> -----Original Message-----
> From: Tim Harvey <tharvey@xxxxxxxxxxxxx>
> Sent: 2025年6月24日 1:33
> To: Manivannan Sadhasivam <mani@xxxxxxxxxx>
> Cc: Hongxing Zhu <hongxing.zhu@xxxxxxx>; Frank Li <frank.li@xxxxxxx>;
> l.stach@xxxxxxxxxxxxxx; lpieralisi@xxxxxxxxxx; kwilczynski@xxxxxxxxxx;
> robh@xxxxxxxxxx; bhelgaas@xxxxxxxxxx; shawnguo@xxxxxxxxxx;
> s.hauer@xxxxxxxxxxxxxx; kernel@xxxxxxxxxxxxxx; festevam@xxxxxxxxx;
> linux-pci@xxxxxxxxxxxxxxx; linux-arm-kernel@xxxxxxxxxxxxxxxxxxx;
> imx@xxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx
> Subject: Re: [PATCH v3 1/2] PCI: imx6: Remove apps_reset toggle in
> _core_reset functions
>
> On Mon, Jun 23, 2025 at 4:42 AM Manivannan Sadhasivam <mani@xxxxxxxxxx>
> wrote:
> >
> > On Mon, Jun 16, 2025 at 04:57:41PM +0800, Richard Zhu wrote:
> > > apps_reset is LTSSM_EN on i.MX7, i.MX8MQ, i.MX8MM and i.MX8MP
> platforms.
> > > Since the assertion/de-assertion of apps_reset(LTSSM_EN bit) had
> > > been wrappered in imx_pcie_ltssm_enable() and
> > > imx_pcie_ltssm_disable();
> > >
> >
> > What about other i.MX chipsets like 6Q and its cousins? Wouldn't this
> > change affect them since they treat 'apps_reset' differently?
> >
> > - Mani
Hi Mani:
Sorry to reply late.
Only i.MX7D, i.MX8MQ, i.MX8MM, and i.MX8MP have the apps_reset. No problems
are found with this change in my local tests on these four platforms.
With this change, the assertion/deassertion of ltssm_en bit are unified into
imx_pcie_ltssm_enable() and imx_pcie_ltssm_disable() functions, aligned with
other i.MX platforms.
Best Regards
Richard Zhu
>
> Hi Main,
>
> This patch effectively brings back the behavior prior to commit
> ef61c7d8d032 ("PCI: imx6: Deassert apps_reset in
> imx_pcie_deassert_core_reset()") which caused the original regressions.
>
> To ease your concerns I've tested this patch on top of v6.16-rc3 with the
> following IMX6 boards I have here with and without a PCI device
> attached:
> imx6q-gw51xx - no switch
> imx6q-gw54xx - switch
>
> I only have imx6qdl/imx8mm/imx8mp boards to test with.
>
> From what I can tell it doesn't look like the original patch that added the
> 'symmetric' apps_reset de-assert was necessarily well tested. It started out
> being added because as far as I can tell it 'looked' like the right thing to do [1].
> You requested changes to the commit log for wording [2],[3] but I'm unclear
> that anyone tested this.
>
Hi Tim:
Thanks for your explains and tests.
Best Regards
Richard Zhu
> Best Regards,
>
> Tim
> [1]
> https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fpatch
> work.kernel.org%2Fproject%2Flinux-pci%2Fpatch%2F1727148464-14341-6-gi
> t-send-email-hongxing.zhu%40nxp.com%2F&data=05%7C02%7Chongxing.zhu
> %40nxp.com%7Cdac60212c1c94cb24d7508ddb27c0f92%7C686ea1d3bc2b4c
> 6fa92cd99c5c301635%7C0%7C0%7C638862968081967795%7CUnknown%7
> CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJ
> XaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=8I
> 7qGI92BfsbAERknsVjZO6cI527Enxgiiw%2FVatI7h4%3D&reserved=0
> [2]
> https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fpatch
> work.kernel.org%2Fproject%2Flinux-pci%2Fpatch%2F1728981213-8771-6-git
> -send-email-hongxing.zhu%40nxp.com%2F&data=05%7C02%7Chongxing.zhu
> %40nxp.com%7Cdac60212c1c94cb24d7508ddb27c0f92%7C686ea1d3bc2b4c
> 6fa92cd99c5c301635%7C0%7C0%7C638862968081993466%7CUnknown%7
> CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJ
> XaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=lh
> ER8F9ApfpgQcVCyMOBYho%2BXvlp79re4jX5C5gj1XY%3D&reserved=0
> [3]
> https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fpatch
> work.kernel.org%2Fproject%2Flinux-pci%2Fpatch%2F20241101070610.1267
> 391-6-hongxing.zhu%40nxp.com%2F&data=05%7C02%7Chongxing.zhu%40nx
> p.com%7Cdac60212c1c94cb24d7508ddb27c0f92%7C686ea1d3bc2b4c6fa92c
> d99c5c301635%7C0%7C0%7C638862968082007982%7CUnknown%7CTWFp
> bGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4z
> MiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&sdata=kp4EqnV
> lfUsVq4k9UV33LSpiwn%2F2OJlPzu2PApAttNs%3D&reserved=0
