCH v2 04/70] radix tree test suite: Add keme_cache_alloc_bulk() support Liam R. Howlett
2021-01-12 16:11 ` [PATCH v2 05/70] Maple Tree: Add new data structure Liam R. Howlett
2021-01-12 16:11 ` [PATCH v2 06/70] mm: Start tracking VMAs with maple tree Liam R. Howlett
2021-01-12 16:11 ` [PATCH v2 07/70] mm/mmap: Introduce unlock_range() for code cleanup Liam R. Howlett
2021-01-12 16:11 ` [PATCH v2 08/70] mm/mmap: Change find_vma() to use the maple tree Liam R. Howlett
2021-01-12 21:01 ` Randy Dunlap
2021-01-12 16:11 ` [PATCH v2 09/70] mm/mmap: Change find_vma_prev() to use " Liam R. Howlett
2021-01-12 21:03 ` Randy Dunlap
2021-01-12 16:11 ` [PATCH v2 10/70] mm/mmap: Change unmapped_area and unmapped_area_topdown " Liam R. Howlett
2021-01-12 21:05 ` Randy Dunlap
2021-01-12 16:11 ` [PATCH v2 11/70] kernel/fork: Convert dup_mmap " Liam R. Howlett
2021-01-12 16:11 ` [PATCH v2 12/70] mm: Remove rb tree Liam R. Howlett
2021-01-12 16:11 ` [PATCH v2 13/70] mm/gup: Add mm_populate_vma() for use when the vma is known Liam R. Howlett
2021-01-12 16:11 ` [PATCH v2 14/70] mm/mmap: Change do_brk_flags() to expand existing VMA and add do_brk_munmap() Liam R. Howlett
2021-01-12 21:15 ` Randy Dunlap
2021-01-13 14:20 ` Liam R. Howlett [this message]
2021-01-12 16:11 ` [PATCH v2 15/70] mm/mmap: Change vm_brk_flags() to use mm_populate_vma() Liam R. Howlett
2021-01-12 16:11 ` [PATCH v2 16/70] mm: Move find_vma_intersection to mmap.c and change implementation to maple tree Liam R. Howlett
2021-01-12 16:11 ` [PATCH v2 17/70] mm/mmap: Change mmap_region to use maple tree state Liam R. Howlett
2021-01-12 16:11 ` [PATCH v2 18/70] mm/mmap: Drop munmap_vma_range() Liam R. Howlett
2021-01-12 16:11 ` [PATCH v2 19/70] mm: Remove vmacache Liam R. Howlett
2021-01-12 16:11 ` [PATCH v2 20/70] mm/mmap: Change __do_munmap() to avoid unnecessary lookups Liam R. Howlett
2021-01-12 16:11 ` [PATCH v2 21/70] mm/mmap: Move mmap_region() below do_munmap() Liam R. Howlett
2021-01-12 16:11 ` [PATCH v2 22/70] mm/mmap: Add do_mas_munmap() and wraper for __do_munmap() Liam R. Howlett
2021-01-12 16:11 ` [PATCH v2 23/70] mmap: Use find_vma_intersection in do_mmap() for overlap Liam R. Howlett
2021-01-12 16:11 ` [PATCH v2 24/70] mmap: Remove __do_munmap() in favour of do_mas_munmap() Liam R. Howlett
2021-01-12 16:11 ` [PATCH v2 25/70] mm/mmap: Change do_brk_munmap() to use do_mas_align_munmap() Liam R. Howlett
2021-01-12 16:11 ` [PATCH v2 26/70] mmap: make remove_vma_list() inline Liam R. Howlett
2021-01-12 16:11 ` [PATCH v2 27/70] mm: Introduce vma_next() and vma_prev() Liam R. Howlett
2021-01-12 16:11 ` [PATCH v2 28/70] arch/arm64: Remove mmap linked list from vdso Liam R. Howlett
2021-01-12 16:11 ` [PATCH v2 29/70] arch/parsic: Remove mmap linked list from kernel/cache Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 30/70] arch/powerpc: Remove mmap linked list from mm/book2s32/tlb Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 31/70] arch/powerpc: Remove mmap linked list from mm/book2s32/subpage_prot Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 32/70] arch/powerpc: Optimize cell spu task sync Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 33/70] arch/s390: Use maple tree iterators instead of linked list Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 34/70] arch/um: " Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 35/70] arch/x86: Use maple tree iterators for vdso/vma Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 36/70] arch/xtensa: Use maple tree iterators for unmapped area Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 37/70] drivers/misc/cxl: Use maple tree iterators for cxl_prefault_vma() Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 38/70] drivers/oprofile: Lookup address in tree instead of linked list Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 39/70] drivers/tee/optee: Use maple tree iterators for __check_mem_type() Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 40/70] fs/binfmt_elf: Use maple tree iterators for fill_files_note() Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 41/70] fs/coredump: Use maple tree iterators in place of linked list Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 42/70] fs/exec: Use vma_next() instead " Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 43/70] fs/proc/base: Use maple tree iterators in place " Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 44/70] fs/proc/task_mmu: Stop using linked list and highest_vm_end Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 45/70] fs/userfaultfd: Stop using vma linked list Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 46/70] ipc/shm: Stop using the " Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 47/70] kernel/acct: Use maple tree iterators instead of " Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 48/70] kernel/events/core: " Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 49/70] kernel/events/uprobes: " Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 50/70] kernel/sched/fair: " Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 51/70] kernel/sys: " Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 52/70] mm/gup: Use maple tree navigation " Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 53/70] mm/huge_memory: Use vma_next() instead of vma " Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 54/70] mm/khugepaged: Use maple tree iterators " Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 55/70] mm/ksm: " Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 56/70] mm/madvise: Use vma_next " Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 57/70] mm/memcontrol: Stop using mm->highest_vm_end Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 58/70] mm/mempolicy: Use maple tree iterators instead of vma linked list Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 59/70] mm/mlock: " Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 60/70] mm/mprotect: Use maple tree navigation " Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 61/70] mm/mremap: Use vma_next() " Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 62/70] mm/msync: " Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 63/70] mm/nommu: Use maple tree iterators " Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 64/70] mm/oom_kill: " Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 65/70] mm/pagewalk: Use vma_next() " Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 66/70] mm/swapfile: Use maple tree iterator " Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 67/70] mm/nommu: Stop inserting into the " Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 68/70] mm/util: Remove __vma_link_list() and __vma_unlink_list() Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 69/70] mm: Remove vma linked list Liam R. Howlett
2021-01-12 16:12 ` [PATCH v2 70/70] mm/mmap: Convert __insert_vm_struct to use mas, convert vma_link to use vma_mas_link() Liam R. Howlett
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20210113142022.rbxbb77saxioednq@revolver \
--to=liam.howlett@oracle.com \
--cc=akpm@google.com \
--cc=axelrasmussen@google.com \
--cc=dave@stgolabs.net \
--cc=jglisse@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=maple-tree@lists.infradead.org \
--cc=paulmck@kernel.org \
--cc=peterz@infradead.org \
--cc=rdunlap@infradead.org \
--cc=riel@surriel.com \
--cc=rientjes@google.com \
--cc=songliubraving@fb.com \
--cc=surenb@google.com \
--cc=vbabka@suse.cz \
--cc=willy@infradead.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).