Linux PCI subsystem development
 help / color / mirror / Atom feed
* [PATCH] PCI: fu740: Check return value of clk_prepare_enable()
@ 2026-11-16 20:49 Jayaraj Devaraj
  2026-11-16 21:03 ` sashiko-bot
  0 siblings, 1 reply; 2+ messages in thread
From: Jayaraj Devaraj @ 2026-11-16 20:49 UTC (permalink / raw)
  To: Manivannan Sadhasivam, Krzysztof Wilczyński,
	Lorenzo Pieralisi, Bjorn Helgaas
  Cc: Paul Walmsley, Greentime Hu, Samuel Holland, Rob Herring,
	Philipp Zabel, linux-pci, linux-riscv, linux-kernel,
	Jayaraj Devaraj

fu740_pcie_host_init() ignores the return value of the second
clk_prepare_enable() call on pcie_aux, so a failure to enable the
clock is silently ignored and the controller is programmed anyway.

Check the return value and propagate the error.

Signed-off-by: Jayaraj Devaraj <jayarajdmit@gmail.com>
---
 drivers/pci/controller/dwc/pcie-fu740.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/drivers/pci/controller/dwc/pcie-fu740.c b/drivers/pci/controller/dwc/pcie-fu740.c
index 00ec69a09164..b2eded23b52c 100644
--- a/drivers/pci/controller/dwc/pcie-fu740.c
+++ b/drivers/pci/controller/dwc/pcie-fu740.c
@@ -270,7 +270,11 @@ static int fu740_pcie_host_init(struct dw_pcie_rp *pp)
 	/* Clear hold_phy_rst */
 	writel_relaxed(0x0, afp->mgmt_base + PCIEX8MGMT_APP_HOLD_PHY_RST);
 	/* Enable pcieauxclk */
-	clk_prepare_enable(afp->pcie_aux);
+	ret = clk_prepare_enable(afp->pcie_aux);
+	if (ret) {
+		dev_err(dev, "unable to re-enable pcie_aux clock\n");
+		return ret;
+	}
 	/* Set RC mode */
 	writel_relaxed(0x4, afp->mgmt_base + PCIEX8MGMT_DEVICE_TYPE);
 
-- 
2.34.1


^ permalink raw reply related	[flat|nested] 2+ messages in thread

* Re: [PATCH] PCI: fu740: Check return value of clk_prepare_enable()
  2026-11-16 20:49 [PATCH] PCI: fu740: Check return value of clk_prepare_enable() Jayaraj Devaraj
@ 2026-11-16 21:03 ` sashiko-bot
  0 siblings, 0 replies; 2+ messages in thread
From: sashiko-bot @ 2026-11-16 21:03 UTC (permalink / raw)
  To: Jayaraj Devaraj; +Cc: linux-pci

Thank you for your contribution! Sashiko AI review found 1 potential
issue to consider:
- [Low] PCI: fu740: missing Fixes trailer forfu740_pcie_host_init()
  clock check
It also noticed 1 related pre-existing issue this patch does not
introduce:
- [Medium] PCI: fu740: missing .deinit cleanup and unchecked
  clk_prepare_enable() [1]
--

commit 04ba94c4c15cc3b4e7a1a6837f3336a1c0b55bce
Author: Jayaraj Devaraj <jayarajdmit@gmail.com>

> PCI: fu740: Check return value of clk_prepare_enable()
>
> fu740_pcie_host_init() ignores the return value of the second
> clk_prepare_enable() call on pcie_aux, so a failure to enable the
> clock is silently ignored and the controller is programmed anyway.
>
> Check the return value and propagate the error.
>
> Signed-off-by: Jayaraj Devaraj <jayarajdmit@gmail.com>

[Severity: Low]
Since this fixes a missing error check in fu740_pcie_host_init(), should it
include:

  Fixes: e7e21b3a339b ("PCI: fu740: Add SiFive FU740 PCIe host controller driver")

> diff --git a/drivers/pci/controller/dwc/pcie-fu740.c b/drivers/pci/controller/dwc/pcie-fu740.c
> index 4908077e58c4c..c9bb4c4fba23f 100644
> --- a/drivers/pci/controller/dwc/pcie-fu740.c
> +++ b/drivers/pci/controller/dwc/pcie-fu740.c
> @@ -270,7 +270,11 @@ static int fu740_pcie_host_init(struct dw_pcie_rp *pp)
>  	/* Clear hold_phy_rst */
>  	writel_relaxed(0x0, afp->mgmt_base + PCIEX8MGMT_APP_HOLD_PHY_RST);
>  	/* Enable pcieauxclk */
> -	clk_prepare_enable(afp->pcie_aux);
> +	ret = clk_prepare_enable(afp->pcie_aux);
> +	if (ret) {
> +		dev_err(dev, "unable to re-enable pcie_aux clock\n");
> +		return ret;
> +	}
>  	/* Set RC mode */
>  	writel_relaxed(0x4, afp->mgmt_base + PCIEX8MGMT_DEVICE_TYPE);
>  

[Severity: Medium]
This problem wasn't introduced by this patch, but does returning here or
failing later in dw_pcie_host_init() leak the clock and reset state
[1]?

Earlier in fu740_pcie_host_init(), reset_control_deassert(afp->rst)
deasserts the reset line, so returning when the second
clk_prepare_enable(afp->pcie_aux) fails leaves afp->rst deasserted.
In addition, fu740_pcie_host_ops has no .deinit callback, so if a later
step in dw_pcie_host_init() fails after fu740_pcie_host_init() succeeds,
afp->pcie_aux remains enabled and afp->rst remains deasserted.

[1] https://sashiko.dev/#/bug/linux-e8a8fccf-83d6-4777-a6b0-51ba01cc3409

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261116204922.159384-1-jayarajdmit@gmail.com?part=1

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2026-11-16 21:03 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-11-16 20:49 [PATCH] PCI: fu740: Check return value of clk_prepare_enable() Jayaraj Devaraj
2026-11-16 21:03 ` sashiko-bot

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox