diff mbox series

pci: imx: disable imx6sdl LTSSM upon driver remove

Message ID 20210416203041.5035-1-tharvey@gateworks.com
State Accepted
Commit dd8c32410e47880961a867ab143e48183b4c0493
Delegated to: Stefano Babic
Headers show
Series pci: imx: disable imx6sdl LTSSM upon driver remove | expand

Commit Message

Tim Harvey April 16, 2021, 8:30 p.m. UTC
commit 6ecbe1375671 ("drivers: pci: imx: add imx_pcie_remove function")
attempted to resolve an issue caused by MX6QDL not having a proper
intneral PCIe core reset and thus hanging during kernel init if the
bootloader had enabled PCI.

The issue exists for IMX6Q, IMX6D, IXM6S, and IMX6DL. Fix the case for
IMX6S and IMX6DL getting missed.

This fixes IMX6S and IMX6DL with PCI enabled in U-Boot booting for
Linux v4.11+.

Signed-off-by: Tim Harvey <tharvey@gateworks.com>
---
 drivers/pci/pcie_imx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Comments

Fabio Estevam April 16, 2021, 8:33 p.m. UTC | #1
Hi Tim,

On Fri, Apr 16, 2021 at 5:30 PM Tim Harvey <tharvey@gateworks.com> wrote:
>
> commit 6ecbe1375671 ("drivers: pci: imx: add imx_pcie_remove function")
> attempted to resolve an issue caused by MX6QDL not having a proper
> intneral PCIe core reset and thus hanging during kernel init if the
> bootloader had enabled PCI.
>
> The issue exists for IMX6Q, IMX6D, IXM6S, and IMX6DL. Fix the case for
> IMX6S and IMX6DL getting missed.
>
> This fixes IMX6S and IMX6DL with PCI enabled in U-Boot booting for
> Linux v4.11+.
>
> Signed-off-by: Tim Harvey <tharvey@gateworks.com>

Thanks for the fix:

Reviewed-by: Fabio Estevam <festevam@gmail.com>
Fabio Estevam April 16, 2021, 8:41 p.m. UTC | #2
Hi Tim,

Some more comments.

On Fri, Apr 16, 2021 at 5:30 PM Tim Harvey <tharvey@gateworks.com> wrote:
>
> commit 6ecbe1375671 ("drivers: pci: imx: add imx_pcie_remove function")
> attempted to resolve an issue caused by MX6QDL not having a proper
> intneral PCIe core reset and thus hanging during kernel init if the

"internal"

> --- a/drivers/pci/pcie_imx.c
> +++ b/drivers/pci/pcie_imx.c
> @@ -473,7 +473,7 @@ static int imx6_pcie_assert_core_reset(struct imx_pcie_priv *priv,
>          * If both LTSSM_ENABLE and REF_SSP_ENABLE are active we have a strong
>          * indication that the bootloader activated the link.
>          */
> -       if (is_mx6dq() && prepare_for_boot) {
> +       if ((is_mx6dq() || is_mx6sdl()) && prepare_for_boot) {

It seems that i.MX6QP is missing. What about adding is_mx6dqp() too?
Tim Harvey April 16, 2021, 8:48 p.m. UTC | #3
On Fri, Apr 16, 2021 at 1:41 PM Fabio Estevam <festevam@gmail.com> wrote:
>
> Hi Tim,
>
> Some more comments.
>
> On Fri, Apr 16, 2021 at 5:30 PM Tim Harvey <tharvey@gateworks.com> wrote:
> >
> > commit 6ecbe1375671 ("drivers: pci: imx: add imx_pcie_remove function")
> > attempted to resolve an issue caused by MX6QDL not having a proper
> > intneral PCIe core reset and thus hanging during kernel init if the
>
> "internal"
>

Fabio,

Thanks, I will correct this for a v2

> > --- a/drivers/pci/pcie_imx.c
> > +++ b/drivers/pci/pcie_imx.c
> > @@ -473,7 +473,7 @@ static int imx6_pcie_assert_core_reset(struct imx_pcie_priv *priv,
> >          * If both LTSSM_ENABLE and REF_SSP_ENABLE are active we have a strong
> >          * indication that the bootloader activated the link.
> >          */
> > -       if (is_mx6dq() && prepare_for_boot) {
> > +       if ((is_mx6dq() || is_mx6sdl()) && prepare_for_boot) {
>
> It seems that i.MX6QP is missing. What about adding is_mx6dqp() too?

The QP has a soft reset bit such that this isn't an issue. For
reference I'm basing this off the kernel 'fixup' that was removed in
v4.11 which can be found here:
https://elixir.bootlin.com/linux/v4.10.17/source/drivers/pci/host/pci-imx6.c#L269

I'm the one who authored the original patch and it was my mistake way
back then to use 'is_mx6dq()' in U-Boot thinking that it covered all
the early IMX6's that lacked a PCIe soft reset.

Tim
Fabio Estevam April 16, 2021, 9:12 p.m. UTC | #4
On Fri, Apr 16, 2021 at 5:48 PM Tim Harvey <tharvey@gateworks.com> wrote:

> The QP has a soft reset bit such that this isn't an issue. For

Yes, I missed that the i.MX6QP has fixed this issue.

Thanks
Stefano Babic May 2, 2021, 11:43 a.m. UTC | #5
> commit 6ecbe1375671 ("drivers: pci: imx: add imx_pcie_remove function")
> attempted to resolve an issue caused by MX6QDL not having a proper
> intneral PCIe core reset and thus hanging during kernel init if the
> bootloader had enabled PCI.
> The issue exists for IMX6Q, IMX6D, IXM6S, and IMX6DL. Fix the case for
> IMX6S and IMX6DL getting missed.
> This fixes IMX6S and IMX6DL with PCI enabled in U-Boot booting for
> Linux v4.11+.
> Signed-off-by: Tim Harvey <tharvey@gateworks.com>
> Reviewed-by: Fabio Estevam <festevam@gmail.com>
Applied to u-boot-imx, master, thanks !

Best regards,
Stefano Babic
diff mbox series

Patch

diff --git a/drivers/pci/pcie_imx.c b/drivers/pci/pcie_imx.c
index d9c2325a22..73875e00db 100644
--- a/drivers/pci/pcie_imx.c
+++ b/drivers/pci/pcie_imx.c
@@ -473,7 +473,7 @@  static int imx6_pcie_assert_core_reset(struct imx_pcie_priv *priv,
 	 * If both LTSSM_ENABLE and REF_SSP_ENABLE are active we have a strong
 	 * indication that the bootloader activated the link.
 	 */
-	if (is_mx6dq() && prepare_for_boot) {
+	if ((is_mx6dq() || is_mx6sdl()) && prepare_for_boot) {
 		u32 val, gpr1, gpr12;
 
 		gpr1 = readl(&iomuxc_regs->gpr[1]);