| 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
| ||
|
Message-ID: <13640a07-7395-4521-9c5d-748599202361@roeck-us.net>
Date: Mon, 18 Mar 2024 17:46:48 -0700
From: Guenter Roeck <linux@...ck-us.net>
To: Peter Yin <peteryin.openbmc@...il.com>, patrick@...cx.xyz,
Rob Herring <robh+dt@...nel.org>,
Krzysztof Kozlowski <krzysztof.kozlowski+dt@...aro.org>,
Conor Dooley <conor+dt@...nel.org>, Joel Stanley <joel@....id.au>,
Andrew Jeffery <andrew@...econstruct.com.au>,
Wim Van Sebroeck <wim@...ux-watchdog.org>, devicetree@...r.kernel.org,
linux-arm-kernel@...ts.infradead.org, linux-aspeed@...ts.ozlabs.org,
linux-kernel@...r.kernel.org, linux-watchdog@...r.kernel.org
Subject: Re: [PATCH v2] drivers: watchdog: ast2600 support bootstatus
On 3/17/24 22:52, Peter Yin wrote:
> Add WDIOF_EXTERN1 and WDIOF_CARDRESET bootstatus in ast2600
>
> Regarding the AST2600 specification, the WDTn Timeout Status Register
> (WDT10) has bit 1 reserved. To verify the second boot source,
> we need to check SEC14 bit 12 and bit 13.
> The bits 8-23 in the WDTn Timeout Status Register are the Watchdog
> Event Count, which we can use to verify WDIOF_EXTERN1.
>
> Signed-off-by: Peter Yin <peteryin.openbmc@...il.com>
You'll have to separate dts and yaml file changes from driver changes.
> ---
> Change log:
>
> v1 -> v2
> - Add comment and support WDIOF_CARDRESET in ast2600
>
> v1
> - Patch 0001 - Add WDIOF_EXTERN1 bootstatus
> ---
> arch/arm/boot/dts/aspeed/aspeed-g6.dtsi | 8 ++---
> drivers/watchdog/aspeed_wdt.c | 45 ++++++++++++++++++++++---
> 2 files changed, 44 insertions(+), 9 deletions(-)
>
> diff --git a/arch/arm/boot/dts/aspeed/aspeed-g6.dtsi b/arch/arm/boot/dts/aspeed/aspeed-g6.dtsi
> index e0b44498269f..23ae7f0430e9 100644
> --- a/arch/arm/boot/dts/aspeed/aspeed-g6.dtsi
> +++ b/arch/arm/boot/dts/aspeed/aspeed-g6.dtsi
> @@ -556,24 +556,24 @@ uart5: serial@...84000 {
>
> wdt1: watchdog@...85000 {
> compatible = "aspeed,ast2600-wdt";
> - reg = <0x1e785000 0x40>;
> + reg = <0x1e785000 0x40>, <0x1e6f2000 0x20>;
> };
>
> wdt2: watchdog@...85040 {
> compatible = "aspeed,ast2600-wdt";
> - reg = <0x1e785040 0x40>;
> + reg = <0x1e785040 0x40>, <0x1e6f2000 0x020>;
> status = "disabled";
> };
>
> wdt3: watchdog@...85080 {
> compatible = "aspeed,ast2600-wdt";
> - reg = <0x1e785080 0x40>;
> + reg = <0x1e785080 0x40>, <0x1e6f2000 0x020>;
> status = "disabled";
> };
>
> wdt4: watchdog@...850c0 {
> compatible = "aspeed,ast2600-wdt";
> - reg = <0x1e7850C0 0x40>;
> + reg = <0x1e7850C0 0x40>, <0x1e6f2000 0x020>;
> status = "disabled";
> };
>
> diff --git a/drivers/watchdog/aspeed_wdt.c b/drivers/watchdog/aspeed_wdt.c
> index b4773a6aaf8c..65118e461130 100644
> --- a/drivers/watchdog/aspeed_wdt.c
> +++ b/drivers/watchdog/aspeed_wdt.c
> @@ -33,6 +33,7 @@ struct aspeed_wdt {
> void __iomem *base;
> u32 ctrl;
> const struct aspeed_wdt_config *cfg;
> + void __iomem *sec_base;
> };
>
> static const struct aspeed_wdt_config ast2400_config = {
> @@ -82,6 +83,15 @@ MODULE_DEVICE_TABLE(of, aspeed_wdt_of_table);
> #define WDT_RESET_MASK1 0x1c
> #define WDT_RESET_MASK2 0x20
