All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Marek Behún" <kabel@kernel.org>
To: Stefan Roese <sr@denx.de>
Cc: u-boot@lists.denx.de, pali@kernel.org,
	"Marek Behún" <marek.behun@nic.cz>
Subject: [PATCH u-boot-marvell 4/6] arm: a37xx: pci: Handle propagation of CRSSVE bit from PCIe Root Port
Date: Sat, 25 Sep 2021 22:37:04 +0200	[thread overview]
Message-ID: <20210925203706.20322-5-kabel@kernel.org> (raw)
In-Reply-To: <20210925203706.20322-1-kabel@kernel.org>

From: Pali Rohár <pali@kernel.org>

Now that PCI Bridge (PCIe Root Port) for Aardvark is emulated in U-Boot,
add support for handling and propagation of CRSSVE bit.

When CRSSVE bit is unset (default), driver has to reissue config
read/write request on CRS response.

CRSSVE bit is supported only when CRSVIS bit is provided in read-only
Root Capabilities register. So manually inject this CRSVIS bit into read
response for that register.

Signed-off-by: Pali Rohár <pali@kernel.org>
Reviewed-by: Marek Behún <marek.behun@nic.cz>
---
 drivers/pci/pci-aardvark.c | 26 ++++++++++++++++++++++----
 include/pci.h              |  4 ++++
 2 files changed, 26 insertions(+), 4 deletions(-)

diff --git a/drivers/pci/pci-aardvark.c b/drivers/pci/pci-aardvark.c
index 8c025dc45d..53e9e23d4a 100644
--- a/drivers/pci/pci-aardvark.c
+++ b/drivers/pci/pci-aardvark.c
@@ -41,6 +41,7 @@
 #define     PCIE_CORE_CMD_MEM_IO_REQ_EN				BIT(2)
 #define PCIE_CORE_DEV_REV_REG					0x8
 #define PCIE_CORE_EXP_ROM_BAR_REG				0x30
+#define PCIE_CORE_PCIEXP_CAP_OFF				0xc0
 #define PCIE_CORE_DEV_CTRL_STATS_REG				0xc8
 #define     PCIE_CORE_DEV_CTRL_STATS_RELAX_ORDER_DISABLE	(0 << 4)
 #define     PCIE_CORE_DEV_CTRL_STATS_SNOOP_DISABLE		(0 << 11)
@@ -201,6 +202,7 @@ struct pcie_advk {
 	struct udevice *dev;
 	struct gpio_desc reset_gpio;
 	u32            cfgcache[0x34 - 0x10];
+	bool           cfgcrssve;
 };
 
 static inline void advk_writel(struct pcie_advk *pcie, uint val, uint reg)
@@ -413,6 +415,18 @@ static int pcie_advk_read_config(const struct udevice *bus, pci_dev_t bdf,
 			data |= PCI_HEADER_TYPE_BRIDGE << 16;
 		}
 
+		if ((offset & ~3) == PCIE_CORE_PCIEXP_CAP_OFF + PCI_EXP_RTCTL) {
+			/* CRSSVE bit is stored only in cache */
+			if (pcie->cfgcrssve)
+				data |= PCI_EXP_RTCTL_CRSSVE;
+		}
+
+		if ((offset & ~3) == PCIE_CORE_PCIEXP_CAP_OFF +
+				     (PCI_EXP_RTCAP & ~3)) {
+			/* CRS is emulated below, so set CRSVIS capability */
+			data |= PCI_EXP_RTCAP_CRSVIS << 16;
+		}
+
 		*valuep = pci_conv_32_to_size(data, offset, size);
 
