From: Leonardo Bras <leonardo@linux.ibm.com>
To: Christophe Leroy <christophe.leroy@c-s.fr>,
Benjamin Herrenschmidt <benh@kernel.crashing.org>,
Paul Mackerras <paulus@samba.org>,
Michael Ellerman <mpe@ellerman.id.au>,
Arnd Bergmann <arnd@arndb.de>,
Andrew Morton <akpm@linux-foundation.org>,
"Aneesh Kumar K.V" <aneesh.kumar@linux.ibm.com>,
Nicholas Piggin <npiggin@gmail.com>,
Steven Price <steven.price@arm.com>,
Robin Murphy <robin.murphy@arm.com>,
Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>,
Balbir Singh <bsingharora@gmail.com>,
Reza Arbab <arbab@linux.ibm.com>,
Thomas Gleixner <tglx@linutronix.de>,
Allison Randal <allison@lohutok.net>,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
Mike Rapoport <rppt@linux.ibm.com>,
Michal Suchanek <msuchanek@suse.de>
Cc: linux-arch@vger.kernel.org, linux-mm@kvack.org,
linuxppc-dev@lists.ozlabs.org, linux-kernel@vger.kernel.org,
kvm-ppc@vger.kernel.org
Subject: Re: [PATCH v6 03/11] powerpc/mm: Adds arch-specificic functions to track lockless pgtable walks
Date: Fri, 07 Feb 2020 01:38:39 -0300 [thread overview]
Message-ID: <efcf780cee767d0f4b06b56e216725c6bd8d60d4.camel@linux.ibm.com> (raw)
In-Reply-To: <1311ce1c-7e5a-f7c4-2ab2-c03e124ca1c1@c-s.fr>
[-- Attachment #1: Type: text/plain, Size: 7765 bytes --]
On Thu, 2020-02-06 at 06:46 +0100, Christophe Leroy wrote:
>
> Le 06/02/2020 à 04:08, Leonardo Bras a écrit :
> > On powerpc, we need to do some lockless pagetable walks from functions
> > that already have disabled interrupts, specially from real mode with
> > MSR[EE=0].
> >
> > In these contexts, disabling/enabling interrupts can be very troubling.
>
> When interrupts are already disabled, the flag returned when disabling
> it will be such that when we restore it later, interrupts remain
> disabled, so what's the problem ?
There are places in code, like patch 8, where it explicitly avoids
doing irq_save/restore by using a function parameter (realmode).
http://patchwork.ozlabs.org/patch/1234130/
I am not sure why it's that way, but I decided to keep it as is.
It was introduced by Aneesh Kumar in 2015
(691e95fd7396905a38d98919e9c150dbc3ea21a3).
> > So, this arch-specific implementation features functions with an extra
> > argument that allows interrupt enable/disable to be skipped:
> > __begin_lockless_pgtbl_walk() and __end_lockless_pgtbl_walk().
> >
> > Functions similar to the generic ones are also exported, by calling
> > the above functions with parameter {en,dis}able_irq = true.
> >
> > Signed-off-by: Leonardo Bras <leonardo@linux.ibm.com>
> > ---
> > arch/powerpc/include/asm/book3s/64/pgtable.h | 6 ++
> > arch/powerpc/mm/book3s64/pgtable.c | 86 +++++++++++++++++++-
> > 2 files changed, 91 insertions(+), 1 deletion(-)
> >
> > diff --git a/arch/powerpc/include/asm/book3s/64/pgtable.h b/arch/powerpc/include/asm/book3s/64/pgtable.h
> > index 201a69e6a355..78f6ffb1bb3e 100644
> > --- a/arch/powerpc/include/asm/book3s/64/pgtable.h
> > +++ b/arch/powerpc/include/asm/book3s/64/pgtable.h
> > @@ -1375,5 +1375,11 @@ static inline bool pgd_is_leaf(pgd_t pgd)
> > return !!(pgd_raw(pgd) & cpu_to_be64(_PAGE_PTE));
> > }
> >
> > +#define __HAVE_ARCH_LOCKLESS_PGTBL_WALK_CONTROL
> > +unsigned long begin_lockless_pgtbl_walk(void);
> > +unsigned long __begin_lockless_pgtbl_walk(bool disable_irq);
> > +void end_lockless_pgtbl_walk(unsigned long irq_mask);
> > +void __end_lockless_pgtbl_walk(unsigned long irq_mask, bool enable_irq);
> > +
>
> Why not make them static inline just like the generic ones ?
>
Sure, can be done. It would save some function calls.
For that I will define the per-cpu variable in .c and declare it in .h
All new function can be moved to .h, while changing adding the inline
modifier.
