Skip to contentSkip to conversation

ARM: dts: imx51: Fix memory node duplication

PATCH In review Sasha Levin 6 participants 220 messages Nov 22, 2019 – Dec 18, 2019 Page 1 of 2

Subsystems (36)
lore
Share & tools
lore

Sasha Levin

author maintainer
#0 lore raw

PATCH 8/219 ARM: dts: imx51: Fix memory node duplication

From: Fabio Estevam <[email protected]>

[ Upstream commit 6a9681168b83c62abfa457c709f2f4b126bd6b92 ]

Boards based on imx51 have duplicate memory nodes:

- One coming from the board dts file: memory@

- One coming from the imx51.dtsi file.

Fix the duplication by removing the memory node from the dtsi file and by adding 'device_type = "memory";' in the board dts.

---
 arch/arm/boot/dts/imx51-apf51.dts                 | 1 +
 arch/arm/boot/dts/imx51-babbage.dts               | 1 +
 arch/arm/boot/dts/imx51-digi-connectcore-som.dtsi | 1 +
 arch/arm/boot/dts/imx51-eukrea-cpuimx51.dtsi      | 1 +
 arch/arm/boot/dts/imx51-ts4800.dts                | 1 +
 arch/arm/boot/dts/imx51-zii-rdu1.dts              | 1 +
 arch/arm/boot/dts/imx51-zii-scu2-mezz.dts         | 1 +
 arch/arm/boot/dts/imx51-zii-scu3-esb.dts          | 1 +
 arch/arm/boot/dts/imx51.dtsi                      | 2 --
 9 files changed, 8 insertions(+), 2 deletions(-)
arch/arm/boot/dts/imx51-apf51.dts +1 -0
index 79d80036f74de..1eddf2908b3f2 100644
@@ -22,6 +22,7 @@
 	compatible = "armadeus,imx51-apf51", "fsl,imx51";
 
 	memory@90000000 {
+		device_type = "memory";
 		reg = <0x90000000 0x20000000>;
 	};
 
arch/arm/boot/dts/imx51-babbage.dts +1 -0
index ba60b0cb3cc13..99191466a8085 100644
@@ -15,6 +15,7 @@
 	};
 
 	memory@90000000 {
+		device_type = "memory";
 		reg = <0x90000000 0x20000000>;
 	};
 
arch/arm/boot/dts/imx51-digi-connectcore-som.dtsi +1 -0
index 5761a66e8a0d3..82d8df097ef1f 100644
@@ -17,6 +17,7 @@
 	compatible = "digi,connectcore-ccxmx51-som", "fsl,imx51";
 
 	memory@90000000 {
+		device_type = "memory";
 		reg = <0x90000000 0x08000000>;
 	};
 };
arch/arm/boot/dts/imx51-eukrea-cpuimx51.dtsi +1 -0
index f8902a338e49a..2e3125391bc49 100644
@@ -23,6 +23,7 @@
 	compatible = "eukrea,cpuimx51", "fsl,imx51";
 
 	memory@90000000 {
+		device_type = "memory";
 		reg = <0x90000000 0x10000000>; /* 256M */
 	};
 };
arch/arm/boot/dts/imx51-ts4800.dts +1 -0
index 39eb067904c3d..4344632f79400 100644
@@ -18,6 +18,7 @@
 	};
 
 	memory@90000000 {
+		device_type = "memory";
 		reg = <0x90000000 0x10000000>;
 	};
 
arch/arm/boot/dts/imx51-zii-rdu1.dts +1 -0
index 6e80254c4562a..d0e0eb9c9adfa 100644
@@ -53,6 +53,7 @@
 
 	/* Will be filled by the bootloader */
 	memory@90000000 {
+		device_type = "memory";
 		reg = <0x90000000 0>;
 	};
 
arch/arm/boot/dts/imx51-zii-scu2-mezz.dts +1 -0
index 26cf08549df40..f5b2d768fe47f 100644
@@ -18,6 +18,7 @@
 
 	/* Will be filled by the bootloader */
 	memory@90000000 {
+		device_type = "memory";
 		reg = <0x90000000 0>;
 	};
 
arch/arm/boot/dts/imx51-zii-scu3-esb.dts +1 -0
index e6ebac8f43e4f..ad90d66ccca6c 100644
@@ -18,6 +18,7 @@
 
 	/* Will be filled by the bootloader */
 	memory@90000000 {
+		device_type = "memory";
 		reg = <0x90000000 0>;
 	};
 
arch/arm/boot/dts/imx51.dtsi +0 -2
index ef2abc0978439..81f60c96a2e41 100644
@@ -16,10 +16,8 @@
 	 * The decompressor and also some bootloaders rely on a
 	 * pre-existing /chosen node to be available to insert the
 	 * command line and merge other ATAGS info.
-	 * Also for U-Boot there must be a pre-existing /memory node.
 	 */
 	chosen {};
-	memory { device_type = "memory"; };
 
 	aliases {
 		ethernet0 = &fec;
Signature · 1 line
2.20.1
author maintainer
#1 lore raw

PATCH 9/219 ARM: dts: imx53: Fix memory node duplication

From: Fabio Estevam <[email protected]>

[ Upstream commit e8fd17b900a4a1e3a8bef7b44727cbad35db05a7 ]

Boards based on imx53 have duplicate memory nodes:

- One coming from the board dts file: memory@

- One coming from the imx53.dtsi file.

Fix the duplication by removing the memory node from the dtsi file and by adding 'device_type = "memory";' in the board dts.

---
 arch/arm/boot/dts/imx53-ard.dts         | 1 +
 arch/arm/boot/dts/imx53-cx9020.dts      | 1 +
 arch/arm/boot/dts/imx53-m53.dtsi        | 1 +
 arch/arm/boot/dts/imx53-qsb-common.dtsi | 1 +
 arch/arm/boot/dts/imx53-smd.dts         | 1 +
 arch/arm/boot/dts/imx53-tqma53.dtsi     | 1 +
 arch/arm/boot/dts/imx53-tx53.dtsi       | 1 +
 arch/arm/boot/dts/imx53-usbarmory.dts   | 1 +
 arch/arm/boot/dts/imx53.dtsi            | 2 --
 9 files changed, 8 insertions(+), 2 deletions(-)
arch/arm/boot/dts/imx53-ard.dts +1 -0
index 117bd002dd1d1..7d5a48250f867 100644
@@ -19,6 +19,7 @@
 	compatible = "fsl,imx53-ard", "fsl,imx53";
 
 	memory@70000000 {
+		device_type = "memory";
 		reg = <0x70000000 0x40000000>;
 	};
 
arch/arm/boot/dts/imx53-cx9020.dts +1 -0
index cf70ebc4399a2..c875e23ee45fb 100644
@@ -22,6 +22,7 @@
 	};
 
 	memory@70000000 {
+		device_type = "memory";
 		reg = <0x70000000 0x20000000>,
 		      <0xb0000000 0x20000000>;
 	};
arch/arm/boot/dts/imx53-m53.dtsi +1 -0
index ce45f08e30514..db2e5bce9b6a1 100644
@@ -16,6 +16,7 @@
 	compatible = "aries,imx53-m53", "denx,imx53-m53", "fsl,imx53";
 
 	memory@70000000 {
+		device_type = "memory";
 		reg = <0x70000000 0x20000000>,
 		      <0xb0000000 0x20000000>;
 	};
arch/arm/boot/dts/imx53-qsb-common.dtsi +1 -0
index 50dde84b72ed7..f00dda334976a 100644
@@ -11,6 +11,7 @@
 	};
 
 	memory@70000000 {
+		device_type = "memory";
 		reg = <0x70000000 0x20000000>,
 		      <0xb0000000 0x20000000>;
 	};
arch/arm/boot/dts/imx53-smd.dts +1 -0
index 462071c9ddd73..09071ca11c6cf 100644
@@ -12,6 +12,7 @@
 	compatible = "fsl,imx53-smd", "fsl,imx53";
 
 	memory@70000000 {
+		device_type = "memory";
 		reg = <0x70000000 0x40000000>;
 	};
 
arch/arm/boot/dts/imx53-tqma53.dtsi +1 -0
index a72b8981fc3bd..c77d58f06c949 100644
@@ -17,6 +17,7 @@
 	compatible = "tq,tqma53", "fsl,imx53";
 
 	memory@70000000 {
+		device_type = "memory";
 		reg = <0x70000000 0x40000000>; /* Up to 1GiB */
 	};
 
arch/arm/boot/dts/imx53-tx53.dtsi +1 -0
index 54cf3e67069a9..4ab135906949f 100644
@@ -51,6 +51,7 @@
 
 	/* Will be filled by the bootloader */
 	memory@70000000 {
+		device_type = "memory";
 		reg = <0x70000000 0>;
 	};
 
arch/arm/boot/dts/imx53-usbarmory.dts +1 -0
index f6268d0ded296..ee6263d1c2d3d 100644
@@ -58,6 +58,7 @@
 	};
 
 	memory@70000000 {
+		device_type = "memory";
 		reg = <0x70000000 0x20000000>;
 	};
 
arch/arm/boot/dts/imx53.dtsi +0 -2
index b6b0818343c4e..8accbe16b7584 100644
@@ -23,10 +23,8 @@
 	 * The decompressor and also some bootloaders rely on a
 	 * pre-existing /chosen node to be available to insert the
 	 * command line and merge other ATAGS info.
-	 * Also for U-Boot there must be a pre-existing /memory node.
 	 */
 	chosen {};
-	memory { device_type = "memory"; };
 
 	aliases {
 		ethernet0 = &fec;
Signature · 1 line
2.20.1
author maintainer
#2 lore raw

PATCH 10/219 ARM: dts: imx31: Fix memory node duplication

In reply to Sasha Levin

From: Fabio Estevam <[email protected]>

[ Upstream commit 013d37e4707e24c7b9bc3fc55aeda55ce9c2b262 ]

Boards based on imx31 have duplicate memory nodes:

- One coming from the board dts file: memory@

- One coming from the imx31.dtsi file.

Fix the duplication by removing the memory node from the dtsi file and by adding 'device_type = "memory";' in the board dts.

---
 arch/arm/boot/dts/imx31-bug.dts  | 1 +
 arch/arm/boot/dts/imx31-lite.dts | 1 +
 arch/arm/boot/dts/imx31.dtsi     | 2 --
 3 files changed, 2 insertions(+), 2 deletions(-)
arch/arm/boot/dts/imx31-bug.dts +1 -0
index 6ee4ff8e4e8f0..9eb960cc02cc5 100644
@@ -17,6 +17,7 @@
 	compatible = "buglabs,imx31-bug", "fsl,imx31";
 
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x8000000>; /* 128M */
 	};
 };
arch/arm/boot/dts/imx31-lite.dts +1 -0
index db52ddccabc33..d17abdfb6330c 100644
@@ -18,6 +18,7 @@
 	};
 
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x8000000>;
 	};
 
arch/arm/boot/dts/imx31.dtsi +0 -2
index ca1419ca303c3..2fc64d2c7c88e 100644
@@ -10,10 +10,8 @@
 	 * The decompressor and also some bootloaders rely on a
 	 * pre-existing /chosen node to be available to insert the
 	 * command line and merge other ATAGS info.
-	 * Also for U-Boot there must be a pre-existing /memory node.
 	 */
 	chosen {};
-	memory { device_type = "memory"; };
 
 	aliases {
 		gpio0 = &gpio1;
Signature · 1 line
2.20.1
author maintainer
#3 lore raw

PATCH 11/219 ARM: dts: imx35: Fix memory node duplication

In reply to Sasha Levin

From: Fabio Estevam <[email protected]>

[ Upstream commit 8721610a6c2b8c42fc57819d8c3bfbb9166f95a3 ]

Boards based on imx35 have duplicate memory nodes:

- One coming from the board dts file: memory@

- One coming from the imx35.dtsi file.

Fix the duplication by removing the memory node from the dtsi file and by adding 'device_type = "memory";' in the board dts.

---
 arch/arm/boot/dts/imx35-eukrea-cpuimx35.dtsi | 1 +
 arch/arm/boot/dts/imx35-pdk.dts              | 1 +
 arch/arm/boot/dts/imx35.dtsi                 | 2 --
 3 files changed, 2 insertions(+), 2 deletions(-)
arch/arm/boot/dts/imx35-eukrea-cpuimx35.dtsi +1 -0
index ba39d938f2891..5f8a47a9fcd40 100644
@@ -18,6 +18,7 @@
 	compatible = "eukrea,cpuimx35", "fsl,imx35";
 
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x8000000>; /* 128M */
 	};
 };
arch/arm/boot/dts/imx35-pdk.dts +1 -0
index df613e88fd2c1..ddce0a844758b 100644
@@ -11,6 +11,7 @@
 	compatible = "fsl,imx35-pdk", "fsl,imx35";
 
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x8000000>,
 		      <0x90000000 0x8000000>;
 	};
arch/arm/boot/dts/imx35.dtsi +0 -2
index 1c50b785cad47..b36b97b655dda 100644
@@ -13,10 +13,8 @@
 	 * The decompressor and also some bootloaders rely on a
 	 * pre-existing /chosen node to be available to insert the
 	 * command line and merge other ATAGS info.
-	 * Also for U-Boot there must be a pre-existing /memory node.
 	 */
 	chosen {};
-	memory { device_type = "memory"; };
 
 	aliases {
 		ethernet0 = &fec;
Signature · 1 line
2.20.1
author maintainer
#4 lore raw

PATCH 12/219 ARM: dts: imx7: Fix memory node duplication

In reply to Sasha Levin

From: Fabio Estevam <[email protected]>

[ Upstream commit 29988e867cb17de7119e971f9acfad2c3fccdb47 ]

Boards based on imx7 have duplicate memory nodes:

- One coming from the board dts file: memory@

- One coming from the imx7s.dtsi file.

Fix the duplication by removing the memory node from the dtsi file and by adding 'device_type = "memory";' in the board dts.

---
 arch/arm/boot/dts/imx7d-cl-som-imx7.dts   | 3 ++-
 arch/arm/boot/dts/imx7d-colibri-emmc.dtsi | 1 +
 arch/arm/boot/dts/imx7d-colibri.dtsi      | 1 +
 arch/arm/boot/dts/imx7d-nitrogen7.dts     | 1 +
 arch/arm/boot/dts/imx7d-pico.dtsi         | 1 +
 arch/arm/boot/dts/imx7d-sdb.dts           | 1 +
 arch/arm/boot/dts/imx7s-colibri.dtsi      | 1 +
 arch/arm/boot/dts/imx7s-warp.dts          | 1 +
 arch/arm/boot/dts/imx7s.dtsi              | 2 --
 9 files changed, 9 insertions(+), 3 deletions(-)
arch/arm/boot/dts/imx7d-cl-som-imx7.dts +2 -1
index 584418f517a88..62d5e9a4a7818 100644
@@ -19,6 +19,7 @@
 	compatible = "compulab,cl-som-imx7", "fsl,imx7d";
 
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x10000000>; /* 256 MB - minimal configuration */
 	};
 
@@ -284,4 +285,4 @@
 			MX7D_PAD_LPSR_GPIO1_IO05__GPIO1_IO5	0x14 /* OTG PWREN */
 		>;
 	};
-};
\ No newline at end of file
+};
arch/arm/boot/dts/imx7d-colibri-emmc.dtsi +1 -0
index 04d24ee17b142..898f4b8d7421f 100644
@@ -8,6 +8,7 @@
 
 / {
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x40000000>;
 	};
 };
arch/arm/boot/dts/imx7d-colibri.dtsi +1 -0
index d9f8fb69511b6..e2e327f437e35 100644
@@ -45,6 +45,7 @@
 
 / {
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x20000000>;
 	};
 };
arch/arm/boot/dts/imx7d-nitrogen7.dts +1 -0
index 177d21fdeb288..6b4acea1ef795 100644
@@ -12,6 +12,7 @@
 	compatible = "boundary,imx7d-nitrogen7", "fsl,imx7d";
 
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x40000000>;
 	};
 
arch/arm/boot/dts/imx7d-pico.dtsi +1 -0
index f27b3849d3ff3..934a019f341e4 100644
@@ -49,6 +49,7 @@
 	compatible = "technexion,imx7d-pico", "fsl,imx7d";
 
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x80000000>;
 	};
 
arch/arm/boot/dts/imx7d-sdb.dts +1 -0
index c9b3c60b0eb22..317f1bcc56e2a 100644
@@ -15,6 +15,7 @@
 	};
 
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x80000000>;
 	};
 
arch/arm/boot/dts/imx7s-colibri.dtsi +1 -0
index fe8344cee8641..1fb1ec5d3d707 100644
@@ -45,6 +45,7 @@
 
 / {
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x10000000>;
 	};
 };
arch/arm/boot/dts/imx7s-warp.dts +1 -0
index fa390da636de7..97d5c711eb0ca 100644
@@ -51,6 +51,7 @@
 	compatible = "warp,imx7s-warp", "fsl,imx7s";
 
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x20000000>;
 	};
 
arch/arm/boot/dts/imx7s.dtsi +0 -2
index 90f5bdfa9b3ce..7eaf96b425bed 100644
@@ -17,10 +17,8 @@
 	 * The decompressor and also some bootloaders rely on a
 	 * pre-existing /chosen node to be available to insert the
 	 * command line and merge other ATAGS info.
-	 * Also for U-Boot there must be a pre-existing /memory node.
 	 */
 	chosen {};
-	memory { device_type = "memory"; };
 
 	aliases {
 		gpio0 = &gpio1;
Signature · 1 line
2.20.1
author maintainer
#5 lore raw

PATCH 13/219 ARM: dts: imx6ul: Fix memory node duplication

In reply to Sasha Levin

From: Fabio Estevam <[email protected]>

[ Upstream commit 750d8df6e7b269b828f66631a1d39ea027afc92a ]

Boards based on imx6ul have duplicate memory nodes:

- One coming from the board dts file: memory@

- One coming from the imx6ul.dtsi file.

Fix the duplication by removing the memory node from the dtsi file and by adding 'device_type = "memory";' in the board dts.

---
 arch/arm/boot/dts/imx6ul-14x14-evk.dtsi        | 1 +
 arch/arm/boot/dts/imx6ul-geam.dts              | 1 +
 arch/arm/boot/dts/imx6ul-isiot.dtsi            | 1 +
 arch/arm/boot/dts/imx6ul-litesom.dtsi          | 1 +
 arch/arm/boot/dts/imx6ul-opos6ul.dtsi          | 1 +
 arch/arm/boot/dts/imx6ul-pico-hobbit.dts       | 1 +
 arch/arm/boot/dts/imx6ul-tx6ul.dtsi            | 1 +
 arch/arm/boot/dts/imx6ul.dtsi                  | 2 --
 arch/arm/boot/dts/imx6ull-colibri-nonwifi.dtsi | 1 +
 arch/arm/boot/dts/imx6ull-colibri-wifi.dtsi    | 1 +
 10 files changed, 9 insertions(+), 2 deletions(-)
arch/arm/boot/dts/imx6ul-14x14-evk.dtsi +1 -0
index 32a07232c0345..8180211265592 100644
@@ -12,6 +12,7 @@
 	};
 
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x20000000>;
 	};
 
arch/arm/boot/dts/imx6ul-geam.dts +1 -0
index d81d20f8fc8dd..85cfad080f15c 100644
@@ -51,6 +51,7 @@
 	compatible = "engicam,imx6ul-geam", "fsl,imx6ul";
 
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x08000000>;
 	};
 
arch/arm/boot/dts/imx6ul-isiot.dtsi +1 -0
index cd99285511544..1cb52744f58ad 100644
@@ -46,6 +46,7 @@
 
 / {
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x20000000>;
 	};
 
arch/arm/boot/dts/imx6ul-litesom.dtsi +1 -0
index 8f775f6974d1c..8d6893210842b 100644
@@ -48,6 +48,7 @@
 	compatible = "grinn,imx6ul-litesom", "fsl,imx6ul";
 
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x20000000>;
 	};
 };
arch/arm/boot/dts/imx6ul-opos6ul.dtsi +1 -0
index a031bee311df4..cf7faf4b9c47e 100644
@@ -49,6 +49,7 @@
 
 / {
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0>; /* will be filled by U-Boot */
 	};
 
arch/arm/boot/dts/imx6ul-pico-hobbit.dts +1 -0
index 0c09420f99512..797262d2f27fd 100644
@@ -53,6 +53,7 @@
 
 	/* Will be filled by the bootloader */
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0>;
 	};
 
arch/arm/boot/dts/imx6ul-tx6ul.dtsi +1 -0
index 02b5ba42cd591..bb6dbfd5546b4 100644
@@ -71,6 +71,7 @@
 	};
 
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0>; /* will be filled by U-Boot */
 	};
 
arch/arm/boot/dts/imx6ul.dtsi +0 -2
index 336cdead3da54..50834a43e5fb2 100644
@@ -15,10 +15,8 @@
 	 * The decompressor and also some bootloaders rely on a
 	 * pre-existing /chosen node to be available to insert the
 	 * command line and merge other ATAGS info.
-	 * Also for U-Boot there must be a pre-existing /memory node.
 	 */
 	chosen {};
-	memory { device_type = "memory"; };
 
 	aliases {
 		ethernet0 = &fec1;
arch/arm/boot/dts/imx6ull-colibri-nonwifi.dtsi +1 -0
index 10ab4697950f5..fb213bec46543 100644
@@ -7,6 +7,7 @@
 
 / {
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x10000000>;
 	};
 };
arch/arm/boot/dts/imx6ull-colibri-wifi.dtsi +1 -0
index 183193e8580dd..038d8c90f6dfe 100644
@@ -7,6 +7,7 @@
 
 / {
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x20000000>;
 	};
 
Signature · 7 lines
2.20.1


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
author maintainer
#6 lore raw

PATCH 14/219 ARM: dts: imx6sx: Fix memory node duplication

In reply to Sasha Levin

From: Fabio Estevam <[email protected]>

[ Upstream commit 216f35fedd8688c8b654ebfbad18c6e64713fad7 ]

Boards based on imx6sx have duplicate memory nodes:

- One coming from the board dts file: memory@

- One coming from the imx6sx.dtsi file.

Fix the duplication by removing the memory node from the dtsi file and by adding 'device_type = "memory";' in the board dts.

---
 arch/arm/boot/dts/imx6sx-nitrogen6sx.dts         | 1 +
 arch/arm/boot/dts/imx6sx-sabreauto.dts           | 1 +
 arch/arm/boot/dts/imx6sx-sdb.dtsi                | 1 +
 arch/arm/boot/dts/imx6sx-softing-vining-2000.dts | 1 +
 arch/arm/boot/dts/imx6sx-udoo-neo-basic.dts      | 1 +
 arch/arm/boot/dts/imx6sx-udoo-neo-extended.dts   | 1 +
 arch/arm/boot/dts/imx6sx-udoo-neo-full.dts       | 1 +
 arch/arm/boot/dts/imx6sx.dtsi                    | 2 --
 8 files changed, 7 insertions(+), 2 deletions(-)
arch/arm/boot/dts/imx6sx-nitrogen6sx.dts +1 -0
index adb5cc7d8ce2f..832b5c5d7441a 100644
@@ -12,6 +12,7 @@
 	compatible = "boundary,imx6sx-nitrogen6sx", "fsl,imx6sx";
 
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x40000000>;
 	};
 
arch/arm/boot/dts/imx6sx-sabreauto.dts +1 -0
index 841a27f3198ff..48aede543612b 100644
@@ -11,6 +11,7 @@
 	compatible = "fsl,imx6sx-sabreauto", "fsl,imx6sx";
 
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x80000000>;
 	};
 
arch/arm/boot/dts/imx6sx-sdb.dtsi +1 -0
index f8f31872fa144..7f5ede5ca4c30 100644
@@ -21,6 +21,7 @@
 	};
 
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x40000000>;
 	};
 
arch/arm/boot/dts/imx6sx-softing-vining-2000.dts +1 -0
index 252175b592475..2bc51623a8060 100644
@@ -21,6 +21,7 @@
 	};
 
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x40000000>;
 	};
 
arch/arm/boot/dts/imx6sx-udoo-neo-basic.dts +1 -0
index 40ccdf43dffc5..db0feb9b9f5d7 100644
@@ -49,6 +49,7 @@
 	compatible = "udoo,neobasic", "fsl,imx6sx";
 
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x20000000>;
 	};
 };
arch/arm/boot/dts/imx6sx-udoo-neo-extended.dts +1 -0
index 42bfc8f8f7f6b..5c7a2bb9141cb 100644
@@ -49,6 +49,7 @@
 	compatible = "udoo,neoextended", "fsl,imx6sx";
 
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x40000000>;
 	};
 };
arch/arm/boot/dts/imx6sx-udoo-neo-full.dts +1 -0
index c84c877f09d49..13dfe2afaba56 100644
@@ -49,6 +49,7 @@
 	compatible = "udoo,neofull", "fsl,imx6sx";
 
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x40000000>;
 	};
 };
arch/arm/boot/dts/imx6sx.dtsi +0 -2
index 7b62e6fb47ebe..ae0728df542e9 100644
@@ -15,10 +15,8 @@
 	 * The decompressor and also some bootloaders rely on a
 	 * pre-existing /chosen node to be available to insert the
 	 * command line and merge other ATAGS info.
-	 * Also for U-Boot there must be a pre-existing /memory node.
 	 */
 	chosen {};
-	memory { device_type = "memory"; };
 
 	aliases {
 		can0 = &flexcan1;
Signature · 1 line
2.20.1
author maintainer
#7 lore raw

PATCH 15/219 ARM: dts: imx6sl: Fix memory node duplication

In reply to Sasha Levin

From: Fabio Estevam <[email protected]>

[ Upstream commit 7fa8ab65ee15e386558ac5e971004712da91e2dd ]

Boards based on imx6sl have duplicate memory nodes:

- One coming from the board dts file: memory@

- One coming from the imx6sl.dtsi file.

Fix the duplication by removing the memory node from the dtsi file and by adding 'device_type = "memory";' in the board dts.

---
 arch/arm/boot/dts/imx6sl-evk.dts  | 1 +
 arch/arm/boot/dts/imx6sl-warp.dts | 1 +
 arch/arm/boot/dts/imx6sl.dtsi     | 2 --
 arch/arm/boot/dts/imx6sll-evk.dts | 1 +
 4 files changed, 3 insertions(+), 2 deletions(-)
arch/arm/boot/dts/imx6sl-evk.dts +1 -0
index 679b4482ab13a..f7a48e4622e1b 100644
@@ -17,6 +17,7 @@
 	};
 
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x40000000>;
 	};
 
arch/arm/boot/dts/imx6sl-warp.dts +1 -0
index 404e602e67813..408da704c459b 100644
@@ -55,6 +55,7 @@
 	compatible = "warp,imx6sl-warp", "fsl,imx6sl";
 
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x20000000>;
 	};
 
arch/arm/boot/dts/imx6sl.dtsi +0 -2
index 2fa88c6f18820..55d1872aa81a8 100644
@@ -13,10 +13,8 @@
 	 * The decompressor and also some bootloaders rely on a
 	 * pre-existing /chosen node to be available to insert the
 	 * command line and merge other ATAGS info.
-	 * Also for U-Boot there must be a pre-existing /memory node.
 	 */
 	chosen {};
-	memory { device_type = "memory"; };
 
 	aliases {
 		ethernet0 = &fec;
arch/arm/boot/dts/imx6sll-evk.dts +1 -0
index c8e115564ba2c..0c2406ac8a638 100644
@@ -20,6 +20,7 @@
 	};
 
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x80000000>;
 	};
 
Signature · 1 line
2.20.1
author maintainer
#8 lore raw

PATCH 16/219 ARM: dts: imx50: Fix memory node duplication

In reply to Sasha Levin

From: Fabio Estevam <[email protected]>

[ Upstream commit aab5e3ea95b958cf22a24e756a84e635bdb081c1 ]

imx50-evk has duplicate memory nodes:

- One coming from the board dts file: memory@

- One coming from the imx50.dtsi file.

Fix the duplication by removing the memory node from the dtsi file and by adding 'device_type = "memory";' in the board dts.

---
 arch/arm/boot/dts/imx50-evk.dts | 1 +
 arch/arm/boot/dts/imx50.dtsi    | 2 --
 2 files changed, 1 insertion(+), 2 deletions(-)
arch/arm/boot/dts/imx50-evk.dts +1 -0
index 682a99783ee69..a25da415cb02e 100644
@@ -12,6 +12,7 @@
 	compatible = "fsl,imx50-evk", "fsl,imx50";
 
 	memory@70000000 {
+		device_type = "memory";
 		reg = <0x70000000 0x80000000>;
 	};
 };
arch/arm/boot/dts/imx50.dtsi +0 -2
index ab522c2da6df6..9e9e92acceb27 100644
@@ -22,10 +22,8 @@
 	 * The decompressor and also some bootloaders rely on a
 	 * pre-existing /chosen node to be available to insert the
 	 * command line and merge other ATAGS info.
-	 * Also for U-Boot there must be a pre-existing /memory node.
 	 */
 	chosen {};
-	memory { device_type = "memory"; };
 
 	aliases {
 		ethernet0 = &fec;
Signature · 1 line
2.20.1
author maintainer
#9 lore raw

PATCH 17/219 ARM: dts: imx23: Fix memory node duplication

In reply to Sasha Levin

From: Fabio Estevam <[email protected]>

[ Upstream commit b629e83520fafe6f4c2f3e8c88c78a496fc4987c ]

Boards based on imx23 have duplicate memory nodes:

- One coming from the board dts file: memory@

- One coming from the imx23.dtsi file.

Fix the duplication by removing the memory node from the dtsi file and by adding 'device_type = "memory";' in the board dts.

---
 arch/arm/boot/dts/imx23-evk.dts           | 1 +
 arch/arm/boot/dts/imx23-olinuxino.dts     | 1 +
 arch/arm/boot/dts/imx23-sansa.dts         | 1 +
 arch/arm/boot/dts/imx23-stmp378x_devb.dts | 1 +
 arch/arm/boot/dts/imx23-xfi3.dts          | 1 +
 arch/arm/boot/dts/imx23.dtsi              | 2 --
 6 files changed, 5 insertions(+), 2 deletions(-)
arch/arm/boot/dts/imx23-evk.dts +1 -0
index ad2ae25b7b4db..aca27aa2d44bd 100644
@@ -10,6 +10,7 @@
 	compatible = "fsl,imx23-evk", "fsl,imx23";
 
 	memory@40000000 {
+		device_type = "memory";
 		reg = <0x40000000 0x08000000>;
 	};
 
arch/arm/boot/dts/imx23-olinuxino.dts +1 -0
index e9351774c6199..109f51603d45e 100644
@@ -20,6 +20,7 @@
 	compatible = "olimex,imx23-olinuxino", "fsl,imx23";
 
 	memory@40000000 {
+		device_type = "memory";
 		reg = <0x40000000 0x04000000>;
 	};
 
arch/arm/boot/dts/imx23-sansa.dts +1 -0
index 67de7863ad795..fa22fd9b24129 100644
@@ -50,6 +50,7 @@
 	compatible = "sandisk,sansa_fuze_plus", "fsl,imx23";
 
 	memory@40000000 {
+		device_type = "memory";
 		reg = <0x40000000 0x04000000>;
 	};
 
arch/arm/boot/dts/imx23-stmp378x_devb.dts +1 -0
index 95c7b918f6d60..aab029349420d 100644
@@ -17,6 +17,7 @@
 	compatible = "fsl,stmp378x-devb", "fsl,imx23";
 
 	memory@40000000 {
+		device_type = "memory";
 		reg = <0x40000000 0x04000000>;
 	};
 
arch/arm/boot/dts/imx23-xfi3.dts +1 -0
index 9616e500b9961..2b5df8dfd3ff3 100644
@@ -49,6 +49,7 @@
 	compatible = "creative,x-fi3", "fsl,imx23";
 
 	memory@40000000 {
+		device_type = "memory";
 		reg = <0x40000000 0x04000000>;
 	};
 
arch/arm/boot/dts/imx23.dtsi +0 -2
index 71bfd2b15609a..aaaa987d8eff9 100644
@@ -13,10 +13,8 @@
 	 * The decompressor and also some bootloaders rely on a
 	 * pre-existing /chosen node to be available to insert the
 	 * command line and merge other ATAGS info.
-	 * Also for U-Boot there must be a pre-existing /memory node.
 	 */
 	chosen {};
-	memory { device_type = "memory"; };
 
 	aliases {
 		gpio0 = &gpio0;
Signature · 7 lines
2.20.1


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
author maintainer
#10 lore raw

PATCH 18/219 ARM: dts: imx1: Fix memory node duplication

In reply to Sasha Levin

From: Fabio Estevam <[email protected]>

[ Upstream commit 62864d5665c4fc636d3021f829b3ac00fa058e30 ]

Boards based on imx1 have duplicate memory nodes:

- One coming from the board dts file: memory@

- One coming from the imx1.dtsi file.

Fix the duplication by removing the memory node from the dtsi file and by adding 'device_type = "memory";' in the board dts.

---
 arch/arm/boot/dts/imx1-ads.dts     | 1 +
 arch/arm/boot/dts/imx1-apf9328.dts | 1 +
 arch/arm/boot/dts/imx1.dtsi        | 2 --
 3 files changed, 2 insertions(+), 2 deletions(-)
arch/arm/boot/dts/imx1-ads.dts +1 -0
index a1d81badb5c8a..119b19ba53b6d 100644
@@ -21,6 +21,7 @@
 	};
 
 	memory@8000000 {
+		device_type = "memory";
 		reg = <0x08000000 0x04000000>;
 	};
 };
arch/arm/boot/dts/imx1-apf9328.dts +1 -0
index 11515c0cb195c..ee4b1b106b1ae 100644
@@ -21,6 +21,7 @@
 	};
 
 	memory@8000000 {
+		device_type = "memory";
 		reg = <0x08000000 0x00800000>;
 	};
 };
arch/arm/boot/dts/imx1.dtsi +0 -2
index 3edc7b5550d88..2b6e77029de4d 100644
@@ -15,10 +15,8 @@
 	 * The decompressor and also some bootloaders rely on a
 	 * pre-existing /chosen node to be available to insert the
 	 * command line and merge other ATAGS info.
-	 * Also for U-Boot there must be a pre-existing /memory node.
 	 */
 	chosen {};
-	memory { device_type = "memory"; };
 
 	aliases {
 		gpio0 = &gpio1;
Signature · 1 line
2.20.1
author maintainer
#11 lore raw

PATCH 19/219 ARM: dts: imx27: Fix memory node duplication

In reply to Sasha Levin

From: Fabio Estevam <[email protected]>

[ Upstream commit 38715dcd49b4430ac5b6bc1293278d91a4d32bd5 ]

Boards based on imx27 have duplicate memory nodes:

- One coming from the board dts file: memory@

- One coming from the imx27.dtsi file.

Fix the duplication by removing the memory node from the dtsi file and by adding 'device_type = "memory";' in the board dts.

---
 arch/arm/boot/dts/imx27-apf27.dts                 | 1 +
 arch/arm/boot/dts/imx27-eukrea-cpuimx27.dtsi      | 1 +
 arch/arm/boot/dts/imx27-pdk.dts                   | 1 +
 arch/arm/boot/dts/imx27-phytec-phycard-s-som.dtsi | 1 +
 arch/arm/boot/dts/imx27-phytec-phycore-som.dtsi   | 1 +
 arch/arm/boot/dts/imx27.dtsi                      | 2 --
 6 files changed, 5 insertions(+), 2 deletions(-)
arch/arm/boot/dts/imx27-apf27.dts +1 -0
index 3eddd805a793a..f635d5c5029c4 100644
@@ -20,6 +20,7 @@
 	compatible = "armadeus,imx27-apf27", "fsl,imx27";
 
 	memory@a0000000 {
+		device_type = "memory";
 		reg = <0xa0000000 0x04000000>;
 	};
 };
arch/arm/boot/dts/imx27-eukrea-cpuimx27.dtsi +1 -0
index 9c455dcbe6ebf..c85f9d01768a1 100644
@@ -17,6 +17,7 @@
 	compatible = "eukrea,cpuimx27", "fsl,imx27";
 
 	memory@a0000000 {
+		device_type = "memory";
 		reg = <0xa0000000 0x04000000>;
 	};
 
arch/arm/boot/dts/imx27-pdk.dts +1 -0
index f9a882d991329..35123b7cb6b3e 100644
@@ -10,6 +10,7 @@
 	compatible = "fsl,imx27-pdk", "fsl,imx27";
 
 	memory@a0000000 {
+		device_type = "memory";
 		reg = <0xa0000000 0x08000000>;
 	};
 
arch/arm/boot/dts/imx27-phytec-phycard-s-som.dtsi +1 -0
index cbad7c88c58cc..b0b4f7c00246d 100644
@@ -18,6 +18,7 @@
 	compatible = "phytec,imx27-pca100", "fsl,imx27";
 
 	memory@a0000000 {
+		device_type = "memory";
 		reg = <0xa0000000 0x08000000>; /* 128MB */
 	};
 };
arch/arm/boot/dts/imx27-phytec-phycore-som.dtsi +1 -0
index ec466b4bfd410..0935e1400e5d2 100644
@@ -17,6 +17,7 @@
 	compatible = "phytec,imx27-pcm038", "fsl,imx27";
 
 	memory@a0000000 {
+		device_type = "memory";
 		reg = <0xa0000000 0x08000000>;
 	};
 
arch/arm/boot/dts/imx27.dtsi +0 -2
index 753d88df16274..39e75b997bdc8 100644
@@ -16,10 +16,8 @@
 	 * The decompressor and also some bootloaders rely on a
 	 * pre-existing /chosen node to be available to insert the
 	 * command line and merge other ATAGS info.
-	 * Also for U-Boot there must be a pre-existing /memory node.
 	 */
 	chosen {};
-	memory { device_type = "memory"; };
 
 	aliases {
 		ethernet0 = &fec;
Signature · 7 lines
2.20.1


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
author maintainer
#12 lore raw

PATCH 20/219 ARM: dts: imx25: Fix memory node duplication

In reply to Sasha Levin

From: Fabio Estevam <[email protected]>

[ Upstream commit 59d8bb363f563e4a147a291037bf979cb8ff9a59 ]

Boards based on imx25 have duplicate memory nodes:

- One coming from the board dts file: memory@

- One coming from the imx25.dtsi file.

Fix the duplication by removing the memory node from the dtsi file and by adding 'device_type = "memory";' in the board dts.

---
 arch/arm/boot/dts/imx25-eukrea-cpuimx25.dtsi | 1 +
 arch/arm/boot/dts/imx25-karo-tx25.dts        | 1 +
 arch/arm/boot/dts/imx25-pdk.dts              | 1 +
 arch/arm/boot/dts/imx25.dtsi                 | 2 --
 4 files changed, 3 insertions(+), 2 deletions(-)
arch/arm/boot/dts/imx25-eukrea-cpuimx25.dtsi +1 -0
index e316fe08837a3..e4d7da267532d 100644
@@ -18,6 +18,7 @@
 	compatible = "eukrea,cpuimx25", "fsl,imx25";
 
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x4000000>; /* 64M */
 	};
 };
arch/arm/boot/dts/imx25-karo-tx25.dts +1 -0
index 5cb6967866c0a..f37e9a75a3ca7 100644
@@ -37,6 +37,7 @@
 	};
 
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x02000000 0x90000000 0x02000000>;
 	};
 };
arch/arm/boot/dts/imx25-pdk.dts +1 -0
index a5626b46ac4e1..f8544a9e46330 100644
@@ -12,6 +12,7 @@
 	compatible = "fsl,imx25-pdk", "fsl,imx25";
 
 	memory@80000000 {
+		device_type = "memory";
 		reg = <0x80000000 0x4000000>;
 	};
 
arch/arm/boot/dts/imx25.dtsi +0 -2
index 85c15ee632727..8c8ad80de4614 100644
@@ -12,10 +12,8 @@
 	 * The decompressor and also some bootloaders rely on a
 	 * pre-existing /chosen node to be available to insert the
 	 * command line and merge other ATAGS info.
-	 * Also for U-Boot there must be a pre-existing /memory node.
 	 */
 	chosen {};
-	memory { device_type = "memory"; };
 
 	aliases {
 		ethernet0 = &fec;
Signature · 7 lines
2.20.1


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
author maintainer
#13 lore raw

PATCH 21/219 ARM: dts: imx53-voipac-dmm-668: Fix memory node duplication

In reply to Sasha Levin

From: Fabio Estevam <[email protected]>

[ Upstream commit 998a84c27a7f3f9133d32af64e19c05cec161a1a ]

imx53-voipac-dmm-668 has two memory nodes, but the correct representation would be to use a single one with two reg entries - one for each RAM chip select, so fix it accordingly.

---
 arch/arm/boot/dts/imx53-voipac-dmm-668.dtsi | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)
arch/arm/boot/dts/imx53-voipac-dmm-668.dtsi +2 -6
index f83a8c62ea531..d595034f3f1bf 100644
@@ -17,12 +17,8 @@
 
 	memory@70000000 {
 		device_type = "memory";
-		reg = <0x70000000 0x20000000>;
-	};
-
-	memory@b0000000 {
-		device_type = "memory";
-		reg = <0xb0000000 0x20000000>;
+		reg = <0x70000000 0x20000000>,
+		      <0xb0000000 0x20000000>;
 	};
 
 	regulators {
Signature · 1 line
2.20.1
author maintainer
#14 lore raw

PATCH 22/219 parisc: Fix serio address output

In reply to Sasha Levin

From: Helge Deller <[email protected]>

[ Upstream commit 785145171d17af2554128becd6a7c8f89e101141 ]

We want the hpa addresses printed in the serio modules, not some virtual ioremap()ed address, e.g.:

serio: gsc-ps2-keyboard port at 0xf0108000 irq 22 @ 2:0:11
serio: gsc-ps2-mouse port at 0xf0108100 irq 22 @ 2:0:12

---
 drivers/input/serio/gscps2.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/input/serio/gscps2.c +2 -2
index 49d8d53e50b7b..96f9b5397367f 100644
@@ -381,9 +381,9 @@ static int __init gscps2_probe(struct parisc_device *dev)
 		goto fail;
 #endif
 
-	printk(KERN_INFO "serio: %s port at 0x%p irq %d @ %s\n",
+	pr_info("serio: %s port at 0x%08lx irq %d @ %s\n",
 		ps2port->port->name,
-		ps2port->addr,
+		hpa,
 		ps2port->padev->irq,
 		ps2port->port->phys);
 
Signature · 1 line
2.20.1
author maintainer
#15 lore raw

PATCH 23/219 parisc: Fix HP SDC hpa address output

In reply to Sasha Levin

From: Helge Deller <[email protected]>

[ Upstream commit c4bff35ca1bfba886da6223c9fed76a2b1382b8e ]

Show the hpa address of the HP SDC instead of a hashed value, e.g.:
HP SDC: HP SDC at 0xf0201000, IRQ 23 (NMI IRQ 24)

---
 drivers/input/serio/hp_sdc.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/input/serio/hp_sdc.c +2 -2
index 0b8a25c58d02e..654252361653d 100644
@@ -884,8 +884,8 @@ static int __init hp_sdc_init(void)
 			"HP SDC NMI", &hp_sdc))
 		goto err2;
 
-	printk(KERN_INFO PREFIX "HP SDC at 0x%p, IRQ %d (NMI IRQ %d)\n",
-	       (void *)hp_sdc.base_io, hp_sdc.irq, hp_sdc.nmi);
+	pr_info(PREFIX "HP SDC at 0x%08lx, IRQ %d (NMI IRQ %d)\n",
+	       hp_sdc.base_io, hp_sdc.irq, hp_sdc.nmi);
 
 	hp_sdc_status_in8();
 	hp_sdc_data_in8();
Signature · 1 line
2.20.1
author maintainer
#16 lore raw

PATCH 24/219 ARM: dts: Fix hsi gdd range for omap4

In reply to Sasha Levin

From: Tony Lindgren <[email protected]>

[ Upstream commit e9e685480b74aef3f3d0967dadb52eea3ff625d2 ]

While reviewing the missing mcasp ranges I noticed omap4 hsi range for gdd is wrong so let's fix it.