 		return 0;
@@ -423,13 +437,14 @@ static int pcie_advk_read_config(const struct udevice *bus, pci_dev_t bdf,
 	 * OS is allowed only for 4-byte PCI_VENDOR_ID config read request and
 	 * only when CRSSVE bit in Root Port PCIe device is enabled. In all
 	 * other error PCIe Root Complex must return all-ones.
-	 * Aardvark HW does not have Root Port PCIe device and U-Boot does not
-	 * implement emulation of this device.
+	 *
 	 * U-Boot currently does not support handling of CRS return value for
 	 * PCI_VENDOR_ID config read request and also does not set CRSSVE bit.
-	 * Therefore disable returning CRS response for now.
+	 * So it means that pcie->cfgcrssve is false. But the code is prepared
+	 * for returning CRS, so that if U-Boot does support CRS in the future,
+	 * it will work for Aardvark.
 	 */
-	allow_crs = false;
+	allow_crs = pcie->cfgcrssve;
 
 	if (advk_readl(pcie, PIO_START)) {
 		dev_err(pcie->dev,
@@ -583,6 +598,9 @@ static int pcie_advk_write_config(struct udevice *bus, pci_dev_t bdf,
 		    (offset == PCI_PRIMARY_BUS && size != PCI_SIZE_8))
 			pcie->sec_busno = (data >> 8) & 0xff;
 
+		if ((offset & ~3) == PCIE_CORE_PCIEXP_CAP_OFF + PCI_EXP_RTCTL)
+			pcie->cfgcrssve = data & PCI_EXP_RTCTL_CRSSVE;
+
 		return 0;
 	}
 
diff --git a/include/pci.h b/include/pci.h
index 0fc22adffd..69eafeb4b9 100644
--- a/include/pci.h
+++ b/include/pci.h
@@ -495,6 +495,10 @@
 #define  PCI_EXP_LNKSTA_DLLLA	0x2000	/* Data Link Layer Link Active */
 #define PCI_EXP_SLTCAP		20	/* Slot Capabilities */
 #define  PCI_EXP_SLTCAP_PSN	0xfff80000 /* Physical Slot Number */
+#define PCI_EXP_RTCTL		28	/* Root Control */
+#define  PCI_EXP_RTCTL_CRSSVE	0x0010	/* CRS Software Visibility Enable */
+#define PCI_EXP_RTCAP		30	/* Root Capabilities */
+#define  PCI_EXP_RTCAP_CRSVIS	0x0001	/* CRS Software Visibility capability */
 #define PCI_EXP_DEVCAP2		36	/* Device Capabilities 2 */
 #define  PCI_EXP_DEVCAP2_ARI	0x00000020 /* ARI Forwarding Supported */
 #define PCI_EXP_DEVCTL2		40	/* Device Control 2 */
-- 
2.32.0


  parent reply	other threads:[~2021-09-25 20:38 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-09-25 20:37 [PATCH u-boot-marvell 0/6] A3720 PCIe enhancements Marek Behún
2021-09-25 20:37 ` [PATCH u-boot-marvell 1/6] arm: a37xx: pci: Fix pcie_advk_link_up() Marek Behún
2021-09-25 20:37 ` [PATCH u-boot-marvell 2/6] arm: a37xx: pci: Add support for accessing PCI Bridge on root bus Marek Behún
2021-09-25 20:37 ` [PATCH u-boot-marvell 3/6] arm: a37xx: pci: Do not automatically enable bus mastering on PCI Bridge Marek Behún
2021-09-25 20:37 ` Marek Behún [this message]
2021-09-25 20:37 ` [PATCH u-boot-marvell 5/6] arm: a37xx: pci: Cosmetic change Marek Behún
2021-09-25 20:37 ` [PATCH u-boot-marvell 6/6] arm: a37xx: pci: Update private structure documentation Marek Behún
2021-09-25 21:50   ` Pali Rohár

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20210925203706.20322-5-kabel@kernel.org \
    --to=kabel@kernel.org \
    --cc=marek.behun@nic.cz \
    --cc=pali@kernel.org \
    --cc=sr@denx.de \
    --cc=u-boot@lists.denx.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.