From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-ig0-f180.google.com (mail-ig0-f180.google.com [209.85.213.180]) by kanga.kvack.org (Postfix) with ESMTP id 78E386B0038 for ; Sun, 27 Sep 2015 01:31:54 -0400 (EDT) Received: by igbkq10 with SMTP id kq10so34029915igb.0 for ; Sat, 26 Sep 2015 22:31:54 -0700 (PDT) Received: from mail-io0-x230.google.com (mail-io0-x230.google.com. [2607:f8b0:4001:c06::230]) by mx.google.com with ESMTPS id i10si7637192ioo.115.2015.09.26.22.31.53 for (version=TLSv1.2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Sat, 26 Sep 2015 22:31:53 -0700 (PDT) Received: by ioiz6 with SMTP id z6so146178519ioi.2 for ; Sat, 26 Sep 2015 22:31:53 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: References: Date: Sun, 27 Sep 2015 07:31:53 +0200 Message-ID: Subject: Re: [PATCH V4 1/2] ACPI / EC: Fix broken 64bit big-endian users of 'global_lock' From: Jiri Slaby Content-Type: multipart/alternative; boundary=001a11402966fe4f180520b3e5c2 Sender: owner-linux-mm@kvack.org List-ID: To: Viresh Kumar Cc: iommu@lists.linux-foundation.org, linux-acpi@vger.kernel.org, linux-bluetooth@vger.kernel.org, linux-scsi@vger.kernel.org, Intel Linux Wireless , alsa-devel@alsa-project.org, linux-usb@vger.kernel.org, linaro-kernel@lists.linaro.org, linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org, linux-edac@vger.kernel.org, QCA ath9k Development , linux-wireless@vger.kernel.org, linux-mm@kvack.org, linux-arm-kernel@lists.infradead.org, Greg Kroah-Hartman , netdev@vger.kernel.org --001a11402966fe4f180520b3e5c2 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Dne 25. 9. 2015 18:42 napsal u=C5=BEivatel "Viresh Kumar" < viresh.kumar@linaro.org>: > > global_lock is defined as an unsigned long and accessing only its lower > 32 bits from sysfs is incorrect, as we need to consider other 32 bits > for big endian 64 bit systems. There are no such platforms yet, but the > code needs to be robust for such a case. > > Fix that by passing a local variable to debugfs_create_bool() and > assigning its value to global_lock later. But this has to crash whenever the file is read as val's storage is gone at that moment already, right? > Signed-off-by: Viresh Kumar > --- > V3->V4: > - Create a local variable instead of changing type of global_lock > (Rafael) > - Drop the stable tag > - BCC'd a lot of people (rather than cc'ing them) to make sure > - the series reaches them > - mailing lists do not block the patchset due to long cc list > - and we don't spam the BCC'd people for every reply > --- > drivers/acpi/ec_sys.c | 6 ++++-- > 1 file changed, 4 insertions(+), 2 deletions(-) > > diff --git a/drivers/acpi/ec_sys.c b/drivers/acpi/ec_sys.c > index b4c216bab22b..b44b91331a56 100644 > --- a/drivers/acpi/ec_sys.c > +++ b/drivers/acpi/ec_sys.c > @@ -110,6 +110,7 @@ static int acpi_ec_add_debugfs(struct acpi_ec *ec, unsigned int ec_device_count) > struct dentry *dev_dir; > char name[64]; > umode_t mode =3D 0400; > + u32 val; > > if (ec_device_count =3D=3D 0) { > acpi_ec_debugfs_dir =3D debugfs_create_dir("ec", NULL); > @@ -127,10 +128,11 @@ static int acpi_ec_add_debugfs(struct acpi_ec *ec, unsigned int ec_device_count) > > if (!debugfs_create_x32("gpe", 0444, dev_dir, (u32 *)&first_ec->gpe)) > goto error; > - if (!debugfs_create_bool("use_global_lock", 0444, dev_dir, > - (u32 *)&first_ec->global_lock)) > + if (!debugfs_create_bool("use_global_lock", 0444, dev_dir, &val)) > goto error; > > + first_ec->global_lock =3D val; > + > if (write_support) > mode =3D 0600; > if (!debugfs_create_file("io", mode, dev_dir, ec, &acpi_ec_io_ops)) > -- > 2.4.0 > --001a11402966fe4f180520b3e5c2 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable


Dne 25. 9. 2015 18:42 napsal u=C5=BEivatel "Viresh Kumar" <viresh.kumar@linaro.org>: >
> global_lock is defined as an unsigned long and accessing only its lowe= r
> 32 bits from sysfs is incorrect, as we need to consider other 32 bits<= br> > for big endian 64 bit systems. There are no such platforms yet, but th= e
> code needs to be robust for such a case.
>
> Fix that by passing a local variable to debugfs_create_bool() and
> assigning its value to global_lock later.

But this has to crash whenever the file is read as val's= storage is gone at that moment already, right?

> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
> ---
> V3->V4:
> - Create a local variable instead of changing type of global_lock
> =C2=A0 (Rafael)
> - Drop the stable tag
> - BCC'd a lot of people (rather than cc'ing them) to make sure=
> =C2=A0 - the series reaches them
> =C2=A0 - mailing lists do not block the patchset due to long cc list > =C2=A0 - and we don't spam the BCC'd people for every reply > ---
> =C2=A0drivers/acpi/ec_sys.c | 6 ++++--
> =C2=A01 file changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/acpi/ec_sys.c b/drivers/acpi/ec_sys.c
> index b4c216bab22b..b44b91331a56 100644
> --- a/drivers/acpi/ec_sys.c
> +++ b/drivers/acpi/ec_sys.c
> @@ -110,6 +110,7 @@ static int acpi_ec_add_debugfs(struct acpi_ec *ec,= unsigned int ec_device_count)
> =C2=A0 =C2=A0 =C2=A0 =C2=A0 struct dentry *dev_dir;
> =C2=A0 =C2=A0 =C2=A0 =C2=A0 char name[64];
> =C2=A0 =C2=A0 =C2=A0 =C2=A0 umode_t mode =3D 0400;
> +=C2=A0 =C2=A0 =C2=A0 =C2=A0u32 val;
>
> =C2=A0 =C2=A0 =C2=A0 =C2=A0 if (ec_device_count =3D=3D 0) {
> =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 acpi_ec_debugf= s_dir =3D debugfs_create_dir("ec", NULL);
> @@ -127,10 +128,11 @@ static int acpi_ec_add_debugfs(struct acpi_ec *e= c, unsigned int ec_device_count)
>
> =C2=A0 =C2=A0 =C2=A0 =C2=A0 if (!debugfs_create_x32("gpe", 0= 444, dev_dir, (u32 *)&first_ec->gpe))
> =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 goto error; > -=C2=A0 =C2=A0 =C2=A0 =C2=A0if (!debugfs_create_bool("use_global_= lock", 0444, dev_dir,
> -=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0= =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 (u32 *)&first_ec->global_= lock))
> +=C2=A0 =C2=A0 =C2=A0 =C2=A0if (!debugfs_create_bool("use_global_= lock", 0444, dev_dir, &val))
> =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 goto error; >
> +=C2=A0 =C2=A0 =C2=A0 =C2=A0first_ec->global_lock =3D val;
> +
> =C2=A0 =C2=A0 =C2=A0 =C2=A0 if (write_support)
> =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 mode =3D 0600;=
> =C2=A0 =C2=A0 =C2=A0 =C2=A0 if (!debugfs_create_file("io", m= ode, dev_dir, ec, &acpi_ec_io_ops))
> --
> 2.4.0
>

--001a11402966fe4f180520b3e5c2-- -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org