[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAGXu5jKnf7NebFQp-w9gka9nR2V4Jc3aSfNK2yvvrYa=UUEh-g@mail.gmail.com>
Date: Fri, 10 Feb 2017 13:55:03 -0800
From: Kees Cook <keescook@...omium.org>
To: Jess Frazelle <me@...sfraz.com>
Cc: "kernel-hardening@...ts.openwall.com" <kernel-hardening@...ts.openwall.com>
Subject: Re: [PATCH 1/4] irq: set {msi_domain,syscore}_ops
as __ro_after_init
On Fri, Feb 10, 2017 at 2:08 AM, Jess Frazelle <me@...sfraz.com> wrote:
> Marked msi_domain_ops structs as __ro_after_init when called only during init.
> Marked syscore_ops structs as __ro_after_init when register_syscore_ops was
> called only during init. Most of the caller functions were already annotated as
> __init.
> This protects the data structure from accidental corruption.
>
> Suggested-by: Kees Cook <keescook@...omium.org>
> Signed-off-by: Jess Frazelle <me@...sfraz.com>
> ---
> drivers/staging/fsl-mc/bus/irq-gic-v3-its-fsl-mc-msi.c | 2 +-
> kernel/irq/generic-chip.c | 2 +-
> kernel/irq/msi.c | 2 +-
> kernel/irq/pm.c | 2 +-
> 4 files changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/staging/fsl-mc/bus/irq-gic-v3-its-fsl-mc-msi.c b/drivers/staging/fsl-mc/bus/irq-gic-v3-its-fsl-mc-msi.c
> index 6b1cd574644f..0e2c1b5e13b7 100644
> --- a/drivers/staging/fsl-mc/bus/irq-gic-v3-its-fsl-mc-msi.c
> +++ b/drivers/staging/fsl-mc/bus/irq-gic-v3-its-fsl-mc-msi.c
> @@ -51,7 +51,7 @@ static int its_fsl_mc_msi_prepare(struct irq_domain *msi_domain,
> return msi_info->ops->msi_prepare(msi_domain->parent, dev, nvec, info);
> }
>
> -static struct msi_domain_ops its_fsl_mc_msi_ops = {
> +static struct msi_domain_ops its_fsl_mc_msi_ops __ro_after_init = {
> .msi_prepare = its_fsl_mc_msi_prepare,
> };
>
> diff --git a/kernel/irq/generic-chip.c b/kernel/irq/generic-chip.c
> index ee32870079c9..cca63dbaabea 100644
> --- a/kernel/irq/generic-chip.c
> +++ b/kernel/irq/generic-chip.c
> @@ -623,7 +623,7 @@ static void irq_gc_shutdown(void)
> }
> }
>
> -static struct syscore_ops irq_gc_syscore_ops = {
> +static struct syscore_ops irq_gc_syscore_ops __ro_after_init = {
> .suspend = irq_gc_suspend,
> .resume = irq_gc_resume,
> .shutdown = irq_gc_shutdown,
> diff --git a/kernel/irq/msi.c b/kernel/irq/msi.c
> index ee230063f033..0e5b723f710f 100644
> --- a/kernel/irq/msi.c
> +++ b/kernel/irq/msi.c
> @@ -217,7 +217,7 @@ static int msi_domain_ops_check(struct irq_domain *domain,
> return 0;
> }
>
> -static struct msi_domain_ops msi_domain_ops_default = {
> +static struct msi_domain_ops msi_domain_ops_default __ro_after_init = {
> .get_hwirq = msi_domain_ops_get_hwirq,
> .msi_init = msi_domain_ops_init,
> .msi_check = msi_domain_ops_check,
> diff --git a/kernel/irq/pm.c b/kernel/irq/pm.c
> index cea1de0161f1..d6b889bed323 100644
> --- a/kernel/irq/pm.c
> +++ b/kernel/irq/pm.c
> @@ -185,7 +185,7 @@ static void irq_pm_syscore_resume(void)
> resume_irqs(true);
> }
>
> -static struct syscore_ops irq_pm_syscore_ops = {
> +static struct syscore_ops irq_pm_syscore_ops __ro_after_init = {
> .resume = irq_pm_syscore_resume,
> };
Cool! How'd you end up choosing these? Did you just go looking for
one-sided initializations? (i.e. register_syscore_ops() without
unregister_syscore_ops() call?)
(It may help the commit message to explicitly state that
unregister_syscore_ops() is never called on these ops.)
-Kees
--
Kees Cook
Pixel Security
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.
