[PATCH] PCI: rockchip: Fix order of rockchip_pci_core_rsts

Jensen Huang posted 1 patch 8 months, 3 weeks ago
drivers/pci/controller/pcie-rockchip.h | 10 +++++++---
1 file changed, 7 insertions(+), 3 deletions(-)
[PATCH] PCI: rockchip: Fix order of rockchip_pci_core_rsts
Posted by Jensen Huang 8 months, 3 weeks ago
The order of rockchip_pci_core_rsts follows the previous comments suggesting
to avoid reordering. However, reset_control_bulk_deassert() applies resets in
reverse, which may lead to the link downgrading to 2.5 GT/s.

This patch restores the deassert order and comments for core_rsts, introduced in
commit 58c6990c5ee7 ("PCI: rockchip: Improve the deassert sequence of four reset pins").

Tested on NanoPC-T4 with Samsung 970 Pro.

Fixes: 18715931a5c0 ("PCI: rockchip: Simplify reset control handling by using reset_control_bulk*() function")
Signed-off-by: Jensen Huang <jensenhuang@friendlyarm.com>
---
 drivers/pci/controller/pcie-rockchip.h | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)

diff --git a/drivers/pci/controller/pcie-rockchip.h b/drivers/pci/controller/pcie-rockchip.h
index 11def598534b..4f63a03d535c 100644
--- a/drivers/pci/controller/pcie-rockchip.h
+++ b/drivers/pci/controller/pcie-rockchip.h
@@ -320,11 +320,15 @@ static const char * const rockchip_pci_pm_rsts[] = {
 	"aclk",
 };
 
+/*
+ * Please don't reorder the deassert sequence of the following
+ * four reset pins.
+ */
 static const char * const rockchip_pci_core_rsts[] = {
-	"mgmt-sticky",
-	"core",
-	"mgmt",
 	"pipe",
+	"mgmt",
+	"core",
+	"mgmt-sticky",
 };
 
 struct rockchip_pcie {
-- 
2.49.0-1
Re: [PATCH] PCI: rockchip: Fix order of rockchip_pci_core_rsts
Posted by Bjorn Helgaas 7 months, 4 weeks ago
On Fri, Mar 28, 2025 at 06:58:22PM +0800, Jensen Huang wrote:
> The order of rockchip_pci_core_rsts follows the previous comments suggesting
> to avoid reordering. However, reset_control_bulk_deassert() applies resets in
> reverse, which may lead to the link downgrading to 2.5 GT/s.
> 
> This patch restores the deassert order and comments for core_rsts, introduced in
> commit 58c6990c5ee7 ("PCI: rockchip: Improve the deassert sequence of four reset pins").
> 
> Tested on NanoPC-T4 with Samsung 970 Pro.
> 
> Fixes: 18715931a5c0 ("PCI: rockchip: Simplify reset control handling by using reset_control_bulk*() function")
> Signed-off-by: Jensen Huang <jensenhuang@friendlyarm.com>

Thanks for the fix!  It looks like 18715931a5c0 appeared in v6.14, so
we should probably add a stable tag so it gets backported there?

> ---
>  drivers/pci/controller/pcie-rockchip.h | 10 +++++++---
>  1 file changed, 7 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/pci/controller/pcie-rockchip.h b/drivers/pci/controller/pcie-rockchip.h
> index 11def598534b..4f63a03d535c 100644
> --- a/drivers/pci/controller/pcie-rockchip.h
> +++ b/drivers/pci/controller/pcie-rockchip.h
> @@ -320,11 +320,15 @@ static const char * const rockchip_pci_pm_rsts[] = {
>  	"aclk",
>  };
>  
> +/*
> + * Please don't reorder the deassert sequence of the following
> + * four reset pins.
> + */
>  static const char * const rockchip_pci_core_rsts[] = {
> -	"mgmt-sticky",
> -	"core",
> -	"mgmt",
>  	"pipe",
> +	"mgmt",
> +	"core",
> +	"mgmt-sticky",
>  };
>  
>  struct rockchip_pcie {
> -- 
> 2.49.0-1
>
Re: [PATCH] PCI: rockchip: Fix order of rockchip_pci_core_rsts
Posted by Anand Moon 8 months ago
Hi Jensen,

