linux-kernel - Re: [PATCH v4 3/5] watchdog: iTCO_wdt: Add PMC specific noreboot update api

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

Date:   Sun, 2 Apr 2017 17:04:57 +0300
From:   Andy Shevchenko <andy.shevchenko@...il.com>
To:     Kuppuswamy Sathyanarayanan 
        <sathyanarayanan.kuppuswamy@...ux.intel.com>
Cc:     Andy Shevchenko <andy@...radead.org>,
        Zha Qipeng <qipeng.zha@...el.com>,
        "dvhart@...radead.org" <dvhart@...radead.org>,
        Guenter Roeck <linux@...ck-us.net>,
        Wim Van Sebroeck <wim@...ana.be>, sathyaosid@...il.com,
        David Box <david.e.box@...ux.intel.com>,
        Rajneesh Bhardwaj <rajneesh.bhardwaj@...el.com>,
        Platform Driver <platform-driver-x86@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        linux-watchdog@...r.kernel.org
Subject: Re: [PATCH v4 3/5] watchdog: iTCO_wdt: Add PMC specific noreboot
 update api

On Sat, Apr 1, 2017 at 2:27 AM, Kuppuswamy Sathyanarayanan
<sathyanarayanan.kuppuswamy@...ux.intel.com> wrote:
> In some SOCs, setting noreboot bit needs modification to

SoCs.

Perhaps you can create a wikipage to share with your team what style
issues usually needs to be addressed.
One of them is a proper capitalization in abbreviations / code names.

> PMC GC registers. But not all PMC drivers allow other drivers
> to memory map their GC region. This could create mem request
> conflict in watchdog driver. So this patch adds facility to allow
> PMC drivers to pass noreboot update function to watchdog
> drivers via platform data.

> --- a/drivers/watchdog/iTCO_wdt.c
> +++ b/drivers/watchdog/iTCO_wdt.c
> @@ -100,6 +100,8 @@ struct iTCO_wdt_private {
>          */
>         struct resource *gcs_pmc_res;
>         unsigned long __iomem *gcs_pmc;

> +       /* pmc specific api to update noreboot flag */

PMC
API

> +       int (*update_noreboot_flag)(bool status);
>         /* the lock for io operations */
>         spinlock_t io_lock;
>         /* the PCI-device */
> @@ -176,9 +178,13 @@ static void iTCO_wdt_set_NO_REBOOT_bit(struct iTCO_wdt_private *p)
>
>         /* Set the NO_REBOOT bit: this disables reboots */
>         if (p->iTCO_version >= 2) {
> -               val32 = readl(p->gcs_pmc);
> -               val32 |= no_reboot_bit(p);
> -               writel(val32, p->gcs_pmc);
> +               if (p->update_noreboot_flag)

> +                       p->update_noreboot_flag(1);

1 -> true for sake of consistency.

> +               else {
> +                       val32 = readl(p->gcs_pmc);
> +                       val32 |= no_reboot_bit(p);
> +                       writel(val32, p->gcs_pmc);
> +               }
>         } else if (p->iTCO_version == 1) {
>                 pci_read_config_dword(p->pci_dev, 0xd4, &val32);
>                 val32 |= no_reboot_bit(p);
> @@ -193,11 +199,14 @@ static int iTCO_wdt_unset_NO_REBOOT_bit(struct iTCO_wdt_private *p)
>
>         /* Unset the NO_REBOOT bit: this enables reboots */
>         if (p->iTCO_version >= 2) {
> -               val32 = readl(p->gcs_pmc);
> -               val32 &= ~enable_bit;
> -               writel(val32, p->gcs_pmc);
> -
> -               val32 = readl(p->gcs_pmc);
> +               if (p->update_noreboot_flag)

> +                       return p->update_noreboot_flag(0);

0 -> false.

> +               else {

> +                       val32 = readl(p->gcs_pmc);
> +                       val32 &= ~enable_bit;
> +                       writel(val32, p->gcs_pmc);
> +                       val32 = readl(p->gcs_pmc);

This and similar above code might be split to a helper and you may
assign it once. In such case you will not need a special flag anymore.

Helpers split might be done as a preparatory separate patch.

-- 
With Best Regards,
Andy Shevchenko