From: Andi Kleen <andi@firstfloor.org>
To: Anton Salikhmetov <salikhmetov@gmail.com>
Cc: linux-mm@kvack.org, jakob@unthought.net,
linux-kernel@vger.kernel.org, valdis.kletnieks@vt.edu,
riel@redhat.com, ksm@42.dk, staubach@redhat.com,
jesper.juhl@gmail.com, torvalds@osdl.org
Subject: Re: [PATCH -v7 2/2] Update ctime and mtime for memory-mapped files
Date: 22 Jan 2008 05:39:43 +0100 [thread overview]
Message-ID: <p737ii2o4mo.fsf@crumb.suse.de> (raw)
In-Reply-To: <12009619584168-git-send-email-salikhmetov@gmail.com>
Anton Salikhmetov <salikhmetov@gmail.com> writes:
You should probably put your design document somewhere in Documentation
with a patch.
> + * Scan the PTEs for pages belonging to the VMA and mark them read-only.
> + * It will force a pagefault on the next write access.
> + */
> +static void vma_wrprotect(struct vm_area_struct *vma)
> +{
> + unsigned long addr;
> +
> + for (addr = vma->vm_start; addr < vma->vm_end; addr += PAGE_SIZE) {
> + spinlock_t *ptl;
> + pgd_t *pgd = pgd_offset(vma->vm_mm, addr);
> + pud_t *pud = pud_offset(pgd, addr);
> + pmd_t *pmd = pmd_offset(pud, addr);
> + pte_t *pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl);
This means on i386 with highmem ptes you will map/flush tlb/unmap each
PTE individually. You will do 512 times as much work as really needed
per PTE leaf page.
The performance critical address space walkers use a different design
pattern that avoids this.
> + if (pte_dirty(*pte) && pte_write(*pte)) {
> + pte_t entry = ptep_clear_flush(vma, addr, pte);
Flushing TLBs unbatched can also be very expensive because if the MM is
shared by several CPUs you'll have a inter-processor interrupt for
each iteration. They are quite costly even on smaller systems.
It would be better if you did a single flush_tlb_range() at the end.
This means on x86 this will currently always do a full flush, but that's
still better than really slowing down in the heavily multithreaded case.
-Andi
next prev parent reply other threads:[~2008-01-22 4:39 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-01-22 0:32 [PATCH -v7 0/2] Fixing the issue with memory-mapped file times Anton Salikhmetov
2008-01-22 0:32 ` [PATCH -v7 1/2] Massive code cleanup of sys_msync() Anton Salikhmetov
2008-01-22 0:32 ` [PATCH -v7 2/2] Update ctime and mtime for memory-mapped files Anton Salikhmetov
2008-01-22 1:40 ` Jesper Juhl
2008-01-22 1:51 ` Anton Salikhmetov
2008-01-22 1:54 ` Jesper Juhl
2008-01-22 1:57 ` Anton Salikhmetov
2008-01-22 2:18 ` Jesper Juhl
2008-01-22 2:07 ` Anton Salikhmetov
2008-01-22 2:16 ` Jesper Juhl
2008-01-22 2:16 ` Linus Torvalds
2008-01-22 2:39 ` Anton Salikhmetov
2008-01-22 8:52 ` Miklos Szeredi
2008-01-22 4:39 ` Andi Kleen [this message]
2008-01-22 1:34 ` [PATCH -v7 0/2] Fixing the issue with memory-mapped file times Jesper Juhl
2008-01-22 1:40 ` Anton Salikhmetov
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=p737ii2o4mo.fsf@crumb.suse.de \
--to=andi@firstfloor.org \
--cc=jakob@unthought.net \
--cc=jesper.juhl@gmail.com \
--cc=ksm@42.dk \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=riel@redhat.com \
--cc=salikhmetov@gmail.com \
--cc=staubach@redhat.com \
--cc=torvalds@osdl.org \
--cc=valdis.kletnieks@vt.edu \
/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).