I'm not aware of any omap4 devices in mainline kernel though that use hsi though.

Fixes 84badc5ec5fc ("ARM: dts: omap4: Move l4 child devices to probe

them with ti-sysc")

---
 arch/arm/boot/dts/omap4-l4.dtsi | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/arm/boot/dts/omap4-l4.dtsi +2 -2
index 6eb26b837446c..5059ecac44787 100644
@@ -196,12 +196,12 @@
 			clock-names = "fck";
 			#address-cells = <1>;
 			#size-cells = <1>;
-			ranges = <0x0 0x58000 0x4000>;
+			ranges = <0x0 0x58000 0x5000>;
 
 			hsi: hsi@0 {
 				compatible = "ti,omap4-hsi";
 				reg = <0x0 0x4000>,
-				      <0x4a05c000 0x1000>;
+				      <0x5000 0x1000>;
 				reg-names = "sys", "gdd";
 
 				clocks = <&l3_init_clkctrl OMAP4_HSI_CLKCTRL 0>;
Signature · 1 line
2.20.1
author maintainer
#17 lore raw

PATCH 25/219 arm64: mm: Prevent mismatched 52-bit VA support

In reply to Sasha Levin

From: Steve Capper <[email protected]>

[ Upstream commit a96a33b1ca57dbea4285893dedf290aeb8eb090b ]

For cases where there is a mismatch in ARMv8.2-LVA support between CPUs we have to be careful in allowing secondary CPUs to boot if 52-bit virtual addresses have already been enabled on the boot CPU.

This patch adds code to the secondary startup path. If the boot CPU has enabled 52-bit VAs then ID_AA64MMFR2_EL1 is checked to see if the secondary can also enable 52-bit support. If not, the secondary is prevented from booting and an error message is displayed indicating why.

Technically this patch could be implemented using the cpufeature code when considering 52-bit userspace support. However, we employ low level checks here as the cpufeature code won't be able to run if we have mismatched 52-bit kernel va support.

---
 arch/arm64/kernel/head.S | 26 ++++++++++++++++++++++++++
 arch/arm64/kernel/smp.c  |  5 +++++
 2 files changed, 31 insertions(+)
arch/arm64/kernel/head.S +26 -0
index 77ca59598c8b1..06058fba5f86c 100644
@@ -703,6 +703,7 @@ secondary_startup:
 	/*
 	 * Common entry point for secondary CPUs.
 	 */
+	bl	__cpu_secondary_check52bitva
 	bl	__cpu_setup			// initialise processor
 	bl	__enable_mmu
 	ldr	x8, =__secondary_switched
@@ -779,6 +780,31 @@ ENTRY(__enable_mmu)
 	ret
 ENDPROC(__enable_mmu)
 
+ENTRY(__cpu_secondary_check52bitva)
+#ifdef CONFIG_ARM64_52BIT_VA
+	ldr_l	x0, vabits_user
+	cmp	x0, #52
+	b.ne	2f
+
+	mrs_s	x0, SYS_ID_AA64MMFR2_EL1
+	and	x0, x0, #(0xf << ID_AA64MMFR2_LVA_SHIFT)
+	cbnz	x0, 2f
+
+	adr_l	x0, va52mismatch
+	mov	w1, #1
+	strb	w1, [x0]
+	dmb	sy
+	dc	ivac, x0	// Invalidate potentially stale cache line
+
+	update_early_cpu_boot_status CPU_STUCK_IN_KERNEL, x0, x1
+1:	wfe
+	wfi
+	b	1b
+
+#endif
+2:	ret
+ENDPROC(__cpu_secondary_check52bitva)
+
 __no_granule_support:
 	/* Indicate that this CPU can't boot and is stuck in the kernel */
 	update_early_cpu_boot_status CPU_STUCK_IN_KERNEL, x1, x2
arch/arm64/kernel/smp.c +5 -0
index 25fcd22a4bb24..2c31d65bc541b 100644
@@ -108,6 +108,7 @@ static int boot_secondary(unsigned int cpu, struct task_struct *idle)
 }
 
 static DECLARE_COMPLETION(cpu_running);
+bool va52mismatch __ro_after_init;
 
 int __cpu_up(unsigned int cpu, struct task_struct *idle)
 {
@@ -137,6 +138,10 @@ int __cpu_up(unsigned int cpu, struct task_struct *idle)
 
 		if (!cpu_online(cpu)) {
 			pr_crit("CPU%u: failed to come online\n", cpu);
+
+			if (IS_ENABLED(CONFIG_ARM64_52BIT_VA) && va52mismatch)
+				pr_crit("CPU%u: does not support 52-bit VAs\n", cpu);
+
 			ret = -EIO;
 		}
 	} else {
Signature · 7 lines
2.20.1


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
author maintainer
#18 lore raw

PATCH 26/219 arm64: smp: Handle errors reported by the firmware

In reply to Sasha Levin

From: Suzuki K Poulose <[email protected]>

[ Upstream commit f357b3a7e17af7736d67d8267edc1ed3d1dd9391 ]

The __cpu_up() routine ignores the errors reported by the firmware for a CPU bringup operation and looks for the error status set by the booting CPU. If the CPU never entered the kernel, we could end up in assuming stale error status, which otherwise would have been set/cleared appropriately by the booting CPU.

---
 arch/arm64/kernel/smp.c | 1 +
 1 file changed, 1 insertion(+)
arch/arm64/kernel/smp.c +1 -0
index 2c31d65bc541b..52aa51f6310b0 100644
@@ -146,6 +146,7 @@ int __cpu_up(unsigned int cpu, struct task_struct *idle)
 		}
 	} else {
 		pr_err("CPU%u: failed to boot: %d\n", cpu, ret);
+		return ret;
 	}
 
 	secondary_data.task = NULL;
Signature · 7 lines
2.20.1


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
author maintainer
#19 lore raw

PATCH 27/219 bus: ti-sysc: Check for no-reset and no-idle flags at the child level

In reply to Sasha Levin

From: Tony Lindgren <[email protected]>

[ Upstream commit 4014c08ba39476a18af546186da625a6833a1529 ]

With ti-sysc, we need to now have the device tree properties for ti,no-reset-on-init and ti,no-idle-on-init at the module level instead of the child device level.

Let's check for these properties at the child device level to enable quirks, and warn about moving the properties to the module level.

Otherwise am335x-evm based boards tagging gpio1 with ti,no-reset-on-init will have their DDR power disabled if wired up in such a tricky way.

Note that this should not be an issue for earlier kernels as we don't rely on this until the dts files have been updated to probe with ti-sysc interconnect target driver.

---
 drivers/bus/ti-sysc.c | 32 +++++++++++++++++++++++++-------
 1 file changed, 25 insertions(+), 7 deletions(-)
drivers/bus/ti-sysc.c +25 -7
index e95b26319cd91..5b31131d0cba2 100644
@@ -91,6 +91,9 @@ struct sysc {
 	struct delayed_work idle_work;
 };
 
+static void sysc_parse_dts_quirks(struct sysc *ddata, struct device_node *np,
+				  bool is_child);
+
 void sysc_write(struct sysc *ddata, int offset, u32 value)
 {
 	writel_relaxed(value, ddata->module_va + offset);
@@ -374,6 +377,7 @@ static int sysc_check_one_child(struct sysc *ddata,
 		dev_warn(ddata->dev, "really a child ti,hwmods property?");
 
 	sysc_check_quirk_stdout(ddata, np);
+	sysc_parse_dts_quirks(ddata, np, true);
 
 	return 0;
 }
@@ -1343,23 +1347,37 @@ static const struct sysc_dts_quirk sysc_dts_quirks[] = {
 	  .mask = SYSC_QUIRK_NO_RESET_ON_INIT, },
 };
 
-static int sysc_init_dts_quirks(struct sysc *ddata)
+static void sysc_parse_dts_quirks(struct sysc *ddata, struct device_node *np,
+				  bool is_child)
 {
-	struct device_node *np = ddata->dev->of_node;
 	const struct property *prop;
-	int i, len, error;
-	u32 val;
-
-	ddata->legacy_mode = of_get_property(np, "ti,hwmods", NULL);
+	int i, len;
 
 	for (i = 0; i < ARRAY_SIZE(sysc_dts_quirks); i++) {
-		prop = of_get_property(np, sysc_dts_quirks[i].name, &len);
+		const char *name = sysc_dts_quirks[i].name;
+
+		prop = of_get_property(np, name, &len);
 		if (!prop)
 			continue;
 
 		ddata->cfg.quirks |= sysc_dts_quirks[i].mask;
+		if (is_child) {
+			dev_warn(ddata->dev,
+				 "dts flag should be at module level for %s\n",
+				 name);
+		}
 	}
+}
+
+static int sysc_init_dts_quirks(struct sysc *ddata)
+{
+	struct device_node *np = ddata->dev->of_node;
+	int error;
+	u32 val;
+
+	ddata->legacy_mode = of_get_property(np, "ti,hwmods", NULL);
 
+	sysc_parse_dts_quirks(ddata, np, false);
 	error = of_property_read_u32(np, "ti,sysc-delay-us", &val);
 	if (!error) {
 		if (val > 255) {
Signature · 1 line
2.20.1
author maintainer
#20 lore raw

PATCH 28/219 platform/x86: mlx-platform: Fix LED configuration

In reply to Sasha Levin

From: Vadim Pasternak <[email protected]>

[ Upstream commit 440f343df1996302d9a3904647ff11b689bf27bc ]

Exchange LED configuration between msn201x and next generation systems types.

Bug was introduced when LED driver activation was added to mlx-platform. LED configuration for the three new system MQMB7, MSN37, MSN34 was assigned to MSN21 and vice versa. This bug affects MSN21 only and likely requires backport to v4.19.

Fixes 1189456b1cce ("platform/x86: mlx-platform: Add LED platform driver activation")
---
 drivers/platform/x86/mlx-platform.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/platform/x86/mlx-platform.c +2 -2
index 742a0c2179256..ab1aedd72d340 100644
@@ -1421,7 +1421,7 @@ static int __init mlxplat_dmi_msn201x_matched(const struct dmi_system_id *dmi)
 	mlxplat_hotplug = &mlxplat_mlxcpld_msn201x_data;
 	mlxplat_hotplug->deferred_nr =
 		mlxplat_default_channels[i - 1][MLXPLAT_CPLD_GRP_CHNL_NUM - 1];
-	mlxplat_led = &mlxplat_default_ng_led_data;
+	mlxplat_led = &mlxplat_msn21xx_led_data;
 	mlxplat_regs_io = &mlxplat_msn21xx_regs_io_data;
 
 	return 1;
@@ -1439,7 +1439,7 @@ static int __init mlxplat_dmi_qmb7xx_matched(const struct dmi_system_id *dmi)
 	mlxplat_hotplug = &mlxplat_mlxcpld_default_ng_data;
 	mlxplat_hotplug->deferred_nr =
 		mlxplat_msn21xx_channels[MLXPLAT_CPLD_GRP_CHNL_NUM - 1];
-	mlxplat_led = &mlxplat_msn21xx_led_data;
+	mlxplat_led = &mlxplat_default_ng_led_data;
 	mlxplat_fan = &mlxplat_default_fan_data;
 
 	return 1;
Signature · 1 line
2.20.1
author maintainer
#21 lore raw

PATCH 29/219 ARM: OMAP1: fix USB configuration for device-only setups

In reply to Sasha Levin

From: Aaro Koskinen <[email protected]>

[ Upstream commit c7b7b5cbd0c859b1546a5a3455d457708bdadf4c ]

Currently we do USB configuration only if the host mode (CONFIG_USB) is enabled. But it should be done also in the case of device-only setups, so change the condition to CONFIG_USB_SUPPORT. This allows to use omap_udc on Palm Tungsten E.

---
 arch/arm/mach-omap1/Makefile           | 2 +-
 arch/arm/mach-omap1/include/mach/usb.h | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
arch/arm/mach-omap1/Makefile +1 -1
index e8ccf51c6f292..ec0235899de20 100644
@@ -25,7 +25,7 @@ obj-y					+= $(i2c-omap-m) $(i2c-omap-y)
 
 led-y := leds.o
 
-usb-fs-$(CONFIG_USB)			:= usb.o
+usb-fs-$(CONFIG_USB_SUPPORT)		:= usb.o
 obj-y					+= $(usb-fs-m) $(usb-fs-y)
 
 # Specific board support
arch/arm/mach-omap1/include/mach/usb.h +1 -1
index 77867778d4ec7..5429d86c7190d 100644
@@ -11,7 +11,7 @@
 
 #include <linux/platform_data/usb-omap1.h>
 
-#if IS_ENABLED(CONFIG_USB)
+#if IS_ENABLED(CONFIG_USB_SUPPORT)
 void omap1_usb_init(struct omap_usb_config *pdata);
 #else
 static inline void omap1_usb_init(struct omap_usb_config *pdata)
Signature · 1 line
2.20.1
author maintainer
#22 lore raw

PATCH 30/219 RDMA/hns: Fix the bug while use multi-hop of pbl

In reply to Sasha Levin

From: Lijun Ou <[email protected]>

[ Upstream commit 4af07f01f7a787ba5158352b98c9e3cb74995a1c ]

It will prevent multiply overflow when defines the pbl for u64 type.

---
 drivers/infiniband/hw/hns/hns_roce_mr.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/infiniband/hw/hns/hns_roce_mr.c +2 -2
index 41a538d23b802..c68596d4e8037 100644
@@ -1017,14 +1017,14 @@ struct ib_mr *hns_roce_reg_user_mr(struct ib_pd *pd, u64 start, u64 length,
 			goto err_umem;
 		}
 	} else {
-		int pbl_size = 1;
+		u64 pbl_size = 1;
 
 		bt_size = (1 << (hr_dev->caps.pbl_ba_pg_sz + PAGE_SHIFT)) / 8;
 		for (i = 0; i < hr_dev->caps.pbl_hop_num; i++)
 			pbl_size *= bt_size;
 		if (n > pbl_size) {
 			dev_err(dev,
-			    " MR len %lld err. MR page num is limited to %d!\n",
+			    " MR len %lld err. MR page num is limited to %lld!\n",
 			    length, pbl_size);
 			ret = -EINVAL;
 			goto err_umem;
Signature · 1 line
2.20.1
author maintainer
#23 lore raw

PATCH 31/219 arm64: preempt: Fix big-endian when checking preempt count in assembly

In reply to Sasha Levin

From: Will Deacon <[email protected]>

[ Upstream commit 7faa313f05cad184e8b17750f0cbe5216ac6debb ]

Commit 396244692232 ("arm64: preempt: Provide our own implementation of asm/preempt.h") extended the preempt count field in struct thread_info to 64 bits, so that it consists of a 32-bit count plus a 32-bit flag indicating whether or not the current task needs rescheduling.

Whilst the asm-offsets definition of TSK_TI_PREEMPT was updated to point to this new field, the assembly usage was left untouched meaning that a 32-bit load from TSK_TI_PREEMPT on a big-endian machine actually returns the reschedule flag instead of the count.

Whilst we could fix this by pointing TSK_TI_PREEMPT at the count field, we're actually better off reworking the two assembly users so that they operate on the whole 64-bit value in favour of inspecting the thread flags separately in order to determine whether a reschedule is needed.

---
 arch/arm64/include/asm/assembler.h | 8 +++-----
 arch/arm64/kernel/entry.S          | 6 ++----
 2 files changed, 5 insertions(+), 9 deletions(-)
arch/arm64/include/asm/assembler.h +3 -5
index 5a97ac8531682..0c100506a29aa 100644
@@ -683,11 +683,9 @@ USER(\label, ic	ivau, \tmp2)			// invalidate I line PoU
 	.macro		if_will_cond_yield_neon
 #ifdef CONFIG_PREEMPT
 	get_thread_info	x0
-	ldr		w1, [x0, #TSK_TI_PREEMPT]
-	ldr		x0, [x0, #TSK_TI_FLAGS]
-	cmp		w1, #PREEMPT_DISABLE_OFFSET
-	csel		x0, x0, xzr, eq
-	tbnz		x0, #TIF_NEED_RESCHED, .Lyield_\@	// needs rescheduling?
+	ldr		x0, [x0, #TSK_TI_PREEMPT]
+	sub		x0, x0, #PREEMPT_DISABLE_OFFSET
+	cbz		x0, .Lyield_\@
 	/* fall through to endif_yield_neon */
 	.subsection	1
 .Lyield_\@ :
arch/arm64/kernel/entry.S +2 -4
index 5f800384cb9a8..bb68323530458 100644
@@ -622,10 +622,8 @@ el1_irq:
 	irq_handler
 
 #ifdef CONFIG_PREEMPT
-	ldr	w24, [tsk, #TSK_TI_PREEMPT]	// get preempt count
-	cbnz	w24, 1f				// preempt count != 0
-	ldr	x0, [tsk, #TSK_TI_FLAGS]	// get flags
-	tbz	x0, #TIF_NEED_RESCHED, 1f	// needs rescheduling?
+	ldr	x24, [tsk, #TSK_TI_PREEMPT]	// get preempt count
+	cbnz	x24, 1f				// preempt count != 0
 	bl	el1_preempt
 1:
 #endif
Signature · 7 lines
2.20.1


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
#24 lore raw

Re: [PATCH AUTOSEL 4.19 031/219] arm64: preempt: Fix big-endian when checking preempt count in assembly

Quoted 53 lines · Sasha Levin

From: Will Deacon <[email protected]>

[ Upstream commit 7faa313f05cad184e8b17750f0cbe5216ac6debb ]

Commit 396244692232 ("arm64: preempt: Provide our own implementation of asm/preempt.h") extended the preempt count field in struct thread_info to 64 bits, so that it consists of a 32-bit count plus a 32-bit flag indicating whether or not the current task needs rescheduling.

Whilst the asm-offsets definition of TSK_TI_PREEMPT was updated to point to this new field, the assembly usage was left untouched meaning that a 32-bit load from TSK_TI_PREEMPT on a big-endian machine actually returns the reschedule flag instead of the count.

Whilst we could fix this by pointing TSK_TI_PREEMPT at the count field, we're actually better off reworking the two assembly users so that they operate on the whole 64-bit value in favour of inspecting the thread flags separately in order to determine whether a reschedule is needed.

---
 arch/arm64/include/asm/assembler.h | 8 +++-----
 arch/arm64/kernel/entry.S          | 6 ++----
 2 files changed, 5 insertions(+), 9 deletions(-)
arch/arm64/include/asm/assembler.h +3 -5
index 5a97ac8531682..0c100506a29aa 100644
@@ -683,11 +683,9 @@ USER(\label, ic	ivau, \tmp2)			// invalidate I line PoU
 	.macro		if_will_cond_yield_neon
 #ifdef CONFIG_PREEMPT
 	get_thread_info	x0
-	ldr		w1, [x0, #TSK_TI_PREEMPT]
-	ldr		x0, [x0, #TSK_TI_FLAGS]
-	cmp		w1, #PREEMPT_DISABLE_OFFSET
-	csel		x0, x0, xzr, eq
-	tbnz		x0, #TIF_NEED_RESCHED, .Lyield_\@	// needs rescheduling?
+	ldr		x0, [x0, #TSK_TI_PREEMPT]
+	sub		x0, x0, #PREEMPT_DISABLE_OFFSET
+	cbz		x0, .Lyield_\@
 	/* fall through to endif_yield_neon */
 	.subsection	1
 .Lyield_\@ :
arch/arm64/kernel/entry.S +0 -0
index 5f800384cb9a8..bb68323530458 100644

arch/arm64/kernel/entry.S el1_irq: · line 627

Quoted 10 lines
arch/arm64/kernel/entry.S +2 -4
@@ -622,10 +622,8 @@ el1_irq:
 	irq_handler
 
 #ifdef CONFIG_PREEMPT
-	ldr	w24, [tsk, #TSK_TI_PREEMPT]	// get preempt count
-	cbnz	w24, 1f				// preempt count != 0
-	ldr	x0, [tsk, #TSK_TI_FLAGS]	// get flags
-	tbz	x0, #TIF_NEED_RESCHED, 1f	// needs rescheduling?
+	ldr	x24, [tsk, #TSK_TI_PREEMPT]	// get preempt count
+	cbnz	x24, 1f				// preempt count != 0
 	bl	el1_preempt

While updating 4.19-rt, I stumbled on this change to arm64 backport. And was confused by it, but looking deeper, this is something that breaks without having 396244692232f ("arm64: preempt: Provide our own implementation of asm/preempt.h").

That commit inverts the TIF_NEED_RESCHED meaning where set means we don't need to resched, and clear means we need to resched. This way we can combine the preempt count with the need resched flag test as they share the same 64bit word. A 0 means we need to preempt (as NEED_RESCHED being zero means we need to resched, and this also means preempt_count is zero). If the TIF_NEED_RESCHED bit is set, that means we don't need to resched, and if preempt count is something other than zero, we don't need to resched, and since those two are together by commit 396244692232f, we can just test #TSK_TI_PREEMPT. But because that commit does not exist in 4.19, we can't remove the TIF_NEED_RESCHED check, that this backport does, and then breaks the kernel!

-- Steve

Quoted 4 lines
arch/arm64/kernel/entry.S +0 -0
 1:
 #endif
Signature · 1 line
2.20.1
#25 lore raw

Re: [PATCH AUTOSEL 4.19 031/219] arm64: preempt: Fix big-endian when checking preempt count in assembly

Quoted 70 lines · Steven Rostedt
Quoted 53 lines · Sasha Levin

From: Will Deacon <[email protected]>

[ Upstream commit 7faa313f05cad184e8b17750f0cbe5216ac6debb ]

Commit 396244692232 ("arm64: preempt: Provide our own implementation of asm/preempt.h") extended the preempt count field in struct thread_info to 64 bits, so that it consists of a 32-bit count plus a 32-bit flag indicating whether or not the current task needs rescheduling.

Whilst the asm-offsets definition of TSK_TI_PREEMPT was updated to point to this new field, the assembly usage was left untouched meaning that a 32-bit load from TSK_TI_PREEMPT on a big-endian machine actually returns the reschedule flag instead of the count.

Whilst we could fix this by pointing TSK_TI_PREEMPT at the count field, we're actually better off reworking the two assembly users so that they operate on the whole 64-bit value in favour of inspecting the thread flags separately in order to determine whether a reschedule is needed.

---
 arch/arm64/include/asm/assembler.h | 8 +++-----
 arch/arm64/kernel/entry.S          | 6 ++----
 2 files changed, 5 insertions(+), 9 deletions(-)
arch/arm64/include/asm/assembler.h +3 -5
index 5a97ac8531682..0c100506a29aa 100644
@@ -683,11 +683,9 @@ USER(\label, ic	ivau, \tmp2)			// invalidate I line PoU
 	.macro		if_will_cond_yield_neon
 #ifdef CONFIG_PREEMPT
 	get_thread_info	x0
-	ldr		w1, [x0, #TSK_TI_PREEMPT]
-	ldr		x0, [x0, #TSK_TI_FLAGS]
-	cmp		w1, #PREEMPT_DISABLE_OFFSET
-	csel		x0, x0, xzr, eq
-	tbnz		x0, #TIF_NEED_RESCHED, .Lyield_\@	// needs rescheduling?
+	ldr		x0, [x0, #TSK_TI_PREEMPT]
+	sub		x0, x0, #PREEMPT_DISABLE_OFFSET
+	cbz		x0, .Lyield_\@
 	/* fall through to endif_yield_neon */
 	.subsection	1
 .Lyield_\@ :
arch/arm64/kernel/entry.S +0 -0
index 5f800384cb9a8..bb68323530458 100644

arch/arm64/kernel/entry.S el1_irq: · line 627

Quoted 10 lines
arch/arm64/kernel/entry.S +2 -4
@@ -622,10 +622,8 @@ el1_irq:
 	irq_handler
 
 #ifdef CONFIG_PREEMPT
-	ldr	w24, [tsk, #TSK_TI_PREEMPT]	// get preempt count
-	cbnz	w24, 1f				// preempt count != 0
-	ldr	x0, [tsk, #TSK_TI_FLAGS]	// get flags
-	tbz	x0, #TIF_NEED_RESCHED, 1f	// needs rescheduling?
+	ldr	x24, [tsk, #TSK_TI_PREEMPT]	// get preempt count
+	cbnz	x24, 1f				// preempt count != 0
 	bl	el1_preempt

While updating 4.19-rt, I stumbled on this change to arm64 backport. And was confused by it, but looking deeper, this is something that breaks without having 396244692232f ("arm64: preempt: Provide our own implementation of asm/preempt.h").

Quoted 11 lines

That commit inverts the TIF_NEED_RESCHED meaning where set means we don't need to resched, and clear means we need to resched. This way we can combine the preempt count with the need resched flag test as they share the same 64bit word. A 0 means we need to preempt (as NEED_RESCHED being zero means we need to resched, and this also means preempt_count is zero). If the TIF_NEED_RESCHED bit is set, that means we don't need to resched, and if preempt count is something other than zero, we don't need to resched, and since those two are together by commit 396244692232f, we can just test #TSK_TI_PREEMPT. But because that commit does not exist in 4.19, we can't remove the TIF_NEED_RESCHED check, that this backport does, and then breaks the kernel!

Yup, without 396244692232 this commit makes no sense. That's why I didn't CC stable or add a Fixes tag :(

Will

#26 lore raw

Re: [PATCH AUTOSEL 4.19 031/219] arm64: preempt: Fix big-endian when checking preempt count in assembly

Quoted 2 lines · On Mon, 16 Dec 2019 09:45:24 +0000 Will Deacon <[email protected]> wrote:

Yup, without 396244692232 this commit makes no sense. That's why I didn't CC stable or add a Fixes tag :(

Yes, please have stable revert this change.

Thanks!

-- Steve

#27 lore raw

Re: [PATCH AUTOSEL 4.19 031/219] arm64: preempt: Fix big-endian when checking preempt count in assembly

Quoted 7 lines · Steven Rostedt
Quoted 2 lines · On Mon, 16 Dec 2019 09:45:24 +0000 Will Deacon <[email protected]> wrote:

Yup, without 396244692232 this commit makes no sense. That's why I didn't CC stable or add a Fixes tag :(

Yes, please have stable revert this change.

Now reverted, thanks.

greg k-h

#28 lore raw

Re: [PATCH AUTOSEL 4.19 031/219] arm64: preempt: Fix big-endian when checking preempt count in assembly

In reply to Will Deacon
Quoted 2 lines · On Mon, 16 Dec 2019 09:45:24 +0000 Will Deacon <[email protected]> wrote:

Yup, without 396244692232 this commit makes no sense. That's why I didn't CC stable or add a Fixes tag :(

I'm wondering if we should add a tag to state "not to be backported", to tell auto select to ignore a patch that appears to be a fix.

-- Steve

#29 lore raw

Re: [PATCH AUTOSEL 4.19 031/219] arm64: preempt: Fix big-endian when checking preempt count in assembly

Quoted 8 lines · Steven Rostedt
Quoted 2 lines · On Mon, 16 Dec 2019 09:45:24 +0000 Will Deacon <[email protected]> wrote:

Yup, without 396244692232 this commit makes no sense. That's why I didn't CC stable or add a Fixes tag :(

I'm wondering if we should add a tag to state "not to be backported", to tell auto select to ignore a patch that appears to be a fix.

I'd be in favour of that!

Will

author maintainer
#30 lore raw

PATCH 32/219 RDMA/vmw_pvrdma: Use atomic memory allocation in create AH

In reply to Sasha Levin

From: Gal Pressman <[email protected]>

[ Upstream commit a276a4d93bf1580d737f38d1810e5f4b166f3edd ]

Create address handle callback should not sleep, use GFP_ATOMIC instead of GFP_KERNEL for memory allocation.

Fixes 29c8d9eba550 ("IB: Add vmw_pvrdma driver")
Cc Adit Ranadive <[email protected]>
---
 drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/infiniband/hw/vmw_pvrdma/pvrdma_verbs.c +1 -1
index b65d10b0a8759..f4cb5cf26006f 100644
@@ -555,7 +555,7 @@ struct ib_ah *pvrdma_create_ah(struct ib_pd *pd, struct rdma_ah_attr *ah_attr,
 	if (!atomic_add_unless(&dev->num_ahs, 1, dev->dsr->caps.max_ah))
 		return ERR_PTR(-ENOMEM);
 
-	ah = kzalloc(sizeof(*ah), GFP_KERNEL);
+	ah = kzalloc(sizeof(*ah), GFP_ATOMIC);
 	if (!ah) {
 		atomic_dec(&dev->num_ahs);
 		return ERR_PTR(-ENOMEM);
Signature · 1 line
2.20.1
author maintainer
#31 lore raw

PATCH 33/219 PM / AVS: SmartReflex: NULL check before some freeing functions is not needed

In reply to Sasha Levin

From: Thomas Meyer <[email protected]>

[ Upstream commit 14d338a857f05f894ba3badd9e6d3039c68b8180 ]

NULL check before some freeing functions is not needed.

---
 drivers/power/avs/smartreflex.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/power/avs/smartreflex.c +1 -2
index 1360a7fa542c5..8760477d0e8af 100644
@@ -1010,8 +1010,7 @@ static int omap_sr_remove(struct platform_device *pdev)
 
 	if (sr_info->autocomp_active)
 		sr_stop_vddautocomp(sr_info);
-	if (sr_info->dbg_dir)
-		debugfs_remove_recursive(sr_info->dbg_dir);
+	debugfs_remove_recursive(sr_info->dbg_dir);
 
 	pm_runtime_disable(&pdev->dev);
 	list_del(&sr_info->node);
Signature · 1 line
2.20.1
author maintainer
#32 lore raw

PATCH 34/219 xfs: zero length symlinks are not valid

In reply to Sasha Levin

From: Dave Chinner <[email protected]>

[ Upstream commit 43feeea88c9cb2955b9f7ba8152ec5abeea42810 ]

A log recovery failure has been reproduced where a symlink inode has a zero length in extent form. It was caused by a shutdown during a combined fstress+fsmark workload.

The underlying problem is the issue in xfs_inactive_symlink(): the inode is unlocked between the symlink inactivation/truncation and the inode being freed. This opens a window for the inode to be written to disk before it xfs_ifree() removes it from the unlinked list, marks it free in the inobt and zeros the mode.

For shortform inodes, the fix is simple. xfs_ifree() clears the data fork state, so there's no need to do it in xfs_inactive_symlink(). This means the shortform fork verifier will not see a zero length data fork as it mirrors the inode size through to xfs_ifree()), and hence if the inode gets written back and the fork verifiers are run they will still see a fork that matches the on-disk inode size.

For extent form (remote) symlinks, it is a little more tricky. Here we explicitly set the inode size to zero, so the above race can lead to zero length symlinks on disk. Because the inode is unlinked at this point (i.e. on the unlinked list) and unreferenced, it can never be seen again by a user. Hence when we set the inode size to zeor, also change the type to S_IFREG. xfs_ifree() expects S_IFREG inodes to be of zero length, and so this avoids all the problems of zero length symlinks ever hitting the disk. It also avoids the problem of needing to handle zero length symlink inodes in log recovery to replay the extent free intents and the remaining deferops to free the extents the symlink used.

Also add a couple of asserts to warn us if zero length symlinks end up in either the symlink create or inactivation paths.

---
 fs/xfs/libxfs/xfs_symlink_remote.c | 14 +++++++++----
 fs/xfs/xfs_symlink.c               | 33 +++++++++++++++---------------
 2 files changed, 27 insertions(+), 20 deletions(-)
fs/xfs/libxfs/xfs_symlink_remote.c +10 -4
index 95374ab2dee73..77d80106f989a 100644
@@ -199,7 +199,10 @@ xfs_symlink_local_to_remote(
 					ifp->if_bytes - 1);
 }
 
-/* Verify the consistency of an inline symlink. */
+/*
+ * Verify the in-memory consistency of an inline symlink data fork. This
+ * does not do on-disk format checks.
+ */
 xfs_failaddr_t
 xfs_symlink_shortform_verify(
 	struct xfs_inode	*ip)
@@ -215,9 +218,12 @@ xfs_symlink_shortform_verify(
 	size = ifp->if_bytes;
 	endp = sfp + size;
 
-	/* Zero length symlinks can exist while we're deleting a remote one. */
-	if (size == 0)
-		return NULL;
+	/*
+	 * Zero length symlinks should never occur in memory as they are
+	 * never alllowed to exist on disk.
+	 */
+	if (!size)
+		return __this_address;
 
 	/* No negative sizes or overly long symlink targets. */
 	if (size < 0 || size > XFS_SYMLINK_MAXLEN)
fs/xfs/xfs_symlink.c +17 -16
index a3e98c64b6e38..b2c1177c717ff 100644
@@ -192,6 +192,7 @@ xfs_symlink(
 	pathlen = strlen(target_path);
 	if (pathlen >= XFS_SYMLINK_MAXLEN)      /* total string too long */
 		return -ENAMETOOLONG;
+	ASSERT(pathlen > 0);
 
 	udqp = gdqp = NULL;
 	prid = xfs_get_initial_prid(dp);
@@ -378,6 +379,12 @@ xfs_symlink(
 
 /*
  * Free a symlink that has blocks associated with it.
+ *
+ * Note: zero length symlinks are not allowed to exist. When we set the size to
+ * zero, also change it to a regular file so that it does not get written to
+ * disk as a zero length symlink. The inode is on the unlinked list already, so
+ * userspace cannot find this inode anymore, so this change is not user visible
+ * but allows us to catch corrupt zero-length symlinks in the verifiers.
  */
 STATIC int
 xfs_inactive_symlink_rmt(
@@ -412,13 +419,14 @@ xfs_inactive_symlink_rmt(
 	xfs_trans_ijoin(tp, ip, 0);
 
 	/*
-	 * Lock the inode, fix the size, and join it to the transaction.
-	 * Hold it so in the normal path, we still have it locked for
-	 * the second transaction.  In the error paths we need it
+	 * Lock the inode, fix the size, turn it into a regular file and join it
+	 * to the transaction.  Hold it so in the normal path, we still have it
+	 * locked for the second transaction.  In the error paths we need it
 	 * held so the cancel won't rele it, see below.
 	 */
 	size = (int)ip->i_d.di_size;
 	ip->i_d.di_size = 0;
+	VFS_I(ip)->i_mode = (VFS_I(ip)->i_mode & ~S_IFMT) | S_IFREG;
 	xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
 	/*
 	 * Find the block(s) so we can inval and unmap them.
@@ -494,17 +502,10 @@ xfs_inactive_symlink(
 		return -EIO;
 
 	xfs_ilock(ip, XFS_ILOCK_EXCL);
-
-	/*
-	 * Zero length symlinks _can_ exist.
-	 */
 	pathlen = (int)ip->i_d.di_size;
-	if (!pathlen) {
-		xfs_iunlock(ip, XFS_ILOCK_EXCL);
-		return 0;
-	}
+	ASSERT(pathlen);
 
-	if (pathlen < 0 || pathlen > XFS_SYMLINK_MAXLEN) {
+	if (pathlen <= 0 || pathlen > XFS_SYMLINK_MAXLEN) {
 		xfs_alert(mp, "%s: inode (0x%llx) bad symlink length (%d)",
 			 __func__, (unsigned long long)ip->i_ino, pathlen);
 		xfs_iunlock(ip, XFS_ILOCK_EXCL);
@@ -512,12 +513,12 @@ xfs_inactive_symlink(
 		return -EFSCORRUPTED;
 	}
 
+	/*
+	 * Inline fork state gets removed by xfs_difree() so we have nothing to
+	 * do here in that case.
+	 */
 	if (ip->i_df.if_flags & XFS_IFINLINE) {
-		if (ip->i_df.if_bytes > 0) 
-			xfs_idata_realloc(ip, -(ip->i_df.if_bytes),
-					  XFS_DATA_FORK);
 		xfs_iunlock(ip, XFS_ILOCK_EXCL);
-		ASSERT(ip->i_df.if_bytes == 0);
 		return 0;
 	}
 
Signature · 1 line
2.20.1
author maintainer
#33 lore raw

PATCH 35/219 ARM: ks8695: fix section mismatch warning

In reply to Sasha Levin

From: Arnd Bergmann <[email protected]>

[ Upstream commit 4aa64677330beeeed721b4b122884dabad845d66 ]

WARNING: vmlinux.o(.text+0x13250): Section mismatch in reference from the function acs5k_i2c_init() to the (unknown reference) .init.data:(unknown) The function acs5k_i2c_init() references
the (unknown reference) __initdata (unknown).
This is often because acs5k_i2c_init lacks a __initdata
annotation or the annotation of (unknown) is wrong.

---
 arch/arm/mach-ks8695/board-acs5k.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm/mach-ks8695/board-acs5k.c +1 -1
index ef835d82cdb95..5783062224c39 100644
@@ -100,7 +100,7 @@ static struct i2c_board_info acs5k_i2c_devs[] __initdata = {
 	},
 };
 
-static void acs5k_i2c_init(void)
+static void __init acs5k_i2c_init(void)
 {
 	/* The gpio interface */
 	gpiod_add_lookup_table(&acs5k_i2c_gpiod_table);
Signature · 7 lines
2.20.1


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
author maintainer
#34 lore raw

PATCH 36/219 ACPI / LPSS: Ignore acpi_device_fix_up_power() return value

In reply to Sasha Levin

From: Hans de Goede <[email protected]>

[ Upstream commit 1a2fa02f7489dc4d746f2a15fb77b3ce1affade8 ]

Ignore acpi_device_fix_up_power() return value. If we return an error we end up with acpi_default_enumeration() still creating a platform- device for the device and we end up with the device still being used but without the special LPSS related handling which is not useful.

Specicifically ignoring the error fixes the touchscreen no longer working after a suspend/resume on a Prowise PT301 tablet.

This tablet has a broken _PS0 method on the touchscreen's I2C controller, causing acpi_device_fix_up_power() to fail, causing fallback to standard platform-dev handling and specifically causing acpi_lpss_save/restore_ctx to not run.

The I2C controllers _PS0 method does actually turn on the device, but then does some more nonsense which fails when run during early boot trying to use I2C opregion handling on another not-yet registered I2C controller.

---
 drivers/acpi/acpi_lpss.c | 7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)
drivers/acpi/acpi_lpss.c +1 -6
index c651e206d7960..d79245e9dff72 100644
@@ -637,12 +637,7 @@ static int acpi_lpss_create_device(struct acpi_device *adev,
 	 * have _PS0 and _PS3 without _PSC (and no power resources), so
 	 * acpi_bus_init_power() will assume that the BIOS has put them into D0.
 	 */
-	ret = acpi_device_fix_up_power(adev);
-	if (ret) {
-		/* Skip the device, but continue the namespace scan. */
-		ret = 0;
-		goto err_out;
-	}
+	acpi_device_fix_up_power(adev);
 
 	adev->driver_data = pdata;
 	pdev = acpi_create_platform_device(adev, dev_desc->properties);
Signature · 1 line
2.20.1
author maintainer
#35 lore raw

PATCH 37/219 scsi: lpfc: Enable Management features for IF_TYPE=6

In reply to Sasha Levin

From: James Smart <[email protected]>

[ Upstream commit 719162bd5bb968203397b9b1d0dd30a9797bbd09 ]

Addition of support for if_type=6 missed several checks for interface type, resulting in the failure of several key management features such as firmware dump and loopback testing.

Correct the checks on the if_type so that both SLI4 IF_TYPE's 2 and 6 are supported.