> > #endif /* __ASSEMBLY__ */
> > #endif /* _ASM_POWERPC_BOOK3S_64_PGTABLE_H_ */
> > diff --git a/arch/powerpc/mm/book3s64/pgtable.c b/arch/powerpc/mm/book3s64/pgtable.c
> > index 2bf7e1b4fd82..535613030363 100644
> > --- a/arch/powerpc/mm/book3s64/pgtable.c
> > +++ b/arch/powerpc/mm/book3s64/pgtable.c
> > @@ -82,6 +82,7 @@ static void do_nothing(void *unused)
> > {
> >
> > }
> > +
>
> Is this blank line related to the patch ?
>
Nope, just something I 'fixed' while reading, and gone past my pre-send
patch reviewing. If it bothers, I can remove.
> > /*
> > * Serialize against find_current_mm_pte which does lock-less
> > * lookup in page tables with local interrupts disabled. For huge pages
> > @@ -98,6 +99,89 @@ void serialize_against_pte_lookup(struct mm_struct *mm)
> > smp_call_function_many(mm_cpumask(mm), do_nothing, NULL, 1);
> > }
> >
> > +/* begin_lockless_pgtbl_walk: Must be inserted before a function call that does
> > + * lockless pagetable walks, such as __find_linux_pte().
> > + * This version allows setting disable_irq=false, so irqs are not touched, which
> > + * is quite useful for running when ints are already disabled (like real-mode)
> > + */
> > +inline
> > +unsigned long __begin_lockless_pgtbl_walk(bool disable_irq)
> > +{
> > + unsigned long irq_mask = 0;
> > +
> > + /*
> > + * Interrupts must be disabled during the lockless page table walk.
> > + * That's because the deleting or splitting involves flushing TLBs,
> > + * which in turn issues interrupts, that will block when disabled.
> > + *
> > + * When this function is called from realmode with MSR[EE=0],
> > + * it's not needed to touch irq, since it's already disabled.
> > + */
> > + if (disable_irq)
> > + local_irq_save(irq_mask);
> > +
> > + /*
> > + * This memory barrier pairs with any code that is either trying to
> > + * delete page tables, or split huge pages. Without this barrier,
> > + * the page tables could be read speculatively outside of interrupt
> > + * disabling or reference counting.
> > + */
> > + smp_mb();
> > +
> > + return irq_mask;
> > +}
> > +EXPORT_SYMBOL(__begin_lockless_pgtbl_walk);
> > +
> > +/* begin_lockless_pgtbl_walk: Must be inserted before a function call that does
> > + * lockless pagetable walks, such as __find_linux_pte().
> > + * This version is used by generic code, and always assume irqs will be disabled
> > + */
> > +unsigned long begin_lockless_pgtbl_walk(void)
> > +{
> > + return __begin_lockless_pgtbl_walk(true);
> > +}
> > +EXPORT_SYMBOL(begin_lockless_pgtbl_walk);
>
> Even more than begin_lockless_pgtbl_walk(), this one is worth being
> static inline in the H file.
Same as above, can be done.
>
> > +
> > +/*
> > + * __end_lockless_pgtbl_walk: Must be inserted after the last use of a pointer
> > + * returned by a lockless pagetable walk, such as __find_linux_pte()
> > + * This version allows setting enable_irq=false, so irqs are not touched, which
> > + * is quite useful for running when ints are already disabled (like real-mode)
> > + */
> > +inline void __end_lockless_pgtbl_walk(unsigned long irq_mask, bool enable_irq)
> > +{
> > + /*
> > + * This memory barrier pairs with any code that is either trying to
> > + * delete page tables, or split huge pages. Without this barrier,
> > + * the page tables could be read speculatively outside of interrupt
> > + * disabling or reference counting.
> > + */
> > + smp_mb();
> > +
> > + /*
> > + * Interrupts must be disabled during the lockless page table walk.
> > + * That's because the deleting or splitting involves flushing TLBs,
> > + * which in turn issues interrupts, that will block when disabled.
> > + *
> > + * When this function is called from realmode with MSR[EE=0],
> > + * it's not needed to touch irq, since it's already disabled.
> > + */
> > + if (enable_irq)
> > + local_irq_restore(irq_mask);
> > +}
> > +EXPORT_SYMBOL(__end_lockless_pgtbl_walk);
> > +
> > +/*
> > + * end_lockless_pgtbl_walk: Must be inserted after the last use of a pointer
> > + * returned by a lockless pagetable walk, such as __find_linux_pte()
> > + * This version is used by generic code, and always assume irqs will be enabled
> > + */
> > +void end_lockless_pgtbl_walk(unsigned long irq_mask)
> > +{
> > + __end_lockless_pgtbl_walk(irq_mask, true);
> > +}
> > +EXPORT_SYMBOL(end_lockless_pgtbl_walk);
> > +
> > /*
> > * We use this to invalidate a pmdp entry before switching from a
> > * hugepte to regular pmd entry.