>
> +/*
> + * Only Ast2600 support
> + */
> +#define WDT_EVENT_COUNTER_MASK (0xFFF << 8)
> +#define WDT_SECURE_ENGINE_STATUS (0x14)
> +#define ABR_IMAGE_SOURCE BIT(12)
> +#define ABR_IMAGE_SOURCE_SPI BIT(13)
> +#define SECOND_BOOT_ENABLE BIT(14)
> +
> /*
> * WDT_RESET_WIDTH controls the characteristics of the external pulse (if
> * enabled), specifically:
> @@ -313,6 +323,7 @@ static int aspeed_wdt_probe(struct platform_device *pdev)
> const char *reset_type;
> u32 duration;
> u32 status;
> + u32 sec_st;
> int ret;
>
> wdt = devm_kzalloc(dev, sizeof(*wdt), GFP_KERNEL);
> @@ -330,6 +341,12 @@ static int aspeed_wdt_probe(struct platform_device *pdev)
> if (IS_ERR(wdt->base))
> return PTR_ERR(wdt->base);
>
> + if (of_device_is_compatible(np, "aspeed,ast2600-wdt")) {
> + wdt->sec_base = devm_platform_ioremap_resource(pdev, 1);
> + if (IS_ERR(wdt->sec_base))
> + return PTR_ERR(wdt->sec_base);
> + }
> +
> wdt->wdd.info = &aspeed_wdt_info;
>
> if (wdt->cfg->irq_mask) {
> @@ -459,12 +476,30 @@ static int aspeed_wdt_probe(struct platform_device *pdev)
> }
>
> status = readl(wdt->base + WDT_TIMEOUT_STATUS);
> - if (status & WDT_TIMEOUT_STATUS_BOOT_SECONDARY) {
> - wdt->wdd.bootstatus = WDIOF_CARDRESET;
>
> - if (of_device_is_compatible(np, "aspeed,ast2400-wdt") ||
> - of_device_is_compatible(np, "aspeed,ast2500-wdt"))
> - wdt->wdd.groups = bswitch_groups;
> + if (of_device_is_compatible(np, "aspeed,ast2600-wdt")) {
> + /*
> + * The WDTn Timeout Status Register bit 1 is reserved.
> + * To verify the second boot source,
> + * we need to check SEC14 bit 12 and bit 13.
> + */
> + sec_st = readl(wdt->sec_base + WDT_SECURE_ENGINE_STATUS);
> + if( sec_st & SECOND_BOOT_ENABLE)
> + if (sec_st & ABR_IMAGE_SOURCE ||
> + sec_st & ABR_IMAGE_SOURCE_SPI)
I am sure that checkpatch as something to say here. Either case, I would very
much prefer a single if() statement such as
if (sec_st & SECOND_BOOT_ENABLE &&
sec_st & (ABR_IMAGE_SOURCE | ABR_IMAGE_SOURCE_SPI))
> + wdt->wdd.bootstatus |= WDIOF_CARDRESET;
> +
> + /*
> + * To check Watchdog Event Count for WDIOF_EXTERN1
> + */
> + if (status & WDT_EVENT_COUNTER_MASK) {
> + wdt->wdd.bootstatus |= WDIOF_EXTERN1;
> + }
Unnecessary { }
.. but does this really indicate that there was a reset due to some event ?
This reads three 8-bit counters. Wouldn't it make more sense to check bit 0
instead ?
I am also not sure if reading the watchdog status from WDT_SECURE_ENGINE_STATUS
adds any value over the status reported in the watchdog status register.
You'll have to explain why the added complexity is necessary or even adds
value.
Never mind, though ...
Looking into the datasheets, the current code is quite completely wrong anyway.
Bit 1 of the status register indicates on ast2500 if the boot was from the second
boot source. It does not indicate that the most recent reset was triggered by
the watchdog. The code should just be changed to set WDIOF_CARDRESET if bit 0
of the status register is set. The boot source is out of scope for the watchdog
status bits.
Thanks,
Guenter
Powered by blists - more mailing lists