---
 drivers/scsi/lpfc/lpfc_attr.c    | 4 ++--
 drivers/scsi/lpfc/lpfc_bsg.c     | 6 +++---
 drivers/scsi/lpfc/lpfc_els.c     | 2 +-
 drivers/scsi/lpfc/lpfc_hbadisc.c | 2 +-
 4 files changed, 7 insertions(+), 7 deletions(-)
drivers/scsi/lpfc/lpfc_attr.c +2 -2
index 55cd96e2469c6..3f69a5e4e470a 100644
@@ -1332,7 +1332,7 @@ lpfc_sli4_pdev_reg_request(struct lpfc_hba *phba, uint32_t opcode)
 		return -EACCES;
 
 	if ((phba->sli_rev < LPFC_SLI_REV4) ||
-	    (bf_get(lpfc_sli_intf_if_type, &phba->sli4_hba.sli_intf) !=
+	    (bf_get(lpfc_sli_intf_if_type, &phba->sli4_hba.sli_intf) <
 	     LPFC_SLI_INTF_IF_TYPE_2))
 		return -EPERM;
 
@@ -4264,7 +4264,7 @@ lpfc_link_speed_store(struct device *dev, struct device_attribute *attr,
 	uint32_t prev_val, if_type;
 
 	if_type = bf_get(lpfc_sli_intf_if_type, &phba->sli4_hba.sli_intf);
-	if (if_type == LPFC_SLI_INTF_IF_TYPE_2 &&
+	if (if_type >= LPFC_SLI_INTF_IF_TYPE_2 &&
 	    phba->hba_flag & HBA_FORCED_LINK_SPEED)
 		return -EPERM;
 
drivers/scsi/lpfc/lpfc_bsg.c +3 -3
index 90745feca8080..99aea52e584b0 100644
@@ -2221,7 +2221,7 @@ lpfc_bsg_diag_loopback_mode(struct bsg_job *job)
 
 	if (phba->sli_rev < LPFC_SLI_REV4)
 		rc = lpfc_sli3_bsg_diag_loopback_mode(phba, job);
-	else if (bf_get(lpfc_sli_intf_if_type, &phba->sli4_hba.sli_intf) ==
+	else if (bf_get(lpfc_sli_intf_if_type, &phba->sli4_hba.sli_intf) >=
 		 LPFC_SLI_INTF_IF_TYPE_2)
 		rc = lpfc_sli4_bsg_diag_loopback_mode(phba, job);
 	else
@@ -2261,7 +2261,7 @@ lpfc_sli4_bsg_diag_mode_end(struct bsg_job *job)
 
 	if (phba->sli_rev < LPFC_SLI_REV4)
 		return -ENODEV;
-	if (bf_get(lpfc_sli_intf_if_type, &phba->sli4_hba.sli_intf) !=
+	if (bf_get(lpfc_sli_intf_if_type, &phba->sli4_hba.sli_intf) <
 	    LPFC_SLI_INTF_IF_TYPE_2)
 		return -ENODEV;
 
@@ -2353,7 +2353,7 @@ lpfc_sli4_bsg_link_diag_test(struct bsg_job *job)
 		rc = -ENODEV;
 		goto job_error;
 	}
-	if (bf_get(lpfc_sli_intf_if_type, &phba->sli4_hba.sli_intf) !=
+	if (bf_get(lpfc_sli_intf_if_type, &phba->sli4_hba.sli_intf) <
 	    LPFC_SLI_INTF_IF_TYPE_2) {
 		rc = -ENODEV;
 		goto job_error;
drivers/scsi/lpfc/lpfc_els.c +1 -1
index 3651c0fc88197..d0cd368790a64 100644
@@ -5536,7 +5536,7 @@ lpfc_els_rcv_rdp(struct lpfc_vport *vport, struct lpfc_iocbq *cmdiocb,
 	struct ls_rjt stat;
 
 	if (phba->sli_rev < LPFC_SLI_REV4 ||
-	    bf_get(lpfc_sli_intf_if_type, &phba->sli4_hba.sli_intf) !=
+	    bf_get(lpfc_sli_intf_if_type, &phba->sli4_hba.sli_intf) <
 						LPFC_SLI_INTF_IF_TYPE_2) {
 		rjt_err = LSRJT_UNABLE_TPC;
 		rjt_expl = LSEXP_REQ_UNSUPPORTED;
drivers/scsi/lpfc/lpfc_hbadisc.c +1 -1
index ccdd82b1123f7..1a6da347f00ad 100644
@@ -4751,7 +4751,7 @@ lpfc_unreg_rpi(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp)
 				if (phba->sli_rev == LPFC_SLI_REV4 &&
 				    (!(vport->load_flag & FC_UNLOADING)) &&
 				    (bf_get(lpfc_sli_intf_if_type,
-				     &phba->sli4_hba.sli_intf) ==
+				     &phba->sli4_hba.sli_intf) >=
 				      LPFC_SLI_INTF_IF_TYPE_2) &&
 				    (kref_read(&ndlp->kref) > 0)) {
 					mbox->context1 = lpfc_nlp_get(ndlp);
Signature · 1 line
2.20.1
author maintainer
#36 lore raw

PATCH 38/219 scsi: qla2xxx: Fix NPIV handling for FC-NVMe

In reply to Sasha Levin

From: Himanshu Madhani <[email protected]>

[ Upstream commit 5e6803b409ba3c18434de6693062d98a470bcb1e ]

This patch fixes issues with NPIV port with FC-NVMe. Clean up code for remoteport delete and also call nvme_delete when deleting VPs.

---
 drivers/scsi/qla2xxx/qla_nvme.c | 16 +++-------------
 drivers/scsi/qla2xxx/qla_os.c   |  2 ++
 2 files changed, 5 insertions(+), 13 deletions(-)
drivers/scsi/qla2xxx/qla_nvme.c +3 -13
index e6545cb9a2c19..5590d6e8b5762 100644
@@ -474,21 +474,10 @@ static int qla_nvme_post_cmd(struct nvme_fc_local_port *lport,
 	int rval = -ENODEV;
 	srb_t *sp;
 	struct qla_qpair *qpair = hw_queue_handle;
-	struct nvme_private *priv;
+	struct nvme_private *priv = fd->private;
 	struct qla_nvme_rport *qla_rport = rport->private;
 
-	if (!fd || !qpair) {
-		ql_log(ql_log_warn, NULL, 0x2134,
-		    "NO NVMe request or Queue Handle\n");
-		return rval;
-	}
-
-	priv = fd->private;
 	fcport = qla_rport->fcport;
-	if (!fcport) {
-		ql_log(ql_log_warn, NULL, 0x210e, "No fcport ptr\n");
-		return rval;
-	}
 
 	vha = fcport->vha;
 
@@ -517,6 +506,7 @@ static int qla_nvme_post_cmd(struct nvme_fc_local_port *lport,
 	sp->name = "nvme_cmd";
 	sp->done = qla_nvme_sp_done;
 	sp->qpair = qpair;
+	sp->vha = vha;
 	nvme = &sp->u.iocb_cmd;
 	nvme->u.nvme.desc = fd;
 
@@ -564,7 +554,7 @@ static void qla_nvme_remoteport_delete(struct nvme_fc_remote_port *rport)
 		schedule_work(&fcport->free_work);
 	}
 
-	fcport->nvme_flag &= ~(NVME_FLAG_REGISTERED | NVME_FLAG_DELETING);
+	fcport->nvme_flag &= ~NVME_FLAG_DELETING;
 	ql_log(ql_log_info, fcport->vha, 0x2110,
 	    "remoteport_delete of %p completed.\n", fcport);
 }
drivers/scsi/qla2xxx/qla_os.c +2 -0
index 18ee614fe07f5..fb6b5de098f58 100644
@@ -3537,6 +3537,8 @@ qla2x00_delete_all_vps(struct qla_hw_data *ha, scsi_qla_host_t *base_vha)
 		spin_unlock_irqrestore(&ha->vport_slock, flags);
 		mutex_unlock(&ha->vport_lock);
 
+		qla_nvme_delete(vha);
+
 		fc_vport_terminate(vha->fc_vport);
 		scsi_host_put(vha->host);
 
Signature · 1 line
2.20.1
author maintainer
#37 lore raw

PATCH 39/219 scsi: qla2xxx: Fix for FC-NVMe discovery for NPIV port

In reply to Sasha Levin

From: Giridhar Malavali <[email protected]>

[ Upstream commit 835aa4f2691e4ed4ed16de81f3cabf17a87a164f ]

This patch fixes NVMe discovery by setting SKIP_PRLI flag, so that PRLI is driven by driver and is retried when the NPIV port is detected to have NVMe capability.

---
 drivers/scsi/qla2xxx/qla_attr.c |  2 ++
 drivers/scsi/qla2xxx/qla_init.c | 10 ++++------
 2 files changed, 6 insertions(+), 6 deletions(-)
drivers/scsi/qla2xxx/qla_attr.c +2 -0
index 15d493f30810f..3e9c49b3184f1 100644
@@ -2161,6 +2161,8 @@ qla24xx_vport_delete(struct fc_vport *fc_vport)
 	    test_bit(FCPORT_UPDATE_NEEDED, &vha->dpc_flags))
 		msleep(1000);
 
+	qla_nvme_delete(vha);
+
 	qla24xx_disable_vp(vha);
 	qla2x00_wait_for_sess_deletion(vha);
 
drivers/scsi/qla2xxx/qla_init.c +4 -6
index bee9cfb291529..1ae6bb6e70415 100644
@@ -242,15 +242,13 @@ qla2x00_async_login(struct scsi_qla_host *vha, fc_port_t *fcport,
 	qla2x00_init_timer(sp, qla2x00_get_async_timeout(vha) + 2);
 
 	sp->done = qla2x00_async_login_sp_done;
-	if (N2N_TOPO(fcport->vha->hw) && fcport_is_bigger(fcport)) {
+	if (N2N_TOPO(fcport->vha->hw) && fcport_is_bigger(fcport))
 		lio->u.logio.flags |= SRB_LOGIN_PRLI_ONLY;
-	} else {
+	else
 		lio->u.logio.flags |= SRB_LOGIN_COND_PLOGI;
 
-		if (fcport->fc4f_nvme)
-			lio->u.logio.flags |= SRB_LOGIN_SKIP_PRLI;
-
-	}
+	if (fcport->fc4f_nvme)
+		lio->u.logio.flags |= SRB_LOGIN_SKIP_PRLI;
 
 	ql_dbg(ql_dbg_disc, vha, 0x2072,
 	    "Async-login - %8phC hdl=%x, loopid=%x portid=%02x%02x%02x "
Signature · 1 line
2.20.1
author maintainer
#38 lore raw

PATCH 40/219 nvme: provide fallback for discard alloc failure

In reply to Sasha Levin

From: Jens Axboe <[email protected]>

[ Upstream commit cb5b7262b011cfb793519bf97e54dff5282da23c ]

When boxes are run near (or to) OOM, we have a problem with the discard page allocation in nvme. If we fail allocating the special page, we return busy, and it'll get retried. But since ordering is honored for dispatch requests, we can keep retrying this same IO and failing. Behind that IO could be requests that want to free memory, but they never get the chance.

Allocate a fixed discard page per controller for a safe fallback, and use that if the initial allocation fails.

---
 drivers/nvme/host/core.c | 41 ++++++++++++++++++++++++++++++++++------
 drivers/nvme/host/nvme.h |  3 +++
 2 files changed, 38 insertions(+), 6 deletions(-)
drivers/nvme/host/core.c +35 -6
index 5d0f99bcc987f..ced70d37e0f9b 100644
@@ -551,9 +551,19 @@ static blk_status_t nvme_setup_discard(struct nvme_ns *ns, struct request *req,
 	struct nvme_dsm_range *range;
 	struct bio *bio;
 
-	range = kmalloc_array(segments, sizeof(*range), GFP_ATOMIC);
-	if (!range)
-		return BLK_STS_RESOURCE;
+	range = kmalloc_array(segments, sizeof(*range),
+				GFP_ATOMIC | __GFP_NOWARN);
+	if (!range) {
+		/*
+		 * If we fail allocation our range, fallback to the controller
+		 * discard page. If that's also busy, it's safe to return
+		 * busy, as we know we can make progress once that's freed.
+		 */
+		if (test_and_set_bit_lock(0, &ns->ctrl->discard_page_busy))
+			return BLK_STS_RESOURCE;
+
+		range = page_address(ns->ctrl->discard_page);
+	}
 
 	__rq_for_each_bio(bio, req) {
 		u64 slba = nvme_block_nr(ns, bio->bi_iter.bi_sector);
@@ -568,7 +578,10 @@ static blk_status_t nvme_setup_discard(struct nvme_ns *ns, struct request *req,
 	}
 
 	if (WARN_ON_ONCE(n != segments)) {
-		kfree(range);
+		if (virt_to_page(range) == ns->ctrl->discard_page)
+			clear_bit_unlock(0, &ns->ctrl->discard_page_busy);
+		else
+			kfree(range);
 		return BLK_STS_IOERR;
 	}
 
@@ -653,8 +666,13 @@ void nvme_cleanup_cmd(struct request *req)
 				blk_rq_bytes(req) >> ns->lba_shift);
 	}
 	if (req->rq_flags & RQF_SPECIAL_PAYLOAD) {
-		kfree(page_address(req->special_vec.bv_page) +
-		      req->special_vec.bv_offset);
+		struct nvme_ns *ns = req->rq_disk->private_data;
+		struct page *page = req->special_vec.bv_page;
+
+		if (page == ns->ctrl->discard_page)
+			clear_bit_unlock(0, &ns->ctrl->discard_page_busy);
+		else
+			kfree(page_address(page) + req->special_vec.bv_offset);
 	}
 }
 EXPORT_SYMBOL_GPL(nvme_cleanup_cmd);
@@ -3551,6 +3569,7 @@ static void nvme_free_ctrl(struct device *dev)
 	ida_simple_remove(&nvme_instance_ida, ctrl->instance);
 	kfree(ctrl->effects);
 	nvme_mpath_uninit(ctrl);
+	kfree(ctrl->discard_page);
 
 	if (subsys) {
 		mutex_lock(&subsys->lock);
@@ -3592,6 +3611,14 @@ int nvme_init_ctrl(struct nvme_ctrl *ctrl, struct device *dev,
 	memset(&ctrl->ka_cmd, 0, sizeof(ctrl->ka_cmd));
 	ctrl->ka_cmd.common.opcode = nvme_admin_keep_alive;
 
+	BUILD_BUG_ON(NVME_DSM_MAX_RANGES * sizeof(struct nvme_dsm_range) >
+			PAGE_SIZE);
+	ctrl->discard_page = alloc_page(GFP_KERNEL);
+	if (!ctrl->discard_page) {
+		ret = -ENOMEM;
+		goto out;
+	}
+
 	ret = ida_simple_get(&nvme_instance_ida, 0, 0, GFP_KERNEL);
 	if (ret < 0)
 		goto out;
@@ -3629,6 +3656,8 @@ int nvme_init_ctrl(struct nvme_ctrl *ctrl, struct device *dev,
 out_release_instance:
 	ida_simple_remove(&nvme_instance_ida, ctrl->instance);
 out:
+	if (ctrl->discard_page)
+		__free_page(ctrl->discard_page);
 	return ret;
 }
 EXPORT_SYMBOL_GPL(nvme_init_ctrl);
drivers/nvme/host/nvme.h +3 -0
index 2653e1f4196d5..cc4273f119894 100644
@@ -238,6 +238,9 @@ struct nvme_ctrl {
 	u16 maxcmd;
 	int nr_reconnects;
 	struct nvmf_ctrl_options *opts;
+
+	struct page *discard_page;
+	unsigned long discard_page_busy;
 };
 
 struct nvme_subsystem {
Signature · 1 line
2.20.1
author maintainer
#39 lore raw

PATCH 41/219 s390/zcrypt: make sysfs reset attribute trigger queue reset

In reply to Sasha Levin

From: Harald Freudenberger <[email protected]>

[ Upstream commit 42a87d4103ae365e18c3be1333592ab583b8ad9d ]

Until now there is no way to reset a AP queue or card. Driving a card or queue offline and online again does only toggle the 'software' online state. The only way to trigger a (hardware) reset is by running hot-unplug/hot-plug for example on the HMC.

This patch makes the queue reset attribute in sysfs writable. Writing into this attribute triggers a reset on the AP queue's state machine. So the AP queue is flushed and state machine runs through the initial states which cause a reset (PQAP(RAPQ)) and a re-registration to interrupts (PQAP(AQIC)) if available.

The reset sysfs attribute is writable by root only. So only an administrator is allowed to initiate a reset of AP queues. Please note that the queue's counter values are left untouched by the reset.

---
 drivers/s390/crypto/ap_queue.c | 23 ++++++++++++++++++++++-
 1 file changed, 22 insertions(+), 1 deletion(-)
drivers/s390/crypto/ap_queue.c +22 -1
index 0aa4b3ccc948c..576ac08777c50 100644
@@ -14,6 +14,9 @@
 #include <asm/facility.h>
 
 #include "ap_bus.h"
+#include "ap_debug.h"
+
+static void __ap_flush_queue(struct ap_queue *aq);
 
 /**
  * ap_queue_enable_interruption(): Enable interruption on an AP queue.
@@ -541,7 +544,25 @@ static ssize_t reset_show(struct device *dev,
 	return rc;
 }
 
-static DEVICE_ATTR_RO(reset);
+static ssize_t reset_store(struct device *dev,
+			   struct device_attribute *attr,
+			   const char *buf, size_t count)
+{
+	struct ap_queue *aq = to_ap_queue(dev);
+
+	spin_lock_bh(&aq->lock);
+	__ap_flush_queue(aq);
+	aq->state = AP_STATE_RESET_START;
+	ap_wait(ap_sm_event(aq, AP_EVENT_POLL));
+	spin_unlock_bh(&aq->lock);
+
+	AP_DBF(DBF_INFO, "reset queue=%02x.%04x triggered by user\n",
+	       AP_QID_CARD(aq->qid), AP_QID_QUEUE(aq->qid));
+
+	return count;
+}
+
+static DEVICE_ATTR_RW(reset);
 
 static ssize_t interrupt_show(struct device *dev,
 			      struct device_attribute *attr, char *buf)
Signature · 1 line
2.20.1
author maintainer
#40 lore raw

PATCH 42/219 crypto: user - support incremental algorithm dumps

In reply to Sasha Levin

From: Eric Biggers <[email protected]>

[ Upstream commit 0ac6b8fb23c724b015d9ca70a89126e8d1563166 ]

CRYPTO_MSG_GETALG in NLM_F_DUMP mode sometimes doesn't return all registered crypto algorithms, because it doesn't support incremental dumps. crypto_dump_report() only permits itself to be called once, yet the netlink subsystem allocates at most ~64 KiB for the skb being dumped to. Thus only the first recvmsg() returns data, and it may only include a subset of the crypto algorithms even if the user buffer passed to recvmsg() is large enough to hold all of them.

Fix this by using one of the arguments in the netlink_callback structure to keep track of the current position in the algorithm list. Then userspace can do multiple recvmsg() on the socket after sending the dump request. This is the way netlink dumps work elsewhere in the kernel; it's unclear why this was different (probably just an oversight).

Also fix an integer overflow when calculating the dump buffer size hint.

Fixes a38f7907b926 ("crypto: Add userspace configuration API")
---
 crypto/crypto_user.c | 37 ++++++++++++++++++++-----------------
 1 file changed, 20 insertions(+), 17 deletions(-)
crypto/crypto_user.c +20 -17
index 3cca814348a26..74cb166097cd2 100644
@@ -296,30 +296,33 @@ static int crypto_report(struct sk_buff *in_skb, struct nlmsghdr *in_nlh,
 
 static int crypto_dump_report(struct sk_buff *skb, struct netlink_callback *cb)
 {
-	struct crypto_alg *alg;
+	const size_t start_pos = cb->args[0];
+	size_t pos = 0;
 	struct crypto_dump_info info;
-	int err;
-
-	if (cb->args[0])
-		goto out;
-
-	cb->args[0] = 1;
+	struct crypto_alg *alg;
+	int res;
 
 	info.in_skb = cb->skb;
 	info.out_skb = skb;
 	info.nlmsg_seq = cb->nlh->nlmsg_seq;
 	info.nlmsg_flags = NLM_F_MULTI;
 
+	down_read(&crypto_alg_sem);
 	list_for_each_entry(alg, &crypto_alg_list, cra_list) {
-		err = crypto_report_alg(alg, &info);
-		if (err)
-			goto out_err;
+		if (pos >= start_pos) {
+			res = crypto_report_alg(alg, &info);
+			if (res == -EMSGSIZE)
+				break;
+			if (res)
+				goto out;
+		}
+		pos++;
 	}
-
+	cb->args[0] = pos;
+	res = skb->len;
 out:
-	return skb->len;
-out_err:
-	return err;
+	up_read(&crypto_alg_sem);
+	return res;
 }
 
 static int crypto_dump_report_done(struct netlink_callback *cb)
@@ -503,7 +506,7 @@ static int crypto_user_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh,
 	if ((type == (CRYPTO_MSG_GETALG - CRYPTO_MSG_BASE) &&
 	    (nlh->nlmsg_flags & NLM_F_DUMP))) {
 		struct crypto_alg *alg;
-		u16 dump_alloc = 0;
+		unsigned long dump_alloc = 0;
 
 		if (link->dump == NULL)
 			return -EINVAL;
@@ -511,16 +514,16 @@ static int crypto_user_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh,
 		down_read(&crypto_alg_sem);
 		list_for_each_entry(alg, &crypto_alg_list, cra_list)
 			dump_alloc += CRYPTO_REPORT_MAXSIZE;
+		up_read(&crypto_alg_sem);
 
 		{
 			struct netlink_dump_control c = {
 				.dump = link->dump,
 				.done = link->done,
-				.min_dump_alloc = dump_alloc,
+				.min_dump_alloc = min(dump_alloc, 65535UL),
 			};
 			err = netlink_dump_start(crypto_nlsk, skb, nlh, &c);
 		}
-		up_read(&crypto_alg_sem);
 
 		return err;
 	}
Signature · 1 line
2.20.1
author maintainer
#41 lore raw

PATCH 43/219 arm64: dts: renesas: draak: Fix CVBS input

In reply to Sasha Levin

From: Laurent Pinchart <[email protected]>

[ Upstream commit 6f61a2c8f1f6163c7e08c77c5f71df0427e4d2f6 ]

A typo in the adv7180 DT node prevents successful probing of the VIN. Fix it.

Fixes 6a0942c20f5c ("arm64: dts: renesas: draak: Describe CVBS input")
---
 arch/arm64/boot/dts/renesas/r8a77995-draak.dts | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm64/boot/dts/renesas/r8a77995-draak.dts +1 -1
index a8e8f2669d4c5..1b8f19ee257f0 100644
@@ -188,7 +188,7 @@
 		compatible = "adi,adv7180cp";
 		reg = <0x20>;
 
-		port {
+		ports {
 			#address-cells = <1>;
 			#size-cells = <0>;
 
Signature · 1 line
2.20.1
author maintainer
#42 lore raw

PATCH 44/219 mwifiex: fix potential NULL dereference and use after free

In reply to Sasha Levin

From: Pan Bian <[email protected]>

[ Upstream commit 1dcd9429212b98bea87fc6ec92fb50bf5953eb47 ]

There are two defects: (1) passing a NULL bss to mwifiex_save_hidden_ssid_channels will result in NULL dereference,
(2) using bss after dropping the reference to it via cfg80211_put_bss. To fix them, the patch moves the buggy code to the branch that bss is not NULL and puts it before cfg80211_put_bss.

---
 drivers/net/wireless/marvell/mwifiex/scan.c | 18 ++++++++++--------
 1 file changed, 10 insertions(+), 8 deletions(-)
drivers/net/wireless/marvell/mwifiex/scan.c +10 -8
index ed27147efcb37..dd02bbd9544e7 100644
@@ -1906,15 +1906,17 @@ mwifiex_parse_single_response_buf(struct mwifiex_private *priv, u8 **bss_info,
 					    ETH_ALEN))
 					mwifiex_update_curr_bss_params(priv,
 								       bss);
-				cfg80211_put_bss(priv->wdev.wiphy, bss);
-			}
 
-			if ((chan->flags & IEEE80211_CHAN_RADAR) ||
-			    (chan->flags & IEEE80211_CHAN_NO_IR)) {
-				mwifiex_dbg(adapter, INFO,
-					    "radar or passive channel %d\n",
-					    channel);
-				mwifiex_save_hidden_ssid_channels(priv, bss);
+				if ((chan->flags & IEEE80211_CHAN_RADAR) ||
+				    (chan->flags & IEEE80211_CHAN_NO_IR)) {
+					mwifiex_dbg(adapter, INFO,
+						    "radar or passive channel %d\n",
+						    channel);
+					mwifiex_save_hidden_ssid_channels(priv,
+									  bss);
+				}
+
+				cfg80211_put_bss(priv->wdev.wiphy, bss);
 			}
 		}
 	} else {
Signature · 1 line
2.20.1
author maintainer
#43 lore raw

PATCH 45/219 mwifiex: debugfs: correct histogram spacing, formatting

In reply to Sasha Levin

From: Brian Norris <[email protected]>

[ Upstream commit 4cb777c64e030778c569f605398d7604d8aabc0f ]

Currently, snippets of this file look like:

rx rates (in Mbps): 0=1M 1=2M2=5.5M 3=11M 4=6M 5=9M 6=12M
7=18M 8=24M 9=36M 10=48M 11=54M12-27=MCS0-15(BW20) 28-43=MCS0-15(BW40) 44-53=MCS0-9(VHT:BW20)54-63=MCS0-9(VHT:BW40)64-73=MCS0-9(VHT:BW80)
...
noise_flr[--96dBm] = 22
noise_flr[--95dBm] = 149
noise_flr[--94dBm] = 9
noise_flr[--93dBm] = 2

We're missing some spaces, and we're adding a minus sign ('-') on values that are already negative signed integers.

---
 drivers/net/wireless/marvell/mwifiex/debugfs.c | 14 ++++++--------
 1 file changed, 6 insertions(+), 8 deletions(-)
drivers/net/wireless/marvell/mwifiex/debugfs.c +6 -8
index cce70252fd96b..cbe4493b32664 100644
@@ -273,15 +273,13 @@ mwifiex_histogram_read(struct file *file, char __user *ubuf,
 		     "total samples = %d\n",
 		     atomic_read(&phist_data->num_samples));
 
-	p += sprintf(p, "rx rates (in Mbps): 0=1M   1=2M");
-	p += sprintf(p, "2=5.5M  3=11M   4=6M   5=9M  6=12M\n");
-	p += sprintf(p, "7=18M  8=24M  9=36M  10=48M  11=54M");
-	p += sprintf(p, "12-27=MCS0-15(BW20) 28-43=MCS0-15(BW40)\n");
+	p += sprintf(p,
+		     "rx rates (in Mbps): 0=1M   1=2M 2=5.5M  3=11M   4=6M   5=9M  6=12M\n"
+		     "7=18M  8=24M  9=36M  10=48M  11=54M 12-27=MCS0-15(BW20) 28-43=MCS0-15(BW40)\n");
 
 	if (ISSUPP_11ACENABLED(priv->adapter->fw_cap_info)) {
-		p += sprintf(p, "44-53=MCS0-9(VHT:BW20)");
-		p += sprintf(p, "54-63=MCS0-9(VHT:BW40)");
-		p += sprintf(p, "64-73=MCS0-9(VHT:BW80)\n\n");
+		p += sprintf(p,
+			     "44-53=MCS0-9(VHT:BW20) 54-63=MCS0-9(VHT:BW40) 64-73=MCS0-9(VHT:BW80)\n\n");
 	} else {
 		p += sprintf(p, "\n");
 	}
@@ -310,7 +308,7 @@ mwifiex_histogram_read(struct file *file, char __user *ubuf,
 	for (i = 0; i < MWIFIEX_MAX_NOISE_FLR; i++) {
 		value = atomic_read(&phist_data->noise_flr[i]);
 		if (value)
-			p += sprintf(p, "noise_flr[-%02ddBm] = %d\n",
+			p += sprintf(p, "noise_flr[%02ddBm] = %d\n",
 				(int)(i-128), value);
 	}
 	for (i = 0; i < MWIFIEX_MAX_SIG_STRENGTH; i++) {
Signature · 1 line
2.20.1
author maintainer
#44 lore raw

PATCH 46/219 brcmfmac: set F2 watermark to 256 for 4373

In reply to Sasha Levin

From: Wright Feng <[email protected]>

[ Upstream commit e1a08730eeb0bad4d82c3bc40e74854872de618d ]

We got SDIO_CRC_ERROR with 4373 on SDR104 when doing bi-directional throughput test. Enable watermark to 256 to guarantee the operation stability.

---
 .../broadcom/brcm80211/brcmfmac/sdio.c        | 26 +++++++++++++++++--
 1 file changed, 24 insertions(+), 2 deletions(-)
drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c +24 -2
index 53e4962ceb8ae..e487dd78cc024 100644
@@ -49,6 +49,10 @@
 #define DCMD_RESP_TIMEOUT	msecs_to_jiffies(2500)
 #define CTL_DONE_TIMEOUT	msecs_to_jiffies(2500)
 
+/* watermark expressed in number of words */
+#define DEFAULT_F2_WATERMARK    0x8
+#define CY_4373_F2_WATERMARK    0x40
+
 #ifdef DEBUG
 
 #define BRCMF_TRAP_INFO_SIZE	80
@@ -138,6 +142,8 @@ struct rte_console {
 /* 1: isolate internal sdio signals, put external pads in tri-state; requires
  * sdio bus power cycle to clear (rev 9) */
 #define SBSDIO_DEVCTL_PADS_ISO		0x08
+/* 1: enable F2 Watermark */
+#define SBSDIO_DEVCTL_F2WM_ENAB		0x10
 /* Force SD->SB reset mapping (rev 11) */
 #define SBSDIO_DEVCTL_SB_RST_CTL	0x30
 /*   Determined by CoreControl bit */
@@ -4060,6 +4066,7 @@ static void brcmf_sdio_firmware_callback(struct device *dev, int err,
 	void *nvram;
 	u32 nvram_len;
 	u8 saveclk;
+	u8 devctl;
 
 	brcmf_dbg(TRACE, "Enter: dev=%s, err=%d\n", dev_name(dev), err);
 
@@ -4115,8 +4122,23 @@ static void brcmf_sdio_firmware_callback(struct device *dev, int err,
 		brcmf_sdiod_writel(sdiod, core->base + SD_REG(hostintmask),
 				   bus->hostintmask, NULL);
 
-
-		brcmf_sdiod_writeb(sdiod, SBSDIO_WATERMARK, 8, &err);
+		switch (sdiod->func1->device) {
+		case SDIO_DEVICE_ID_CYPRESS_4373:
+			brcmf_dbg(INFO, "set F2 watermark to 0x%x*4 bytes\n",
+				  CY_4373_F2_WATERMARK);
+			brcmf_sdiod_writeb(sdiod, SBSDIO_WATERMARK,
+					   CY_4373_F2_WATERMARK, &err);
+			devctl = brcmf_sdiod_readb(sdiod, SBSDIO_DEVICE_CTL,
+						   &err);
+			devctl |= SBSDIO_DEVCTL_F2WM_ENAB;
+			brcmf_sdiod_writeb(sdiod, SBSDIO_DEVICE_CTL, devctl,
+					   &err);
+			break;
+		default:
+			brcmf_sdiod_writeb(sdiod, SBSDIO_WATERMARK,
+					   DEFAULT_F2_WATERMARK, &err);
+			break;
+		}
 	} else {
 		/* Disable F2 again */
 		sdio_disable_func(sdiod->func2);
Signature · 1 line
2.20.1
author maintainer
#45 lore raw

PATCH 47/219 brcmfmac: set SDIO F1 MesBusyCtrl for CYW4373

In reply to Sasha Levin

From: Madhan Mohan R <[email protected]>

[ Upstream commit 58e4bbea0c1d9b5ace11df968c5dc096ce052a73 ]

Along with F2 watermark (existing) configuration, F1 MesBusyCtrl should be enabled & sdio device RX FIFO watermark should be configured to avoid overflow errors.

---
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c | 3 +++
 drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.h | 9 ++++++++-
 2 files changed, 11 insertions(+), 1 deletion(-)
drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c +3 -0
index e487dd78cc024..abaed2fa2defd 100644
@@ -4133,6 +4133,9 @@ static void brcmf_sdio_firmware_callback(struct device *dev, int err,
 			devctl |= SBSDIO_DEVCTL_F2WM_ENAB;
 			brcmf_sdiod_writeb(sdiod, SBSDIO_DEVICE_CTL, devctl,
 					   &err);
+			brcmf_sdiod_writeb(sdiod, SBSDIO_FUNC1_MESBUSYCTRL,
+					   CY_4373_F2_WATERMARK |
+					   SBSDIO_MESBUSYCTRL_ENAB, &err);
 			break;
 		default:
 			brcmf_sdiod_writeb(sdiod, SBSDIO_WATERMARK,
drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.h +8 -1
index 7faed831f07d5..34b031154da93 100644
@@ -77,7 +77,7 @@
 #define SBSDIO_GPIO_OUT			0x10006
 /* gpio enable */
 #define SBSDIO_GPIO_EN			0x10007
-/* rev < 7, watermark for sdio device */
+/* rev < 7, watermark for sdio device TX path */
 #define SBSDIO_WATERMARK		0x10008
 /* control busy signal generation */
 #define SBSDIO_DEVICE_CTL		0x10009
@@ -104,6 +104,13 @@
 #define SBSDIO_FUNC1_RFRAMEBCHI		0x1001C
 /* MesBusyCtl (rev 11) */
 #define SBSDIO_FUNC1_MESBUSYCTRL	0x1001D
+/* Watermark for sdio device RX path */
+#define SBSDIO_MESBUSY_RXFIFO_WM_MASK	0x7F
+#define SBSDIO_MESBUSY_RXFIFO_WM_SHIFT	0
+/* Enable busy capability for MES access */
+#define SBSDIO_MESBUSYCTRL_ENAB		0x80
+#define SBSDIO_MESBUSYCTRL_ENAB_SHIFT	7
+
 /* Sdio Core Rev 12 */
 #define SBSDIO_FUNC1_WAKEUPCTRL		0x1001E
 #define SBSDIO_FUNC1_WCTRL_ALPWAIT_MASK		0x1
Signature · 1 line
2.20.1
author maintainer
#46 lore raw

PATCH 48/219 rtl818x: fix potential use after free

In reply to Sasha Levin

From: Pan Bian <[email protected]>

[ Upstream commit afbb1947db94eacc5a13302eee88a9772fb78935 ]

entry is released via usb_put_urb just after calling usb_submit_urb. However, entry is used if the submission fails, resulting in a use after free bug. The patch fixes this.

---
 drivers/net/wireless/realtek/rtl818x/rtl8187/dev.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/wireless/realtek/rtl818x/rtl8187/dev.c +2 -1
index 9a1d15b3ce453..518caaaf8a987 100644
@@ -444,12 +444,13 @@ static int rtl8187_init_urbs(struct ieee80211_hw *dev)
 		skb_queue_tail(&priv->rx_queue, skb);
 		usb_anchor_urb(entry, &priv->anchored);
 		ret = usb_submit_urb(entry, GFP_KERNEL);
-		usb_put_urb(entry);
 		if (ret) {
 			skb_unlink(skb, &priv->rx_queue);
 			usb_unanchor_urb(entry);
+			usb_put_urb(entry);
 			goto err;
 		}
+		usb_put_urb(entry);
 	}
 	return ret;
 
Signature · 1 line
2.20.1
author maintainer
#47 lore raw

PATCH 49/219 bcache: do not check if debug dentry is ERR or NULL explicitly on remove

In reply to Sasha Levin

From: Shenghui Wang <[email protected]>

[ Upstream commit ae17102316550b4b230a283febe31b2a9ff30084 ]

debugfs_remove and debugfs_remove_recursive will check if the dentry pointer is NULL or ERR, and will do nothing in that case.

Remove the check in cache_set_free and bch_debug_init.

---
 drivers/md/bcache/debug.c | 3 +--
 drivers/md/bcache/super.c | 3 +--
 2 files changed, 2 insertions(+), 4 deletions(-)
drivers/md/bcache/debug.c +1 -2
index 06da66b2488ae..8c53d874ada4a 100644
@@ -249,8 +249,7 @@ void bch_debug_init_cache_set(struct cache_set *c)
 
 void bch_debug_exit(void)
 {
-	if (!IS_ERR_OR_NULL(bcache_debug))
-		debugfs_remove_recursive(bcache_debug);
+	debugfs_remove_recursive(bcache_debug);
 }
 
 void __init bch_debug_init(struct kobject *kobj)
drivers/md/bcache/super.c +1 -2
index 2321643974dab..3124399e2607a 100644
@@ -1485,8 +1485,7 @@ static void cache_set_free(struct closure *cl)
 	struct cache *ca;
 	unsigned int i;
 
-	if (!IS_ERR_OR_NULL(c->debug))
-		debugfs_remove(c->debug);
+	debugfs_remove(c->debug);
 
 	bch_open_buckets_free(c);
 	bch_btree_cache_free(c);
Signature · 1 line
2.20.1
author maintainer
#48 lore raw

PATCH 50/219 bcache: do not mark writeback_running too early

In reply to Sasha Levin

From: Shenghui Wang <[email protected]>

[ Upstream commit 79b791466e525c98f6aeee9acf5726e7b27f4231 ]

A fresh backing device is not attached to any cache_set, and has no writeback kthread created until first attached to some cache_set.

But bch_cached_dev_writeback_init run
"

	dc->writeback_running		= true;
	WARN_ON(test_and_clear_bit(BCACHE_DEV_WB_RUNNING,
			&dc->disk.flags));

"
for any newly formatted backing devices.

For a fresh standalone backing device, we can get something like following even if no writeback kthread created:
------------------------
/sys/block/bcache0/bcache# cat writeback_running
1
/sys/block/bcache0/bcache# cat writeback_rate_debug
rate: 512.0k/sec
dirty: 0.0k
target: 0.0k
proportional: 0.0k
integral: 0.0k
change: 0.0k/sec
next io: -15427384ms

The none ZERO fields are misleading as no alive writeback kthread yet.

Set dc->writeback_running false as no writeback thread created in bch_cached_dev_writeback_init().

We have writeback thread created and woken up in bch_cached_dev_writeback _start(). Set dc->writeback_running true before bch_writeback_queue() called, as a writeback thread will check if dc->writeback_running is true before writing back dirty data, and hung if false detected.

After the change, we can get the following output for a fresh standalone backing device:
-----------------------
/sys/block/bcache0/bcache$ cat writeback_running
0
/sys/block/bcache0/bcache# cat writeback_rate_debug
rate: 0.0k/sec
dirty: 0.0k
target: 0.0k
proportional: 0.0k
integral: 0.0k
change: 0.0k/sec
next io: 0ms

v1 -> v2:
Set dc->writeback_running before bch_writeback_queue() called,

---
 drivers/md/bcache/writeback.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/md/bcache/writeback.c +2 -1
index ba5395fd386d5..b5fc3c6c7178e 100644
@@ -781,7 +781,7 @@ void bch_cached_dev_writeback_init(struct cached_dev *dc)
 	bch_keybuf_init(&dc->writeback_keys);
 
 	dc->writeback_metadata		= true;
-	dc->writeback_running		= true;
+	dc->writeback_running		= false;
 	dc->writeback_percent		= 10;
 	dc->writeback_delay		= 30;
 	atomic_long_set(&dc->writeback_rate.rate, 1024);
@@ -810,6 +810,7 @@ int bch_cached_dev_writeback_start(struct cached_dev *dc)
 		destroy_workqueue(dc->writeback_write_wq);
 		return PTR_ERR(dc->writeback_thread);
 	}
+	dc->writeback_running = true;
 
 	WARN_ON(test_and_set_bit(BCACHE_DEV_WB_RUNNING, &dc->disk.flags));
 	schedule_delayed_work(&dc->writeback_rate_update,
Signature · 1 line
2.20.1
author maintainer
#49 lore raw

PATCH 51/219 xfs: require both realtime inodes to mount

In reply to Sasha Levin

From: "Darrick J. Wong" <[email protected]>

[ Upstream commit 64bafd2f1e484e27071e7584642005d56516cb77 ]

Since mkfs always formats the filesystem with the realtime bitmap and summary inodes immediately after the root directory, we should expect that both of them are present and loadable, even if there isn't a realtime volume attached. There's no reason to skip this if rbmino == NULLFSINO; in fact, this causes an immediate crash if the there /is/ a realtime volume and someone writes to it.

---
 fs/xfs/xfs_rtalloc.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)
fs/xfs/xfs_rtalloc.c +1 -3
index 926ed314ffba1..484eb0adcefb2 100644
@@ -1198,13 +1198,11 @@ xfs_rtmount_inodes(
 	xfs_sb_t	*sbp;
 
 	sbp = &mp->m_sb;
-	if (sbp->sb_rbmino == NULLFSINO)
-		return 0;
 	error = xfs_iget(mp, NULL, sbp->sb_rbmino, 0, 0, &mp->m_rbmip);
 	if (error)
 		return error;
 	ASSERT(mp->m_rbmip != NULL);
-	ASSERT(sbp->sb_rsumino != NULLFSINO);
+
 	error = xfs_iget(mp, NULL, sbp->sb_rsumino, 0, 0, &mp->m_rsumip);
 	if (error) {
 		xfs_irele(mp->m_rbmip);
Signature · 1 line
2.20.1
author maintainer
#50 lore raw

PATCH 52/219 nvme: fix kernel paging oops

In reply to Sasha Levin

From: Sagi Grimberg <[email protected]>

[ Upstream commit 092ff0520070fad8407b196f3bb6156ce77a6f98 ]

free the controller discard_page correctly.

Fixes cb5b7262b011 ("nvme: provide fallback for discard alloc failure")
---
 drivers/nvme/host/core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/nvme/host/core.c +1 -1
index ced70d37e0f9b..e9060317c1a7b 100644
@@ -3569,7 +3569,7 @@ static void nvme_free_ctrl(struct device *dev)
 	ida_simple_remove(&nvme_instance_ida, ctrl->instance);
 	kfree(ctrl->effects);
 	nvme_mpath_uninit(ctrl);
-	kfree(ctrl->discard_page);
+	__free_page(ctrl->discard_page);
 
 	if (subsys) {
 		mutex_lock(&subsys->lock);
Signature · 7 lines
2.20.1


_______________________________________________
Linux-nvme mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-nvme
author maintainer
#51 lore raw

PATCH 53/219 ubifs: Fix default compression selection in ubifs

In reply to Sasha Levin

From: Gabor Juhos <[email protected]>

[ Upstream commit d62e98ed1efcaa94caa004f622944afdce5f1c3c ]

When ubifs is build without the LZO compressor and no compressor is given the creation of the default file system will fail. before selection the LZO compressor check if it is present and if not fall back to the zlib or none.

---
 fs/ubifs/sb.c | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)
fs/ubifs/sb.c +12 -1
index bf17f58908ff9..11dc3977fb64f 100644
@@ -63,6 +63,17 @@
 /* Default time granularity in nanoseconds */
 #define DEFAULT_TIME_GRAN 1000000000
 
+static int get_default_compressor(struct ubifs_info *c)
+{
+	if (ubifs_compr_present(c, UBIFS_COMPR_LZO))
+		return UBIFS_COMPR_LZO;
+
+	if (ubifs_compr_present(c, UBIFS_COMPR_ZLIB))
+		return UBIFS_COMPR_ZLIB;
+
+	return UBIFS_COMPR_NONE;
+}
+
 /**
  * create_default_filesystem - format empty UBI volume.
  * @c: UBIFS file-system description object
@@ -186,7 +197,7 @@ static int create_default_filesystem(struct ubifs_info *c)
 	if (c->mount_opts.override_compr)
 		sup->default_compr = cpu_to_le16(c->mount_opts.compr_type);
 	else
-		sup->default_compr = cpu_to_le16(UBIFS_COMPR_LZO);
+		sup->default_compr = cpu_to_le16(get_default_compressor(c));
 
 	generate_random_uuid(sup->uuid);
 
Signature · 6 lines
2.20.1


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/
author maintainer
#52 lore raw

PATCH 54/219 ubi: Put MTD device after it is not used

In reply to Sasha Levin

From: Pan Bian <[email protected]>

[ Upstream commit b95f83ab762dd6211351b9140f99f43644076ca8 ]

The MTD device reference is dropped via put_mtd_device, however its field ->index is read and passed to ubi_msg. To fix this, the patch moves the reference dropping after calling ubi_msg.