On Fri, 28 Mar 2025 at 16:29, Jensen Huang <jensenhuang@friendlyarm.com> wrote:
>
> The order of rockchip_pci_core_rsts follows the previous comments suggesting
> to avoid reordering. However, reset_control_bulk_deassert() applies resets in
> reverse, which may lead to the link downgrading to 2.5 GT/s.
>
> This patch restores the deassert order and comments for core_rsts, introduced in
> commit 58c6990c5ee7 ("PCI: rockchip: Improve the deassert sequence of four reset pins").
>
> Tested on NanoPC-T4 with Samsung 970 Pro.
>
> Fixes: 18715931a5c0 ("PCI: rockchip: Simplify reset control handling by using reset_control_bulk*() function")
> Signed-off-by: Jensen Huang <jensenhuang@friendlyarm.com>
> ---
>  drivers/pci/controller/pcie-rockchip.h | 10 +++++++---
>  1 file changed, 7 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/pci/controller/pcie-rockchip.h b/drivers/pci/controller/pcie-rockchip.h
> index 11def598534b..4f63a03d535c 100644
> --- a/drivers/pci/controller/pcie-rockchip.h
> +++ b/drivers/pci/controller/pcie-rockchip.h
> @@ -320,11 +320,15 @@ static const char * const rockchip_pci_pm_rsts[] = {
>         "aclk",
>  };
>
> +/*
> + * Please don't reorder the deassert sequence of the following
> + * four reset pins.
> + */
>  static const char * const rockchip_pci_core_rsts[] = {
> -       "mgmt-sticky",
> -       "core",
> -       "mgmt",
>         "pipe",
> +       "mgmt",
> +       "core",
> +       "mgmt-sticky",
>  };

Thanks for this Fix. Could you add my if you can?

Reviewed-by: Anand Moon <linux.amoon@gmail.com>

Thanks
-Anand
Re: [PATCH] PCI: rockchip: Fix order of rockchip_pci_core_rsts
Posted by Manivannan Sadhasivam 8 months ago
On Fri, 28 Mar 2025 18:58:22 +0800, Jensen Huang wrote:
> The order of rockchip_pci_core_rsts follows the previous comments suggesting
> to avoid reordering. However, reset_control_bulk_deassert() applies resets in
> reverse, which may lead to the link downgrading to 2.5 GT/s.
> 
> This patch restores the deassert order and comments for core_rsts, introduced in
> commit 58c6990c5ee7 ("PCI: rockchip: Improve the deassert sequence of four reset pins").
> 
> [...]

Applied, thanks!

[1/1] PCI: rockchip: Fix order of rockchip_pci_core_rsts
      commit: 84d79f3304645d6e87b936d2bf8b8310798efec2

Best regards,
-- 
Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Re: [PATCH] PCI: rockchip: Fix order of rockchip_pci_core_rsts
Posted by Shawn Lin 8 months, 2 weeks ago
在 2025/03/28 星期五 18:58, Jensen Huang 写道:
> The order of rockchip_pci_core_rsts follows the previous comments suggesting
> to avoid reordering. However, reset_control_bulk_deassert() applies resets in
> reverse, which may lead to the link downgrading to 2.5 GT/s.
> 
> This patch restores the deassert order and comments for core_rsts, introduced in
> commit 58c6990c5ee7 ("PCI: rockchip: Improve the deassert sequence of four reset pins").
> 
> Tested on NanoPC-T4 with Samsung 970 Pro.

Acked-by:  Shawn Lin <shawn.lin@rock-chips.com>

> 
> Fixes: 18715931a5c0 ("PCI: rockchip: Simplify reset control handling by using reset_control_bulk*() function")
> Signed-off-by: Jensen Huang <jensenhuang@friendlyarm.com>
> ---
>   drivers/pci/controller/pcie-rockchip.h | 10 +++++++---
>   1 file changed, 7 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/pci/controller/pcie-rockchip.h b/drivers/pci/controller/pcie-rockchip.h
> index 11def598534b..4f63a03d535c 100644
> --- a/drivers/pci/controller/pcie-rockchip.h
> +++ b/drivers/pci/controller/pcie-rockchip.h
> @@ -320,11 +320,15 @@ static const char * const rockchip_pci_pm_rsts[] = {
>   	"aclk",
>   };
>   
> +/*
> + * Please don't reorder the deassert sequence of the following
> + * four reset pins.
> + */
>   static const char * const rockchip_pci_core_rsts[] = {
> -	"mgmt-sticky",
> -	"core",
> -	"mgmt",
>   	"pipe",
> +	"mgmt",
> +	"core",
> +	"mgmt-sticky",
>   };
>   
>   struct rockchip_pcie {
Re: [PATCH] PCI: rockchip: Fix order of rockchip_pci_core_rsts
Posted by Manivannan Sadhasivam 8 months, 2 weeks ago
On Fri, Mar 28, 2025 at 06:58:22PM +0800, Jensen Huang wrote:
> The order of rockchip_pci_core_rsts follows the previous comments suggesting
> to avoid reordering. However, reset_control_bulk_deassert() applies resets in
> reverse, which may lead to the link downgrading to 2.5 GT/s.
> 

Oops! I failed to spot it...

> This patch restores the deassert order and comments for core_rsts, introduced in
> commit 58c6990c5ee7 ("PCI: rockchip: Improve the deassert sequence of four reset pins").
> 
> Tested on NanoPC-T4 with Samsung 970 Pro.
>

Thanks for the fix.

> Fixes: 18715931a5c0 ("PCI: rockchip: Simplify reset control handling by using reset_control_bulk*() function")
> Signed-off-by: Jensen Huang <jensenhuang@friendlyarm.com>

Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>

- Mani

-- 
மணிவண்ணன் சதாசிவம்