linux-kernel - Re: [PATCH v3 6/5] AHCI: Do not read HOST_IRQ

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20140919183950.GJ8793@agordeev.usersys.redhat.com>
Date:	Fri, 19 Sep 2014 19:39:51 +0100
From:	Alexander Gordeev <agordeev@...hat.com>
To:	Dan Williams <dan.j.williams@...il.com>
Cc:	Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
	IDE/ATA development list <linux-ide@...r.kernel.org>,
	"Jiang, Dave" <dave.jiang@...el.com>
Subject: Re: [PATCH v3 6/5] AHCI: Do not read HOST_IRQ_STAT register in
 multi-MSI mode

On Fri, Sep 19, 2014 at 09:34:34AM -0700, Dan Williams wrote:
> On Fri, Sep 19, 2014 at 1:25 AM, Alexander Gordeev <agordeev@...hat.com> wrote:
> > As described in AHCI v1.0 specification chapter 10.6.2.2
> > "Multiple MSI Based Messages" generation of interrupts
> > is not controlled through the HOST_IRQ_STAT register.
> >
> > Considering MMIO access is expensive remove unnecessary
> > reading and writing of HOST_IRQ_STAT register.
> >
> > Further, serializing access to the host data is no longer
> > needed and the interrupt service routine can avoid competing
> > on the host lock.
> >
> > Signed-off-by: Alexander Gordeev <agordeev@...hat.com>
> > Suggested-by: "Jiang, Dave" <dave.jiang@...el.com>
> > Cc: linux-ide@...r.kernel.org
> > ---
> >  drivers/ata/ahci.h    |  1 +
> >  drivers/ata/libahci.c | 54 ++++++++-------------------------------------------
> >  2 files changed, 9 insertions(+), 46 deletions(-)
> >
> > diff --git a/drivers/ata/ahci.h b/drivers/ata/ahci.h
> > index c12f590..b8e117a 100644
> > --- a/drivers/ata/ahci.h
> > +++ b/drivers/ata/ahci.h
> > @@ -305,6 +305,7 @@ struct ahci_port_priv {
> >         unsigned int            ncq_saw_dmas:1;
> >         unsigned int            ncq_saw_sdb:1;
> >         u32                     intr_status;    /* interrupts to handle */
> > +       spinlock_t              intr_lock;      /* protects intr_status */
> 
> Why introduce a new lock?  Can't we switch to per-ata port locking
> rather than ata_host locking?

We could. But this case hardware context interrupt handler would compete
with threads/softriqs, which is exactly what I tried to avoid. With the
separate lock we *only* update ahci_port_priv::intr_status with interrupts
disabled.

-- 
Regards,
Alexander Gordeev
agordeev@...hat.com
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/