---
 drivers/mtd/ubi/build.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/mtd/ubi/build.c +1 -1
index d2a726654ff11..c120c8761fcd2 100644
@@ -1101,10 +1101,10 @@ int ubi_detach_mtd_dev(int ubi_num, int anyway)
 	ubi_wl_close(ubi);
 	ubi_free_internal_volumes(ubi);
 	vfree(ubi->vtbl);
-	put_mtd_device(ubi->mtd);
 	vfree(ubi->peb_buf);
 	vfree(ubi->fm_buf);
 	ubi_msg(ubi, "mtd%d is detached", ubi->mtd->index);
+	put_mtd_device(ubi->mtd);
 	put_device(&ubi->dev);
 	return 0;
 }
Signature · 1 line
2.20.1
author maintainer
#53 lore raw

PATCH 55/219 ubi: Do not drop UBI device reference before using

In reply to Sasha Levin

From: Pan Bian <[email protected]>

[ Upstream commit e542087701f09418702673631a908429feb3eae0 ]

The UBI device reference is dropped but then the device is used as a parameter of ubi_err. The bug is introduced in changing ubi_err's behavior. The old ubi_err does not require a UBI device as its first parameter, but the new one does.

Fixes 32608703310 ("UBI: Extend UBI layer debug/messaging capabilities")
---
 drivers/mtd/ubi/kapi.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/mtd/ubi/kapi.c +1 -1
index e9e9ecbcedcc3..0b8f0c46268da 100644
@@ -227,9 +227,9 @@ struct ubi_volume_desc *ubi_open_volume(int ubi_num, int vol_id, int mode)
 out_free:
 	kfree(desc);
 out_put_ubi:
-	ubi_put_device(ubi);
 	ubi_err(ubi, "cannot open device %d, volume %d, error %d",
 		ubi_num, vol_id, err);
+	ubi_put_device(ubi);
 	return ERR_PTR(err);
 }
 EXPORT_SYMBOL_GPL(ubi_open_volume);
Signature · 1 line
2.20.1
author maintainer
#54 lore raw

PATCH 56/219 microblaze: adjust the help to the real behavior

In reply to Sasha Levin

From: Masahiro Yamada <[email protected]>

[ Upstream commit bafcc61d998c1ca18f556d92a0e95335ac68c7da ]

"make ARCH=microblaze help" mentions simpleImage.<dt>.unstrip, but it is not a real Make target. It does not work because Makefile assumes "system.unstrip" is the name of DT.

$ make ARCH=microblaze CROSS_COMPILE=microblaze-linux- simpleImage.system.unstrip
  [ snip ]
make[1]: *** No rule to make target 'arch/microblaze/boot/dts/system.unstrip.dtb', needed by 'arch/microblaze/boot/dts/system.dtb'.  Stop.
make: *** [Makefile;1060: arch/microblaze/boot/dts] Error 2
make: *** Waiting for unfinished jobs....

simpleImage.<dt> works like a phony target that generates multiple images. Reflect the real behavior. I removed the DT directory path information because it is already explained a few lines below.

While I am here, I deleted the redundant *_defconfig explanation.

The top-level Makefile caters to list available defconfig files:

mmu_defconfig - Build for mmu
nommu_defconfig - Build for nommu

---
 arch/microblaze/Makefile | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)
arch/microblaze/Makefile +5 -5
index 4f3ab57072652..eecf37276521c 100644
@@ -91,11 +91,11 @@ define archhelp
   echo '* linux.bin    - Create raw binary'
   echo '  linux.bin.gz - Create compressed raw binary'
   echo '  linux.bin.ub - Create U-Boot wrapped raw binary'
-  echo '  simpleImage.<dt> - ELF image with $(arch)/boot/dts/<dt>.dts linked in'
-  echo '                   - stripped elf with fdt blob'
-  echo '  simpleImage.<dt>.unstrip - full ELF image with fdt blob'
-  echo '  *_defconfig      - Select default config from arch/microblaze/configs'
-  echo ''
+  echo '  simpleImage.<dt> - Create the following images with <dt>.dtb linked in'
+  echo '                    simpleImage.<dt>        : raw image'
+  echo '                    simpleImage.<dt>.ub     : raw image with U-Boot header'
+  echo '                    simpleImage.<dt>.unstrip: ELF (identical to vmlinux)'
+  echo '                    simpleImage.<dt>.strip  : stripped ELF'
   echo '  Targets with <dt> embed a device tree blob inside the image'
   echo '  These targets support board with firmware that does not'
   echo '  support passing a device tree directly. Replace <dt> with the'
Signature · 1 line
2.20.1
author maintainer
#55 lore raw

PATCH 57/219 microblaze: move "... is ready" messages to arch/microblaze/Makefile

In reply to Sasha Levin

From: Masahiro Yamada <[email protected]>

[ Upstream commit 2e14f94cf4bc2f15ca5362e81ca3a987c79e3062 ]

To prepare for more fixes, move this to arch/microblaze/Makefile. Otherwise, the same "... is ready" would be printed multiple times.

---
 arch/microblaze/Makefile      | 2 ++
 arch/microblaze/boot/Makefile | 4 ----
 2 files changed, 2 insertions(+), 4 deletions(-)
arch/microblaze/Makefile +2 -0
index eecf37276521c..b9808ddaf985f 100644
@@ -83,9 +83,11 @@ archclean:
 
 linux.bin linux.bin.gz linux.bin.ub: vmlinux
 	$(Q)$(MAKE) $(build)=$(boot) $(boot)/$@
+	@echo 'Kernel: $(boot)/$@ is ready' ' (#'`cat .version`')'
 
 simpleImage.%: vmlinux
 	$(Q)$(MAKE) $(build)=$(boot) $(boot)/$@
+	@echo 'Kernel: $(boot)/$@ is ready' ' (#'`cat .version`')'
 
 define archhelp
   echo '* linux.bin    - Create raw binary'
arch/microblaze/boot/Makefile +0 -4
index 600e5a198bd2a..96eefdca0d9b3 100644
@@ -9,15 +9,12 @@ OBJCOPYFLAGS := -R .note -R .comment -R .note.gnu.build-id -O binary
 
 $(obj)/linux.bin: vmlinux FORCE
 	$(call if_changed,objcopy)
-	@echo 'Kernel: $@ is ready' ' (#'`cat .version`')'
 
 $(obj)/linux.bin.ub: $(obj)/linux.bin FORCE
 	$(call if_changed,uimage)
-	@echo 'Kernel: $@ is ready' ' (#'`cat .version`')'
 
 $(obj)/linux.bin.gz: $(obj)/linux.bin FORCE
 	$(call if_changed,gzip)
-	@echo 'Kernel: $@ is ready' ' (#'`cat .version`')'
 
 quiet_cmd_cp = CP      $< $@$2
 	cmd_cp = cat $< >$@$2 || (rm -f $@ && echo false)
@@ -35,6 +32,5 @@ $(obj)/simpleImage.%: vmlinux FORCE
 	$(call if_changed,objcopy)
 	$(call if_changed,uimage)
 	$(call if_changed,strip,.strip)
-	@echo 'Kernel: $(UIMAGE_OUT) is ready' ' (#'`cat .version`')'
 
 clean-files += simpleImage.*.unstrip linux.bin.ub
Signature · 1 line
2.20.1
author maintainer
#56 lore raw

PATCH 58/219 microblaze: fix multiple bugs in arch/microblaze/boot/Makefile

In reply to Sasha Levin

From: Masahiro Yamada <[email protected]>

[ Upstream commit 4722a3e6b716d9d4594c3cf3856b03bbd24a59a8 ]

This commit fixes some build issues.

The first issue is the breakage of linux.bin.ub target since commit ece97f3a5fb5 ("microblaze: Fix simpleImage format generation") because the addition of UIMAGE_{IN,OUT} affected it.

make ARCH=microblaze CROSS_COMPILE=microblaze-linux- linux.bin.ub
[ snip ]
OBJCOPY arch/microblaze/boot/linux.bin
UIMAGE arch/microblaze/boot/linux.bin.ub.ub
/usr/bin/mkimage: Can't open arch/microblaze/boot/linux.bin.ub: No such file or directory make[1]: *** [arch/microblaze/boot/Makefile;14: arch/microblaze/boot/linux.bin.ub] Error 1
make: *** [arch/microblaze/Makefile;83: linux.bin.ub] Error 2

The second issue is the use of the "if_changed" multiple times for the same target.

As commit 92a4728608a8 ("x86/boot: Fix if_changed build flip/flop bug") pointed out, this never works properly. Moreover, generating multiple images as a side-effect is confusing.

Let's split the build recipe for each image.

simpleImage.<dt>*.unstrip is just a copy of vmlinux.

simpleImage.<dt> and simpleImage.<dt>.ub are created in the same way as linux.bin and linux.bin.ub, respectively.

I kept simpleImage.* recipes independent of linux.bin.* ones to not change the behavior.

Lastly, this commit fixes "make ARCH=microblaze clean". Previously, it only cleaned up the unstrip image. Now, all the simpleImage files are cleaned.

---
 arch/microblaze/Makefile      |  2 +-
 arch/microblaze/boot/Makefile | 19 +++++++++----------
 2 files changed, 10 insertions(+), 11 deletions(-)
arch/microblaze/Makefile +1 -1
index b9808ddaf985f..548bac6c60f8c 100644
@@ -86,7 +86,7 @@ linux.bin linux.bin.gz linux.bin.ub: vmlinux
 	@echo 'Kernel: $(boot)/$@ is ready' ' (#'`cat .version`')'
 
 simpleImage.%: vmlinux
-	$(Q)$(MAKE) $(build)=$(boot) $(boot)/$@
+	$(Q)$(MAKE) $(build)=$(boot) $(addprefix $(boot)/$@., ub unstrip strip)
 	@echo 'Kernel: $(boot)/$@ is ready' ' (#'`cat .version`')'
 
 define archhelp
arch/microblaze/boot/Makefile +9 -10
index 96eefdca0d9b3..cff570a719461 100644
@@ -3,7 +3,7 @@
 # arch/microblaze/boot/Makefile
 #
 
-targets := linux.bin linux.bin.gz linux.bin.ub simpleImage.%
+targets := linux.bin linux.bin.gz linux.bin.ub simpleImage.*
 
 OBJCOPYFLAGS := -R .note -R .comment -R .note.gnu.build-id -O binary
 
@@ -16,21 +16,20 @@ $(obj)/linux.bin.ub: $(obj)/linux.bin FORCE
 $(obj)/linux.bin.gz: $(obj)/linux.bin FORCE
 	$(call if_changed,gzip)
 
-quiet_cmd_cp = CP      $< $@$2
-	cmd_cp = cat $< >$@$2 || (rm -f $@ && echo false)
-
 quiet_cmd_strip = STRIP   $< $@$2
 	cmd_strip = $(STRIP) -K microblaze_start -K _end -K __log_buf \
 				-K _fdt_start $< -o $@$2
 
 UIMAGE_LOADADDR = $(CONFIG_KERNEL_BASE_ADDR)
-UIMAGE_IN = $@
-UIMAGE_OUT = [email protected]
 
-$(obj)/simpleImage.%: vmlinux FORCE
-	$(call if_changed,cp,.unstrip)
+$(obj)/simpleImage.$(DTB): vmlinux FORCE
 	$(call if_changed,objcopy)
+
+$(obj)/simpleImage.$(DTB).ub: $(obj)/simpleImage.$(DTB) FORCE
 	$(call if_changed,uimage)
-	$(call if_changed,strip,.strip)
 
-clean-files += simpleImage.*.unstrip linux.bin.ub
+$(obj)/simpleImage.$(DTB).unstrip: vmlinux FORCE
+	$(call if_changed,shipped)
+
+$(obj)/simpleImage.$(DTB).strip: vmlinux FORCE
+	$(call if_changed,strip)
Signature · 1 line
2.20.1
author maintainer
#57 lore raw

PATCH 59/219 iwlwifi: move iwl_nvm_check_version() into dvm

In reply to Sasha Levin

From: Luca Coelho <[email protected]>

[ Upstream commit 64866e5da1eabd0c52ff45029b245f5465920031 ]

This function is only half-used by mvm (i.e. only the nvm_version part matters, since the calibration version is irrelevant), so it's pointless to export it from iwlwifi. If mvm uses this function, it has the additional complexity of setting the calib version to a bogus value on all cfg structs.

To avoid this, move the function to dvm and make a simple comparison of the nvm_version in mvm instead.

---
 drivers/net/wireless/intel/iwlwifi/dvm/main.c | 17 +++++++++++++++++
 .../wireless/intel/iwlwifi/iwl-eeprom-parse.c | 19 -------------------
 .../wireless/intel/iwlwifi/iwl-eeprom-parse.h |  5 ++---
 drivers/net/wireless/intel/iwlwifi/mvm/fw.c   |  4 +++-
 4 files changed, 22 insertions(+), 23 deletions(-)
drivers/net/wireless/intel/iwlwifi/dvm/main.c +17 -0
index 030482b357a3c..06dd4e81b7374 100644
@@ -1227,6 +1227,23 @@ static int iwl_eeprom_init_hw_params(struct iwl_priv *priv)
 	return 0;
 }
 
+static int iwl_nvm_check_version(struct iwl_nvm_data *data,
+				 struct iwl_trans *trans)
+{
+	if (data->nvm_version >= trans->cfg->nvm_ver ||
+	    data->calib_version >= trans->cfg->nvm_calib_ver) {
+		IWL_DEBUG_INFO(trans, "device EEPROM VER=0x%x, CALIB=0x%x\n",
+			       data->nvm_version, data->calib_version);
+		return 0;
+	}
+
+	IWL_ERR(trans,
+		"Unsupported (too old) EEPROM VER=0x%x < 0x%x CALIB=0x%x < 0x%x\n",
+		data->nvm_version, trans->cfg->nvm_ver,
+		data->calib_version,  trans->cfg->nvm_calib_ver);
+	return -EINVAL;
+}
+
 static struct iwl_op_mode *iwl_op_mode_dvm_start(struct iwl_trans *trans,
 						 const struct iwl_cfg *cfg,
 						 const struct iwl_fw *fw,
drivers/net/wireless/intel/iwlwifi/iwl-eeprom-parse.c +0 -19
index a4c96215933ba..a59bab8345f4e 100644
@@ -928,22 +928,3 @@ iwl_parse_eeprom_data(struct device *dev, const struct iwl_cfg *cfg,
 	return NULL;
 }
 IWL_EXPORT_SYMBOL(iwl_parse_eeprom_data);
-
-/* helper functions */
-int iwl_nvm_check_version(struct iwl_nvm_data *data,
-			     struct iwl_trans *trans)
-{
-	if (data->nvm_version >= trans->cfg->nvm_ver ||
-	    data->calib_version >= trans->cfg->nvm_calib_ver) {
-		IWL_DEBUG_INFO(trans, "device EEPROM VER=0x%x, CALIB=0x%x\n",
-			       data->nvm_version, data->calib_version);
-		return 0;
-	}
-
-	IWL_ERR(trans,
-		"Unsupported (too old) EEPROM VER=0x%x < 0x%x CALIB=0x%x < 0x%x\n",
-		data->nvm_version, trans->cfg->nvm_ver,
-		data->calib_version,  trans->cfg->nvm_calib_ver);
-	return -EINVAL;
-}
-IWL_EXPORT_SYMBOL(iwl_nvm_check_version);
drivers/net/wireless/intel/iwlwifi/iwl-eeprom-parse.h +2 -3
index 8be50ed12300f..c59dd47cf15d3 100644
@@ -7,6 +7,7 @@
  *
  * Copyright(c) 2008 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2015 Intel Mobile Communications GmbH
+ * Copyright (C) 2018 Intel Corporation
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -33,6 +34,7 @@
  *
  * Copyright(c) 2005 - 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2015 Intel Mobile Communications GmbH
+ * Copyright (C) 2018 Intel Corporation
  * All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
@@ -122,9 +124,6 @@ struct iwl_nvm_data *
 iwl_parse_eeprom_data(struct device *dev, const struct iwl_cfg *cfg,
 		      const u8 *eeprom, size_t eeprom_size);
 
-int iwl_nvm_check_version(struct iwl_nvm_data *data,
-			  struct iwl_trans *trans);
-
 int iwl_init_sband_channels(struct iwl_nvm_data *data,
 			    struct ieee80211_supported_band *sband,
 			    int n_channels, enum nl80211_band band);
drivers/net/wireless/intel/iwlwifi/mvm/fw.c +3 -1
index 2eba6d6f367f8..9808d954dca29 100644
@@ -547,7 +547,9 @@ int iwl_run_init_mvm_ucode(struct iwl_mvm *mvm, bool read_nvm)
 	if (mvm->nvm_file_name)
 		iwl_mvm_load_nvm_to_nic(mvm);
 
-	WARN_ON(iwl_nvm_check_version(mvm->nvm_data, mvm->trans));
+	WARN_ONCE(mvm->nvm_data->nvm_version < mvm->trans->cfg->nvm_ver,
+		  "Too old NVM version (0x%0x, required = 0x%0x)",
+		  mvm->nvm_data->nvm_version, mvm->trans->cfg->nvm_ver);
 
 	/*
 	 * abort after reading the nvm in case RF Kill is on, we will complete
Signature · 1 line
2.20.1
author maintainer
#58 lore raw

PATCH 60/219 iwlwifi: mvm: force TCM re-evaluation on TCM resume

In reply to Sasha Levin

From: Avraham Stern <[email protected]>

[ Upstream commit 7bc2468277033e05401d5f8fd48a772f407338c2 ]

When traffic load is not low or low latency is active, TCM schedules re-evaluation work so in case traffic stops TCM will detect that traffic load has become low or that low latency is no longer active. However, if TCM is paused when the re-evaluation work runs, it does not re-evaluate and the re-evaluation work is no longer scheduled. As a result, TCM will not indicate that low latency is no longer active or that traffic load is low when traffic stops.

Fix this by forcing TCM re-evaluation when TCM is resumed in case low latency is active or traffic load is not low.

---
 drivers/net/wireless/intel/iwlwifi/mvm/utils.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)
drivers/net/wireless/intel/iwlwifi/mvm/utils.c +14 -0
index 6a5349401aa99..00712205c05f2 100644
@@ -1804,6 +1804,7 @@ void iwl_mvm_pause_tcm(struct iwl_mvm *mvm, bool with_cancel)
 void iwl_mvm_resume_tcm(struct iwl_mvm *mvm)
 {
 	int mac;
+	bool low_latency = false;
 
 	spin_lock_bh(&mvm->tcm.lock);
 	mvm->tcm.ts = jiffies;
@@ -1815,10 +1816,23 @@ void iwl_mvm_resume_tcm(struct iwl_mvm *mvm)
 		memset(&mdata->tx.pkts, 0, sizeof(mdata->tx.pkts));
 		memset(&mdata->rx.airtime, 0, sizeof(mdata->rx.airtime));
 		memset(&mdata->tx.airtime, 0, sizeof(mdata->tx.airtime));
+
+		if (mvm->tcm.result.low_latency[mac])
+			low_latency = true;
 	}
 	/* The TCM data needs to be reset before "paused" flag changes */
 	smp_mb();
 	mvm->tcm.paused = false;
+
+	/*
+	 * if the current load is not low or low latency is active, force
+	 * re-evaluation to cover the case of no traffic.
+	 */
+	if (mvm->tcm.result.global_load > IWL_MVM_TRAFFIC_LOW)
+		schedule_delayed_work(&mvm->tcm.work, MVM_TCM_PERIOD);
+	else if (low_latency)
+		schedule_delayed_work(&mvm->tcm.work, MVM_LL_PERIOD);
+
 	spin_unlock_bh(&mvm->tcm.lock);
 }
 
Signature · 1 line
2.20.1
author maintainer
#59 lore raw

PATCH 61/219 iwlwifi: pcie: fix erroneous print

In reply to Sasha Levin

From: Sara Sharon <[email protected]>

[ Upstream commit 0916224eaa77bff0fbbc747961d550ff8db45457 ]

When removing the driver, the following flow can happen:
1. host command is in progress, for example at index 68.
2. RX interrupt is received with the response.
3. Before it is processed, the remove flow kicks in, and calls iwl_pcie_txq_unmap. The function cleans all DMA, and promotes the read pointer to 69.
4. RX thread proceeds with the processing, and is calling iwl_pcie_cmdq_reclaim, which will print this error:
iwl_pcie_cmdq_reclaim: Read index for DMA queue txq id (0), index 4 is out of range [0-256] 69 69.

Detect this situation, and avoid the print. Change it to warning while at it, to make such issues more noticeable in the future.

---
 drivers/net/wireless/intel/iwlwifi/pcie/tx.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)
drivers/net/wireless/intel/iwlwifi/pcie/tx.c +5 -5
index 42fdb7970cfdc..7cba4c37c66fd 100644
@@ -1247,11 +1247,11 @@ static void iwl_pcie_cmdq_reclaim(struct iwl_trans *trans, int txq_id, int idx)
 
 	if (idx >= trans->cfg->base_params->max_tfd_queue_size ||
 	    (!iwl_queue_used(txq, idx))) {
-		IWL_ERR(trans,
-			"%s: Read index for DMA queue txq id (%d), index %d is out of range [0-%d] %d %d.\n",
-			__func__, txq_id, idx,
-			trans->cfg->base_params->max_tfd_queue_size,
-			txq->write_ptr, txq->read_ptr);
+		WARN_ONCE(test_bit(txq_id, trans_pcie->queue_used),
+			  "%s: Read index for DMA queue txq id (%d), index %d is out of range [0-%d] %d %d.\n",
+			  __func__, txq_id, idx,
+			  trans->cfg->base_params->max_tfd_queue_size,
+			  txq->write_ptr, txq->read_ptr);
 		return;
 	}
 
Signature · 1 line
2.20.1
author maintainer
#60 lore raw

PATCH 62/219 iwlwifi: pcie: set cmd_len in the correct place

In reply to Sasha Levin

From: Sara Sharon <[email protected]>

[ Upstream commit 956343a61226e1af3d146386f70a059feb567d0c ]

command len is set too early in the code, since when building AMSDU, the size changes. This causes the byte count table to have the wrong size.

Fixes a0ec0169b7a9 ("iwlwifi: support new tx api")
---
 .../net/wireless/intel/iwlwifi/pcie/tx-gen2.c | 24 +++++++++----------
 1 file changed, 12 insertions(+), 12 deletions(-)
drivers/net/wireless/intel/iwlwifi/pcie/tx-gen2.c +12 -12
index b99f33ff91230..a9e42f0437ccb 100644
@@ -555,18 +555,6 @@ int iwl_trans_pcie_gen2_tx(struct iwl_trans *trans, struct sk_buff *skb,
 
 	spin_lock(&txq->lock);
 
-	if (trans->cfg->device_family >= IWL_DEVICE_FAMILY_22560) {
-		struct iwl_tx_cmd_gen3 *tx_cmd_gen3 =
-			(void *)dev_cmd->payload;
-
-		cmd_len = le16_to_cpu(tx_cmd_gen3->len);
-	} else {
-		struct iwl_tx_cmd_gen2 *tx_cmd_gen2 =
-			(void *)dev_cmd->payload;
-
-		cmd_len = le16_to_cpu(tx_cmd_gen2->len);
-	}
-
 	if (iwl_queue_space(trans, txq) < txq->high_mark) {
 		iwl_stop_queue(trans, txq);
 
@@ -604,6 +592,18 @@ int iwl_trans_pcie_gen2_tx(struct iwl_trans *trans, struct sk_buff *skb,
 		return -1;
 	}
 
+	if (trans->cfg->device_family >= IWL_DEVICE_FAMILY_22560) {
+		struct iwl_tx_cmd_gen3 *tx_cmd_gen3 =
+			(void *)dev_cmd->payload;
+
+		cmd_len = le16_to_cpu(tx_cmd_gen3->len);
+	} else {
+		struct iwl_tx_cmd_gen2 *tx_cmd_gen2 =
+			(void *)dev_cmd->payload;
+
+		cmd_len = le16_to_cpu(tx_cmd_gen2->len);
+	}
+
 	/* Set up entry for this TFD in Tx byte-count array */
 	iwl_pcie_gen2_update_byte_tbl(trans_pcie, txq, cmd_len,
 				      iwl_pcie_gen2_get_num_tbs(trans, tfd));
Signature · 1 line
2.20.1
author maintainer
#61 lore raw

PATCH 63/219 gpio: pca953x: Fix AI overflow on PCAL6524

In reply to Sasha Levin

From: Marek Vasut <[email protected]>

[ Upstream commit 92f45ebe68181c2d7f76633ffae55bc9447d62cd ]

The PCAL_PINCTRL_MASK is too large. The extended register block on PCAL6524, which is the largest chip with this block, has the block limited to address range 0x40..0x7f. This is because the bit 7 in the command register is used for the Address Increment functionality.

Trim the mask to 0x60 to match the datasheet and to prevent accidental overwrite of the AI bit.

---
 drivers/gpio/gpio-pca953x.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpio/gpio-pca953x.c +1 -1
index e0657fc72d31f..0232c25a15864 100644
@@ -58,7 +58,7 @@
 #define PCA_GPIO_MASK		0x00FF
 
 #define PCAL_GPIO_MASK		0x1f
-#define PCAL_PINCTRL_MASK	0xe0
+#define PCAL_PINCTRL_MASK	0x60
 
 #define PCA_INT			0x0100
 #define PCA_PCAL		0x0200
Signature · 1 line
2.20.1
author maintainer
#62 lore raw

PATCH 64/219 gpiolib: Fix return value of gpio_to_desc() stub if !GPIOLIB

In reply to Sasha Levin

From: Krzysztof Kozlowski <[email protected]>

[ Upstream commit c5510b8dafce5f3f5a039c9b262ebcae0092c462 ]

If CONFIG_GPOILIB is not set, the stub of gpio_to_desc() should return the same type of error as regular version: NULL. All the callers compare the return value of gpio_to_desc() against NULL, so returned ERR_PTR would be treated as non-error case leading to dereferencing of error value.

Fixes 79a9becda894 ("gpiolib: export descriptor-based GPIO interface")
---
 include/linux/gpio/consumer.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
include/linux/gpio/consumer.h +1 -1
index 412098b24f58b..8dfd8300d9c31 100644
@@ -475,7 +475,7 @@ static inline int gpiod_set_consumer_name(struct gpio_desc *desc,
 
 static inline struct gpio_desc *gpio_to_desc(unsigned gpio)
 {
-	return ERR_PTR(-EINVAL);
+	return NULL;
 }
 
 static inline int desc_to_gpio(const struct gpio_desc *desc)
Signature · 1 line
2.20.1
author maintainer
#63 lore raw

PATCH 65/219 kvm: vmx: Set IA32_TSC_AUX for legacy mode guests

In reply to Sasha Levin

From: Jim Mattson <[email protected]>

[ Upstream commit 0023ef39dc35c773c436eaa46ca539a26b308b55 ]

RDTSCP is supported in legacy mode as well as long mode. The IA32_TSC_AUX MSR should be set to the correct guest value before entering any guest that supports RDTSCP.

Fixes 4e47c7a6d714 ("KVM: VMX: Add instruction rdtscp support for guest")
---
 arch/x86/kvm/vmx.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
arch/x86/kvm/vmx.c +3 -3
index 4eda2a9c234a6..b892f345f6317 100644
@@ -3404,9 +3404,6 @@ static void setup_msrs(struct vcpu_vmx *vmx)
 		index = __find_msr_index(vmx, MSR_CSTAR);
 		if (index >= 0)
 			move_msr_up(vmx, index, save_nmsrs++);
-		index = __find_msr_index(vmx, MSR_TSC_AUX);
-		if (index >= 0 && guest_cpuid_has(&vmx->vcpu, X86_FEATURE_RDTSCP))
-			move_msr_up(vmx, index, save_nmsrs++);
 		/*
 		 * MSR_STAR is only needed on long mode guests, and only
 		 * if efer.sce is enabled.
@@ -3419,6 +3416,9 @@ static void setup_msrs(struct vcpu_vmx *vmx)
 	index = __find_msr_index(vmx, MSR_EFER);
 	if (index >= 0 && update_transition_efer(vmx, index))
 		move_msr_up(vmx, index, save_nmsrs++);
+	index = __find_msr_index(vmx, MSR_TSC_AUX);
+	if (index >= 0 && guest_cpuid_has(&vmx->vcpu, X86_FEATURE_RDTSCP))
+		move_msr_up(vmx, index, save_nmsrs++);
 
 	vmx->save_nmsrs = save_nmsrs;
 	vmx->guest_msrs_dirty = true;
Signature · 1 line
2.20.1
author maintainer
#64 lore raw

PATCH 66/219 crypto/chelsio/chtls: listen fails with multiadapt

In reply to Sasha Levin

From: Atul Gupta <[email protected]>

[ Upstream commit 6422ccc5fbefbd219f3fab133f698e58f5aa44eb ]

listen fails when more than one tls capable device is registered. tls_hw_hash is called for each dev which loops again for each cdev_list causing listen failure. Hence call chtls_listen_start/stop for specific device than loop over all devices.

---
 drivers/crypto/chelsio/chtls/chtls.h      |  5 +++
 drivers/crypto/chelsio/chtls/chtls_main.c | 50 ++++++++++++++---------
 2 files changed, 36 insertions(+), 19 deletions(-)
drivers/crypto/chelsio/chtls/chtls.h +5 -0
index 7725b6ee14efb..fcb6747ed29ea 100644
@@ -153,6 +153,11 @@ struct chtls_dev {
 	unsigned int cdev_state;
 };
 
+struct chtls_listen {
+	struct chtls_dev *cdev;
+	struct sock *sk;
+};
+
 struct chtls_hws {
 	struct sk_buff_head sk_recv_queue;
 	u8 txqid;
drivers/crypto/chelsio/chtls/chtls_main.c +31 -19

99 lines not on this page · raw message

Signature · 1 line
2.20.1
author maintainer
#65 lore raw

PATCH 67/219 VSOCK: bind to random port for VMADDR_PORT_ANY

In reply to Sasha Levin

From: Lepton Wu <[email protected]>

[ Upstream commit 8236b08cf50f85bbfaf48910a0b3ee68318b7c4b ]

The old code always starts from fixed port for VMADDR_PORT_ANY. Sometimes when VMM crashed, there is still orphaned vsock which is waiting for close timer, then it could cause connection time out for new started VM if they are trying to connect to same port with same guest cid since the new packets could hit that orphaned vsock. We could also fix this by doing more in vhost_vsock_reset_orphans, but any way, it should be better to start from a random local port instead of a fixed one.

---
 net/vmw_vsock/af_vsock.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
net/vmw_vsock/af_vsock.c +6 -1
index b2724fb06bbbb..62ec9182f234d 100644
@@ -107,6 +107,7 @@
 #include <linux/mutex.h>
 #include <linux/net.h>
 #include <linux/poll.h>
+#include <linux/random.h>
 #include <linux/skbuff.h>
 #include <linux/smp.h>
 #include <linux/socket.h>
@@ -480,9 +481,13 @@ static void vsock_pending_work(struct work_struct *work)
 static int __vsock_bind_stream(struct vsock_sock *vsk,
 			       struct sockaddr_vm *addr)
 {
-	static u32 port = LAST_RESERVED_PORT + 1;
+	static u32 port = 0;
 	struct sockaddr_vm new_addr;
 
+	if (!port)
+		port = LAST_RESERVED_PORT + 1 +
+			prandom_u32_max(U32_MAX - LAST_RESERVED_PORT);
+
 	vsock_addr_init(&new_addr, addr->svm_cid, addr->svm_port);
 
 	if (addr->svm_port == VMADDR_PORT_ANY) {
Signature · 1 line
2.20.1
author maintainer
#66 lore raw

PATCH 68/219 mmc: meson-gx: make sure the descriptor is stopped on errors

In reply to Sasha Levin

From: Jerome Brunet <[email protected]>

[ Upstream commit 18f92bc02f1739b5c4d5b70009fbb7eada45bca3 ]

On errors, if we don't stop the descriptor chain, it may continue to run and raise IRQ after we have called mmc_request_done(). This is bad because we won't be able to get cmd anymore and properly deal with the IRQ.

This patch makes sure the descriptor chain is stopped before calling mmc_request_done()

Fixes 79ed05e329c3 ("mmc: meson-gx: add support for descriptor chain mode")
---
 drivers/mmc/host/meson-gx-mmc.c | 73 ++++++++++++++++++++++++++++-----
 1 file changed, 63 insertions(+), 10 deletions(-)
drivers/mmc/host/meson-gx-mmc.c +63 -10
index ddd98cdd33bcd..72f34a58928ca 100644
@@ -21,6 +21,7 @@
 #include <linux/kernel.h>
 #include <linux/module.h>
 #include <linux/init.h>
+#include <linux/delay.h>
 #include <linux/device.h>
 #include <linux/of_device.h>
 #include <linux/platform_device.h>
@@ -90,9 +91,11 @@
 #define   CFG_CLK_ALWAYS_ON BIT(18)
 #define   CFG_CHK_DS BIT(20)
 #define   CFG_AUTO_CLK BIT(23)
+#define   CFG_ERR_ABORT BIT(27)
 
 #define SD_EMMC_STATUS 0x48
 #define   STATUS_BUSY BIT(31)
+#define   STATUS_DESC_BUSY BIT(30)
 #define   STATUS_DATI GENMASK(23, 16)
 
 #define SD_EMMC_IRQ_EN 0x4c
@@ -930,6 +933,7 @@ static void meson_mmc_start_cmd(struct mmc_host *mmc, struct mmc_command *cmd)
 
 	cmd_cfg |= FIELD_PREP(CMD_CFG_CMD_INDEX_MASK, cmd->opcode);
 	cmd_cfg |= CMD_CFG_OWNER;  /* owned by CPU */
+	cmd_cfg |= CMD_CFG_ERROR; /* stop in case of error */
 
 	meson_mmc_set_response_bits(cmd, &cmd_cfg);
 
@@ -1024,6 +1028,17 @@ static irqreturn_t meson_mmc_irq(int irq, void *dev_id)
 	u32 irq_en, status, raw_status;
 	irqreturn_t ret = IRQ_NONE;
 
+	irq_en = readl(host->regs + SD_EMMC_IRQ_EN);
+	raw_status = readl(host->regs + SD_EMMC_STATUS);
+	status = raw_status & irq_en;
+
+	if (!status) {
+		dev_dbg(host->dev,
+			"Unexpected IRQ! irq_en 0x%08x - status 0x%08x\n",
+			 irq_en, raw_status);
+		return IRQ_NONE;
+	}
+
 	if (WARN_ON(!host) || WARN_ON(!host->cmd))
 		return IRQ_NONE;
 
@@ -1031,22 +1046,18 @@ static irqreturn_t meson_mmc_irq(int irq, void *dev_id)
 
 	cmd = host->cmd;
 	data = cmd->data;
-	irq_en = readl(host->regs + SD_EMMC_IRQ_EN);
-	raw_status = readl(host->regs + SD_EMMC_STATUS);
-	status = raw_status & irq_en;
-
 	cmd->error = 0;
 	if (status & IRQ_CRC_ERR) {
 		dev_dbg(host->dev, "CRC Error - status 0x%08x\n", status);
 		cmd->error = -EILSEQ;
-		ret = IRQ_HANDLED;
+		ret = IRQ_WAKE_THREAD;
 		goto out;
 	}
 
 	if (status & IRQ_TIMEOUTS) {
 		dev_dbg(host->dev, "Timeout - status 0x%08x\n", status);
 		cmd->error = -ETIMEDOUT;
-		ret = IRQ_HANDLED;
+		ret = IRQ_WAKE_THREAD;
 		goto out;
 	}
 
@@ -1071,17 +1082,49 @@ static irqreturn_t meson_mmc_irq(int irq, void *dev_id)
 	/* ack all enabled interrupts */
 	writel(irq_en, host->regs + SD_EMMC_STATUS);
 
+	if (cmd->error) {
+		/* Stop desc in case of errors */
+		u32 start = readl(host->regs + SD_EMMC_START);
+
+		start &= ~START_DESC_BUSY;
+		writel(start, host->regs + SD_EMMC_START);
+	}
+
 	if (ret == IRQ_HANDLED)
 		meson_mmc_request_done(host->mmc, cmd->mrq);
-	else if (ret == IRQ_NONE)
-		dev_warn(host->dev,
-			 "Unexpected IRQ! status=0x%08x, irq_en=0x%08x\n",
-			 raw_status, irq_en);
 
 	spin_unlock(&host->lock);
 	return ret;
 }
 
+static int meson_mmc_wait_desc_stop(struct meson_host *host)
+{
+	int loop;
+	u32 status;
+
+	/*
+	 * It may sometimes take a while for it to actually halt. Here, we
+	 * are giving it 5ms to comply
+	 *
+	 * If we don't confirm the descriptor is stopped, it might raise new
+	 * IRQs after we have called mmc_request_done() which is bad.
+	 */
+	for (loop = 50; loop; loop--) {
+		status = readl(host->regs + SD_EMMC_STATUS);
+		if (status & (STATUS_BUSY | STATUS_DESC_BUSY))
+			udelay(100);
+		else
+			break;
+	}
+
+	if (status & (STATUS_BUSY | STATUS_DESC_BUSY)) {
+		dev_err(host->dev, "Timed out waiting for host to stop\n");
+		return -ETIMEDOUT;
+	}
+
+	return 0;
+}
+
 static irqreturn_t meson_mmc_irq_thread(int irq, void *dev_id)
 {
 	struct meson_host *host = dev_id;
@@ -1092,6 +1135,13 @@ static irqreturn_t meson_mmc_irq_thread(int irq, void *dev_id)
 	if (WARN_ON(!cmd))
 		return IRQ_NONE;
 
+	if (cmd->error) {
+		meson_mmc_wait_desc_stop(host);
+		meson_mmc_request_done(host->mmc, cmd->mrq);
+
+		return IRQ_HANDLED;
+	}
+
 	data = cmd->data;
 	if (meson_mmc_bounce_buf_read(data)) {
 		xfer_bytes = data->blksz * data->blocks;
@@ -1132,6 +1182,9 @@ static void meson_mmc_cfg_init(struct meson_host *host)
 	cfg |= FIELD_PREP(CFG_RC_CC_MASK, ilog2(SD_EMMC_CFG_CMD_GAP));
 	cfg |= FIELD_PREP(CFG_BLK_LEN_MASK, ilog2(SD_EMMC_CFG_BLK_SIZE));
 
+	/* abort chain on R/W errors */
+	cfg |= CFG_ERR_ABORT;
+
 	writel(cfg, host->regs + SD_EMMC_CFG);
 }
 
Signature · 1 line
2.20.1
author maintainer
#67 lore raw

PATCH 69/219 mtd: rawnand: sunxi: Write pageprog related opcodes to WCMD_SET

In reply to Sasha Levin

From: Boris Brezillon <[email protected]>

[ Upstream commit 732774437ae01d9882e60314e303898e63c7f038 ]

The opcodes used by the controller when doing batched page prog should be written in NFC_REG_WCMD_SET not FC_REG_RCMD_SET. Luckily, the default NFC_REG_WCMD_SET value matches the one we set in the driver which explains why we didn't notice the problem.

Fixes 614049a8d904 ("mtd: nand: sunxi: add support for DMA assisted operations")
---
 drivers/mtd/nand/raw/sunxi_nand.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/mtd/nand/raw/sunxi_nand.c +1 -1
index 1f0b7ee38df56..5b5f4d25a3e12 100644
@@ -1397,7 +1397,7 @@ static int sunxi_nfc_hw_ecc_write_page_dma(struct mtd_info *mtd,
 	sunxi_nfc_randomizer_enable(mtd);
 
 	writel((NAND_CMD_RNDIN << 8) | NAND_CMD_PAGEPROG,
-	       nfc->regs + NFC_REG_RCMD_SET);
+	       nfc->regs + NFC_REG_WCMD_SET);
 
 	dma_async_issue_pending(nfc->dmac);
 
Signature · 1 line
2.20.1
author maintainer
#68 lore raw

PATCH 70/219 usb: ehci-omap: Fix deferred probe for phy handling

In reply to Sasha Levin

From: Roger Quadros <[email protected]>

[ Upstream commit 8dc7623bf608495b6e6743e805807c7840673573 ]

PHY model is being used on omap5 platforms even if port mode is not OMAP_EHCI_PORT_MODE_PHY. So don't guess if PHY is required or not based on PHY mode.

If PHY is provided in device tree, it must be required. So, if devm_usb_get_phy_by_phandle() gives us an error code other than -ENODEV (no PHY) then error out.

This fixes USB Ethernet on omap5-uevm if PHY happens to probe after EHCI thus causing a -EPROBE_DEFER.

---
 drivers/usb/host/ehci-omap.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
drivers/usb/host/ehci-omap.c +4 -3
index 7e4c13346a1ee..7d20296cbe9f9 100644
@@ -159,11 +159,12 @@ static int ehci_hcd_omap_probe(struct platform_device *pdev)
 		/* get the PHY device */
 		phy = devm_usb_get_phy_by_phandle(dev, "phys", i);
 		if (IS_ERR(phy)) {
-			/* Don't bail out if PHY is not absolutely necessary */
-			if (pdata->port_mode[i] != OMAP_EHCI_PORT_MODE_PHY)
+			ret = PTR_ERR(phy);
+			if (ret == -ENODEV) { /* no PHY */
+				phy = NULL;
 				continue;
+			}
 
-			ret = PTR_ERR(phy);
 			if (ret != -EPROBE_DEFER)
 				dev_err(dev, "Can't get PHY for port %d: %d\n",
 					i, ret);
Signature · 1 line
2.20.1
author maintainer
#69 lore raw

PATCH 71/219 btrfs: Check for missing device before bio submission in btrfs_map_bio

In reply to Sasha Levin

From: Nikolay Borisov <[email protected]>

[ Upstream commit fc8a168aa9ab1680c2bd52bf9db7c994e0f2524f ]

Before btrfs_map_bio submits all stripe bios it does a number of checks to ensure the device for every stripe is present. However, it doesn't do a DEV_STATE_MISSING check, instead this is relegated to the lower level btrfs_schedule_bio (in the async submission case, sync submission doesn't check DEV_STATE_MISSING at all). Additionally btrfs_schedule_bios does the duplicate device->bdev check which has already been performed in btrfs_map_bio.

This patch moves the DEV_STATE_MISSING check in btrfs_map_bio and removes the duplicate device->bdev check. Doing so ensures that no bio cloning/submission happens for both async/sync requests in the face of missing device. This makes the async io submission path slightly shorter in terms of instruction count. No functional changes.

