* [PATCH V1 00/10] PCI: tegra: Add Tegra234 PCIe support
@ 2022-02-05 16:21 ` Vidya Sagar
0 siblings, 0 replies; 46+ messages in thread
From: Vidya Sagar @ 2022-02-05 16:21 UTC (permalink / raw)
To: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh
Cc: kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, vidyas, sagar.tv
Tegra234 has a total of 11 PCIe controllers based on Synopsys DesignWare core.
There are three Universal PHY (UPHY) blocks (viz. HSIO, NVHS and GBE) with
each block supporting 8 lanes respectively. Controllers:0~4 use UPHY lanes
from HSIO block, Controllers:5,6 use UPHY lanes from NVHS block and
Controllers:7~10 use UPHY lanes from GBE block. Lane mapping in each block
is controlled in XBAR module by BPMP-FW. Since PCIe core has PIPE interface,
a glue module called PIPE-to-UPHY (P2U) is used to connect each UPHY lane
(applicable to all three UPHY bricks i.e. HSIO/NVHS/GBE) to PCIe controller.
This patch series
- Adds support for Tegra234 in the existing P2U PHY driver
- Adds support for Tegra234 in the existing PCIe platform controller driver
- Adds device tree nodes each PCIe controllers
- Enables nodes applicable to P3737-0000 platform
Testing done on P3737-0000 platform
- PCIe link is up with on-board Broadcom WiFi controller
- PCIe link is up with NVMe drive connected to M.2 Key-M slot and its
functionality is verified
- PCIe link is up with a variety of cards (NICs and USB3.0 add-on cards)
and their functionality is verified
Vidya Sagar (10):
dt-bindings: Add Tegra234 PCIe clocks and resets
dt-bindings: power: Add Tegra234 PCIe power domains
dt-bindings: memory: Add Tegra234 PCIe memory
dt-bindings: PHY: P2U: Add support for Tegra234 P2U block
dt-bindings: PCI: tegra: Add device tree support for Tegra234
arm64: tegra: Add P2U and PCIe controller nodes to Tegra234 DT
arm64: tegra: Enable PCIe slots in P3737-0000 board
phy: tegra: Add PCIe PIPE2UPHY support for Tegra234
PCI: Disable MSI for Tegra234 root ports
PCI: tegra: Add Tegra234 PCIe support
.../bindings/pci/nvidia,tegra194-pcie.txt | 106 ++-
.../bindings/phy/phy-tegra194-p2u.yaml | 17 +-
.../nvidia/tegra234-p3737-0000+p3701-0000.dts | 26 +
arch/arm64/boot/dts/nvidia/tegra234.dtsi | 743 ++++++++++++++++++
drivers/pci/controller/dwc/pcie-tegra194.c | 409 +++++++---
drivers/pci/quirks.c | 9 +
drivers/phy/tegra/phy-tegra194-p2u.c | 48 +-
include/dt-bindings/clock/tegra234-clock.h | 25 +-
include/dt-bindings/memory/tegra234-mc.h | 64 ++
.../dt-bindings/power/tegra234-powergate.h | 20 +
include/dt-bindings/reset/tegra234-reset.h | 27 +-
11 files changed, 1390 insertions(+), 104 deletions(-)
create mode 100644 include/dt-bindings/power/tegra234-powergate.h
--
2.17.1
^ permalink raw reply [flat|nested] 46+ messages in thread
* [PATCH V1 00/10] PCI: tegra: Add Tegra234 PCIe support
@ 2022-02-05 16:21 ` Vidya Sagar
0 siblings, 0 replies; 46+ messages in thread
From: Vidya Sagar @ 2022-02-05 16:21 UTC (permalink / raw)
To: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh
Cc: kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, vidyas, sagar.tv
Tegra234 has a total of 11 PCIe controllers based on Synopsys DesignWare core.
There are three Universal PHY (UPHY) blocks (viz. HSIO, NVHS and GBE) with
each block supporting 8 lanes respectively. Controllers:0~4 use UPHY lanes
from HSIO block, Controllers:5,6 use UPHY lanes from NVHS block and
Controllers:7~10 use UPHY lanes from GBE block. Lane mapping in each block
is controlled in XBAR module by BPMP-FW. Since PCIe core has PIPE interface,
a glue module called PIPE-to-UPHY (P2U) is used to connect each UPHY lane
(applicable to all three UPHY bricks i.e. HSIO/NVHS/GBE) to PCIe controller.
This patch series
- Adds support for Tegra234 in the existing P2U PHY driver
- Adds support for Tegra234 in the existing PCIe platform controller driver
- Adds device tree nodes each PCIe controllers
- Enables nodes applicable to P3737-0000 platform
Testing done on P3737-0000 platform
- PCIe link is up with on-board Broadcom WiFi controller
- PCIe link is up with NVMe drive connected to M.2 Key-M slot and its
functionality is verified
- PCIe link is up with a variety of cards (NICs and USB3.0 add-on cards)
and their functionality is verified
Vidya Sagar (10):
dt-bindings: Add Tegra234 PCIe clocks and resets
dt-bindings: power: Add Tegra234 PCIe power domains
dt-bindings: memory: Add Tegra234 PCIe memory
dt-bindings: PHY: P2U: Add support for Tegra234 P2U block
dt-bindings: PCI: tegra: Add device tree support for Tegra234
arm64: tegra: Add P2U and PCIe controller nodes to Tegra234 DT
arm64: tegra: Enable PCIe slots in P3737-0000 board
phy: tegra: Add PCIe PIPE2UPHY support for Tegra234
PCI: Disable MSI for Tegra234 root ports
PCI: tegra: Add Tegra234 PCIe support
.../bindings/pci/nvidia,tegra194-pcie.txt | 106 ++-
.../bindings/phy/phy-tegra194-p2u.yaml | 17 +-
.../nvidia/tegra234-p3737-0000+p3701-0000.dts | 26 +
arch/arm64/boot/dts/nvidia/tegra234.dtsi | 743 ++++++++++++++++++
drivers/pci/controller/dwc/pcie-tegra194.c | 409 +++++++---
drivers/pci/quirks.c | 9 +
drivers/phy/tegra/phy-tegra194-p2u.c | 48 +-
include/dt-bindings/clock/tegra234-clock.h | 25 +-
include/dt-bindings/memory/tegra234-mc.h | 64 ++
.../dt-bindings/power/tegra234-powergate.h | 20 +
include/dt-bindings/reset/tegra234-reset.h | 27 +-
11 files changed, 1390 insertions(+), 104 deletions(-)
create mode 100644 include/dt-bindings/power/tegra234-powergate.h
--
2.17.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply [flat|nested] 46+ messages in thread
* [PATCH V1 01/10] dt-bindings: Add Tegra234 PCIe clocks and resets
2022-02-05 16:21 ` Vidya Sagar
@ 2022-02-05 16:21 ` Vidya Sagar
-1 siblings, 0 replies; 46+ messages in thread
From: Vidya Sagar @ 2022-02-05 16:21 UTC (permalink / raw)
To: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh
Cc: kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, vidyas, sagar.tv
Add the clocks and resets used by the PCIe hardware found on
Tegra234 SoCs.
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
---
include/dt-bindings/clock/tegra234-clock.h | 25 +++++++++++++++++++-
include/dt-bindings/reset/tegra234-reset.h | 27 +++++++++++++++++++++-
2 files changed, 50 insertions(+), 2 deletions(-)
diff --git a/include/dt-bindings/clock/tegra234-clock.h b/include/dt-bindings/clock/tegra234-clock.h
index 8d7e66e1b6ef..106a310a6075 100644
--- a/include/dt-bindings/clock/tegra234-clock.h
+++ b/include/dt-bindings/clock/tegra234-clock.h
@@ -1,5 +1,5 @@
/* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (c) 2018-2019, NVIDIA CORPORATION. All rights reserved. */
+/* Copyright (c) 2018-2022, NVIDIA CORPORATION. All rights reserved. */
#ifndef DT_BINDINGS_CLOCK_TEGRA234_CLOCK_H
#define DT_BINDINGS_CLOCK_TEGRA234_CLOCK_H
@@ -24,8 +24,31 @@
#define TEGRA234_CLK_SDMMC4 123U
/** @brief output of mux controlled by CLK_RST_CONTROLLER_CLK_SOURCE_UARTA */
#define TEGRA234_CLK_UARTA 155U
+/** @brief output of gate CLK_ENB_PEX1_CORE_6 */
+#define TEGRA234_CLK_PEX1_C6_CORE 161U
+/** @brief output of gate CLK_ENB_PEX2_CORE_7 */
+#define TEGRA234_CLK_PEX2_C7_CORE 171U
+/** @brief output of gate CLK_ENB_PEX2_CORE_8 */
+#define TEGRA234_CLK_PEX2_C8_CORE 172U
+/** @brief output of gate CLK_ENB_PEX2_CORE_9 */
+#define TEGRA234_CLK_PEX2_C9_CORE 173U
+/** @brief output of gate CLK_ENB_PEX2_CORE_10 */
+#define TEGRA234_CLK_PEX2_C10_CORE 187U
/** @brief CLK_RST_CONTROLLER_CLK_SOURCE_SDMMC_LEGACY_TM switch divider output */
#define TEGRA234_CLK_SDMMC_LEGACY_TM 219U
+/** @brief output of gate CLK_ENB_PEX0_CORE_0 */
+#define TEGRA234_CLK_PEX0_C0_CORE 220U
+/** @brief output of gate CLK_ENB_PEX0_CORE_1 */
+#define TEGRA234_CLK_PEX0_C1_CORE 221U
+/** @brief output of gate CLK_ENB_PEX0_CORE_2 */
+#define TEGRA234_CLK_PEX0_C2_CORE 222U
+/** @brief output of gate CLK_ENB_PEX0_CORE_3 */
+#define TEGRA234_CLK_PEX0_C3_CORE 223U
+/** @brief output of gate CLK_ENB_PEX0_CORE_4 */
+#define TEGRA234_CLK_PEX0_C4_CORE 224U
+/** @brief output of gate CLK_ENB_PEX1_CORE_5 */
+#define TEGRA234_CLK_PEX1_C5_CORE 225U
+
/** @brief PLL controlled by CLK_RST_CONTROLLER_PLLC4_BASE */
#define TEGRA234_CLK_PLLC4 237U
/** @brief 32K input clock provided by PMIC */
diff --git a/include/dt-bindings/reset/tegra234-reset.h b/include/dt-bindings/reset/tegra234-reset.h
index 50e13bced642..2a0c5faf0a7e 100644
--- a/include/dt-bindings/reset/tegra234-reset.h
+++ b/include/dt-bindings/reset/tegra234-reset.h
@@ -1,5 +1,5 @@
/* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (c) 2018-2020, NVIDIA CORPORATION. All rights reserved. */
+/* Copyright (c) 2018-2022, NVIDIA CORPORATION. All rights reserved. */
#ifndef DT_BINDINGS_RESET_TEGRA234_RESET_H
#define DT_BINDINGS_RESET_TEGRA234_RESET_H
@@ -10,8 +10,33 @@
* @brief Identifiers for Resets controllable by firmware
* @{
*/
+#define TEGRA234_RESET_PEX1_CORE_6 11U
+#define TEGRA234_RESET_PEX1_CORE_6_APB 12U
+#define TEGRA234_RESET_PEX1_COMMON_APB 13U
+#define TEGRA234_RESET_PEX2_CORE_7 14U
+#define TEGRA234_RESET_PEX2_CORE_7_APB 15U
+#define TEGRA234_RESET_PEX2_CORE_8 25U
+#define TEGRA234_RESET_PEX2_CORE_8_APB 26U
+#define TEGRA234_RESET_PEX2_CORE_9 27U
+#define TEGRA234_RESET_PEX2_CORE_9_APB 28U
+#define TEGRA234_RESET_PEX2_CORE_10 56U
+#define TEGRA234_RESET_PEX2_CORE_10_APB 57U
+#define TEGRA234_RESET_PEX2_COMMON_APB 58U
#define TEGRA234_RESET_SDMMC4 85U
#define TEGRA234_RESET_UARTA 100U
+#define TEGRA234_RESET_PEX0_CORE_0 116U
+#define TEGRA234_RESET_PEX0_CORE_1 117U
+#define TEGRA234_RESET_PEX0_CORE_2 118U
+#define TEGRA234_RESET_PEX0_CORE_3 119U
+#define TEGRA234_RESET_PEX0_CORE_4 120U
+#define TEGRA234_RESET_PEX0_CORE_0_APB 121U
+#define TEGRA234_RESET_PEX0_CORE_1_APB 122U
+#define TEGRA234_RESET_PEX0_CORE_2_APB 123U
+#define TEGRA234_RESET_PEX0_CORE_3_APB 124U
+#define TEGRA234_RESET_PEX0_CORE_4_APB 125U
+#define TEGRA234_RESET_PEX0_COMMON_APB 126U
+#define TEGRA234_RESET_PEX1_CORE_5 129U
+#define TEGRA234_RESET_PEX1_CORE_5_APB 130U
/** @} */
--
2.17.1
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [PATCH V1 01/10] dt-bindings: Add Tegra234 PCIe clocks and resets
@ 2022-02-05 16:21 ` Vidya Sagar
0 siblings, 0 replies; 46+ messages in thread
From: Vidya Sagar @ 2022-02-05 16:21 UTC (permalink / raw)
To: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh
Cc: kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, vidyas, sagar.tv
Add the clocks and resets used by the PCIe hardware found on
Tegra234 SoCs.
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
---
include/dt-bindings/clock/tegra234-clock.h | 25 +++++++++++++++++++-
include/dt-bindings/reset/tegra234-reset.h | 27 +++++++++++++++++++++-
2 files changed, 50 insertions(+), 2 deletions(-)
diff --git a/include/dt-bindings/clock/tegra234-clock.h b/include/dt-bindings/clock/tegra234-clock.h
index 8d7e66e1b6ef..106a310a6075 100644
--- a/include/dt-bindings/clock/tegra234-clock.h
+++ b/include/dt-bindings/clock/tegra234-clock.h
@@ -1,5 +1,5 @@
/* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (c) 2018-2019, NVIDIA CORPORATION. All rights reserved. */
+/* Copyright (c) 2018-2022, NVIDIA CORPORATION. All rights reserved. */
#ifndef DT_BINDINGS_CLOCK_TEGRA234_CLOCK_H
#define DT_BINDINGS_CLOCK_TEGRA234_CLOCK_H
@@ -24,8 +24,31 @@
#define TEGRA234_CLK_SDMMC4 123U
/** @brief output of mux controlled by CLK_RST_CONTROLLER_CLK_SOURCE_UARTA */
#define TEGRA234_CLK_UARTA 155U
+/** @brief output of gate CLK_ENB_PEX1_CORE_6 */
+#define TEGRA234_CLK_PEX1_C6_CORE 161U
+/** @brief output of gate CLK_ENB_PEX2_CORE_7 */
+#define TEGRA234_CLK_PEX2_C7_CORE 171U
+/** @brief output of gate CLK_ENB_PEX2_CORE_8 */
+#define TEGRA234_CLK_PEX2_C8_CORE 172U
+/** @brief output of gate CLK_ENB_PEX2_CORE_9 */
+#define TEGRA234_CLK_PEX2_C9_CORE 173U
+/** @brief output of gate CLK_ENB_PEX2_CORE_10 */
+#define TEGRA234_CLK_PEX2_C10_CORE 187U
/** @brief CLK_RST_CONTROLLER_CLK_SOURCE_SDMMC_LEGACY_TM switch divider output */
#define TEGRA234_CLK_SDMMC_LEGACY_TM 219U
+/** @brief output of gate CLK_ENB_PEX0_CORE_0 */
+#define TEGRA234_CLK_PEX0_C0_CORE 220U
+/** @brief output of gate CLK_ENB_PEX0_CORE_1 */
+#define TEGRA234_CLK_PEX0_C1_CORE 221U
+/** @brief output of gate CLK_ENB_PEX0_CORE_2 */
+#define TEGRA234_CLK_PEX0_C2_CORE 222U
+/** @brief output of gate CLK_ENB_PEX0_CORE_3 */
+#define TEGRA234_CLK_PEX0_C3_CORE 223U
+/** @brief output of gate CLK_ENB_PEX0_CORE_4 */
+#define TEGRA234_CLK_PEX0_C4_CORE 224U
+/** @brief output of gate CLK_ENB_PEX1_CORE_5 */
+#define TEGRA234_CLK_PEX1_C5_CORE 225U
+
/** @brief PLL controlled by CLK_RST_CONTROLLER_PLLC4_BASE */
#define TEGRA234_CLK_PLLC4 237U
/** @brief 32K input clock provided by PMIC */
diff --git a/include/dt-bindings/reset/tegra234-reset.h b/include/dt-bindings/reset/tegra234-reset.h
index 50e13bced642..2a0c5faf0a7e 100644
--- a/include/dt-bindings/reset/tegra234-reset.h
+++ b/include/dt-bindings/reset/tegra234-reset.h
@@ -1,5 +1,5 @@
/* SPDX-License-Identifier: GPL-2.0 */
-/* Copyright (c) 2018-2020, NVIDIA CORPORATION. All rights reserved. */
+/* Copyright (c) 2018-2022, NVIDIA CORPORATION. All rights reserved. */
#ifndef DT_BINDINGS_RESET_TEGRA234_RESET_H
#define DT_BINDINGS_RESET_TEGRA234_RESET_H
@@ -10,8 +10,33 @@
* @brief Identifiers for Resets controllable by firmware
* @{
*/
+#define TEGRA234_RESET_PEX1_CORE_6 11U
+#define TEGRA234_RESET_PEX1_CORE_6_APB 12U
+#define TEGRA234_RESET_PEX1_COMMON_APB 13U
+#define TEGRA234_RESET_PEX2_CORE_7 14U
+#define TEGRA234_RESET_PEX2_CORE_7_APB 15U
+#define TEGRA234_RESET_PEX2_CORE_8 25U
+#define TEGRA234_RESET_PEX2_CORE_8_APB 26U
+#define TEGRA234_RESET_PEX2_CORE_9 27U
+#define TEGRA234_RESET_PEX2_CORE_9_APB 28U
+#define TEGRA234_RESET_PEX2_CORE_10 56U
+#define TEGRA234_RESET_PEX2_CORE_10_APB 57U
+#define TEGRA234_RESET_PEX2_COMMON_APB 58U
#define TEGRA234_RESET_SDMMC4 85U
#define TEGRA234_RESET_UARTA 100U
+#define TEGRA234_RESET_PEX0_CORE_0 116U
+#define TEGRA234_RESET_PEX0_CORE_1 117U
+#define TEGRA234_RESET_PEX0_CORE_2 118U
+#define TEGRA234_RESET_PEX0_CORE_3 119U
+#define TEGRA234_RESET_PEX0_CORE_4 120U
+#define TEGRA234_RESET_PEX0_CORE_0_APB 121U
+#define TEGRA234_RESET_PEX0_CORE_1_APB 122U
+#define TEGRA234_RESET_PEX0_CORE_2_APB 123U
+#define TEGRA234_RESET_PEX0_CORE_3_APB 124U
+#define TEGRA234_RESET_PEX0_CORE_4_APB 125U
+#define TEGRA234_RESET_PEX0_COMMON_APB 126U
+#define TEGRA234_RESET_PEX1_CORE_5 129U
+#define TEGRA234_RESET_PEX1_CORE_5_APB 130U
/** @} */
--
2.17.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [PATCH V1 02/10] dt-bindings: power: Add Tegra234 PCIe power domains
2022-02-05 16:21 ` Vidya Sagar
@ 2022-02-05 16:21 ` Vidya Sagar
-1 siblings, 0 replies; 46+ messages in thread
From: Vidya Sagar @ 2022-02-05 16:21 UTC (permalink / raw)
To: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh
Cc: kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, vidyas, sagar.tv
Add power domain IDs for the four PCIe power partitions found on
Tegra234.
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
---
.../dt-bindings/power/tegra234-powergate.h | 20 +++++++++++++++++++
1 file changed, 20 insertions(+)
create mode 100644 include/dt-bindings/power/tegra234-powergate.h
diff --git a/include/dt-bindings/power/tegra234-powergate.h b/include/dt-bindings/power/tegra234-powergate.h
new file mode 100644
index 000000000000..e989f84b24f6
--- /dev/null
+++ b/include/dt-bindings/power/tegra234-powergate.h
@@ -0,0 +1,20 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) 2022, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __ABI_MACH_T234_POWERGATE_T234_H_
+#define __ABI_MACH_T234_POWERGATE_T234_H_
+
+#define TEGRA234_POWER_DOMAIN_PCIEX8A 5U
+#define TEGRA234_POWER_DOMAIN_PCIEX4A 6U
+#define TEGRA234_POWER_DOMAIN_PCIEX4BA 7U
+#define TEGRA234_POWER_DOMAIN_PCIEX4BB 8U
+#define TEGRA234_POWER_DOMAIN_PCIEX1A 9U
+#define TEGRA234_POWER_DOMAIN_PCIEX4CA 13U
+#define TEGRA234_POWER_DOMAIN_PCIEX4CB 14U
+#define TEGRA234_POWER_DOMAIN_PCIEX4CC 15U
+#define TEGRA234_POWER_DOMAIN_PCIEX8B 16U
+#define TEGRA234_POWER_DOMAIN_MGBEA 17U
+#define TEGRA234_POWER_DOMAIN_MGBEB 18U
+#define TEGRA234_POWER_DOMAIN_MGBEC 19U
+
+#endif
--
2.17.1
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [PATCH V1 02/10] dt-bindings: power: Add Tegra234 PCIe power domains
@ 2022-02-05 16:21 ` Vidya Sagar
0 siblings, 0 replies; 46+ messages in thread
From: Vidya Sagar @ 2022-02-05 16:21 UTC (permalink / raw)
To: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh
Cc: kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, vidyas, sagar.tv
Add power domain IDs for the four PCIe power partitions found on
Tegra234.
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
---
.../dt-bindings/power/tegra234-powergate.h | 20 +++++++++++++++++++
1 file changed, 20 insertions(+)
create mode 100644 include/dt-bindings/power/tegra234-powergate.h
diff --git a/include/dt-bindings/power/tegra234-powergate.h b/include/dt-bindings/power/tegra234-powergate.h
new file mode 100644
index 000000000000..e989f84b24f6
--- /dev/null
+++ b/include/dt-bindings/power/tegra234-powergate.h
@@ -0,0 +1,20 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright (c) 2022, NVIDIA CORPORATION. All rights reserved. */
+
+#ifndef __ABI_MACH_T234_POWERGATE_T234_H_
+#define __ABI_MACH_T234_POWERGATE_T234_H_
+
+#define TEGRA234_POWER_DOMAIN_PCIEX8A 5U
+#define TEGRA234_POWER_DOMAIN_PCIEX4A 6U
+#define TEGRA234_POWER_DOMAIN_PCIEX4BA 7U
+#define TEGRA234_POWER_DOMAIN_PCIEX4BB 8U
+#define TEGRA234_POWER_DOMAIN_PCIEX1A 9U
+#define TEGRA234_POWER_DOMAIN_PCIEX4CA 13U
+#define TEGRA234_POWER_DOMAIN_PCIEX4CB 14U
+#define TEGRA234_POWER_DOMAIN_PCIEX4CC 15U
+#define TEGRA234_POWER_DOMAIN_PCIEX8B 16U
+#define TEGRA234_POWER_DOMAIN_MGBEA 17U
+#define TEGRA234_POWER_DOMAIN_MGBEB 18U
+#define TEGRA234_POWER_DOMAIN_MGBEC 19U
+
+#endif
--
2.17.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [PATCH V1 03/10] dt-bindings: memory: Add Tegra234 PCIe memory
2022-02-05 16:21 ` Vidya Sagar
@ 2022-02-05 16:21 ` Vidya Sagar
-1 siblings, 0 replies; 46+ messages in thread
From: Vidya Sagar @ 2022-02-05 16:21 UTC (permalink / raw)
To: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh
Cc: kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, vidyas, sagar.tv
Add the memory client and stream ID definitions for the PCIe hardware
found on Tegra234 SoCs.
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
---
include/dt-bindings/memory/tegra234-mc.h | 64 ++++++++++++++++++++++++
1 file changed, 64 insertions(+)
diff --git a/include/dt-bindings/memory/tegra234-mc.h b/include/dt-bindings/memory/tegra234-mc.h
index 2662f70c15c6..60017684858a 100644
--- a/include/dt-bindings/memory/tegra234-mc.h
+++ b/include/dt-bindings/memory/tegra234-mc.h
@@ -7,15 +7,53 @@
#define TEGRA234_SID_INVALID 0x00
#define TEGRA234_SID_PASSTHROUGH 0x7f
+/* NISO0 stream IDs */
+#define TEGRA234_SID_PCIE0 0x12U
+#define TEGRA234_SID_PCIE4 0x13U
+#define TEGRA234_SID_PCIE5 0x14U
+#define TEGRA234_SID_PCIE6 0x15U
+#define TEGRA234_SID_PCIE9 0x1FU
/* NISO1 stream IDs */
#define TEGRA234_SID_SDMMC4 0x02
+#define TEGRA234_SID_PCIE1 0x5U
+#define TEGRA234_SID_PCIE2 0x6U
+#define TEGRA234_SID_PCIE3 0x7U
+#define TEGRA234_SID_PCIE7 0x8U
+#define TEGRA234_SID_PCIE8 0x9U
+#define TEGRA234_SID_PCIE10 0xBU
#define TEGRA234_SID_BPMP 0x10
/*
* memory client IDs
*/
+/* PCIE6 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE6AR 0x28
+/* PCIE6 write clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE6AW 0x29
+/* PCIE7 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE7AR 0x2a
+/* PCIE7 write clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE7AW 0x30
+/* PCIE8 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE8AR 0x32
+/* PCIE8 write clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE8AW 0x3b
+/* PCIE9 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE9AR 0x3c
+/* PCIE6r1 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE6AR1 0x3d
+/* PCIE9 write clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE9AW 0x3e
+/* PCIE10 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE10AR 0x3f
+/* PCIE10 write clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE10AW 0x40
+/* PCIE10r1 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE10AR1 0x48
+/* PCIE7r1 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE7AR1 0x49
/* sdmmcd memory read client */
#define TEGRA234_MEMORY_CLIENT_SDMMCRAB 0x63
/* sdmmcd memory write client */
@@ -28,5 +66,31 @@
#define TEGRA234_MEMORY_CLIENT_BPMPDMAR 0x95
/* BPMPDMA write client */
#define TEGRA234_MEMORY_CLIENT_BPMPDMAW 0x96
+/* PCIE0 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE0R 0xd8
+/* PCIE0 write clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE0W 0xd9
+/* PCIE1 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE1R 0xda
+/* PCIE1 write clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE1W 0xdb
+/* PCIE2 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE2AR 0xdc
+/* PCIE2 write clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE2AW 0xdd
+/* PCIE3 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE3R 0xde
+/* PCIE3 write clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE3W 0xdf
+/* PCIE4 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE4R 0xe0
+/* PCIE4 write clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE4W 0xe1
+/* PCIE5 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE5R 0xe2
+/* PCIE5 write clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE5W 0xe3
+/* PCIE5r1 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE5R1 0xef
#endif
--
2.17.1
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [PATCH V1 03/10] dt-bindings: memory: Add Tegra234 PCIe memory
@ 2022-02-05 16:21 ` Vidya Sagar
0 siblings, 0 replies; 46+ messages in thread
From: Vidya Sagar @ 2022-02-05 16:21 UTC (permalink / raw)
To: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh
Cc: kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, vidyas, sagar.tv
Add the memory client and stream ID definitions for the PCIe hardware
found on Tegra234 SoCs.
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
---
include/dt-bindings/memory/tegra234-mc.h | 64 ++++++++++++++++++++++++
1 file changed, 64 insertions(+)
diff --git a/include/dt-bindings/memory/tegra234-mc.h b/include/dt-bindings/memory/tegra234-mc.h
index 2662f70c15c6..60017684858a 100644
--- a/include/dt-bindings/memory/tegra234-mc.h
+++ b/include/dt-bindings/memory/tegra234-mc.h
@@ -7,15 +7,53 @@
#define TEGRA234_SID_INVALID 0x00
#define TEGRA234_SID_PASSTHROUGH 0x7f
+/* NISO0 stream IDs */
+#define TEGRA234_SID_PCIE0 0x12U
+#define TEGRA234_SID_PCIE4 0x13U
+#define TEGRA234_SID_PCIE5 0x14U
+#define TEGRA234_SID_PCIE6 0x15U
+#define TEGRA234_SID_PCIE9 0x1FU
/* NISO1 stream IDs */
#define TEGRA234_SID_SDMMC4 0x02
+#define TEGRA234_SID_PCIE1 0x5U
+#define TEGRA234_SID_PCIE2 0x6U
+#define TEGRA234_SID_PCIE3 0x7U
+#define TEGRA234_SID_PCIE7 0x8U
+#define TEGRA234_SID_PCIE8 0x9U
+#define TEGRA234_SID_PCIE10 0xBU
#define TEGRA234_SID_BPMP 0x10
/*
* memory client IDs
*/
+/* PCIE6 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE6AR 0x28
+/* PCIE6 write clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE6AW 0x29
+/* PCIE7 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE7AR 0x2a
+/* PCIE7 write clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE7AW 0x30
+/* PCIE8 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE8AR 0x32
+/* PCIE8 write clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE8AW 0x3b
+/* PCIE9 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE9AR 0x3c
+/* PCIE6r1 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE6AR1 0x3d
+/* PCIE9 write clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE9AW 0x3e
+/* PCIE10 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE10AR 0x3f
+/* PCIE10 write clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE10AW 0x40
+/* PCIE10r1 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE10AR1 0x48
+/* PCIE7r1 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE7AR1 0x49
/* sdmmcd memory read client */
#define TEGRA234_MEMORY_CLIENT_SDMMCRAB 0x63
/* sdmmcd memory write client */
@@ -28,5 +66,31 @@
#define TEGRA234_MEMORY_CLIENT_BPMPDMAR 0x95
/* BPMPDMA write client */
#define TEGRA234_MEMORY_CLIENT_BPMPDMAW 0x96
+/* PCIE0 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE0R 0xd8
+/* PCIE0 write clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE0W 0xd9
+/* PCIE1 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE1R 0xda
+/* PCIE1 write clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE1W 0xdb
+/* PCIE2 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE2AR 0xdc
+/* PCIE2 write clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE2AW 0xdd
+/* PCIE3 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE3R 0xde
+/* PCIE3 write clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE3W 0xdf
+/* PCIE4 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE4R 0xe0
+/* PCIE4 write clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE4W 0xe1
+/* PCIE5 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE5R 0xe2
+/* PCIE5 write clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE5W 0xe3
+/* PCIE5r1 read clients */
+#define TEGRA234_MEMORY_CLIENT_PCIE5R1 0xef
#endif
--
2.17.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [PATCH V1 04/10] dt-bindings: PHY: P2U: Add support for Tegra234 P2U block
2022-02-05 16:21 ` Vidya Sagar
@ 2022-02-05 16:21 ` Vidya Sagar
-1 siblings, 0 replies; 46+ messages in thread
From: Vidya Sagar @ 2022-02-05 16:21 UTC (permalink / raw)
To: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh
Cc: kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, vidyas, sagar.tv
Add support for Tegra234 P2U (PIPE to UPHY) module block which is a glue
module instantiated once for each PCIe lane between Synopsys DesignWare
core based PCIe IP and Universal PHY block.
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
---
.../bindings/phy/phy-tegra194-p2u.yaml | 17 +++++++++++++----
1 file changed, 13 insertions(+), 4 deletions(-)
diff --git a/Documentation/devicetree/bindings/phy/phy-tegra194-p2u.yaml b/Documentation/devicetree/bindings/phy/phy-tegra194-p2u.yaml
index 9a89d05efbda..6ba1f69b1126 100644
--- a/Documentation/devicetree/bindings/phy/phy-tegra194-p2u.yaml
+++ b/Documentation/devicetree/bindings/phy/phy-tegra194-p2u.yaml
@@ -4,7 +4,7 @@
$id: "http://devicetree.org/schemas/phy/phy-tegra194-p2u.yaml#"
$schema: "http://devicetree.org/meta-schemas/core.yaml#"
-title: NVIDIA Tegra194 P2U binding
+title: NVIDIA Tegra194 & Tegra234 P2U binding
maintainers:
- Thierry Reding <treding@nvidia.com>
@@ -12,13 +12,17 @@ maintainers:
description: >
Tegra194 has two PHY bricks namely HSIO (High Speed IO) and NVHS (NVIDIA High
Speed) each interfacing with 12 and 8 P2U instances respectively.
+ Tegra234 has three PHY bricks namesly HSIO, NVHS and GBE (Gigabit Ethernet)
+ each interfacing with 8, 8 and 8 P2U instances respectively.
A P2U instance is a glue logic between Synopsys DesignWare Core PCIe IP's PIPE
- interface and PHY of HSIO/NVHS bricks. Each P2U instance represents one PCIe
- lane.
+ interface and PHY of HSIO/NVHS/GBE bricks. Each P2U instance represents one
+ PCIe lane.
properties:
compatible:
- const: nvidia,tegra194-p2u
+ oneOf:
+ - const: nvidia,tegra194-p2u
+ - const: nvidia,tegra234-p2u
reg:
maxItems: 1
@@ -28,6 +32,11 @@ properties:
items:
- const: ctl
+ nvidia,skip-sz-protect-en:
+ description: Should be present if two PCIe retimers are present between
+ the root port and its immediate downstream device.
+ type: boolean
+
'#phy-cells':
const: 0
--
2.17.1
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [PATCH V1 04/10] dt-bindings: PHY: P2U: Add support for Tegra234 P2U block
@ 2022-02-05 16:21 ` Vidya Sagar
0 siblings, 0 replies; 46+ messages in thread
From: Vidya Sagar @ 2022-02-05 16:21 UTC (permalink / raw)
To: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh
Cc: kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, vidyas, sagar.tv
Add support for Tegra234 P2U (PIPE to UPHY) module block which is a glue
module instantiated once for each PCIe lane between Synopsys DesignWare
core based PCIe IP and Universal PHY block.
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
---
.../bindings/phy/phy-tegra194-p2u.yaml | 17 +++++++++++++----
1 file changed, 13 insertions(+), 4 deletions(-)
diff --git a/Documentation/devicetree/bindings/phy/phy-tegra194-p2u.yaml b/Documentation/devicetree/bindings/phy/phy-tegra194-p2u.yaml
index 9a89d05efbda..6ba1f69b1126 100644
--- a/Documentation/devicetree/bindings/phy/phy-tegra194-p2u.yaml
+++ b/Documentation/devicetree/bindings/phy/phy-tegra194-p2u.yaml
@@ -4,7 +4,7 @@
$id: "http://devicetree.org/schemas/phy/phy-tegra194-p2u.yaml#"
$schema: "http://devicetree.org/meta-schemas/core.yaml#"
-title: NVIDIA Tegra194 P2U binding
+title: NVIDIA Tegra194 & Tegra234 P2U binding
maintainers:
- Thierry Reding <treding@nvidia.com>
@@ -12,13 +12,17 @@ maintainers:
description: >
Tegra194 has two PHY bricks namely HSIO (High Speed IO) and NVHS (NVIDIA High
Speed) each interfacing with 12 and 8 P2U instances respectively.
+ Tegra234 has three PHY bricks namesly HSIO, NVHS and GBE (Gigabit Ethernet)
+ each interfacing with 8, 8 and 8 P2U instances respectively.
A P2U instance is a glue logic between Synopsys DesignWare Core PCIe IP's PIPE
- interface and PHY of HSIO/NVHS bricks. Each P2U instance represents one PCIe
- lane.
+ interface and PHY of HSIO/NVHS/GBE bricks. Each P2U instance represents one
+ PCIe lane.
properties:
compatible:
- const: nvidia,tegra194-p2u
+ oneOf:
+ - const: nvidia,tegra194-p2u
+ - const: nvidia,tegra234-p2u
reg:
maxItems: 1
@@ -28,6 +32,11 @@ properties:
items:
- const: ctl
+ nvidia,skip-sz-protect-en:
+ description: Should be present if two PCIe retimers are present between
+ the root port and its immediate downstream device.
+ type: boolean
+
'#phy-cells':
const: 0
--
2.17.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [PATCH V1 05/10] dt-bindings: PCI: tegra: Add device tree support for Tegra234
2022-02-05 16:21 ` Vidya Sagar
@ 2022-02-05 16:21 ` Vidya Sagar
-1 siblings, 0 replies; 46+ messages in thread
From: Vidya Sagar @ 2022-02-05 16:21 UTC (permalink / raw)
To: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh
Cc: kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, vidyas, sagar.tv
Add support for Tegra234 PCIe controllers. These controllers are based
on Synopsys DesignWare core IP.
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
---
.../bindings/pci/nvidia,tegra194-pcie.txt | 106 +++++++++++++++++-
1 file changed, 103 insertions(+), 3 deletions(-)
diff --git a/Documentation/devicetree/bindings/pci/nvidia,tegra194-pcie.txt b/Documentation/devicetree/bindings/pci/nvidia,tegra194-pcie.txt
index 8e4f9bfb316d..ae9555a50fb0 100644
--- a/Documentation/devicetree/bindings/pci/nvidia,tegra194-pcie.txt
+++ b/Documentation/devicetree/bindings/pci/nvidia,tegra194-pcie.txt
@@ -10,6 +10,8 @@ Required properties:
- power-domains: A phandle to the node that controls power to the respective
PCIe controller and a specifier name for the PCIe controller. Following are
the specifiers for the different PCIe controllers
+ Tegra194:
+ ---------
TEGRA194_POWER_DOMAIN_PCIEX8B: C0
TEGRA194_POWER_DOMAIN_PCIEX1A: C1
TEGRA194_POWER_DOMAIN_PCIEX1A: C2
@@ -18,6 +20,21 @@ Required properties:
TEGRA194_POWER_DOMAIN_PCIEX8A: C5
these specifiers are defined in
"include/dt-bindings/power/tegra194-powergate.h" file.
+ Tegra234:
+ ---------
+ TEGRA234_POWER_DOMAIN_PCIEX4BA: C0
+ TEGRA234_POWER_DOMAIN_PCIEX1A : C1
+ TEGRA234_POWER_DOMAIN_PCIEX1A : C2
+ TEGRA234_POWER_DOMAIN_PCIEX1A : C3
+ TEGRA234_POWER_DOMAIN_PCIEX4BB: C4
+ TEGRA234_POWER_DOMAIN_PCIEX8A : C5
+ TEGRA234_POWER_DOMAIN_PCIEX4A : C6
+ TEGRA234_POWER_DOMAIN_PCIEX8B : C7
+ TEGRA234_POWER_DOMAIN_PCIEX4CA: C8
+ TEGRA234_POWER_DOMAIN_PCIEX4CB: C9
+ TEGRA234_POWER_DOMAIN_PCIEX4CC: C10
+ these specifiers are defined in
+ "include/dt-bindings/power/tegra234-powergate.h" file.
- reg: A list of physical base address and length pairs for each set of
controller registers. Must contain an entry for each entry in the reg-names
property.
@@ -47,16 +64,33 @@ Required properties:
"p2u-N": where N ranges from 0 to one less than the total number of lanes
- nvidia,bpmp: Must contain a pair of phandle to BPMP controller node followed
by controller-id. Following are the controller ids for each controller.
+ Tegra194:
+ ---------
0: C0
1: C1
2: C2
3: C3
4: C4
5: C5
+ Tegra194:
+ ---------
+ 0 : C0
+ 1 : C1
+ 2 : C2
+ 3 : C3
+ 4 : C4
+ 5 : C5
+ 6 : C6
+ 7 : C7
+ 8 : C8
+ 9 : C9
+ 10: C10
- vddio-pex-ctl-supply: Regulator supply for PCIe side band signals
RC mode:
-- compatible: Tegra19x must contain "nvidia,tegra194-pcie"
+- compatible: Must be
+ - "nvidia,tegra194-pcie" for Tegra194
+ - "nvidia,tegra234-pcie" for Tegra234
- device_type: Must be "pci" for RC mode
- interrupt-names: Must include the following entries:
"msi": The Tegra interrupt that is asserted when an MSI is received
@@ -99,7 +133,8 @@ In Tegra194, Only controllers C0, C4 & C5 support EP mode.
Optional properties:
- pinctrl-names: A list of pinctrl state names.
- It is mandatory for C5 controller and optional for other controllers.
+ It is mandatory for C5 controller in Tegra194 and optional for other
+ controllers.
- "default": Configures PCIe I/O for proper operation.
- pinctrl-0: phandle for the 'default' state of pin configuration.
It is mandatory for C5 controller and optional for other controllers.
@@ -108,7 +143,7 @@ Optional properties:
improve performance when a platform is designed in such a way that it
satisfies at least one of the following conditions thereby enabling root
port to exchange optimum number of FC (Flow Control) credits with
- downstream devices
+ downstream devices. This is applicable for only Tegra194.
1. If C0/C4/C5 run at x1/x2 link widths (irrespective of speed and MPS)
2. If C0/C1/C2/C3/C4/C5 operate at their respective max link widths and
a) speed is Gen-2 and MPS is 256B
@@ -119,6 +154,9 @@ Optional properties:
specified in microseconds
- nvidia,aspm-l0s-entrance-latency-us: ASPM L0s entrance latency to be
specified in microseconds
+- nvidia,enable-srns: This boolean property needs to be present if the
+ controller is configured to operate in SRNS (Separate Reference Clocks with
+ No Spread-Spectrum Clocking). This is applicable only for Tegra234.
RC mode:
- vpcie3v3-supply: A phandle to the regulator node that supplies 3.3V to the slot
@@ -127,6 +165,9 @@ RC mode:
- vpcie12v-supply: A phandle to the regulator node that supplies 12V to the slot
if the platform has one such slot. (Ex:- x16 slot owned by C5 controller
in p2972-0000 platform).
+- nvidia,enable-ext-refclk: This boolean property needs to be present if the
+ controller is configured to use the reference clocking coming in from an
+ external clock source instead of using the internal clock source.
EP mode:
- nvidia,refclk-select-gpios: Must contain a phandle to a GPIO controller
@@ -243,3 +284,62 @@ Tegra194 EP mode:
phy-names = "p2u-0", "p2u-1", "p2u-2", "p2u-3", "p2u-4",
"p2u-5", "p2u-6", "p2u-7";
};
+
+Tegra234 RP mode:
+-----------------
+
+ pcie@141a0000 {
+ compatible = "nvidia,tegra234-pcie";
+ power-domains = <&bpmp TEGRA234_POWER_DOMAIN_PCIEX8A>;
+ reg = <0x00 0x141a0000 0x0 0x00020000>, /* appl registers (128K) */
+ <0x00 0x3a000000 0x0 0x00040000>, /* configuration space (256K) */
+ <0x00 0x3a040000 0x0 0x00040000>, /* iATU_DMA reg space (256K) */
+ <0x00 0x3a080000 0x0 0x00040000>; /* DBI reg space (256K) */
+ reg-names = "appl", "config", "atu_dma", "dbi";
+
+ status = "disabled";
+
+ #address-cells = <3>;
+ #size-cells = <2>;
+ device_type = "pci";
+ num-lanes = <8>;
+ num-viewport = <8>;
+ linux,pci-domain = <5>;
+
+ clocks = <&bpmp TEGRA234_CLK_PEX1_C5_CORE>;
+ clock-names = "core";
+
+ resets = <&bpmp TEGRA234_RESET_PEX1_CORE_5_APB>,
+ <&bpmp TEGRA234_RESET_PEX1_CORE_5>;
+ reset-names = "apb", "core";
+
+ interrupts = <GIC_SPI 53 IRQ_TYPE_LEVEL_HIGH>, /* controller interrupt */
+ <GIC_SPI 54 IRQ_TYPE_LEVEL_HIGH>; /* MSI interrupt */
+ interrupt-names = "intr", "msi";
+
+ #interrupt-cells = <1>;
+ interrupt-map-mask = <0 0 0 0>;
+ interrupt-map = <0 0 0 0 &gic GIC_SPI 53 IRQ_TYPE_LEVEL_HIGH>;
+
+ nvidia,bpmp = <&bpmp 5>;
+
+ nvidia,aspm-cmrt-us = <60>;
+ nvidia,aspm-pwr-on-t-us = <20>;
+ nvidia,aspm-l0s-entrance-latency-us = <3>;
+
+ bus-range = <0x0 0xff>;
+
+ ranges = <0x43000000 0x27 0x40000000 0x27 0x40000000 0x3 0xe8000000>, /* prefetchable memory (16000 MB) */
+ <0x02000000 0x0 0x40000000 0x2b 0x28000000 0x0 0x08000000>, /* non-prefetchable memory (128 MB) */
+ <0x01000000 0x0 0x3a100000 0x00 0x3a100000 0x0 0x00100000>; /* downstream I/O (1 MB) */
+
+ interconnects = <&mc TEGRA234_MEMORY_CLIENT_PCIE5R &emc>,
+ <&mc TEGRA234_MEMORY_CLIENT_PCIE5W &emc>;
+ interconnect-names = "dma-mem", "write";
+
+ phys = <&p2u_nvhs_0>, <&p2u_nvhs_1>, <&p2u_nvhs_2>,
+ <&p2u_nvhs_3>, <&p2u_nvhs_4>, <&p2u_nvhs_5>,
+ <&p2u_nvhs_6>, <&p2u_nvhs_7>;
+ phy-names = "p2u-0", "p2u-1", "p2u-2", "p2u-3", "p2u-4",
+ "p2u-5", "p2u-6", "p2u-7";
+ };
--
2.17.1
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [PATCH V1 05/10] dt-bindings: PCI: tegra: Add device tree support for Tegra234
@ 2022-02-05 16:21 ` Vidya Sagar
0 siblings, 0 replies; 46+ messages in thread
From: Vidya Sagar @ 2022-02-05 16:21 UTC (permalink / raw)
To: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh
Cc: kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, vidyas, sagar.tv
Add support for Tegra234 PCIe controllers. These controllers are based
on Synopsys DesignWare core IP.
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
---
.../bindings/pci/nvidia,tegra194-pcie.txt | 106 +++++++++++++++++-
1 file changed, 103 insertions(+), 3 deletions(-)
diff --git a/Documentation/devicetree/bindings/pci/nvidia,tegra194-pcie.txt b/Documentation/devicetree/bindings/pci/nvidia,tegra194-pcie.txt
index 8e4f9bfb316d..ae9555a50fb0 100644
--- a/Documentation/devicetree/bindings/pci/nvidia,tegra194-pcie.txt
+++ b/Documentation/devicetree/bindings/pci/nvidia,tegra194-pcie.txt
@@ -10,6 +10,8 @@ Required properties:
- power-domains: A phandle to the node that controls power to the respective
PCIe controller and a specifier name for the PCIe controller. Following are
the specifiers for the different PCIe controllers
+ Tegra194:
+ ---------
TEGRA194_POWER_DOMAIN_PCIEX8B: C0
TEGRA194_POWER_DOMAIN_PCIEX1A: C1
TEGRA194_POWER_DOMAIN_PCIEX1A: C2
@@ -18,6 +20,21 @@ Required properties:
TEGRA194_POWER_DOMAIN_PCIEX8A: C5
these specifiers are defined in
"include/dt-bindings/power/tegra194-powergate.h" file.
+ Tegra234:
+ ---------
+ TEGRA234_POWER_DOMAIN_PCIEX4BA: C0
+ TEGRA234_POWER_DOMAIN_PCIEX1A : C1
+ TEGRA234_POWER_DOMAIN_PCIEX1A : C2
+ TEGRA234_POWER_DOMAIN_PCIEX1A : C3
+ TEGRA234_POWER_DOMAIN_PCIEX4BB: C4
+ TEGRA234_POWER_DOMAIN_PCIEX8A : C5
+ TEGRA234_POWER_DOMAIN_PCIEX4A : C6
+ TEGRA234_POWER_DOMAIN_PCIEX8B : C7
+ TEGRA234_POWER_DOMAIN_PCIEX4CA: C8
+ TEGRA234_POWER_DOMAIN_PCIEX4CB: C9
+ TEGRA234_POWER_DOMAIN_PCIEX4CC: C10
+ these specifiers are defined in
+ "include/dt-bindings/power/tegra234-powergate.h" file.
- reg: A list of physical base address and length pairs for each set of
controller registers. Must contain an entry for each entry in the reg-names
property.
@@ -47,16 +64,33 @@ Required properties:
"p2u-N": where N ranges from 0 to one less than the total number of lanes
- nvidia,bpmp: Must contain a pair of phandle to BPMP controller node followed
by controller-id. Following are the controller ids for each controller.
+ Tegra194:
+ ---------
0: C0
1: C1
2: C2
3: C3
4: C4
5: C5
+ Tegra194:
+ ---------
+ 0 : C0
+ 1 : C1
+ 2 : C2
+ 3 : C3
+ 4 : C4
+ 5 : C5
+ 6 : C6
+ 7 : C7
+ 8 : C8
+ 9 : C9
+ 10: C10
- vddio-pex-ctl-supply: Regulator supply for PCIe side band signals
RC mode:
-- compatible: Tegra19x must contain "nvidia,tegra194-pcie"
+- compatible: Must be
+ - "nvidia,tegra194-pcie" for Tegra194
+ - "nvidia,tegra234-pcie" for Tegra234
- device_type: Must be "pci" for RC mode
- interrupt-names: Must include the following entries:
"msi": The Tegra interrupt that is asserted when an MSI is received
@@ -99,7 +133,8 @@ In Tegra194, Only controllers C0, C4 & C5 support EP mode.
Optional properties:
- pinctrl-names: A list of pinctrl state names.
- It is mandatory for C5 controller and optional for other controllers.
+ It is mandatory for C5 controller in Tegra194 and optional for other
+ controllers.
- "default": Configures PCIe I/O for proper operation.
- pinctrl-0: phandle for the 'default' state of pin configuration.
It is mandatory for C5 controller and optional for other controllers.
@@ -108,7 +143,7 @@ Optional properties:
improve performance when a platform is designed in such a way that it
satisfies at least one of the following conditions thereby enabling root
port to exchange optimum number of FC (Flow Control) credits with
- downstream devices
+ downstream devices. This is applicable for only Tegra194.
1. If C0/C4/C5 run at x1/x2 link widths (irrespective of speed and MPS)
2. If C0/C1/C2/C3/C4/C5 operate at their respective max link widths and
a) speed is Gen-2 and MPS is 256B
@@ -119,6 +154,9 @@ Optional properties:
specified in microseconds
- nvidia,aspm-l0s-entrance-latency-us: ASPM L0s entrance latency to be
specified in microseconds
+- nvidia,enable-srns: This boolean property needs to be present if the
+ controller is configured to operate in SRNS (Separate Reference Clocks with
+ No Spread-Spectrum Clocking). This is applicable only for Tegra234.
RC mode:
- vpcie3v3-supply: A phandle to the regulator node that supplies 3.3V to the slot
@@ -127,6 +165,9 @@ RC mode:
- vpcie12v-supply: A phandle to the regulator node that supplies 12V to the slot
if the platform has one such slot. (Ex:- x16 slot owned by C5 controller
in p2972-0000 platform).
+- nvidia,enable-ext-refclk: This boolean property needs to be present if the
+ controller is configured to use the reference clocking coming in from an
+ external clock source instead of using the internal clock source.
EP mode:
- nvidia,refclk-select-gpios: Must contain a phandle to a GPIO controller
@@ -243,3 +284,62 @@ Tegra194 EP mode:
phy-names = "p2u-0", "p2u-1", "p2u-2", "p2u-3", "p2u-4",
"p2u-5", "p2u-6", "p2u-7";
};
+
+Tegra234 RP mode:
+-----------------
+
+ pcie@141a0000 {
+ compatible = "nvidia,tegra234-pcie";
+ power-domains = <&bpmp TEGRA234_POWER_DOMAIN_PCIEX8A>;
+ reg = <0x00 0x141a0000 0x0 0x00020000>, /* appl registers (128K) */
+ <0x00 0x3a000000 0x0 0x00040000>, /* configuration space (256K) */
+ <0x00 0x3a040000 0x0 0x00040000>, /* iATU_DMA reg space (256K) */
+ <0x00 0x3a080000 0x0 0x00040000>; /* DBI reg space (256K) */
+ reg-names = "appl", "config", "atu_dma", "dbi";
+
+ status = "disabled";
+
+ #address-cells = <3>;
+ #size-cells = <2>;
+ device_type = "pci";
+ num-lanes = <8>;
+ num-viewport = <8>;
+ linux,pci-domain = <5>;
+
+ clocks = <&bpmp TEGRA234_CLK_PEX1_C5_CORE>;
+ clock-names = "core";
+
+ resets = <&bpmp TEGRA234_RESET_PEX1_CORE_5_APB>,
+ <&bpmp TEGRA234_RESET_PEX1_CORE_5>;
+ reset-names = "apb", "core";
+
+ interrupts = <GIC_SPI 53 IRQ_TYPE_LEVEL_HIGH>, /* controller interrupt */
+ <GIC_SPI 54 IRQ_TYPE_LEVEL_HIGH>; /* MSI interrupt */
+ interrupt-names = "intr", "msi";
+
+ #interrupt-cells = <1>;
+ interrupt-map-mask = <0 0 0 0>;
+ interrupt-map = <0 0 0 0 &gic GIC_SPI 53 IRQ_TYPE_LEVEL_HIGH>;
+
+ nvidia,bpmp = <&bpmp 5>;
+
+ nvidia,aspm-cmrt-us = <60>;
+ nvidia,aspm-pwr-on-t-us = <20>;
+ nvidia,aspm-l0s-entrance-latency-us = <3>;
+
+ bus-range = <0x0 0xff>;
+
+ ranges = <0x43000000 0x27 0x40000000 0x27 0x40000000 0x3 0xe8000000>, /* prefetchable memory (16000 MB) */
+ <0x02000000 0x0 0x40000000 0x2b 0x28000000 0x0 0x08000000>, /* non-prefetchable memory (128 MB) */
+ <0x01000000 0x0 0x3a100000 0x00 0x3a100000 0x0 0x00100000>; /* downstream I/O (1 MB) */
+
+ interconnects = <&mc TEGRA234_MEMORY_CLIENT_PCIE5R &emc>,
+ <&mc TEGRA234_MEMORY_CLIENT_PCIE5W &emc>;
+ interconnect-names = "dma-mem", "write";
+
+ phys = <&p2u_nvhs_0>, <&p2u_nvhs_1>, <&p2u_nvhs_2>,
+ <&p2u_nvhs_3>, <&p2u_nvhs_4>, <&p2u_nvhs_5>,
+ <&p2u_nvhs_6>, <&p2u_nvhs_7>;
+ phy-names = "p2u-0", "p2u-1", "p2u-2", "p2u-3", "p2u-4",
+ "p2u-5", "p2u-6", "p2u-7";
+ };
--
2.17.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [PATCH V1 06/10] arm64: tegra: Add P2U and PCIe controller nodes to Tegra234 DT
2022-02-05 16:21 ` Vidya Sagar
@ 2022-02-05 16:21 ` Vidya Sagar
-1 siblings, 0 replies; 46+ messages in thread
From: Vidya Sagar @ 2022-02-05 16:21 UTC (permalink / raw)
To: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh
Cc: kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, vidyas, sagar.tv
Add P2U (PIPE to UPHY) and PCIe controller nodes to device tree.
The Tegra234 SoC contains 10 PCIe controllers and 24 P2U instances
grouped into three different PHY bricks namely High-Speed IO (HSIO-8 P2Us)
NVIDIA High Speed (NVHS-8 P2Us) and Gigabit Ethernet (GBE-8 P2Us)
respectively.
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
---
arch/arm64/boot/dts/nvidia/tegra234.dtsi | 743 +++++++++++++++++++++++
1 file changed, 743 insertions(+)
diff --git a/arch/arm64/boot/dts/nvidia/tegra234.dtsi b/arch/arm64/boot/dts/nvidia/tegra234.dtsi
index 6b6f15804a1a..f51ff940a25e 100644
--- a/arch/arm64/boot/dts/nvidia/tegra234.dtsi
+++ b/arch/arm64/boot/dts/nvidia/tegra234.dtsi
@@ -4,6 +4,7 @@
#include <dt-bindings/interrupt-controller/arm-gic.h>
#include <dt-bindings/mailbox/tegra186-hsp.h>
#include <dt-bindings/memory/tegra234-mc.h>
+#include <dt-bindings/power/tegra234-powergate.h>
#include <dt-bindings/reset/tegra234-reset.h>
/ {
@@ -197,6 +198,198 @@
#mbox-cells = <2>;
};
+ p2u_hsio_0: phy@3e00000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03e00000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_hsio_1: phy@3e10000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03e10000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_hsio_2: phy@3e20000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03e20000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_hsio_3: phy@3e30000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03e30000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_hsio_4: phy@3e40000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03e40000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_hsio_5: phy@3e50000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03e50000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_hsio_6: phy@3e60000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03e60000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_hsio_7: phy@3e70000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03e70000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_nvhs_0: phy@3e90000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03e90000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_nvhs_1: phy@3ea0000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03ea0000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_nvhs_2: phy@3eb0000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03eb0000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_nvhs_3: phy@3ec0000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03ec0000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_nvhs_4: phy@3ed0000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03ed0000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_nvhs_5: phy@3ee0000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03ee0000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_nvhs_6: phy@3ef0000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03ef0000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_nvhs_7: phy@3f00000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03f00000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_gbe_0: phy@3f20000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03f20000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_gbe_1: phy@3f30000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03f30000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_gbe_2: phy@3f40000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03f40000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_gbe_3: phy@3f50000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03f50000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_gbe_4: phy@3f60000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03f60000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_gbe_5: phy@3f70000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03f70000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_gbe_6: phy@3f80000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03f80000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_gbe_7: phy@3f90000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03f90000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
hsp_aon: hsp@c150000 {
compatible = "nvidia,tegra234-hsp", "nvidia,tegra194-hsp";
reg = <0x0c150000 0x90000>;
@@ -263,6 +456,556 @@
};
};
+ pcie@140a0000 {
+ compatible = "nvidia,tegra234-pcie";
+ power-domains = <&bpmp TEGRA234_POWER_DOMAIN_PCIEX4CA>;
+ reg = <0x00 0x140a0000 0x0 0x00020000>, /* appl registers (128K) */
+ <0x00 0x2a000000 0x0 0x00040000>, /* configuration space (256K) */
+ <0x00 0x2a040000 0x0 0x00040000>, /* iATU_DMA reg space (256K) */
+ <0x00 0x2a080000 0x0 0x00040000>; /* DBI reg space (256K) */
+ reg-names = "appl", "config", "atu_dma", "dbi";
+
+ status = "disabled";
+
+ #address-cells = <3>;
+ #size-cells = <2>;
+ device_type = "pci";
+ num-lanes = <4>;
+ num-viewport = <8>;
+ linux,pci-domain = <8>;
+
+ clocks = <&bpmp TEGRA234_CLK_PEX2_C8_CORE>;
+ clock-names = "core";
+
+ resets = <&bpmp TEGRA234_RESET_PEX2_CORE_8_APB>,
+ <&bpmp TEGRA234_RESET_PEX2_CORE_8>;
+ reset-names = "apb", "core";
+
+ interrupts = <GIC_SPI 356 IRQ_TYPE_LEVEL_HIGH>, /* controller interrupt */
+ <GIC_SPI 357 IRQ_TYPE_LEVEL_HIGH>; /* MSI interrupt */
+ interrupt-names = "intr", "msi";
+
+ #interrupt-cells = <1>;
+ interrupt-map-mask = <0 0 0 0>;
+ interrupt-map = <0 0 0 0 &gic GIC_SPI 356 IRQ_TYPE_LEVEL_HIGH>;
+
+ nvidia,bpmp = <&bpmp 8>;
+
+ nvidia,aspm-cmrt-us = <60>;
+ nvidia,aspm-pwr-on-t-us = <20>;
+ nvidia,aspm-l0s-entrance-latency-us = <3>;
+
+ bus-range = <0x0 0xff>;
+
+ ranges = <0x43000000 0x32 0x40000000 0x32 0x40000000 0x2 0xe8000000>, /* prefetchable memory (11904 MB) */
+ <0x02000000 0x0 0x40000000 0x35 0x28000000 0x0 0x08000000>, /* non-prefetchable memory (128 MB) */
+ <0x01000000 0x0 0x2a100000 0x00 0x2a100000 0x0 0x00100000>; /* downstream I/O (1 MB) */
+
+ interconnects = <&mc TEGRA234_MEMORY_CLIENT_PCIE8AR &emc>,
+ <&mc TEGRA234_MEMORY_CLIENT_PCIE8AW &emc>;
+ interconnect-names = "dma-mem", "write";
+ };
+
+ pcie@140c0000 {
+ compatible = "nvidia,tegra234-pcie";
+ power-domains = <&bpmp TEGRA234_POWER_DOMAIN_PCIEX4CB>;
+ reg = <0x00 0x140c0000 0x0 0x00020000>, /* appl registers (128K) */
+ <0x00 0x2c000000 0x0 0x00040000>, /* configuration space (256K) */
+ <0x00 0x2c040000 0x0 0x00040000>, /* iATU_DMA reg space (256K) */
+ <0x00 0x2c080000 0x0 0x00040000>; /* DBI reg space (256K) */
+ reg-names = "appl", "config", "atu_dma", "dbi";
+
+ status = "disabled";
+
+ #address-cells = <3>;
+ #size-cells = <2>;
+ device_type = "pci";
+ num-lanes = <4>;
+ num-viewport = <8>;
+ linux,pci-domain = <9>;
+
+ clocks = <&bpmp TEGRA234_CLK_PEX2_C9_CORE>;
+ clock-names = "core";
+
+ resets = <&bpmp TEGRA234_RESET_PEX2_CORE_9_APB>,
+ <&bpmp TEGRA234_RESET_PEX2_CORE_9>;
+ reset-names = "apb", "core";
+
+ interrupts = <GIC_SPI 358 IRQ_TYPE_LEVEL_HIGH>, /* controller interrupt */
+ <GIC_SPI 359 IRQ_TYPE_LEVEL_HIGH>; /* MSI interrupt */
+ interrupt-names = "intr", "msi";
+
+ #interrupt-cells = <1>;
+ interrupt-map-mask = <0 0 0 0>;
+ interrupt-map = <0 0 0 0 &gic GIC_SPI 358 IRQ_TYPE_LEVEL_HIGH>;
+
+ nvidia,bpmp = <&bpmp 9>;
+
+ nvidia,aspm-cmrt-us = <60>;
+ nvidia,aspm-pwr-on-t-us = <20>;
+ nvidia,aspm-l0s-entrance-latency-us = <3>;
+
+ bus-range = <0x0 0xff>;
+
+ ranges = <0x43000000 0x35 0x40000000 0x35 0x40000000 0x2 0xe8000000>, /* prefetchable memory (11904 MB) */
+ <0x02000000 0x0 0x40000000 0x38 0x28000000 0x0 0x08000000>, /* non-prefetchable memory (128 MB) */
+ <0x01000000 0x0 0x2c100000 0x00 0x2c100000 0x0 0x00100000>; /* downstream I/O (1 MB) */
+
+ interconnects = <&mc TEGRA234_MEMORY_CLIENT_PCIE9AR &emc>,
+ <&mc TEGRA234_MEMORY_CLIENT_PCIE9AW &emc>;
+ interconnect-names = "dma-mem", "write";
+ };
+
+ pcie@140e0000 {
+ compatible = "nvidia,tegra234-pcie";
+ power-domains = <&bpmp TEGRA234_POWER_DOMAIN_PCIEX4CC>;
+ reg = <0x00 0x140e0000 0x0 0x00020000>, /* appl registers (128K) */
+ <0x00 0x2e000000 0x0 0x00040000>, /* configuration space (256K) */
+ <0x00 0x2e040000 0x0 0x00040000>, /* iATU_DMA reg space (256K) */
+ <0x00 0x2e080000 0x0 0x00040000>; /* DBI reg space (256K) */
+ reg-names = "appl", "config", "atu_dma", "dbi";
+
+ status = "disabled";
+
+ #address-cells = <3>;
+ #size-cells = <2>;
+ device_type = "pci";
+ num-lanes = <4>;
+ num-viewport = <8>;
+ linux,pci-domain = <10>;
+
+ clocks = <&bpmp TEGRA234_CLK_PEX2_C10_CORE>;
+ clock-names = "core";
+
+ resets = <&bpmp TEGRA234_RESET_PEX2_CORE_10_APB>,
+ <&bpmp TEGRA234_RESET_PEX2_CORE_10>;
+ reset-names = "apb", "core";
+
+ interrupts = <GIC_SPI 360 IRQ_TYPE_LEVEL_HIGH>, /* controller interrupt */
+ <GIC_SPI 361 IRQ_TYPE_LEVEL_HIGH>; /* MSI interrupt */
+ interrupt-names = "intr", "msi";
+
+ #interrupt-cells = <1>;
+ interrupt-map-mask = <0 0 0 0>;
+ interrupt-map = <0 0 0 0 &gic GIC_SPI 360 IRQ_TYPE_LEVEL_HIGH>;
+
+ nvidia,bpmp = <&bpmp 10>;
+
+ nvidia,aspm-cmrt-us = <60>;
+ nvidia,aspm-pwr-on-t-us = <20>;
+ nvidia,aspm-l0s-entrance-latency-us = <3>;
+
+ bus-range = <0x0 0xff>;
+
+ ranges = <0x43000000 0x38 0x40000000 0x38 0x40000000 0x2 0xe8000000>, /* prefetchable memory (11904 MB) */
+ <0x02000000 0x0 0x40000000 0x3b 0x28000000 0x0 0x08000000>, /* non-prefetchable memory (128 MB) */
+ <0x01000000 0x0 0x2e100000 0x00 0x2e100000 0x0 0x00100000>; /* downstream I/O (1 MB) */
+
+ interconnects = <&mc TEGRA234_MEMORY_CLIENT_PCIE10AR &emc>,
+ <&mc TEGRA234_MEMORY_CLIENT_PCIE10AW &emc>;
+ interconnect-names = "dma-mem", "write";
+ };
+
+ pcie@14100000 {
+ compatible = "nvidia,tegra234-pcie";
+ power-domains = <&bpmp TEGRA234_POWER_DOMAIN_PCIEX1A>;
+ reg = <0x00 0x14100000 0x0 0x00020000>, /* appl registers (128K) */
+ <0x00 0x30000000 0x0 0x00040000>, /* configuration space (256K) */
+ <0x00 0x30040000 0x0 0x00040000>, /* iATU_DMA reg space (256K) */
+ <0x00 0x30080000 0x0 0x00040000>; /* DBI reg space (256K) */
+ reg-names = "appl", "config", "atu_dma", "dbi";
+
+ status = "disabled";
+
+ #address-cells = <3>;
+ #size-cells = <2>;
+ device_type = "pci";
+ num-lanes = <1>;
+ num-viewport = <8>;
+ linux,pci-domain = <1>;
+
+ clocks = <&bpmp TEGRA234_CLK_PEX0_C1_CORE>;
+ clock-names = "core";
+
+ resets = <&bpmp TEGRA234_RESET_PEX0_CORE_1_APB>,
+ <&bpmp TEGRA234_RESET_PEX0_CORE_1>;
+ reset-names = "apb", "core";
+
+ interrupts = <GIC_SPI 45 IRQ_TYPE_LEVEL_HIGH>, /* controller interrupt */
+ <GIC_SPI 46 IRQ_TYPE_LEVEL_HIGH>; /* MSI interrupt */
+ interrupt-names = "intr", "msi";
+
+ #interrupt-cells = <1>;
+ interrupt-map-mask = <0 0 0 0>;
+ interrupt-map = <0 0 0 0 &gic GIC_SPI 45 IRQ_TYPE_LEVEL_HIGH>;
+
+ nvidia,bpmp = <&bpmp 1>;
+
+ nvidia,aspm-cmrt-us = <60>;
+ nvidia,aspm-pwr-on-t-us = <20>;
+ nvidia,aspm-l0s-entrance-latency-us = <3>;
+
+ bus-range = <0x0 0xff>;
+
+ ranges = <0x43000000 0x20 0x80000000 0x20 0x80000000 0x0 0x28000000>, /* prefetchable memory (640 MB) */
+ <0x02000000 0x0 0x40000000 0x20 0xa8000000 0x0 0x08000000>, /* non-prefetchable memory (128 MB) */
+ <0x01000000 0x0 0x30100000 0x00 0x30100000 0x0 0x00100000>; /* downstream I/O (1 MB) */
+
+ interconnects = <&mc TEGRA234_MEMORY_CLIENT_PCIE1R &emc>,
+ <&mc TEGRA234_MEMORY_CLIENT_PCIE1W &emc>;
+ interconnect-names = "dma-mem", "write";
+ };
+
+ pcie@14120000 {
+ compatible = "nvidia,tegra234-pcie";
+ power-domains = <&bpmp TEGRA234_POWER_DOMAIN_PCIEX1A>;
+ reg = <0x00 0x14120000 0x0 0x00020000>, /* appl registers (128K) */
+ <0x00 0x32000000 0x0 0x00040000>, /* configuration space (256K) */
+ <0x00 0x32040000 0x0 0x00040000>, /* iATU_DMA reg space (256K) */
+ <0x00 0x32080000 0x0 0x00040000>; /* DBI reg space (256K) */
+ reg-names = "appl", "config", "atu_dma", "dbi";
+
+ status = "disabled";
+
+ #address-cells = <3>;
+ #size-cells = <2>;
+ device_type = "pci";
+ num-lanes = <1>;
+ num-viewport = <8>;
+ linux,pci-domain = <2>;
+
+ clocks = <&bpmp TEGRA234_CLK_PEX0_C2_CORE>;
+ clock-names = "core";
+
+ resets = <&bpmp TEGRA234_RESET_PEX0_CORE_2_APB>,
+ <&bpmp TEGRA234_RESET_PEX0_CORE_2>;
+ reset-names = "apb", "core";
+
+ interrupts = <GIC_SPI 47 IRQ_TYPE_LEVEL_HIGH>, /* controller interrupt */
+ <GIC_SPI 48 IRQ_TYPE_LEVEL_HIGH>; /* MSI interrupt */
+ interrupt-names = "intr", "msi";
+
+ #interrupt-cells = <1>;
+ interrupt-map-mask = <0 0 0 0>;
+ interrupt-map = <0 0 0 0 &gic GIC_SPI 47 IRQ_TYPE_LEVEL_HIGH>;
+
+ nvidia,bpmp = <&bpmp 2>;
+
+ nvidia,aspm-cmrt-us = <60>;
+ nvidia,aspm-pwr-on-t-us = <20>;
+ nvidia,aspm-l0s-entrance-latency-us = <3>;
+
+ bus-range = <0x0 0xff>;
+
+ ranges = <0x43000000 0x20 0xc0000000 0x20 0xc0000000 0x0 0x28000000>, /* prefetchable memory (640 MB) */
+ <0x02000000 0x0 0x40000000 0x20 0xe8000000 0x0 0x08000000>, /* non-prefetchable memory (128 MB) */
+ <0x01000000 0x0 0x32100000 0x00 0x32100000 0x0 0x00100000>; /* downstream I/O (1 MB) */
+
+ interconnects = <&mc TEGRA234_MEMORY_CLIENT_PCIE2AR &emc>,
+ <&mc TEGRA234_MEMORY_CLIENT_PCIE2AW &emc>;
+ interconnect-names = "dma-mem", "write";
+ };
+
+ pcie@14140000 {
+ compatible = "nvidia,tegra234-pcie";
+ power-domains = <&bpmp TEGRA234_POWER_DOMAIN_PCIEX1A>;
+ reg = <0x00 0x14140000 0x0 0x00020000>, /* appl registers (128K) */
+ <0x00 0x34000000 0x0 0x00040000>, /* configuration space (256K) */
+ <0x00 0x34040000 0x0 0x00040000>, /* iATU_DMA reg space (256K) */
+ <0x00 0x34080000 0x0 0x00040000>; /* DBI reg space (256K) */
+ reg-names = "appl", "config", "atu_dma", "dbi";
+
+ status = "disabled";
+
+ #address-cells = <3>;
+ #size-cells = <2>;
+ device_type = "pci";
+ num-lanes = <1>;
+ num-viewport = <8>;
+ linux,pci-domain = <3>;
+
+ clocks = <&bpmp TEGRA234_CLK_PEX0_C3_CORE>;
+ clock-names = "core";
+
+ resets = <&bpmp TEGRA234_RESET_PEX0_CORE_3_APB>,
+ <&bpmp TEGRA234_RESET_PEX0_CORE_3>;
+ reset-names = "apb", "core";
+
+ interrupts = <GIC_SPI 49 IRQ_TYPE_LEVEL_HIGH>, /* controller interrupt */
+ <GIC_SPI 50 IRQ_TYPE_LEVEL_HIGH>; /* MSI interrupt */
+ interrupt-names = "intr", "msi";
+
+ #interrupt-cells = <1>;
+ interrupt-map-mask = <0 0 0 0>;
+ interrupt-map = <0 0 0 0 &gic GIC_SPI 49 IRQ_TYPE_LEVEL_HIGH>;
+
+ nvidia,bpmp = <&bpmp 3>;
+
+ nvidia,aspm-cmrt-us = <60>;
+ nvidia,aspm-pwr-on-t-us = <20>;
+ nvidia,aspm-l0s-entrance-latency-us = <3>;
+
+ bus-range = <0x0 0xff>;
+
+ ranges = <0x43000000 0x21 0x00000000 0x21 0x00000000 0x0 0x28000000>, /* prefetchable memory (640 MB) */
+ <0x02000000 0x0 0x40000000 0x21 0xe8000000 0x0 0x08000000>, /* non-prefetchable memory (128 MB) */
+ <0x01000000 0x0 0x34100000 0x00 0x34100000 0x0 0x00100000>; /* downstream I/O (1 MB) */
+
+ interconnects = <&mc TEGRA234_MEMORY_CLIENT_PCIE3R &emc>,
+ <&mc TEGRA234_MEMORY_CLIENT_PCIE3W &emc>;
+ interconnect-names = "dma-mem", "write";
+ };
+
+ pcie@14160000 {
+ compatible = "nvidia,tegra234-pcie";
+ power-domains = <&bpmp TEGRA234_POWER_DOMAIN_PCIEX4BB>;
+ reg = <0x00 0x14160000 0x0 0x00020000>, /* appl registers (128K) */
+ <0x00 0x36000000 0x0 0x00040000>, /* configuration space (256K) */
+ <0x00 0x36040000 0x0 0x00040000>, /* iATU_DMA reg space (256K) */
+ <0x00 0x36080000 0x0 0x00040000>; /* DBI reg space (256K) */
+ reg-names = "appl", "config", "atu_dma", "dbi";
+
+ status = "disabled";
+
+ #address-cells = <3>;
+ #size-cells = <2>;
+ device_type = "pci";
+ num-lanes = <4>;
+ num-viewport = <8>;
+ linux,pci-domain = <4>;
+
+ clocks = <&bpmp TEGRA234_CLK_PEX0_C4_CORE>;
+ clock-names = "core";
+
+ resets = <&bpmp TEGRA234_RESET_PEX0_CORE_4_APB>,
+ <&bpmp TEGRA234_RESET_PEX0_CORE_4>;
+ reset-names = "apb", "core";
+
+ interrupts = <GIC_SPI 51 IRQ_TYPE_LEVEL_HIGH>, /* controller interrupt */
+ <GIC_SPI 52 IRQ_TYPE_LEVEL_HIGH>; /* MSI interrupt */
+ interrupt-names = "intr", "msi";
+
+ #interrupt-cells = <1>;
+ interrupt-map-mask = <0 0 0 0>;
+ interrupt-map = <0 0 0 0 &gic GIC_SPI 51 IRQ_TYPE_LEVEL_HIGH>;
+
+ nvidia,bpmp = <&bpmp 4>;
+
+ nvidia,aspm-cmrt-us = <60>;
+ nvidia,aspm-pwr-on-t-us = <20>;
+ nvidia,aspm-l0s-entrance-latency-us = <3>;
+
+ bus-range = <0x0 0xff>;
+
+ ranges = <0x43000000 0x21 0x40000000 0x21 0x40000000 0x2 0xe8000000>, /* prefetchable memory (11904 MB) */
+ <0x02000000 0x0 0x40000000 0x24 0x28000000 0x0 0x08000000>, /* non-prefetchable memory (128 MB) */
+ <0x01000000 0x0 0x36100000 0x00 0x36100000 0x0 0x00100000>; /* downstream I/O (1 MB) */
+
+ interconnects = <&mc TEGRA234_MEMORY_CLIENT_PCIE4R &emc>,
+ <&mc TEGRA234_MEMORY_CLIENT_PCIE4W &emc>;
+ interconnect-names = "dma-mem", "write";
+ };
+
+ pcie@14180000 {
+ compatible = "nvidia,tegra234-pcie";
+ power-domains = <&bpmp TEGRA234_POWER_DOMAIN_PCIEX4BA>;
+ reg = <0x00 0x14180000 0x0 0x00020000>, /* appl registers (128K) */
+ <0x00 0x38000000 0x0 0x00040000>, /* configuration space (256K) */
+ <0x00 0x38040000 0x0 0x00040000>, /* iATU_DMA reg space (256K) */