> > @@ -487,7 +571,7 @@ static int __init setup_disable_tlbie(char *str)
> > tlbie_capable = false;
> > tlbie_enabled = false;
> >
> > - return 1;
> > + return 1;
>
> Is that related to this patch at all ?
Nope, just another something I 'fixed' while reading, and gone past my
pre-send patch reviewing. If it bothers, I can remove.
>
> > }
> > __setup("disable_tlbie", setup_disable_tlbie);
> >
> >
>
> Christophe
Found other unbalanced begin/end in kvmppc_do_h_enter(). I will change
that too.
Thanks for the feedback,
Leonardo Bras
[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 833 bytes --]
next prev parent reply other threads:[~2020-02-07 4:41 UTC|newest]
Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-02-06 3:08 [PATCH v6 00/11] Introduces new functions for tracking lockless pagetable walks Leonardo Bras
2020-02-06 3:08 ` [PATCH v6 01/11] asm-generic/pgtable: Adds generic functions to track lockless pgtable walks Leonardo Bras
2020-02-06 5:54 ` Christophe Leroy
2020-02-07 2:19 ` Leonardo Bras
2020-02-07 5:39 ` kbuild test robot
2020-02-06 3:08 ` [PATCH v6 02/11] mm/gup: Use functions to track lockless pgtbl walks on gup_pgd_range Leonardo Bras
2020-02-06 3:25 ` Leonardo Bras
2020-02-07 22:54 ` John Hubbard
2020-02-17 20:55 ` Leonardo Bras
2020-10-15 14:46 ` Michal Suchánek
2020-10-16 3:27 ` Aneesh Kumar K.V
2020-02-07 1:19 ` kbuild test robot
2020-02-07 8:01 ` kbuild test robot
2020-02-06 3:08 ` [PATCH v6 03/11] powerpc/mm: Adds arch-specificic functions to track lockless pgtable walks Leonardo Bras
2020-02-06 5:46 ` Christophe Leroy
2020-02-07 4:38 ` Leonardo Bras [this message]
2020-02-17 20:32 ` Leonardo Bras
2020-02-06 3:08 ` [PATCH v6 04/11] powerpc/mce_power: Use functions to track lockless pgtbl walks Leonardo Bras
2020-02-06 5:48 ` Christophe Leroy
2020-02-07 4:00 ` Leonardo Bras
2020-02-06 3:08 ` [PATCH v6 05/11] powerpc/perf: " Leonardo Bras
2020-02-06 3:08 ` [PATCH v6 06/11] powerpc/mm/book3s64/hash: " Leonardo Bras
2020-02-06 6:06 ` Christophe Leroy
2020-02-07 3:49 ` Leonardo Bras
2020-02-06 3:08 ` [PATCH v6 07/11] powerpc/kvm/e500: " Leonardo Bras
2020-02-06 6:18 ` Christophe Leroy
2020-02-07 3:10 ` Leonardo Bras
2020-02-06 3:08 ` [PATCH v6 08/11] powerpc/kvm/book3s_hv: " Leonardo Bras
2020-02-06 3:08 ` [PATCH v6 09/11] powerpc/kvm/book3s_64: " Leonardo Bras
2020-02-06 3:08 ` [PATCH v6 10/11] powerpc/mm: Adds counting method to track lockless pagetable walks Leonardo Bras
2020-02-06 6:23 ` Christophe Leroy
2020-02-07 1:56 ` Leonardo Bras
2020-02-06 3:09 ` [PATCH v6 11/11] powerpc/mm/book3s64/pgtable: Uses counting method to skip serializing Leonardo Bras
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=efcf780cee767d0f4b06b56e216725c6bd8d60d4.camel@linux.ibm.com \
--to=leonardo@linux.ibm.com \
--cc=akpm@linux-foundation.org \
--cc=allison@lohutok.net \
--cc=aneesh.kumar@linux.ibm.com \
--cc=arbab@linux.ibm.com \
--cc=arnd@arndb.de \
--cc=benh@kernel.crashing.org \
--cc=bsingharora@gmail.com \
--cc=christophe.leroy@c-s.fr \
--cc=gregkh@linuxfoundation.org \
--cc=kvm-ppc@vger.kernel.org \
--cc=linux-arch@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=linuxppc-dev@lists.ozlabs.org \
--cc=mahesh@linux.vnet.ibm.com \
--cc=mpe@ellerman.id.au \
--cc=msuchanek@suse.de \
--cc=npiggin@gmail.com \
--cc=paulus@samba.org \
--cc=robin.murphy@arm.com \
--cc=rppt@linux.ibm.com \
--cc=steven.price@arm.com \
--cc=tglx@linutronix.de \
/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).