---
 fs/btrfs/volumes.c | 9 ++-------
 1 file changed, 2 insertions(+), 7 deletions(-)
fs/btrfs/volumes.c +2 -7
index a8297e7489d98..f84c18e86c816 100644
@@ -6106,12 +6106,6 @@ static noinline void btrfs_schedule_bio(struct btrfs_device *device,
 	int should_queue = 1;
 	struct btrfs_pending_bios *pending_bios;
 
-	if (test_bit(BTRFS_DEV_STATE_MISSING, &device->dev_state) ||
-	    !device->bdev) {
-		bio_io_error(bio);
-		return;
-	}
-
 	/* don't bother with additional async steps for reads, right now */
 	if (bio_op(bio) == REQ_OP_READ) {
 		btrfsic_submit_bio(bio);
@@ -6240,7 +6234,8 @@ blk_status_t btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio,
 
 	for (dev_nr = 0; dev_nr < total_devs; dev_nr++) {
 		dev = bbio->stripes[dev_nr].dev;
-		if (!dev || !dev->bdev ||
+		if (!dev || !dev->bdev || test_bit(BTRFS_DEV_STATE_MISSING,
+						   &dev->dev_state) ||
 		    (bio_op(first_bio) == REQ_OP_WRITE &&
 		    !test_bit(BTRFS_DEV_STATE_WRITEABLE, &dev->dev_state))) {
 			bbio_error(bbio, first_bio, logical);
Signature · 1 line
2.20.1
#70 lore raw

Re: [PATCH AUTOSEL 4.19 071/219] btrfs: Check for missing device before bio submission in btrfs_map_bio

Quoted 12 lines · Sasha Levin

From: Nikolay Borisov <[email protected]>

[ Upstream commit fc8a168aa9ab1680c2bd52bf9db7c994e0f2524f ]

Before btrfs_map_bio submits all stripe bios it does a number of checks to ensure the device for every stripe is present. However, it doesn't do a DEV_STATE_MISSING check, instead this is relegated to the lower level btrfs_schedule_bio (in the async submission case, sync submission doesn't check DEV_STATE_MISSING at all). Additionally btrfs_schedule_bios does the duplicate device->bdev check which has already been performed in btrfs_map_bio.

Quoted 5 lines

This patch moves the DEV_STATE_MISSING check in btrfs_map_bio and removes the duplicate device->bdev check. Doing so ensures that no bio cloning/submission happens for both async/sync requests in the face of missing device. This makes the async io submission path slightly shorter in terms of instruction count. No functional changes.

                                 ^^^^^^^^^^^^^^^^^^^^^

This should be a strong indication that the patch is not suitable for autosel/stable, please remove it. Thanks.

author maintainer
#71 lore raw

PATCH 72/219 btrfs: fix ncopies raid_attr for RAID56

In reply to Sasha Levin

From: Hans van Kranenburg <[email protected]>

[ Upstream commit da612e31aee51bd13231c78a47c714b543bd3ad8 ]

RAID5 and RAID6 profile store one copy of the data, not 2 or 3. These values are not yet used anywhere so there's no change.

---
 fs/btrfs/volumes.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
fs/btrfs/volumes.c +2 -2
index f84c18e86c816..5bbcdcff68a9e 100644
@@ -96,7 +96,7 @@ const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES] = {
 		.devs_min	= 2,
 		.tolerated_failures = 1,
 		.devs_increment	= 1,
-		.ncopies	= 2,
+		.ncopies	= 1,
 		.raid_name	= "raid5",
 		.bg_flag	= BTRFS_BLOCK_GROUP_RAID5,
 		.mindev_error	= BTRFS_ERROR_DEV_RAID5_MIN_NOT_MET,
@@ -108,7 +108,7 @@ const struct btrfs_raid_attr btrfs_raid_array[BTRFS_NR_RAID_TYPES] = {
 		.devs_min	= 3,
 		.tolerated_failures = 2,
 		.devs_increment	= 1,
-		.ncopies	= 3,
+		.ncopies	= 1,
 		.raid_name	= "raid6",
 		.bg_flag	= BTRFS_BLOCK_GROUP_RAID6,
 		.mindev_error	= BTRFS_ERROR_DEV_RAID6_MIN_NOT_MET,
Signature · 1 line
2.20.1
author maintainer
#72 lore raw

PATCH 73/219 btrfs: dev-replace: set result code of cancel by status of scrub

In reply to Sasha Levin

From: Anand Jain <[email protected]>

[ Upstream commit b47dda2ef6d793b67fd5979032dcd106e3f0a5c9 ]

The device-replace needs to check the result code of the scrub workers in btrfs_dev_replace_cancel and distinguish if successful cancel operation and when the there was no operation running.

If btrfs_scrub_cancel() fails, return BTRFS_IOCTL_DEV_REPLACE_RESULT_NOT_STARTED so that user can try to cancel the replace again.

---
 fs/btrfs/dev-replace.c | 21 ++++++++++++++-------
 1 file changed, 14 insertions(+), 7 deletions(-)
fs/btrfs/dev-replace.c +14 -7
index 23b13fbecdc22..96763805787ed 100644
@@ -810,16 +810,23 @@ int btrfs_dev_replace_cancel(struct btrfs_fs_info *fs_info)
 		btrfs_dev_replace_write_unlock(dev_replace);
 		break;
 	case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED:
-		result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NO_ERROR;
 		tgt_device = dev_replace->tgtdev;
 		src_device = dev_replace->srcdev;
 		btrfs_dev_replace_write_unlock(dev_replace);
-		btrfs_scrub_cancel(fs_info);
-		/* btrfs_dev_replace_finishing() will handle the cleanup part */
-		btrfs_info_in_rcu(fs_info,
-			"dev_replace from %s (devid %llu) to %s canceled",
-			btrfs_dev_name(src_device), src_device->devid,
-			btrfs_dev_name(tgt_device));
+		ret = btrfs_scrub_cancel(fs_info);
+		if (ret < 0) {
+			result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NOT_STARTED;
+		} else {
+			result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NO_ERROR;
+			/*
+			 * btrfs_dev_replace_finishing() will handle the
+			 * cleanup part
+			 */
+			btrfs_info_in_rcu(fs_info,
+				"dev_replace from %s (devid %llu) to %s canceled",
+				btrfs_dev_name(src_device), src_device->devid,
+				btrfs_dev_name(tgt_device));
+		}
 		break;
 	case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED:
 		/*
Signature · 1 line
2.20.1
author maintainer
#73 lore raw

PATCH 74/219 Btrfs: allow clear_extent_dirty() to receive a cached extent state record

In reply to Sasha Levin

From: Filipe Manana <[email protected]>

[ Upstream commit 0e6ec385b55f6001da8c6b1532494241e52c550d ]

We can have a lot freed extents during the life span of transaction, so the red black tree that keeps track of the ranges of each freed extent (fs_info->freed_extents[]) can get quite big. When finishing a transaction commit we find each range, process it (discard the extents, unpin them) and then remove it from the red black tree.

We can use an extent state record as a cache when searching for a range, so that when we clean the range we can use the cached extent state we passed to the search function instead of iterating the red black tree again. Doing things as fast as possible when finishing a transaction (in state TRANS_STATE_UNBLOCKED) is convenient as it reduces the time we block another task that wants to commit the next transaction.

So change clear_extent_dirty() to allow an optional extent state record to be passed as an argument, which will be passed down to __clear_extent_bit.

---
 fs/btrfs/disk-io.c     | 7 +++++--
 fs/btrfs/extent-tree.c | 7 +++++--
 fs/btrfs/extent_io.h   | 4 ++--
 3 files changed, 12 insertions(+), 6 deletions(-)
fs/btrfs/disk-io.c +5 -2
index b2dc613ebed2c..96296dc7d2ea4 100644
@@ -4350,6 +4350,8 @@ static int btrfs_destroy_pinned_extent(struct btrfs_fs_info *fs_info,
 	unpin = pinned_extents;
 again:
 	while (1) {
+		struct extent_state *cached_state = NULL;
+
 		/*
 		 * The btrfs_finish_extent_commit() may get the same range as
 		 * ours between find_first_extent_bit and clear_extent_dirty.
@@ -4358,13 +4360,14 @@ static int btrfs_destroy_pinned_extent(struct btrfs_fs_info *fs_info,
 		 */
 		mutex_lock(&fs_info->unused_bg_unpin_mutex);
 		ret = find_first_extent_bit(unpin, 0, &start, &end,
-					    EXTENT_DIRTY, NULL);
+					    EXTENT_DIRTY, &cached_state);
 		if (ret) {
 			mutex_unlock(&fs_info->unused_bg_unpin_mutex);
 			break;
 		}
 
-		clear_extent_dirty(unpin, start, end);
+		clear_extent_dirty(unpin, start, end, &cached_state);
+		free_extent_state(cached_state);
 		btrfs_error_unpin_extent_range(fs_info, start, end);
 		mutex_unlock(&fs_info->unused_bg_unpin_mutex);
 		cond_resched();
fs/btrfs/extent-tree.c +5 -2
index 024dd336b20ae..4bda5c09cdfe7 100644
@@ -6618,9 +6618,11 @@ int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans)
 		unpin = &fs_info->freed_extents[0];
 
 	while (!trans->aborted) {
+		struct extent_state *cached_state = NULL;
+
 		mutex_lock(&fs_info->unused_bg_unpin_mutex);
 		ret = find_first_extent_bit(unpin, 0, &start, &end,
-					    EXTENT_DIRTY, NULL);
+					    EXTENT_DIRTY, &cached_state);
 		if (ret) {
 			mutex_unlock(&fs_info->unused_bg_unpin_mutex);
 			break;
@@ -6630,9 +6632,10 @@ int btrfs_finish_extent_commit(struct btrfs_trans_handle *trans)
 			ret = btrfs_discard_extent(fs_info, start,
 						   end + 1 - start, NULL);
 
-		clear_extent_dirty(unpin, start, end);
+		clear_extent_dirty(unpin, start, end, &cached_state);
 		unpin_extent_range(fs_info, start, end, true);
 		mutex_unlock(&fs_info->unused_bg_unpin_mutex);
+		free_extent_state(cached_state);
 		cond_resched();
 	}
 
fs/btrfs/extent_io.h +2 -2
index ed27becd963c5..a3598b24441e1 100644
@@ -348,11 +348,11 @@ static inline int set_extent_dirty(struct extent_io_tree *tree, u64 start,
 }
 
 static inline int clear_extent_dirty(struct extent_io_tree *tree, u64 start,
-		u64 end)
+				     u64 end, struct extent_state **cached)
 {
 	return clear_extent_bit(tree, start, end,
 				EXTENT_DIRTY | EXTENT_DELALLOC |
-				EXTENT_DO_ACCOUNTING, 0, 0, NULL);
+				EXTENT_DO_ACCOUNTING, 0, 0, cached);
 }
 
 int convert_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
Signature · 1 line
2.20.1
author maintainer
#74 lore raw

PATCH 75/219 btrfs: only track ref_heads in delayed_ref_updates

In reply to Sasha Levin

From: Josef Bacik <[email protected]>

[ Upstream commit 158ffa364bf723fa1ef128060646d23dc3942994 ]

We use this number to figure out how many delayed refs to run, but __btrfs_run_delayed_refs really only checks every time we need a new delayed ref head, so we always run at least one ref head completely no matter what the number of items on it. Fix the accounting to only be adjusted when we add/remove a ref head.

In addition to using this number to limit the number of delayed refs run, a future patch is also going to use it to calculate the amount of space required for delayed refs space reservation.

---
 fs/btrfs/delayed-ref.c | 3 ---
 1 file changed, 3 deletions(-)
fs/btrfs/delayed-ref.c +0 -3
index 62ff545ba1f71..7e5c81e80e15d 100644
@@ -234,8 +234,6 @@ static inline void drop_delayed_ref(struct btrfs_trans_handle *trans,
 	ref->in_tree = 0;
 	btrfs_put_delayed_ref(ref);
 	atomic_dec(&delayed_refs->num_entries);
-	if (trans->delayed_ref_updates)
-		trans->delayed_ref_updates--;
 }
 
 static bool merge_ref(struct btrfs_trans_handle *trans,
@@ -446,7 +444,6 @@ static int insert_delayed_ref(struct btrfs_trans_handle *trans,
 	if (ref->action == BTRFS_ADD_DELAYED_REF)
 		list_add_tail(&ref->add_list, &href->ref_add_list);
 	atomic_inc(&root->num_entries);
-	trans->delayed_ref_updates++;
 	spin_unlock(&href->lock);
 	return ret;
 }
Signature · 1 line
2.20.1
author maintainer
#75 lore raw

PATCH 76/219 serial: sh-sci: Fix crash in rx_timer_fn() on PIO fallback

In reply to Sasha Levin

From: Geert Uytterhoeven <[email protected]>

[ Upstream commit 2e948218b7c1262a3830823d6620eb227e3d4e3a ]

When falling back to PIO, active_rx must be set to a different value than cookie_rx[i], else sci_dma_rx_find_active() will incorrectly find a match, leading to a NULL pointer dereference in rx_timer_fn() later.

Use zero instead, which is the same value as after driver initialization.

---
 drivers/tty/serial/sh-sci.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/tty/serial/sh-sci.c +1 -1
index 5550289e6678b..9e1a6af23ca2b 100644
@@ -1359,7 +1359,7 @@ static int sci_submit_rx(struct sci_port *s, bool port_lock_held)
 		dmaengine_terminate_async(chan);
 	for (i = 0; i < 2; i++)
 		s->cookie_rx[i] = -EINVAL;
-	s->active_rx = -EINVAL;
+	s->active_rx = 0;
 	s->chan_rx = NULL;
 	sci_start_rx(port);
 	if (!port_lock_held)
Signature · 1 line
2.20.1
author maintainer
#76 lore raw

PATCH 77/219 HID: intel-ish-hid: fixes incorrect error handling

In reply to Sasha Levin

From: Pan Bian <[email protected]>

[ Upstream commit 6e0856d317440a950b17c00a9283114f025e5699 ]

The memory chunk allocated by hid_allocate_device() should be released by hid_destroy_device(), not kfree().

Fixes 0b28cb4bcb1("HID: intel-ish-hid: ISH HID client driver")
---
 drivers/hid/intel-ish-hid/ishtp-hid.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/hid/intel-ish-hid/ishtp-hid.c +1 -1
index cd23903ddcf19..e918d78e541c0 100644
@@ -222,7 +222,7 @@ int ishtp_hid_probe(unsigned int cur_hid_dev,
 err_hid_device:
 	kfree(hid_data);
 err_hid_data:
-	kfree(hid);
+	hid_destroy_device(hid);
 	return rv;
 }
 
Signature · 1 line
2.20.1
author maintainer
#77 lore raw

PATCH 78/219 gpio: raspberrypi-exp: decrease refcount on firmware dt node

In reply to Sasha Levin

From: Nicolas Saenz Julienne <[email protected]>

[ Upstream commit 85af74c474b21940e88483fd48f6094145c89d97 ]

We're getting a reference RPi's firmware node in order to be able to communicate with it's driver. We should decrease the reference count on the dt node after being done with it.

Fixes a98d90e7d588 ("gpio: raspberrypi-exp: Driver for RPi3 GPIO expander via mailbox service")
---
 drivers/gpio/gpio-raspberrypi-exp.c | 1 +
 1 file changed, 1 insertion(+)
drivers/gpio/gpio-raspberrypi-exp.c +1 -0
index d6d36d537e373..b77ea16ffa031 100644
@@ -206,6 +206,7 @@ static int rpi_exp_gpio_probe(struct platform_device *pdev)
 	}
 
 	fw = rpi_firmware_get(fw_node);
+	of_node_put(fw_node);
 	if (!fw)
 		return -EPROBE_DEFER;
 
Signature · 1 line
2.20.1
author maintainer
#78 lore raw

PATCH 79/219 serial: 8250: Rate limit serial port rx interrupts during input overruns

In reply to Sasha Levin

From: Darwin Dingel <[email protected]>

[ Upstream commit 6d7f677a2afa1c82d7fc7af7f9159cbffd5dc010 ]

When a serial port gets faulty or gets flooded with inputs, its interrupt handler starts to work double time to get the characters to the workqueue for the tty layer to handle them. When this busy time on the serial/tty subsystem happens during boot, where it is also busy on the userspace trying to initialise, some processes can continuously get preempted and will be on hold until the interrupts subside.

The fix is to backoff on processing received characters for a specified amount of time when an input overrun is seen (received a new character before the previous one is processed). This only stops receive and will continue to transmit characters to serial port. After the backoff period is done, it receive will be re-enabled. This is optional and will only be enabled by setting 'overrun-throttle-ms' in the dts.

---
 drivers/tty/serial/8250/8250_core.c | 25 +++++++++++++++++++++++++
 drivers/tty/serial/8250/8250_fsl.c  | 23 ++++++++++++++++++++++-
 drivers/tty/serial/8250/8250_of.c   |  5 +++++
 include/linux/serial_8250.h         |  4 ++++
 4 files changed, 56 insertions(+), 1 deletion(-)
drivers/tty/serial/8250/8250_core.c +25 -0
index 8fe3d0ed229ed..0e65d4261f94c 100644
@@ -946,6 +946,21 @@ static struct uart_8250_port *serial8250_find_match_or_unused(struct uart_port *
 	return NULL;
 }
 
+static void serial_8250_overrun_backoff_work(struct work_struct *work)
+{
+	struct uart_8250_port *up =
+	    container_of(to_delayed_work(work), struct uart_8250_port,
+			 overrun_backoff);
+	struct uart_port *port = &up->port;
+	unsigned long flags;
+
+	spin_lock_irqsave(&port->lock, flags);
+	up->ier |= UART_IER_RLSI | UART_IER_RDI;
+	up->port.read_status_mask |= UART_LSR_DR;
+	serial_out(up, UART_IER, up->ier);
+	spin_unlock_irqrestore(&port->lock, flags);
+}
+
 /**
  *	serial8250_register_8250_port - register a serial port
  *	@up: serial port template
@@ -1060,6 +1075,16 @@ int serial8250_register_8250_port(struct uart_8250_port *up)
 			ret = 0;
 		}
 	}
+
+	/* Initialise interrupt backoff work if required */
+	if (up->overrun_backoff_time_ms > 0) {
+		uart->overrun_backoff_time_ms = up->overrun_backoff_time_ms;
+		INIT_DELAYED_WORK(&uart->overrun_backoff,
+				  serial_8250_overrun_backoff_work);
+	} else {
+		uart->overrun_backoff_time_ms = 0;
+	}
+
 	mutex_unlock(&serial_mutex);
 
 	return ret;
drivers/tty/serial/8250/8250_fsl.c +22 -1
index 6640a4c7ddd1d..bb9571eed275d 100644
@@ -45,8 +45,29 @@ int fsl8250_handle_irq(struct uart_port *port)
 
 	lsr = orig_lsr = up->port.serial_in(&up->port, UART_LSR);
 
-	if (lsr & (UART_LSR_DR | UART_LSR_BI))
+	/* Process incoming characters first */
+	if ((lsr & (UART_LSR_DR | UART_LSR_BI)) &&
+	    (up->ier & (UART_IER_RLSI | UART_IER_RDI))) {
 		lsr = serial8250_rx_chars(up, lsr);
+	}
+
+	/* Stop processing interrupts on input overrun */
+	if ((orig_lsr & UART_LSR_OE) && (up->overrun_backoff_time_ms > 0)) {
+		unsigned long delay;
+
+		up->ier = port->serial_in(port, UART_IER);
+		if (up->ier & (UART_IER_RLSI | UART_IER_RDI)) {
+			port->ops->stop_rx(port);
+		} else {
+			/* Keep restarting the timer until
+			 * the input overrun subsides.
+			 */
+			cancel_delayed_work(&up->overrun_backoff);
+		}
+
+		delay = msecs_to_jiffies(up->overrun_backoff_time_ms);
+		schedule_delayed_work(&up->overrun_backoff, delay);
+	}
 
 	serial8250_modem_status(up);
 
drivers/tty/serial/8250/8250_of.c +5 -0
index 98125de2f0a6c..2488de1c4bc4b 100644
@@ -244,6 +244,11 @@ static int of_platform_serial_probe(struct platform_device *ofdev)
 	if (of_property_read_bool(ofdev->dev.of_node, "auto-flow-control"))
 		port8250.capabilities |= UART_CAP_AFE;
 
+	if (of_property_read_u32(ofdev->dev.of_node,
+			"overrun-throttle-ms",
+			&port8250.overrun_backoff_time_ms) != 0)
+		port8250.overrun_backoff_time_ms = 0;
+
 	ret = serial8250_register_8250_port(&port8250);
 	if (ret < 0)
 		goto err_dispose;
include/linux/serial_8250.h +4 -0
index 18e21427bce43..5a655ba8d2730 100644
@@ -134,6 +134,10 @@ struct uart_8250_port {
 	void			(*dl_write)(struct uart_8250_port *, int);
 
 	struct uart_8250_em485 *em485;
+
+	/* Serial port overrun backoff */
+	struct delayed_work overrun_backoff;
+	u32 overrun_backoff_time_ms;
 };
 
 static inline struct uart_8250_port *up_to_u8250p(struct uart_port *up)
Signature · 1 line
2.20.1
author maintainer
#79 lore raw

PATCH 80/219 kprobes/x86/xen: blacklist non-attachable xen interrupt functions

In reply to Sasha Levin

From: Andrea Righi <[email protected]>

[ Upstream commit bf9445a33ae6ac2f0822d2f1ce1365408387d568 ]

Blacklist symbols in Xen probe-prohibited areas, so that user can see these prohibited symbols in debugfs.

See also: a50480cb6d61.

---
 arch/x86/xen/xen-asm_64.S | 2 ++
 1 file changed, 2 insertions(+)
arch/x86/xen/xen-asm_64.S +2 -0
index 3a6feed76dfc1..a93d8a7cef26c 100644
@@ -12,6 +12,7 @@
 #include <asm/segment.h>
 #include <asm/asm-offsets.h>
 #include <asm/thread_info.h>
+#include <asm/asm.h>
 
 #include <xen/interface/xen.h>
 
@@ -24,6 +25,7 @@ ENTRY(xen_\name)
 	pop %r11
 	jmp  \name
 END(xen_\name)
+_ASM_NOKPROBE(xen_\name)
 .endm
 
 xen_pv_trap divide_error
Signature · 7 lines
2.20.1


_______________________________________________
Xen-devel mailing list
[email protected]
https://lists.xenproject.org/mailman/listinfo/xen-devel
author maintainer
#80 lore raw

PATCH 81/219 xen/pciback: Check dev_data before using it

In reply to Sasha Levin

From: Ross Lagerwall <[email protected]>

[ Upstream commit 1669907e3d1abfa3f7586e2d55dbbc117b5adba2 ]

If pcistub_init_device fails, the release function will be called with dev_data set to NULL. Check it before using it to avoid a NULL pointer dereference.

---
 drivers/xen/xen-pciback/pci_stub.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/xen/xen-pciback/pci_stub.c +2 -1
index 59661db144e51..097410a7cdb74 100644
@@ -106,7 +106,8 @@ static void pcistub_device_release(struct kref *kref)
 	 * is called from "unbind" which takes a device_lock mutex.
 	 */
 	__pci_reset_function_locked(dev);
-	if (pci_load_and_free_saved_state(dev, &dev_data->pci_saved_state))
+	if (dev_data &&
+	    pci_load_and_free_saved_state(dev, &dev_data->pci_saved_state))
 		dev_info(&dev->dev, "Could not reload PCI state\n");
 	else
 		pci_restore_state(dev);
Signature · 1 line
2.20.1
author maintainer
#81 lore raw

PATCH 82/219 kprobes: Blacklist symbols in arch-defined prohibited area

In reply to Sasha Levin

From: Masami Hiramatsu <[email protected]>

[ Upstream commit fb1a59fae8baa3f3c69b72a87ff94fc4fa5683ec ]

Blacklist symbols in arch-defined probe-prohibited areas. With this change, user can see all symbols which are prohibited to probe in debugfs.

All archtectures which have custom prohibit areas should define its own arch_populate_kprobe_blacklist() function, but unless that, all symbols marked __kprobes are blacklisted.

---
 include/linux/kprobes.h |  3 ++
 kernel/kprobes.c        | 67 ++++++++++++++++++++++++++++++++---------
 2 files changed, 56 insertions(+), 14 deletions(-)
include/linux/kprobes.h +3 -0
index 32cae0f35b9d4..9adb92ad24d3f 100644
@@ -243,10 +243,13 @@ extern int arch_init_kprobes(void);
 extern void show_registers(struct pt_regs *regs);
 extern void kprobes_inc_nmissed_count(struct kprobe *p);
 extern bool arch_within_kprobe_blacklist(unsigned long addr);
+extern int arch_populate_kprobe_blacklist(void);
 extern bool arch_kprobe_on_func_entry(unsigned long offset);
 extern bool kprobe_on_func_entry(kprobe_opcode_t *addr, const char *sym, unsigned long offset);
 
 extern bool within_kprobe_blacklist(unsigned long addr);
+extern int kprobe_add_ksym_blacklist(unsigned long entry);
+extern int kprobe_add_area_blacklist(unsigned long start, unsigned long end);
 
 struct kprobe_insn_cache {
 	struct mutex mutex;
kernel/kprobes.c +53 -14

85 lines not on this page · raw message

Signature · 1 line
2.20.1
author maintainer
#82 lore raw

PATCH 83/219 kprobes/x86: Show x86-64 specific blacklisted symbols correctly

In reply to Sasha Levin

From: Masami Hiramatsu <[email protected]>

[ Upstream commit fe6e65615415987629a2dda583b4495677d8c388 ]

Show x86-64 specific blacklisted symbols in debugfs.

Since x86-64 prohibits probing on symbols which are in entry text, those should be shown.

---
 arch/x86/kernel/kprobes/core.c | 6 ++++++
 1 file changed, 6 insertions(+)
arch/x86/kernel/kprobes/core.c +6 -0
index e83a057564d1b..173e915e11d54 100644
@@ -1140,6 +1140,12 @@ bool arch_within_kprobe_blacklist(unsigned long addr)
 		is_in_entry_trampoline_section;
 }
 
+int __init arch_populate_kprobe_blacklist(void)
+{
+	return kprobe_add_area_blacklist((unsigned long)__entry_text_start,
+					 (unsigned long)__entry_text_end);
+}
+
 int __init arch_init_kprobes(void)
 {
 	return 0;
Signature · 1 line
2.20.1
author maintainer
#83 lore raw

PATCH 84/219 vfio-mdev/samples: Use u8 instead of char for handle functions

In reply to Sasha Levin

From: Nathan Chancellor <[email protected]>

[ Upstream commit 8ba35b3a0046d6573c98f00461d9bd1b86250d35 ]

Clang warns:

samples/vfio-mdev/mtty.c:592:39: warning: implicit conversion from 'int' to 'char' changes value from 162 to -94 [-Wconstant-conversion]

                *buf = UART_MSR_DSR | UART_MSR_DDSR | UART_MSR_DCD;
                     ~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~

1 warning generated.

Turns out that all uses of buf in this function ultimately end up stored or cast to an unsigned type. Just use u8, which has the same number of bits but can store this larger number so Clang no longer warns.

---
 samples/vfio-mdev/mtty.c | 26 +++++++++++++-------------
 1 file changed, 13 insertions(+), 13 deletions(-)
samples/vfio-mdev/mtty.c +13 -13
index 7abb79d8313d9..f6732aa16bb1f 100644
@@ -171,7 +171,7 @@ static struct mdev_state *find_mdev_state_by_uuid(uuid_le uuid)
 	return NULL;
 }
 
-void dump_buffer(char *buf, uint32_t count)
+void dump_buffer(u8 *buf, uint32_t count)
 {
 #if defined(DEBUG)
 	int i;
@@ -250,7 +250,7 @@ static void mtty_create_config_space(struct mdev_state *mdev_state)
 }
 
 static void handle_pci_cfg_write(struct mdev_state *mdev_state, u16 offset,
-				 char *buf, u32 count)
+				 u8 *buf, u32 count)
 {
 	u32 cfg_addr, bar_mask, bar_index = 0;
 
@@ -304,7 +304,7 @@ static void handle_pci_cfg_write(struct mdev_state *mdev_state, u16 offset,
 }
 
 static void handle_bar_write(unsigned int index, struct mdev_state *mdev_state,
-				u16 offset, char *buf, u32 count)
+				u16 offset, u8 *buf, u32 count)
 {
 	u8 data = *buf;
 
@@ -475,7 +475,7 @@ static void handle_bar_write(unsigned int index, struct mdev_state *mdev_state,
 }
 
 static void handle_bar_read(unsigned int index, struct mdev_state *mdev_state,
-			    u16 offset, char *buf, u32 count)
+			    u16 offset, u8 *buf, u32 count)
 {
 	/* Handle read requests by guest */
 	switch (offset) {
@@ -650,7 +650,7 @@ static void mdev_read_base(struct mdev_state *mdev_state)
 	}
 }
 
-static ssize_t mdev_access(struct mdev_device *mdev, char *buf, size_t count,
+static ssize_t mdev_access(struct mdev_device *mdev, u8 *buf, size_t count,
 			   loff_t pos, bool is_write)
 {
 	struct mdev_state *mdev_state;
@@ -698,7 +698,7 @@ static ssize_t mdev_access(struct mdev_device *mdev, char *buf, size_t count,
 #if defined(DEBUG_REGS)
 			pr_info("%s: BAR%d  WR @0x%llx %s val:0x%02x dlab:%d\n",
 				__func__, index, offset, wr_reg[offset],
-				(u8)*buf, mdev_state->s[index].dlab);
+				*buf, mdev_state->s[index].dlab);
 #endif
 			handle_bar_write(index, mdev_state, offset, buf, count);
 		} else {
@@ -708,7 +708,7 @@ static ssize_t mdev_access(struct mdev_device *mdev, char *buf, size_t count,
 #if defined(DEBUG_REGS)
 			pr_info("%s: BAR%d  RD @0x%llx %s val:0x%02x dlab:%d\n",
 				__func__, index, offset, rd_reg[offset],
-				(u8)*buf, mdev_state->s[index].dlab);
+				*buf, mdev_state->s[index].dlab);
 #endif
 		}
 		break;
@@ -827,7 +827,7 @@ ssize_t mtty_read(struct mdev_device *mdev, char __user *buf, size_t count,
 		if (count >= 4 && !(*ppos % 4)) {
 			u32 val;
 
-			ret =  mdev_access(mdev, (char *)&val, sizeof(val),
+			ret =  mdev_access(mdev, (u8 *)&val, sizeof(val),
 					   *ppos, false);
 			if (ret <= 0)
 				goto read_err;
@@ -839,7 +839,7 @@ ssize_t mtty_read(struct mdev_device *mdev, char __user *buf, size_t count,
 		} else if (count >= 2 && !(*ppos % 2)) {
 			u16 val;
 
-			ret = mdev_access(mdev, (char *)&val, sizeof(val),
+			ret = mdev_access(mdev, (u8 *)&val, sizeof(val),
 					  *ppos, false);
 			if (ret <= 0)
 				goto read_err;
@@ -851,7 +851,7 @@ ssize_t mtty_read(struct mdev_device *mdev, char __user *buf, size_t count,
 		} else {
 			u8 val;
 
-			ret = mdev_access(mdev, (char *)&val, sizeof(val),
+			ret = mdev_access(mdev, (u8 *)&val, sizeof(val),
 					  *ppos, false);
 			if (ret <= 0)
 				goto read_err;
@@ -889,7 +889,7 @@ ssize_t mtty_write(struct mdev_device *mdev, const char __user *buf,
 			if (copy_from_user(&val, buf, sizeof(val)))
 				goto write_err;
 
-			ret = mdev_access(mdev, (char *)&val, sizeof(val),
+			ret = mdev_access(mdev, (u8 *)&val, sizeof(val),
 					  *ppos, true);
 			if (ret <= 0)
 				goto write_err;
@@ -901,7 +901,7 @@ ssize_t mtty_write(struct mdev_device *mdev, const char __user *buf,
 			if (copy_from_user(&val, buf, sizeof(val)))
 				goto write_err;
 
-			ret = mdev_access(mdev, (char *)&val, sizeof(val),
+			ret = mdev_access(mdev, (u8 *)&val, sizeof(val),
 					  *ppos, true);
 			if (ret <= 0)
 				goto write_err;
@@ -913,7 +913,7 @@ ssize_t mtty_write(struct mdev_device *mdev, const char __user *buf,
 			if (copy_from_user(&val, buf, sizeof(val)))
 				goto write_err;
 
-			ret = mdev_access(mdev, (char *)&val, sizeof(val),
+			ret = mdev_access(mdev, (u8 *)&val, sizeof(val),
 					  *ppos, true);
 			if (ret <= 0)
 				goto write_err;
Signature · 1 line
2.20.1
author maintainer
#84 lore raw

PATCH 85/219 memory: omap-gpmc: Get the header of the enum

In reply to Sasha Levin

From: Linus Walleij <[email protected]>

[ Upstream commit a0752e9c3097b2c4fccd618802938e0951038dfa ]

Commit 21abf103818a
("gpio: Pass a flag to gpiochip_request_own_desc()") started to pass an enum gpiod_flags but this file is not including the header file that defines that enum and the compiler spits:

drivers/memory/omap-gpmc.c: In function
'gpmc_probe_generic_child':
drivers/memory/omap-gpmc.c:2174:9: error: type of formal

				   parameter 4 is incomplete
         0);
         ^
Cc Ladislav Michl <[email protected]>
Cc Janusz Krzysztofik <[email protected]>
Fixes 21abf103818a ("gpio: Pass a flag to gpiochip_request_own_desc()")
---
 drivers/memory/omap-gpmc.c | 1 +
 1 file changed, 1 insertion(+)
drivers/memory/omap-gpmc.c +1 -0
index c215287e80cf3..1c6a7c16e0c17 100644
@@ -21,6 +21,7 @@
 #include <linux/spinlock.h>
 #include <linux/io.h>
 #include <linux/gpio/driver.h>
+#include <linux/gpio/consumer.h> /* GPIO descriptor enum */
 #include <linux/interrupt.h>
 #include <linux/irqdomain.h>
 #include <linux/platform_device.h>
Signature · 1 line
2.20.1
author maintainer
#85 lore raw

PATCH 86/219 pinctrl: xway: fix gpio-hog related boot issues

In reply to Sasha Levin

From: Martin Schiller <[email protected]>

[ Upstream commit 9b4924da4711674e62d97d4f5360446cc78337af ]

This patch is based on commit a86caa9ba5d7 ("pinctrl: msm: fix gpio-hog related boot issues").

It fixes the issue that the gpio ranges needs to be defined before gpiochip_add().

Therefore, we also have to swap the order of registering the pinctrl driver and registering the gpio chip.

You also have to add the "gpio-ranges" property to the pinctrl device node to get it finally working.

---
 drivers/pinctrl/pinctrl-xway.c | 39 +++++++++++++++++++++++-----------
 1 file changed, 27 insertions(+), 12 deletions(-)
drivers/pinctrl/pinctrl-xway.c +27 -12
index 93f8bd04e7fe6..ae74b260b014b 100644
@@ -1746,14 +1746,6 @@ static int pinmux_xway_probe(struct platform_device *pdev)
 	}
 	xway_pctrl_desc.pins = xway_info.pads;
 
-	/* register the gpio chip */
-	xway_chip.parent = &pdev->dev;
-	ret = devm_gpiochip_add_data(&pdev->dev, &xway_chip, NULL);
-	if (ret) {
-		dev_err(&pdev->dev, "Failed to register gpio chip\n");
-		return ret;
-	}
-
 	/* setup the data needed by pinctrl */
 	xway_pctrl_desc.name	= dev_name(&pdev->dev);
 	xway_pctrl_desc.npins	= xway_chip.ngpio;
@@ -1775,10 +1767,33 @@ static int pinmux_xway_probe(struct platform_device *pdev)
 		return ret;
 	}
 
-	/* finish with registering the gpio range in pinctrl */
-	xway_gpio_range.npins = xway_chip.ngpio;
-	xway_gpio_range.base = xway_chip.base;
-	pinctrl_add_gpio_range(xway_info.pctrl, &xway_gpio_range);
+	/* register the gpio chip */
+	xway_chip.parent = &pdev->dev;
+	xway_chip.owner = THIS_MODULE;
+	xway_chip.of_node = pdev->dev.of_node;
+	ret = devm_gpiochip_add_data(&pdev->dev, &xway_chip, NULL);
+	if (ret) {
+		dev_err(&pdev->dev, "Failed to register gpio chip\n");
+		return ret;
+	}
+
+	/*
+	 * For DeviceTree-supported systems, the gpio core checks the
+	 * pinctrl's device node for the "gpio-ranges" property.
+	 * If it is present, it takes care of adding the pin ranges
+	 * for the driver. In this case the driver can skip ahead.
+	 *
+	 * In order to remain compatible with older, existing DeviceTree
+	 * files which don't set the "gpio-ranges" property or systems that
+	 * utilize ACPI the driver has to call gpiochip_add_pin_range().
+	 */
+	if (!of_property_read_bool(pdev->dev.of_node, "gpio-ranges")) {
+		/* finish with registering the gpio range in pinctrl */
+		xway_gpio_range.npins = xway_chip.ngpio;
+		xway_gpio_range.base = xway_chip.base;
+		pinctrl_add_gpio_range(xway_info.pctrl, &xway_gpio_range);
+	}
+
 	dev_info(&pdev->dev, "Init done\n");
 	return 0;
 }
Signature · 1 line
2.20.1
author maintainer
#86 lore raw

PATCH 87/219 net/mlx5: Continue driver initialization despite debugfs failure

In reply to Sasha Levin

From: Leon Romanovsky <[email protected]>

[ Upstream commit 199fa087dc6b503baad06712716fac645a983e8a ]

The failure to create debugfs entry is unpleasant event, but not enough to abort drier initialization. Align the mlx5_core code to debugfs design and continue execution whenever debugfs_create_dir() successes or not.

Fixes e126ba97dba9 ("mlx5: Add driver for Mellanox Connect-IB adapters")
---
 drivers/net/ethernet/mellanox/mlx5/core/main.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/main.c +3 -5
index 231ed508c240a..5fac00ea62457 100644
@@ -859,11 +859,9 @@ static int mlx5_pci_init(struct mlx5_core_dev *dev, struct mlx5_priv *priv)
 
 	priv->numa_node = dev_to_node(&dev->pdev->dev);
 
-	priv->dbg_root = debugfs_create_dir(dev_name(&pdev->dev), mlx5_debugfs_root);
-	if (!priv->dbg_root) {
-		dev_err(&pdev->dev, "Cannot create debugfs dir, aborting\n");
-		return -ENOMEM;
-	}
+	if (mlx5_debugfs_root)
+		priv->dbg_root =
+			debugfs_create_dir(pci_name(pdev), mlx5_debugfs_root);
 
 	err = mlx5_pci_enable_device(dev);
 	if (err) {
Signature · 1 line
2.20.1
author maintainer
#87 lore raw

PATCH 88/219 netfilter: nf_nat_sip: fix RTP/RTCP source port translations

In reply to Sasha Levin

From: Alin Nastac <[email protected]>

[ Upstream commit 8294059931448aa1ca112615bdffa3eab552c382 ]

Each media stream negotiation between 2 SIP peers will trigger creation of 4 different expectations (2 RTP and 2 RTCP):
- INVITE will create expectations for the media packets sent by the called peer
- reply to the INVITE will create expectations for media packets sent by the caller

The dport used by these expectations usually match the ones selected by the SIP peers, but they might get translated due to conflicts with another expectation. When such event occur, it is important to do this translation in both directions, dport translation on the receiving path and sport translation on the sending path.

This commit fixes the sport translation when the peer requiring it is also the one that starts the media stream. In this scenario, first media stream packet is forwarded from LAN to WAN and will rely on nf_nat_sip_expected() to do the necessary sport translation. However, the expectation matched by this packet does not contain the necessary information for doing SNAT, this data being stored in the paired expectation created by the sender's SIP message (INVITE or reply to it).

---
 net/netfilter/nf_nat_sip.c | 39 ++++++++++++++++++++++++++++++++++----
 1 file changed, 35 insertions(+), 4 deletions(-)
net/netfilter/nf_nat_sip.c +35 -4
index 1f30860749817..aa1be643d7a09 100644
@@ -18,6 +18,7 @@
 
 #include <net/netfilter/nf_nat.h>
 #include <net/netfilter/nf_nat_helper.h>
+#include <net/netfilter/nf_conntrack_core.h>
 #include <net/netfilter/nf_conntrack_helper.h>
 #include <net/netfilter/nf_conntrack_expect.h>
 #include <net/netfilter/nf_conntrack_seqadj.h>
@@ -316,6 +317,9 @@ static void nf_nat_sip_seq_adjust(struct sk_buff *skb, unsigned int protoff,
 static void nf_nat_sip_expected(struct nf_conn *ct,
 				struct nf_conntrack_expect *exp)
 {
+	struct nf_conn_help *help = nfct_help(ct->master);
+	struct nf_conntrack_expect *pair_exp;
+	int range_set_for_snat = 0;
 	struct nf_nat_range2 range;
 
 	/* This must be a fresh one. */
@@ -327,15 +331,42 @@ static void nf_nat_sip_expected(struct nf_conn *ct,
 	range.min_addr = range.max_addr = exp->saved_addr;
 	nf_nat_setup_info(ct, &range, NF_NAT_MANIP_DST);
 
-	/* Change src to where master sends to, but only if the connection
-	 * actually came from the same source. */
-	if (nf_inet_addr_cmp(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3,
+	/* Do media streams SRC manip according with the parameters
+	 * found in the paired expectation.
+	 */
+	if (exp->class != SIP_EXPECT_SIGNALLING) {
+		spin_lock_bh(&nf_conntrack_expect_lock);
+		hlist_for_each_entry(pair_exp, &help->expectations, lnode) {
+			if (pair_exp->tuple.src.l3num == nf_ct_l3num(ct) &&
+			    pair_exp->tuple.dst.protonum == ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.dst.protonum &&
+			    nf_inet_addr_cmp(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3, &pair_exp->saved_addr) &&
+			    ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u.all == pair_exp->saved_proto.all) {
+				range.flags = (NF_NAT_RANGE_MAP_IPS | NF_NAT_RANGE_PROTO_SPECIFIED);
+				range.min_proto.all = range.max_proto.all = pair_exp->tuple.dst.u.all;
+				range.min_addr = range.max_addr = pair_exp->tuple.dst.u3;
+				range_set_for_snat = 1;
+				break;
+			}
+		}
+		spin_unlock_bh(&nf_conntrack_expect_lock);
+	}
+
+	/* When no paired expectation has been found, change src to
+	 * where master sends to, but only if the connection actually came
+	 * from the same source.
+	 */
+	if (!range_set_for_snat &&
+	    nf_inet_addr_cmp(&ct->tuplehash[IP_CT_DIR_ORIGINAL].tuple.src.u3,
 			     &ct->master->tuplehash[exp->dir].tuple.src.u3)) {
 		range.flags = NF_NAT_RANGE_MAP_IPS;
 		range.min_addr = range.max_addr
 			= ct->master->tuplehash[!exp->dir].tuple.dst.u3;
-		nf_nat_setup_info(ct, &range, NF_NAT_MANIP_SRC);
+		range_set_for_snat = 1;
 	}
+
+	/* Perform SRC manip. */
+	if (range_set_for_snat)
+		nf_nat_setup_info(ct, &range, NF_NAT_MANIP_SRC);
 }
 
 static unsigned int nf_nat_sip_expect(struct sk_buff *skb, unsigned int protoff,
Signature · 1 line
2.20.1
author maintainer
#88 lore raw

PATCH 89/219 exofs_mount(): fix leaks on failure exits

In reply to Sasha Levin

From: Al Viro <[email protected]>

[ Upstream commit 26cb5a328c6b2bda9e859307ce4cfc60df3a2c28 ]

... and don't abuse mount_nodev(), while we are at it.

