Re: [PATCH] thermal: intel: Fix unchecked MSR issue

From: Rafael J. Wysocki
Date: Tue Apr 11 2023 - 12:18:19 EST


On Mon, Apr 10, 2023 at 7:35 PM Srinivas Pandruvada
<srinivas.pandruvada@xxxxxxxxxxxxxxx> wrote:
>
> Some older processors don't allow BIT(13) and BIT(15) in the current
> mask set by "THERM_STATUS_CLEAR_CORE_MASK". This results in:
>
> unchecked MSR access error: WRMSR to 0x19c (tried to
> write 0x000000000000aaa8) at rIP: 0xffffffff816f66a6
> (throttle_active_work+0xa6/0x1d0)
>
> To avoid unchecked MSR issues, check cpuid for each feature and then
> form core mask. Do the same for package mask set by
> "THERM_STATUS_CLEAR_PKG_MASK".
>
> Introduce functions thermal_intr_core_clear_mask() and
> thermal_intr_pkg_clear_mask()

I've renamed these two functions to
thermal_intr_init_core_clear_mask() and
thermal_intr_init_pkg_clear_mask(), respectively.

> to set core and package mask respectively.
> These functions are called during initialization.
>
> Fixes: 6fe1e64b6026 ("thermal: intel: Prevent accidental clearing of HFI status")
> Reported-by: Rui Salvaterra <rsalvaterra@xxxxxxxxx>
> Link: https://lore.kernel.org/lkml/cdf43fb423368ee3994124a9e8c9b4f8d00712c6.camel@xxxxxxxxxxxxxxx/T/
> Tested-by: Rui Salvaterra <rsalvaterra@xxxxxxxxx>
> Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@xxxxxxxxxxxxxxx>
> Cc: stable@xxxxxxxxxx # 6.2+
> ---
> drivers/thermal/intel/therm_throt.c | 73 ++++++++++++++++++++++++++---
> 1 file changed, 66 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/thermal/intel/therm_throt.c b/drivers/thermal/intel/therm_throt.c
> index 2e22bb82b738..d5047676f3d2 100644
> --- a/drivers/thermal/intel/therm_throt.c
> +++ b/drivers/thermal/intel/therm_throt.c
> @@ -193,8 +193,67 @@ static const struct attribute_group thermal_attr_group = {
> #define THERM_THROT_POLL_INTERVAL HZ
> #define THERM_STATUS_PROCHOT_LOG BIT(1)
>
> -#define THERM_STATUS_CLEAR_CORE_MASK (BIT(1) | BIT(3) | BIT(5) | BIT(7) | BIT(9) | BIT(11) | BIT(13) | BIT(15))
> -#define THERM_STATUS_CLEAR_PKG_MASK (BIT(1) | BIT(3) | BIT(5) | BIT(7) | BIT(9) | BIT(11))
> +static u64 def_therm_core_clear_mask;
> +static u64 def_therm_pkg_clear_mask;

And I've renamed these two variables to therm_intr_core_clear_mask and
therm_intr_pkg_clear_mask, respectively.

Also I've changed the subject (to "thermal: intel: Avoid updating
unsupported THERM_STATUS_CLEAR mask bits") and made some assorted
changelog edits.

With the above changes, the patch has been queued up for 6.3-rc7.

> +
> +static void thermal_intr_core_clear_mask(void)
> +{
> + if (def_therm_core_clear_mask)
> + return;
> +
> + /*
> + * Reference: Intel SDM Volume 4
> + * "Table 2-2. IA-32 Architectural MSRs", MSR 0x19C
> + * IA32_THERM_STATUS.
> + */
> +
> + /*
> + * Bit 1, 3, 5: CPUID.01H:EDX[22] = 1. This driver will not
> + * enable interrupts, when 0 as it checks for X86_FEATURE_ACPI.
> + */
> + def_therm_core_clear_mask = (BIT(1) | BIT(3) | BIT(5));
> +
> + /*
> + * Bit 7 and 9: Thermal Threshold #1 and #2 log
> + * If CPUID.01H:ECX[8] = 1
> + */
> + if (boot_cpu_has(X86_FEATURE_TM2))
> + def_therm_core_clear_mask |= (BIT(7) | BIT(9));
> +
> + /* Bit 11: Power Limitation log (R/WC0) If CPUID.06H:EAX[4] = 1 */
> + if (boot_cpu_has(X86_FEATURE_PLN))
> + def_therm_core_clear_mask |= BIT(11);
> +
> + /*
> + * Bit 13: Current Limit log (R/WC0) If CPUID.06H:EAX[7] = 1
> + * Bit 15: Cross Domain Limit log (R/WC0) If CPUID.06H:EAX[7] = 1
> + */
> + if (boot_cpu_has(X86_FEATURE_HWP))
> + def_therm_core_clear_mask |= (BIT(13) | BIT(15));
> +}
> +
> +static void thermal_intr_pkg_clear_mask(void)
> +{
> + if (def_therm_pkg_clear_mask)
> + return;
> +
> + /*
> + * Reference: Intel SDM Volume 4
> + * "Table 2-2. IA-32 Architectural MSRs", MSR 0x1B1
> + * IA32_PACKAGE_THERM_STATUS.
> + */
> +
> + /* All bits except BIT 26 depends on CPUID.06H: EAX[6] = 1 */
> + if (boot_cpu_has(X86_FEATURE_PTS))
> + def_therm_pkg_clear_mask = (BIT(1) | BIT(3) | BIT(5) | BIT(7) | BIT(9) | BIT(11));
> +
> + /*
> + * Intel SDM Volume 2A: Thermal and Power Management Leaf
> + * Bit 26: CPUID.06H: EAX[19] = 1
> + */
> + if (boot_cpu_has(X86_FEATURE_HFI))
> + def_therm_pkg_clear_mask |= BIT(26);
> +}
>
> /*
> * Clear the bits in package thermal status register for bit = 1
> @@ -207,13 +266,10 @@ void thermal_clear_package_intr_status(int level, u64 bit_mask)
>
> if (level == CORE_LEVEL) {
> msr = MSR_IA32_THERM_STATUS;
> - msr_val = THERM_STATUS_CLEAR_CORE_MASK;
> + msr_val = def_therm_core_clear_mask;
> } else {
> msr = MSR_IA32_PACKAGE_THERM_STATUS;
> - msr_val = THERM_STATUS_CLEAR_PKG_MASK;
> - if (boot_cpu_has(X86_FEATURE_HFI))
> - msr_val |= BIT(26);
> -
> + msr_val = def_therm_pkg_clear_mask;
> }
>
> msr_val &= ~bit_mask;
> @@ -708,6 +764,9 @@ void intel_init_thermal(struct cpuinfo_x86 *c)
> h = THERMAL_APIC_VECTOR | APIC_DM_FIXED | APIC_LVT_MASKED;
> apic_write(APIC_LVTTHMR, h);
>
> + thermal_intr_core_clear_mask();
> + thermal_intr_pkg_clear_mask();
> +
> rdmsr(MSR_IA32_THERM_INTERRUPT, l, h);
> if (cpu_has(c, X86_FEATURE_PLN) && !int_pln_enable)
> wrmsr(MSR_IA32_THERM_INTERRUPT,
> --
> 2.39.1
>