lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20250819-pci-pwrctrl-perst-v1-1-4b74978d2007@oss.qualcomm.com>
Date: Tue, 19 Aug 2025 12:44:50 +0530
From: Manivannan Sadhasivam via B4 Relay <devnull+manivannan.sadhasivam.oss.qualcomm.com@...nel.org>
To: Manivannan Sadhasivam <mani@...nel.org>, 
 Lorenzo Pieralisi <lpieralisi@...nel.org>, 
 Krzysztof WilczyƄski <kwilczynski@...nel.org>, 
 Rob Herring <robh@...nel.org>, Bjorn Helgaas <bhelgaas@...gle.com>, 
 Bartosz Golaszewski <brgl@...ev.pl>, Saravana Kannan <saravanak@...gle.com>
Cc: linux-pci@...r.kernel.org, linux-arm-msm@...r.kernel.org, 
 linux-kernel@...r.kernel.org, devicetree@...r.kernel.org, 
 Krishna Chaitanya Chundru <krishna.chundru@....qualcomm.com>, 
 Brian Norris <briannorris@...omium.org>, 
 Manivannan Sadhasivam <manivannan.sadhasivam@....qualcomm.com>
Subject: [PATCH 1/6] PCI: qcom: Wait for PCIE_RESET_CONFIG_WAIT_MS after
 PERST# deassert

From: Manivannan Sadhasivam <manivannan.sadhasivam@....qualcomm.com>

PCIe spec r6.0, sec 6.6.1 mandates waiting for 100ms before deasserting
PERST# if the downstream port does not support Link speeds greater than
5.0 GT/s. But in practice, this delay seem to be required irrespective of
the supported link speed as it gives the endpoints enough time to
initialize.

Hence, add the delay by reusing the PCIE_RESET_CONFIG_WAIT_MS definition if
the linkup_irq is not supported. If the linkup_irq is supported, the driver
already waits for 100ms in the IRQ handler post link up.

Also, remove the redundant comment for PCIE_T_PVPERL_MS. Finally, the
PERST_DELAY_US sleep can be moved to PERST# assert where it should be.

Fixes: 82a823833f4e ("PCI: qcom: Add Qualcomm PCIe controller driver")
Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@....qualcomm.com>
---
 drivers/pci/controller/dwc/pcie-qcom.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/pci/controller/dwc/pcie-qcom.c b/drivers/pci/controller/dwc/pcie-qcom.c
index 294babe1816e4d0c2b2343fe22d89af72afcd6cd..bcd080315d70e64eafdefd852740fe07df3dbe75 100644
--- a/drivers/pci/controller/dwc/pcie-qcom.c
+++ b/drivers/pci/controller/dwc/pcie-qcom.c
@@ -302,20 +302,22 @@ static void qcom_perst_assert(struct qcom_pcie *pcie, bool assert)
 	else
 		list_for_each_entry(port, &pcie->ports, list)
 			gpiod_set_value_cansleep(port->reset, val);
-
-	usleep_range(PERST_DELAY_US, PERST_DELAY_US + 500);
 }
 
 static void qcom_ep_reset_assert(struct qcom_pcie *pcie)
 {
 	qcom_perst_assert(pcie, true);
+	usleep_range(PERST_DELAY_US, PERST_DELAY_US + 500);
 }
 
 static void qcom_ep_reset_deassert(struct qcom_pcie *pcie)
 {
-	/* Ensure that PERST has been asserted for at least 100 ms */
+	struct dw_pcie_rp *pp = &pcie->pci->pp;
+
 	msleep(PCIE_T_PVPERL_MS);
 	qcom_perst_assert(pcie, false);
+	if (!pp->use_linkup_irq)
+		msleep(PCIE_RESET_CONFIG_WAIT_MS);
 }
 
 static int qcom_pcie_start_link(struct dw_pcie *pci)

-- 
2.45.2



Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