Some bug fixes when DWC generic suspend/resume callbacks are used.
Drop the first patch of v3 patch-set, since the use case of this patch had been
covered by #3 commit "PCI: dwc: Clean up some unnecessary codes in dw_pcie_suspend_noirq()".
To be simple, re-format the codes, drop the first patch of v3 patch-set, and
only keep last two patches of v3 in v4 or later
Here is the discussion [1] and final solution [2] of the codes clean up commit.
[1] https://patchwork.kernel.org/project/linux-pci/patch/1721628913-1449-1-git-send-email-hongxing.zhu@nxp.com/
[2] https://patchwork.kernel.org/project/linux-pci/patch/20241126073909.4058733-1-hongxing.zhu@nxp.com/
v4 changes:
Drop the first patch("PCI: dwc: Fix resume failure if no EP is connected on some platforms")
in v3, since it's use-case had been covered by #3 patch of v3.
Add one Fixes tag into "PCI: dwc: Clean up some unnecessary codes in dw_pcie_suspend_noirq()".
Refer to Damien's comments, let ret test go inside the "else" and remove the
initialization of ret to 0 declaration. Thanks.
v3 changes:
Regarding the discussion listed above[2].
Resend the patch-set after adding one more codes clean up patch together.
v2 changes:
Thanks for Manivannan's review.
- Refine the subject of second patch and add
"Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>"
- In first patch, update the commit message and move one comment into
proper place.
BTW, Manivannan found another potential issue that suspend is entry but
the link is in L1SS stat in v1 review. This is a new story. And it's better
to be verified and fixed by another commit later.
v2: https://patchwork.kernel.org/project/linux-pci/cover/1728539269-1861-1-git-send-email-hongxing.zhu@nxp.com/
[PATCH v4 1/2] PCI: dwc: Always stop link in the
[PATCH v4 2/2] PCI: dwc: Clean up some unnecessary codes in
drivers/pci/controller/dwc/pcie-designware-host.c | 28 ++++++++++++++++++----------
drivers/pci/controller/dwc/pcie-designware.h | 1 +
2 files changed, 19 insertions(+), 10 deletions(-)