+ <0x00 0x38080000 0x0 0x00040000>; /* DBI reg space (256K) */
+ reg-names = "appl", "config", "atu_dma", "dbi";
+
+ status = "disabled";
+
+ #address-cells = <3>;
+ #size-cells = <2>;
+ device_type = "pci";
+ num-lanes = <4>;
+ num-viewport = <8>;
+ linux,pci-domain = <0>;
+
+ clocks = <&bpmp TEGRA234_CLK_PEX0_C0_CORE>;
+ clock-names = "core";
+
+ resets = <&bpmp TEGRA234_RESET_PEX0_CORE_0_APB>,
+ <&bpmp TEGRA234_RESET_PEX0_CORE_0>;
+ reset-names = "apb", "core";
+
+ interrupts = <GIC_SPI 72 IRQ_TYPE_LEVEL_HIGH>, /* controller interrupt */
+ <GIC_SPI 73 IRQ_TYPE_LEVEL_HIGH>; /* MSI interrupt */
+ interrupt-names = "intr", "msi";
+
+ #interrupt-cells = <1>;
+ interrupt-map-mask = <0 0 0 0>;
+ interrupt-map = <0 0 0 0 &gic GIC_SPI 72 IRQ_TYPE_LEVEL_HIGH>;
+
+ nvidia,bpmp = <&bpmp 0>;
+
+ nvidia,aspm-cmrt-us = <60>;
+ nvidia,aspm-pwr-on-t-us = <20>;
+ nvidia,aspm-l0s-entrance-latency-us = <3>;
+
+ bus-range = <0x0 0xff>;
+
+ ranges = <0x43000000 0x24 0x40000000 0x24 0x40000000 0x2 0xe8000000>, /* prefetchable memory (11904 MB) */
+ <0x02000000 0x0 0x40000000 0x27 0x28000000 0x0 0x08000000>, /* non-prefetchable memory (128 MB) */
+ <0x01000000 0x0 0x38100000 0x00 0x38100000 0x0 0x00100000>; /* downstream I/O (1 MB) */
+
+ interconnects = <&mc TEGRA234_MEMORY_CLIENT_PCIE0R &emc>,
+ <&mc TEGRA234_MEMORY_CLIENT_PCIE0W &emc>;
+ interconnect-names = "dma-mem", "write";
+ };
+
+ pcie@141a0000 {
+ compatible = "nvidia,tegra234-pcie";
+ power-domains = <&bpmp TEGRA234_POWER_DOMAIN_PCIEX8A>;
+ reg = <0x00 0x141a0000 0x0 0x00020000>, /* appl registers (128K) */
+ <0x00 0x3a000000 0x0 0x00040000>, /* configuration space (256K) */
+ <0x00 0x3a040000 0x0 0x00040000>, /* iATU_DMA reg space (256K) */
+ <0x00 0x3a080000 0x0 0x00040000>; /* DBI reg space (256K) */
+ reg-names = "appl", "config", "atu_dma", "dbi";
+
+ status = "disabled";
+
+ #address-cells = <3>;
+ #size-cells = <2>;
+ device_type = "pci";
+ num-lanes = <8>;
+ num-viewport = <8>;
+ linux,pci-domain = <5>;
+
+ clocks = <&bpmp TEGRA234_CLK_PEX1_C5_CORE>;
+ clock-names = "core";
+
+ resets = <&bpmp TEGRA234_RESET_PEX1_CORE_5_APB>,
+ <&bpmp TEGRA234_RESET_PEX1_CORE_5>;
+ reset-names = "apb", "core";
+
+ interrupts = <GIC_SPI 53 IRQ_TYPE_LEVEL_HIGH>, /* controller interrupt */
+ <GIC_SPI 54 IRQ_TYPE_LEVEL_HIGH>; /* MSI interrupt */
+ interrupt-names = "intr", "msi";
+
+ #interrupt-cells = <1>;
+ interrupt-map-mask = <0 0 0 0>;
+ interrupt-map = <0 0 0 0 &gic GIC_SPI 53 IRQ_TYPE_LEVEL_HIGH>;
+
+ nvidia,bpmp = <&bpmp 5>;
+
+ nvidia,aspm-cmrt-us = <60>;
+ nvidia,aspm-pwr-on-t-us = <20>;
+ nvidia,aspm-l0s-entrance-latency-us = <3>;
+
+ bus-range = <0x0 0xff>;
+
+ ranges = <0x43000000 0x27 0x40000000 0x27 0x40000000 0x3 0xe8000000>, /* prefetchable memory (16000 MB) */
+ <0x02000000 0x0 0x40000000 0x2b 0x28000000 0x0 0x08000000>, /* non-prefetchable memory (128 MB) */
+ <0x01000000 0x0 0x3a100000 0x00 0x3a100000 0x0 0x00100000>; /* downstream I/O (1 MB) */
+
+ interconnects = <&mc TEGRA234_MEMORY_CLIENT_PCIE5R &emc>,
+ <&mc TEGRA234_MEMORY_CLIENT_PCIE5W &emc>;
+ interconnect-names = "dma-mem", "write";
+ };
+
+ pcie@141c0000 {
+ compatible = "nvidia,tegra234-pcie";
+ power-domains = <&bpmp TEGRA234_POWER_DOMAIN_PCIEX4A>;
+ reg = <0x00 0x141c0000 0x0 0x00020000>, /* appl registers (128K) */
+ <0x00 0x3c000000 0x0 0x00040000>, /* configuration space (256K) */
+ <0x00 0x3c040000 0x0 0x00040000>, /* iATU_DMA reg space (256K) */
+ <0x00 0x3c080000 0x0 0x00040000>; /* DBI reg space (256K) */
+ reg-names = "appl", "config", "atu_dma", "dbi";
+
+ status = "disabled";
+
+ #address-cells = <3>;
+ #size-cells = <2>;
+ device_type = "pci";
+ num-lanes = <4>;
+ num-viewport = <8>;
+ linux,pci-domain = <6>;
+
+ clocks = <&bpmp TEGRA234_CLK_PEX1_C6_CORE>;
+ clock-names = "core";
+
+ resets = <&bpmp TEGRA234_RESET_PEX1_CORE_6_APB>,
+ <&bpmp TEGRA234_RESET_PEX1_CORE_6>;
+ reset-names = "apb", "core";
+
+ interrupts = <GIC_SPI 352 IRQ_TYPE_LEVEL_HIGH>, /* controller interrupt */
+ <GIC_SPI 353 IRQ_TYPE_LEVEL_HIGH>; /* MSI interrupt */
+ interrupt-names = "intr", "msi";
+
+ #interrupt-cells = <1>;
+ interrupt-map-mask = <0 0 0 0>;
+ interrupt-map = <0 0 0 0 &gic GIC_SPI 352 IRQ_TYPE_LEVEL_HIGH>;
+
+ nvidia,bpmp = <&bpmp 6>;
+
+ nvidia,aspm-cmrt-us = <60>;
+ nvidia,aspm-pwr-on-t-us = <20>;
+ nvidia,aspm-l0s-entrance-latency-us = <3>;
+
+ bus-range = <0x0 0xff>;
+
+ ranges = <0x43000000 0x2b 0x40000000 0x2b 0x40000000 0x2 0xe8000000>, /* prefetchable memory (11904 MB) */
+ <0x02000000 0x0 0x40000000 0x2e 0x28000000 0x0 0x08000000>, /* non-prefetchable memory (128 MB) */
+ <0x01000000 0x0 0x3c100000 0x00 0x3c100000 0x0 0x00100000>; /* downstream I/O (1 MB) */
+
+ interconnects = <&mc TEGRA234_MEMORY_CLIENT_PCIE6AR &emc>,
+ <&mc TEGRA234_MEMORY_CLIENT_PCIE6AW &emc>;
+ interconnect-names = "dma-mem", "write";
+ };
+
+ pcie@141e0000 {
+ compatible = "nvidia,tegra234-pcie";
+ power-domains = <&bpmp TEGRA234_POWER_DOMAIN_PCIEX8B>;
+ reg = <0x00 0x141e0000 0x0 0x00020000>, /* appl registers (128K) */
+ <0x00 0x3e000000 0x0 0x00040000>, /* configuration space (256K) */
+ <0x00 0x3e040000 0x0 0x00040000>, /* iATU_DMA reg space (256K) */
+ <0x00 0x3e080000 0x0 0x00040000>; /* DBI reg space (256K) */
+ reg-names = "appl", "config", "atu_dma", "dbi";
+
+ status = "disabled";
+
+ #address-cells = <3>;
+ #size-cells = <2>;
+ device_type = "pci";
+ num-lanes = <8>;
+ num-viewport = <8>;
+ linux,pci-domain = <7>;
+
+ clocks = <&bpmp TEGRA234_CLK_PEX2_C7_CORE>;
+ clock-names = "core";
+
+ resets = <&bpmp TEGRA234_RESET_PEX2_CORE_7_APB>,
+ <&bpmp TEGRA234_RESET_PEX2_CORE_7>;
+ reset-names = "apb", "core";
+
+ interrupts = <GIC_SPI 354 IRQ_TYPE_LEVEL_HIGH>, /* controller interrupt */
+ <GIC_SPI 355 IRQ_TYPE_LEVEL_HIGH>; /* MSI interrupt */
+ interrupt-names = "intr", "msi";
+
+ #interrupt-cells = <1>;
+ interrupt-map-mask = <0 0 0 0>;
+ interrupt-map = <0 0 0 0 &gic GIC_SPI 354 IRQ_TYPE_LEVEL_HIGH>;
+
+ nvidia,bpmp = <&bpmp 7>;
+
+ nvidia,aspm-cmrt-us = <60>;
+ nvidia,aspm-pwr-on-t-us = <20>;
+ nvidia,aspm-l0s-entrance-latency-us = <3>;
+
+ bus-range = <0x0 0xff>;
+
+ ranges = <0x43000000 0x2e 0x40000000 0x2e 0x40000000 0x3 0xe8000000>, /* prefetchable memory (16000 MB) */
+ <0x02000000 0x0 0x40000000 0x32 0x28000000 0x0 0x08000000>, /* non-prefetchable memory (128 MB) */
+ <0x01000000 0x0 0x3e100000 0x00 0x3e100000 0x0 0x00100000>; /* downstream I/O (1 MB) */
+
+ interconnects = <&mc TEGRA234_MEMORY_CLIENT_PCIE7AR &emc>,
+ <&mc TEGRA234_MEMORY_CLIENT_PCIE7AW &emc>;
+ interconnect-names = "dma-mem", "write";
+ };
+
sram@40000000 {
compatible = "nvidia,tegra234-sysram", "mmio-sram";
reg = <0x0 0x40000000 0x0 0x80000>;
--
2.17.1
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [PATCH V1 06/10] arm64: tegra: Add P2U and PCIe controller nodes to Tegra234 DT
@ 2022-02-05 16:21 ` Vidya Sagar
0 siblings, 0 replies; 46+ messages in thread
From: Vidya Sagar @ 2022-02-05 16:21 UTC (permalink / raw)
To: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh
Cc: kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, vidyas, sagar.tv
Add P2U (PIPE to UPHY) and PCIe controller nodes to device tree.
The Tegra234 SoC contains 10 PCIe controllers and 24 P2U instances
grouped into three different PHY bricks namely High-Speed IO (HSIO-8 P2Us)
NVIDIA High Speed (NVHS-8 P2Us) and Gigabit Ethernet (GBE-8 P2Us)
respectively.
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
---
arch/arm64/boot/dts/nvidia/tegra234.dtsi | 743 +++++++++++++++++++++++
1 file changed, 743 insertions(+)
diff --git a/arch/arm64/boot/dts/nvidia/tegra234.dtsi b/arch/arm64/boot/dts/nvidia/tegra234.dtsi
index 6b6f15804a1a..f51ff940a25e 100644
--- a/arch/arm64/boot/dts/nvidia/tegra234.dtsi
+++ b/arch/arm64/boot/dts/nvidia/tegra234.dtsi
@@ -4,6 +4,7 @@
#include <dt-bindings/interrupt-controller/arm-gic.h>
#include <dt-bindings/mailbox/tegra186-hsp.h>
#include <dt-bindings/memory/tegra234-mc.h>
+#include <dt-bindings/power/tegra234-powergate.h>
#include <dt-bindings/reset/tegra234-reset.h>
/ {
@@ -197,6 +198,198 @@
#mbox-cells = <2>;
};
+ p2u_hsio_0: phy@3e00000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03e00000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_hsio_1: phy@3e10000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03e10000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_hsio_2: phy@3e20000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03e20000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_hsio_3: phy@3e30000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03e30000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_hsio_4: phy@3e40000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03e40000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_hsio_5: phy@3e50000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03e50000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_hsio_6: phy@3e60000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03e60000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_hsio_7: phy@3e70000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03e70000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_nvhs_0: phy@3e90000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03e90000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_nvhs_1: phy@3ea0000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03ea0000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_nvhs_2: phy@3eb0000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03eb0000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_nvhs_3: phy@3ec0000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03ec0000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_nvhs_4: phy@3ed0000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03ed0000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_nvhs_5: phy@3ee0000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03ee0000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_nvhs_6: phy@3ef0000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03ef0000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_nvhs_7: phy@3f00000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03f00000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_gbe_0: phy@3f20000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03f20000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_gbe_1: phy@3f30000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03f30000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_gbe_2: phy@3f40000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03f40000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_gbe_3: phy@3f50000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03f50000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_gbe_4: phy@3f60000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03f60000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_gbe_5: phy@3f70000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03f70000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_gbe_6: phy@3f80000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03f80000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
+ p2u_gbe_7: phy@3f90000 {
+ compatible = "nvidia,tegra234-p2u";
+ reg = <0x03f90000 0x10000>;
+ reg-names = "ctl";
+
+ #phy-cells = <0>;
+ };
+
hsp_aon: hsp@c150000 {
compatible = "nvidia,tegra234-hsp", "nvidia,tegra194-hsp";
reg = <0x0c150000 0x90000>;
@@ -263,6 +456,556 @@
};
};
+ pcie@140a0000 {
+ compatible = "nvidia,tegra234-pcie";
+ power-domains = <&bpmp TEGRA234_POWER_DOMAIN_PCIEX4CA>;
+ reg = <0x00 0x140a0000 0x0 0x00020000>, /* appl registers (128K) */
+ <0x00 0x2a000000 0x0 0x00040000>, /* configuration space (256K) */
+ <0x00 0x2a040000 0x0 0x00040000>, /* iATU_DMA reg space (256K) */
+ <0x00 0x2a080000 0x0 0x00040000>; /* DBI reg space (256K) */
+ reg-names = "appl", "config", "atu_dma", "dbi";
+
+ status = "disabled";
+
+ #address-cells = <3>;
+ #size-cells = <2>;
+ device_type = "pci";
+ num-lanes = <4>;
+ num-viewport = <8>;
+ linux,pci-domain = <8>;
+
+ clocks = <&bpmp TEGRA234_CLK_PEX2_C8_CORE>;
+ clock-names = "core";
+
+ resets = <&bpmp TEGRA234_RESET_PEX2_CORE_8_APB>,
+ <&bpmp TEGRA234_RESET_PEX2_CORE_8>;
+ reset-names = "apb", "core";
+
+ interrupts = <GIC_SPI 356 IRQ_TYPE_LEVEL_HIGH>, /* controller interrupt */
+ <GIC_SPI 357 IRQ_TYPE_LEVEL_HIGH>; /* MSI interrupt */
+ interrupt-names = "intr", "msi";
+
+ #interrupt-cells = <1>;
+ interrupt-map-mask = <0 0 0 0>;
+ interrupt-map = <0 0 0 0 &gic GIC_SPI 356 IRQ_TYPE_LEVEL_HIGH>;
+
+ nvidia,bpmp = <&bpmp 8>;
+
+ nvidia,aspm-cmrt-us = <60>;
+ nvidia,aspm-pwr-on-t-us = <20>;
+ nvidia,aspm-l0s-entrance-latency-us = <3>;
+
+ bus-range = <0x0 0xff>;
+
+ ranges = <0x43000000 0x32 0x40000000 0x32 0x40000000 0x2 0xe8000000>, /* prefetchable memory (11904 MB) */
+ <0x02000000 0x0 0x40000000 0x35 0x28000000 0x0 0x08000000>, /* non-prefetchable memory (128 MB) */
+ <0x01000000 0x0 0x2a100000 0x00 0x2a100000 0x0 0x00100000>; /* downstream I/O (1 MB) */
+
+ interconnects = <&mc TEGRA234_MEMORY_CLIENT_PCIE8AR &emc>,
+ <&mc TEGRA234_MEMORY_CLIENT_PCIE8AW &emc>;
+ interconnect-names = "dma-mem", "write";
+ };
+
+ pcie@140c0000 {
+ compatible = "nvidia,tegra234-pcie";
+ power-domains = <&bpmp TEGRA234_POWER_DOMAIN_PCIEX4CB>;
+ reg = <0x00 0x140c0000 0x0 0x00020000>, /* appl registers (128K) */
+ <0x00 0x2c000000 0x0 0x00040000>, /* configuration space (256K) */
+ <0x00 0x2c040000 0x0 0x00040000>, /* iATU_DMA reg space (256K) */
+ <0x00 0x2c080000 0x0 0x00040000>; /* DBI reg space (256K) */
+ reg-names = "appl", "config", "atu_dma", "dbi";
+
+ status = "disabled";
+
+ #address-cells = <3>;
+ #size-cells = <2>;
+ device_type = "pci";
+ num-lanes = <4>;
+ num-viewport = <8>;
+ linux,pci-domain = <9>;
+
+ clocks = <&bpmp TEGRA234_CLK_PEX2_C9_CORE>;
+ clock-names = "core";
+
+ resets = <&bpmp TEGRA234_RESET_PEX2_CORE_9_APB>,
+ <&bpmp TEGRA234_RESET_PEX2_CORE_9>;
+ reset-names = "apb", "core";
+
+ interrupts = <GIC_SPI 358 IRQ_TYPE_LEVEL_HIGH>, /* controller interrupt */
+ <GIC_SPI 359 IRQ_TYPE_LEVEL_HIGH>; /* MSI interrupt */
+ interrupt-names = "intr", "msi";
+
+ #interrupt-cells = <1>;
+ interrupt-map-mask = <0 0 0 0>;
+ interrupt-map = <0 0 0 0 &gic GIC_SPI 358 IRQ_TYPE_LEVEL_HIGH>;
+
+ nvidia,bpmp = <&bpmp 9>;
+
+ nvidia,aspm-cmrt-us = <60>;
+ nvidia,aspm-pwr-on-t-us = <20>;
+ nvidia,aspm-l0s-entrance-latency-us = <3>;
+
+ bus-range = <0x0 0xff>;
+
+ ranges = <0x43000000 0x35 0x40000000 0x35 0x40000000 0x2 0xe8000000>, /* prefetchable memory (11904 MB) */
+ <0x02000000 0x0 0x40000000 0x38 0x28000000 0x0 0x08000000>, /* non-prefetchable memory (128 MB) */
+ <0x01000000 0x0 0x2c100000 0x00 0x2c100000 0x0 0x00100000>; /* downstream I/O (1 MB) */
+
+ interconnects = <&mc TEGRA234_MEMORY_CLIENT_PCIE9AR &emc>,
+ <&mc TEGRA234_MEMORY_CLIENT_PCIE9AW &emc>;
+ interconnect-names = "dma-mem", "write";
+ };
+
+ pcie@140e0000 {
+ compatible = "nvidia,tegra234-pcie";
+ power-domains = <&bpmp TEGRA234_POWER_DOMAIN_PCIEX4CC>;
+ reg = <0x00 0x140e0000 0x0 0x00020000>, /* appl registers (128K) */
+ <0x00 0x2e000000 0x0 0x00040000>, /* configuration space (256K) */
+ <0x00 0x2e040000 0x0 0x00040000>, /* iATU_DMA reg space (256K) */
+ <0x00 0x2e080000 0x0 0x00040000>; /* DBI reg space (256K) */
+ reg-names = "appl", "config", "atu_dma", "dbi";
+
+ status = "disabled";
+
+ #address-cells = <3>;
+ #size-cells = <2>;
+ device_type = "pci";
+ num-lanes = <4>;
+ num-viewport = <8>;
+ linux,pci-domain = <10>;
+
+ clocks = <&bpmp TEGRA234_CLK_PEX2_C10_CORE>;
+ clock-names = "core";
+
+ resets = <&bpmp TEGRA234_RESET_PEX2_CORE_10_APB>,
+ <&bpmp TEGRA234_RESET_PEX2_CORE_10>;
+ reset-names = "apb", "core";
+
+ interrupts = <GIC_SPI 360 IRQ_TYPE_LEVEL_HIGH>, /* controller interrupt */
+ <GIC_SPI 361 IRQ_TYPE_LEVEL_HIGH>; /* MSI interrupt */
+ interrupt-names = "intr", "msi";
+
+ #interrupt-cells = <1>;
+ interrupt-map-mask = <0 0 0 0>;
+ interrupt-map = <0 0 0 0 &gic GIC_SPI 360 IRQ_TYPE_LEVEL_HIGH>;
+
+ nvidia,bpmp = <&bpmp 10>;
+
+ nvidia,aspm-cmrt-us = <60>;
+ nvidia,aspm-pwr-on-t-us = <20>;
+ nvidia,aspm-l0s-entrance-latency-us = <3>;
+
+ bus-range = <0x0 0xff>;
+
+ ranges = <0x43000000 0x38 0x40000000 0x38 0x40000000 0x2 0xe8000000>, /* prefetchable memory (11904 MB) */
+ <0x02000000 0x0 0x40000000 0x3b 0x28000000 0x0 0x08000000>, /* non-prefetchable memory (128 MB) */
+ <0x01000000 0x0 0x2e100000 0x00 0x2e100000 0x0 0x00100000>; /* downstream I/O (1 MB) */
+
+ interconnects = <&mc TEGRA234_MEMORY_CLIENT_PCIE10AR &emc>,
+ <&mc TEGRA234_MEMORY_CLIENT_PCIE10AW &emc>;
+ interconnect-names = "dma-mem", "write";
+ };
+
+ pcie@14100000 {
+ compatible = "nvidia,tegra234-pcie";
+ power-domains = <&bpmp TEGRA234_POWER_DOMAIN_PCIEX1A>;
+ reg = <0x00 0x14100000 0x0 0x00020000>, /* appl registers (128K) */
+ <0x00 0x30000000 0x0 0x00040000>, /* configuration space (256K) */
+ <0x00 0x30040000 0x0 0x00040000>, /* iATU_DMA reg space (256K) */
+ <0x00 0x30080000 0x0 0x00040000>; /* DBI reg space (256K) */
+ reg-names = "appl", "config", "atu_dma", "dbi";
+
+ status = "disabled";
+
+ #address-cells = <3>;
+ #size-cells = <2>;
+ device_type = "pci";
+ num-lanes = <1>;
+ num-viewport = <8>;
+ linux,pci-domain = <1>;
+
+ clocks = <&bpmp TEGRA234_CLK_PEX0_C1_CORE>;
+ clock-names = "core";
+
+ resets = <&bpmp TEGRA234_RESET_PEX0_CORE_1_APB>,
+ <&bpmp TEGRA234_RESET_PEX0_CORE_1>;
+ reset-names = "apb", "core";
+
+ interrupts = <GIC_SPI 45 IRQ_TYPE_LEVEL_HIGH>, /* controller interrupt */
+ <GIC_SPI 46 IRQ_TYPE_LEVEL_HIGH>; /* MSI interrupt */
+ interrupt-names = "intr", "msi";
+
+ #interrupt-cells = <1>;
+ interrupt-map-mask = <0 0 0 0>;
+ interrupt-map = <0 0 0 0 &gic GIC_SPI 45 IRQ_TYPE_LEVEL_HIGH>;
+
+ nvidia,bpmp = <&bpmp 1>;
+
+ nvidia,aspm-cmrt-us = <60>;
+ nvidia,aspm-pwr-on-t-us = <20>;
+ nvidia,aspm-l0s-entrance-latency-us = <3>;
+
+ bus-range = <0x0 0xff>;
+
+ ranges = <0x43000000 0x20 0x80000000 0x20 0x80000000 0x0 0x28000000>, /* prefetchable memory (640 MB) */
+ <0x02000000 0x0 0x40000000 0x20 0xa8000000 0x0 0x08000000>, /* non-prefetchable memory (128 MB) */
+ <0x01000000 0x0 0x30100000 0x00 0x30100000 0x0 0x00100000>; /* downstream I/O (1 MB) */
+
+ interconnects = <&mc TEGRA234_MEMORY_CLIENT_PCIE1R &emc>,
+ <&mc TEGRA234_MEMORY_CLIENT_PCIE1W &emc>;
+ interconnect-names = "dma-mem", "write";
+ };
+
+ pcie@14120000 {
+ compatible = "nvidia,tegra234-pcie";
+ power-domains = <&bpmp TEGRA234_POWER_DOMAIN_PCIEX1A>;
+ reg = <0x00 0x14120000 0x0 0x00020000>, /* appl registers (128K) */
+ <0x00 0x32000000 0x0 0x00040000>, /* configuration space (256K) */
+ <0x00 0x32040000 0x0 0x00040000>, /* iATU_DMA reg space (256K) */
+ <0x00 0x32080000 0x0 0x00040000>; /* DBI reg space (256K) */
+ reg-names = "appl", "config", "atu_dma", "dbi";
+
+ status = "disabled";
+
+ #address-cells = <3>;
+ #size-cells = <2>;
+ device_type = "pci";
+ num-lanes = <1>;
+ num-viewport = <8>;
+ linux,pci-domain = <2>;
+
+ clocks = <&bpmp TEGRA234_CLK_PEX0_C2_CORE>;
+ clock-names = "core";
+
+ resets = <&bpmp TEGRA234_RESET_PEX0_CORE_2_APB>,
+ <&bpmp TEGRA234_RESET_PEX0_CORE_2>;
+ reset-names = "apb", "core";
+
+ interrupts = <GIC_SPI 47 IRQ_TYPE_LEVEL_HIGH>, /* controller interrupt */
+ <GIC_SPI 48 IRQ_TYPE_LEVEL_HIGH>; /* MSI interrupt */
+ interrupt-names = "intr", "msi";
+
+ #interrupt-cells = <1>;
+ interrupt-map-mask = <0 0 0 0>;
+ interrupt-map = <0 0 0 0 &gic GIC_SPI 47 IRQ_TYPE_LEVEL_HIGH>;
+
+ nvidia,bpmp = <&bpmp 2>;
+
+ nvidia,aspm-cmrt-us = <60>;
+ nvidia,aspm-pwr-on-t-us = <20>;
+ nvidia,aspm-l0s-entrance-latency-us = <3>;
+
+ bus-range = <0x0 0xff>;
+
+ ranges = <0x43000000 0x20 0xc0000000 0x20 0xc0000000 0x0 0x28000000>, /* prefetchable memory (640 MB) */
+ <0x02000000 0x0 0x40000000 0x20 0xe8000000 0x0 0x08000000>, /* non-prefetchable memory (128 MB) */
+ <0x01000000 0x0 0x32100000 0x00 0x32100000 0x0 0x00100000>; /* downstream I/O (1 MB) */
+
+ interconnects = <&mc TEGRA234_MEMORY_CLIENT_PCIE2AR &emc>,
+ <&mc TEGRA234_MEMORY_CLIENT_PCIE2AW &emc>;
+ interconnect-names = "dma-mem", "write";
+ };
+
+ pcie@14140000 {
+ compatible = "nvidia,tegra234-pcie";
+ power-domains = <&bpmp TEGRA234_POWER_DOMAIN_PCIEX1A>;
+ reg = <0x00 0x14140000 0x0 0x00020000>, /* appl registers (128K) */
+ <0x00 0x34000000 0x0 0x00040000>, /* configuration space (256K) */
+ <0x00 0x34040000 0x0 0x00040000>, /* iATU_DMA reg space (256K) */
+ <0x00 0x34080000 0x0 0x00040000>; /* DBI reg space (256K) */
+ reg-names = "appl", "config", "atu_dma", "dbi";
+
+ status = "disabled";
+
+ #address-cells = <3>;
+ #size-cells = <2>;
+ device_type = "pci";
+ num-lanes = <1>;
+ num-viewport = <8>;
+ linux,pci-domain = <3>;
+
+ clocks = <&bpmp TEGRA234_CLK_PEX0_C3_CORE>;
+ clock-names = "core";
+
+ resets = <&bpmp TEGRA234_RESET_PEX0_CORE_3_APB>,
+ <&bpmp TEGRA234_RESET_PEX0_CORE_3>;
+ reset-names = "apb", "core";
+
+ interrupts = <GIC_SPI 49 IRQ_TYPE_LEVEL_HIGH>, /* controller interrupt */
+ <GIC_SPI 50 IRQ_TYPE_LEVEL_HIGH>; /* MSI interrupt */
+ interrupt-names = "intr", "msi";
+
+ #interrupt-cells = <1>;
+ interrupt-map-mask = <0 0 0 0>;
+ interrupt-map = <0 0 0 0 &gic GIC_SPI 49 IRQ_TYPE_LEVEL_HIGH>;
+
+ nvidia,bpmp = <&bpmp 3>;
+
+ nvidia,aspm-cmrt-us = <60>;
+ nvidia,aspm-pwr-on-t-us = <20>;
+ nvidia,aspm-l0s-entrance-latency-us = <3>;
+
+ bus-range = <0x0 0xff>;
+
+ ranges = <0x43000000 0x21 0x00000000 0x21 0x00000000 0x0 0x28000000>, /* prefetchable memory (640 MB) */
+ <0x02000000 0x0 0x40000000 0x21 0xe8000000 0x0 0x08000000>, /* non-prefetchable memory (128 MB) */
+ <0x01000000 0x0 0x34100000 0x00 0x34100000 0x0 0x00100000>; /* downstream I/O (1 MB) */
+
+ interconnects = <&mc TEGRA234_MEMORY_CLIENT_PCIE3R &emc>,
+ <&mc TEGRA234_MEMORY_CLIENT_PCIE3W &emc>;
+ interconnect-names = "dma-mem", "write";
+ };
+
+ pcie@14160000 {
+ compatible = "nvidia,tegra234-pcie";
+ power-domains = <&bpmp TEGRA234_POWER_DOMAIN_PCIEX4BB>;
+ reg = <0x00 0x14160000 0x0 0x00020000>, /* appl registers (128K) */
+ <0x00 0x36000000 0x0 0x00040000>, /* configuration space (256K) */
+ <0x00 0x36040000 0x0 0x00040000>, /* iATU_DMA reg space (256K) */
+ <0x00 0x36080000 0x0 0x00040000>; /* DBI reg space (256K) */
+ reg-names = "appl", "config", "atu_dma", "dbi";
+
+ status = "disabled";
+
+ #address-cells = <3>;
+ #size-cells = <2>;
+ device_type = "pci";
+ num-lanes = <4>;
+ num-viewport = <8>;
+ linux,pci-domain = <4>;
+
+ clocks = <&bpmp TEGRA234_CLK_PEX0_C4_CORE>;
+ clock-names = "core";
+
+ resets = <&bpmp TEGRA234_RESET_PEX0_CORE_4_APB>,
+ <&bpmp TEGRA234_RESET_PEX0_CORE_4>;
+ reset-names = "apb", "core";
+
+ interrupts = <GIC_SPI 51 IRQ_TYPE_LEVEL_HIGH>, /* controller interrupt */
+ <GIC_SPI 52 IRQ_TYPE_LEVEL_HIGH>; /* MSI interrupt */
+ interrupt-names = "intr", "msi";
+
+ #interrupt-cells = <1>;
+ interrupt-map-mask = <0 0 0 0>;
+ interrupt-map = <0 0 0 0 &gic GIC_SPI 51 IRQ_TYPE_LEVEL_HIGH>;
+
+ nvidia,bpmp = <&bpmp 4>;
+
+ nvidia,aspm-cmrt-us = <60>;
+ nvidia,aspm-pwr-on-t-us = <20>;
+ nvidia,aspm-l0s-entrance-latency-us = <3>;
+
+ bus-range = <0x0 0xff>;
+
+ ranges = <0x43000000 0x21 0x40000000 0x21 0x40000000 0x2 0xe8000000>, /* prefetchable memory (11904 MB) */
+ <0x02000000 0x0 0x40000000 0x24 0x28000000 0x0 0x08000000>, /* non-prefetchable memory (128 MB) */
+ <0x01000000 0x0 0x36100000 0x00 0x36100000 0x0 0x00100000>; /* downstream I/O (1 MB) */
+
+ interconnects = <&mc TEGRA234_MEMORY_CLIENT_PCIE4R &emc>,
+ <&mc TEGRA234_MEMORY_CLIENT_PCIE4W &emc>;
+ interconnect-names = "dma-mem", "write";
+ };
+
+ pcie@14180000 {
+ compatible = "nvidia,tegra234-pcie";
+ power-domains = <&bpmp TEGRA234_POWER_DOMAIN_PCIEX4BA>;
+ reg = <0x00 0x14180000 0x0 0x00020000>, /* appl registers (128K) */
+ <0x00 0x38000000 0x0 0x00040000>, /* configuration space (256K) */
+ <0x00 0x38040000 0x0 0x00040000>, /* iATU_DMA reg space (256K) */
+ <0x00 0x38080000 0x0 0x00040000>; /* DBI reg space (256K) */
+ reg-names = "appl", "config", "atu_dma", "dbi";
+
+ status = "disabled";
+
+ #address-cells = <3>;
+ #size-cells = <2>;
+ device_type = "pci";
+ num-lanes = <4>;
+ num-viewport = <8>;
+ linux,pci-domain = <0>;
+
+ clocks = <&bpmp TEGRA234_CLK_PEX0_C0_CORE>;
+ clock-names = "core";
+
+ resets = <&bpmp TEGRA234_RESET_PEX0_CORE_0_APB>,
+ <&bpmp TEGRA234_RESET_PEX0_CORE_0>;
+ reset-names = "apb", "core";
+
+ interrupts = <GIC_SPI 72 IRQ_TYPE_LEVEL_HIGH>, /* controller interrupt */
+ <GIC_SPI 73 IRQ_TYPE_LEVEL_HIGH>; /* MSI interrupt */
+ interrupt-names = "intr", "msi";
+
+ #interrupt-cells = <1>;
+ interrupt-map-mask = <0 0 0 0>;
+ interrupt-map = <0 0 0 0 &gic GIC_SPI 72 IRQ_TYPE_LEVEL_HIGH>;
+
+ nvidia,bpmp = <&bpmp 0>;
+
+ nvidia,aspm-cmrt-us = <60>;
+ nvidia,aspm-pwr-on-t-us = <20>;
+ nvidia,aspm-l0s-entrance-latency-us = <3>;
+
+ bus-range = <0x0 0xff>;
+
+ ranges = <0x43000000 0x24 0x40000000 0x24 0x40000000 0x2 0xe8000000>, /* prefetchable memory (11904 MB) */
+ <0x02000000 0x0 0x40000000 0x27 0x28000000 0x0 0x08000000>, /* non-prefetchable memory (128 MB) */
+ <0x01000000 0x0 0x38100000 0x00 0x38100000 0x0 0x00100000>; /* downstream I/O (1 MB) */
+
+ interconnects = <&mc TEGRA234_MEMORY_CLIENT_PCIE0R &emc>,
+ <&mc TEGRA234_MEMORY_CLIENT_PCIE0W &emc>;
+ interconnect-names = "dma-mem", "write";
+ };
+
+ pcie@141a0000 {
+ compatible = "nvidia,tegra234-pcie";
+ power-domains = <&bpmp TEGRA234_POWER_DOMAIN_PCIEX8A>;
+ reg = <0x00 0x141a0000 0x0 0x00020000>, /* appl registers (128K) */
+ <0x00 0x3a000000 0x0 0x00040000>, /* configuration space (256K) */
+ <0x00 0x3a040000 0x0 0x00040000>, /* iATU_DMA reg space (256K) */
+ <0x00 0x3a080000 0x0 0x00040000>; /* DBI reg space (256K) */
+ reg-names = "appl", "config", "atu_dma", "dbi";
+
+ status = "disabled";
+
+ #address-cells = <3>;
+ #size-cells = <2>;
+ device_type = "pci";
+ num-lanes = <8>;
+ num-viewport = <8>;
+ linux,pci-domain = <5>;
+
+ clocks = <&bpmp TEGRA234_CLK_PEX1_C5_CORE>;
+ clock-names = "core";
+
+ resets = <&bpmp TEGRA234_RESET_PEX1_CORE_5_APB>,
+ <&bpmp TEGRA234_RESET_PEX1_CORE_5>;
+ reset-names = "apb", "core";
+
+ interrupts = <GIC_SPI 53 IRQ_TYPE_LEVEL_HIGH>, /* controller interrupt */
+ <GIC_SPI 54 IRQ_TYPE_LEVEL_HIGH>; /* MSI interrupt */
+ interrupt-names = "intr", "msi";
+
+ #interrupt-cells = <1>;
+ interrupt-map-mask = <0 0 0 0>;
+ interrupt-map = <0 0 0 0 &gic GIC_SPI 53 IRQ_TYPE_LEVEL_HIGH>;
+
+ nvidia,bpmp = <&bpmp 5>;
+
+ nvidia,aspm-cmrt-us = <60>;
+ nvidia,aspm-pwr-on-t-us = <20>;
+ nvidia,aspm-l0s-entrance-latency-us = <3>;
+
+ bus-range = <0x0 0xff>;
+
+ ranges = <0x43000000 0x27 0x40000000 0x27 0x40000000 0x3 0xe8000000>, /* prefetchable memory (16000 MB) */
+ <0x02000000 0x0 0x40000000 0x2b 0x28000000 0x0 0x08000000>, /* non-prefetchable memory (128 MB) */
+ <0x01000000 0x0 0x3a100000 0x00 0x3a100000 0x0 0x00100000>; /* downstream I/O (1 MB) */
+
+ interconnects = <&mc TEGRA234_MEMORY_CLIENT_PCIE5R &emc>,
+ <&mc TEGRA234_MEMORY_CLIENT_PCIE5W &emc>;
+ interconnect-names = "dma-mem", "write";
+ };
+
+ pcie@141c0000 {
+ compatible = "nvidia,tegra234-pcie";
+ power-domains = <&bpmp TEGRA234_POWER_DOMAIN_PCIEX4A>;
+ reg = <0x00 0x141c0000 0x0 0x00020000>, /* appl registers (128K) */
+ <0x00 0x3c000000 0x0 0x00040000>, /* configuration space (256K) */
+ <0x00 0x3c040000 0x0 0x00040000>, /* iATU_DMA reg space (256K) */
+ <0x00 0x3c080000 0x0 0x00040000>; /* DBI reg space (256K) */
+ reg-names = "appl", "config", "atu_dma", "dbi";
+
+ status = "disabled";
+
+ #address-cells = <3>;
+ #size-cells = <2>;
+ device_type = "pci";
+ num-lanes = <4>;
+ num-viewport = <8>;
+ linux,pci-domain = <6>;
+
+ clocks = <&bpmp TEGRA234_CLK_PEX1_C6_CORE>;
+ clock-names = "core";
+
+ resets = <&bpmp TEGRA234_RESET_PEX1_CORE_6_APB>,
+ <&bpmp TEGRA234_RESET_PEX1_CORE_6>;
+ reset-names = "apb", "core";
+
+ interrupts = <GIC_SPI 352 IRQ_TYPE_LEVEL_HIGH>, /* controller interrupt */
+ <GIC_SPI 353 IRQ_TYPE_LEVEL_HIGH>; /* MSI interrupt */
+ interrupt-names = "intr", "msi";
+
+ #interrupt-cells = <1>;
+ interrupt-map-mask = <0 0 0 0>;
+ interrupt-map = <0 0 0 0 &gic GIC_SPI 352 IRQ_TYPE_LEVEL_HIGH>;
+
+ nvidia,bpmp = <&bpmp 6>;
+
+ nvidia,aspm-cmrt-us = <60>;
+ nvidia,aspm-pwr-on-t-us = <20>;
+ nvidia,aspm-l0s-entrance-latency-us = <3>;
+
+ bus-range = <0x0 0xff>;
+
+ ranges = <0x43000000 0x2b 0x40000000 0x2b 0x40000000 0x2 0xe8000000>, /* prefetchable memory (11904 MB) */
+ <0x02000000 0x0 0x40000000 0x2e 0x28000000 0x0 0x08000000>, /* non-prefetchable memory (128 MB) */
+ <0x01000000 0x0 0x3c100000 0x00 0x3c100000 0x0 0x00100000>; /* downstream I/O (1 MB) */
+
+ interconnects = <&mc TEGRA234_MEMORY_CLIENT_PCIE6AR &emc>,
+ <&mc TEGRA234_MEMORY_CLIENT_PCIE6AW &emc>;
+ interconnect-names = "dma-mem", "write";
+ };
+
+ pcie@141e0000 {
+ compatible = "nvidia,tegra234-pcie";
+ power-domains = <&bpmp TEGRA234_POWER_DOMAIN_PCIEX8B>;
+ reg = <0x00 0x141e0000 0x0 0x00020000>, /* appl registers (128K) */
+ <0x00 0x3e000000 0x0 0x00040000>, /* configuration space (256K) */
+ <0x00 0x3e040000 0x0 0x00040000>, /* iATU_DMA reg space (256K) */
+ <0x00 0x3e080000 0x0 0x00040000>; /* DBI reg space (256K) */
+ reg-names = "appl", "config", "atu_dma", "dbi";
+
+ status = "disabled";
+
+ #address-cells = <3>;
+ #size-cells = <2>;
+ device_type = "pci";
+ num-lanes = <8>;
+ num-viewport = <8>;
+ linux,pci-domain = <7>;
+
+ clocks = <&bpmp TEGRA234_CLK_PEX2_C7_CORE>;
+ clock-names = "core";
+
+ resets = <&bpmp TEGRA234_RESET_PEX2_CORE_7_APB>,
+ <&bpmp TEGRA234_RESET_PEX2_CORE_7>;
+ reset-names = "apb", "core";
+
+ interrupts = <GIC_SPI 354 IRQ_TYPE_LEVEL_HIGH>, /* controller interrupt */
+ <GIC_SPI 355 IRQ_TYPE_LEVEL_HIGH>; /* MSI interrupt */
+ interrupt-names = "intr", "msi";
+
+ #interrupt-cells = <1>;
+ interrupt-map-mask = <0 0 0 0>;
+ interrupt-map = <0 0 0 0 &gic GIC_SPI 354 IRQ_TYPE_LEVEL_HIGH>;
+
+ nvidia,bpmp = <&bpmp 7>;
+
+ nvidia,aspm-cmrt-us = <60>;
+ nvidia,aspm-pwr-on-t-us = <20>;
+ nvidia,aspm-l0s-entrance-latency-us = <3>;
+
+ bus-range = <0x0 0xff>;
+
+ ranges = <0x43000000 0x2e 0x40000000 0x2e 0x40000000 0x3 0xe8000000>, /* prefetchable memory (16000 MB) */
+ <0x02000000 0x0 0x40000000 0x32 0x28000000 0x0 0x08000000>, /* non-prefetchable memory (128 MB) */
+ <0x01000000 0x0 0x3e100000 0x00 0x3e100000 0x0 0x00100000>; /* downstream I/O (1 MB) */
+
+ interconnects = <&mc TEGRA234_MEMORY_CLIENT_PCIE7AR &emc>,
+ <&mc TEGRA234_MEMORY_CLIENT_PCIE7AW &emc>;
+ interconnect-names = "dma-mem", "write";
+ };
+
sram@40000000 {
compatible = "nvidia,tegra234-sysram", "mmio-sram";
reg = <0x0 0x40000000 0x0 0x80000>;
--
2.17.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [PATCH V1 07/10] arm64: tegra: Enable PCIe slots in P3737-0000 board
2022-02-05 16:21 ` Vidya Sagar
@ 2022-02-05 16:21 ` Vidya Sagar
-1 siblings, 0 replies; 46+ messages in thread
From: Vidya Sagar @ 2022-02-05 16:21 UTC (permalink / raw)
To: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh
Cc: kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, vidyas, sagar.tv
Enable PCIe controller nodes to enable respective PCIe slots on
P3737-0000 board. Following is the ownership of slots by different
PCIe controllers.