---
 fs/exofs/super.c | 37 +++++++++++++++++++++++++++++--------
 1 file changed, 29 insertions(+), 8 deletions(-)
fs/exofs/super.c +29 -8
index 7d61e3fa378c2..8fb98bb942372 100644
@@ -705,21 +705,18 @@ static int exofs_read_lookup_dev_table(struct exofs_sb_info *sbi,
 /*
  * Read the superblock from the OSD and fill in the fields
  */
-static int exofs_fill_super(struct super_block *sb, void *data, int silent)
+static int exofs_fill_super(struct super_block *sb,
+				struct exofs_mountopt *opts,
+				struct exofs_sb_info *sbi,
+				int silent)
 {
 	struct inode *root;
-	struct exofs_mountopt *opts = data;
-	struct exofs_sb_info *sbi;	/*extended info                  */
 	struct osd_dev *od;		/* Master device                 */
 	struct exofs_fscb fscb;		/*on-disk superblock info        */
 	struct ore_comp comp;
 	unsigned table_count;
 	int ret;
 
-	sbi = kzalloc(sizeof(*sbi), GFP_KERNEL);
-	if (!sbi)
-		return -ENOMEM;
-
 	/* use mount options to fill superblock */
 	if (opts->is_osdname) {
 		struct osd_dev_info odi = {.systemid_len = 0};
@@ -863,7 +860,9 @@ static struct dentry *exofs_mount(struct file_system_type *type,
 			  int flags, const char *dev_name,
 			  void *data)
 {
+	struct super_block *s;
 	struct exofs_mountopt opts;
+	struct exofs_sb_info *sbi;
 	int ret;
 
 	ret = parse_options(data, &opts);
@@ -872,9 +871,31 @@ static struct dentry *exofs_mount(struct file_system_type *type,
 		return ERR_PTR(ret);
 	}
 
+	sbi = kzalloc(sizeof(*sbi), GFP_KERNEL);
+	if (!sbi) {
+		kfree(opts.dev_name);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	s = sget(type, NULL, set_anon_super, flags, NULL);
+
+	if (IS_ERR(s)) {
+		kfree(opts.dev_name);
+		kfree(sbi);
+		return ERR_CAST(s);
+	}
+
 	if (!opts.dev_name)
 		opts.dev_name = dev_name;
-	return mount_nodev(type, flags, &opts, exofs_fill_super);
+
+
+	ret = exofs_fill_super(s, &opts, sbi, flags & SB_SILENT ? 1 : 0);
+	if (ret) {
+		deactivate_locked_super(s);
+		return ERR_PTR(ret);
+	}
+	s->s_flags |= SB_ACTIVE;
+	return dget(s->s_root);
 }
 
 /*
Signature · 1 line
2.20.1
author maintainer
#89 lore raw

PATCH 90/219 bnxt_en: Return linux standard errors in bnxt_ethtool.c

In reply to Sasha Levin

From: Vasundhara Volam <[email protected]>

[ Upstream commit 7c675421afef18253a86ffc383f57bc15ef32ea8 ]

Currently firmware specific errors are returned directly in flash_device and reset ethtool hooks. Modify it to return linux standard errors to userspace when flashing operations fail.

---
 .../net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 56 +++++++++++++------
 1 file changed, 39 insertions(+), 17 deletions(-)
drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +39 -17
index da9b876899963..eaa285bf908b9 100644
@@ -1444,14 +1444,22 @@ static int bnxt_flash_nvram(struct net_device *dev,
 	rc = hwrm_send_message(bp, &req, sizeof(req), FLASH_NVRAM_TIMEOUT);
 	dma_free_coherent(&bp->pdev->dev, data_len, kmem, dma_handle);
 
+	if (rc == HWRM_ERR_CODE_RESOURCE_ACCESS_DENIED) {
+		netdev_info(dev,
+			    "PF does not have admin privileges to flash the device\n");
+		rc = -EACCES;
+	} else if (rc) {
+		rc = -EIO;
+	}
 	return rc;
 }
 
 static int bnxt_firmware_reset(struct net_device *dev,
 			       u16 dir_type)
 {
-	struct bnxt *bp = netdev_priv(dev);
 	struct hwrm_fw_reset_input req = {0};
+	struct bnxt *bp = netdev_priv(dev);
+	int rc;
 
 	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_FW_RESET, -1, -1);
 
@@ -1491,7 +1499,15 @@ static int bnxt_firmware_reset(struct net_device *dev,
 		return -EINVAL;
 	}
 
-	return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	rc = hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	if (rc == HWRM_ERR_CODE_RESOURCE_ACCESS_DENIED) {
+		netdev_info(dev,
+			    "PF does not have admin privileges to reset the device\n");
+		rc = -EACCES;
+	} else if (rc) {
+		rc = -EIO;
+	}
+	return rc;
 }
 
 static int bnxt_flash_firmware(struct net_device *dev,
@@ -1698,9 +1714,9 @@ static int bnxt_flash_package_from_file(struct net_device *dev,
 	struct hwrm_nvm_install_update_output *resp = bp->hwrm_cmd_resp_addr;
 	struct hwrm_nvm_install_update_input install = {0};
 	const struct firmware *fw;
+	int rc, hwrm_err = 0;
 	u32 item_len;
 	u16 index;
-	int rc;
 
 	bnxt_hwrm_fw_set_time(bp);
 
@@ -1743,15 +1759,16 @@ static int bnxt_flash_package_from_file(struct net_device *dev,
 			memcpy(kmem, fw->data, fw->size);
 			modify.host_src_addr = cpu_to_le64(dma_handle);
 
-			rc = hwrm_send_message(bp, &modify, sizeof(modify),
-					       FLASH_PACKAGE_TIMEOUT);
+			hwrm_err = hwrm_send_message(bp, &modify,
+						     sizeof(modify),
+						     FLASH_PACKAGE_TIMEOUT);
 			dma_free_coherent(&bp->pdev->dev, fw->size, kmem,
 					  dma_handle);
 		}
 	}
 	release_firmware(fw);
-	if (rc)
-		return rc;
+	if (rc || hwrm_err)
+		goto err_exit;
 
 	if ((install_type & 0xffff) == 0)
 		install_type >>= 16;
@@ -1759,12 +1776,10 @@ static int bnxt_flash_package_from_file(struct net_device *dev,
 	install.install_type = cpu_to_le32(install_type);
 
 	mutex_lock(&bp->hwrm_cmd_lock);
-	rc = _hwrm_send_message(bp, &install, sizeof(install),
-				INSTALL_PACKAGE_TIMEOUT);
-	if (rc) {
-		rc = -EOPNOTSUPP;
+	hwrm_err = _hwrm_send_message(bp, &install, sizeof(install),
+				      INSTALL_PACKAGE_TIMEOUT);
+	if (hwrm_err)
 		goto flash_pkg_exit;
-	}
 
 	if (resp->error_code) {
 		u8 error_code = ((struct hwrm_err_output *)resp)->cmd_err;
@@ -1772,12 +1787,11 @@ static int bnxt_flash_package_from_file(struct net_device *dev,
 		if (error_code == NVM_INSTALL_UPDATE_CMD_ERR_CODE_FRAG_ERR) {
 			install.flags |= cpu_to_le16(
 			       NVM_INSTALL_UPDATE_REQ_FLAGS_ALLOWED_TO_DEFRAG);
-			rc = _hwrm_send_message(bp, &install, sizeof(install),
-						INSTALL_PACKAGE_TIMEOUT);
-			if (rc) {
-				rc = -EOPNOTSUPP;
+			hwrm_err = _hwrm_send_message(bp, &install,
+						      sizeof(install),
+						      INSTALL_PACKAGE_TIMEOUT);
+			if (hwrm_err)
 				goto flash_pkg_exit;
-			}
 		}
 	}
 
@@ -1788,6 +1802,14 @@ static int bnxt_flash_package_from_file(struct net_device *dev,
 	}
 flash_pkg_exit:
 	mutex_unlock(&bp->hwrm_cmd_lock);
+err_exit:
+	if (hwrm_err == HWRM_ERR_CODE_RESOURCE_ACCESS_DENIED) {
+		netdev_info(dev,
+			    "PF does not have admin privileges to flash the device\n");
+		rc = -EACCES;
+	} else if (hwrm_err) {
+		rc = -EOPNOTSUPP;
+	}
 	return rc;
 }
 
Signature · 1 line
2.20.1
author maintainer
#90 lore raw

PATCH 91/219 bnxt_en: Save ring statistics before reset.

In reply to Sasha Levin

From: Michael Chan <[email protected]>

[ Upstream commit b8875ca356f1c0b17ec68be6666269373a62288e ]

With the current driver, the statistics reported by .ndo_get_stats64() are reset when the device goes down. Store a snapshot of the rtnl_link_stats64 before shutdown. This snapshot is added to the current counters in .ndo_get_stats64() so that the counters will not get reset when the device is down.

---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 57 +++++++++++++++++------
 drivers/net/ethernet/broadcom/bnxt/bnxt.h |  1 +
 2 files changed, 44 insertions(+), 14 deletions(-)
drivers/net/ethernet/broadcom/bnxt/bnxt.c +43 -14
index c54a74de7b088..2f61175f5655a 100644
@@ -7148,6 +7148,9 @@ static bool bnxt_drv_busy(struct bnxt *bp)
 		test_bit(BNXT_STATE_READ_STATS, &bp->state));
 }
 
+static void bnxt_get_ring_stats(struct bnxt *bp,
+				struct rtnl_link_stats64 *stats);
+
 static void __bnxt_close_nic(struct bnxt *bp, bool irq_re_init,
 			     bool link_re_init)
 {
@@ -7173,6 +7176,9 @@ static void __bnxt_close_nic(struct bnxt *bp, bool irq_re_init,
 	del_timer_sync(&bp->timer);
 	bnxt_free_skbs(bp);
 
+	/* Save ring stats before shutdown */
+	if (bp->bnapi)
+		bnxt_get_ring_stats(bp, &bp->net_stats_prev);
 	if (irq_re_init) {
 		bnxt_free_irq(bp);
 		bnxt_del_napi(bp);
@@ -7234,23 +7240,12 @@ static int bnxt_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
 	return -EOPNOTSUPP;
 }
 
-static void
-bnxt_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats)
+static void bnxt_get_ring_stats(struct bnxt *bp,
+				struct rtnl_link_stats64 *stats)
 {
-	u32 i;
-	struct bnxt *bp = netdev_priv(dev);
+	int i;
 
-	set_bit(BNXT_STATE_READ_STATS, &bp->state);
-	/* Make sure bnxt_close_nic() sees that we are reading stats before
-	 * we check the BNXT_STATE_OPEN flag.
-	 */
-	smp_mb__after_atomic();
-	if (!test_bit(BNXT_STATE_OPEN, &bp->state)) {
-		clear_bit(BNXT_STATE_READ_STATS, &bp->state);
-		return;
-	}
 
-	/* TODO check if we need to synchronize with bnxt_close path */
 	for (i = 0; i < bp->cp_nr_rings; i++) {
 		struct bnxt_napi *bnapi = bp->bnapi[i];
 		struct bnxt_cp_ring_info *cpr = &bnapi->cp_ring;
@@ -7279,6 +7274,40 @@ bnxt_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats)
 
 		stats->tx_dropped += le64_to_cpu(hw_stats->tx_drop_pkts);
 	}
+}
+
+static void bnxt_add_prev_stats(struct bnxt *bp,
+				struct rtnl_link_stats64 *stats)
+{
+	struct rtnl_link_stats64 *prev_stats = &bp->net_stats_prev;
+
+	stats->rx_packets += prev_stats->rx_packets;
+	stats->tx_packets += prev_stats->tx_packets;
+	stats->rx_bytes += prev_stats->rx_bytes;
+	stats->tx_bytes += prev_stats->tx_bytes;
+	stats->rx_missed_errors += prev_stats->rx_missed_errors;
+	stats->multicast += prev_stats->multicast;
+	stats->tx_dropped += prev_stats->tx_dropped;
+}
+
+static void
+bnxt_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats)
+{
+	struct bnxt *bp = netdev_priv(dev);
+
+	set_bit(BNXT_STATE_READ_STATS, &bp->state);
+	/* Make sure bnxt_close_nic() sees that we are reading stats before
+	 * we check the BNXT_STATE_OPEN flag.
+	 */
+	smp_mb__after_atomic();
+	if (!test_bit(BNXT_STATE_OPEN, &bp->state)) {
+		clear_bit(BNXT_STATE_READ_STATS, &bp->state);
+		*stats = bp->net_stats_prev;
+		return;
+	}
+
+	bnxt_get_ring_stats(bp, stats);
+	bnxt_add_prev_stats(bp, stats);
 
 	if (bp->flags & BNXT_FLAG_PORT_STATS) {
 		struct rx_port_stats *rx = bp->hw_rx_port_stats;
drivers/net/ethernet/broadcom/bnxt/bnxt.h +1 -0
index cf2d4a6583d55..f9e253b705ece 100644
@@ -1302,6 +1302,7 @@ struct bnxt {
 	void			*hwrm_cmd_resp_addr;
 	dma_addr_t		hwrm_cmd_resp_dma_addr;
 
+	struct rtnl_link_stats64	net_stats_prev;
 	struct rx_port_stats	*hw_rx_port_stats;
 	struct tx_port_stats	*hw_tx_port_stats;
 	struct rx_port_stats_ext	*hw_rx_port_stats_ext;
Signature · 1 line
2.20.1
author maintainer
#91 lore raw

PATCH 92/219 bnxt_en: query force speeds before disabling autoneg mode.

In reply to Sasha Levin

From: Vasundhara Volam <[email protected]>

[ Upstream commit 56d374624778652d2a999e18c87a25338b127b41 ]

With autoneg enabled, PHY loopback test fails. To disable autoneg, driver needs to send a valid forced speed to FW. FW is not sending async event for invalid speeds. To fix this, query forced speeds and send the correct speed when disabling autoneg mode.

---
 .../net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 22 ++++++++++++++++++-
 1 file changed, 21 insertions(+), 1 deletion(-)
drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +21 -1
index eaa285bf908b9..2240c23b0a4c9 100644
@@ -2390,17 +2390,37 @@ static int bnxt_hwrm_mac_loopback(struct bnxt *bp, bool enable)
 	return hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
 }
 
+static int bnxt_query_force_speeds(struct bnxt *bp, u16 *force_speeds)
+{
+	struct hwrm_port_phy_qcaps_output *resp = bp->hwrm_cmd_resp_addr;
+	struct hwrm_port_phy_qcaps_input req = {0};
+	int rc;
+
+	bnxt_hwrm_cmd_hdr_init(bp, &req, HWRM_PORT_PHY_QCAPS, -1, -1);
+	mutex_lock(&bp->hwrm_cmd_lock);
+	rc = _hwrm_send_message(bp, &req, sizeof(req), HWRM_CMD_TIMEOUT);
+	if (!rc)
+		*force_speeds = le16_to_cpu(resp->supported_speeds_force_mode);
+
+	mutex_unlock(&bp->hwrm_cmd_lock);
+	return rc;
+}
+
 static int bnxt_disable_an_for_lpbk(struct bnxt *bp,
 				    struct hwrm_port_phy_cfg_input *req)
 {
 	struct bnxt_link_info *link_info = &bp->link_info;
-	u16 fw_advertising = link_info->advertising;
+	u16 fw_advertising;
 	u16 fw_speed;
 	int rc;
 
 	if (!link_info->autoneg)
 		return 0;
 
+	rc = bnxt_query_force_speeds(bp, &fw_advertising);
+	if (rc)
+		return rc;
+
 	fw_speed = PORT_PHY_CFG_REQ_FORCE_LINK_SPEED_1GB;
 	if (netif_carrier_ok(bp->dev))
 		fw_speed = bp->link_info.link_speed;
Signature · 1 line
2.20.1
author maintainer
#92 lore raw

PATCH 93/219 KVM: s390: unregister debug feature on failing arch init

In reply to Sasha Levin

From: Michael Mueller <[email protected]>

[ Upstream commit 308c3e6673b012beecb96ef04cc65f4a0e7cdd99 ]

Make sure the debug feature and its allocated resources get released upon unsuccessful architecture initialization.

A related indication of the issue will be reported as kernel message.

Message-Id: <[email protected]>

---
 arch/s390/kvm/kvm-s390.c | 17 ++++++++++++++---
 1 file changed, 14 insertions(+), 3 deletions(-)
arch/s390/kvm/kvm-s390.c +14 -3
index 3c317bc6b7997..db3196aebaa1c 100644
@@ -416,19 +416,30 @@ static void kvm_s390_cpu_feat_init(void)
 
 int kvm_arch_init(void *opaque)
 {
+	int rc;
+
 	kvm_s390_dbf = debug_register("kvm-trace", 32, 1, 7 * sizeof(long));
 	if (!kvm_s390_dbf)
 		return -ENOMEM;
 
 	if (debug_register_view(kvm_s390_dbf, &debug_sprintf_view)) {
-		debug_unregister(kvm_s390_dbf);
-		return -ENOMEM;
+		rc = -ENOMEM;
+		goto out_debug_unreg;
 	}
 
 	kvm_s390_cpu_feat_init();
 
 	/* Register floating interrupt controller interface. */
-	return kvm_register_device_ops(&kvm_flic_ops, KVM_DEV_TYPE_FLIC);
+	rc = kvm_register_device_ops(&kvm_flic_ops, KVM_DEV_TYPE_FLIC);
+	if (rc) {
+		pr_err("Failed to register FLIC rc=%d\n", rc);
+		goto out_debug_unreg;
+	}
+	return 0;
+
+out_debug_unreg:
+	debug_unregister(kvm_s390_dbf);
+	return rc;
 }
 
 void kvm_arch_exit(void)
Signature · 1 line
2.20.1
author maintainer
#93 lore raw

PATCH 94/219 pinctrl: sh-pfc: r8a77990: Fix MOD_SEL0 SEL_I2C1 field width

In reply to Sasha Levin

From: Geert Uytterhoeven <[email protected]>

[ Upstream commit 755a5b805fa7ff22e2934d67501efd92109f41ea ]

The SEL_I2C1 (MOD_SEL0[21:20]) field in Module Select Register 0 has a width of 2 bits, i.e. it allows programming one out of 4 different configurations.
However, the MOD_SEL0_21_20 macro contains 8 values instead of 4, overflowing into the subsequent fields in the register, and thus breaking the configuration of the latter.

Fix this by dropping the bogus last 4 values, including the non-existent SEL_I2C1_4 configuration.

Fixes 6d4036a1e3b3ac0f ("pinctrl: sh-pfc: Initial R8A77990 PFC support")
---
 drivers/pinctrl/sh-pfc/pfc-r8a77990.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/pinctrl/sh-pfc/pfc-r8a77990.c +1 -1
index b81c807ac54d5..d2fcf7f7b9668 100644
@@ -395,7 +395,7 @@ FM(IP12_31_28)	IP12_31_28	FM(IP13_31_28)	IP13_31_28	FM(IP14_31_28)	IP14_31_28	FM
 #define MOD_SEL0_24		FM(SEL_HSCIF0_0)		FM(SEL_HSCIF0_1)
 #define MOD_SEL0_23		FM(SEL_HSCIF1_0)		FM(SEL_HSCIF1_1)
 #define MOD_SEL0_22		FM(SEL_HSCIF2_0)		FM(SEL_HSCIF2_1)
-#define MOD_SEL0_21_20		FM(SEL_I2C1_0)			FM(SEL_I2C1_1)			FM(SEL_I2C1_2)			FM(SEL_I2C1_3)		FM(SEL_I2C1_4)		F_(0, 0)	F_(0, 0)	F_(0, 0)
+#define MOD_SEL0_21_20		FM(SEL_I2C1_0)			FM(SEL_I2C1_1)			FM(SEL_I2C1_2)			FM(SEL_I2C1_3)
 #define MOD_SEL0_19_18_17	FM(SEL_I2C2_0)			FM(SEL_I2C2_1)			FM(SEL_I2C2_2)			FM(SEL_I2C2_3)		FM(SEL_I2C2_4)		F_(0, 0)	F_(0, 0)	F_(0, 0)
 #define MOD_SEL0_16		FM(SEL_NDFC_0)			FM(SEL_NDFC_1)
 #define MOD_SEL0_15		FM(SEL_PWM0_0)			FM(SEL_PWM0_1)
Signature · 1 line
2.20.1
author maintainer
#94 lore raw

PATCH 95/219 pinctrl: sh-pfc: sh7264: Fix PFCR3 and PFCR0 register configuration

In reply to Sasha Levin

From: Geert Uytterhoeven <[email protected]>

[ Upstream commit 1b99d0c80bbe1810572c2cb77b90f67886adfa8d ]

The Port F Control Register 3 (PFCR3) contains only a single field. However, counting from left to right, it is the fourth field, not the first field.
Insert the missing dummy configuration values (3 fields of 16 values) to fix this.

The descriptor for the Port F Control Register 0 (PFCR0) lacks the description for the 4th field (PF0 Mode, PF0MD[2:0]). Add the missing configuration values to fix this.

Fixes a8d42fc4217b1ea1 ("sh-pfc: Add sh7264 pinmux support")
---
 drivers/pinctrl/sh-pfc/pfc-sh7264.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)
drivers/pinctrl/sh-pfc/pfc-sh7264.c +7 -2
index 8070765311dbf..e1c34e19222ee 100644
@@ -1716,6 +1716,9 @@ static const struct pinmux_cfg_reg pinmux_config_regs[] = {
 	},
 
 	{ PINMUX_CFG_REG("PFCR3", 0xfffe38a8, 16, 4) {
+		0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+		0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
+		0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
 		PF12MD_000, PF12MD_001, 0, PF12MD_011,
 		PF12MD_100, PF12MD_101, 0, 0,
 		0, 0, 0, 0, 0, 0, 0, 0 }
@@ -1759,8 +1762,10 @@ static const struct pinmux_cfg_reg pinmux_config_regs[] = {
 		0, 0, 0, 0, 0, 0, 0, 0,
 		PF1MD_000, PF1MD_001, PF1MD_010, PF1MD_011,
 		PF1MD_100, PF1MD_101, 0, 0,
-		0, 0, 0, 0, 0, 0, 0, 0
-	 }
+		0, 0, 0, 0, 0, 0, 0, 0,
+		PF0MD_000, PF0MD_001, PF0MD_010, PF0MD_011,
+		PF0MD_100, PF0MD_101, 0, 0,
+		0, 0, 0, 0, 0, 0, 0, 0 }
 	},
 
 	{ PINMUX_CFG_REG("PFIOR0", 0xfffe38b2, 16, 1) {
Signature · 1 line
2.20.1
author maintainer
#95 lore raw

PATCH 96/219 pinctrl: sh-pfc: sh7734: Fix shifted values in IPSR10

In reply to Sasha Levin

From: Geert Uytterhoeven <[email protected]>

[ Upstream commit 054f2400f706327f96770219c3065b5131f8f154 ]

Some values in the Peripheral Function Select Register 10 descriptor are shifted by one position, which may cause a peripheral function to be programmed incorrectly.

Fixing this makes all HSCIF0 pins use Function 4 (value 3), like was already the case for the HSCK0 pin in field IP10[5:3].

Fixes ac1ebc2190f575fc ("sh-pfc: Add sh7734 pinmux support")
---
 drivers/pinctrl/sh-pfc/pfc-sh7734.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)
drivers/pinctrl/sh-pfc/pfc-sh7734.c +8 -8
index 6502e676d3686..33232041ee86d 100644
@@ -2213,22 +2213,22 @@ static const struct pinmux_cfg_reg pinmux_config_regs[] = {
 	    /* IP10_22 [1] */
 		FN_CAN_CLK_A, FN_RX4_D,
 	    /* IP10_21_19 [3] */
-		FN_AUDIO_CLKOUT, FN_TX1_E, FN_HRTS0_C, FN_FSE_B,
-		FN_LCD_M_DISP_B, 0, 0, 0,
+		FN_AUDIO_CLKOUT, FN_TX1_E, 0, FN_HRTS0_C, FN_FSE_B,
+		FN_LCD_M_DISP_B, 0, 0,
 	    /* IP10_18_16 [3] */
-		FN_AUDIO_CLKC, FN_SCK1_E, FN_HCTS0_C, FN_FRB_B,
-		FN_LCD_VEPWC_B, 0, 0, 0,
+		FN_AUDIO_CLKC, FN_SCK1_E, 0, FN_HCTS0_C, FN_FRB_B,
+		FN_LCD_VEPWC_B, 0, 0,
 	    /* IP10_15 [1] */
 		FN_AUDIO_CLKB_A, FN_LCD_CLK_B,
 	    /* IP10_14_12 [3] */
 		FN_AUDIO_CLKA_A, FN_VI1_CLK_B, FN_SCK1_D, FN_IECLK_B,
 		FN_LCD_FLM_B, 0, 0, 0,
 	    /* IP10_11_9 [3] */
-		FN_SSI_SDATA3, FN_VI1_7_B, FN_HTX0_C, FN_FWE_B,
-		FN_LCD_CL2_B, 0, 0, 0,
+		FN_SSI_SDATA3, FN_VI1_7_B, 0, FN_HTX0_C, FN_FWE_B,
+		FN_LCD_CL2_B, 0, 0,
 	    /* IP10_8_6 [3] */
-		FN_SSI_SDATA2, FN_VI1_6_B, FN_HRX0_C, FN_FRE_B,
-		FN_LCD_CL1_B, 0, 0, 0,
+		FN_SSI_SDATA2, FN_VI1_6_B, 0, FN_HRX0_C, FN_FRE_B,
+		FN_LCD_CL1_B, 0, 0,
 	    /* IP10_5_3 [3] */
 		FN_SSI_WS23, FN_VI1_5_B, FN_TX1_D, FN_HSCK0_C, FN_FALE_B,
 		FN_LCD_DON_B, 0, 0, 0,
Signature · 1 line
2.20.1
author maintainer
#96 lore raw

PATCH 97/219 HID: doc: fix wrong data structure reference for UHID_OUTPUT

In reply to Sasha Levin

From: Peter Hutterer <[email protected]>

[ Upstream commit 46b14eef59a8157138dc02f916a7f97c73b3ec53 ]

---
 Documentation/hid/uhid.txt | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
Documentation/hid/uhid.txt +1 -1
index c8656dd029a91..958fff9453044 100644
@@ -160,7 +160,7 @@ them but you should handle them according to your needs.
   UHID_OUTPUT:
   This is sent if the HID device driver wants to send raw data to the I/O
   device on the interrupt channel. You should read the payload and forward it to
-  the device. The payload is of type "struct uhid_data_req".
+  the device. The payload is of type "struct uhid_output_req".
   This may be received even though you haven't received UHID_OPEN, yet.
 
   UHID_GET_REPORT:
Signature · 1 line
2.20.1
author maintainer
#97 lore raw

PATCH 98/219 dm flakey: Properly corrupt multi-page bios.

In reply to Sasha Levin

From: Sweet Tea <[email protected]>

[ Upstream commit a00f5276e26636cbf72f24f79831026d2e2868e7 ]

The flakey target is documented to be able to corrupt the Nth byte in a bio, but does not corrupt byte indices after the first biovec in the bio. Change the corrupting function to actually corrupt the Nth byte no matter in which biovec that index falls.

A test device generating two-page bios, atop a flakey device configured to corrupt a byte index on the second page, verified both the failure to corrupt before this patch and the expected corruption after this change.

---
 drivers/md/dm-flakey.c | 33 ++++++++++++++++++++++-----------
 1 file changed, 22 insertions(+), 11 deletions(-)
drivers/md/dm-flakey.c +22 -11
index b86d2439ffc76..2fcf62fb2844f 100644
@@ -287,20 +287,31 @@ static void flakey_map_bio(struct dm_target *ti, struct bio *bio)
 
 static void corrupt_bio_data(struct bio *bio, struct flakey_c *fc)
 {
-	unsigned bio_bytes = bio_cur_bytes(bio);
-	char *data = bio_data(bio);
+	unsigned int corrupt_bio_byte = fc->corrupt_bio_byte - 1;
+
+	struct bvec_iter iter;
+	struct bio_vec bvec;
+
+	if (!bio_has_data(bio))
+		return;
 
 	/*
-	 * Overwrite the Nth byte of the data returned.
+	 * Overwrite the Nth byte of the bio's data, on whichever page
+	 * it falls.
 	 */
-	if (data && bio_bytes >= fc->corrupt_bio_byte) {
-		data[fc->corrupt_bio_byte - 1] = fc->corrupt_bio_value;
-
-		DMDEBUG("Corrupting data bio=%p by writing %u to byte %u "
-			"(rw=%c bi_opf=%u bi_sector=%llu cur_bytes=%u)\n",
-			bio, fc->corrupt_bio_value, fc->corrupt_bio_byte,
-			(bio_data_dir(bio) == WRITE) ? 'w' : 'r', bio->bi_opf,
-			(unsigned long long)bio->bi_iter.bi_sector, bio_bytes);
+	bio_for_each_segment(bvec, bio, iter) {
+		if (bio_iter_len(bio, iter) > corrupt_bio_byte) {
+			char *segment = (page_address(bio_iter_page(bio, iter))
+					 + bio_iter_offset(bio, iter));
+			segment[corrupt_bio_byte] = fc->corrupt_bio_value;
+			DMDEBUG("Corrupting data bio=%p by writing %u to byte %u "
+				"(rw=%c bi_opf=%u bi_sector=%llu size=%u)\n",
+				bio, fc->corrupt_bio_value, fc->corrupt_bio_byte,
+				(bio_data_dir(bio) == WRITE) ? 'w' : 'r', bio->bi_opf,
+				(unsigned long long)bio->bi_iter.bi_sector, bio->bi_iter.bi_size);
+			break;
+		}
+		corrupt_bio_byte -= bio_iter_len(bio, iter);
 	}
 }
 
Signature · 1 line
2.20.1
author maintainer
#98 lore raw

PATCH 99/219 gfs2: take jdata unstuff into account in do_grow

In reply to Sasha Levin

From: Bob Peterson <[email protected]>

[ Upstream commit bc0205612bbd4dd4026d4ba6287f5643c37366ec ]

Before this patch, function do_grow would not reserve enough journal blocks in the transaction to unstuff jdata files while growing them. This patch adds the logic to add one more block if the file to grow is jdata.

---
 fs/gfs2/bmap.c | 2 ++
 1 file changed, 2 insertions(+)
fs/gfs2/bmap.c +2 -0
index 52feccedd7a44..096b479721395 100644
@@ -2122,6 +2122,8 @@ static int do_grow(struct inode *inode, u64 size)
 	}
 
 	error = gfs2_trans_begin(sdp, RES_DINODE + RES_STATFS + RES_RG_BIT +
+				 (unstuff &&
+				  gfs2_is_jdata(ip) ? RES_JDATA : 0) +
 				 (sdp->sd_args.ar_quota == GFS2_QUOTA_OFF ?
 				  0 : RES_QUOTA), 0);
 	if (error)
Signature · 1 line
2.20.1
author maintainer
#99 lore raw

PATCH 100/219 dm raid: fix false -EBUSY when handling check/repair message

In reply to Sasha Levin

From: Heinz Mauelshagen <[email protected]>

[ Upstream commit 74694bcbdf7e28a5ad548cdda9ac56d30be00d13 ]

Sending a check/repair message infrequently leads to -EBUSY instead of properly identifying an active resync. This occurs because raid_message() is testing recovery bits in a racy way.

Fix by calling decipher_sync_action() from raid_message() to properly identify the idle state of the RAID device.

---
 drivers/md/dm-raid.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/md/dm-raid.c +1 -2
index b78a8a4d061ca..416998523d455 100644
@@ -3690,8 +3690,7 @@ static int raid_message(struct dm_target *ti, unsigned int argc, char **argv,
 			set_bit(MD_RECOVERY_INTR, &mddev->recovery);
 			md_reap_sync_thread(mddev);
 		}
-	} else if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery) ||
-		   test_bit(MD_RECOVERY_NEEDED, &mddev->recovery))
+	} else if (decipher_sync_action(mddev, mddev->recovery) != st_idle)
 		return -EBUSY;
 	else if (!strcasecmp(argv[0], "resync"))
 		; /* MD_RECOVERY_NEEDED set below */
Signature · 1 line
2.20.1
author maintainer
#100 lore raw

PATCH 101/219 xfs: Align compat attrlist_by_handle with native implementation.

In reply to Sasha Levin

From: Nick Bowler <[email protected]>

[ Upstream commit c456d64449efe37da50832b63d91652a85ea1d20 ]

While inspecting the ioctl implementations, I noticed that the compat implementation of XFS_IOC_ATTRLIST_BY_HANDLE does not do exactly the same thing as the native implementation. Specifically, the "cursor" does not appear to be written out to userspace on the compat path, like it is on the native path.

This adjusts the compat implementation to copy out the cursor just like the native implementation does. The attrlist cursor does not require any special compat handling. This fixes xfstests xfs/269 on both IA-32 and x32 userspace, when running on an amd64 kernel.

Fixes 0facef7fb053b ("xfs: in _attrlist_by_handle, copy the cursor back to userspace")
---
 fs/xfs/xfs_ioctl32.c | 6 ++++++
 1 file changed, 6 insertions(+)
fs/xfs/xfs_ioctl32.c +6 -0
index fba115f4103ac..4c34efcbf7e80 100644
@@ -336,6 +336,7 @@ xfs_compat_attrlist_by_handle(
 {
 	int			error;
 	attrlist_cursor_kern_t	*cursor;
+	compat_xfs_fsop_attrlist_handlereq_t __user *p = arg;
 	compat_xfs_fsop_attrlist_handlereq_t al_hreq;
 	struct dentry		*dentry;
 	char			*kbuf;
@@ -370,6 +371,11 @@ xfs_compat_attrlist_by_handle(
 	if (error)
 		goto out_kfree;
 
+	if (copy_to_user(&p->pos, cursor, sizeof(attrlist_cursor_kern_t))) {
+		error = -EFAULT;
+		goto out_kfree;
+	}
+
 	if (copy_to_user(compat_ptr(al_hreq.buffer), kbuf, al_hreq.buflen))
 		error = -EFAULT;
 
Signature · 1 line
2.20.1
author maintainer
#101 lore raw

PATCH 102/219 xfs: Fix bulkstat compat ioctls on x32 userspace.

In reply to Sasha Levin

From: Nick Bowler <[email protected]>

[ Upstream commit 7ca860e3c1a74ad6bd8949364073ef1044cad758 ]

The bulkstat family of ioctls are problematic on x32, because there is a mixup of native 32-bit and 64-bit conventions. The xfs_fsop_bulkreq struct contains pointers and 32-bit integers so that matches the native 32-bit layout, and that means the ioctl implementation goes into the regular compat path on x32.

However, the 'ubuffer' member of that struct in turn refers to either struct xfs_inogrp or xfs_bstat (or an array of these). On x32, those structures match the native 64-bit layout. The compat implementation writes out the 32-bit version of these structures. This is not the expected format for x32 userspace, causing problems.

Fortunately the functions which actually output these xfs_inogrp and xfs_bstat structures have an easy way to select which output format is required, so we just need a little tweak to select the right format on x32.

---
 fs/xfs/xfs_ioctl32.c | 34 ++++++++++++++++++++++++++++++----
 1 file changed, 30 insertions(+), 4 deletions(-)
fs/xfs/xfs_ioctl32.c +30 -4
index 4c34efcbf7e80..b044f7d36782c 100644
@@ -241,6 +241,32 @@ xfs_compat_ioc_bulkstat(
 	int			done;
 	int			error;
 
+	/*
+	 * Output structure handling functions.  Depending on the command,
+	 * either the xfs_bstat and xfs_inogrp structures are written out
+	 * to userpace memory via bulkreq.ubuffer.  Normally the compat
+	 * functions and structure size are the correct ones to use ...
+	 */
+	inumbers_fmt_pf inumbers_func = xfs_inumbers_fmt_compat;
+	bulkstat_one_pf	bs_one_func = xfs_bulkstat_one_compat;
+	size_t bs_one_size = sizeof(struct compat_xfs_bstat);
+
+#ifdef CONFIG_X86_X32
+	if (in_x32_syscall()) {
+		/*
+		 * ... but on x32 the input xfs_fsop_bulkreq has pointers
+		 * which must be handled in the "compat" (32-bit) way, while
+		 * the xfs_bstat and xfs_inogrp structures follow native 64-
+		 * bit layout convention.  So adjust accordingly, otherwise
+		 * the data written out in compat layout will not match what
+		 * x32 userspace expects.
+		 */
+		inumbers_func = xfs_inumbers_fmt;
+		bs_one_func = xfs_bulkstat_one;
+		bs_one_size = sizeof(struct xfs_bstat);
+	}
+#endif
+
 	/* done = 1 if there are more stats to get and if bulkstat */
 	/* should be called again (unused here, but used in dmapi) */
 
@@ -272,15 +298,15 @@ xfs_compat_ioc_bulkstat(
 
 	if (cmd == XFS_IOC_FSINUMBERS_32) {
 		error = xfs_inumbers(mp, &inlast, &count,
-				bulkreq.ubuffer, xfs_inumbers_fmt_compat);
+				bulkreq.ubuffer, inumbers_func);
 	} else if (cmd == XFS_IOC_FSBULKSTAT_SINGLE_32) {
 		int res;
 
-		error = xfs_bulkstat_one_compat(mp, inlast, bulkreq.ubuffer,
-				sizeof(compat_xfs_bstat_t), NULL, &res);
+		error = bs_one_func(mp, inlast, bulkreq.ubuffer,
+				bs_one_size, NULL, &res);
 	} else if (cmd == XFS_IOC_FSBULKSTAT_32) {
 		error = xfs_bulkstat(mp, &inlast, &count,
-			xfs_bulkstat_one_compat, sizeof(compat_xfs_bstat_t),
+			bs_one_func, bs_one_size,
 			bulkreq.ubuffer, &done);
 	} else
 		error = -EINVAL;
Signature · 1 line
2.20.1
author maintainer
#102 lore raw

PATCH 103/219 IB/qib: Fix an error code in qib_sdma_verbs_send()

In reply to Sasha Levin

From: Dan Carpenter <[email protected]>

[ Upstream commit 5050ae5fa3d54c8e83e1e447cc7e3591110a7f57 ]

We accidentally return success on this error path.

Fixes f931551bafe1 ("IB/qib: Add new qib driver for QLogic PCIe InfiniBand adapters")
---
 drivers/infiniband/hw/qib/qib_sdma.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/infiniband/hw/qib/qib_sdma.c +3 -1
index d0723d4aef5c9..7424e88b0d918 100644
@@ -576,8 +576,10 @@ int qib_sdma_verbs_send(struct qib_pportdata *ppd,
 		dw = (len + 3) >> 2;
 		addr = dma_map_single(&ppd->dd->pcidev->dev, sge->vaddr,
 				      dw << 2, DMA_TO_DEVICE);
-		if (dma_mapping_error(&ppd->dd->pcidev->dev, addr))
+		if (dma_mapping_error(&ppd->dd->pcidev->dev, addr)) {
+			ret = -ENOMEM;
 			goto unmap;
+		}
 		sdmadesc[0] = 0;
 		make_sdma_desc(ppd, sdmadesc, (u64) addr, dw, dwoffset);
 		/* SDmaUseLargeBuf has to be set in every descriptor */
Signature · 1 line
2.20.1
author maintainer
#103 lore raw

PATCH 104/219 clocksource/drivers/fttmr010: Fix invalid interrupt register access

In reply to Sasha Levin

From: Tao Ren <[email protected]>

[ Upstream commit 86fe57fc47b17b3528fa5497fc57e158d846c4ea ]

TIMER_INTR_MASK register (Base Address of Timer + 0x38) is not designed for masking interrupts on ast2500 chips, and it's not even listed in ast2400 datasheet, so it's not safe to access TIMER_INTR_MASK on aspeed chips.

Similarly, TIMER_INTR_STATE register (Base Address of Timer + 0x34) is not interrupt status register on ast2400 and ast2500 chips. Although there is no side effect to reset the register in fttmr010_common_init(), it's just misleading to do so.

Besides, "count_down" is renamed to "is_aspeed" in "fttmr010" structure, and more comments are added so the code is more readble.

