[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: Thu, 4 Apr 2019 09:56:48 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: Khalid Aziz <khalid.aziz@...cle.com>
Cc: juergh@...il.com, tycho@...ho.ws, jsteckli@...zon.de,
ak@...ux.intel.com, liran.alon@...cle.com, keescook@...gle.com,
konrad.wilk@...cle.com, deepa.srinivasan@...cle.com,
chris.hyser@...cle.com, tyhicks@...onical.com, dwmw@...zon.co.uk,
andrew.cooper3@...rix.com, jcm@...hat.com,
boris.ostrovsky@...cle.com, kanth.ghatraju@...cle.com,
joao.m.martins@...cle.com, jmattson@...gle.com,
pradeep.vincent@...cle.com, john.haxby@...cle.com,
tglx@...utronix.de, kirill.shutemov@...ux.intel.com, hch@....de,
steven.sistare@...cle.com, labbott@...hat.com, luto@...nel.org,
dave.hansen@...el.com, aaron.lu@...el.com,
akpm@...ux-foundation.org, alexander.h.duyck@...ux.intel.com,
amir73il@...il.com, andreyknvl@...gle.com,
aneesh.kumar@...ux.ibm.com, anthony.yznaga@...cle.com,
ard.biesheuvel@...aro.org, arnd@...db.de, arunks@...eaurora.org,
ben@...adent.org.uk, bigeasy@...utronix.de, bp@...en8.de,
brgl@...ev.pl, catalin.marinas@....com, corbet@....net,
cpandya@...eaurora.org, daniel.vetter@...ll.ch,
dan.j.williams@...el.com, gregkh@...uxfoundation.org, guro@...com,
hannes@...xchg.org, hpa@...or.com, iamjoonsoo.kim@....com,
james.morse@....com, jannh@...gle.com, jgross@...e.com,
jkosina@...e.cz, jmorris@...ei.org, joe@...ches.com,
jrdr.linux@...il.com, jroedel@...e.de, keith.busch@...el.com,
khlebnikov@...dex-team.ru, logang@...tatee.com,
marco.antonio.780@...il.com, mark.rutland@....com,
mgorman@...hsingularity.net, mhocko@...e.com, mhocko@...e.cz,
mike.kravetz@...cle.com, mingo@...hat.com, mst@...hat.com,
m.szyprowski@...sung.com, npiggin@...il.com, osalvador@...e.de,
paulmck@...ux.vnet.ibm.com, pavel.tatashin@...rosoft.com,
rdunlap@...radead.org, richard.weiyang@...il.com, riel@...riel.com,
rientjes@...gle.com, robin.murphy@....com, rostedt@...dmis.org,
rppt@...ux.vnet.ibm.com, sai.praneeth.prakhya@...el.com,
serge@...lyn.com, steve.capper@....com, thymovanbeers@...il.com,
vbabka@...e.cz, will.deacon@....com, willy@...radead.org,
yang.shi@...ux.alibaba.com, yaojun8558363@...il.com,
ying.huang@...el.com, zhangshaokun@...ilicon.com,
iommu@...ts.linux-foundation.org, x86@...nel.org,
linux-arm-kernel@...ts.infradead.org, linux-doc@...r.kernel.org,
linux-kernel@...r.kernel.org, linux-mm@...ck.org,
linux-security-module@...r.kernel.org,
Khalid Aziz <khalid@...ehiking.org>,
kernel-hardening@...ts.openwall.com,
"Vasileios P . Kemerlis" <vpk@...columbia.edu>,
Juerg Haefliger <juerg.haefliger@...onical.com>,
David Woodhouse <dwmw2@...radead.org>
Subject: Re: [RFC PATCH v9 11/13] xpfo, mm: optimize spinlock usage in
xpfo_kunmap
On Wed, Apr 03, 2019 at 11:34:12AM -0600, Khalid Aziz wrote:
> From: Julian Stecklina <jsteckli@...zon.de>
>
> Only the xpfo_kunmap call that needs to actually unmap the page
> needs to be serialized. We need to be careful to handle the case,
> where after the atomic decrement of the mapcount, a xpfo_kmap
> increased the mapcount again. In this case, we can safely skip
> modifying the page table.
>
> Model-checked with up to 4 concurrent callers with Spin.
>
> Signed-off-by: Julian Stecklina <jsteckli@...zon.de>
> Signed-off-by: Khalid Aziz <khalid.aziz@...cle.com>
> Cc: Khalid Aziz <khalid@...ehiking.org>
> Cc: x86@...nel.org
> Cc: kernel-hardening@...ts.openwall.com
> Cc: Vasileios P. Kemerlis <vpk@...columbia.edu>
> Cc: Juerg Haefliger <juerg.haefliger@...onical.com>
> Cc: Tycho Andersen <tycho@...ho.ws>
> Cc: Marco Benatto <marco.antonio.780@...il.com>
> Cc: David Woodhouse <dwmw2@...radead.org>
> ---
> include/linux/xpfo.h | 24 +++++++++++++++---------
> 1 file changed, 15 insertions(+), 9 deletions(-)
>
> diff --git a/include/linux/xpfo.h b/include/linux/xpfo.h
> index 2318c7eb5fb7..37e7f52fa6ce 100644
> --- a/include/linux/xpfo.h
> +++ b/include/linux/xpfo.h
> @@ -61,6 +61,7 @@ static inline void xpfo_kmap(void *kaddr, struct page *page)
> static inline void xpfo_kunmap(void *kaddr, struct page *page)
> {
> unsigned long flags;
> + bool flush_tlb = false;
>
> if (!static_branch_unlikely(&xpfo_inited))
> return;
> @@ -72,18 +73,23 @@ static inline void xpfo_kunmap(void *kaddr, struct page *page)
> * The page is to be allocated back to user space, so unmap it from
> * the kernel, flush the TLB and tag it as a user page.
> */
> - spin_lock_irqsave(&page->xpfo_lock, flags);
> -
> if (atomic_dec_return(&page->xpfo_mapcount) == 0) {
> -#ifdef CONFIG_XPFO_DEBUG
> - WARN_ON(PageXpfoUnmapped(page));
> -#endif
> - SetPageXpfoUnmapped(page);
> - set_kpte(kaddr, page, __pgprot(0));
> - xpfo_flush_kernel_tlb(page, 0);
> + spin_lock_irqsave(&page->xpfo_lock, flags);
> +
> + /*
> + * In the case, where we raced with kmap after the
> + * atomic_dec_return, we must not nuke the mapping.
> + */
> + if (atomic_read(&page->xpfo_mapcount) == 0) {
> + SetPageXpfoUnmapped(page);
> + set_kpte(kaddr, page, __pgprot(0));
> + flush_tlb = true;
> + }
> + spin_unlock_irqrestore(&page->xpfo_lock, flags);
> }
>
> - spin_unlock_irqrestore(&page->xpfo_lock, flags);
> + if (flush_tlb)
> + xpfo_flush_kernel_tlb(page, 0);
> }
This doesn't help with the TLB invalidation issue, AFAICT this is still
completely buggered. kunmap_atomic() can be called from IRQ context.
Powered by blists - more mailing lists
Confused about mailing lists and their use?
Read about mailing lists on Wikipedia
and check out these
guidelines on proper formatting of your messages.