Controller-1 : On-board Broadcom WiFi controller
Controller-4 : M.2 Key-M slot
Controller-5 : CEM form-factor x8 slot
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
---
.../nvidia/tegra234-p3737-0000+p3701-0000.dts | 26 +++++++++++++++++++
1 file changed, 26 insertions(+)
diff --git a/arch/arm64/boot/dts/nvidia/tegra234-p3737-0000+p3701-0000.dts b/arch/arm64/boot/dts/nvidia/tegra234-p3737-0000+p3701-0000.dts
index efbbb878ba5a..b819e1133bc4 100644
--- a/arch/arm64/boot/dts/nvidia/tegra234-p3737-0000+p3701-0000.dts
+++ b/arch/arm64/boot/dts/nvidia/tegra234-p3737-0000+p3701-0000.dts
@@ -21,4 +21,30 @@
serial {
status = "okay";
};
+
+ pcie@14100000 {
+ status = "okay";
+
+ phys = <&p2u_hsio_3>;
+ phy-names = "p2u-0";
+ };
+
+ pcie@14160000 {
+ status = "okay";
+
+ phys = <&p2u_hsio_4>, <&p2u_hsio_5>, <&p2u_hsio_6>,
+ <&p2u_hsio_7>;
+ phy-names = "p2u-0", "p2u-1", "p2u-2", "p2u-3";
+ };
+
+ pcie@141a0000 {
+ status = "okay";
+
+ phys = <&p2u_nvhs_0>, <&p2u_nvhs_1>, <&p2u_nvhs_2>,
+ <&p2u_nvhs_3>, <&p2u_nvhs_4>, <&p2u_nvhs_5>,
+ <&p2u_nvhs_6>, <&p2u_nvhs_7>;
+ phy-names = "p2u-0", "p2u-1", "p2u-2", "p2u-3", "p2u-4",
+ "p2u-5", "p2u-6", "p2u-7";
+ };
+
};
--
2.17.1
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [PATCH V1 07/10] arm64: tegra: Enable PCIe slots in P3737-0000 board
@ 2022-02-05 16:21 ` Vidya Sagar
0 siblings, 0 replies; 46+ messages in thread
From: Vidya Sagar @ 2022-02-05 16:21 UTC (permalink / raw)
To: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh
Cc: kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, vidyas, sagar.tv
Enable PCIe controller nodes to enable respective PCIe slots on
P3737-0000 board. Following is the ownership of slots by different
PCIe controllers.
Controller-1 : On-board Broadcom WiFi controller
Controller-4 : M.2 Key-M slot
Controller-5 : CEM form-factor x8 slot
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
---
.../nvidia/tegra234-p3737-0000+p3701-0000.dts | 26 +++++++++++++++++++
1 file changed, 26 insertions(+)
diff --git a/arch/arm64/boot/dts/nvidia/tegra234-p3737-0000+p3701-0000.dts b/arch/arm64/boot/dts/nvidia/tegra234-p3737-0000+p3701-0000.dts
index efbbb878ba5a..b819e1133bc4 100644
--- a/arch/arm64/boot/dts/nvidia/tegra234-p3737-0000+p3701-0000.dts
+++ b/arch/arm64/boot/dts/nvidia/tegra234-p3737-0000+p3701-0000.dts
@@ -21,4 +21,30 @@
serial {
status = "okay";
};
+
+ pcie@14100000 {
+ status = "okay";
+
+ phys = <&p2u_hsio_3>;
+ phy-names = "p2u-0";
+ };
+
+ pcie@14160000 {
+ status = "okay";
+
+ phys = <&p2u_hsio_4>, <&p2u_hsio_5>, <&p2u_hsio_6>,
+ <&p2u_hsio_7>;
+ phy-names = "p2u-0", "p2u-1", "p2u-2", "p2u-3";
+ };
+
+ pcie@141a0000 {
+ status = "okay";
+
+ phys = <&p2u_nvhs_0>, <&p2u_nvhs_1>, <&p2u_nvhs_2>,
+ <&p2u_nvhs_3>, <&p2u_nvhs_4>, <&p2u_nvhs_5>,
+ <&p2u_nvhs_6>, <&p2u_nvhs_7>;
+ phy-names = "p2u-0", "p2u-1", "p2u-2", "p2u-3", "p2u-4",
+ "p2u-5", "p2u-6", "p2u-7";
+ };
+
};
--
2.17.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [PATCH V1 08/10] phy: tegra: Add PCIe PIPE2UPHY support for Tegra234
2022-02-05 16:21 ` Vidya Sagar
@ 2022-02-05 16:21 ` Vidya Sagar
-1 siblings, 0 replies; 46+ messages in thread
From: Vidya Sagar @ 2022-02-05 16:21 UTC (permalink / raw)
To: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh
Cc: kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, vidyas, sagar.tv
Synopsys DesignWare core based PCIe controllers in Tegra234 SoC
interface with Universal PHY (UPHY) module through a PIPE2UPHY (P2U)
module. For each PCIe lane of a controller, there is a P2U unit
instantiated at hardware level. This driver provides support for the
programming required for each P2U that is going to be used for a PCIe
controller.
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
---
drivers/phy/tegra/phy-tegra194-p2u.c | 48 +++++++++++++++++++++++++++-
1 file changed, 47 insertions(+), 1 deletion(-)
diff --git a/drivers/phy/tegra/phy-tegra194-p2u.c b/drivers/phy/tegra/phy-tegra194-p2u.c
index 3ee02b9eb04f..1415ca71de38 100644
--- a/drivers/phy/tegra/phy-tegra194-p2u.c
+++ b/drivers/phy/tegra/phy-tegra194-p2u.c
@@ -2,7 +2,7 @@
/*
* P2U (PIPE to UPHY) driver for Tegra T194 SoC
*
- * Copyright (C) 2019 NVIDIA Corporation.
+ * Copyright (C) 2019-2022 NVIDIA Corporation.
*
* Author: Vidya Sagar <vidyas@nvidia.com>
*/
@@ -14,6 +14,9 @@
#include <linux/of_platform.h>
#include <linux/phy/phy.h>
+#define P2U_CONTROL_CMN 0x74
+#define P2U_CONTROL_CMN_SKP_SIZE_PROTECTION_EN BIT(20)
+
#define P2U_PERIODIC_EQ_CTRL_GEN3 0xc0
#define P2U_PERIODIC_EQ_CTRL_GEN3_PERIODIC_EQ_EN BIT(0)
#define P2U_PERIODIC_EQ_CTRL_GEN3_INIT_PRESET_EQ_TRAIN_EN BIT(1)
@@ -24,8 +27,17 @@
#define P2U_RX_DEBOUNCE_TIME_DEBOUNCE_TIMER_MASK 0xffff
#define P2U_RX_DEBOUNCE_TIME_DEBOUNCE_TIMER_VAL 160
+#define P2U_DIR_SEARCH_CTRL 0xd4
+#define P2U_DIR_SEARCH_CTRL_GEN4_FINE_GRAIN_SEARCH_TWICE BIT(18)
+
+struct tegra_p2u_of_data {
+ bool one_dir_search;
+};
+
struct tegra_p2u {
void __iomem *base;
+ bool skip_sz_protection_en; /* Needed to support two retimers */
+ struct tegra_p2u_of_data *of_data;
};
static inline void p2u_writel(struct tegra_p2u *phy, const u32 value,
@@ -44,6 +56,12 @@ static int tegra_p2u_power_on(struct phy *x)
struct tegra_p2u *phy = phy_get_drvdata(x);
u32 val;
+ if (phy->skip_sz_protection_en) {
+ val = p2u_readl(phy, P2U_CONTROL_CMN);
+ val |= P2U_CONTROL_CMN_SKP_SIZE_PROTECTION_EN;
+ p2u_writel(phy, val, P2U_CONTROL_CMN);
+ }
+
val = p2u_readl(phy, P2U_PERIODIC_EQ_CTRL_GEN3);
val &= ~P2U_PERIODIC_EQ_CTRL_GEN3_PERIODIC_EQ_EN;
val |= P2U_PERIODIC_EQ_CTRL_GEN3_INIT_PRESET_EQ_TRAIN_EN;
@@ -58,6 +76,12 @@ static int tegra_p2u_power_on(struct phy *x)
val |= P2U_RX_DEBOUNCE_TIME_DEBOUNCE_TIMER_VAL;
p2u_writel(phy, val, P2U_RX_DEBOUNCE_TIME);
+ if (phy->of_data->one_dir_search) {
+ val = p2u_readl(phy, P2U_DIR_SEARCH_CTRL);
+ val &= ~P2U_DIR_SEARCH_CTRL_GEN4_FINE_GRAIN_SEARCH_TWICE;
+ p2u_writel(phy, val, P2U_DIR_SEARCH_CTRL);
+ }
+
return 0;
}
@@ -77,10 +101,19 @@ static int tegra_p2u_probe(struct platform_device *pdev)
if (!phy)
return -ENOMEM;
+ phy->of_data =
+ (struct tegra_p2u_of_data *)of_device_get_match_data(dev);
+ if (!phy->of_data)
+ return -EINVAL;
+
phy->base = devm_platform_ioremap_resource_byname(pdev, "ctl");
if (IS_ERR(phy->base))
return PTR_ERR(phy->base);
+ phy->skip_sz_protection_en =
+ of_property_read_bool(dev->of_node,
+ "nvidia,skip-sz-protect-en");
+
platform_set_drvdata(pdev, phy);
generic_phy = devm_phy_create(dev, NULL, &ops);
@@ -96,9 +129,22 @@ static int tegra_p2u_probe(struct platform_device *pdev)
return 0;
}
+static const struct tegra_p2u_of_data tegra194_p2u_of_data = {
+ .one_dir_search = false,
+};
+
+static const struct tegra_p2u_of_data tegra234_p2u_of_data = {
+ .one_dir_search = true,
+};
+
static const struct of_device_id tegra_p2u_id_table[] = {
{
.compatible = "nvidia,tegra194-p2u",
+ .data = &tegra194_p2u_of_data,
+ },
+ {
+ .compatible = "nvidia,tegra234-p2u",
+ .data = &tegra234_p2u_of_data,
},
{}
};
--
2.17.1
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [PATCH V1 08/10] phy: tegra: Add PCIe PIPE2UPHY support for Tegra234
@ 2022-02-05 16:21 ` Vidya Sagar
0 siblings, 0 replies; 46+ messages in thread
From: Vidya Sagar @ 2022-02-05 16:21 UTC (permalink / raw)
To: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh
Cc: kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, vidyas, sagar.tv
Synopsys DesignWare core based PCIe controllers in Tegra234 SoC
interface with Universal PHY (UPHY) module through a PIPE2UPHY (P2U)
module. For each PCIe lane of a controller, there is a P2U unit
instantiated at hardware level. This driver provides support for the
programming required for each P2U that is going to be used for a PCIe
controller.
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
---
drivers/phy/tegra/phy-tegra194-p2u.c | 48 +++++++++++++++++++++++++++-
1 file changed, 47 insertions(+), 1 deletion(-)
diff --git a/drivers/phy/tegra/phy-tegra194-p2u.c b/drivers/phy/tegra/phy-tegra194-p2u.c
index 3ee02b9eb04f..1415ca71de38 100644
--- a/drivers/phy/tegra/phy-tegra194-p2u.c
+++ b/drivers/phy/tegra/phy-tegra194-p2u.c
@@ -2,7 +2,7 @@
/*
* P2U (PIPE to UPHY) driver for Tegra T194 SoC
*
- * Copyright (C) 2019 NVIDIA Corporation.
+ * Copyright (C) 2019-2022 NVIDIA Corporation.
*
* Author: Vidya Sagar <vidyas@nvidia.com>
*/
@@ -14,6 +14,9 @@
#include <linux/of_platform.h>
#include <linux/phy/phy.h>
+#define P2U_CONTROL_CMN 0x74
+#define P2U_CONTROL_CMN_SKP_SIZE_PROTECTION_EN BIT(20)
+
#define P2U_PERIODIC_EQ_CTRL_GEN3 0xc0
#define P2U_PERIODIC_EQ_CTRL_GEN3_PERIODIC_EQ_EN BIT(0)
#define P2U_PERIODIC_EQ_CTRL_GEN3_INIT_PRESET_EQ_TRAIN_EN BIT(1)
@@ -24,8 +27,17 @@
#define P2U_RX_DEBOUNCE_TIME_DEBOUNCE_TIMER_MASK 0xffff
#define P2U_RX_DEBOUNCE_TIME_DEBOUNCE_TIMER_VAL 160
+#define P2U_DIR_SEARCH_CTRL 0xd4
+#define P2U_DIR_SEARCH_CTRL_GEN4_FINE_GRAIN_SEARCH_TWICE BIT(18)
+
+struct tegra_p2u_of_data {
+ bool one_dir_search;
+};
+
struct tegra_p2u {
void __iomem *base;
+ bool skip_sz_protection_en; /* Needed to support two retimers */
+ struct tegra_p2u_of_data *of_data;
};
static inline void p2u_writel(struct tegra_p2u *phy, const u32 value,
@@ -44,6 +56,12 @@ static int tegra_p2u_power_on(struct phy *x)
struct tegra_p2u *phy = phy_get_drvdata(x);
u32 val;
+ if (phy->skip_sz_protection_en) {
+ val = p2u_readl(phy, P2U_CONTROL_CMN);
+ val |= P2U_CONTROL_CMN_SKP_SIZE_PROTECTION_EN;
+ p2u_writel(phy, val, P2U_CONTROL_CMN);
+ }
+
val = p2u_readl(phy, P2U_PERIODIC_EQ_CTRL_GEN3);
val &= ~P2U_PERIODIC_EQ_CTRL_GEN3_PERIODIC_EQ_EN;
val |= P2U_PERIODIC_EQ_CTRL_GEN3_INIT_PRESET_EQ_TRAIN_EN;
@@ -58,6 +76,12 @@ static int tegra_p2u_power_on(struct phy *x)
val |= P2U_RX_DEBOUNCE_TIME_DEBOUNCE_TIMER_VAL;
p2u_writel(phy, val, P2U_RX_DEBOUNCE_TIME);
+ if (phy->of_data->one_dir_search) {
+ val = p2u_readl(phy, P2U_DIR_SEARCH_CTRL);
+ val &= ~P2U_DIR_SEARCH_CTRL_GEN4_FINE_GRAIN_SEARCH_TWICE;
+ p2u_writel(phy, val, P2U_DIR_SEARCH_CTRL);
+ }
+
return 0;
}
@@ -77,10 +101,19 @@ static int tegra_p2u_probe(struct platform_device *pdev)
if (!phy)
return -ENOMEM;
+ phy->of_data =
+ (struct tegra_p2u_of_data *)of_device_get_match_data(dev);
+ if (!phy->of_data)
+ return -EINVAL;
+
phy->base = devm_platform_ioremap_resource_byname(pdev, "ctl");
if (IS_ERR(phy->base))
return PTR_ERR(phy->base);
+ phy->skip_sz_protection_en =
+ of_property_read_bool(dev->of_node,
+ "nvidia,skip-sz-protect-en");
+
platform_set_drvdata(pdev, phy);
generic_phy = devm_phy_create(dev, NULL, &ops);
@@ -96,9 +129,22 @@ static int tegra_p2u_probe(struct platform_device *pdev)
return 0;
}
+static const struct tegra_p2u_of_data tegra194_p2u_of_data = {
+ .one_dir_search = false,
+};
+
+static const struct tegra_p2u_of_data tegra234_p2u_of_data = {
+ .one_dir_search = true,
+};
+
static const struct of_device_id tegra_p2u_id_table[] = {
{
.compatible = "nvidia,tegra194-p2u",
+ .data = &tegra194_p2u_of_data,
+ },
+ {
+ .compatible = "nvidia,tegra234-p2u",
+ .data = &tegra234_p2u_of_data,
},
{}
};
--
2.17.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [PATCH V1 09/10] PCI: Disable MSI for Tegra234 root ports
2022-02-05 16:21 ` Vidya Sagar
@ 2022-02-05 16:21 ` Vidya Sagar
-1 siblings, 0 replies; 46+ messages in thread
From: Vidya Sagar @ 2022-02-05 16:21 UTC (permalink / raw)
To: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh
Cc: kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, vidyas, sagar.tv
Tegra234 PCIe rootports don't generate MSI interrupts for PME and AER
events. Since PCIe spec (Ref: r4.0 sec 7.7.1.2 and 7.7.2.2) doesn't support
using a mix of INTx and MSI/MSI-X, MSI needs to be disabled to avoid root
ports service drivers registering their respective ISRs with MSI interrupt
and to let only INTx be used for all events.
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
---
drivers/pci/quirks.c | 9 +++++++++
1 file changed, 9 insertions(+)
diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
index d2dd6a6cda60..3ac5c45e61a1 100644
--- a/drivers/pci/quirks.c
+++ b/drivers/pci/quirks.c
@@ -2747,6 +2747,15 @@ DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x10e5,
DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x10e6,
PCI_CLASS_BRIDGE_PCI, 8,
pci_quirk_nvidia_tegra_disable_rp_msi);
+DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x229a,
+ PCI_CLASS_BRIDGE_PCI, 8,
+ pci_quirk_nvidia_tegra_disable_rp_msi);
+DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x229c,
+ PCI_CLASS_BRIDGE_PCI, 8,
+ pci_quirk_nvidia_tegra_disable_rp_msi);
+DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x229e,
+ PCI_CLASS_BRIDGE_PCI, 8,
+ pci_quirk_nvidia_tegra_disable_rp_msi);
/*
* Some versions of the MCP55 bridge from Nvidia have a legacy IRQ routing
--
2.17.1
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [PATCH V1 09/10] PCI: Disable MSI for Tegra234 root ports
@ 2022-02-05 16:21 ` Vidya Sagar
0 siblings, 0 replies; 46+ messages in thread
From: Vidya Sagar @ 2022-02-05 16:21 UTC (permalink / raw)
To: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh
Cc: kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, vidyas, sagar.tv
Tegra234 PCIe rootports don't generate MSI interrupts for PME and AER
events. Since PCIe spec (Ref: r4.0 sec 7.7.1.2 and 7.7.2.2) doesn't support
using a mix of INTx and MSI/MSI-X, MSI needs to be disabled to avoid root
ports service drivers registering their respective ISRs with MSI interrupt
and to let only INTx be used for all events.
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
---
drivers/pci/quirks.c | 9 +++++++++
1 file changed, 9 insertions(+)
diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
index d2dd6a6cda60..3ac5c45e61a1 100644
--- a/drivers/pci/quirks.c
+++ b/drivers/pci/quirks.c
@@ -2747,6 +2747,15 @@ DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x10e5,
DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x10e6,
PCI_CLASS_BRIDGE_PCI, 8,
pci_quirk_nvidia_tegra_disable_rp_msi);
+DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x229a,
+ PCI_CLASS_BRIDGE_PCI, 8,
+ pci_quirk_nvidia_tegra_disable_rp_msi);
+DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x229c,
+ PCI_CLASS_BRIDGE_PCI, 8,
+ pci_quirk_nvidia_tegra_disable_rp_msi);
+DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x229e,
+ PCI_CLASS_BRIDGE_PCI, 8,
+ pci_quirk_nvidia_tegra_disable_rp_msi);
/*
* Some versions of the MCP55 bridge from Nvidia have a legacy IRQ routing
--
2.17.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [PATCH V1 10/10] PCI: tegra: Add Tegra234 PCIe support
2022-02-05 16:21 ` Vidya Sagar
@ 2022-02-05 16:21 ` Vidya Sagar
-1 siblings, 0 replies; 46+ messages in thread
From: Vidya Sagar @ 2022-02-05 16:21 UTC (permalink / raw)
To: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh
Cc: kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, vidyas, sagar.tv
Add support for Synopsys DesignWare core IP based PCIe host controllers
present in the Tegra234 SoC.
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
---
drivers/pci/controller/dwc/pcie-tegra194.c | 409 ++++++++++++++++-----
1 file changed, 315 insertions(+), 94 deletions(-)
diff --git a/drivers/pci/controller/dwc/pcie-tegra194.c b/drivers/pci/controller/dwc/pcie-tegra194.c
index b1b5f836a806..512ab0767f11 100644
--- a/drivers/pci/controller/dwc/pcie-tegra194.c
+++ b/drivers/pci/controller/dwc/pcie-tegra194.c
@@ -1,8 +1,10 @@
// SPDX-License-Identifier: GPL-2.0+
/*
- * PCIe host controller driver for Tegra194 SoC
+ * PCIe host controller driver for the following SoCs
+ * Tegra194
+ * Tegra234
*
- * Copyright (C) 2019 NVIDIA Corporation.
+ * Copyright (C) 2019-2022 NVIDIA Corporation.
*
* Author: Vidya Sagar <vidyas@nvidia.com>
*/
@@ -35,6 +37,9 @@
#include <soc/tegra/bpmp-abi.h>
#include "../../pci.h"
+#define TEGRA194_DWC_IP_VER 0x490A
+#define TEGRA234_DWC_IP_VER 0x562A
+
#define APPL_PINMUX 0x0
#define APPL_PINMUX_PEX_RST BIT(0)
#define APPL_PINMUX_CLKREQ_OVERRIDE_EN BIT(2)
@@ -49,6 +54,7 @@
#define APPL_CTRL_HW_HOT_RST_MODE_MASK GENMASK(1, 0)
#define APPL_CTRL_HW_HOT_RST_MODE_SHIFT 22
#define APPL_CTRL_HW_HOT_RST_MODE_IMDT_RST 0x1
+#define APPL_CTRL_HW_HOT_RST_MODE_IMDT_RST_LTSSM_EN 0x2
#define APPL_INTR_EN_L0_0 0x8
#define APPL_INTR_EN_L0_0_LINK_STATE_INT_EN BIT(0)
@@ -245,6 +251,18 @@ static const unsigned int pcie_gen_freq[] = {
GEN4_CORE_CLK_FREQ
};
+struct tegra_pcie_of_data {
+ u32 version;
+ enum dw_pcie_device_mode mode;
+ bool msix_doorbell_access_fixup;
+ bool sbr_reset_fixup;
+ bool l1ss_exit_fixup;
+ bool ltr_req_fixup;
+ u32 cdm_chk_int_en;
+ u32 gen4_preset_vec;
+ u8 n_fts[2];
+};
+
struct tegra194_pcie {
struct device *dev;
struct resource *appl_res;
@@ -257,12 +275,14 @@ struct tegra194_pcie {
struct dw_pcie pci;
struct tegra_bpmp *bpmp;
- enum dw_pcie_device_mode mode;
+ struct tegra_pcie_of_data *of_data;
bool supports_clkreq;
bool enable_cdm_check;
+ bool enable_srns;
bool link_state;
bool update_fc_fixup;
+ bool enable_ext_refclk;
u8 init_link_width;
u32 msi_ctrl_int;
u32 num_lanes;
@@ -286,13 +306,10 @@ struct tegra194_pcie {
struct gpio_desc *pex_rst_gpiod;
struct gpio_desc *pex_refclk_sel_gpiod;
unsigned int pex_rst_irq;
+ bool pex_rst_irq_enabled;
int ep_state;
};
-struct tegra194_pcie_of_data {
- enum dw_pcie_device_mode mode;
-};
-
static inline struct tegra194_pcie *to_tegra_pcie(struct dw_pcie *pci)
{
return container_of(pci, struct tegra194_pcie, pci);
@@ -352,15 +369,15 @@ static irqreturn_t tegra_pcie_rp_irq_handler(int irq, void *arg)
struct tegra194_pcie *pcie = arg;
struct dw_pcie *pci = &pcie->pci;
struct pcie_port *pp = &pci->pp;
- u32 val, tmp;
+ u32 val, status_l0, status_l1;
u16 val_w;
- val = appl_readl(pcie, APPL_INTR_STATUS_L0);
- if (val & APPL_INTR_STATUS_L0_LINK_STATE_INT) {
- val = appl_readl(pcie, APPL_INTR_STATUS_L1_0_0);
- if (val & APPL_INTR_STATUS_L1_0_0_LINK_REQ_RST_NOT_CHGED) {
- appl_writel(pcie, val, APPL_INTR_STATUS_L1_0_0);
-
+ status_l0 = appl_readl(pcie, APPL_INTR_STATUS_L0);
+ if (status_l0 & APPL_INTR_STATUS_L0_LINK_STATE_INT) {
+ status_l1 = appl_readl(pcie, APPL_INTR_STATUS_L1_0_0);
+ appl_writel(pcie, status_l1, APPL_INTR_STATUS_L1_0_0);
+ if (pcie->of_data->sbr_reset_fixup &&
+ status_l1 & APPL_INTR_STATUS_L1_0_0_LINK_REQ_RST_NOT_CHGED) {
/* SBR & Surprise Link Down WAR */
val = appl_readl(pcie, APPL_CAR_RESET_OVRD);
val &= ~APPL_CAR_RESET_OVRD_CYA_OVERRIDE_CORE_RST_N;
@@ -376,15 +393,21 @@ static irqreturn_t tegra_pcie_rp_irq_handler(int irq, void *arg)
}
}
- if (val & APPL_INTR_STATUS_L0_INT_INT) {
- val = appl_readl(pcie, APPL_INTR_STATUS_L1_8_0);
- if (val & APPL_INTR_STATUS_L1_8_0_AUTO_BW_INT_STS) {
+ if (status_l0 & APPL_INTR_STATUS_L0_INT_INT) {
+ status_l1 = appl_readl(pcie, APPL_INTR_STATUS_L1_8_0);
+ if (status_l1 & APPL_INTR_STATUS_L1_8_0_AUTO_BW_INT_STS) {
appl_writel(pcie,
APPL_INTR_STATUS_L1_8_0_AUTO_BW_INT_STS,
APPL_INTR_STATUS_L1_8_0);
apply_bad_link_workaround(pp);
}
- if (val & APPL_INTR_STATUS_L1_8_0_BW_MGT_INT_STS) {
+ if (status_l1 & APPL_INTR_STATUS_L1_8_0_BW_MGT_INT_STS) {
+ val_w = dw_pcie_readw_dbi(pci, pcie->pcie_cap_base +
+ PCI_EXP_LNKSTA);
+ val_w |= PCI_EXP_LNKSTA_LBMS;
+ dw_pcie_writew_dbi(pci, pcie->pcie_cap_base +
+ PCI_EXP_LNKSTA, val_w);
+
appl_writel(pcie,
APPL_INTR_STATUS_L1_8_0_BW_MGT_INT_STS,
APPL_INTR_STATUS_L1_8_0);
@@ -396,25 +419,24 @@ static irqreturn_t tegra_pcie_rp_irq_handler(int irq, void *arg)
}
}
- val = appl_readl(pcie, APPL_INTR_STATUS_L0);
- if (val & APPL_INTR_STATUS_L0_CDM_REG_CHK_INT) {
- val = appl_readl(pcie, APPL_INTR_STATUS_L1_18);
- tmp = dw_pcie_readl_dbi(pci, PCIE_PL_CHK_REG_CONTROL_STATUS);
- if (val & APPL_INTR_STATUS_L1_18_CDM_REG_CHK_CMPLT) {
+ if (status_l0 & APPL_INTR_STATUS_L0_CDM_REG_CHK_INT) {
+ status_l1 = appl_readl(pcie, APPL_INTR_STATUS_L1_18);
+ val = dw_pcie_readl_dbi(pci, PCIE_PL_CHK_REG_CONTROL_STATUS);
+ if (status_l1 & APPL_INTR_STATUS_L1_18_CDM_REG_CHK_CMPLT) {
dev_info(pci->dev, "CDM check complete\n");
- tmp |= PCIE_PL_CHK_REG_CHK_REG_COMPLETE;
+ val |= PCIE_PL_CHK_REG_CHK_REG_COMPLETE;
}
- if (val & APPL_INTR_STATUS_L1_18_CDM_REG_CHK_CMP_ERR) {
+ if (status_l1 & APPL_INTR_STATUS_L1_18_CDM_REG_CHK_CMP_ERR) {
dev_err(pci->dev, "CDM comparison mismatch\n");
- tmp |= PCIE_PL_CHK_REG_CHK_REG_COMPARISON_ERROR;
+ val |= PCIE_PL_CHK_REG_CHK_REG_COMPARISON_ERROR;
}
- if (val & APPL_INTR_STATUS_L1_18_CDM_REG_CHK_LOGIC_ERR) {
+ if (status_l1 & APPL_INTR_STATUS_L1_18_CDM_REG_CHK_LOGIC_ERR) {
dev_err(pci->dev, "CDM Logic error\n");
- tmp |= PCIE_PL_CHK_REG_CHK_REG_LOGIC_ERROR;
+ val |= PCIE_PL_CHK_REG_CHK_REG_LOGIC_ERROR;
}
- dw_pcie_writel_dbi(pci, PCIE_PL_CHK_REG_CONTROL_STATUS, tmp);
- tmp = dw_pcie_readl_dbi(pci, PCIE_PL_CHK_REG_ERR_ADDR);
- dev_err(pci->dev, "CDM Error Address Offset = 0x%08X\n", tmp);
+ dw_pcie_writel_dbi(pci, PCIE_PL_CHK_REG_CONTROL_STATUS, val);
+ val = dw_pcie_readl_dbi(pci, PCIE_PL_CHK_REG_ERR_ADDR);
+ dev_err(pci->dev, "CDM Error Address Offset = 0x%08X\n", val);
}
return IRQ_HANDLED;
@@ -456,6 +478,9 @@ static irqreturn_t tegra_pcie_ep_irq_thread(int irq, void *arg)
PCI_EXP_LNKSTA_CLS;
clk_set_rate(pcie->core_clk, pcie_gen_freq[speed - 1]);
+ if (!pcie->of_data->ltr_req_fixup)
+ return IRQ_HANDLED;
+
/* If EP doesn't advertise L1SS, just return */
val = dw_pcie_readl_dbi(pci, pcie->cfg_link_cap_l1sub);
if (!(val & (PCI_L1SS_CAP_ASPM_L1_1 | PCI_L1SS_CAP_ASPM_L1_2)))
@@ -540,13 +565,18 @@ static irqreturn_t tegra_pcie_ep_hard_irq(int irq, void *arg)
static int tegra194_pcie_rd_own_conf(struct pci_bus *bus, u32 devfn, int where,
int size, u32 *val)
{
+ struct pcie_port *pp = bus->sysdata;
+ struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
+ struct tegra194_pcie *pcie = to_tegra_pcie(pci);
+
/*
* This is an endpoint mode specific register happen to appear even
* when controller is operating in root port mode and system hangs
* when it is accessed with link being in ASPM-L1 state.