---
 drivers/clocksource/timer-fttmr010.c | 73 ++++++++++++++++------------
 1 file changed, 42 insertions(+), 31 deletions(-)
drivers/clocksource/timer-fttmr010.c +42 -31
index cf93f6419b514..fadff7915dd9c 100644
@@ -21,7 +21,7 @@
 #include <linux/delay.h>
 
 /*
- * Register definitions for the timers
+ * Register definitions common for all the timer variants.
  */
 #define TIMER1_COUNT		(0x00)
 #define TIMER1_LOAD		(0x04)
@@ -36,9 +36,10 @@
 #define TIMER3_MATCH1		(0x28)
 #define TIMER3_MATCH2		(0x2c)
 #define TIMER_CR		(0x30)
-#define TIMER_INTR_STATE	(0x34)
-#define TIMER_INTR_MASK		(0x38)
 
+/*
+ * Control register (TMC30) bit fields for fttmr010/gemini/moxart timers.
+ */
 #define TIMER_1_CR_ENABLE	BIT(0)
 #define TIMER_1_CR_CLOCK	BIT(1)
 #define TIMER_1_CR_INT		BIT(2)
@@ -53,8 +54,9 @@
 #define TIMER_3_CR_UPDOWN	BIT(11)
 
 /*
- * The Aspeed AST2400 moves bits around in the control register
- * and lacks bits for setting the timer to count upwards.
+ * Control register (TMC30) bit fields for aspeed ast2400/ast2500 timers.
+ * The aspeed timers move bits around in the control register and lacks
+ * bits for setting the timer to count upwards.
  */
 #define TIMER_1_CR_ASPEED_ENABLE	BIT(0)
 #define TIMER_1_CR_ASPEED_CLOCK		BIT(1)
@@ -66,6 +68,18 @@
 #define TIMER_3_CR_ASPEED_CLOCK		BIT(9)
 #define TIMER_3_CR_ASPEED_INT		BIT(10)
 
+/*
+ * Interrupt status/mask register definitions for fttmr010/gemini/moxart
+ * timers.
+ * The registers don't exist and they are not needed on aspeed timers
+ * because:
+ *   - aspeed timer overflow interrupt is controlled by bits in Control
+ *     Register (TMC30).
+ *   - aspeed timers always generate interrupt when either one of the
+ *     Match registers equals to Status register.
+ */
+#define TIMER_INTR_STATE	(0x34)
+#define TIMER_INTR_MASK		(0x38)
 #define TIMER_1_INT_MATCH1	BIT(0)
 #define TIMER_1_INT_MATCH2	BIT(1)
 #define TIMER_1_INT_OVERFLOW	BIT(2)
@@ -80,7 +94,7 @@
 struct fttmr010 {
 	void __iomem *base;
 	unsigned int tick_rate;
-	bool count_down;
+	bool is_aspeed;
 	u32 t1_enable_val;
 	struct clock_event_device clkevt;
 #ifdef CONFIG_ARM
@@ -130,7 +144,7 @@ static int fttmr010_timer_set_next_event(unsigned long cycles,
 	cr &= ~fttmr010->t1_enable_val;
 	writel(cr, fttmr010->base + TIMER_CR);
 
-	if (fttmr010->count_down) {
+	if (fttmr010->is_aspeed) {
 		/*
 		 * ASPEED Timer Controller will load TIMER1_LOAD register
 		 * into TIMER1_COUNT register when the timer is re-enabled.
@@ -175,16 +189,17 @@ static int fttmr010_timer_set_oneshot(struct clock_event_device *evt)
 
 	/* Setup counter start from 0 or ~0 */
 	writel(0, fttmr010->base + TIMER1_COUNT);
-	if (fttmr010->count_down)
+	if (fttmr010->is_aspeed) {
 		writel(~0, fttmr010->base + TIMER1_LOAD);
-	else
+	} else {
 		writel(0, fttmr010->base + TIMER1_LOAD);
 
-	/* Enable interrupt */
-	cr = readl(fttmr010->base + TIMER_INTR_MASK);
-	cr &= ~(TIMER_1_INT_OVERFLOW | TIMER_1_INT_MATCH2);
-	cr |= TIMER_1_INT_MATCH1;
-	writel(cr, fttmr010->base + TIMER_INTR_MASK);
+		/* Enable interrupt */
+		cr = readl(fttmr010->base + TIMER_INTR_MASK);
+		cr &= ~(TIMER_1_INT_OVERFLOW | TIMER_1_INT_MATCH2);
+		cr |= TIMER_1_INT_MATCH1;
+		writel(cr, fttmr010->base + TIMER_INTR_MASK);
+	}
 
 	return 0;
 }
@@ -201,9 +216,8 @@ static int fttmr010_timer_set_periodic(struct clock_event_device *evt)
 	writel(cr, fttmr010->base + TIMER_CR);
 
 	/* Setup timer to fire at 1/HZ intervals. */
-	if (fttmr010->count_down) {
+	if (fttmr010->is_aspeed) {
 		writel(period, fttmr010->base + TIMER1_LOAD);
-		writel(0, fttmr010->base + TIMER1_MATCH1);
 	} else {
 		cr = 0xffffffff - (period - 1);
 		writel(cr, fttmr010->base + TIMER1_COUNT);
@@ -281,23 +295,21 @@ static int __init fttmr010_common_init(struct device_node *np, bool is_aspeed)
 	}
 
 	/*
-	 * The Aspeed AST2400 moves bits around in the control register,
-	 * otherwise it works the same.
+	 * The Aspeed timers move bits around in the control register.
 	 */
 	if (is_aspeed) {
 		fttmr010->t1_enable_val = TIMER_1_CR_ASPEED_ENABLE |
 			TIMER_1_CR_ASPEED_INT;
-		/* Downward not available */
-		fttmr010->count_down = true;
+		fttmr010->is_aspeed = true;
 	} else {
 		fttmr010->t1_enable_val = TIMER_1_CR_ENABLE | TIMER_1_CR_INT;
-	}
 
-	/*
-	 * Reset the interrupt mask and status
-	 */
-	writel(TIMER_INT_ALL_MASK, fttmr010->base + TIMER_INTR_MASK);
-	writel(0, fttmr010->base + TIMER_INTR_STATE);
+		/*
+		 * Reset the interrupt mask and status
+		 */
+		writel(TIMER_INT_ALL_MASK, fttmr010->base + TIMER_INTR_MASK);
+		writel(0, fttmr010->base + TIMER_INTR_STATE);
+	}
 
 	/*
 	 * Enable timer 1 count up, timer 2 count up, except on Aspeed,
@@ -306,9 +318,8 @@ static int __init fttmr010_common_init(struct device_node *np, bool is_aspeed)
 	if (is_aspeed)
 		val = TIMER_2_CR_ASPEED_ENABLE;
 	else {
-		val = TIMER_2_CR_ENABLE;
-		if (!fttmr010->count_down)
-			val |= TIMER_1_CR_UPDOWN | TIMER_2_CR_UPDOWN;
+		val = TIMER_2_CR_ENABLE | TIMER_1_CR_UPDOWN |
+			TIMER_2_CR_UPDOWN;
 	}
 	writel(val, fttmr010->base + TIMER_CR);
 
@@ -321,7 +332,7 @@ static int __init fttmr010_common_init(struct device_node *np, bool is_aspeed)
 	writel(0, fttmr010->base + TIMER2_MATCH1);
 	writel(0, fttmr010->base + TIMER2_MATCH2);
 
-	if (fttmr010->count_down) {
+	if (fttmr010->is_aspeed) {
 		writel(~0, fttmr010->base + TIMER2_LOAD);
 		clocksource_mmio_init(fttmr010->base + TIMER2_COUNT,
 				      "FTTMR010-TIMER2",
@@ -371,7 +382,7 @@ static int __init fttmr010_common_init(struct device_node *np, bool is_aspeed)
 
 #ifdef CONFIG_ARM
 	/* Also use this timer for delays */
-	if (fttmr010->count_down)
+	if (fttmr010->is_aspeed)
 		fttmr010->delay_timer.read_current_timer =
 			fttmr010_read_current_timer_down;
 	else
Signature · 1 line
2.20.1
author maintainer
#104 lore raw

PATCH 105/219 vxlan: Fix error path in __vxlan_dev_create()

In reply to Sasha Levin

From: Petr Machata <[email protected]>

[ Upstream commit 6db9246871394b3a136cd52001a0763676563840 ]

When a failure occurs in rtnl_configure_link(), the current code calls unregister_netdevice() to roll back the earlier call to register_netdevice(), and jumps to errout, which calls vxlan_fdb_destroy().

However unregister_netdevice() calls transitively ndo_uninit, which is vxlan_uninit(), and that already takes care of deleting the default FDB entry by calling vxlan_fdb_delete_default(). Since the entry added earlier in __vxlan_dev_create() is exactly the default entry, the cleanup code in the errout block always leads to double free and thus a panic.

Besides, since vxlan_fdb_delete_default() always destroys the FDB entry with notification enabled, the deletion of the default entry is notified even before the addition was notified.

Instead, move the unregister_netdevice() call after the manual destroy, which solves both problems.

Fixes 0241b836732f ("vxlan: fix default fdb entry netlink notify ordering during netdev create")
---
 drivers/net/vxlan.c | 13 ++++++++++---
 1 file changed, 10 insertions(+), 3 deletions(-)
drivers/net/vxlan.c +10 -3
index 2a536f84d5f69..d8a56df3933f0 100644
@@ -3213,6 +3213,7 @@ static int __vxlan_dev_create(struct net *net, struct net_device *dev,
 	struct vxlan_net *vn = net_generic(net, vxlan_net_id);
 	struct vxlan_dev *vxlan = netdev_priv(dev);
 	struct vxlan_fdb *f = NULL;
+	bool unregister = false;
 	int err;
 
 	err = vxlan_dev_configure(net, dev, conf, false, extack);
@@ -3238,12 +3239,11 @@ static int __vxlan_dev_create(struct net *net, struct net_device *dev,
 	err = register_netdevice(dev);
 	if (err)
 		goto errout;
+	unregister = true;
 
 	err = rtnl_configure_link(dev, NULL);
-	if (err) {
-		unregister_netdevice(dev);
+	if (err)
 		goto errout;
-	}
 
 	/* notify default fdb entry */
 	if (f)
@@ -3251,9 +3251,16 @@ static int __vxlan_dev_create(struct net *net, struct net_device *dev,
 
 	list_add(&vxlan->next, &vn->vxlan_list);
 	return 0;
+
 errout:
+	/* unregister_netdevice() destroys the default FDB entry with deletion
+	 * notification. But the addition notification was not sent yet, so
+	 * destroy the entry by hand here.
+	 */
 	if (f)
 		vxlan_fdb_destroy(vxlan, f, false);
+	if (unregister)
+		unregister_netdevice(dev);
 	return err;
 }
 
Signature · 1 line
2.20.1
author maintainer
#105 lore raw

PATCH 106/219 powerpc/book3s/32: fix number of bats in p/v_block_mapped()

In reply to Sasha Levin

From: Christophe Leroy <[email protected]>

[ Upstream commit e93ba1b7eb5b188c749052df7af1c90821c5f320 ]

This patch fixes the loop in p_block_mapped() and v_block_mapped() to scan the entire bat_addrs[] array.

---
 arch/powerpc/mm/ppc_mmu_32.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/powerpc/mm/ppc_mmu_32.c +2 -2
index bea6c544e38f9..06783270a1242 100644
@@ -52,7 +52,7 @@ struct batrange {		/* stores address ranges mapped by BATs */
 phys_addr_t v_block_mapped(unsigned long va)
 {
 	int b;
-	for (b = 0; b < 4; ++b)
+	for (b = 0; b < ARRAY_SIZE(bat_addrs); ++b)
 		if (va >= bat_addrs[b].start && va < bat_addrs[b].limit)
 			return bat_addrs[b].phys + (va - bat_addrs[b].start);
 	return 0;
@@ -64,7 +64,7 @@ phys_addr_t v_block_mapped(unsigned long va)
 unsigned long p_block_mapped(phys_addr_t pa)
 {
 	int b;
-	for (b = 0; b < 4; ++b)
+	for (b = 0; b < ARRAY_SIZE(bat_addrs); ++b)
 		if (pa >= bat_addrs[b].phys
 	    	    && pa < (bat_addrs[b].limit-bat_addrs[b].start)
 		              +bat_addrs[b].phys)
Signature · 1 line
2.20.1
author maintainer
#106 lore raw

PATCH 107/219 powerpc/xmon: fix dump_segments()

In reply to Sasha Levin

From: Christophe Leroy <[email protected]>

[ Upstream commit 32c8c4c621897199e690760c2d57054f8b84b6e6 ]

mfsrin() takes segment num from bits 31-28 (IBM bits 0-3).

---
 arch/powerpc/xmon/xmon.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/powerpc/xmon/xmon.c +1 -1
index bb5db7bfd8539..f0fa22e7d36c7 100644
@@ -3493,7 +3493,7 @@ void dump_segments(void)
 
 	printf("sr0-15 =");
 	for (i = 0; i < 16; ++i)
-		printf(" %x", mfsrin(i));
+		printf(" %x", mfsrin(i << 28));
 	printf("\n");
 }
 #endif
Signature · 1 line
2.20.1
author maintainer
#107 lore raw

PATCH 108/219 drivers/regulator: fix a missing check of return value

In reply to Sasha Levin

From: Kangjie Lu <[email protected]>

[ Upstream commit 966e927bf8cc6a44f8b72582a1d6d3ffc73b12ad ]

If palmas_smps_read() fails, we should not use the read data in "reg" which may contain random value. The fix inserts a check for the return value of palmas_smps_read(): If it fails, we return the error code upstream and stop using "reg".

---
 drivers/regulator/palmas-regulator.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/regulator/palmas-regulator.c +4 -1
index bb5ab7d78895b..c2cc392a27d40 100644
@@ -443,13 +443,16 @@ static int palmas_ldo_write(struct palmas *palmas, unsigned int reg,
 static int palmas_set_mode_smps(struct regulator_dev *dev, unsigned int mode)
 {
 	int id = rdev_get_id(dev);
+	int ret;
 	struct palmas_pmic *pmic = rdev_get_drvdata(dev);
 	struct palmas_pmic_driver_data *ddata = pmic->palmas->pmic_ddata;
 	struct palmas_regs_info *rinfo = &ddata->palmas_regs_info[id];
 	unsigned int reg;
 	bool rail_enable = true;
 
-	palmas_smps_read(pmic->palmas, rinfo->ctrl_addr, &reg);
+	ret = palmas_smps_read(pmic->palmas, rinfo->ctrl_addr, &reg);
+	if (ret)
+		return ret;
 
 	reg &= ~PALMAS_SMPS12_CTRL_MODE_ACTIVE_MASK;
 
Signature · 1 line
2.20.1
author maintainer
#108 lore raw

PATCH 109/219 Bluetooth: hci_bcm: Handle specific unknown packets after firmware loading

In reply to Sasha Levin

From: Jonathan Bakker <[email protected]>

[ Upstream commit 22bba80500fdf624a7cfbb65fdfa97a038ae224d ]

The Broadcom controller on aries S5PV210 boards sends out a couple of unknown packets after the firmware is loaded. This will cause logging of errors such as:
Bluetooth: hci0: Frame reassembly failed (-84)

This is probably also the case with other boards, as there are related Android userspace patches for custom ROMs such as
https://review.lineageos.org/#/c/LineageOS/android_system_bt/+/142721/
Since this appears to be intended behaviour, treated them as diagnostic packets.

Note that this is another variant of commit 01d5e44ace8a ("Bluetooth: hci_bcm: Handle empty packet after firmware loading")

---
 drivers/bluetooth/hci_bcm.c | 22 ++++++++++++++++++++++
 1 file changed, 22 insertions(+)
drivers/bluetooth/hci_bcm.c +22 -0
index aa6b7ed9fdf12..59e5fc5eec8f8 100644
@@ -51,6 +51,12 @@
 #define BCM_LM_DIAG_PKT 0x07
 #define BCM_LM_DIAG_SIZE 63
 
+#define BCM_TYPE49_PKT 0x31
+#define BCM_TYPE49_SIZE 0
+
+#define BCM_TYPE52_PKT 0x34
+#define BCM_TYPE52_SIZE 0
+
 #define BCM_AUTOSUSPEND_DELAY	5000 /* default autosleep delay */
 
 /**
@@ -564,12 +570,28 @@ static int bcm_setup(struct hci_uart *hu)
 	.lsize = 0, \
 	.maxlen = BCM_NULL_SIZE
 
+#define BCM_RECV_TYPE49 \
+	.type = BCM_TYPE49_PKT, \
+	.hlen = BCM_TYPE49_SIZE, \
+	.loff = 0, \
+	.lsize = 0, \
+	.maxlen = BCM_TYPE49_SIZE
+
+#define BCM_RECV_TYPE52 \
+	.type = BCM_TYPE52_PKT, \
+	.hlen = BCM_TYPE52_SIZE, \
+	.loff = 0, \
+	.lsize = 0, \
+	.maxlen = BCM_TYPE52_SIZE
+
 static const struct h4_recv_pkt bcm_recv_pkts[] = {
 	{ H4_RECV_ACL,      .recv = hci_recv_frame },
 	{ H4_RECV_SCO,      .recv = hci_recv_frame },
 	{ H4_RECV_EVENT,    .recv = hci_recv_frame },
 	{ BCM_RECV_LM_DIAG, .recv = hci_recv_diag  },
 	{ BCM_RECV_NULL,    .recv = hci_recv_diag  },
+	{ BCM_RECV_TYPE49,  .recv = hci_recv_diag  },
+	{ BCM_RECV_TYPE52,  .recv = hci_recv_diag  },
 };
 
 static int bcm_recv(struct hci_uart *hu, const void *data, int count)
Signature · 1 line
2.20.1
author maintainer
#109 lore raw

PATCH 110/219 serial: max310x: Fix tx_empty() callback

In reply to Sasha Levin

From: Alexander Shiyan <[email protected]>

[ Upstream commit a8da3c7873ea57acb8f9cea58c0af477522965aa ]

Function max310x_tx_empty() accesses the IRQSTS register, which is cleared by IC when reading, so if there is an interrupt status, we will lose it. This patch implement the transmitter check only by the current FIFO level.

---
 drivers/tty/serial/max310x.c | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)
drivers/tty/serial/max310x.c +2 -5
index bd3e6cf81af5c..0c35c3c5e3734 100644
@@ -844,12 +844,9 @@ static void max310x_wq_proc(struct work_struct *ws)
 
 static unsigned int max310x_tx_empty(struct uart_port *port)
 {
-	unsigned int lvl, sts;
+	u8 lvl = max310x_port_read(port, MAX310X_TXFIFOLVL_REG);
 
-	lvl = max310x_port_read(port, MAX310X_TXFIFOLVL_REG);
-	sts = max310x_port_read(port, MAX310X_IRQSTS_REG);
-
-	return ((sts & MAX310X_IRQ_TXEMPTY_BIT) && !lvl) ? TIOCSER_TEMT : 0;
+	return lvl ? 0 : TIOCSER_TEMT;
 }
 
 static unsigned int max310x_get_mctrl(struct uart_port *port)
Signature · 1 line
2.20.1
author maintainer
#110 lore raw

PATCH 111/219 openrisc: Fix broken paths to arch/or32

In reply to Sasha Levin

From: Geert Uytterhoeven <[email protected]>

[ Upstream commit 57ce8ba0fd3a95bf29ed741df1c52bd591bf43ff ]

OpenRISC was mainlined as "openrisc", not "or32". vmlinux.lds is generated from vmlinux.lds.S.

---
 arch/openrisc/kernel/entry.S | 2 +-
 arch/openrisc/kernel/head.S  | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
arch/openrisc/kernel/entry.S +1 -1
index 0c826ad6e994c..ee6159d2ed22e 100644
@@ -240,7 +240,7 @@ handler:							;\
  *	 occured. in fact they never do. if you need them use
  *	 values saved on stack (for SPR_EPC, SPR_ESR) or content
  *       of r4 (for SPR_EEAR). for details look at EXCEPTION_HANDLE()
- *       in 'arch/or32/kernel/head.S'
+ *       in 'arch/openrisc/kernel/head.S'
  */
 
 /* =====================================================[ exceptions] === */
arch/openrisc/kernel/head.S +1 -1
index 9fc6b60140f00..31ed257ff0618 100644
@@ -1728,7 +1728,7 @@ _string_nl:
 
 /*
  * .data section should be page aligned
- *	(look into arch/or32/kernel/vmlinux.lds)
+ *	(look into arch/openrisc/kernel/vmlinux.lds.S)
  */
 	.section .data,"aw"
 	.align	8192
Signature · 1 line
2.20.1
author maintainer
#111 lore raw

PATCH 112/219 RDMA/srp: Propagate ib_post_send() failures to the SCSI mid-layer

In reply to Sasha Levin

From: Bart Van Assche <[email protected]>

[ Upstream commit 2ee00f6a98c36f7e4ba07cc33f24cc5a69060cc9 ]

This patch avoids that the SCSI mid-layer keeps retrying forever if ib_post_send() fails. This was discovered while testing immediate data support and passing a too large num_sge value to ib_post_send().

---
 drivers/infiniband/ulp/srp/ib_srp.c | 1 +
 1 file changed, 1 insertion(+)
drivers/infiniband/ulp/srp/ib_srp.c +1 -0
index bc6a44a16445c..03ee53adaacd2 100644
@@ -2357,6 +2357,7 @@ static int srp_queuecommand(struct Scsi_Host *shost, struct scsi_cmnd *scmnd)
 
 	if (srp_post_send(ch, iu, len)) {
 		shost_printk(KERN_ERR, target->scsi_host, PFX "Send failed\n");
+		scmnd->result = DID_ERROR << 16;
 		goto err_unmap;
 	}
 
Signature · 1 line
2.20.1
author maintainer
#112 lore raw

PATCH 113/219 scsi: qla2xxx: deadlock by configfs_depend_item

In reply to Sasha Levin

From: Anatoliy Glagolev <[email protected]>

[ Upstream commit 17b18eaa6f59044a5172db7d07149e31ede0f920 ]

The intent of invoking configfs_depend_item in commit 7474f52a82d51 ("tcm_qla2xxx: Perform configfs depend/undepend for base_tpg") was to prevent a physical Fibre Channel port removal when virtual (NPIV) ports announced through that physical port are active. The change does not work as expected: it makes enabled physical port dependent on target configfs subsystem (the port's parent), something the configfs guarantees anyway.

Besides, scheduling work in a worker thread and waiting for the work's completion is not really a valid workaround for the requirement not to call configfs_depend_item from a configfs callback: the call occasionally deadlocks.

Thus, removing configfs_depend_item calls does not break anything and fixes the deadlock problem.

---
 drivers/scsi/qla2xxx/tcm_qla2xxx.c | 48 +++++-------------------------
 drivers/scsi/qla2xxx/tcm_qla2xxx.h |  3 --
 2 files changed, 8 insertions(+), 43 deletions(-)
drivers/scsi/qla2xxx/tcm_qla2xxx.c +8 -40
index b8c1a739dfbd1..654e1af7f542c 100644
@@ -926,38 +926,14 @@ static ssize_t tcm_qla2xxx_tpg_enable_show(struct config_item *item,
 			atomic_read(&tpg->lport_tpg_enabled));
 }
 
-static void tcm_qla2xxx_depend_tpg(struct work_struct *work)
-{
-	struct tcm_qla2xxx_tpg *base_tpg = container_of(work,
-				struct tcm_qla2xxx_tpg, tpg_base_work);
-	struct se_portal_group *se_tpg = &base_tpg->se_tpg;
-	struct scsi_qla_host *base_vha = base_tpg->lport->qla_vha;
-
-	if (!target_depend_item(&se_tpg->tpg_group.cg_item)) {
-		atomic_set(&base_tpg->lport_tpg_enabled, 1);
-		qlt_enable_vha(base_vha);
-	}
-	complete(&base_tpg->tpg_base_comp);
-}
-
-static void tcm_qla2xxx_undepend_tpg(struct work_struct *work)
-{
-	struct tcm_qla2xxx_tpg *base_tpg = container_of(work,
-				struct tcm_qla2xxx_tpg, tpg_base_work);
-	struct se_portal_group *se_tpg = &base_tpg->se_tpg;
-	struct scsi_qla_host *base_vha = base_tpg->lport->qla_vha;
-
-	if (!qlt_stop_phase1(base_vha->vha_tgt.qla_tgt)) {
-		atomic_set(&base_tpg->lport_tpg_enabled, 0);
-		target_undepend_item(&se_tpg->tpg_group.cg_item);
-	}
-	complete(&base_tpg->tpg_base_comp);
-}
-
 static ssize_t tcm_qla2xxx_tpg_enable_store(struct config_item *item,
 		const char *page, size_t count)
 {
 	struct se_portal_group *se_tpg = to_tpg(item);
+	struct se_wwn *se_wwn = se_tpg->se_tpg_wwn;
+	struct tcm_qla2xxx_lport *lport = container_of(se_wwn,
+			struct tcm_qla2xxx_lport, lport_wwn);
+	struct scsi_qla_host *vha = lport->qla_vha;
 	struct tcm_qla2xxx_tpg *tpg = container_of(se_tpg,
 			struct tcm_qla2xxx_tpg, se_tpg);
 	unsigned long op;
@@ -976,24 +952,16 @@ static ssize_t tcm_qla2xxx_tpg_enable_store(struct config_item *item,
 		if (atomic_read(&tpg->lport_tpg_enabled))
 			return -EEXIST;
 
-		INIT_WORK(&tpg->tpg_base_work, tcm_qla2xxx_depend_tpg);
+		atomic_set(&tpg->lport_tpg_enabled, 1);
+		qlt_enable_vha(vha);
 	} else {
 		if (!atomic_read(&tpg->lport_tpg_enabled))
 			return count;
 
-		INIT_WORK(&tpg->tpg_base_work, tcm_qla2xxx_undepend_tpg);
+		atomic_set(&tpg->lport_tpg_enabled, 0);
+		qlt_stop_phase1(vha->vha_tgt.qla_tgt);
 	}
-	init_completion(&tpg->tpg_base_comp);
-	schedule_work(&tpg->tpg_base_work);
-	wait_for_completion(&tpg->tpg_base_comp);
 
-	if (op) {
-		if (!atomic_read(&tpg->lport_tpg_enabled))
-			return -ENODEV;
-	} else {
-		if (atomic_read(&tpg->lport_tpg_enabled))
-			return -EPERM;
-	}
 	return count;
 }
 
drivers/scsi/qla2xxx/tcm_qla2xxx.h +0 -3
index 7550ba2831c36..147cf6c903666 100644
@@ -48,9 +48,6 @@ struct tcm_qla2xxx_tpg {
 	struct tcm_qla2xxx_tpg_attrib tpg_attrib;
 	/* Returned by tcm_qla2xxx_make_tpg() */
 	struct se_portal_group se_tpg;
-	/* Items for dealing with configfs_depend_item */
-	struct completion tpg_base_comp;
-	struct work_struct tpg_base_work;
 };
 
 struct tcm_qla2xxx_fc_loopid {
Signature · 1 line
2.20.1
author maintainer
#113 lore raw

PATCH 114/219 scsi: csiostor: fix incorrect dma device in case of vport

In reply to Sasha Levin

From: Varun Prakash <[email protected]>

[ Upstream commit 9934613edcb40b92a216122876cd3b7e76d08390 ]

In case of ->vport_create() call scsi_add_host_with_dma() instead of scsi_add_host() to pass correct dma device.

---
 drivers/scsi/csiostor/csio_init.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/scsi/csiostor/csio_init.c +1 -1
index ed2dae657964b..1793981337dd9 100644
@@ -649,7 +649,7 @@ csio_shost_init(struct csio_hw *hw, struct device *dev,
 	if (csio_lnode_init(ln, hw, pln))
 		goto err_shost_put;
 
-	if (scsi_add_host(shost, dev))
+	if (scsi_add_host_with_dma(shost, dev, &hw->pdev->dev))
 		goto err_lnode_exit;
 
 	return ln;
Signature · 1 line
2.20.1
author maintainer
#114 lore raw

PATCH 115/219 brcmfmac: Fix access point mode

In reply to Sasha Levin

From: Stefan Wahren <[email protected]>

[ Upstream commit 861cb5eb467f5e38dce1aabe4e8db379255bd89b ]

Since commit 1204aa17f3b4 ("brcmfmac: set WIPHY_FLAG_HAVE_AP_SME flag") the Raspberry Pi 3 A+ (BCM43455) isn't able to operate in AP mode with hostapd (device_ap_sme=1 use_monitor=0):

brcmfmac: brcmf_cfg80211_stop_ap: setting AP mode failed -52

So add the missing mgmt_stypes for AP mode to fix this.

Fixes 1204aa17f3b4 ("brcmfmac: set WIPHY_FLAG_HAVE_AP_SME flag")
---
 .../wireless/broadcom/brcm80211/brcmfmac/cfg80211.c    | 10 ++++++++++
 1 file changed, 10 insertions(+)
drivers/net/wireless/broadcom/brcm80211/brcmfmac/cfg80211.c +10 -0
index c7c520f327f2b..bbdc6000afb9b 100644
@@ -6314,6 +6314,16 @@ brcmf_txrx_stypes[NUM_NL80211_IFTYPES] = {
 		.tx = 0xffff,
 		.rx = BIT(IEEE80211_STYPE_ACTION >> 4) |
 		      BIT(IEEE80211_STYPE_PROBE_REQ >> 4)
+	},
+	[NL80211_IFTYPE_AP] = {
+		.tx = 0xffff,
+		.rx = BIT(IEEE80211_STYPE_ASSOC_REQ >> 4) |
+		      BIT(IEEE80211_STYPE_REASSOC_REQ >> 4) |
+		      BIT(IEEE80211_STYPE_PROBE_REQ >> 4) |
+		      BIT(IEEE80211_STYPE_DISASSOC >> 4) |
+		      BIT(IEEE80211_STYPE_AUTH >> 4) |
+		      BIT(IEEE80211_STYPE_DEAUTH >> 4) |
+		      BIT(IEEE80211_STYPE_ACTION >> 4)
 	}
 };
 
Signature · 1 line
2.20.1
author maintainer
#115 lore raw

PATCH 116/219 ath6kl: Only use match sets when firmware supports it

In reply to Sasha Levin

From: Kyle Roeschley <[email protected]>

[ Upstream commit fb376a495fbdb886f38cfaf5a3805401b9e46f13 ]

Commit dd45b7598f1c ("ath6kl: Include match ssid list in scheduled scan") merged the probed and matched SSID lists before sending them to the firmware. In the process, it assumed match set support is always available in ath6kl_set_probed_ssids, which breaks scans for hidden SSIDs. Now, check that the firmware supports matching SSIDs in scheduled scans before setting MATCH_SSID_FLAG.

Fixes dd45b7598f1c ("ath6kl: Include match ssid list in scheduled scan")
---
 drivers/net/wireless/ath/ath6kl/cfg80211.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/ath/ath6kl/cfg80211.c +1 -1
index e121187f371ff..6c98d7903ffb6 100644
@@ -939,7 +939,7 @@ static int ath6kl_set_probed_ssids(struct ath6kl *ar,
 		else
 			ssid_list[i].flag = ANY_SSID_FLAG;
 
-		if (n_match_ssid == 0)
+		if (ar->wiphy->max_match_sets != 0 && n_match_ssid == 0)
 			ssid_list[i].flag |= MATCH_SSID_FLAG;
 	}
 
Signature · 1 line
2.20.1
author maintainer
#116 lore raw

PATCH 117/219 ath6kl: Fix off by one error in scan completion

In reply to Sasha Levin

From: Kyle Roeschley <[email protected]>

[ Upstream commit 5803c12816c43bd09e5f4247dd9313c2d9a2c41b ]

When ath6kl was reworked to share code between regular and scheduled scans in commit 3b8ffc6a22ba ("ath6kl: Configure probed SSID list consistently"), probed SSID entry changed from 1-index to 0-indexed. However, ath6kl_cfg80211_scan_complete_event() was missed in that change. Fix its indexing so that we correctly clear out the probed SSID list.

---
 drivers/net/wireless/ath/ath6kl/cfg80211.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/ath/ath6kl/cfg80211.c +1 -1
index 6c98d7903ffb6..d7c626d9594e1 100644
@@ -1093,7 +1093,7 @@ void ath6kl_cfg80211_scan_complete_event(struct ath6kl_vif *vif, bool aborted)
 	if (vif->scan_req->n_ssids && vif->scan_req->ssids[0].ssid_len) {
 		for (i = 0; i < vif->scan_req->n_ssids; i++) {
 			ath6kl_wmi_probedssid_cmd(ar->wmi, vif->fw_vif_idx,
-						  i + 1, DISABLE_SSID_FLAG,
+						  i, DISABLE_SSID_FLAG,
 						  0, NULL);
 		}
 	}
Signature · 1 line
2.20.1
author maintainer
#117 lore raw

PATCH 118/219 powerpc/perf: Fix unit_sel/cache_sel checks

In reply to Sasha Levin

From: Madhavan Srinivasan <[email protected]>

[ Upstream commit 2d46d4877b1afd14059393a48bdb8ce27955174c ]

Raw event code has couple of fields "unit" and "cache" in it, to capture the "unit" to monitor for a given pmcxsel and cache reload qualifier to program in MMCR1.

isa207_get_constraint() refers "unit" field to update the MMCRC (L2/L3) Event bus control fields with "cache" bits of the raw event code. These are power8 specific and not supported by PowerISA v3.0 pmu. So wrap the checks to be power8 specific. Also, "cache" bit field is referred to update MMCR1[16:17] and this check can be power8 specific.

Fixes 7ffd948fae4cd ('powerpc/perf: factor out power8 pmu functions')
---
 arch/powerpc/perf/isa207-common.c | 25 ++++++++++++++++++-------
 arch/powerpc/perf/isa207-common.h |  4 ++--
 2 files changed, 20 insertions(+), 9 deletions(-)
arch/powerpc/perf/isa207-common.c +18 -7
index 6a2f65d3d088c..053b8e9aa9e75 100644
@@ -148,6 +148,14 @@ static bool is_thresh_cmp_valid(u64 event)
 	return true;
 }
 
+static unsigned int dc_ic_rld_quad_l1_sel(u64 event)
+{
+	unsigned int cache;
+
+	cache = (event >> EVENT_CACHE_SEL_SHIFT) & MMCR1_DC_IC_QUAL_MASK;
+	return cache;
+}
+
 static inline u64 isa207_find_source(u64 idx, u32 sub_idx)
 {
 	u64 ret = PERF_MEM_NA;
@@ -288,10 +296,10 @@ int isa207_get_constraint(u64 event, unsigned long *maskp, unsigned long *valp)
 		 * have a cache selector of zero. The bank selector (bit 3) is
 		 * irrelevant, as long as the rest of the value is 0.
 		 */
-		if (cache & 0x7)
+		if (!cpu_has_feature(CPU_FTR_ARCH_300) && (cache & 0x7))
 			return -1;
 
-	} else if (event & EVENT_IS_L1) {
+	} else if (cpu_has_feature(CPU_FTR_ARCH_300) || (event & EVENT_IS_L1)) {
 		mask  |= CNST_L1_QUAL_MASK;
 		value |= CNST_L1_QUAL_VAL(cache);
 	}
@@ -394,11 +402,14 @@ int isa207_compute_mmcr(u64 event[], int n_ev,
 		/* In continuous sampling mode, update SDAR on TLB miss */
 		mmcra_sdar_mode(event[i], &mmcra);
 
-		if (event[i] & EVENT_IS_L1) {
-			cache = event[i] >> EVENT_CACHE_SEL_SHIFT;
-			mmcr1 |= (cache & 1) << MMCR1_IC_QUAL_SHIFT;
-			cache >>= 1;
-			mmcr1 |= (cache & 1) << MMCR1_DC_QUAL_SHIFT;
+		if (cpu_has_feature(CPU_FTR_ARCH_300)) {
+			cache = dc_ic_rld_quad_l1_sel(event[i]);
+			mmcr1 |= (cache) << MMCR1_DC_IC_QUAL_SHIFT;
+		} else {
+			if (event[i] & EVENT_IS_L1) {
+				cache = dc_ic_rld_quad_l1_sel(event[i]);
+				mmcr1 |= (cache) << MMCR1_DC_IC_QUAL_SHIFT;
+			}
 		}
 
 		if (is_event_marked(event[i])) {
arch/powerpc/perf/isa207-common.h +2 -2
index 0028f4b9490db..e5a621699a6d8 100644
@@ -163,8 +163,8 @@
 #define MMCR1_COMBINE_SHIFT(pmc)	(35 - ((pmc) - 1))
 #define MMCR1_PMCSEL_SHIFT(pmc)		(24 - (((pmc) - 1)) * 8)
 #define MMCR1_FAB_SHIFT			36
-#define MMCR1_DC_QUAL_SHIFT		47
-#define MMCR1_IC_QUAL_SHIFT		46
+#define MMCR1_DC_IC_QUAL_MASK		0x3
+#define MMCR1_DC_IC_QUAL_SHIFT		46
 
 /* MMCR1 Combine bits macro for power9 */
 #define p9_MMCR1_COMBINE_SHIFT(pmc)	(38 - ((pmc - 1) * 2))
Signature · 1 line
2.20.1
author maintainer
#118 lore raw

PATCH 119/219 powerpc/32: Avoid unsupported flags with clang

In reply to Sasha Levin

From: Joel Stanley <[email protected]>

[ Upstream commit 72e7bcc2cdf82bf03caaa5e6c9b0134c2fc2ee7d ]

When building for ppc32 with clang these flags are unsupported:

-ffixed-r2 and -mmultiple

llvm's lib/Target/PowerPC/PPCRegisterInfo.cpp marks r2 as reserved on when building for SVR4ABI and !ppc64:

// The SVR4 ABI reserves r2 and r13

  if (Subtarget.isSVR4ABI()) {
    // We only reserve r2 if we need to use the TOC pointer. If we have no
    // explicit uses of the TOC pointer (meaning we're a leaf function with
    // no constant-pool loads, etc.) and we have no potential uses inside an
    // inline asm block, then we can treat r2 has an ordinary callee-saved
    // register.
    const PPCFunctionInfo *FuncInfo = MF.getInfo<PPCFunctionInfo>();
    if (!TM.isPPC64() || FuncInfo->usesTOCBasePtr() || MF.hasInlineAsm())
      markSuperRegs(Reserved, PPC::R2);  // System-reserved register
    markSuperRegs(Reserved, PPC::R13); // Small Data Area pointer register

}

This means we can safely omit -ffixed-r2 when building for 32-bit targets.

The -mmultiple/-mno-multiple flags are not supported by clang, so platforms that might support multiple miss out on using multiple word instructions.

We wrap these flags in cc-option so that when Clang gains support the kernel will be able use these flags.

Clang 8 can then build a ppc44x_defconfig which boots in Qemu:

make CC=clang-8 ARCH=powerpc CROSS_COMPILE=powerpc-linux-gnu- ppc44x_defconfig ./scripts/config -e CONFIG_DEVTMPFS -d DEVTMPFS_MOUNT
make CC=clang-8 ARCH=powerpc CROSS_COMPILE=powerpc-linux-gnu-

  qemu-system-ppc -M bamboo \
   -kernel arch/powerpc/boot/zImage \
   -dtb arch/powerpc/boot/dts/bamboo.dtb \
   -initrd ~/ppc32-440-rootfs.cpio \
   -nographic -serial stdio -monitor pty -append "console=ttyS0"
---
 arch/powerpc/Makefile | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)
arch/powerpc/Makefile +8 -1
index c4c03992ee828..dfcb698ec8f3b 100644
@@ -145,7 +145,14 @@ endif
 CFLAGS-$(CONFIG_PPC64)	+= $(call cc-option,-mcmodel=medium,$(call cc-option,-mminimal-toc))
 CFLAGS-$(CONFIG_PPC64)	+= $(call cc-option,-mno-pointers-to-nested-functions)
 
-CFLAGS-$(CONFIG_PPC32)	:= -ffixed-r2 $(MULTIPLEWORD)
+# Clang unconditionally reserves r2 on ppc32 and does not support the flag
+# https://bugs.llvm.org/show_bug.cgi?id=39555
+CFLAGS-$(CONFIG_PPC32)	:= $(call cc-option, -ffixed-r2)
+
+# Clang doesn't support -mmultiple / -mno-multiple
+# https://bugs.llvm.org/show_bug.cgi?id=39556
+CFLAGS-$(CONFIG_PPC32)	+= $(call cc-option, $(MULTIPLEWORD))
+
 CFLAGS-$(CONFIG_PPC32)	+= $(call cc-option,-mno-readonly-in-sdata)
 
 ifdef CONFIG_PPC_BOOK3S_64
Signature · 1 line
2.20.1
author maintainer
#119 lore raw

PATCH 120/219 powerpc/prom: fix early DEBUG messages

In reply to Sasha Levin

From: Christophe Leroy <[email protected]>

[ Upstream commit b18f0ae92b0a1db565c3e505fa87b6971ad3b641 ]

This patch fixes early DEBUG messages in prom.c:
- Use %px instead of %p to see the addresses
- Cast memblock_phys_mem_size() with (unsigned long long) to avoid build failure when phys_addr_t is not 64 bits.

---
 arch/powerpc/kernel/prom.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
arch/powerpc/kernel/prom.c +3 -3
index c4d7078e5295f..8e88f78e57dba 100644
@@ -129,7 +129,7 @@ static void __init move_device_tree(void)
 		p = __va(memblock_alloc(size, PAGE_SIZE));
 		memcpy(p, initial_boot_params, size);
 		initial_boot_params = p;
-		DBG("Moved device tree to 0x%p\n", p);
+		DBG("Moved device tree to 0x%px\n", p);
 	}
 
 	DBG("<- move_device_tree\n");
@@ -689,7 +689,7 @@ void __init early_init_devtree(void *params)
 {
 	phys_addr_t limit;
 
-	DBG(" -> early_init_devtree(%p)\n", params);
+	DBG(" -> early_init_devtree(%px)\n", params);
 
 	/* Too early to BUG_ON(), do it by hand */
 	if (!early_init_dt_verify(params))
@@ -749,7 +749,7 @@ void __init early_init_devtree(void *params)
 	memblock_allow_resize();
 	memblock_dump_all();
 
-	DBG("Phys. mem: %llx\n", memblock_phys_mem_size());
+	DBG("Phys. mem: %llx\n", (unsigned long long)memblock_phys_mem_size());
 
 	/* We may need to relocate the flat tree, do it now.
 	 * FIXME .. and the initrd too? */
Signature · 1 line
2.20.1
author maintainer
#120 lore raw

PATCH 121/219 powerpc/mm: Make NULL pointer deferences explicit on bad page faults.

In reply to Sasha Levin

From: Christophe Leroy <[email protected]>

[ Upstream commit 49a502ea23bf9dec47f8f3c3960909ff409cd1bb ]

As several other arches including x86, this patch makes it explicit that a bad page fault is a NULL pointer dereference when the fault address is lower than PAGE_SIZE

In the mean time, this page makes all bad_page_fault() messages shorter so that they remain on one single line. And it prefixes them by "BUG: " so that they get easily grepped.

---
 arch/powerpc/mm/fault.c | 17 +++++++++--------
 1 file changed, 9 insertions(+), 8 deletions(-)
arch/powerpc/mm/fault.c +9 -8
index 365526ee29b88..6e0ff8b600ced 100644
@@ -633,21 +633,22 @@ void bad_page_fault(struct pt_regs *regs, unsigned long address, int sig)
 	switch (TRAP(regs)) {
 	case 0x300:
 	case 0x380:
-		printk(KERN_ALERT "Unable to handle kernel paging request for "
-			"data at address 0x%08lx\n", regs->dar);
+		pr_alert("BUG: %s at 0x%08lx\n",
+			 regs->dar < PAGE_SIZE ? "Kernel NULL pointer dereference" :
+			 "Unable to handle kernel data access", regs->dar);
 		break;
 	case 0x400:
 	case 0x480:
-		printk(KERN_ALERT "Unable to handle kernel paging request for "
-			"instruction fetch\n");
+		pr_alert("BUG: Unable to handle kernel instruction fetch%s",
+			 regs->nip < PAGE_SIZE ? " (NULL pointer?)\n" : "\n");
 		break;
 	case 0x600:
-		printk(KERN_ALERT "Unable to handle kernel paging request for "
-			"unaligned access at address 0x%08lx\n", regs->dar);
+		pr_alert("BUG: Unable to handle kernel unaligned access at 0x%08lx\n",
+			 regs->dar);
 		break;
 	default:
-		printk(KERN_ALERT "Unable to handle kernel paging request for "
-			"unknown fault\n");
+		pr_alert("BUG: Unable to handle unknown paging fault at 0x%08lx\n",
+			 regs->dar);
 		break;
 	}
 	printk(KERN_ALERT "Faulting instruction address: 0x%08lx\n",
Signature · 1 line
2.20.1
author maintainer
#121 lore raw

PATCH 122/219 powerpc/44x/bamboo: Fix PCI range

In reply to Sasha Levin

From: Benjamin Herrenschmidt <[email protected]>

[ Upstream commit 3cfb9ebe906b51f2942b1e251009bb251efd2ba6 ]

The bamboo dts has a bug: it uses a non-naturally aligned range for PCI memory space. This isnt' supported by the code, thus causing PCI to break on this system.

