Re: mt76x2e hardware restart

[Lorenzo Bianconi]

> On 19.09.2019 23:22, Oleksandr Natalenko wrote:
> > It checks for TX hang here:
> > 
> > === mt76x02_mmio.c
> > 557 void mt76x02_wdt_work(struct work_struct *work)
> > 558 {
> > ...
> > 562     mt76x02_check_tx_hang(dev);
> > ===
> 
> I've commented out the watchdog here ^^, and the card is not resetted any
> more, but similarly it stops working shortly after the first client
> connects. So, indeed, it must be some hang in the HW, and wdt seems to do a
> correct job.
> 
> Is it even debuggable/fixable from the driver?

Hi Oleksandr,

sorry for the delay. Felix and me worked on this issue today. Could you please
try if the following patch fixes your issue?

Regards,
Lorenzo

From cf3436c42a297967235a9c9778620c585100529e Mon Sep 17 00:00:00 2001
Message-Id: <cf3436c42a297967235a9c9778620c585100529e.1570897574.git.lorenzo@xxxxxxxxxx>
From: Lorenzo Bianconi <lorenzo@xxxxxxxxxx>
Date: Sat, 12 Oct 2019 17:32:57 +0200
Subject: [PATCH] mt76: mt76x2: disable pcie_aspm by default

On same device (e.g. U7612E-H1) PCIE_ASPM causes continues mcu hangs and
instability. This patch disable PCIE_ASPM by default. This patch has
been successfully tested on U7612E-H1 mini-pice card

Signed-off-by: Felix Fietkau <nbd@xxxxxxxx>
Signed-off-by: Lorenzo Bianconi <lorenzo@xxxxxxxxxx>
---
 drivers/net/wireless/mediatek/mt76/mmio.c     | 48 +++++++++++++++++++
 drivers/net/wireless/mediatek/mt76/mt76.h     |  1 +
 .../net/wireless/mediatek/mt76/mt76x2/pci.c   |  2 +
 3 files changed, 51 insertions(+)

diff --git a/drivers/net/wireless/mediatek/mt76/mmio.c b/drivers/net/wireless/mediatek/mt76/mmio.c
index 1c974df1fe25..8e1dbc1903f3 100644
--- a/drivers/net/wireless/mediatek/mt76/mmio.c
+++ b/drivers/net/wireless/mediatek/mt76/mmio.c
@@ -3,6 +3,9 @@
  * Copyright (C) 2016 Felix Fietkau <nbd@xxxxxxxx>
  */
 
+#include <linux/pci.h>
+#include <linux/pci-aspm.h>
+
 #include "mt76.h"
 #include "trace.h"
 
@@ -78,6 +81,51 @@ void mt76_set_irq_mask(struct mt76_dev *dev, u32 addr,
 }
 EXPORT_SYMBOL_GPL(mt76_set_irq_mask);
 
+void mt76_mmio_disable_aspm(struct pci_dev *pdev)
+{
+	struct pci_dev *parent = pdev->bus->self;
+	u16 aspm_conf, parent_aspm_conf = 0;
+
+	pcie_capability_read_word(pdev, PCI_EXP_LNKCTL, &aspm_conf);
+	aspm_conf &= PCI_EXP_LNKCTL_ASPMC;
+	if (parent) {
+		pcie_capability_read_word(parent, PCI_EXP_LNKCTL,
+					  &parent_aspm_conf);
+		parent_aspm_conf &= PCI_EXP_LNKCTL_ASPMC;
+	}
+
+	if (!aspm_conf && (!parent || !parent_aspm_conf)) {
+		/* aspm already disabled */
+		return;
+	}
+
+	dev_info(&pdev->dev, "disabling ASPM %s %s\n",
+		 (aspm_conf & PCI_EXP_LNKCTL_ASPM_L0S) ? "L0s" : "",
+		 (aspm_conf & PCI_EXP_LNKCTL_ASPM_L1) ? "L1" : "");
+
+#ifdef CONFIG_PCIEASPM
+	pci_disable_link_state(pdev, aspm_conf);
+
+	/* Double-check ASPM control.  If not disabled by the above, the
+	 * BIOS is preventing that from happening (or CONFIG_PCIEASPM is
+	 * not enabled); override by writing PCI config space directly.
+	 */
+	pcie_capability_read_word(pdev, PCI_EXP_LNKCTL, &aspm_conf);
+	if (!(aspm_conf & PCI_EXP_LNKCTL_ASPMC))
+		return;
+#endif /* CONFIG_PCIEASPM */
+
+	/* Both device and parent should have the same ASPM setting.
+	 * Disable ASPM in downstream component first and then upstream.
+	 */
+	pcie_capability_clear_word(pdev, PCI_EXP_LNKCTL, aspm_conf);
+
+	if (parent)
+		pcie_capability_clear_word(parent, PCI_EXP_LNKCTL,
+					   aspm_conf);
+}
+EXPORT_SYMBOL_GPL(mt76_mmio_disable_aspm);
+
 void mt76_mmio_init(struct mt76_dev *dev, void __iomem *regs)
 {
 	static const struct mt76_bus_ops mt76_mmio_ops = {
diff --git a/drivers/net/wireless/mediatek/mt76/mt76.h b/drivers/net/wireless/mediatek/mt76/mt76.h
index 8bcc7f21e83c..e95a5893f93b 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76.h
+++ b/drivers/net/wireless/mediatek/mt76/mt76.h
@@ -596,6 +596,7 @@ bool __mt76_poll_msec(struct mt76_dev *dev, u32 offset, u32 mask, u32 val,
 #define mt76_poll_msec(dev, ...) __mt76_poll_msec(&((dev)->mt76), __VA_ARGS__)
 
 void mt76_mmio_init(struct mt76_dev *dev, void __iomem *regs);
+void mt76_mmio_disable_aspm(struct pci_dev *pdev);
 
 static inline u16 mt76_chip(struct mt76_dev *dev)
 {
diff --git a/drivers/net/wireless/mediatek/mt76/mt76x2/pci.c b/drivers/net/wireless/mediatek/mt76/mt76x2/pci.c
index 6253ec5fbd72..06fb80163c8e 100644
--- a/drivers/net/wireless/mediatek/mt76/mt76x2/pci.c
+++ b/drivers/net/wireless/mediatek/mt76/mt76x2/pci.c
@@ -83,6 +83,8 @@ mt76pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	/* RG_SSUSB_CDR_BR_PE1D = 0x3 */
 	mt76_rmw_field(dev, 0x15c58, 0x3 << 6, 0x3);
 
+	mt76_mmio_disable_aspm(pdev);
+
 	return 0;
 
 error:
-- 
2.21.0

> 
> -- 
>   Oleksandr Natalenko (post-factum)

Attachment: signature.asc
Description: PGP signature