* So skip accessing it altogether
*/
- if (!PCI_SLOT(devfn) && where == PORT_LOGIC_MSIX_DOORBELL) {
+ if (pcie->of_data->msix_doorbell_access_fixup &&
+ !PCI_SLOT(devfn) && where == PORT_LOGIC_MSIX_DOORBELL) {
*val = 0x00000000;
return PCIBIOS_SUCCESSFUL;
}
@@ -557,13 +587,18 @@ static int tegra194_pcie_rd_own_conf(struct pci_bus *bus, u32 devfn, int where,
static int tegra194_pcie_wr_own_conf(struct pci_bus *bus, u32 devfn, int where,
int size, u32 val)
{
+ struct pcie_port *pp = bus->sysdata;
+ struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
+ struct tegra194_pcie *pcie = to_tegra_pcie(pci);
+
/*
* This is an endpoint mode specific register happen to appear even
* when controller is operating in root port mode and system hangs
* when it is accessed with link being in ASPM-L1 state.
* So skip accessing it altogether
*/
- if (!PCI_SLOT(devfn) && where == PORT_LOGIC_MSIX_DOORBELL)
+ if (pcie->of_data->msix_doorbell_access_fixup &&
+ !PCI_SLOT(devfn) && where == PORT_LOGIC_MSIX_DOORBELL)
return PCIBIOS_SUCCESSFUL;
return pci_generic_config_write(bus, devfn, where, size, val);
@@ -711,13 +746,15 @@ static void tegra_pcie_enable_system_interrupts(struct pcie_port *pp)
val |= APPL_INTR_EN_L0_0_LINK_STATE_INT_EN;
appl_writel(pcie, val, APPL_INTR_EN_L0_0);
- val = appl_readl(pcie, APPL_INTR_EN_L1_0_0);
- val |= APPL_INTR_EN_L1_0_0_LINK_REQ_RST_NOT_INT_EN;
- appl_writel(pcie, val, APPL_INTR_EN_L1_0_0);
+ if (pcie->of_data->sbr_reset_fixup) {
+ val = appl_readl(pcie, APPL_INTR_EN_L1_0_0);
+ val |= APPL_INTR_EN_L1_0_0_LINK_REQ_RST_NOT_INT_EN;
+ appl_writel(pcie, val, APPL_INTR_EN_L1_0_0);
+ }
if (pcie->enable_cdm_check) {
val = appl_readl(pcie, APPL_INTR_EN_L0_0);
- val |= APPL_INTR_EN_L0_0_CDM_REG_CHK_INT_EN;
+ val |= pcie->of_data->cdm_chk_int_en;
appl_writel(pcie, val, APPL_INTR_EN_L0_0);
val = appl_readl(pcie, APPL_INTR_EN_L1_18);
@@ -844,7 +881,8 @@ static void config_gen3_gen4_eq_presets(struct tegra194_pcie *pcie)
val = dw_pcie_readl_dbi(pci, GEN3_EQ_CONTROL_OFF);
val &= ~GEN3_EQ_CONTROL_OFF_PSET_REQ_VEC_MASK;
- val |= (0x360 << GEN3_EQ_CONTROL_OFF_PSET_REQ_VEC_SHIFT);
+ val |= (pcie->of_data->gen4_preset_vec <<
+ GEN3_EQ_CONTROL_OFF_PSET_REQ_VEC_SHIFT);
val &= ~GEN3_EQ_CONTROL_OFF_FB_MODE_MASK;
dw_pcie_writel_dbi(pci, GEN3_EQ_CONTROL_OFF, val);
@@ -858,6 +896,7 @@ static int tegra194_pcie_host_init(struct pcie_port *pp)
struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
struct tegra194_pcie *pcie = to_tegra_pcie(pci);
u32 val;
+ u16 val_16;
pp->bridge->ops = &tegra_pci_ops;
@@ -865,6 +904,11 @@ static int tegra194_pcie_host_init(struct pcie_port *pp)
pcie->pcie_cap_base = dw_pcie_find_capability(&pcie->pci,
PCI_CAP_ID_EXP);
+ val_16 = dw_pcie_readw_dbi(pci, pcie->pcie_cap_base + PCI_EXP_DEVCTL);
+ val_16 &= ~PCI_EXP_DEVCTL_PAYLOAD;
+ val_16 |= PCI_EXP_DEVCTL_PAYLOAD_256B;
+ dw_pcie_writew_dbi(pci, pcie->pcie_cap_base + PCI_EXP_DEVCTL, val_16);
+
val = dw_pcie_readl_dbi(pci, PCI_IO_BASE);
val &= ~(IO_BASE_IO_DECODE | IO_BASE_IO_DECODE_BIT8);
dw_pcie_writel_dbi(pci, PCI_IO_BASE, val);
@@ -889,6 +933,15 @@ static int tegra194_pcie_host_init(struct pcie_port *pp)
val |= (pcie->num_lanes << PCI_EXP_LNKSTA_NLW_SHIFT);
dw_pcie_writel_dbi(pci, pcie->pcie_cap_base + PCI_EXP_LNKCAP, val);
+ /* Clear Slot Clock Configuration bit if SRNS configuration */
+ if (pcie->enable_srns) {
+ val_16 = dw_pcie_readw_dbi(pci, pcie->pcie_cap_base +
+ PCI_EXP_LNKSTA);
+ val_16 &= ~PCI_EXP_LNKSTA_SLC;
+ dw_pcie_writew_dbi(pci, pcie->pcie_cap_base + PCI_EXP_LNKSTA,
+ val_16);
+ }
+
config_gen3_gen4_eq_presets(pcie);
init_host_aspm(pcie);
@@ -899,9 +952,11 @@ static int tegra194_pcie_host_init(struct pcie_port *pp)
disable_aspm_l12(pcie);
}
- val = dw_pcie_readl_dbi(pci, GEN3_RELATED_OFF);
- val &= ~GEN3_RELATED_OFF_GEN3_ZRXDC_NONCOMPL;
- dw_pcie_writel_dbi(pci, GEN3_RELATED_OFF, val);
+ if (pcie->of_data->l1ss_exit_fixup) {
+ val = dw_pcie_readl_dbi(pci, GEN3_RELATED_OFF);
+ val &= ~GEN3_RELATED_OFF_GEN3_ZRXDC_NONCOMPL;
+ dw_pcie_writel_dbi(pci, GEN3_RELATED_OFF, val);
+ }
if (pcie->update_fc_fixup) {
val = dw_pcie_readl_dbi(pci, CFG_TIMER_CTRL_MAX_FUNC_NUM_OFF);
@@ -921,8 +976,11 @@ static int tegra194_pcie_start_link(struct dw_pcie *pci)
struct pcie_port *pp = &pci->pp;
bool retry = true;
- if (pcie->mode == DW_PCIE_EP_TYPE) {
- enable_irq(pcie->pex_rst_irq);
+ if (pcie->of_data->mode == DW_PCIE_EP_TYPE) {
+ if (!pcie->pex_rst_irq_enabled) {
+ enable_irq(pcie->pex_rst_irq);
+ pcie->pex_rst_irq_enabled = true;
+ }
return 0;
}
@@ -1113,13 +1171,26 @@ static int tegra194_pcie_parse_dt(struct tegra194_pcie *pcie)
if (of_property_read_bool(np, "nvidia,update-fc-fixup"))
pcie->update_fc_fixup = true;
+ pcie->enable_ext_refclk =
+ of_property_read_bool(pcie->dev->of_node,
+ "nvidia,enable-ext-refclk");
+ /* RP using an external REFCLK is supported only in Tegra234 */
+ if (pcie->of_data->version == TEGRA194_DWC_IP_VER) {
+ if (pcie->of_data->mode == DW_PCIE_RC_TYPE)
+ pcie->enable_ext_refclk = false;
+ else
+ pcie->enable_ext_refclk = true;
+ }
+
pcie->supports_clkreq =
of_property_read_bool(pcie->dev->of_node, "supports-clkreq");
pcie->enable_cdm_check =
of_property_read_bool(np, "snps,enable-cdm-check");
- if (pcie->mode == DW_PCIE_RC_TYPE)
+ pcie->enable_srns = of_property_read_bool(np, "nvidia,enable-srns");
+
+ if (pcie->of_data->mode == DW_PCIE_RC_TYPE)
return 0;
/* Endpoint mode specific DT entries */
@@ -1163,8 +1234,11 @@ static int tegra_pcie_bpmp_set_ctrl_state(struct tegra194_pcie *pcie,
struct tegra_bpmp_message msg;
struct mrq_uphy_request req;
- /* Controller-5 doesn't need to have its state set by BPMP-FW */
- if (pcie->cid == 5)
+ /*
+ * Controller-5 doesn't need to have its state set by BPMP-FW in
+ * Tegra194
+ */
+ if (pcie->cid == 5 && pcie->of_data->version == 0x490A)
return 0;
memset(&req, 0, sizeof(req));
@@ -1330,6 +1404,14 @@ static int tegra_pcie_config_controller(struct tegra194_pcie *pcie,
return ret;
}
+ if (pcie->enable_ext_refclk) {
+ ret = tegra_pcie_bpmp_set_pll_state(pcie, true);
+ if (ret) {
+ dev_err(pcie->dev, "Failed to init UPHY: %d\n", ret);
+ goto fail_pll_init;
+ }
+ }
+
ret = tegra_pcie_enable_slot_regulators(pcie);
if (ret < 0)
goto fail_slot_reg_en;
@@ -1353,11 +1435,13 @@ static int tegra_pcie_config_controller(struct tegra194_pcie *pcie,
goto fail_core_apb_rst;
}
- if (en_hw_hot_rst) {
+ if (en_hw_hot_rst || !pcie->of_data->sbr_reset_fixup) {
/* Enable HW_HOT_RST mode */
val = appl_readl(pcie, APPL_CTRL);
val &= ~(APPL_CTRL_HW_HOT_RST_MODE_MASK <<
APPL_CTRL_HW_HOT_RST_MODE_SHIFT);
+ val |= (APPL_CTRL_HW_HOT_RST_MODE_IMDT_RST_LTSSM_EN <<
+ APPL_CTRL_HW_HOT_RST_MODE_SHIFT);
val |= APPL_CTRL_HW_HOT_RST_EN;
appl_writel(pcie, val, APPL_CTRL);
}
@@ -1384,6 +1468,19 @@ static int tegra_pcie_config_controller(struct tegra194_pcie *pcie,
val |= (APPL_CFG_MISC_ARCACHE_VAL << APPL_CFG_MISC_ARCACHE_SHIFT);
appl_writel(pcie, val, APPL_CFG_MISC);
+ if (pcie->enable_srns || pcie->enable_ext_refclk) {
+ /*
+ * When Tegra PCIe RP is using external clock, it cannot
+ * supply same clock back to EP, which makes it separate clock.
+ * Gate PCIe RP REFCLK out pads when RP & EP are using separate
+ * clock or RP is using external REFCLK.
+ */
+ val = appl_readl(pcie, APPL_PINMUX);
+ val |= APPL_PINMUX_CLK_OUTPUT_IN_OVERRIDE_EN;
+ val &= ~APPL_PINMUX_CLK_OUTPUT_IN_OVERRIDE;
+ appl_writel(pcie, val, APPL_PINMUX);
+ }
+
if (!pcie->supports_clkreq) {
val = appl_readl(pcie, APPL_PINMUX);
val |= APPL_PINMUX_CLKREQ_OVERRIDE_EN;
@@ -1409,6 +1506,9 @@ static int tegra_pcie_config_controller(struct tegra194_pcie *pcie,
fail_reg_en:
tegra_pcie_disable_slot_regulators(pcie);
fail_slot_reg_en:
+ if (pcie->enable_ext_refclk)
+ tegra_pcie_bpmp_set_pll_state(pcie, false);
+fail_pll_init:
tegra_pcie_bpmp_set_ctrl_state(pcie, false);
return ret;
@@ -1436,6 +1536,12 @@ static void tegra_pcie_unconfig_controller(struct tegra194_pcie *pcie)
tegra_pcie_disable_slot_regulators(pcie);
+ if (pcie->enable_ext_refclk) {
+ ret = tegra_pcie_bpmp_set_pll_state(pcie, false);
+ if (ret)
+ dev_err(pcie->dev, "Failed to deinit UPHY: %d\n", ret);
+ }
+
ret = tegra_pcie_bpmp_set_ctrl_state(pcie, false);
if (ret)
dev_err(pcie->dev, "Failed to disable controller %d: %d\n",
@@ -1636,6 +1742,13 @@ static void pex_ep_event_pex_rst_assert(struct tegra194_pcie *pcie)
pm_runtime_put_sync(pcie->dev);
+ if (pcie->enable_ext_refclk) {
+ ret = tegra_pcie_bpmp_set_pll_state(pcie, false);
+ if (ret)
+ dev_err(pcie->dev, "Failed to turn off UPHY: %d\n",
+ ret);
+ }
+
ret = tegra_pcie_bpmp_set_pll_state(pcie, false);
if (ret)
dev_err(pcie->dev, "Failed to turn off UPHY: %d\n", ret);
@@ -1651,6 +1764,7 @@ static void pex_ep_event_pex_rst_deassert(struct tegra194_pcie *pcie)
struct device *dev = pcie->dev;
u32 val;
int ret;
+ u16 val_16;
if (pcie->ep_state == EP_STATE_ENABLED)
return;
@@ -1662,10 +1776,20 @@ static void pex_ep_event_pex_rst_deassert(struct tegra194_pcie *pcie)
return;
}
- ret = tegra_pcie_bpmp_set_pll_state(pcie, true);
+ ret = tegra_pcie_bpmp_set_ctrl_state(pcie, true);
if (ret) {
- dev_err(dev, "Failed to init UPHY for PCIe EP: %d\n", ret);
- goto fail_pll_init;
+ dev_err(pcie->dev, "Failed to enable controller %u: %d\n",
+ pcie->cid, ret);
+ goto fail_set_ctrl_state;
+ }
+
+ if (pcie->enable_ext_refclk) {
+ ret = tegra_pcie_bpmp_set_pll_state(pcie, true);
+ if (ret) {
+ dev_err(dev, "Failed to init UPHY for PCIe EP: %d\n",
+ ret);
+ goto fail_pll_init;
+ }
}
ret = clk_prepare_enable(pcie->core_clk);
@@ -1762,12 +1886,29 @@ static void pex_ep_event_pex_rst_deassert(struct tegra194_pcie *pcie)
disable_aspm_l12(pcie);
}
- val = dw_pcie_readl_dbi(pci, GEN3_RELATED_OFF);
- val &= ~GEN3_RELATED_OFF_GEN3_ZRXDC_NONCOMPL;
- dw_pcie_writel_dbi(pci, GEN3_RELATED_OFF, val);
+ if (pcie->of_data->l1ss_exit_fixup) {
+ val = dw_pcie_readl_dbi(pci, GEN3_RELATED_OFF);
+ val &= ~GEN3_RELATED_OFF_GEN3_ZRXDC_NONCOMPL;
+ dw_pcie_writel_dbi(pci, GEN3_RELATED_OFF, val);
+ }
pcie->pcie_cap_base = dw_pcie_find_capability(&pcie->pci,
PCI_CAP_ID_EXP);
+
+ val_16 = dw_pcie_readw_dbi(pci, pcie->pcie_cap_base + PCI_EXP_DEVCTL);
+ val_16 &= ~PCI_EXP_DEVCTL_PAYLOAD;
+ val_16 |= PCI_EXP_DEVCTL_PAYLOAD_256B;
+ dw_pcie_writew_dbi(pci, pcie->pcie_cap_base + PCI_EXP_DEVCTL, val_16);
+
+ /* Clear Slot Clock Configuration bit if SRNS configuration */
+ if (pcie->enable_srns) {
+ val_16 = dw_pcie_readw_dbi(pci, pcie->pcie_cap_base +
+ PCI_EXP_LNKSTA);
+ val_16 &= ~PCI_EXP_LNKSTA_SLC;
+ dw_pcie_writew_dbi(pci, pcie->pcie_cap_base + PCI_EXP_LNKSTA,
+ val_16);
+ }
+
clk_set_rate(pcie->core_clk, GEN4_CORE_CLK_FREQ);
val = (ep->msi_mem_phys & MSIX_ADDR_MATCH_LOW_OFF_MASK);
@@ -1784,6 +1925,13 @@ static void pex_ep_event_pex_rst_deassert(struct tegra194_pcie *pcie)
dw_pcie_ep_init_notify(ep);
+ /* Send LTR upstream */
+ if (!pcie->of_data->ltr_req_fixup) {
+ val = appl_readl(pcie, APPL_LTR_MSG_2);
+ val |= APPL_LTR_MSG_2_LTR_MSG_REQ_STATE;
+ appl_writel(pcie, val, APPL_LTR_MSG_2);
+ }
+
/* Enable LTSSM */
val = appl_readl(pcie, APPL_CTRL);
val |= APPL_CTRL_LTSSM_EN;
@@ -1804,6 +1952,8 @@ static void pex_ep_event_pex_rst_deassert(struct tegra194_pcie *pcie)
fail_core_clk_enable:
tegra_pcie_bpmp_set_pll_state(pcie, false);
fail_pll_init:
+ tegra_pcie_bpmp_set_ctrl_state(pcie, false);
+fail_set_ctrl_state:
pm_runtime_put_sync(dev);
}
@@ -1933,6 +2083,7 @@ static int tegra_pcie_config_ep(struct tegra194_pcie *pcie,
irq_set_status_flags(pcie->pex_rst_irq, IRQ_NOAUTOEN);
+ pcie->pex_rst_irq_enabled = false;
pcie->ep_state = EP_STATE_DISABLED;
ret = devm_request_threaded_irq(dev, pcie->pex_rst_irq, NULL,
@@ -1979,14 +2130,13 @@ static int tegra194_pcie_probe(struct platform_device *pdev)
pci = &pcie->pci;
pci->dev = &pdev->dev;
pci->ops = &tegra_dw_pcie_ops;
- pci->n_fts[0] = N_FTS_VAL;
- pci->n_fts[1] = FTS_VAL;
- pci->version = 0x490A;
-
+ pcie->dev = &pdev->dev;
+ pcie->of_data = (struct tegra_pcie_of_data *)data;
+ pci->n_fts[0] = pcie->of_data->n_fts[0];
+ pci->n_fts[1] = pcie->of_data->n_fts[1];
+ pci->version = pcie->of_data->version;
pp = &pci->pp;
pp->num_vectors = MAX_MSI_IRQS;
- pcie->dev = &pdev->dev;
- pcie->mode = (enum dw_pcie_device_mode)data->mode;
ret = tegra194_pcie_parse_dt(pcie);
if (ret < 0) {
@@ -2103,7 +2253,7 @@ static int tegra194_pcie_probe(struct platform_device *pdev)
platform_set_drvdata(pdev, pcie);
- switch (pcie->mode) {
+ switch (pcie->of_data->mode) {
case DW_PCIE_RC_TYPE:
ret = devm_request_irq(dev, pp->irq, tegra_pcie_rp_irq_handler,
IRQF_SHARED, "tegra-pcie-intr", pcie);
@@ -2138,7 +2288,8 @@ static int tegra194_pcie_probe(struct platform_device *pdev)
break;
default:
- dev_err(dev, "Invalid PCIe device type %d\n", pcie->mode);
+ dev_err(dev, "Invalid PCIe device type %d\n",
+ pcie->of_data->mode);
}
fail:
@@ -2150,12 +2301,19 @@ static int tegra194_pcie_remove(struct platform_device *pdev)
{
struct tegra194_pcie *pcie = platform_get_drvdata(pdev);
- if (!pcie->link_state)
- return 0;
+ if (pcie->of_data->mode == DW_PCIE_RC_TYPE) {
+ if (!pcie->link_state)
+ return 0;
+
+ debugfs_remove_recursive(pcie->debugfs);
+ tegra_pcie_deinit_controller(pcie);
+ pm_runtime_put_sync(pcie->dev);
+ } else {
+ if (pcie->pex_rst_irq_enabled)
+ disable_irq(pcie->pex_rst_irq);
+ pex_ep_event_pex_rst_assert(pcie);
+ }
- debugfs_remove_recursive(pcie->debugfs);
- tegra_pcie_deinit_controller(pcie);
- pm_runtime_put_sync(pcie->dev);
pm_runtime_disable(pcie->dev);
tegra_bpmp_put(pcie->bpmp);
if (pcie->pex_refclk_sel_gpiod)
@@ -2169,15 +2327,22 @@ static int tegra194_pcie_suspend_late(struct device *dev)
struct tegra194_pcie *pcie = dev_get_drvdata(dev);
u32 val;
+ if (pcie->of_data->mode == DW_PCIE_EP_TYPE) {
+ dev_err(dev, "Failed to Suspend as Tegra PCIe is in EP mode\n");
+ return -EPERM;
+ }
+
if (!pcie->link_state)
return 0;
/* Enable HW_HOT_RST mode */
- val = appl_readl(pcie, APPL_CTRL);
- val &= ~(APPL_CTRL_HW_HOT_RST_MODE_MASK <<
- APPL_CTRL_HW_HOT_RST_MODE_SHIFT);
- val |= APPL_CTRL_HW_HOT_RST_EN;
- appl_writel(pcie, val, APPL_CTRL);
+ if (pcie->of_data->sbr_reset_fixup) {
+ val = appl_readl(pcie, APPL_CTRL);
+ val &= ~(APPL_CTRL_HW_HOT_RST_MODE_MASK <<
+ APPL_CTRL_HW_HOT_RST_MODE_SHIFT);
+ val |= APPL_CTRL_HW_HOT_RST_EN;
+ appl_writel(pcie, val, APPL_CTRL);
+ }
return 0;
}
@@ -2239,22 +2404,19 @@ static int tegra194_pcie_resume_early(struct device *dev)
struct tegra194_pcie *pcie = dev_get_drvdata(dev);
u32 val;
- if (pcie->mode == DW_PCIE_EP_TYPE) {
- dev_err(dev, "Suspend is not supported in EP mode");
- return -ENOTSUPP;
- }
-
if (!pcie->link_state)
return 0;
/* Disable HW_HOT_RST mode */
- val = appl_readl(pcie, APPL_CTRL);
- val &= ~(APPL_CTRL_HW_HOT_RST_MODE_MASK <<
- APPL_CTRL_HW_HOT_RST_MODE_SHIFT);
- val |= APPL_CTRL_HW_HOT_RST_MODE_IMDT_RST <<
- APPL_CTRL_HW_HOT_RST_MODE_SHIFT;
- val &= ~APPL_CTRL_HW_HOT_RST_EN;
- appl_writel(pcie, val, APPL_CTRL);
+ if (pcie->of_data->sbr_reset_fixup) {
+ val = appl_readl(pcie, APPL_CTRL);
+ val &= ~(APPL_CTRL_HW_HOT_RST_MODE_MASK <<
+ APPL_CTRL_HW_HOT_RST_MODE_SHIFT);
+ val |= APPL_CTRL_HW_HOT_RST_MODE_IMDT_RST <<
+ APPL_CTRL_HW_HOT_RST_MODE_SHIFT;
+ val &= ~APPL_CTRL_HW_HOT_RST_EN;
+ appl_writel(pcie, val, APPL_CTRL);
+ }
return 0;
}
@@ -2263,26 +2425,77 @@ static void tegra194_pcie_shutdown(struct platform_device *pdev)
{
struct tegra194_pcie *pcie = platform_get_drvdata(pdev);
- if (!pcie->link_state)
- return;
+ if (pcie->of_data->mode == DW_PCIE_RC_TYPE) {
+ if (!pcie->link_state)
+ return;
- debugfs_remove_recursive(pcie->debugfs);
- tegra_pcie_downstream_dev_to_D0(pcie);
+ debugfs_remove_recursive(pcie->debugfs);
+ tegra_pcie_downstream_dev_to_D0(pcie);
- disable_irq(pcie->pci.pp.irq);
- if (IS_ENABLED(CONFIG_PCI_MSI))
- disable_irq(pcie->pci.pp.msi_irq);
+ disable_irq(pcie->pci.pp.irq);
+ if (IS_ENABLED(CONFIG_PCI_MSI))
+ disable_irq(pcie->pci.pp.msi_irq);
- tegra194_pcie_pme_turnoff(pcie);
- tegra_pcie_unconfig_controller(pcie);
+ tegra194_pcie_pme_turnoff(pcie);
+ tegra_pcie_unconfig_controller(pcie);
+ pm_runtime_put_sync(pcie->dev);
+ } else {
+ if (pcie->pex_rst_irq_enabled)
+ disable_irq(pcie->pex_rst_irq);
+ pex_ep_event_pex_rst_assert(pcie);
+ }
}
-static const struct tegra194_pcie_of_data tegra194_pcie_rc_of_data = {
+static const struct tegra_pcie_of_data tegra194_pcie_rc_of_data = {
+ .version = TEGRA194_DWC_IP_VER,
+ .mode = DW_PCIE_RC_TYPE,
+ .msix_doorbell_access_fixup = true,
+ .sbr_reset_fixup = true,
+ .l1ss_exit_fixup = true,
+ .ltr_req_fixup = false,
+ .cdm_chk_int_en = BIT(19),
+ /* Gen4 - 5, 6, 8 and 9 presets enabled */
+ .gen4_preset_vec = 0x360,
+ .n_fts = { 52, 52 },
+};
+
+static const struct tegra_pcie_of_data tegra194_pcie_ep_of_data = {
+ .version = TEGRA194_DWC_IP_VER,
+ .mode = DW_PCIE_EP_TYPE,
+ .msix_doorbell_access_fixup = false,
+ .sbr_reset_fixup = false,
+ .l1ss_exit_fixup = true,
+ .ltr_req_fixup = true,
+ .cdm_chk_int_en = BIT(19),
+ /* Gen4 - 5, 6, 8 and 9 presets enabled */
+ .gen4_preset_vec = 0x360,
+ .n_fts = { 52, 52 },
+};
+
+static const struct tegra_pcie_of_data tegra234_pcie_rc_of_data = {
+ .version = TEGRA234_DWC_IP_VER,
.mode = DW_PCIE_RC_TYPE,
+ .msix_doorbell_access_fixup = false,
+ .sbr_reset_fixup = false,
+ .l1ss_exit_fixup = false,
+ .ltr_req_fixup = false,
+ .cdm_chk_int_en = BIT(18),
+ /* Gen4 - 6, 8 and 9 presets enabled */
+ .gen4_preset_vec = 0x340,
+ .n_fts = { 52, 80 },
};
-static const struct tegra194_pcie_of_data tegra194_pcie_ep_of_data = {
+static const struct tegra_pcie_of_data tegra234_pcie_ep_of_data = {
+ .version = TEGRA234_DWC_IP_VER,
.mode = DW_PCIE_EP_TYPE,
+ .msix_doorbell_access_fixup = false,
+ .sbr_reset_fixup = false,
+ .l1ss_exit_fixup = false,
+ .ltr_req_fixup = false,
+ .cdm_chk_int_en = BIT(18),
+ /* Gen4 - 6, 8 and 9 presets enabled */
+ .gen4_preset_vec = 0x340,
+ .n_fts = { 52, 80 },
};
static const struct of_device_id tegra194_pcie_of_match[] = {
@@ -2294,6 +2507,14 @@ static const struct of_device_id tegra194_pcie_of_match[] = {
.compatible = "nvidia,tegra194-pcie-ep",
.data = &tegra194_pcie_ep_of_data,
},
+ {
+ .compatible = "nvidia,tegra234-pcie",
+ .data = &tegra234_pcie_rc_of_data,
+ },
+ {
+ .compatible = "nvidia,tegra234-pcie-ep",
+ .data = &tegra234_pcie_ep_of_data,
+ },
{},
};
--
2.17.1
^ permalink raw reply related [flat|nested] 46+ messages in thread
* [PATCH V1 10/10] PCI: tegra: Add Tegra234 PCIe support
@ 2022-02-05 16:21 ` Vidya Sagar
0 siblings, 0 replies; 46+ messages in thread
From: Vidya Sagar @ 2022-02-05 16:21 UTC (permalink / raw)
To: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh
Cc: kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, vidyas, sagar.tv
Add support for Synopsys DesignWare core IP based PCIe host controllers
present in the Tegra234 SoC.
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
---
drivers/pci/controller/dwc/pcie-tegra194.c | 409 ++++++++++++++++-----
1 file changed, 315 insertions(+), 94 deletions(-)
diff --git a/drivers/pci/controller/dwc/pcie-tegra194.c b/drivers/pci/controller/dwc/pcie-tegra194.c
index b1b5f836a806..512ab0767f11 100644
--- a/drivers/pci/controller/dwc/pcie-tegra194.c
+++ b/drivers/pci/controller/dwc/pcie-tegra194.c
@@ -1,8 +1,10 @@
// SPDX-License-Identifier: GPL-2.0+
/*
- * PCIe host controller driver for Tegra194 SoC
+ * PCIe host controller driver for the following SoCs
+ * Tegra194
+ * Tegra234
*
- * Copyright (C) 2019 NVIDIA Corporation.
+ * Copyright (C) 2019-2022 NVIDIA Corporation.
*
* Author: Vidya Sagar <vidyas@nvidia.com>
*/
@@ -35,6 +37,9 @@
#include <soc/tegra/bpmp-abi.h>
#include "../../pci.h"
+#define TEGRA194_DWC_IP_VER 0x490A
+#define TEGRA234_DWC_IP_VER 0x562A
+
#define APPL_PINMUX 0x0
#define APPL_PINMUX_PEX_RST BIT(0)
#define APPL_PINMUX_CLKREQ_OVERRIDE_EN BIT(2)
@@ -49,6 +54,7 @@
#define APPL_CTRL_HW_HOT_RST_MODE_MASK GENMASK(1, 0)
#define APPL_CTRL_HW_HOT_RST_MODE_SHIFT 22
#define APPL_CTRL_HW_HOT_RST_MODE_IMDT_RST 0x1
+#define APPL_CTRL_HW_HOT_RST_MODE_IMDT_RST_LTSSM_EN 0x2
#define APPL_INTR_EN_L0_0 0x8
#define APPL_INTR_EN_L0_0_LINK_STATE_INT_EN BIT(0)
@@ -245,6 +251,18 @@ static const unsigned int pcie_gen_freq[] = {
GEN4_CORE_CLK_FREQ
};
+struct tegra_pcie_of_data {
+ u32 version;
+ enum dw_pcie_device_mode mode;
+ bool msix_doorbell_access_fixup;
+ bool sbr_reset_fixup;
+ bool l1ss_exit_fixup;
+ bool ltr_req_fixup;
+ u32 cdm_chk_int_en;
+ u32 gen4_preset_vec;
+ u8 n_fts[2];
+};
+
struct tegra194_pcie {
struct device *dev;
struct resource *appl_res;
@@ -257,12 +275,14 @@ struct tegra194_pcie {
struct dw_pcie pci;
struct tegra_bpmp *bpmp;
- enum dw_pcie_device_mode mode;
+ struct tegra_pcie_of_data *of_data;
bool supports_clkreq;
bool enable_cdm_check;
+ bool enable_srns;
bool link_state;
bool update_fc_fixup;
+ bool enable_ext_refclk;
u8 init_link_width;
u32 msi_ctrl_int;
u32 num_lanes;
@@ -286,13 +306,10 @@ struct tegra194_pcie {
struct gpio_desc *pex_rst_gpiod;
struct gpio_desc *pex_refclk_sel_gpiod;
unsigned int pex_rst_irq;
+ bool pex_rst_irq_enabled;
int ep_state;
};
-struct tegra194_pcie_of_data {
- enum dw_pcie_device_mode mode;
-};
-
static inline struct tegra194_pcie *to_tegra_pcie(struct dw_pcie *pci)
{
return container_of(pci, struct tegra194_pcie, pci);
@@ -352,15 +369,15 @@ static irqreturn_t tegra_pcie_rp_irq_handler(int irq, void *arg)
struct tegra194_pcie *pcie = arg;
struct dw_pcie *pci = &pcie->pci;
struct pcie_port *pp = &pci->pp;
- u32 val, tmp;
+ u32 val, status_l0, status_l1;
u16 val_w;
- val = appl_readl(pcie, APPL_INTR_STATUS_L0);
- if (val & APPL_INTR_STATUS_L0_LINK_STATE_INT) {
- val = appl_readl(pcie, APPL_INTR_STATUS_L1_0_0);
- if (val & APPL_INTR_STATUS_L1_0_0_LINK_REQ_RST_NOT_CHGED) {
- appl_writel(pcie, val, APPL_INTR_STATUS_L1_0_0);
-
+ status_l0 = appl_readl(pcie, APPL_INTR_STATUS_L0);
+ if (status_l0 & APPL_INTR_STATUS_L0_LINK_STATE_INT) {
+ status_l1 = appl_readl(pcie, APPL_INTR_STATUS_L1_0_0);
+ appl_writel(pcie, status_l1, APPL_INTR_STATUS_L1_0_0);
+ if (pcie->of_data->sbr_reset_fixup &&
+ status_l1 & APPL_INTR_STATUS_L1_0_0_LINK_REQ_RST_NOT_CHGED) {
/* SBR & Surprise Link Down WAR */
val = appl_readl(pcie, APPL_CAR_RESET_OVRD);
val &= ~APPL_CAR_RESET_OVRD_CYA_OVERRIDE_CORE_RST_N;
@@ -376,15 +393,21 @@ static irqreturn_t tegra_pcie_rp_irq_handler(int irq, void *arg)
}
}
- if (val & APPL_INTR_STATUS_L0_INT_INT) {
- val = appl_readl(pcie, APPL_INTR_STATUS_L1_8_0);
- if (val & APPL_INTR_STATUS_L1_8_0_AUTO_BW_INT_STS) {
+ if (status_l0 & APPL_INTR_STATUS_L0_INT_INT) {
+ status_l1 = appl_readl(pcie, APPL_INTR_STATUS_L1_8_0);
+ if (status_l1 & APPL_INTR_STATUS_L1_8_0_AUTO_BW_INT_STS) {
appl_writel(pcie,
APPL_INTR_STATUS_L1_8_0_AUTO_BW_INT_STS,
APPL_INTR_STATUS_L1_8_0);
apply_bad_link_workaround(pp);
}
- if (val & APPL_INTR_STATUS_L1_8_0_BW_MGT_INT_STS) {
+ if (status_l1 & APPL_INTR_STATUS_L1_8_0_BW_MGT_INT_STS) {
+ val_w = dw_pcie_readw_dbi(pci, pcie->pcie_cap_base +
+ PCI_EXP_LNKSTA);
+ val_w |= PCI_EXP_LNKSTA_LBMS;
+ dw_pcie_writew_dbi(pci, pcie->pcie_cap_base +
+ PCI_EXP_LNKSTA, val_w);
+
appl_writel(pcie,
APPL_INTR_STATUS_L1_8_0_BW_MGT_INT_STS,
APPL_INTR_STATUS_L1_8_0);
@@ -396,25 +419,24 @@ static irqreturn_t tegra_pcie_rp_irq_handler(int irq, void *arg)
}
}
- val = appl_readl(pcie, APPL_INTR_STATUS_L0);
- if (val & APPL_INTR_STATUS_L0_CDM_REG_CHK_INT) {
- val = appl_readl(pcie, APPL_INTR_STATUS_L1_18);
- tmp = dw_pcie_readl_dbi(pci, PCIE_PL_CHK_REG_CONTROL_STATUS);
- if (val & APPL_INTR_STATUS_L1_18_CDM_REG_CHK_CMPLT) {
+ if (status_l0 & APPL_INTR_STATUS_L0_CDM_REG_CHK_INT) {
+ status_l1 = appl_readl(pcie, APPL_INTR_STATUS_L1_18);
+ val = dw_pcie_readl_dbi(pci, PCIE_PL_CHK_REG_CONTROL_STATUS);
+ if (status_l1 & APPL_INTR_STATUS_L1_18_CDM_REG_CHK_CMPLT) {
dev_info(pci->dev, "CDM check complete\n");
- tmp |= PCIE_PL_CHK_REG_CHK_REG_COMPLETE;
+ val |= PCIE_PL_CHK_REG_CHK_REG_COMPLETE;
}
- if (val & APPL_INTR_STATUS_L1_18_CDM_REG_CHK_CMP_ERR) {
+ if (status_l1 & APPL_INTR_STATUS_L1_18_CDM_REG_CHK_CMP_ERR) {
dev_err(pci->dev, "CDM comparison mismatch\n");
- tmp |= PCIE_PL_CHK_REG_CHK_REG_COMPARISON_ERROR;
+ val |= PCIE_PL_CHK_REG_CHK_REG_COMPARISON_ERROR;
}
- if (val & APPL_INTR_STATUS_L1_18_CDM_REG_CHK_LOGIC_ERR) {
+ if (status_l1 & APPL_INTR_STATUS_L1_18_CDM_REG_CHK_LOGIC_ERR) {
dev_err(pci->dev, "CDM Logic error\n");
- tmp |= PCIE_PL_CHK_REG_CHK_REG_LOGIC_ERROR;
+ val |= PCIE_PL_CHK_REG_CHK_REG_LOGIC_ERROR;
}
- dw_pcie_writel_dbi(pci, PCIE_PL_CHK_REG_CONTROL_STATUS, tmp);
- tmp = dw_pcie_readl_dbi(pci, PCIE_PL_CHK_REG_ERR_ADDR);
- dev_err(pci->dev, "CDM Error Address Offset = 0x%08X\n", tmp);
+ dw_pcie_writel_dbi(pci, PCIE_PL_CHK_REG_CONTROL_STATUS, val);
+ val = dw_pcie_readl_dbi(pci, PCIE_PL_CHK_REG_ERR_ADDR);
+ dev_err(pci->dev, "CDM Error Address Offset = 0x%08X\n", val);
}
return IRQ_HANDLED;
@@ -456,6 +478,9 @@ static irqreturn_t tegra_pcie_ep_irq_thread(int irq, void *arg)
PCI_EXP_LNKSTA_CLS;
clk_set_rate(pcie->core_clk, pcie_gen_freq[speed - 1]);
+ if (!pcie->of_data->ltr_req_fixup)
+ return IRQ_HANDLED;
+
/* If EP doesn't advertise L1SS, just return */
val = dw_pcie_readl_dbi(pci, pcie->cfg_link_cap_l1sub);
if (!(val & (PCI_L1SS_CAP_ASPM_L1_1 | PCI_L1SS_CAP_ASPM_L1_2)))
@@ -540,13 +565,18 @@ static irqreturn_t tegra_pcie_ep_hard_irq(int irq, void *arg)
static int tegra194_pcie_rd_own_conf(struct pci_bus *bus, u32 devfn, int where,
int size, u32 *val)
{
+ struct pcie_port *pp = bus->sysdata;
+ struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
+ struct tegra194_pcie *pcie = to_tegra_pcie(pci);
+
/*
* This is an endpoint mode specific register happen to appear even
* when controller is operating in root port mode and system hangs
* when it is accessed with link being in ASPM-L1 state.
