PCI: dwc: Advertise L1 PM Substates only if driver requests it

L1 PM Substates require the CLKREQ# signal and may also require
device-specific support.  If CLKREQ# is not supported or driver support is
lacking, enabling L1.1 or L1.2 may cause errors when accessing devices,
e.g.,

  nvme nvme0: controller is down; will reset: CSTS=0xffffffff, PCI_STATUS=0x10

If the kernel is built with CONFIG_PCIEASPM_POWER_SUPERSAVE=y or users
enable L1.x via sysfs, users may trip over these errors even if L1
Substates haven't been enabled by firmware or the driver.

To prevent such errors, disable advertising the L1 PM Substates unless the
driver sets "dw_pcie.l1ss_support" to indicate that it knows CLKREQ# is
present and any device-specific configuration has been done.

Set "dw_pcie.l1ss_support" in tegra194 (if DT includes the
"supports-clkreq' property) and qcom (for cfg_2_7_0, cfg_1_9_0, cfg_1_34_0,
and cfg_sc8280xp controllers) so they can continue to use L1 Substates.

Based on Niklas's patch:
https://patch.msgid.link/20251017163252.598812-2-cassel@kernel.org

[bhelgaas: drop hiding for endpoints]
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Link: https://patch.msgid.link/20251118214312.2598220-2-helgaas@kernel.org

drivers/pci/controller/dwc/pcie-designware-host.c

@@ -1060,6 +1060,8 @@ int dw_pcie_setup_rc(struct dw_pcie_rp *pp)
 		PCI_COMMAND_MASTER | PCI_COMMAND_SERR;
 	dw_pcie_writel_dbi(pci, PCI_COMMAND, val);
 
+	dw_pcie_hide_unsupported_l1ss(pci);
+
 	dw_pcie_config_presets(pp);
 	/*
 	 * If the platform provides its own child bus config accesses, it means

drivers/pci/controller/dwc/pcie-designware.c

@@ -1081,6 +1081,30 @@ void dw_pcie_edma_remove(struct dw_pcie *pci)
 	dw_edma_remove(&pci->edma);
 }
 
+void dw_pcie_hide_unsupported_l1ss(struct dw_pcie *pci)
+{
+	u16 l1ss;
+	u32 l1ss_cap;
+
+	if (pci->l1ss_support)
+		return;
+
+	l1ss = dw_pcie_find_ext_capability(pci, PCI_EXT_CAP_ID_L1SS);
+	if (!l1ss)
+		return;
+
+	/*
+	 * Unless the driver claims "l1ss_support", don't advertise L1 PM
+	 * Substates because they require CLKREQ# and possibly other
+	 * device-specific configuration.
+	 */
+	l1ss_cap = dw_pcie_readl_dbi(pci, l1ss + PCI_L1SS_CAP);
+	l1ss_cap &= ~(PCI_L1SS_CAP_PCIPM_L1_1 | PCI_L1SS_CAP_ASPM_L1_1 |
+		      PCI_L1SS_CAP_PCIPM_L1_2 | PCI_L1SS_CAP_ASPM_L1_2 |
+		      PCI_L1SS_CAP_L1_PM_SS);
+	dw_pcie_writel_dbi(pci, l1ss + PCI_L1SS_CAP, l1ss_cap);
+}
+
 void dw_pcie_setup(struct dw_pcie *pci)
 {
 	u32 val;

drivers/pci/controller/dwc/pcie-designware.h

@@ -516,6 +516,7 @@ struct dw_pcie {
 	int			max_link_speed;
 	u8			n_fts[2];
 	struct dw_edma_chip	edma;
+	bool			l1ss_support;	/* L1 PM Substates support */
 	struct clk_bulk_data	app_clks[DW_PCIE_NUM_APP_CLKS];
 	struct clk_bulk_data	core_clks[DW_PCIE_NUM_CORE_CLKS];
 	struct reset_control_bulk_data	app_rsts[DW_PCIE_NUM_APP_RSTS];
@@ -573,6 +574,7 @@ int dw_pcie_prog_ep_inbound_atu(struct dw_pcie *pci, u8 func_no, int index,
 				int type, u64 parent_bus_addr,
 				u8 bar, size_t size);
 void dw_pcie_disable_atu(struct dw_pcie *pci, u32 dir, int index);
+void dw_pcie_hide_unsupported_l1ss(struct dw_pcie *pci);
 void dw_pcie_setup(struct dw_pcie *pci);
 void dw_pcie_iatu_detect(struct dw_pcie *pci);
 int dw_pcie_edma_detect(struct dw_pcie *pci);

drivers/pci/controller/dwc/pcie-qcom.c

@@ -1067,6 +1067,8 @@ static int qcom_pcie_init_2_7_0(struct qcom_pcie *pcie)
 	val &= ~REQ_NOT_ENTR_L1;
 	writel(val, pcie->parf + PARF_PM_CTRL);
 
+	pci->l1ss_support = true;
+
 	val = readl(pcie->parf + PARF_AXI_MSTR_WR_ADDR_HALT_V2);
 	val |= EN;
 	writel(val, pcie->parf + PARF_AXI_MSTR_WR_ADDR_HALT_V2);

drivers/pci/controller/dwc/pcie-tegra194.c

@@ -703,6 +703,9 @@ static void init_host_aspm(struct tegra_pcie_dw *pcie)
 	val |= (pcie->aspm_pwr_on_t << 19);
 	dw_pcie_writel_dbi(pci, pcie->cfg_link_cap_l1sub, val);
 
+	if (pcie->supports_clkreq)
+		pci->l1ss_support = true;
+
 	/* Program L0s and L1 entrance latencies */
 	val = dw_pcie_readl_dbi(pci, PCIE_PORT_AFR);
 	val &= ~PORT_AFR_L0S_ENTRANCE_LAT_MASK;