* So skip accessing it altogether
*/
- if (!PCI_SLOT(devfn) && where == PORT_LOGIC_MSIX_DOORBELL) {
+ if (pcie->of_data->msix_doorbell_access_fixup &&
+ !PCI_SLOT(devfn) && where == PORT_LOGIC_MSIX_DOORBELL) {
*val = 0x00000000;
return PCIBIOS_SUCCESSFUL;
}
@@ -557,13 +587,18 @@ static int tegra194_pcie_rd_own_conf(struct pci_bus *bus, u32 devfn, int where,
static int tegra194_pcie_wr_own_conf(struct pci_bus *bus, u32 devfn, int where,
int size, u32 val)
{
+ struct pcie_port *pp = bus->sysdata;
+ struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
+ struct tegra194_pcie *pcie = to_tegra_pcie(pci);
+
/*
* This is an endpoint mode specific register happen to appear even
* when controller is operating in root port mode and system hangs
* when it is accessed with link being in ASPM-L1 state.
* So skip accessing it altogether
*/
- if (!PCI_SLOT(devfn) && where == PORT_LOGIC_MSIX_DOORBELL)
+ if (pcie->of_data->msix_doorbell_access_fixup &&
+ !PCI_SLOT(devfn) && where == PORT_LOGIC_MSIX_DOORBELL)
return PCIBIOS_SUCCESSFUL;
return pci_generic_config_write(bus, devfn, where, size, val);
@@ -711,13 +746,15 @@ static void tegra_pcie_enable_system_interrupts(struct pcie_port *pp)
val |= APPL_INTR_EN_L0_0_LINK_STATE_INT_EN;
appl_writel(pcie, val, APPL_INTR_EN_L0_0);
- val = appl_readl(pcie, APPL_INTR_EN_L1_0_0);
- val |= APPL_INTR_EN_L1_0_0_LINK_REQ_RST_NOT_INT_EN;
- appl_writel(pcie, val, APPL_INTR_EN_L1_0_0);
+ if (pcie->of_data->sbr_reset_fixup) {
+ val = appl_readl(pcie, APPL_INTR_EN_L1_0_0);
+ val |= APPL_INTR_EN_L1_0_0_LINK_REQ_RST_NOT_INT_EN;
+ appl_writel(pcie, val, APPL_INTR_EN_L1_0_0);
+ }
if (pcie->enable_cdm_check) {
val = appl_readl(pcie, APPL_INTR_EN_L0_0);
- val |= APPL_INTR_EN_L0_0_CDM_REG_CHK_INT_EN;
+ val |= pcie->of_data->cdm_chk_int_en;
appl_writel(pcie, val, APPL_INTR_EN_L0_0);
val = appl_readl(pcie, APPL_INTR_EN_L1_18);
@@ -844,7 +881,8 @@ static void config_gen3_gen4_eq_presets(struct tegra194_pcie *pcie)
val = dw_pcie_readl_dbi(pci, GEN3_EQ_CONTROL_OFF);
val &= ~GEN3_EQ_CONTROL_OFF_PSET_REQ_VEC_MASK;
- val |= (0x360 << GEN3_EQ_CONTROL_OFF_PSET_REQ_VEC_SHIFT);
+ val |= (pcie->of_data->gen4_preset_vec <<
+ GEN3_EQ_CONTROL_OFF_PSET_REQ_VEC_SHIFT);
val &= ~GEN3_EQ_CONTROL_OFF_FB_MODE_MASK;
dw_pcie_writel_dbi(pci, GEN3_EQ_CONTROL_OFF, val);
@@ -858,6 +896,7 @@ static int tegra194_pcie_host_init(struct pcie_port *pp)
struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
struct tegra194_pcie *pcie = to_tegra_pcie(pci);
u32 val;
+ u16 val_16;
pp->bridge->ops = &tegra_pci_ops;
@@ -865,6 +904,11 @@ static int tegra194_pcie_host_init(struct pcie_port *pp)
pcie->pcie_cap_base = dw_pcie_find_capability(&pcie->pci,
PCI_CAP_ID_EXP);
+ val_16 = dw_pcie_readw_dbi(pci, pcie->pcie_cap_base + PCI_EXP_DEVCTL);
+ val_16 &= ~PCI_EXP_DEVCTL_PAYLOAD;
+ val_16 |= PCI_EXP_DEVCTL_PAYLOAD_256B;
+ dw_pcie_writew_dbi(pci, pcie->pcie_cap_base + PCI_EXP_DEVCTL, val_16);
+
val = dw_pcie_readl_dbi(pci, PCI_IO_BASE);
val &= ~(IO_BASE_IO_DECODE | IO_BASE_IO_DECODE_BIT8);
dw_pcie_writel_dbi(pci, PCI_IO_BASE, val);
@@ -889,6 +933,15 @@ static int tegra194_pcie_host_init(struct pcie_port *pp)
val |= (pcie->num_lanes << PCI_EXP_LNKSTA_NLW_SHIFT);
dw_pcie_writel_dbi(pci, pcie->pcie_cap_base + PCI_EXP_LNKCAP, val);
+ /* Clear Slot Clock Configuration bit if SRNS configuration */
+ if (pcie->enable_srns) {
+ val_16 = dw_pcie_readw_dbi(pci, pcie->pcie_cap_base +
+ PCI_EXP_LNKSTA);
+ val_16 &= ~PCI_EXP_LNKSTA_SLC;
+ dw_pcie_writew_dbi(pci, pcie->pcie_cap_base + PCI_EXP_LNKSTA,
+ val_16);
+ }
+
config_gen3_gen4_eq_presets(pcie);
init_host_aspm(pcie);
@@ -899,9 +952,11 @@ static int tegra194_pcie_host_init(struct pcie_port *pp)
disable_aspm_l12(pcie);
}
- val = dw_pcie_readl_dbi(pci, GEN3_RELATED_OFF);
- val &= ~GEN3_RELATED_OFF_GEN3_ZRXDC_NONCOMPL;
- dw_pcie_writel_dbi(pci, GEN3_RELATED_OFF, val);
+ if (pcie->of_data->l1ss_exit_fixup) {
+ val = dw_pcie_readl_dbi(pci, GEN3_RELATED_OFF);
+ val &= ~GEN3_RELATED_OFF_GEN3_ZRXDC_NONCOMPL;
+ dw_pcie_writel_dbi(pci, GEN3_RELATED_OFF, val);
+ }
if (pcie->update_fc_fixup) {
val = dw_pcie_readl_dbi(pci, CFG_TIMER_CTRL_MAX_FUNC_NUM_OFF);
@@ -921,8 +976,11 @@ static int tegra194_pcie_start_link(struct dw_pcie *pci)
struct pcie_port *pp = &pci->pp;
bool retry = true;
- if (pcie->mode == DW_PCIE_EP_TYPE) {
- enable_irq(pcie->pex_rst_irq);
+ if (pcie->of_data->mode == DW_PCIE_EP_TYPE) {
+ if (!pcie->pex_rst_irq_enabled) {
+ enable_irq(pcie->pex_rst_irq);
+ pcie->pex_rst_irq_enabled = true;
+ }
return 0;
}
@@ -1113,13 +1171,26 @@ static int tegra194_pcie_parse_dt(struct tegra194_pcie *pcie)
if (of_property_read_bool(np, "nvidia,update-fc-fixup"))
pcie->update_fc_fixup = true;
+ pcie->enable_ext_refclk =
+ of_property_read_bool(pcie->dev->of_node,
+ "nvidia,enable-ext-refclk");
+ /* RP using an external REFCLK is supported only in Tegra234 */
+ if (pcie->of_data->version == TEGRA194_DWC_IP_VER) {
+ if (pcie->of_data->mode == DW_PCIE_RC_TYPE)
+ pcie->enable_ext_refclk = false;
+ else
+ pcie->enable_ext_refclk = true;
+ }
+
pcie->supports_clkreq =
of_property_read_bool(pcie->dev->of_node, "supports-clkreq");
pcie->enable_cdm_check =
of_property_read_bool(np, "snps,enable-cdm-check");
- if (pcie->mode == DW_PCIE_RC_TYPE)
+ pcie->enable_srns = of_property_read_bool(np, "nvidia,enable-srns");
+
+ if (pcie->of_data->mode == DW_PCIE_RC_TYPE)
return 0;
/* Endpoint mode specific DT entries */
@@ -1163,8 +1234,11 @@ static int tegra_pcie_bpmp_set_ctrl_state(struct tegra194_pcie *pcie,
struct tegra_bpmp_message msg;
struct mrq_uphy_request req;
- /* Controller-5 doesn't need to have its state set by BPMP-FW */
- if (pcie->cid == 5)
+ /*
+ * Controller-5 doesn't need to have its state set by BPMP-FW in
+ * Tegra194
+ */
+ if (pcie->cid == 5 && pcie->of_data->version == 0x490A)
return 0;
memset(&req, 0, sizeof(req));
@@ -1330,6 +1404,14 @@ static int tegra_pcie_config_controller(struct tegra194_pcie *pcie,
return ret;
}
+ if (pcie->enable_ext_refclk) {
+ ret = tegra_pcie_bpmp_set_pll_state(pcie, true);
+ if (ret) {
+ dev_err(pcie->dev, "Failed to init UPHY: %d\n", ret);
+ goto fail_pll_init;
+ }
+ }
+
ret = tegra_pcie_enable_slot_regulators(pcie);
if (ret < 0)
goto fail_slot_reg_en;
@@ -1353,11 +1435,13 @@ static int tegra_pcie_config_controller(struct tegra194_pcie *pcie,
goto fail_core_apb_rst;
}
- if (en_hw_hot_rst) {
+ if (en_hw_hot_rst || !pcie->of_data->sbr_reset_fixup) {
/* Enable HW_HOT_RST mode */
val = appl_readl(pcie, APPL_CTRL);
val &= ~(APPL_CTRL_HW_HOT_RST_MODE_MASK <<
APPL_CTRL_HW_HOT_RST_MODE_SHIFT);
+ val |= (APPL_CTRL_HW_HOT_RST_MODE_IMDT_RST_LTSSM_EN <<
+ APPL_CTRL_HW_HOT_RST_MODE_SHIFT);
val |= APPL_CTRL_HW_HOT_RST_EN;
appl_writel(pcie, val, APPL_CTRL);
}
@@ -1384,6 +1468,19 @@ static int tegra_pcie_config_controller(struct tegra194_pcie *pcie,
val |= (APPL_CFG_MISC_ARCACHE_VAL << APPL_CFG_MISC_ARCACHE_SHIFT);
appl_writel(pcie, val, APPL_CFG_MISC);
+ if (pcie->enable_srns || pcie->enable_ext_refclk) {
+ /*
+ * When Tegra PCIe RP is using external clock, it cannot
+ * supply same clock back to EP, which makes it separate clock.
+ * Gate PCIe RP REFCLK out pads when RP & EP are using separate
+ * clock or RP is using external REFCLK.
+ */
+ val = appl_readl(pcie, APPL_PINMUX);
+ val |= APPL_PINMUX_CLK_OUTPUT_IN_OVERRIDE_EN;
+ val &= ~APPL_PINMUX_CLK_OUTPUT_IN_OVERRIDE;
+ appl_writel(pcie, val, APPL_PINMUX);
+ }
+
if (!pcie->supports_clkreq) {
val = appl_readl(pcie, APPL_PINMUX);
val |= APPL_PINMUX_CLKREQ_OVERRIDE_EN;
@@ -1409,6 +1506,9 @@ static int tegra_pcie_config_controller(struct tegra194_pcie *pcie,
fail_reg_en:
tegra_pcie_disable_slot_regulators(pcie);
fail_slot_reg_en:
+ if (pcie->enable_ext_refclk)
+ tegra_pcie_bpmp_set_pll_state(pcie, false);
+fail_pll_init:
tegra_pcie_bpmp_set_ctrl_state(pcie, false);
return ret;
@@ -1436,6 +1536,12 @@ static void tegra_pcie_unconfig_controller(struct tegra194_pcie *pcie)
tegra_pcie_disable_slot_regulators(pcie);
+ if (pcie->enable_ext_refclk) {
+ ret = tegra_pcie_bpmp_set_pll_state(pcie, false);
+ if (ret)
+ dev_err(pcie->dev, "Failed to deinit UPHY: %d\n", ret);
+ }
+
ret = tegra_pcie_bpmp_set_ctrl_state(pcie, false);
if (ret)
dev_err(pcie->dev, "Failed to disable controller %d: %d\n",
@@ -1636,6 +1742,13 @@ static void pex_ep_event_pex_rst_assert(struct tegra194_pcie *pcie)
pm_runtime_put_sync(pcie->dev);
+ if (pcie->enable_ext_refclk) {
+ ret = tegra_pcie_bpmp_set_pll_state(pcie, false);
+ if (ret)
+ dev_err(pcie->dev, "Failed to turn off UPHY: %d\n",
+ ret);
+ }
+
ret = tegra_pcie_bpmp_set_pll_state(pcie, false);
if (ret)
dev_err(pcie->dev, "Failed to turn off UPHY: %d\n", ret);
@@ -1651,6 +1764,7 @@ static void pex_ep_event_pex_rst_deassert(struct tegra194_pcie *pcie)
struct device *dev = pcie->dev;
u32 val;
int ret;
+ u16 val_16;
if (pcie->ep_state == EP_STATE_ENABLED)
return;
@@ -1662,10 +1776,20 @@ static void pex_ep_event_pex_rst_deassert(struct tegra194_pcie *pcie)
return;
}
- ret = tegra_pcie_bpmp_set_pll_state(pcie, true);
+ ret = tegra_pcie_bpmp_set_ctrl_state(pcie, true);
if (ret) {
- dev_err(dev, "Failed to init UPHY for PCIe EP: %d\n", ret);
- goto fail_pll_init;
+ dev_err(pcie->dev, "Failed to enable controller %u: %d\n",
+ pcie->cid, ret);
+ goto fail_set_ctrl_state;
+ }
+
+ if (pcie->enable_ext_refclk) {
+ ret = tegra_pcie_bpmp_set_pll_state(pcie, true);
+ if (ret) {
+ dev_err(dev, "Failed to init UPHY for PCIe EP: %d\n",
+ ret);
+ goto fail_pll_init;
+ }
}
ret = clk_prepare_enable(pcie->core_clk);
@@ -1762,12 +1886,29 @@ static void pex_ep_event_pex_rst_deassert(struct tegra194_pcie *pcie)
disable_aspm_l12(pcie);
}
- val = dw_pcie_readl_dbi(pci, GEN3_RELATED_OFF);
- val &= ~GEN3_RELATED_OFF_GEN3_ZRXDC_NONCOMPL;
- dw_pcie_writel_dbi(pci, GEN3_RELATED_OFF, val);
+ if (pcie->of_data->l1ss_exit_fixup) {
+ val = dw_pcie_readl_dbi(pci, GEN3_RELATED_OFF);
+ val &= ~GEN3_RELATED_OFF_GEN3_ZRXDC_NONCOMPL;
+ dw_pcie_writel_dbi(pci, GEN3_RELATED_OFF, val);
+ }
pcie->pcie_cap_base = dw_pcie_find_capability(&pcie->pci,
PCI_CAP_ID_EXP);
+
+ val_16 = dw_pcie_readw_dbi(pci, pcie->pcie_cap_base + PCI_EXP_DEVCTL);
+ val_16 &= ~PCI_EXP_DEVCTL_PAYLOAD;
+ val_16 |= PCI_EXP_DEVCTL_PAYLOAD_256B;
+ dw_pcie_writew_dbi(pci, pcie->pcie_cap_base + PCI_EXP_DEVCTL, val_16);
+
+ /* Clear Slot Clock Configuration bit if SRNS configuration */
+ if (pcie->enable_srns) {
+ val_16 = dw_pcie_readw_dbi(pci, pcie->pcie_cap_base +
+ PCI_EXP_LNKSTA);
+ val_16 &= ~PCI_EXP_LNKSTA_SLC;
+ dw_pcie_writew_dbi(pci, pcie->pcie_cap_base + PCI_EXP_LNKSTA,
+ val_16);
+ }
+
clk_set_rate(pcie->core_clk, GEN4_CORE_CLK_FREQ);
val = (ep->msi_mem_phys & MSIX_ADDR_MATCH_LOW_OFF_MASK);
@@ -1784,6 +1925,13 @@ static void pex_ep_event_pex_rst_deassert(struct tegra194_pcie *pcie)
dw_pcie_ep_init_notify(ep);
+ /* Send LTR upstream */
+ if (!pcie->of_data->ltr_req_fixup) {
+ val = appl_readl(pcie, APPL_LTR_MSG_2);
+ val |= APPL_LTR_MSG_2_LTR_MSG_REQ_STATE;
+ appl_writel(pcie, val, APPL_LTR_MSG_2);
+ }
+
/* Enable LTSSM */
val = appl_readl(pcie, APPL_CTRL);
val |= APPL_CTRL_LTSSM_EN;
@@ -1804,6 +1952,8 @@ static void pex_ep_event_pex_rst_deassert(struct tegra194_pcie *pcie)
fail_core_clk_enable:
tegra_pcie_bpmp_set_pll_state(pcie, false);
fail_pll_init:
+ tegra_pcie_bpmp_set_ctrl_state(pcie, false);
+fail_set_ctrl_state:
pm_runtime_put_sync(dev);
}
@@ -1933,6 +2083,7 @@ static int tegra_pcie_config_ep(struct tegra194_pcie *pcie,
irq_set_status_flags(pcie->pex_rst_irq, IRQ_NOAUTOEN);
+ pcie->pex_rst_irq_enabled = false;
pcie->ep_state = EP_STATE_DISABLED;
ret = devm_request_threaded_irq(dev, pcie->pex_rst_irq, NULL,
@@ -1979,14 +2130,13 @@ static int tegra194_pcie_probe(struct platform_device *pdev)
pci = &pcie->pci;
pci->dev = &pdev->dev;
pci->ops = &tegra_dw_pcie_ops;
- pci->n_fts[0] = N_FTS_VAL;
- pci->n_fts[1] = FTS_VAL;
- pci->version = 0x490A;
-
+ pcie->dev = &pdev->dev;
+ pcie->of_data = (struct tegra_pcie_of_data *)data;
+ pci->n_fts[0] = pcie->of_data->n_fts[0];
+ pci->n_fts[1] = pcie->of_data->n_fts[1];
+ pci->version = pcie->of_data->version;
pp = &pci->pp;
pp->num_vectors = MAX_MSI_IRQS;
- pcie->dev = &pdev->dev;
- pcie->mode = (enum dw_pcie_device_mode)data->mode;
ret = tegra194_pcie_parse_dt(pcie);
if (ret < 0) {
@@ -2103,7 +2253,7 @@ static int tegra194_pcie_probe(struct platform_device *pdev)
platform_set_drvdata(pdev, pcie);
- switch (pcie->mode) {
+ switch (pcie->of_data->mode) {
case DW_PCIE_RC_TYPE:
ret = devm_request_irq(dev, pp->irq, tegra_pcie_rp_irq_handler,
IRQF_SHARED, "tegra-pcie-intr", pcie);
@@ -2138,7 +2288,8 @@ static int tegra194_pcie_probe(struct platform_device *pdev)
break;
default:
- dev_err(dev, "Invalid PCIe device type %d\n", pcie->mode);
+ dev_err(dev, "Invalid PCIe device type %d\n",
+ pcie->of_data->mode);
}
fail:
@@ -2150,12 +2301,19 @@ static int tegra194_pcie_remove(struct platform_device *pdev)
{
struct tegra194_pcie *pcie = platform_get_drvdata(pdev);
- if (!pcie->link_state)
- return 0;
+ if (pcie->of_data->mode == DW_PCIE_RC_TYPE) {
+ if (!pcie->link_state)
+ return 0;
+
+ debugfs_remove_recursive(pcie->debugfs);
+ tegra_pcie_deinit_controller(pcie);
+ pm_runtime_put_sync(pcie->dev);
+ } else {
+ if (pcie->pex_rst_irq_enabled)
+ disable_irq(pcie->pex_rst_irq);
+ pex_ep_event_pex_rst_assert(pcie);
+ }
- debugfs_remove_recursive(pcie->debugfs);
- tegra_pcie_deinit_controller(pcie);
- pm_runtime_put_sync(pcie->dev);
pm_runtime_disable(pcie->dev);
tegra_bpmp_put(pcie->bpmp);
if (pcie->pex_refclk_sel_gpiod)
@@ -2169,15 +2327,22 @@ static int tegra194_pcie_suspend_late(struct device *dev)
struct tegra194_pcie *pcie = dev_get_drvdata(dev);
u32 val;
+ if (pcie->of_data->mode == DW_PCIE_EP_TYPE) {
+ dev_err(dev, "Failed to Suspend as Tegra PCIe is in EP mode\n");
+ return -EPERM;
+ }
+
if (!pcie->link_state)
return 0;
/* Enable HW_HOT_RST mode */
- val = appl_readl(pcie, APPL_CTRL);
- val &= ~(APPL_CTRL_HW_HOT_RST_MODE_MASK <<
- APPL_CTRL_HW_HOT_RST_MODE_SHIFT);
- val |= APPL_CTRL_HW_HOT_RST_EN;
- appl_writel(pcie, val, APPL_CTRL);
+ if (pcie->of_data->sbr_reset_fixup) {
+ val = appl_readl(pcie, APPL_CTRL);
+ val &= ~(APPL_CTRL_HW_HOT_RST_MODE_MASK <<
+ APPL_CTRL_HW_HOT_RST_MODE_SHIFT);
+ val |= APPL_CTRL_HW_HOT_RST_EN;
+ appl_writel(pcie, val, APPL_CTRL);
+ }
return 0;
}
@@ -2239,22 +2404,19 @@ static int tegra194_pcie_resume_early(struct device *dev)
struct tegra194_pcie *pcie = dev_get_drvdata(dev);
u32 val;
- if (pcie->mode == DW_PCIE_EP_TYPE) {
- dev_err(dev, "Suspend is not supported in EP mode");
- return -ENOTSUPP;
- }
-
if (!pcie->link_state)
return 0;
/* Disable HW_HOT_RST mode */
- val = appl_readl(pcie, APPL_CTRL);
- val &= ~(APPL_CTRL_HW_HOT_RST_MODE_MASK <<
- APPL_CTRL_HW_HOT_RST_MODE_SHIFT);
- val |= APPL_CTRL_HW_HOT_RST_MODE_IMDT_RST <<
- APPL_CTRL_HW_HOT_RST_MODE_SHIFT;
- val &= ~APPL_CTRL_HW_HOT_RST_EN;
- appl_writel(pcie, val, APPL_CTRL);
+ if (pcie->of_data->sbr_reset_fixup) {
+ val = appl_readl(pcie, APPL_CTRL);
+ val &= ~(APPL_CTRL_HW_HOT_RST_MODE_MASK <<
+ APPL_CTRL_HW_HOT_RST_MODE_SHIFT);
+ val |= APPL_CTRL_HW_HOT_RST_MODE_IMDT_RST <<
+ APPL_CTRL_HW_HOT_RST_MODE_SHIFT;
+ val &= ~APPL_CTRL_HW_HOT_RST_EN;
+ appl_writel(pcie, val, APPL_CTRL);
+ }
return 0;
}
@@ -2263,26 +2425,77 @@ static void tegra194_pcie_shutdown(struct platform_device *pdev)
{
struct tegra194_pcie *pcie = platform_get_drvdata(pdev);
- if (!pcie->link_state)
- return;
+ if (pcie->of_data->mode == DW_PCIE_RC_TYPE) {
+ if (!pcie->link_state)
+ return;
- debugfs_remove_recursive(pcie->debugfs);
- tegra_pcie_downstream_dev_to_D0(pcie);
+ debugfs_remove_recursive(pcie->debugfs);
+ tegra_pcie_downstream_dev_to_D0(pcie);
- disable_irq(pcie->pci.pp.irq);
- if (IS_ENABLED(CONFIG_PCI_MSI))
- disable_irq(pcie->pci.pp.msi_irq);
+ disable_irq(pcie->pci.pp.irq);
+ if (IS_ENABLED(CONFIG_PCI_MSI))
+ disable_irq(pcie->pci.pp.msi_irq);
- tegra194_pcie_pme_turnoff(pcie);
- tegra_pcie_unconfig_controller(pcie);
+ tegra194_pcie_pme_turnoff(pcie);
+ tegra_pcie_unconfig_controller(pcie);
+ pm_runtime_put_sync(pcie->dev);
+ } else {
+ if (pcie->pex_rst_irq_enabled)
+ disable_irq(pcie->pex_rst_irq);
+ pex_ep_event_pex_rst_assert(pcie);
+ }
}
-static const struct tegra194_pcie_of_data tegra194_pcie_rc_of_data = {
+static const struct tegra_pcie_of_data tegra194_pcie_rc_of_data = {
+ .version = TEGRA194_DWC_IP_VER,
+ .mode = DW_PCIE_RC_TYPE,
+ .msix_doorbell_access_fixup = true,
+ .sbr_reset_fixup = true,
+ .l1ss_exit_fixup = true,
+ .ltr_req_fixup = false,
+ .cdm_chk_int_en = BIT(19),
+ /* Gen4 - 5, 6, 8 and 9 presets enabled */
+ .gen4_preset_vec = 0x360,
+ .n_fts = { 52, 52 },
+};
+
+static const struct tegra_pcie_of_data tegra194_pcie_ep_of_data = {
+ .version = TEGRA194_DWC_IP_VER,
+ .mode = DW_PCIE_EP_TYPE,
+ .msix_doorbell_access_fixup = false,
+ .sbr_reset_fixup = false,
+ .l1ss_exit_fixup = true,
+ .ltr_req_fixup = true,
+ .cdm_chk_int_en = BIT(19),
+ /* Gen4 - 5, 6, 8 and 9 presets enabled */
+ .gen4_preset_vec = 0x360,
+ .n_fts = { 52, 52 },
+};
+
+static const struct tegra_pcie_of_data tegra234_pcie_rc_of_data = {
+ .version = TEGRA234_DWC_IP_VER,
.mode = DW_PCIE_RC_TYPE,
+ .msix_doorbell_access_fixup = false,
+ .sbr_reset_fixup = false,
+ .l1ss_exit_fixup = false,
+ .ltr_req_fixup = false,
+ .cdm_chk_int_en = BIT(18),
+ /* Gen4 - 6, 8 and 9 presets enabled */
+ .gen4_preset_vec = 0x340,
+ .n_fts = { 52, 80 },
};
-static const struct tegra194_pcie_of_data tegra194_pcie_ep_of_data = {
+static const struct tegra_pcie_of_data tegra234_pcie_ep_of_data = {
+ .version = TEGRA234_DWC_IP_VER,
.mode = DW_PCIE_EP_TYPE,
+ .msix_doorbell_access_fixup = false,
+ .sbr_reset_fixup = false,
+ .l1ss_exit_fixup = false,
+ .ltr_req_fixup = false,
+ .cdm_chk_int_en = BIT(18),
+ /* Gen4 - 6, 8 and 9 presets enabled */
+ .gen4_preset_vec = 0x340,
+ .n_fts = { 52, 80 },
};
static const struct of_device_id tegra194_pcie_of_match[] = {
@@ -2294,6 +2507,14 @@ static const struct of_device_id tegra194_pcie_of_match[] = {
.compatible = "nvidia,tegra194-pcie-ep",
.data = &tegra194_pcie_ep_of_data,
},
+ {
+ .compatible = "nvidia,tegra234-pcie",
+ .data = &tegra234_pcie_rc_of_data,
+ },
+ {
+ .compatible = "nvidia,tegra234-pcie-ep",
+ .data = &tegra234_pcie_ep_of_data,
+ },
{},
};
--
2.17.1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply related [flat|nested] 46+ messages in thread
* Re: [PATCH V1 07/10] arm64: tegra: Enable PCIe slots in P3737-0000 board
2022-02-05 16:21 ` Vidya Sagar
@ 2022-02-06 11:29 ` Krzysztof Kozlowski
-1 siblings, 0 replies; 46+ messages in thread
From: Krzysztof Kozlowski @ 2022-02-06 11:29 UTC (permalink / raw)
To: Vidya Sagar, bhelgaas, lorenzo.pieralisi, robh+dt,
thierry.reding, jonathanh
Cc: kishon, vkoul, kw, p.zabel, mperttunen, linux-pci, devicetree,
linux-tegra, linux-kernel, linux-phy, kthota, mmaddireddy,
sagar.tv
On 05/02/2022 17:21, Vidya Sagar wrote:
> Enable PCIe controller nodes to enable respective PCIe slots on
> P3737-0000 board. Following is the ownership of slots by different
> PCIe controllers.
> Controller-1 : On-board Broadcom WiFi controller
> Controller-4 : M.2 Key-M slot
> Controller-5 : CEM form-factor x8 slot
>
> Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
> ---
> .../nvidia/tegra234-p3737-0000+p3701-0000.dts | 26 +++++++++++++++++++
> 1 file changed, 26 insertions(+)
>
> diff --git a/arch/arm64/boot/dts/nvidia/tegra234-p3737-0000+p3701-0000.dts b/arch/arm64/boot/dts/nvidia/tegra234-p3737-0000+p3701-0000.dts
> index efbbb878ba5a..b819e1133bc4 100644
> --- a/arch/arm64/boot/dts/nvidia/tegra234-p3737-0000+p3701-0000.dts
> +++ b/arch/arm64/boot/dts/nvidia/tegra234-p3737-0000+p3701-0000.dts
> @@ -21,4 +21,30 @@
> serial {
> status = "okay";
> };
> +
> + pcie@14100000 {
> + status = "okay";
> +
> + phys = <&p2u_hsio_3>;
> + phy-names = "p2u-0";
> + };
> +
> + pcie@14160000 {
> + status = "okay";
> +
> + phys = <&p2u_hsio_4>, <&p2u_hsio_5>, <&p2u_hsio_6>,
> + <&p2u_hsio_7>;
> + phy-names = "p2u-0", "p2u-1", "p2u-2", "p2u-3";
> + };
> +
> + pcie@141a0000 {
> + status = "okay";
> +
> + phys = <&p2u_nvhs_0>, <&p2u_nvhs_1>, <&p2u_nvhs_2>,
> + <&p2u_nvhs_3>, <&p2u_nvhs_4>, <&p2u_nvhs_5>,
> + <&p2u_nvhs_6>, <&p2u_nvhs_7>;
> + phy-names = "p2u-0", "p2u-1", "p2u-2", "p2u-3", "p2u-4",
> + "p2u-5", "p2u-6", "p2u-7";
> + };
> +
No need for trailing new line.
> };
Best regards,
Krzysztof
^ permalink raw reply [flat|nested] 46+ messages in thread
* Re: [PATCH V1 07/10] arm64: tegra: Enable PCIe slots in P3737-0000 board
@ 2022-02-06 11:29 ` Krzysztof Kozlowski
0 siblings, 0 replies; 46+ messages in thread
From: Krzysztof Kozlowski @ 2022-02-06 11:29 UTC (permalink / raw)
To: Vidya Sagar, bhelgaas, lorenzo.pieralisi, robh+dt,
thierry.reding, jonathanh
Cc: kishon, vkoul, kw, p.zabel, mperttunen, linux-pci, devicetree,
linux-tegra, linux-kernel, linux-phy, kthota, mmaddireddy,
sagar.tv
On 05/02/2022 17:21, Vidya Sagar wrote:
> Enable PCIe controller nodes to enable respective PCIe slots on
> P3737-0000 board. Following is the ownership of slots by different
> PCIe controllers.
> Controller-1 : On-board Broadcom WiFi controller
> Controller-4 : M.2 Key-M slot
> Controller-5 : CEM form-factor x8 slot
>
> Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
> ---
> .../nvidia/tegra234-p3737-0000+p3701-0000.dts | 26 +++++++++++++++++++
> 1 file changed, 26 insertions(+)
>
> diff --git a/arch/arm64/boot/dts/nvidia/tegra234-p3737-0000+p3701-0000.dts b/arch/arm64/boot/dts/nvidia/tegra234-p3737-0000+p3701-0000.dts
> index efbbb878ba5a..b819e1133bc4 100644
> --- a/arch/arm64/boot/dts/nvidia/tegra234-p3737-0000+p3701-0000.dts
> +++ b/arch/arm64/boot/dts/nvidia/tegra234-p3737-0000+p3701-0000.dts
> @@ -21,4 +21,30 @@
> serial {
> status = "okay";
> };
> +
> + pcie@14100000 {
> + status = "okay";
> +
> + phys = <&p2u_hsio_3>;
> + phy-names = "p2u-0";
> + };
> +
> + pcie@14160000 {
> + status = "okay";
> +
> + phys = <&p2u_hsio_4>, <&p2u_hsio_5>, <&p2u_hsio_6>,
> + <&p2u_hsio_7>;
> + phy-names = "p2u-0", "p2u-1", "p2u-2", "p2u-3";
> + };
> +
> + pcie@141a0000 {
> + status = "okay";
> +
> + phys = <&p2u_nvhs_0>, <&p2u_nvhs_1>, <&p2u_nvhs_2>,
> + <&p2u_nvhs_3>, <&p2u_nvhs_4>, <&p2u_nvhs_5>,
> + <&p2u_nvhs_6>, <&p2u_nvhs_7>;
> + phy-names = "p2u-0", "p2u-1", "p2u-2", "p2u-3", "p2u-4",
> + "p2u-5", "p2u-6", "p2u-7";
> + };
> +
No need for trailing new line.
> };
Best regards,
Krzysztof
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply [flat|nested] 46+ messages in thread
* Re: [PATCH V1 03/10] dt-bindings: memory: Add Tegra234 PCIe memory
2022-02-05 16:21 ` Vidya Sagar
@ 2022-02-06 11:33 ` Krzysztof Kozlowski
-1 siblings, 0 replies; 46+ messages in thread
From: Krzysztof Kozlowski @ 2022-02-06 11:33 UTC (permalink / raw)
To: Vidya Sagar, bhelgaas, lorenzo.pieralisi, robh+dt,
thierry.reding, jonathanh
Cc: kishon, vkoul, kw, p.zabel, mperttunen, linux-pci, devicetree,
linux-tegra, linux-kernel, linux-phy, kthota, mmaddireddy,
sagar.tv
On 05/02/2022 17:21, Vidya Sagar wrote:
> Add the memory client and stream ID definitions for the PCIe hardware
> found on Tegra234 SoCs.
I could not find dependencies or merging strategy in cover letter.
Please always describe it, so I don't have to go through all the patches
to figure this out.
>
> Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
> ---
> include/dt-bindings/memory/tegra234-mc.h | 64 ++++++++++++++++++++++++
> 1 file changed, 64 insertions(+)
>
> diff --git a/include/dt-bindings/memory/tegra234-mc.h b/include/dt-bindings/memory/tegra234-mc.h
> index 2662f70c15c6..60017684858a 100644
> --- a/include/dt-bindings/memory/tegra234-mc.h
> +++ b/include/dt-bindings/memory/tegra234-mc.h
> @@ -7,15 +7,53 @@
> #define TEGRA234_SID_INVALID 0x00
> #define TEGRA234_SID_PASSTHROUGH 0x7f
>
> +/* NISO0 stream IDs */
> +#define TEGRA234_SID_PCIE0 0x12U
> +#define TEGRA234_SID_PCIE4 0x13U
> +#define TEGRA234_SID_PCIE5 0x14U
> +#define TEGRA234_SID_PCIE6 0x15U
> +#define TEGRA234_SID_PCIE9 0x1FU
>
> /* NISO1 stream IDs */
> #define TEGRA234_SID_SDMMC4 0x02
> +#define TEGRA234_SID_PCIE1 0x5U
> +#define TEGRA234_SID_PCIE2 0x6U
> +#define TEGRA234_SID_PCIE3 0x7U
> +#define TEGRA234_SID_PCIE7 0x8U
> +#define TEGRA234_SID_PCIE8 0x9U
> +#define TEGRA234_SID_PCIE10 0xBU
I don't see usage of these...
> #define TEGRA234_SID_BPMP 0x10
>
> /*
> * memory client IDs
> */
>
> +/* PCIE6 read clients */
> +#define TEGRA234_MEMORY_CLIENT_PCIE6AR 0x28
I see you use them in DTS but not in mc driver. Don't you miss anything
here?
> +/* PCIE6 write clients */
> +#define TEGRA234_MEMORY_CLIENT_PCIE6AW 0x29
> +/* PCIE7 read clients */
> +#define TEGRA234_MEMORY_
Best regards,
Krzysztof
^ permalink raw reply [flat|nested] 46+ messages in thread
* Re: [PATCH V1 03/10] dt-bindings: memory: Add Tegra234 PCIe memory
@ 2022-02-06 11:33 ` Krzysztof Kozlowski
0 siblings, 0 replies; 46+ messages in thread
From: Krzysztof Kozlowski @ 2022-02-06 11:33 UTC (permalink / raw)
To: Vidya Sagar, bhelgaas, lorenzo.pieralisi, robh+dt,
thierry.reding, jonathanh
Cc: kishon, vkoul, kw, p.zabel, mperttunen, linux-pci, devicetree,
linux-tegra, linux-kernel, linux-phy, kthota, mmaddireddy,
sagar.tv
On 05/02/2022 17:21, Vidya Sagar wrote:
> Add the memory client and stream ID definitions for the PCIe hardware
> found on Tegra234 SoCs.
I could not find dependencies or merging strategy in cover letter.
Please always describe it, so I don't have to go through all the patches
to figure this out.
>
> Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
> ---
> include/dt-bindings/memory/tegra234-mc.h | 64 ++++++++++++++++++++++++
> 1 file changed, 64 insertions(+)
>
> diff --git a/include/dt-bindings/memory/tegra234-mc.h b/include/dt-bindings/memory/tegra234-mc.h
> index 2662f70c15c6..60017684858a 100644
> --- a/include/dt-bindings/memory/tegra234-mc.h
> +++ b/include/dt-bindings/memory/tegra234-mc.h
> @@ -7,15 +7,53 @@
> #define TEGRA234_SID_INVALID 0x00
> #define TEGRA234_SID_PASSTHROUGH 0x7f
>
> +/* NISO0 stream IDs */
> +#define TEGRA234_SID_PCIE0 0x12U
> +#define TEGRA234_SID_PCIE4 0x13U
> +#define TEGRA234_SID_PCIE5 0x14U
> +#define TEGRA234_SID_PCIE6 0x15U
> +#define TEGRA234_SID_PCIE9 0x1FU
>
> /* NISO1 stream IDs */
> #define TEGRA234_SID_SDMMC4 0x02
> +#define TEGRA234_SID_PCIE1 0x5U
> +#define TEGRA234_SID_PCIE2 0x6U
> +#define TEGRA234_SID_PCIE3 0x7U
> +#define TEGRA234_SID_PCIE7 0x8U
> +#define TEGRA234_SID_PCIE8 0x9U
> +#define TEGRA234_SID_PCIE10 0xBU
I don't see usage of these...
> #define TEGRA234_SID_BPMP 0x10
>
> /*
> * memory client IDs
> */
>
> +/* PCIE6 read clients */
> +#define TEGRA234_MEMORY_CLIENT_PCIE6AR 0x28
I see you use them in DTS but not in mc driver. Don't you miss anything
here?
> +/* PCIE6 write clients */
> +#define TEGRA234_MEMORY_CLIENT_PCIE6AW 0x29
> +/* PCIE7 read clients */
> +#define TEGRA234_MEMORY_
Best regards,
Krzysztof
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply [flat|nested] 46+ messages in thread
* Re: [PATCH V1 04/10] dt-bindings: PHY: P2U: Add support for Tegra234 P2U block
2022-02-05 16:21 ` Vidya Sagar
@ 2022-02-07 6:47 ` Raul Tambre
-1 siblings, 0 replies; 46+ messages in thread
From: Raul Tambre @ 2022-02-07 6:47 UTC (permalink / raw)
To: Vidya Sagar, bhelgaas, lorenzo.pieralisi, robh+dt,
thierry.reding, jonathanh
Cc: kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, sagar.tv
On 2022-02-05 18:21, Vidya Sagar wrote:
> Subject:
> [PATCH V1 04/10] dt-bindings: PHY: P2U: Add support for Tegra234 P2U block
> From:
> Vidya Sagar <vidyas@nvidia.com>
> Date:
> 2022-02-05, 18:21
>
> To:
> <bhelgaas@google.com>, <lorenzo.pieralisi@arm.com>,
> <robh+dt@kernel.org>, <thierry.reding@gmail.com>, <jonathanh@nvidia.com>
> CC:
> <kishon@ti.com>, <vkoul@kernel.org>, <kw@linux.com>,
> <krzysztof.kozlowski@canonical.com>, <p.zabel@pengutronix.de>,
> <mperttunen@nvidia.com>, <linux-pci@vger.kernel.org>,
> <devicetree@vger.kernel.org>, <linux-tegra@vger.kernel.org>,
> <linux-kernel@vger.kernel.org>, <linux-phy@lists.infradead.org>,
> <kthota@nvidia.com>, <mmaddireddy@nvidia.com>, <vidyas@nvidia.com>,
> <sagar.tv@gmail.com>
>
>
> Add support for Tegra234 P2U (PIPE to UPHY) module block which is a glue
> module instantiated once for each PCIe lane between Synopsys DesignWare
> core based PCIe IP and Universal PHY block.
>
> Signed-off-by: Vidya Sagar<vidyas@nvidia.com>
> ---
> .../bindings/phy/phy-tegra194-p2u.yaml | 17 +++++++++++++----
> 1 file changed, 13 insertions(+), 4 deletions(-)
>
> diff --git a/Documentation/devicetree/bindings/phy/phy-tegra194-p2u.yaml b/Documentation/devicetree/bindings/phy/phy-tegra194-p2u.yaml
> index 9a89d05efbda..6ba1f69b1126 100644
> --- a/Documentation/devicetree/bindings/phy/phy-tegra194-p2u.yaml
> +++ b/Documentation/devicetree/bindings/phy/phy-tegra194-p2u.yaml
> @@ -4,7 +4,7 @@
> $id:"http://devicetree.org/schemas/phy/phy-tegra194-p2u.yaml#"
> $schema:"http://devicetree.org/meta-schemas/core.yaml#"
>
> -title: NVIDIA Tegra194 P2U binding
> +title: NVIDIA Tegra194 & Tegra234 P2U binding
>
> maintainers:
> - Thierry Reding<treding@nvidia.com>
> @@ -12,13 +12,17 @@ maintainers:
> description: >
> Tegra194 has two PHY bricks namely HSIO (High Speed IO) and NVHS (NVIDIA High
> Speed) each interfacing with 12 and 8 P2U instances respectively.
> + Tegra234 has three PHY bricks namesly HSIO, NVHS and GBE (Gigabit Ethernet)
typo: namely
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply [flat|nested] 46+ messages in thread
* Re: [PATCH V1 04/10] dt-bindings: PHY: P2U: Add support for Tegra234 P2U block
@ 2022-02-07 6:47 ` Raul Tambre
0 siblings, 0 replies; 46+ messages in thread
From: Raul Tambre @ 2022-02-07 6:47 UTC (permalink / raw)
To: Vidya Sagar, bhelgaas, lorenzo.pieralisi, robh+dt,
thierry.reding, jonathanh
Cc: kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, sagar.tv
On 2022-02-05 18:21, Vidya Sagar wrote:
> Subject:
> [PATCH V1 04/10] dt-bindings: PHY: P2U: Add support for Tegra234 P2U block
> From:
> Vidya Sagar <vidyas@nvidia.com>
> Date:
> 2022-02-05, 18:21
>
> To:
> <bhelgaas@google.com>, <lorenzo.pieralisi@arm.com>,
> <robh+dt@kernel.org>, <thierry.reding@gmail.com>, <jonathanh@nvidia.com>
> CC:
> <kishon@ti.com>, <vkoul@kernel.org>, <kw@linux.com>,
> <krzysztof.kozlowski@canonical.com>, <p.zabel@pengutronix.de>,
> <mperttunen@nvidia.com>, <linux-pci@vger.kernel.org>,
> <devicetree@vger.kernel.org>, <linux-tegra@vger.kernel.org>,
> <linux-kernel@vger.kernel.org>, <linux-phy@lists.infradead.org>,
> <kthota@nvidia.com>, <mmaddireddy@nvidia.com>, <vidyas@nvidia.com>,
> <sagar.tv@gmail.com>
>
>
> Add support for Tegra234 P2U (PIPE to UPHY) module block which is a glue
> module instantiated once for each PCIe lane between Synopsys DesignWare
> core based PCIe IP and Universal PHY block.
>
> Signed-off-by: Vidya Sagar<vidyas@nvidia.com>
> ---
> .../bindings/phy/phy-tegra194-p2u.yaml | 17 +++++++++++++----
> 1 file changed, 13 insertions(+), 4 deletions(-)
>
> diff --git a/Documentation/devicetree/bindings/phy/phy-tegra194-p2u.yaml b/Documentation/devicetree/bindings/phy/phy-tegra194-p2u.yaml
> index 9a89d05efbda..6ba1f69b1126 100644
> --- a/Documentation/devicetree/bindings/phy/phy-tegra194-p2u.yaml
> +++ b/Documentation/devicetree/bindings/phy/phy-tegra194-p2u.yaml
> @@ -4,7 +4,7 @@
> $id:"http://devicetree.org/schemas/phy/phy-tegra194-p2u.yaml#"
> $schema:"http://devicetree.org/meta-schemas/core.yaml#"
>
> -title: NVIDIA Tegra194 P2U binding
> +title: NVIDIA Tegra194 & Tegra234 P2U binding
>
> maintainers:
> - Thierry Reding<treding@nvidia.com>
> @@ -12,13 +12,17 @@ maintainers:
> description: >
> Tegra194 has two PHY bricks namely HSIO (High Speed IO) and NVHS (NVIDIA High
> Speed) each interfacing with 12 and 8 P2U instances respectively.
> + Tegra234 has three PHY bricks namesly HSIO, NVHS and GBE (Gigabit Ethernet)
typo: namely
^ permalink raw reply [flat|nested] 46+ messages in thread
* Re: [PATCH V1 09/10] PCI: Disable MSI for Tegra234 root ports
2022-02-05 16:21 ` Vidya Sagar
@ 2022-02-07 17:36 ` Bjorn Helgaas
-1 siblings, 0 replies; 46+ messages in thread
From: Bjorn Helgaas @ 2022-02-07 17:36 UTC (permalink / raw)
To: Vidya Sagar
Cc: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh,
kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, sagar.tv
On Sat, Feb 05, 2022 at 09:51:43PM +0530, Vidya Sagar wrote:
> Tegra234 PCIe rootports don't generate MSI interrupts for PME and AER
> events. Since PCIe spec (Ref: r4.0 sec 7.7.1.2 and 7.7.2.2) doesn't support
> using a mix of INTx and MSI/MSI-X, MSI needs to be disabled to avoid root
> ports service drivers registering their respective ISRs with MSI interrupt
> and to let only INTx be used for all events.
s/rootports/root ports/ to match other usage here.
This argument matches that in 8c7e96d3fe75 ("PCI: Disable MSI for
Tegra root ports") [1], but that's not quite what sec 7.7.1.2 and
7.7.2.2 say. Those sections talk about what happens when both MSI and
MSI-X are disabled:
If MSI and MSI-X are both disabled, the Function requests servicing
using INTx interrupts (if supported).
but they don't say anything about what happens when MSI or MSI-X is
*enabled*.
I think a better citation is PCIe r6.0, sec 6.1.4.3, which says:
While enabled for MSI or MSI-X operation, a Function is prohibited
from using INTx interrupts (if implemented) to request service (MSI,
MSI-X, and INTx are mutually exclusive).
Can you please update the comment in the code and this commit log to
cite PCIe r6.0, sec 6.1.4.3 instead, and to clarify that these Tegra
devices always use INTx for PME and AER, even when MSI/MSI-X is
enabled?
Why do these Tegra quirks use DECLARE_PCI_FIXUP_CLASS_EARLY() instead
of just DECLARE_PCI_FIXUP_EARLY()? quirk_al_msi_disable() uses the
_CLASS version because the same Device ID is used for non-Root Port
devices. Is the same true here, or could these use
DECLARE_PCI_FIXUP_EARLY()?
There are many quirks that disable MSI, and they're a mixture of EARLY
and FINAL. They should probably all be the same.
[1] https://git.kernel.org/linus/8c7e96d3fe75
> Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
> ---
> drivers/pci/quirks.c | 9 +++++++++
> 1 file changed, 9 insertions(+)
>
> diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
> index d2dd6a6cda60..3ac5c45e61a1 100644
> --- a/drivers/pci/quirks.c
> +++ b/drivers/pci/quirks.c
> @@ -2747,6 +2747,15 @@ DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x10e5,
> DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x10e6,
> PCI_CLASS_BRIDGE_PCI, 8,
> pci_quirk_nvidia_tegra_disable_rp_msi);
> +DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x229a,
> + PCI_CLASS_BRIDGE_PCI, 8,
> + pci_quirk_nvidia_tegra_disable_rp_msi);
> +DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x229c,
> + PCI_CLASS_BRIDGE_PCI, 8,
> + pci_quirk_nvidia_tegra_disable_rp_msi);
> +DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x229e,
> + PCI_CLASS_BRIDGE_PCI, 8,
> + pci_quirk_nvidia_tegra_disable_rp_msi);
>
> /*
> * Some versions of the MCP55 bridge from Nvidia have a legacy IRQ routing
> --
> 2.17.1
>
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply [flat|nested] 46+ messages in thread
* Re: [PATCH V1 09/10] PCI: Disable MSI for Tegra234 root ports
@ 2022-02-07 17:36 ` Bjorn Helgaas
0 siblings, 0 replies; 46+ messages in thread
From: Bjorn Helgaas @ 2022-02-07 17:36 UTC (permalink / raw)
To: Vidya Sagar
Cc: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh,
kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, sagar.tv
On Sat, Feb 05, 2022 at 09:51:43PM +0530, Vidya Sagar wrote:
> Tegra234 PCIe rootports don't generate MSI interrupts for PME and AER
> events. Since PCIe spec (Ref: r4.0 sec 7.7.1.2 and 7.7.2.2) doesn't support
> using a mix of INTx and MSI/MSI-X, MSI needs to be disabled to avoid root
> ports service drivers registering their respective ISRs with MSI interrupt
> and to let only INTx be used for all events.
s/rootports/root ports/ to match other usage here.
This argument matches that in 8c7e96d3fe75 ("PCI: Disable MSI for
Tegra root ports") [1], but that's not quite what sec 7.7.1.2 and
7.7.2.2 say. Those sections talk about what happens when both MSI and
MSI-X are disabled:
If MSI and MSI-X are both disabled, the Function requests servicing
using INTx interrupts (if supported).
but they don't say anything about what happens when MSI or MSI-X is
*enabled*.
I think a better citation is PCIe r6.0, sec 6.1.4.3, which says:
While enabled for MSI or MSI-X operation, a Function is prohibited
from using INTx interrupts (if implemented) to request service (MSI,
MSI-X, and INTx are mutually exclusive).
Can you please update the comment in the code and this commit log to
cite PCIe r6.0, sec 6.1.4.3 instead, and to clarify that these Tegra
devices always use INTx for PME and AER, even when MSI/MSI-X is
enabled?
Why do these Tegra quirks use DECLARE_PCI_FIXUP_CLASS_EARLY() instead
of just DECLARE_PCI_FIXUP_EARLY()? quirk_al_msi_disable() uses the
_CLASS version because the same Device ID is used for non-Root Port
devices. Is the same true here, or could these use
DECLARE_PCI_FIXUP_EARLY()?
There are many quirks that disable MSI, and they're a mixture of EARLY
and FINAL. They should probably all be the same.
[1] https://git.kernel.org/linus/8c7e96d3fe75
> Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
> ---
> drivers/pci/quirks.c | 9 +++++++++
> 1 file changed, 9 insertions(+)
>
> diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
> index d2dd6a6cda60..3ac5c45e61a1 100644
> --- a/drivers/pci/quirks.c
> +++ b/drivers/pci/quirks.c
> @@ -2747,6 +2747,15 @@ DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x10e5,
> DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x10e6,
> PCI_CLASS_BRIDGE_PCI, 8,
> pci_quirk_nvidia_tegra_disable_rp_msi);
> +DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x229a,
> + PCI_CLASS_BRIDGE_PCI, 8,
> + pci_quirk_nvidia_tegra_disable_rp_msi);
> +DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x229c,
> + PCI_CLASS_BRIDGE_PCI, 8,
> + pci_quirk_nvidia_tegra_disable_rp_msi);
> +DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x229e,
> + PCI_CLASS_BRIDGE_PCI, 8,
> + pci_quirk_nvidia_tegra_disable_rp_msi);
>
> /*
> * Some versions of the MCP55 bridge from Nvidia have a legacy IRQ routing
> --
> 2.17.1
>
^ permalink raw reply [flat|nested] 46+ messages in thread
* Re: [PATCH V1 10/10] PCI: tegra: Add Tegra234 PCIe support
2022-02-05 16:21 ` Vidya Sagar
@ 2022-02-07 18:19 ` Bjorn Helgaas
-1 siblings, 0 replies; 46+ messages in thread
From: Bjorn Helgaas @ 2022-02-07 18:19 UTC (permalink / raw)
To: Vidya Sagar
Cc: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh,
kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, sagar.tv
On Sat, Feb 05, 2022 at 09:51:44PM +0530, Vidya Sagar wrote:
> Add support for Synopsys DesignWare core IP based PCIe host controllers
> present in the Tegra234 SoC.
Can you please split this into two patches?
- A patch that does the struct tegra194_pcie_of_data to struct
tegra_pcie_of_data conversion but doesn't change any
functionality, and
- A patch that adds Tegra234 support?
It would be really helpful to be able to see the Tegra234 stuff by
itself.
Bjorn
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply [flat|nested] 46+ messages in thread
* Re: [PATCH V1 10/10] PCI: tegra: Add Tegra234 PCIe support
@ 2022-02-07 18:19 ` Bjorn Helgaas
0 siblings, 0 replies; 46+ messages in thread
From: Bjorn Helgaas @ 2022-02-07 18:19 UTC (permalink / raw)
To: Vidya Sagar
Cc: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh,
kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, sagar.tv
On Sat, Feb 05, 2022 at 09:51:44PM +0530, Vidya Sagar wrote:
> Add support for Synopsys DesignWare core IP based PCIe host controllers
> present in the Tegra234 SoC.
Can you please split this into two patches?
- A patch that does the struct tegra194_pcie_of_data to struct
tegra_pcie_of_data conversion but doesn't change any
functionality, and
- A patch that adds Tegra234 support?
It would be really helpful to be able to see the Tegra234 stuff by
itself.
Bjorn
^ permalink raw reply [flat|nested] 46+ messages in thread
* Re: [PATCH V1 01/10] dt-bindings: Add Tegra234 PCIe clocks and resets
2022-02-05 16:21 ` Vidya Sagar
@ 2022-02-11 14:51 ` Rob Herring
-1 siblings, 0 replies; 46+ messages in thread
From: Rob Herring @ 2022-02-11 14:51 UTC (permalink / raw)
To: Vidya Sagar
Cc: sagar.tv, kw, krzysztof.kozlowski, kthota, p.zabel, mmaddireddy,
devicetree, linux-phy, vkoul, jonathanh, robh+dt, bhelgaas,
lorenzo.pieralisi, thierry.reding, linux-kernel, kishon,
linux-pci, linux-tegra, mperttunen
On Sat, 05 Feb 2022 21:51:35 +0530, Vidya Sagar wrote:
> Add the clocks and resets used by the PCIe hardware found on
> Tegra234 SoCs.
>
> Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
> ---
> include/dt-bindings/clock/tegra234-clock.h | 25 +++++++++++++++++++-
> include/dt-bindings/reset/tegra234-reset.h | 27 +++++++++++++++++++++-
> 2 files changed, 50 insertions(+), 2 deletions(-)
>
Acked-by: Rob Herring <robh@kernel.org>
^ permalink raw reply [flat|nested] 46+ messages in thread
* Re: [PATCH V1 01/10] dt-bindings: Add Tegra234 PCIe clocks and resets
@ 2022-02-11 14:51 ` Rob Herring
0 siblings, 0 replies; 46+ messages in thread
From: Rob Herring @ 2022-02-11 14:51 UTC (permalink / raw)
To: Vidya Sagar
Cc: sagar.tv, kw, krzysztof.kozlowski, kthota, p.zabel, mmaddireddy,
devicetree, linux-phy, vkoul, jonathanh, robh+dt, bhelgaas,
lorenzo.pieralisi, thierry.reding, linux-kernel, kishon,
linux-pci, linux-tegra, mperttunen
On Sat, 05 Feb 2022 21:51:35 +0530, Vidya Sagar wrote:
> Add the clocks and resets used by the PCIe hardware found on
> Tegra234 SoCs.
>
> Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
> ---
> include/dt-bindings/clock/tegra234-clock.h | 25 +++++++++++++++++++-
> include/dt-bindings/reset/tegra234-reset.h | 27 +++++++++++++++++++++-
> 2 files changed, 50 insertions(+), 2 deletions(-)
>
Acked-by: Rob Herring <robh@kernel.org>
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply [flat|nested] 46+ messages in thread
* Re: [PATCH V1 02/10] dt-bindings: power: Add Tegra234 PCIe power domains
2022-02-05 16:21 ` Vidya Sagar
@ 2022-02-11 14:52 ` Rob Herring
-1 siblings, 0 replies; 46+ messages in thread
From: Rob Herring @ 2022-02-11 14:52 UTC (permalink / raw)
To: Vidya Sagar
Cc: vkoul, linux-tegra, linux-pci, kthota, mperttunen, bhelgaas,
linux-phy, krzysztof.kozlowski, sagar.tv, mmaddireddy, robh+dt,
devicetree, jonathanh, kishon, lorenzo.pieralisi, linux-kernel,
kw, thierry.reding, p.zabel
On Sat, 05 Feb 2022 21:51:36 +0530, Vidya Sagar wrote:
> Add power domain IDs for the four PCIe power partitions found on
> Tegra234.
>
> Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
> ---
> .../dt-bindings/power/tegra234-powergate.h | 20 +++++++++++++++++++
> 1 file changed, 20 insertions(+)
> create mode 100644 include/dt-bindings/power/tegra234-powergate.h
>
Acked-by: Rob Herring <robh@kernel.org>
^ permalink raw reply [flat|nested] 46+ messages in thread
* Re: [PATCH V1 02/10] dt-bindings: power: Add Tegra234 PCIe power domains
@ 2022-02-11 14:52 ` Rob Herring
0 siblings, 0 replies; 46+ messages in thread
From: Rob Herring @ 2022-02-11 14:52 UTC (permalink / raw)
To: Vidya Sagar
Cc: vkoul, linux-tegra, linux-pci, kthota, mperttunen, bhelgaas,
linux-phy, krzysztof.kozlowski, sagar.tv, mmaddireddy, robh+dt,
devicetree, jonathanh, kishon, lorenzo.pieralisi, linux-kernel,
kw, thierry.reding, p.zabel
On Sat, 05 Feb 2022 21:51:36 +0530, Vidya Sagar wrote:
> Add power domain IDs for the four PCIe power partitions found on
> Tegra234.
>
> Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
> ---
> .../dt-bindings/power/tegra234-powergate.h | 20 +++++++++++++++++++
> 1 file changed, 20 insertions(+)
> create mode 100644 include/dt-bindings/power/tegra234-powergate.h
>
Acked-by: Rob Herring <robh@kernel.org>
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply [flat|nested] 46+ messages in thread
* Re: [PATCH V1 03/10] dt-bindings: memory: Add Tegra234 PCIe memory
2022-02-05 16:21 ` Vidya Sagar
@ 2022-02-11 14:53 ` Rob Herring
-1 siblings, 0 replies; 46+ messages in thread
From: Rob Herring @ 2022-02-11 14:53 UTC (permalink / raw)
To: Vidya Sagar
Cc: robh+dt, kishon, mperttunen, kw, linux-tegra, linux-pci,
linux-kernel, kthota, sagar.tv, vkoul, devicetree, bhelgaas,
lorenzo.pieralisi, mmaddireddy, krzysztof.kozlowski, p.zabel,
thierry.reding, linux-phy, jonathanh
On Sat, 05 Feb 2022 21:51:37 +0530, Vidya Sagar wrote:
> Add the memory client and stream ID definitions for the PCIe hardware
> found on Tegra234 SoCs.
>
> Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
> ---
> include/dt-bindings/memory/tegra234-mc.h | 64 ++++++++++++++++++++++++
> 1 file changed, 64 insertions(+)
>
Acked-by: Rob Herring <robh@kernel.org>
^ permalink raw reply [flat|nested] 46+ messages in thread
* Re: [PATCH V1 03/10] dt-bindings: memory: Add Tegra234 PCIe memory
@ 2022-02-11 14:53 ` Rob Herring
0 siblings, 0 replies; 46+ messages in thread
From: Rob Herring @ 2022-02-11 14:53 UTC (permalink / raw)
To: Vidya Sagar
Cc: robh+dt, kishon, mperttunen, kw, linux-tegra, linux-pci,
linux-kernel, kthota, sagar.tv, vkoul, devicetree, bhelgaas,
lorenzo.pieralisi, mmaddireddy, krzysztof.kozlowski, p.zabel,
thierry.reding, linux-phy, jonathanh
On Sat, 05 Feb 2022 21:51:37 +0530, Vidya Sagar wrote:
> Add the memory client and stream ID definitions for the PCIe hardware
> found on Tegra234 SoCs.
>
> Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
> ---
> include/dt-bindings/memory/tegra234-mc.h | 64 ++++++++++++++++++++++++
> 1 file changed, 64 insertions(+)
>
Acked-by: Rob Herring <robh@kernel.org>
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply [flat|nested] 46+ messages in thread
* Re: [PATCH V1 04/10] dt-bindings: PHY: P2U: Add support for Tegra234 P2U block
2022-02-05 16:21 ` Vidya Sagar
@ 2022-02-11 14:55 ` Rob Herring
-1 siblings, 0 replies; 46+ messages in thread
From: Rob Herring @ 2022-02-11 14:55 UTC (permalink / raw)
To: Vidya Sagar
Cc: bhelgaas, lorenzo.pieralisi, thierry.reding, jonathanh, kishon,
vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen, linux-pci,
devicetree, linux-tegra, linux-kernel, linux-phy, kthota,
mmaddireddy, sagar.tv
On Sat, Feb 05, 2022 at 09:51:38PM +0530, Vidya Sagar wrote:
> Add support for Tegra234 P2U (PIPE to UPHY) module block which is a glue
> module instantiated once for each PCIe lane between Synopsys DesignWare
> core based PCIe IP and Universal PHY block.
>
> Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
> ---
> .../bindings/phy/phy-tegra194-p2u.yaml | 17 +++++++++++++----
> 1 file changed, 13 insertions(+), 4 deletions(-)
>
> diff --git a/Documentation/devicetree/bindings/phy/phy-tegra194-p2u.yaml b/Documentation/devicetree/bindings/phy/phy-tegra194-p2u.yaml
> index 9a89d05efbda..6ba1f69b1126 100644
> --- a/Documentation/devicetree/bindings/phy/phy-tegra194-p2u.yaml
> +++ b/Documentation/devicetree/bindings/phy/phy-tegra194-p2u.yaml
> @@ -4,7 +4,7 @@
> $id: "http://devicetree.org/schemas/phy/phy-tegra194-p2u.yaml#"
> $schema: "http://devicetree.org/meta-schemas/core.yaml#"
>
> -title: NVIDIA Tegra194 P2U binding
> +title: NVIDIA Tegra194 & Tegra234 P2U binding
>
> maintainers:
> - Thierry Reding <treding@nvidia.com>
> @@ -12,13 +12,17 @@ maintainers:
> description: >
> Tegra194 has two PHY bricks namely HSIO (High Speed IO) and NVHS (NVIDIA High
> Speed) each interfacing with 12 and 8 P2U instances respectively.
> + Tegra234 has three PHY bricks namesly HSIO, NVHS and GBE (Gigabit Ethernet)
> + each interfacing with 8, 8 and 8 P2U instances respectively.
> A P2U instance is a glue logic between Synopsys DesignWare Core PCIe IP's PIPE
> - interface and PHY of HSIO/NVHS bricks. Each P2U instance represents one PCIe
> - lane.
> + interface and PHY of HSIO/NVHS/GBE bricks. Each P2U instance represents one
> + PCIe lane.
>
> properties:
> compatible:
> - const: nvidia,tegra194-p2u
> + oneOf:
> + - const: nvidia,tegra194-p2u
> + - const: nvidia,tegra234-p2u
Use 'enum'
>
> reg:
> maxItems: 1
> @@ -28,6 +32,11 @@ properties:
> items:
> - const: ctl
>
> + nvidia,skip-sz-protect-en:
> + description: Should be present if two PCIe retimers are present between
> + the root port and its immediate downstream device.
> + type: boolean
Check your indentation.
This patch should have failed checks for both of these issues. No report
so either this patch couldn't be applied or there another issue. In any
case, you failed to test this yourself.
Rob
^ permalink raw reply [flat|nested] 46+ messages in thread
* Re: [PATCH V1 04/10] dt-bindings: PHY: P2U: Add support for Tegra234 P2U block
@ 2022-02-11 14:55 ` Rob Herring
0 siblings, 0 replies; 46+ messages in thread
From: Rob Herring @ 2022-02-11 14:55 UTC (permalink / raw)
To: Vidya Sagar
Cc: bhelgaas, lorenzo.pieralisi, thierry.reding, jonathanh, kishon,
vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen, linux-pci,
devicetree, linux-tegra, linux-kernel, linux-phy, kthota,
mmaddireddy, sagar.tv
On Sat, Feb 05, 2022 at 09:51:38PM +0530, Vidya Sagar wrote:
> Add support for Tegra234 P2U (PIPE to UPHY) module block which is a glue
> module instantiated once for each PCIe lane between Synopsys DesignWare
> core based PCIe IP and Universal PHY block.
>
> Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
> ---
> .../bindings/phy/phy-tegra194-p2u.yaml | 17 +++++++++++++----
> 1 file changed, 13 insertions(+), 4 deletions(-)
>
> diff --git a/Documentation/devicetree/bindings/phy/phy-tegra194-p2u.yaml b/Documentation/devicetree/bindings/phy/phy-tegra194-p2u.yaml
> index 9a89d05efbda..6ba1f69b1126 100644
> --- a/Documentation/devicetree/bindings/phy/phy-tegra194-p2u.yaml
> +++ b/Documentation/devicetree/bindings/phy/phy-tegra194-p2u.yaml
> @@ -4,7 +4,7 @@
> $id: "http://devicetree.org/schemas/phy/phy-tegra194-p2u.yaml#"
> $schema: "http://devicetree.org/meta-schemas/core.yaml#"
>
> -title: NVIDIA Tegra194 P2U binding
> +title: NVIDIA Tegra194 & Tegra234 P2U binding
>
> maintainers:
> - Thierry Reding <treding@nvidia.com>
> @@ -12,13 +12,17 @@ maintainers:
> description: >
> Tegra194 has two PHY bricks namely HSIO (High Speed IO) and NVHS (NVIDIA High
> Speed) each interfacing with 12 and 8 P2U instances respectively.
> + Tegra234 has three PHY bricks namesly HSIO, NVHS and GBE (Gigabit Ethernet)
> + each interfacing with 8, 8 and 8 P2U instances respectively.
> A P2U instance is a glue logic between Synopsys DesignWare Core PCIe IP's PIPE
> - interface and PHY of HSIO/NVHS bricks. Each P2U instance represents one PCIe
> - lane.
> + interface and PHY of HSIO/NVHS/GBE bricks. Each P2U instance represents one
> + PCIe lane.
>
> properties:
> compatible:
> - const: nvidia,tegra194-p2u
> + oneOf:
> + - const: nvidia,tegra194-p2u
> + - const: nvidia,tegra234-p2u
Use 'enum'
>
> reg:
> maxItems: 1
> @@ -28,6 +32,11 @@ properties:
> items:
> - const: ctl
>
> + nvidia,skip-sz-protect-en:
> + description: Should be present if two PCIe retimers are present between
> + the root port and its immediate downstream device.
> + type: boolean
Check your indentation.
This patch should have failed checks for both of these issues. No report
so either this patch couldn't be applied or there another issue. In any
case, you failed to test this yourself.
Rob
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply [flat|nested] 46+ messages in thread
* Re: [PATCH V1 05/10] dt-bindings: PCI: tegra: Add device tree support for Tegra234
2022-02-05 16:21 ` Vidya Sagar
@ 2022-02-11 14:57 ` Rob Herring
-1 siblings, 0 replies; 46+ messages in thread
From: Rob Herring @ 2022-02-11 14:57 UTC (permalink / raw)
To: Vidya Sagar
Cc: bhelgaas, lorenzo.pieralisi, thierry.reding, jonathanh, kishon,
vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen, linux-pci,
devicetree, linux-tegra, linux-kernel, linux-phy, kthota,
mmaddireddy, sagar.tv
On Sat, Feb 05, 2022 at 09:51:39PM +0530, Vidya Sagar wrote:
> Add support for Tegra234 PCIe controllers. These controllers are based
> on Synopsys DesignWare core IP.
>
> Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
> ---
> .../bindings/pci/nvidia,tegra194-pcie.txt | 106 +++++++++++++++++-
> 1 file changed, 103 insertions(+), 3 deletions(-)
This falls above my threshold of convert this to schema first.
>
> diff --git a/Documentation/devicetree/bindings/pci/nvidia,tegra194-pcie.txt b/Documentation/devicetree/bindings/pci/nvidia,tegra194-pcie.txt
> index 8e4f9bfb316d..ae9555a50fb0 100644
> --- a/Documentation/devicetree/bindings/pci/nvidia,tegra194-pcie.txt
> +++ b/Documentation/devicetree/bindings/pci/nvidia,tegra194-pcie.txt
> @@ -10,6 +10,8 @@ Required properties:
> - power-domains: A phandle to the node that controls power to the respective
> PCIe controller and a specifier name for the PCIe controller. Following are
> the specifiers for the different PCIe controllers
> + Tegra194:
> + ---------
> TEGRA194_POWER_DOMAIN_PCIEX8B: C0
> TEGRA194_POWER_DOMAIN_PCIEX1A: C1
> TEGRA194_POWER_DOMAIN_PCIEX1A: C2
> @@ -18,6 +20,21 @@ Required properties:
> TEGRA194_POWER_DOMAIN_PCIEX8A: C5
> these specifiers are defined in
> "include/dt-bindings/power/tegra194-powergate.h" file.
> + Tegra234:
> + ---------
> + TEGRA234_POWER_DOMAIN_PCIEX4BA: C0
> + TEGRA234_POWER_DOMAIN_PCIEX1A : C1
> + TEGRA234_POWER_DOMAIN_PCIEX1A : C2
> + TEGRA234_POWER_DOMAIN_PCIEX1A : C3
> + TEGRA234_POWER_DOMAIN_PCIEX4BB: C4
> + TEGRA234_POWER_DOMAIN_PCIEX8A : C5
> + TEGRA234_POWER_DOMAIN_PCIEX4A : C6
> + TEGRA234_POWER_DOMAIN_PCIEX8B : C7
> + TEGRA234_POWER_DOMAIN_PCIEX4CA: C8
> + TEGRA234_POWER_DOMAIN_PCIEX4CB: C9
> + TEGRA234_POWER_DOMAIN_PCIEX4CC: C10
> + these specifiers are defined in
> + "include/dt-bindings/power/tegra234-powergate.h" file.
> - reg: A list of physical base address and length pairs for each set of
> controller registers. Must contain an entry for each entry in the reg-names
> property.
> @@ -47,16 +64,33 @@ Required properties:
> "p2u-N": where N ranges from 0 to one less than the total number of lanes
> - nvidia,bpmp: Must contain a pair of phandle to BPMP controller node followed
> by controller-id. Following are the controller ids for each controller.
> + Tegra194:
> + ---------
> 0: C0
> 1: C1
> 2: C2
> 3: C3
> 4: C4
> 5: C5
> + Tegra194:
> + ---------
> + 0 : C0
> + 1 : C1
> + 2 : C2
> + 3 : C3
> + 4 : C4
> + 5 : C5
> + 6 : C6
> + 7 : C7
> + 8 : C8
> + 9 : C9
> + 10: C10
> - vddio-pex-ctl-supply: Regulator supply for PCIe side band signals
>
> RC mode:
> -- compatible: Tegra19x must contain "nvidia,tegra194-pcie"
> +- compatible: Must be
> + - "nvidia,tegra194-pcie" for Tegra194
> + - "nvidia,tegra234-pcie" for Tegra234
> - device_type: Must be "pci" for RC mode
> - interrupt-names: Must include the following entries:
> "msi": The Tegra interrupt that is asserted when an MSI is received
> @@ -99,7 +133,8 @@ In Tegra194, Only controllers C0, C4 & C5 support EP mode.
>
> Optional properties:
> - pinctrl-names: A list of pinctrl state names.
> - It is mandatory for C5 controller and optional for other controllers.
> + It is mandatory for C5 controller in Tegra194 and optional for other
> + controllers.
> - "default": Configures PCIe I/O for proper operation.
> - pinctrl-0: phandle for the 'default' state of pin configuration.
> It is mandatory for C5 controller and optional for other controllers.
> @@ -108,7 +143,7 @@ Optional properties:
> improve performance when a platform is designed in such a way that it
> satisfies at least one of the following conditions thereby enabling root
> port to exchange optimum number of FC (Flow Control) credits with
> - downstream devices
> + downstream devices. This is applicable for only Tegra194.
> 1. If C0/C4/C5 run at x1/x2 link widths (irrespective of speed and MPS)
> 2. If C0/C1/C2/C3/C4/C5 operate at their respective max link widths and
> a) speed is Gen-2 and MPS is 256B
> @@ -119,6 +154,9 @@ Optional properties:
> specified in microseconds
> - nvidia,aspm-l0s-entrance-latency-us: ASPM L0s entrance latency to be
> specified in microseconds
> +- nvidia,enable-srns: This boolean property needs to be present if the
> + controller is configured to operate in SRNS (Separate Reference Clocks with
> + No Spread-Spectrum Clocking). This is applicable only for Tegra234.
>
> RC mode:
> - vpcie3v3-supply: A phandle to the regulator node that supplies 3.3V to the slot
> @@ -127,6 +165,9 @@ RC mode:
> - vpcie12v-supply: A phandle to the regulator node that supplies 12V to the slot
> if the platform has one such slot. (Ex:- x16 slot owned by C5 controller
> in p2972-0000 platform).
> +- nvidia,enable-ext-refclk: This boolean property needs to be present if the
> + controller is configured to use the reference clocking coming in from an
> + external clock source instead of using the internal clock source.
>
> EP mode:
> - nvidia,refclk-select-gpios: Must contain a phandle to a GPIO controller
> @@ -243,3 +284,62 @@ Tegra194 EP mode:
> phy-names = "p2u-0", "p2u-1", "p2u-2", "p2u-3", "p2u-4",
> "p2u-5", "p2u-6", "p2u-7";
> };
> +
> +Tegra234 RP mode:
> +-----------------
> +
> + pcie@141a0000 {
> + compatible = "nvidia,tegra234-pcie";
> + power-domains = <&bpmp TEGRA234_POWER_DOMAIN_PCIEX8A>;
> + reg = <0x00 0x141a0000 0x0 0x00020000>, /* appl registers (128K) */
> + <0x00 0x3a000000 0x0 0x00040000>, /* configuration space (256K) */
> + <0x00 0x3a040000 0x0 0x00040000>, /* iATU_DMA reg space (256K) */
> + <0x00 0x3a080000 0x0 0x00040000>; /* DBI reg space (256K) */
> + reg-names = "appl", "config", "atu_dma", "dbi";
> +
> + status = "disabled";
Why is your example disabled?
> +
> + #address-cells = <3>;
> + #size-cells = <2>;
> + device_type = "pci";
> + num-lanes = <8>;
> + num-viewport = <8>;
> + linux,pci-domain = <5>;
> +
> + clocks = <&bpmp TEGRA234_CLK_PEX1_C5_CORE>;
> + clock-names = "core";
> +
> + resets = <&bpmp TEGRA234_RESET_PEX1_CORE_5_APB>,
> + <&bpmp TEGRA234_RESET_PEX1_CORE_5>;
> + reset-names = "apb", "core";
> +
> + interrupts = <GIC_SPI 53 IRQ_TYPE_LEVEL_HIGH>, /* controller interrupt */
> + <GIC_SPI 54 IRQ_TYPE_LEVEL_HIGH>; /* MSI interrupt */
> + interrupt-names = "intr", "msi";
> +
> + #interrupt-cells = <1>;
> + interrupt-map-mask = <0 0 0 0>;
> + interrupt-map = <0 0 0 0 &gic GIC_SPI 53 IRQ_TYPE_LEVEL_HIGH>;
> +
> + nvidia,bpmp = <&bpmp 5>;
> +
> + nvidia,aspm-cmrt-us = <60>;
> + nvidia,aspm-pwr-on-t-us = <20>;
> + nvidia,aspm-l0s-entrance-latency-us = <3>;
> +
> + bus-range = <0x0 0xff>;
> +
> + ranges = <0x43000000 0x27 0x40000000 0x27 0x40000000 0x3 0xe8000000>, /* prefetchable memory (16000 MB) */
> + <0x02000000 0x0 0x40000000 0x2b 0x28000000 0x0 0x08000000>, /* non-prefetchable memory (128 MB) */
> + <0x01000000 0x0 0x3a100000 0x00 0x3a100000 0x0 0x00100000>; /* downstream I/O (1 MB) */
> +
> + interconnects = <&mc TEGRA234_MEMORY_CLIENT_PCIE5R &emc>,
> + <&mc TEGRA234_MEMORY_CLIENT_PCIE5W &emc>;
> + interconnect-names = "dma-mem", "write";
> +
> + phys = <&p2u_nvhs_0>, <&p2u_nvhs_1>, <&p2u_nvhs_2>,
> + <&p2u_nvhs_3>, <&p2u_nvhs_4>, <&p2u_nvhs_5>,
> + <&p2u_nvhs_6>, <&p2u_nvhs_7>;
> + phy-names = "p2u-0", "p2u-1", "p2u-2", "p2u-3", "p2u-4",
> + "p2u-5", "p2u-6", "p2u-7";
> + };
> --
> 2.17.1
>
>
^ permalink raw reply [flat|nested] 46+ messages in thread
* Re: [PATCH V1 05/10] dt-bindings: PCI: tegra: Add device tree support for Tegra234
@ 2022-02-11 14:57 ` Rob Herring
0 siblings, 0 replies; 46+ messages in thread
From: Rob Herring @ 2022-02-11 14:57 UTC (permalink / raw)
To: Vidya Sagar
Cc: bhelgaas, lorenzo.pieralisi, thierry.reding, jonathanh, kishon,
vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen, linux-pci,
devicetree, linux-tegra, linux-kernel, linux-phy, kthota,
mmaddireddy, sagar.tv
On Sat, Feb 05, 2022 at 09:51:39PM +0530, Vidya Sagar wrote:
> Add support for Tegra234 PCIe controllers. These controllers are based
> on Synopsys DesignWare core IP.
>
> Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
> ---
> .../bindings/pci/nvidia,tegra194-pcie.txt | 106 +++++++++++++++++-
> 1 file changed, 103 insertions(+), 3 deletions(-)
This falls above my threshold of convert this to schema first.
>
> diff --git a/Documentation/devicetree/bindings/pci/nvidia,tegra194-pcie.txt b/Documentation/devicetree/bindings/pci/nvidia,tegra194-pcie.txt
> index 8e4f9bfb316d..ae9555a50fb0 100644
> --- a/Documentation/devicetree/bindings/pci/nvidia,tegra194-pcie.txt
> +++ b/Documentation/devicetree/bindings/pci/nvidia,tegra194-pcie.txt
> @@ -10,6 +10,8 @@ Required properties:
> - power-domains: A phandle to the node that controls power to the respective
> PCIe controller and a specifier name for the PCIe controller. Following are
> the specifiers for the different PCIe controllers
> + Tegra194:
> + ---------
> TEGRA194_POWER_DOMAIN_PCIEX8B: C0
> TEGRA194_POWER_DOMAIN_PCIEX1A: C1
> TEGRA194_POWER_DOMAIN_PCIEX1A: C2
> @@ -18,6 +20,21 @@ Required properties:
> TEGRA194_POWER_DOMAIN_PCIEX8A: C5
> these specifiers are defined in
> "include/dt-bindings/power/tegra194-powergate.h" file.
> + Tegra234:
> + ---------
> + TEGRA234_POWER_DOMAIN_PCIEX4BA: C0
> + TEGRA234_POWER_DOMAIN_PCIEX1A : C1
> + TEGRA234_POWER_DOMAIN_PCIEX1A : C2
> + TEGRA234_POWER_DOMAIN_PCIEX1A : C3
> + TEGRA234_POWER_DOMAIN_PCIEX4BB: C4
> + TEGRA234_POWER_DOMAIN_PCIEX8A : C5
> + TEGRA234_POWER_DOMAIN_PCIEX4A : C6
> + TEGRA234_POWER_DOMAIN_PCIEX8B : C7
> + TEGRA234_POWER_DOMAIN_PCIEX4CA: C8
> + TEGRA234_POWER_DOMAIN_PCIEX4CB: C9
> + TEGRA234_POWER_DOMAIN_PCIEX4CC: C10
> + these specifiers are defined in
> + "include/dt-bindings/power/tegra234-powergate.h" file.
> - reg: A list of physical base address and length pairs for each set of
> controller registers. Must contain an entry for each entry in the reg-names
> property.
> @@ -47,16 +64,33 @@ Required properties:
> "p2u-N": where N ranges from 0 to one less than the total number of lanes
> - nvidia,bpmp: Must contain a pair of phandle to BPMP controller node followed
> by controller-id. Following are the controller ids for each controller.
> + Tegra194:
> + ---------
> 0: C0
> 1: C1
> 2: C2
> 3: C3
> 4: C4
> 5: C5
> + Tegra194:
> + ---------
> + 0 : C0
> + 1 : C1
> + 2 : C2
> + 3 : C3
> + 4 : C4
> + 5 : C5
> + 6 : C6
> + 7 : C7
> + 8 : C8
> + 9 : C9
> + 10: C10
> - vddio-pex-ctl-supply: Regulator supply for PCIe side band signals
>
> RC mode:
> -- compatible: Tegra19x must contain "nvidia,tegra194-pcie"
> +- compatible: Must be
> + - "nvidia,tegra194-pcie" for Tegra194
> + - "nvidia,tegra234-pcie" for Tegra234
> - device_type: Must be "pci" for RC mode
> - interrupt-names: Must include the following entries:
> "msi": The Tegra interrupt that is asserted when an MSI is received
> @@ -99,7 +133,8 @@ In Tegra194, Only controllers C0, C4 & C5 support EP mode.
>
> Optional properties:
> - pinctrl-names: A list of pinctrl state names.
> - It is mandatory for C5 controller and optional for other controllers.
> + It is mandatory for C5 controller in Tegra194 and optional for other
> + controllers.
> - "default": Configures PCIe I/O for proper operation.
> - pinctrl-0: phandle for the 'default' state of pin configuration.
> It is mandatory for C5 controller and optional for other controllers.
> @@ -108,7 +143,7 @@ Optional properties:
> improve performance when a platform is designed in such a way that it
> satisfies at least one of the following conditions thereby enabling root
> port to exchange optimum number of FC (Flow Control) credits with
> - downstream devices
> + downstream devices. This is applicable for only Tegra194.
> 1. If C0/C4/C5 run at x1/x2 link widths (irrespective of speed and MPS)
> 2. If C0/C1/C2/C3/C4/C5 operate at their respective max link widths and
> a) speed is Gen-2 and MPS is 256B
> @@ -119,6 +154,9 @@ Optional properties:
> specified in microseconds
> - nvidia,aspm-l0s-entrance-latency-us: ASPM L0s entrance latency to be
> specified in microseconds
> +- nvidia,enable-srns: This boolean property needs to be present if the
> + controller is configured to operate in SRNS (Separate Reference Clocks with
> + No Spread-Spectrum Clocking). This is applicable only for Tegra234.
>
> RC mode:
> - vpcie3v3-supply: A phandle to the regulator node that supplies 3.3V to the slot
> @@ -127,6 +165,9 @@ RC mode:
> - vpcie12v-supply: A phandle to the regulator node that supplies 12V to the slot
> if the platform has one such slot. (Ex:- x16 slot owned by C5 controller
> in p2972-0000 platform).
> +- nvidia,enable-ext-refclk: This boolean property needs to be present if the
> + controller is configured to use the reference clocking coming in from an
> + external clock source instead of using the internal clock source.
>
> EP mode:
> - nvidia,refclk-select-gpios: Must contain a phandle to a GPIO controller
> @@ -243,3 +284,62 @@ Tegra194 EP mode:
> phy-names = "p2u-0", "p2u-1", "p2u-2", "p2u-3", "p2u-4",
> "p2u-5", "p2u-6", "p2u-7";
> };
> +
> +Tegra234 RP mode:
> +-----------------
> +
> + pcie@141a0000 {
> + compatible = "nvidia,tegra234-pcie";
> + power-domains = <&bpmp TEGRA234_POWER_DOMAIN_PCIEX8A>;
> + reg = <0x00 0x141a0000 0x0 0x00020000>, /* appl registers (128K) */
> + <0x00 0x3a000000 0x0 0x00040000>, /* configuration space (256K) */
> + <0x00 0x3a040000 0x0 0x00040000>, /* iATU_DMA reg space (256K) */
> + <0x00 0x3a080000 0x0 0x00040000>; /* DBI reg space (256K) */
> + reg-names = "appl", "config", "atu_dma", "dbi";
> +
> + status = "disabled";
Why is your example disabled?
> +
> + #address-cells = <3>;
> + #size-cells = <2>;
> + device_type = "pci";
> + num-lanes = <8>;
> + num-viewport = <8>;
> + linux,pci-domain = <5>;
> +
> + clocks = <&bpmp TEGRA234_CLK_PEX1_C5_CORE>;
> + clock-names = "core";
> +
> + resets = <&bpmp TEGRA234_RESET_PEX1_CORE_5_APB>,
> + <&bpmp TEGRA234_RESET_PEX1_CORE_5>;
> + reset-names = "apb", "core";
> +
> + interrupts = <GIC_SPI 53 IRQ_TYPE_LEVEL_HIGH>, /* controller interrupt */
> + <GIC_SPI 54 IRQ_TYPE_LEVEL_HIGH>; /* MSI interrupt */
> + interrupt-names = "intr", "msi";
> +
> + #interrupt-cells = <1>;
> + interrupt-map-mask = <0 0 0 0>;
> + interrupt-map = <0 0 0 0 &gic GIC_SPI 53 IRQ_TYPE_LEVEL_HIGH>;
> +
> + nvidia,bpmp = <&bpmp 5>;
> +
> + nvidia,aspm-cmrt-us = <60>;
> + nvidia,aspm-pwr-on-t-us = <20>;
> + nvidia,aspm-l0s-entrance-latency-us = <3>;
> +
> + bus-range = <0x0 0xff>;
> +
> + ranges = <0x43000000 0x27 0x40000000 0x27 0x40000000 0x3 0xe8000000>, /* prefetchable memory (16000 MB) */
> + <0x02000000 0x0 0x40000000 0x2b 0x28000000 0x0 0x08000000>, /* non-prefetchable memory (128 MB) */
> + <0x01000000 0x0 0x3a100000 0x00 0x3a100000 0x0 0x00100000>; /* downstream I/O (1 MB) */
> +
> + interconnects = <&mc TEGRA234_MEMORY_CLIENT_PCIE5R &emc>,
> + <&mc TEGRA234_MEMORY_CLIENT_PCIE5W &emc>;
> + interconnect-names = "dma-mem", "write";
> +
> + phys = <&p2u_nvhs_0>, <&p2u_nvhs_1>, <&p2u_nvhs_2>,
> + <&p2u_nvhs_3>, <&p2u_nvhs_4>, <&p2u_nvhs_5>,
> + <&p2u_nvhs_6>, <&p2u_nvhs_7>;
> + phy-names = "p2u-0", "p2u-1", "p2u-2", "p2u-3", "p2u-4",
> + "p2u-5", "p2u-6", "p2u-7";
> + };
> --
> 2.17.1
>
>
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply [flat|nested] 46+ messages in thread
* Re: [PATCH V1 03/10] dt-bindings: memory: Add Tegra234 PCIe memory
2022-02-06 11:33 ` Krzysztof Kozlowski
@ 2022-02-24 19:04 ` Thierry Reding
-1 siblings, 0 replies; 46+ messages in thread
From: Thierry Reding @ 2022-02-24 19:04 UTC (permalink / raw)
To: Krzysztof Kozlowski
Cc: Vidya Sagar, bhelgaas, lorenzo.pieralisi, robh+dt, jonathanh,
kishon, vkoul, kw, p.zabel, mperttunen, linux-pci, devicetree,
linux-tegra, linux-kernel, linux-phy, kthota, mmaddireddy,
sagar.tv
[-- Attachment #1: Type: text/plain, Size: 2325 bytes --]
On Sun, Feb 06, 2022 at 12:33:27PM +0100, Krzysztof Kozlowski wrote:
> On 05/02/2022 17:21, Vidya Sagar wrote:
> > Add the memory client and stream ID definitions for the PCIe hardware
> > found on Tegra234 SoCs.
>
> I could not find dependencies or merging strategy in cover letter.
> Please always describe it, so I don't have to go through all the patches
> to figure this out.
>
> >
> > Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
> > ---
> > include/dt-bindings/memory/tegra234-mc.h | 64 ++++++++++++++++++++++++
> > 1 file changed, 64 insertions(+)
> >
> > diff --git a/include/dt-bindings/memory/tegra234-mc.h b/include/dt-bindings/memory/tegra234-mc.h
> > index 2662f70c15c6..60017684858a 100644
> > --- a/include/dt-bindings/memory/tegra234-mc.h
> > +++ b/include/dt-bindings/memory/tegra234-mc.h
> > @@ -7,15 +7,53 @@
> > #define TEGRA234_SID_INVALID 0x00
> > #define TEGRA234_SID_PASSTHROUGH 0x7f
> >
> > +/* NISO0 stream IDs */
> > +#define TEGRA234_SID_PCIE0 0x12U
> > +#define TEGRA234_SID_PCIE4 0x13U
> > +#define TEGRA234_SID_PCIE5 0x14U
> > +#define TEGRA234_SID_PCIE6 0x15U
> > +#define TEGRA234_SID_PCIE9 0x1FU
> >
> > /* NISO1 stream IDs */
> > #define TEGRA234_SID_SDMMC4 0x02
> > +#define TEGRA234_SID_PCIE1 0x5U
> > +#define TEGRA234_SID_PCIE2 0x6U
> > +#define TEGRA234_SID_PCIE3 0x7U
> > +#define TEGRA234_SID_PCIE7 0x8U
> > +#define TEGRA234_SID_PCIE8 0x9U
> > +#define TEGRA234_SID_PCIE10 0xBU
>
> I don't see usage of these...
>
> > #define TEGRA234_SID_BPMP 0x10
> >
> > /*
> > * memory client IDs
> > */
> >
> > +/* PCIE6 read clients */
> > +#define TEGRA234_MEMORY_CLIENT_PCIE6AR 0x28
>
> I see you use them in DTS but not in mc driver. Don't you miss anything
> here?
This is along the same lines as the APE and HDA patches earlier, so I
would expect Vidya to add a memory controller patch that makes use of
these once the initial Tegra234 memory controller patch was merged.
Meanwhile, I've applied this to the Tegra tree, on top of the other
patches that add memory client definitions and resolved the conflicts
that ensued.
That way, by the time we get around to the next cycle all of these
dependencies will exist and applying the memory controller patches
should become easier.
Thierry
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]
^ permalink raw reply [flat|nested] 46+ messages in thread
* Re: [PATCH V1 03/10] dt-bindings: memory: Add Tegra234 PCIe memory
@ 2022-02-24 19:04 ` Thierry Reding
0 siblings, 0 replies; 46+ messages in thread
From: Thierry Reding @ 2022-02-24 19:04 UTC (permalink / raw)
To: Krzysztof Kozlowski
Cc: Vidya Sagar, bhelgaas, lorenzo.pieralisi, robh+dt, jonathanh,
kishon, vkoul, kw, p.zabel, mperttunen, linux-pci, devicetree,
linux-tegra, linux-kernel, linux-phy, kthota, mmaddireddy,
sagar.tv
[-- Attachment #1.1: Type: text/plain, Size: 2325 bytes --]
On Sun, Feb 06, 2022 at 12:33:27PM +0100, Krzysztof Kozlowski wrote:
> On 05/02/2022 17:21, Vidya Sagar wrote:
> > Add the memory client and stream ID definitions for the PCIe hardware
> > found on Tegra234 SoCs.
>
> I could not find dependencies or merging strategy in cover letter.
> Please always describe it, so I don't have to go through all the patches
> to figure this out.
>
> >
> > Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
> > ---
> > include/dt-bindings/memory/tegra234-mc.h | 64 ++++++++++++++++++++++++
> > 1 file changed, 64 insertions(+)
> >
> > diff --git a/include/dt-bindings/memory/tegra234-mc.h b/include/dt-bindings/memory/tegra234-mc.h
> > index 2662f70c15c6..60017684858a 100644
> > --- a/include/dt-bindings/memory/tegra234-mc.h
> > +++ b/include/dt-bindings/memory/tegra234-mc.h
> > @@ -7,15 +7,53 @@
> > #define TEGRA234_SID_INVALID 0x00
> > #define TEGRA234_SID_PASSTHROUGH 0x7f
> >
> > +/* NISO0 stream IDs */
> > +#define TEGRA234_SID_PCIE0 0x12U
> > +#define TEGRA234_SID_PCIE4 0x13U
> > +#define TEGRA234_SID_PCIE5 0x14U
> > +#define TEGRA234_SID_PCIE6 0x15U
> > +#define TEGRA234_SID_PCIE9 0x1FU
> >
> > /* NISO1 stream IDs */
> > #define TEGRA234_SID_SDMMC4 0x02
> > +#define TEGRA234_SID_PCIE1 0x5U
> > +#define TEGRA234_SID_PCIE2 0x6U
> > +#define TEGRA234_SID_PCIE3 0x7U
> > +#define TEGRA234_SID_PCIE7 0x8U
> > +#define TEGRA234_SID_PCIE8 0x9U
> > +#define TEGRA234_SID_PCIE10 0xBU
>
> I don't see usage of these...
>
> > #define TEGRA234_SID_BPMP 0x10
> >
> > /*
> > * memory client IDs
> > */
> >
> > +/* PCIE6 read clients */
> > +#define TEGRA234_MEMORY_CLIENT_PCIE6AR 0x28
>
> I see you use them in DTS but not in mc driver. Don't you miss anything
> here?
This is along the same lines as the APE and HDA patches earlier, so I
would expect Vidya to add a memory controller patch that makes use of
these once the initial Tegra234 memory controller patch was merged.
Meanwhile, I've applied this to the Tegra tree, on top of the other
patches that add memory client definitions and resolved the conflicts
that ensued.
That way, by the time we get around to the next cycle all of these
dependencies will exist and applying the memory controller patches
should become easier.
Thierry
[-- Attachment #1.2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]
[-- Attachment #2: Type: text/plain, Size: 112 bytes --]
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply [flat|nested] 46+ messages in thread
* Re: [PATCH V1 09/10] PCI: Disable MSI for Tegra234 root ports
2022-02-07 17:36 ` Bjorn Helgaas
@ 2022-04-23 7:22 ` Vidya Sagar
-1 siblings, 0 replies; 46+ messages in thread
From: Vidya Sagar @ 2022-04-23 7:22 UTC (permalink / raw)
To: Bjorn Helgaas
Cc: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh,
kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, sagar.tv
On 2/7/2022 11:06 PM, Bjorn Helgaas wrote:
> External email: Use caution opening links or attachments
>
>
> On Sat, Feb 05, 2022 at 09:51:43PM +0530, Vidya Sagar wrote:
>> Tegra234 PCIe rootports don't generate MSI interrupts for PME and AER
>> events. Since PCIe spec (Ref: r4.0 sec 7.7.1.2 and 7.7.2.2) doesn't support
>> using a mix of INTx and MSI/MSI-X, MSI needs to be disabled to avoid root
>> ports service drivers registering their respective ISRs with MSI interrupt
>> and to let only INTx be used for all events.
>
> s/rootports/root ports/ to match other usage here.
>
> This argument matches that in 8c7e96d3fe75 ("PCI: Disable MSI for
> Tegra root ports") [1], but that's not quite what sec 7.7.1.2 and
> 7.7.2.2 say. Those sections talk about what happens when both MSI and
> MSI-X are disabled:
>
> If MSI and MSI-X are both disabled, the Function requests servicing
> using INTx interrupts (if supported).
>
> but they don't say anything about what happens when MSI or MSI-X is
> *enabled*.
>
> I think a better citation is PCIe r6.0, sec 6.1.4.3, which says:
>
> While enabled for MSI or MSI-X operation, a Function is prohibited
> from using INTx interrupts (if implemented) to request service (MSI,
> MSI-X, and INTx are mutually exclusive).
>
> Can you please update the comment in the code and this commit log to
> cite PCIe r6.0, sec 6.1.4.3 instead, and to clarify that these Tegra
> devices always use INTx for PME and AER, even when MSI/MSI-X is
> enabled?
Sure. I would fix this in the next patch set.
>
> Why do these Tegra quirks use DECLARE_PCI_FIXUP_CLASS_EARLY() instead
> of just DECLARE_PCI_FIXUP_EARLY()? quirk_al_msi_disable() uses the
> _CLASS version because the same Device ID is used for non-Root Port
> devices. Is the same true here, or could these use
> DECLARE_PCI_FIXUP_EARLY()?
Tegra's PCIe controllers are also dual mode controllers and MSI works
just fine when they operate in the endpoint mode configuration.
>
> There are many quirks that disable MSI, and they're a mixture of EARLY
> and FINAL. They should probably all be the same.
>
> [1] https://git.kernel.org/linus/8c7e96d3fe75
>
>> Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
>> ---
>> drivers/pci/quirks.c | 9 +++++++++
>> 1 file changed, 9 insertions(+)
>>
>> diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
>> index d2dd6a6cda60..3ac5c45e61a1 100644
>> --- a/drivers/pci/quirks.c
>> +++ b/drivers/pci/quirks.c
>> @@ -2747,6 +2747,15 @@ DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x10e5,
>> DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x10e6,
>> PCI_CLASS_BRIDGE_PCI, 8,
>> pci_quirk_nvidia_tegra_disable_rp_msi);
>> +DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x229a,
>> + PCI_CLASS_BRIDGE_PCI, 8,
>> + pci_quirk_nvidia_tegra_disable_rp_msi);
>> +DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x229c,
>> + PCI_CLASS_BRIDGE_PCI, 8,
>> + pci_quirk_nvidia_tegra_disable_rp_msi);
>> +DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x229e,
>> + PCI_CLASS_BRIDGE_PCI, 8,
>> + pci_quirk_nvidia_tegra_disable_rp_msi);
>>
>> /*
>> * Some versions of the MCP55 bridge from Nvidia have a legacy IRQ routing
>> --
>> 2.17.1
>>
^ permalink raw reply [flat|nested] 46+ messages in thread
* Re: [PATCH V1 09/10] PCI: Disable MSI for Tegra234 root ports
@ 2022-04-23 7:22 ` Vidya Sagar
0 siblings, 0 replies; 46+ messages in thread
From: Vidya Sagar @ 2022-04-23 7:22 UTC (permalink / raw)
To: Bjorn Helgaas
Cc: bhelgaas, lorenzo.pieralisi, robh+dt, thierry.reding, jonathanh,
kishon, vkoul, kw, krzysztof.kozlowski, p.zabel, mperttunen,
linux-pci, devicetree, linux-tegra, linux-kernel, linux-phy,
kthota, mmaddireddy, sagar.tv
On 2/7/2022 11:06 PM, Bjorn Helgaas wrote:
> External email: Use caution opening links or attachments
>
>
> On Sat, Feb 05, 2022 at 09:51:43PM +0530, Vidya Sagar wrote:
>> Tegra234 PCIe rootports don't generate MSI interrupts for PME and AER
>> events. Since PCIe spec (Ref: r4.0 sec 7.7.1.2 and 7.7.2.2) doesn't support
>> using a mix of INTx and MSI/MSI-X, MSI needs to be disabled to avoid root
>> ports service drivers registering their respective ISRs with MSI interrupt
>> and to let only INTx be used for all events.
>
> s/rootports/root ports/ to match other usage here.
>
> This argument matches that in 8c7e96d3fe75 ("PCI: Disable MSI for
> Tegra root ports") [1], but that's not quite what sec 7.7.1.2 and
> 7.7.2.2 say. Those sections talk about what happens when both MSI and
> MSI-X are disabled:
>
> If MSI and MSI-X are both disabled, the Function requests servicing
> using INTx interrupts (if supported).
>
> but they don't say anything about what happens when MSI or MSI-X is
> *enabled*.
>
> I think a better citation is PCIe r6.0, sec 6.1.4.3, which says:
>
> While enabled for MSI or MSI-X operation, a Function is prohibited
> from using INTx interrupts (if implemented) to request service (MSI,
> MSI-X, and INTx are mutually exclusive).
>
> Can you please update the comment in the code and this commit log to
> cite PCIe r6.0, sec 6.1.4.3 instead, and to clarify that these Tegra
> devices always use INTx for PME and AER, even when MSI/MSI-X is
> enabled?
Sure. I would fix this in the next patch set.
>
> Why do these Tegra quirks use DECLARE_PCI_FIXUP_CLASS_EARLY() instead
> of just DECLARE_PCI_FIXUP_EARLY()? quirk_al_msi_disable() uses the
> _CLASS version because the same Device ID is used for non-Root Port
> devices. Is the same true here, or could these use
> DECLARE_PCI_FIXUP_EARLY()?
Tegra's PCIe controllers are also dual mode controllers and MSI works
just fine when they operate in the endpoint mode configuration.
>
> There are many quirks that disable MSI, and they're a mixture of EARLY
> and FINAL. They should probably all be the same.
>
> [1] https://git.kernel.org/linus/8c7e96d3fe75
>
>> Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
>> ---
>> drivers/pci/quirks.c | 9 +++++++++
>> 1 file changed, 9 insertions(+)
>>
>> diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
>> index d2dd6a6cda60..3ac5c45e61a1 100644
>> --- a/drivers/pci/quirks.c
>> +++ b/drivers/pci/quirks.c
>> @@ -2747,6 +2747,15 @@ DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x10e5,
>> DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x10e6,
>> PCI_CLASS_BRIDGE_PCI, 8,
>> pci_quirk_nvidia_tegra_disable_rp_msi);
>> +DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x229a,
>> + PCI_CLASS_BRIDGE_PCI, 8,
>> + pci_quirk_nvidia_tegra_disable_rp_msi);
>> +DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x229c,
>> + PCI_CLASS_BRIDGE_PCI, 8,
>> + pci_quirk_nvidia_tegra_disable_rp_msi);
>> +DECLARE_PCI_FIXUP_CLASS_EARLY(PCI_VENDOR_ID_NVIDIA, 0x229e,
>> + PCI_CLASS_BRIDGE_PCI, 8,
>> + pci_quirk_nvidia_tegra_disable_rp_msi);
>>
>> /*
>> * Some versions of the MCP55 bridge from Nvidia have a legacy IRQ routing
>> --
>> 2.17.1
>>
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
^ permalink raw reply [flat|nested] 46+ messages in thread
end of thread, other threads:[~2022-04-23 7:23 UTC | newest]
Thread overview: 46+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-02-05 16:21 [PATCH V1 00/10] PCI: tegra: Add Tegra234 PCIe support Vidya Sagar
2022-02-05 16:21 ` Vidya Sagar
2022-02-05 16:21 ` [PATCH V1 01/10] dt-bindings: Add Tegra234 PCIe clocks and resets Vidya Sagar
2022-02-05 16:21 ` Vidya Sagar
2022-02-11 14:51 ` Rob Herring
2022-02-11 14:51 ` Rob Herring
2022-02-05 16:21 ` [PATCH V1 02/10] dt-bindings: power: Add Tegra234 PCIe power domains Vidya Sagar
2022-02-05 16:21 ` Vidya Sagar
2022-02-11 14:52 ` Rob Herring
2022-02-11 14:52 ` Rob Herring
2022-02-05 16:21 ` [PATCH V1 03/10] dt-bindings: memory: Add Tegra234 PCIe memory Vidya Sagar
2022-02-05 16:21 ` Vidya Sagar
2022-02-06 11:33 ` Krzysztof Kozlowski
2022-02-06 11:33 ` Krzysztof Kozlowski
2022-02-24 19:04 ` Thierry Reding
2022-02-24 19:04 ` Thierry Reding
2022-02-11 14:53 ` Rob Herring
2022-02-11 14:53 ` Rob Herring
2022-02-05 16:21 ` [PATCH V1 04/10] dt-bindings: PHY: P2U: Add support for Tegra234 P2U block Vidya Sagar
2022-02-05 16:21 ` Vidya Sagar
2022-02-07 6:47 ` Raul Tambre
2022-02-07 6:47 ` Raul Tambre
2022-02-11 14:55 ` Rob Herring
2022-02-11 14:55 ` Rob Herring
2022-02-05 16:21 ` [PATCH V1 05/10] dt-bindings: PCI: tegra: Add device tree support for Tegra234 Vidya Sagar
2022-02-05 16:21 ` Vidya Sagar
2022-02-11 14:57 ` Rob Herring
2022-02-11 14:57 ` Rob Herring
2022-02-05 16:21 ` [PATCH V1 06/10] arm64: tegra: Add P2U and PCIe controller nodes to Tegra234 DT Vidya Sagar
2022-02-05 16:21 ` Vidya Sagar
2022-02-05 16:21 ` [PATCH V1 07/10] arm64: tegra: Enable PCIe slots in P3737-0000 board Vidya Sagar
2022-02-05 16:21 ` Vidya Sagar
2022-02-06 11:29 ` Krzysztof Kozlowski
2022-02-06 11:29 ` Krzysztof Kozlowski
2022-02-05 16:21 ` [PATCH V1 08/10] phy: tegra: Add PCIe PIPE2UPHY support for Tegra234 Vidya Sagar
2022-02-05 16:21 ` Vidya Sagar
2022-02-05 16:21 ` [PATCH V1 09/10] PCI: Disable MSI for Tegra234 root ports Vidya Sagar
2022-02-05 16:21 ` Vidya Sagar
2022-02-07 17:36 ` Bjorn Helgaas
2022-02-07 17:36 ` Bjorn Helgaas
2022-04-23 7:22 ` Vidya Sagar
2022-04-23 7:22 ` Vidya Sagar
2022-02-05 16:21 ` [PATCH V1 10/10] PCI: tegra: Add Tegra234 PCIe support Vidya Sagar
2022-02-05 16:21 ` Vidya Sagar
2022-02-07 18:19 ` Bjorn Helgaas
2022-02-07 18:19 ` Bjorn Helgaas
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.