This is due to the fact that while the chip memory map has 1G reserved for PCI memory, it's only 512M aligned. The code doesn't know how to split that into 2 different PMMs and fails, so limit the region to 512M.

---
 arch/powerpc/boot/dts/bamboo.dts | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
arch/powerpc/boot/dts/bamboo.dts +3 -1
index 538e42b1120d8..b5861fa3836c1 100644
@@ -268,8 +268,10 @@
 			/* Outbound ranges, one memory and one IO,
 			 * later cannot be changed. Chip supports a second
 			 * IO range but we don't use it for now
+			 * The chip also supports a larger memory range but
+			 * it's not naturally aligned, so our code will break
 			 */
-			ranges = <0x02000000 0x00000000 0xa0000000 0x00000000 0xa0000000 0x00000000 0x40000000
+			ranges = <0x02000000 0x00000000 0xa0000000 0x00000000 0xa0000000 0x00000000 0x20000000
 				  0x02000000 0x00000000 0x00000000 0x00000000 0xe0000000 0x00000000 0x00100000
 				  0x01000000 0x00000000 0x00000000 0x00000000 0xe8000000 0x00000000 0x00010000>;
 
Signature · 1 line
2.20.1
author maintainer
#122 lore raw

PATCH 123/219 vfio/spapr_tce: Get rid of possible infinite loop

In reply to Sasha Levin

From: Alexey Kardashevskiy <[email protected]>

[ Upstream commit 517ad4ae8aa93dccdb9a88c27257ecb421c9e848 ]

As a part of cleanup, the SPAPR TCE IOMMU subdriver releases preregistered memory. If there is a bug in memory release, the loop in tce_iommu_release() becomes infinite; this actually happened to me.

This makes the loop finite and prints a warning on every failure to make the code more bug prone.

---
 drivers/vfio/vfio_iommu_spapr_tce.c | 10 +++-------
 1 file changed, 3 insertions(+), 7 deletions(-)
drivers/vfio/vfio_iommu_spapr_tce.c +3 -7
index 96721b154454f..ec53310f16136 100644
@@ -371,6 +371,7 @@ static void tce_iommu_release(void *iommu_data)
 {
 	struct tce_container *container = iommu_data;
 	struct tce_iommu_group *tcegrp;
+	struct tce_iommu_prereg *tcemem, *tmtmp;
 	long i;
 
 	while (tce_groups_attached(container)) {
@@ -393,13 +394,8 @@ static void tce_iommu_release(void *iommu_data)
 		tce_iommu_free_table(container, tbl);
 	}
 
-	while (!list_empty(&container->prereg_list)) {
-		struct tce_iommu_prereg *tcemem;
-
-		tcemem = list_first_entry(&container->prereg_list,
-				struct tce_iommu_prereg, next);
-		WARN_ON_ONCE(tce_iommu_prereg_free(container, tcemem));
-	}
+	list_for_each_entry_safe(tcemem, tmtmp, &container->prereg_list, next)
+		WARN_ON(tce_iommu_prereg_free(container, tcemem));
 
 	tce_iommu_disable(container);
 	if (container->mm)
Signature · 1 line
2.20.1
author maintainer
#123 lore raw

PATCH 124/219 powerpc/powernv/eeh/npu: Fix uninitialized variables in opal_pci_eeh_freeze_status

In reply to Sasha Levin

From: Alexey Kardashevskiy <[email protected]>

[ Upstream commit c20577014f85f36d4e137d3d52a1f61225b4a3d2 ]

The current implementation of the OPAL_PCI_EEH_FREEZE_STATUS call in skiboot's NPU driver does not touch the pci_error_type parameter so it might have garbage but the powernv code analyzes it nevertheless.

This initializes pcierr and fstate to zero in all call sites.

---
 arch/powerpc/platforms/powernv/eeh-powernv.c | 8 ++++----
 arch/powerpc/platforms/powernv/pci-ioda.c    | 4 ++--
 arch/powerpc/platforms/powernv/pci.c         | 4 ++--
 3 files changed, 8 insertions(+), 8 deletions(-)
arch/powerpc/platforms/powernv/eeh-powernv.c +4 -4
index 3c1beae29f2d8..9dd5b8909178b 100644
@@ -578,8 +578,8 @@ static void pnv_eeh_get_phb_diag(struct eeh_pe *pe)
 static int pnv_eeh_get_phb_state(struct eeh_pe *pe)
 {
 	struct pnv_phb *phb = pe->phb->private_data;
-	u8 fstate;
-	__be16 pcierr;
+	u8 fstate = 0;
+	__be16 pcierr = 0;
 	s64 rc;
 	int result = 0;
 
@@ -617,8 +617,8 @@ static int pnv_eeh_get_phb_state(struct eeh_pe *pe)
 static int pnv_eeh_get_pe_state(struct eeh_pe *pe)
 {
 	struct pnv_phb *phb = pe->phb->private_data;
-	u8 fstate;
-	__be16 pcierr;
+	u8 fstate = 0;
+	__be16 pcierr = 0;
 	s64 rc;
 	int result;
 
arch/powerpc/platforms/powernv/pci-ioda.c +2 -2
index 326ca6288bb12..ee63749a2d47e 100644
@@ -605,8 +605,8 @@ static int pnv_ioda_unfreeze_pe(struct pnv_phb *phb, int pe_no, int opt)
 static int pnv_ioda_get_pe_state(struct pnv_phb *phb, int pe_no)
 {
 	struct pnv_ioda_pe *slave, *pe;
-	u8 fstate, state;
-	__be16 pcierr;
+	u8 fstate = 0, state;
+	__be16 pcierr = 0;
 	s64 rc;
 
 	/* Sanity check on PE number */
arch/powerpc/platforms/powernv/pci.c +2 -2
index 13aef2323bbca..db230a35609bf 100644
@@ -602,8 +602,8 @@ static void pnv_pci_handle_eeh_config(struct pnv_phb *phb, u32 pe_no)
 static void pnv_pci_config_check_eeh(struct pci_dn *pdn)
 {
 	struct pnv_phb *phb = pdn->phb->private_data;
-	u8	fstate;
-	__be16	pcierr;
+	u8	fstate = 0;
+	__be16	pcierr = 0;
 	unsigned int pe_no;
 	s64	rc;
 
Signature · 1 line
2.20.1
author maintainer
#124 lore raw

PATCH 125/219 drbd: ignore "all zero" peer volume sizes in handshake

In reply to Sasha Levin

From: Lars Ellenberg <[email protected]>

[ Upstream commit 94c43a13b8d6e3e0dd77b3536b5e04a84936b762 ]

During handshake, if we are diskless ourselves, we used to accept any size presented by the peer.

Which could be zero if that peer was just brought up and connected to us without having a disk attached first, in which case both peers would just "flip" their volume sizes.

Now, even a diskless node will ignore "zero" sizes presented by a diskless peer.

Also a currently Diskless Primary will refuse to shrink during handshake:
it may be frozen, and waiting for a "suitable" local disk or peer to re-appear (on-no-data-accessible suspend-io). If the peer is smaller than what we used to be, it is not suitable.

The logic for a diskless node during handshake is now supposed to be:
believe the peer, if
- I don't have a current size myself
- we agree on the size anyways
- I do have a current size, am Secondary, and he has the only disk
- I do have a current size, am Primary, and he has the only disk, which is larger than my current size

---
 drivers/block/drbd/drbd_receiver.c | 33 +++++++++++++++++++++++++++---
 1 file changed, 30 insertions(+), 3 deletions(-)
drivers/block/drbd/drbd_receiver.c +30 -3
index 3cdadf75c82da..c9e8d61dea248 100644
@@ -3980,6 +3980,7 @@ static int receive_sizes(struct drbd_connection *connection, struct packet_info
 	struct o_qlim *o = (connection->agreed_features & DRBD_FF_WSAME) ? p->qlim : NULL;
 	enum determine_dev_size dd = DS_UNCHANGED;
 	sector_t p_size, p_usize, p_csize, my_usize;
+	sector_t new_size, cur_size;
 	int ldsc = 0; /* local disk size changed */
 	enum dds_flags ddsf;
 
@@ -3987,6 +3988,7 @@ static int receive_sizes(struct drbd_connection *connection, struct packet_info
 	if (!peer_device)
 		return config_unknown_volume(connection, pi);
 	device = peer_device->device;
+	cur_size = drbd_get_capacity(device->this_bdev);
 
 	p_size = be64_to_cpu(p->d_size);
 	p_usize = be64_to_cpu(p->u_size);
@@ -3997,7 +3999,6 @@ static int receive_sizes(struct drbd_connection *connection, struct packet_info
 	device->p_size = p_size;
 
 	if (get_ldev(device)) {
-		sector_t new_size, cur_size;
 		rcu_read_lock();
 		my_usize = rcu_dereference(device->ldev->disk_conf)->disk_size;
 		rcu_read_unlock();
@@ -4015,7 +4016,6 @@ static int receive_sizes(struct drbd_connection *connection, struct packet_info
 		/* Never shrink a device with usable data during connect.
 		   But allow online shrinking if we are connected. */
 		new_size = drbd_new_dev_size(device, device->ldev, p_usize, 0);
-		cur_size = drbd_get_capacity(device->this_bdev);
 		if (new_size < cur_size &&
 		    device->state.disk >= D_OUTDATED &&
 		    device->state.conn < C_CONNECTED) {
@@ -4080,9 +4080,36 @@ static int receive_sizes(struct drbd_connection *connection, struct packet_info
 		 *
 		 * However, if he sends a zero current size,
 		 * take his (user-capped or) backing disk size anyways.
+		 *
+		 * Unless of course he does not have a disk himself.
+		 * In which case we ignore this completely.
 		 */
+		sector_t new_size = p_csize ?: p_usize ?: p_size;
 		drbd_reconsider_queue_parameters(device, NULL, o);
-		drbd_set_my_capacity(device, p_csize ?: p_usize ?: p_size);
+		if (new_size == 0) {
+			/* Ignore, peer does not know nothing. */
+		} else if (new_size == cur_size) {
+			/* nothing to do */
+		} else if (cur_size != 0 && p_size == 0) {
+			drbd_warn(device, "Ignored diskless peer device size (peer:%llu != me:%llu sectors)!\n",
+					(unsigned long long)new_size, (unsigned long long)cur_size);
+		} else if (new_size < cur_size && device->state.role == R_PRIMARY) {
+			drbd_err(device, "The peer's device size is too small! (%llu < %llu sectors); demote me first!\n",
+					(unsigned long long)new_size, (unsigned long long)cur_size);
+			conn_request_state(peer_device->connection, NS(conn, C_DISCONNECTING), CS_HARD);
+			return -EIO;
+		} else {
+			/* I believe the peer, if
+			 *  - I don't have a current size myself
+			 *  - we agree on the size anyways
+			 *  - I do have a current size, am Secondary,
+			 *    and he has the only disk
+			 *  - I do have a current size, am Primary,
+			 *    and he has the only disk,
+			 *    which is larger than my current size
+			 */
+			drbd_set_my_capacity(device, new_size);
+		}
 	}
 
 	if (get_ldev(device)) {
Signature · 1 line
2.20.1
author maintainer
#125 lore raw

PATCH 126/219 drbd: reject attach of unsuitable uuids even if connected

In reply to Sasha Levin

From: Lars Ellenberg <[email protected]>

[ Upstream commit fe43ed97bba3b11521abd934b83ed93143470e4f ]

Multiple failure scenario:
a) all good
Connected Primary/Secondary UpToDate/UpToDate
b) lose disk on Primary,
Connected Primary/Secondary Diskless/UpToDate
c) continue to write to the device,
changes only make it to the Secondary storage.
d) lose disk on Secondary,
Connected Primary/Secondary Diskless/Diskless
e) now try to re-attach on Primary

This would have succeeded before, even though that is clearly the wrong data set to attach to (missing the modifications from c). Because we only compared our "effective" and the "to-be-attached" data generation uuid tags if (device->state.conn < C_CONNECTED).

Fix: change that constraint to (device->state.pdsk != D_UP_TO_DATE) compare the uuids, and reject the attach.

This patch also tries to improve the reverse scenario:
first lose Secondary, then Primary disk,
then try to attach the disk on Secondary.

Before this patch, the attach on the Secondary succeeds, but since commit
drbd: disconnect, if the wrong UUIDs are attached on a connected peer the Primary will notice unsuitable data, and drop the connection hard.

Though unfortunately at a point in time during the handshake where we cannot easily abort the attach on the peer without more refactoring of the handshake.

We now reject any attach to "unsuitable" uuids,
as long as we can see a Primary role,
unless we already have access to "good" data.

---
 drivers/block/drbd/drbd_nl.c       |  6 +++---
 drivers/block/drbd/drbd_receiver.c | 19 +++++++++++++++++++
 2 files changed, 22 insertions(+), 3 deletions(-)
drivers/block/drbd/drbd_nl.c +3 -3
index 319fabdd63a3f..3366c1a91ee00 100644
@@ -1935,9 +1935,9 @@ int drbd_adm_attach(struct sk_buff *skb, struct genl_info *info)
 		}
 	}
 
-	if (device->state.conn < C_CONNECTED &&
-	    device->state.role == R_PRIMARY && device->ed_uuid &&
-	    (device->ed_uuid & ~((u64)1)) != (nbc->md.uuid[UI_CURRENT] & ~((u64)1))) {
+	if (device->state.pdsk != D_UP_TO_DATE && device->ed_uuid &&
+	    (device->state.role == R_PRIMARY || device->state.peer == R_PRIMARY) &&
+            (device->ed_uuid & ~((u64)1)) != (nbc->md.uuid[UI_CURRENT] & ~((u64)1))) {
 		drbd_err(device, "Can only attach to data with current UUID=%016llX\n",
 		    (unsigned long long)device->ed_uuid);
 		retcode = ERR_DATA_NOT_CURRENT;
drivers/block/drbd/drbd_receiver.c +19 -0
index c9e8d61dea248..cbb6ef719978f 100644
@@ -4395,6 +4395,25 @@ static int receive_state(struct drbd_connection *connection, struct packet_info
 	if (peer_state.conn == C_AHEAD)
 		ns.conn = C_BEHIND;
 
+	/* TODO:
+	 * if (primary and diskless and peer uuid != effective uuid)
+	 *     abort attach on peer;
+	 *
+	 * If this node does not have good data, was already connected, but
+	 * the peer did a late attach only now, trying to "negotiate" with me,
+	 * AND I am currently Primary, possibly frozen, with some specific
+	 * "effective" uuid, this should never be reached, really, because
+	 * we first send the uuids, then the current state.
+	 *
+	 * In this scenario, we already dropped the connection hard
+	 * when we received the unsuitable uuids (receive_uuids().
+	 *
+	 * Should we want to change this, that is: not drop the connection in
+	 * receive_uuids() already, then we would need to add a branch here
+	 * that aborts the attach of "unsuitable uuids" on the peer in case
+	 * this node is currently Diskless Primary.
+	 */
+
 	if (device->p_uuid && peer_state.disk >= D_NEGOTIATING &&
 	    get_ldev_if_state(device, D_NEGOTIATING)) {
 		int cr; /* consider resync */
Signature · 1 line
2.20.1
author maintainer
#126 lore raw

PATCH 127/219 drbd: do not block when adjusting "disk-options" while IO is frozen

In reply to Sasha Levin

From: Lars Ellenberg <[email protected]>

[ Upstream commit f708bd08ecbdc23d03aaedf5b3311ebe44cfdb50 ]

"suspending" IO is overloaded.
It can mean "do not allow new requests" (obviously), but it also may mean "must not complete pending IO", for example while the fencing handlers do their arbitration.

When adjusting disk options, we suspend io (disallow new requests), then wait for the activity-log to become unused (drain all IO completions), and possibly replace it with a new activity log of different size.

If the other "suspend IO" aspect is active, pending IO completions won't happen, and we would block forever (unkillable drbdsetup process).

Fix this by skipping the activity log adjustment if the "al-extents" setting did not change. Also, in case it did change, fail early without blocking if it looks like we would block forever.

---
 drivers/block/drbd/drbd_nl.c | 37 ++++++++++++++++++++++++++++--------
 1 file changed, 29 insertions(+), 8 deletions(-)
drivers/block/drbd/drbd_nl.c +29 -8
index 3366c1a91ee00..5b15ffd0c7f57 100644
@@ -1515,6 +1515,30 @@ static void sanitize_disk_conf(struct drbd_device *device, struct disk_conf *dis
 	}
 }
 
+static int disk_opts_check_al_size(struct drbd_device *device, struct disk_conf *dc)
+{
+	int err = -EBUSY;
+
+	if (device->act_log &&
+	    device->act_log->nr_elements == dc->al_extents)
+		return 0;
+
+	drbd_suspend_io(device);
+	/* If IO completion is currently blocked, we would likely wait
+	 * "forever" for the activity log to become unused. So we don't. */
+	if (atomic_read(&device->ap_bio_cnt))
+		goto out;
+
+	wait_event(device->al_wait, lc_try_lock(device->act_log));
+	drbd_al_shrink(device);
+	err = drbd_check_al_size(device, dc);
+	lc_unlock(device->act_log);
+	wake_up(&device->al_wait);
+out:
+	drbd_resume_io(device);
+	return err;
+}
+
 int drbd_adm_disk_opts(struct sk_buff *skb, struct genl_info *info)
 {
 	struct drbd_config_context adm_ctx;
@@ -1577,15 +1601,12 @@ int drbd_adm_disk_opts(struct sk_buff *skb, struct genl_info *info)
 		}
 	}
 
-	drbd_suspend_io(device);
-	wait_event(device->al_wait, lc_try_lock(device->act_log));
-	drbd_al_shrink(device);
-	err = drbd_check_al_size(device, new_disk_conf);
-	lc_unlock(device->act_log);
-	wake_up(&device->al_wait);
-	drbd_resume_io(device);
-
+	err = disk_opts_check_al_size(device, new_disk_conf);
 	if (err) {
+		/* Could be just "busy". Ignore?
+		 * Introduce dedicated error code? */
+		drbd_msg_put_info(adm_ctx.reply_skb,
+			"Try again without changing current al-extents setting");
 		retcode = ERR_NOMEM;
 		goto fail_unlock;
 	}
Signature · 1 line
2.20.1
author maintainer
#127 lore raw

PATCH 128/219 drbd: fix print_st_err()'s prototype to match the definition

In reply to Sasha Levin

From: Luc Van Oostenryck <[email protected]>

[ Upstream commit 2c38f035117331eb78d0504843c79ea7c7fabf37 ]

print_st_err() is defined with its 4th argument taking an 'enum drbd_state_rv' but its prototype use an int for it.

Fix this by using 'enum drbd_state_rv' in the prototype too.

---
 drivers/block/drbd/drbd_state.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/block/drbd/drbd_state.h +1 -1
index ea58301d0895c..b2a390ba73a05 100644
@@ -131,7 +131,7 @@ extern enum drbd_state_rv _drbd_set_state(struct drbd_device *, union drbd_state
 					  enum chg_state_flags,
 					  struct completion *done);
 extern void print_st_err(struct drbd_device *, union drbd_state,
-			union drbd_state, int);
+			union drbd_state, enum drbd_state_rv);
 
 enum drbd_state_rv
 _conn_request_state(struct drbd_connection *connection, union drbd_state mask, union drbd_state val,
Signature · 1 line
2.20.1
author maintainer
#128 lore raw

PATCH 129/219 IB/rxe: Make counters thread safe

In reply to Sasha Levin

From: Parav Pandit <[email protected]>

[ Upstream commit d5108e69fe013ff47ab815b849caba9cc33ca1e5 ]

Current rxe device counters are not thread safe. When multiple QPs are used, they can be racy. Make them thread safe by making it atomic64.

Fixes 0b1e5b99a48b ("IB/rxe: Add port protocol stats")
---
 drivers/infiniband/sw/rxe/rxe_hw_counters.c | 2 +-
 drivers/infiniband/sw/rxe/rxe_verbs.h       | 6 +++---
 2 files changed, 4 insertions(+), 4 deletions(-)
drivers/infiniband/sw/rxe/rxe_hw_counters.c +1 -1
index 6aeb7a165e469..ea4542a9d69e6 100644
@@ -59,7 +59,7 @@ int rxe_ib_get_hw_stats(struct ib_device *ibdev,
 		return -EINVAL;
 
 	for (cnt = 0; cnt  < ARRAY_SIZE(rxe_counter_name); cnt++)
-		stats->value[cnt] = dev->stats_counters[cnt];
+		stats->value[cnt] = atomic64_read(&dev->stats_counters[cnt]);
 
 	return ARRAY_SIZE(rxe_counter_name);
 }
drivers/infiniband/sw/rxe/rxe_verbs.h +3 -3
index 3b731c7682e5b..a61782df4e805 100644
@@ -408,16 +408,16 @@ struct rxe_dev {
 	spinlock_t		mmap_offset_lock; /* guard mmap_offset */
 	int			mmap_offset;
 
-	u64			stats_counters[RXE_NUM_OF_COUNTERS];
+	atomic64_t		stats_counters[RXE_NUM_OF_COUNTERS];
 
 	struct rxe_port		port;
 	struct list_head	list;
 	struct crypto_shash	*tfm;
 };
 
-static inline void rxe_counter_inc(struct rxe_dev *rxe, enum rxe_counters cnt)
+static inline void rxe_counter_inc(struct rxe_dev *rxe, enum rxe_counters index)
 {
-	rxe->stats_counters[cnt]++;
+	atomic64_inc(&rxe->stats_counters[index]);
 }
 
 static inline struct rxe_dev *to_rdev(struct ib_device *dev)
Signature · 1 line
2.20.1
author maintainer
#129 lore raw

PATCH 130/219 bpf/cpumap: make sure frame_size for build_skb is aligned if headroom isn't

In reply to Sasha Levin

From: Jesper Dangaard Brouer <[email protected]>

[ Upstream commit 77ea5f4cbe2084db9ab021ba73fb7eadf1610884 ]

The frame_size passed to build_skb must be aligned, else it is possible that the embedded struct skb_shared_info gets unaligned.

For correctness make sure that xdpf->headroom in included in the alignment. No upstream drivers can hit this, as all XDP drivers provide an aligned headroom. This was discovered when playing with implementing XDP support for mvneta, which have a 2 bytes DSA header, and this Marvell ARM64 platform didn't like doing atomic operations on an unaligned skb_shinfo(skb)->dataref addresses.

Fixes 1c601d829ab0 ("bpf: cpumap xdp_buff to skb conversion and allocation")
---
 kernel/bpf/cpumap.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
kernel/bpf/cpumap.c +1 -1
index 24aac0d0f4127..8974b3755670e 100644
@@ -183,7 +183,7 @@ static struct sk_buff *cpu_map_build_skb(struct bpf_cpu_map_entry *rcpu,
 	 * is not at a fixed memory location, with mixed length
 	 * packets, which is bad for cache-line hotness.
 	 */
-	frame_size = SKB_DATA_ALIGN(xdpf->len) + xdpf->headroom +
+	frame_size = SKB_DATA_ALIGN(xdpf->len + xdpf->headroom) +
 		SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
 
 	pkt_data_start = xdpf->data - xdpf->headroom;
Signature · 1 line
2.20.1
author maintainer
#130 lore raw

PATCH 131/219 regulator: tps65910: fix a missing check of return value

In reply to Sasha Levin

From: Kangjie Lu <[email protected]>

[ Upstream commit cd07e3701fa6a4c68f8493ee1d12caa18d46ec6a ]

tps65910_reg_set_bits() may fail. The fix checks if it fails, and if so, returns with its error code.

---
 drivers/regulator/tps65910-regulator.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/regulator/tps65910-regulator.c +3 -1
index 02ccdaa226a73..5ebb6ee73f077 100644
@@ -1102,8 +1102,10 @@ static int tps65910_probe(struct platform_device *pdev)
 	platform_set_drvdata(pdev, pmic);
 
 	/* Give control of all register to control port */
-	tps65910_reg_set_bits(pmic->mfd, TPS65910_DEVCTRL,
+	err = tps65910_reg_set_bits(pmic->mfd, TPS65910_DEVCTRL,
 				DEVCTRL_SR_CTL_I2C_SEL_MASK);
+	if (err < 0)
+		return err;
 
 	switch (tps65910_chip_id(tps65910)) {
 	case TPS65910:
Signature · 1 line
2.20.1
author maintainer
#131 lore raw

PATCH 132/219 powerpc/83xx: handle machine check caused by watchdog timer

In reply to Sasha Levin

From: Christophe Leroy <[email protected]>

[ Upstream commit 0deae39cec6dab3a66794f3e9e83ca4dc30080f1 ]

When the watchdog timer is set in interrupt mode, it causes a machine check when it times out. The purpose of this mode is to ease debugging, not to crash the kernel and reboot the machine.

This patch implements a special handling for that, in order to not crash the kernel if the watchdog times out while in interrupt or within the idle task.

[scottwood: added missing #include]
---
 arch/powerpc/include/asm/cputable.h |  1 +
 arch/powerpc/include/asm/reg.h      |  2 ++
 arch/powerpc/kernel/cputable.c      | 10 ++++++----
 arch/powerpc/platforms/83xx/misc.c  | 17 +++++++++++++++++
 4 files changed, 26 insertions(+), 4 deletions(-)
arch/powerpc/include/asm/cputable.h +1 -0
index 3ce690e5f345c..59b35b93eadec 100644
@@ -44,6 +44,7 @@ extern int machine_check_e500(struct pt_regs *regs);
 extern int machine_check_e200(struct pt_regs *regs);
 extern int machine_check_47x(struct pt_regs *regs);
 int machine_check_8xx(struct pt_regs *regs);
+int machine_check_83xx(struct pt_regs *regs);
 
 extern void cpu_down_flush_e500v2(void);
 extern void cpu_down_flush_e500mc(void);
arch/powerpc/include/asm/reg.h +2 -0
index 640a4d818772a..af99716615122 100644
@@ -768,6 +768,8 @@
 #define   SRR1_PROGTRAP		0x00020000 /* Trap */
 #define   SRR1_PROGADDR		0x00010000 /* SRR0 contains subsequent addr */
 
+#define   SRR1_MCE_MCP		0x00080000 /* Machine check signal caused interrupt */
+
 #define SPRN_HSRR0	0x13A	/* Save/Restore Register 0 */
 #define SPRN_HSRR1	0x13B	/* Save/Restore Register 1 */
 #define   HSRR1_DENORM		0x00100000 /* Denorm exception */
arch/powerpc/kernel/cputable.c +6 -4
index 2da01340c84c3..1eab54bc6ee93 100644
@@ -1141,6 +1141,7 @@ static struct cpu_spec __initdata cpu_specs[] = {
 		.machine_check		= machine_check_generic,
 		.platform		= "ppc603",
 	},
+#ifdef CONFIG_PPC_83xx
 	{	/* e300c1 (a 603e core, plus some) on 83xx */
 		.pvr_mask		= 0x7fff0000,
 		.pvr_value		= 0x00830000,
@@ -1151,7 +1152,7 @@ static struct cpu_spec __initdata cpu_specs[] = {
 		.icache_bsize		= 32,
 		.dcache_bsize		= 32,
 		.cpu_setup		= __setup_cpu_603,
-		.machine_check		= machine_check_generic,
+		.machine_check		= machine_check_83xx,
 		.platform		= "ppc603",
 	},
 	{	/* e300c2 (an e300c1 core, plus some, minus FPU) on 83xx */
@@ -1165,7 +1166,7 @@ static struct cpu_spec __initdata cpu_specs[] = {
 		.icache_bsize		= 32,
 		.dcache_bsize		= 32,
 		.cpu_setup		= __setup_cpu_603,
-		.machine_check		= machine_check_generic,
+		.machine_check		= machine_check_83xx,
 		.platform		= "ppc603",
 	},
 	{	/* e300c3 (e300c1, plus one IU, half cache size) on 83xx */
@@ -1179,7 +1180,7 @@ static struct cpu_spec __initdata cpu_specs[] = {
 		.icache_bsize		= 32,
 		.dcache_bsize		= 32,
 		.cpu_setup		= __setup_cpu_603,
-		.machine_check		= machine_check_generic,
+		.machine_check		= machine_check_83xx,
 		.num_pmcs		= 4,
 		.oprofile_cpu_type	= "ppc/e300",
 		.oprofile_type		= PPC_OPROFILE_FSL_EMB,
@@ -1196,12 +1197,13 @@ static struct cpu_spec __initdata cpu_specs[] = {
 		.icache_bsize		= 32,
 		.dcache_bsize		= 32,
 		.cpu_setup		= __setup_cpu_603,
-		.machine_check		= machine_check_generic,
+		.machine_check		= machine_check_83xx,
 		.num_pmcs		= 4,
 		.oprofile_cpu_type	= "ppc/e300",
 		.oprofile_type		= PPC_OPROFILE_FSL_EMB,
 		.platform		= "ppc603",
 	},
+#endif
 	{	/* default match, we assume split I/D cache & TB (non-601)... */
 		.pvr_mask		= 0x00000000,
 		.pvr_value		= 0x00000000,
arch/powerpc/platforms/83xx/misc.c +17 -0
index d75c9816a5c92..2b6589fe812dd 100644
@@ -14,6 +14,7 @@
 #include <linux/of_platform.h>
 #include <linux/pci.h>
 
+#include <asm/debug.h>
 #include <asm/io.h>
 #include <asm/hw_irq.h>
 #include <asm/ipic.h>
@@ -150,3 +151,19 @@ void __init mpc83xx_setup_arch(void)
 
 	mpc83xx_setup_pci();
 }
+
+int machine_check_83xx(struct pt_regs *regs)
+{
+	u32 mask = 1 << (31 - IPIC_MCP_WDT);
+
+	if (!(regs->msr & SRR1_MCE_MCP) || !(ipic_get_mcp_status() & mask))
+		return machine_check_generic(regs);
+	ipic_clear_mcp_status(mask);
+
+	if (debugger_fault_handler(regs))
+		return 1;
+
+	die("Watchdog NMI Reset", regs, 0);
+
+	return 1;
+}
Signature · 1 line
2.20.1
author maintainer
#132 lore raw

PATCH 133/219 powerpc/pseries: Fix node leak in update_lmb_associativity_index()

In reply to Sasha Levin

From: Michael Ellerman <[email protected]>

[ Upstream commit 47918bc68b7427e961035949cc1501a864578a69 ]

In update_lmb_associativity_index() we lookup dr_node using of_find_node_by_path() which takes a reference for us. In the non-error case we forget to drop the reference. Note that find_aa_index() does modify properties of the node, but doesn't need an extra reference held once it's returned.

---
 arch/powerpc/platforms/pseries/hotplug-memory.c | 1 +
 1 file changed, 1 insertion(+)
arch/powerpc/platforms/pseries/hotplug-memory.c +1 -0
index f99cd31b6fd1a..8a1c2c8963214 100644
@@ -202,6 +202,7 @@ static u32 lookup_lmb_associativity_index(struct drmem_lmb *lmb)
 
 	aa_index = find_aa_index(dr_node, ala_prop, lmb_assoc);
 
+	of_node_put(dr_node);
 	dlpar_free_cc_nodes(lmb_node);
 	return aa_index;
 }
Signature · 1 line
2.20.1
author maintainer
#133 lore raw

PATCH 134/219 powerpc: Fix HMIs on big-endian with CONFIG_RELOCATABLE=y

In reply to Sasha Levin

From: Benjamin Herrenschmidt <[email protected]>

[ Upstream commit 505a314fb28ce122091691c51426fa85c084e115 ]

HMIs will crash the kernel due to

BRANCH_LINK_TO_FAR(hmi_exception_realmode)

Calling into the OPD instead of the actual code.

Fixes 2337d207288f ("powerpc/64: CONFIG_RELOCATABLE support for hmi interrupts")
[mpe: Use DOTSYM() rather than #ifdef]
---
 arch/powerpc/kernel/exceptions-64s.S | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/powerpc/kernel/exceptions-64s.S +1 -1
index 90af86f143a91..e1dab9b1e4478 100644
@@ -1123,7 +1123,7 @@ TRAMP_REAL_BEGIN(hmi_exception_early)
 	EXCEPTION_PROLOG_COMMON_2(PACA_EXGEN)
 	EXCEPTION_PROLOG_COMMON_3(0xe60)
 	addi	r3,r1,STACK_FRAME_OVERHEAD
-	BRANCH_LINK_TO_FAR(hmi_exception_realmode) /* Function call ABI */
+	BRANCH_LINK_TO_FAR(DOTSYM(hmi_exception_realmode)) /* Function call ABI */
 	cmpdi	cr0,r3,0
 
 	/* Windup the stack. */
Signature · 1 line
2.20.1
author maintainer
#134 lore raw

PATCH 135/219 crypto: mxc-scc - fix build warnings on ARM64

In reply to Sasha Levin

From: Fabio Estevam <[email protected]>

[ Upstream commit 2326828ee40357b3d2b1359b8ca7526af201495b ]

The following build warnings are seen when building for ARM64 allmodconfig:

drivers/crypto/mxc-scc.c:181:20: warning: format '%d' expects argument of type 'int', but argument 5 has type 'size_t' {aka 'long unsigned int'} [-Wformat=] drivers/crypto/mxc-scc.c:186:21: warning: format '%d' expects argument of type 'int', but argument 4 has type 'size_t' {aka 'long unsigned int'} [-Wformat=] drivers/crypto/mxc-scc.c:277:21: warning: format '%d' expects argument of type 'int', but argument 4 has type 'size_t' {aka 'long unsigned int'} [-Wformat=] drivers/crypto/mxc-scc.c:339:3: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast] drivers/crypto/mxc-scc.c:340:3: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast]

Fix them by using the %zu specifier to print a size_t variable and using a plain %x to print the result of a readl().

---
 drivers/crypto/mxc-scc.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)
drivers/crypto/mxc-scc.c +6 -6
index e01c46387df8d..519086730791b 100644
@@ -178,12 +178,12 @@ static int mxc_scc_get_data(struct mxc_scc_ctx *ctx,
 	else
 		from = scc->black_memory;
 
-	dev_dbg(scc->dev, "pcopy: from 0x%p %d bytes\n", from,
+	dev_dbg(scc->dev, "pcopy: from 0x%p %zu bytes\n", from,
 		ctx->dst_nents * 8);
 	len = sg_pcopy_from_buffer(ablkreq->dst, ctx->dst_nents,
 				   from, ctx->size, ctx->offset);
 	if (!len) {
-		dev_err(scc->dev, "pcopy err from 0x%p (len=%d)\n", from, len);
+		dev_err(scc->dev, "pcopy err from 0x%p (len=%zu)\n", from, len);
 		return -EINVAL;
 	}
 
@@ -274,7 +274,7 @@ static int mxc_scc_put_data(struct mxc_scc_ctx *ctx,
 	len = sg_pcopy_to_buffer(req->src, ctx->src_nents,
 				 to, len, ctx->offset);
 	if (!len) {
-		dev_err(scc->dev, "pcopy err to 0x%p (len=%d)\n", to, len);
+		dev_err(scc->dev, "pcopy err to 0x%p (len=%zu)\n", to, len);
 		return -EINVAL;
 	}
 
@@ -335,9 +335,9 @@ static void mxc_scc_ablkcipher_next(struct mxc_scc_ctx *ctx,
 		return;
 	}
 
-	dev_dbg(scc->dev, "Start encryption (0x%p/0x%p)\n",
-		(void *)readl(scc->base + SCC_SCM_RED_START),
-		(void *)readl(scc->base + SCC_SCM_BLACK_START));
+	dev_dbg(scc->dev, "Start encryption (0x%x/0x%x)\n",
+		readl(scc->base + SCC_SCM_RED_START),
+		readl(scc->base + SCC_SCM_BLACK_START));
 
 	/* clear interrupt control registers */
 	writel(SCC_SCM_INTR_CTRL_CLR_INTR,
Signature · 1 line
2.20.1
author maintainer
#135 lore raw

PATCH 136/219 MIPS: BCM63XX: fix switch core reset on BCM6368

In reply to Sasha Levin

From: Jonas Gorski <[email protected]>

[ Upstream commit 8a38dacf87180738d42b058334c951eba15d2d47 ]

The Ethernet Switch core mask was set to 0, causing the switch core to be not reset on BCM6368 on boot. Provide the proper mask so the switch core gets reset to a known good state.

Fixes 799faa626c71 ("MIPS: BCM63XX: add core reset helper")
Cc Ralf Baechle <[email protected]>
Cc James Hogan <[email protected]>
Cc Florian Fainelli <[email protected]>
---
 arch/mips/bcm63xx/reset.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/mips/bcm63xx/reset.c +1 -1
index a2af38cf28a70..64574e74cb236 100644
@@ -120,7 +120,7 @@
 #define BCM6368_RESET_DSL	0
 #define BCM6368_RESET_SAR	SOFTRESET_6368_SAR_MASK
 #define BCM6368_RESET_EPHY	SOFTRESET_6368_EPHY_MASK
-#define BCM6368_RESET_ENETSW	0
+#define BCM6368_RESET_ENETSW	SOFTRESET_6368_ENETSW_MASK
 #define BCM6368_RESET_PCM	SOFTRESET_6368_PCM_MASK
 #define BCM6368_RESET_MPI	SOFTRESET_6368_MPI_MASK
 #define BCM6368_RESET_PCIE	0
Signature · 7 lines
2.20.1


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
author maintainer
#136 lore raw

PATCH 137/219 pwm: clps711x: Fix period calculation

In reply to Sasha Levin

From: Alexander Shiyan <[email protected]>

[ Upstream commit b0f17570b8203c22f139459c86cfbaa0311313ed ]

Commit e39c0df1be5a ("pwm: Introduce the pwm_args concept") has changed the variable for the period for clps711x-pwm driver, so now pwm_get/set_period() works with pwm->state.period variable instead of pwm->args.period.
This patch changes the period variable in other places where it is used.

---
 drivers/pwm/pwm-clps711x.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/pwm/pwm-clps711x.c +2 -2
index 26ec24e457b12..7e16b7def0dcb 100644
@@ -48,7 +48,7 @@ static void clps711x_pwm_update_val(struct clps711x_chip *priv, u32 n, u32 v)
 static unsigned int clps711x_get_duty(struct pwm_device *pwm, unsigned int v)
 {
 	/* Duty cycle 0..15 max */
-	return DIV_ROUND_CLOSEST(v * 0xf, pwm_get_period(pwm));
+	return DIV_ROUND_CLOSEST(v * 0xf, pwm->args.period);
 }
 
 static int clps711x_pwm_request(struct pwm_chip *chip, struct pwm_device *pwm)
@@ -71,7 +71,7 @@ static int clps711x_pwm_config(struct pwm_chip *chip, struct pwm_device *pwm,
 	struct clps711x_chip *priv = to_clps711x_chip(chip);
 	unsigned int duty;
 
-	if (period_ns != pwm_get_period(pwm))
+	if (period_ns != pwm->args.period)
 		return -EINVAL;
 
 	duty = clps711x_get_duty(pwm, duty_ns);
Signature · 1 line
2.20.1
author maintainer
#137 lore raw

PATCH 138/219 net/netlink_compat: Fix a missing check of nla_parse_nested

In reply to Sasha Levin

From: Aditya Pakki <[email protected]>

[ Upstream commit 89dfd0083751d00d5d7ead36f6d8b045bf89c5e1 ]

In tipc_nl_compat_sk_dump(), if nla_parse_nested() fails, it could return an error. To be consistent with other invocations of the function call, on error, the fix passes the return value upstream.

---
 net/tipc/netlink_compat.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)
net/tipc/netlink_compat.c +5 -2
index 318c541970ecd..6494d6b5e1b24 100644
@@ -1030,8 +1030,11 @@ static int tipc_nl_compat_sk_dump(struct tipc_nl_compat_msg *msg,
 		u32 node;
 		struct nlattr *con[TIPC_NLA_CON_MAX + 1];
 
-		nla_parse_nested(con, TIPC_NLA_CON_MAX,
-				 sock[TIPC_NLA_SOCK_CON], NULL, NULL);
+		err = nla_parse_nested(con, TIPC_NLA_CON_MAX,
+				       sock[TIPC_NLA_SOCK_CON], NULL, NULL);
+
+		if (err)
+			return err;
 
 		node = nla_get_u32(con[TIPC_NLA_CON_NODE]);
 		tipc_tlv_sprintf(msg->rep, "  connected to <%u.%u.%u:%u>",
Signature · 1 line
2.20.1
author maintainer
#138 lore raw

PATCH 139/219 net/net_namespace: Check the return value of register_pernet_subsys()

In reply to Sasha Levin

From: Aditya Pakki <[email protected]>

[ Upstream commit 0eb987c874dc93f9c9d85a6465dbde20fdd3884c ]

In net_ns_init(), register_pernet_subsys() could fail while registering network namespace subsystems. The fix checks the return value and sends a panic() on failure.

---
 net/core/net_namespace.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
net/core/net_namespace.c +2 -1
index 6dab186d4b8f6..c60123dff8039 100644
@@ -913,7 +913,8 @@ static int __init net_ns_init(void)
 	init_net_initialized = true;
 	up_write(&pernet_ops_rwsem);
 
-	register_pernet_subsys(&net_ns_ops);
+	if (register_pernet_subsys(&net_ns_ops))
+		panic("Could not register network namespace subsystems");
 
 	rtnl_register(PF_UNSPEC, RTM_NEWNSID, rtnl_net_newid, NULL,
 		      RTNL_FLAG_DOIT_UNLOCKED);
Signature · 1 line
2.20.1
author maintainer
#139 lore raw

PATCH 140/219 libceph: drop last_piece logic from write_partial_message_data()

In reply to Sasha Levin

From: Ilya Dryomov <[email protected]>

[ Upstream commit 1f6b821aef78e3d79e8d598ae59fc7e23fb6c563 ]

last_piece is for the last piece in the current data item, not in the entire data payload of the message. This is harmful for messages with multiple data items. On top of that, we don't need to signal the end of a data payload either because it is always followed by a footer.

We used to signal "more" unconditionally, until commit fe38a2b67bc6 ("libceph: start defining message data cursor"). Part of a large series, it introduced cursor->last_piece and also mistakenly inverted the hint by passing last_piece for "more". This was corrected with commit c2cfa1940097 ("libceph: Fix ceph_tcp_sendpage()'s more boolean usage").

As it is, last_piece is not helping at all: because Nagle algorithm is disabled, for a simple message with two 512-byte data items we end up emitting three packets: front + first data item, second data item and footer. Go back to the original pre-fe38a2b67bc6 behavior -- a single packet in most cases.

---
 net/ceph/messenger.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)
net/ceph/messenger.c +3 -5
index f7d7f32ac673c..6514816947fbe 100644
@@ -1612,7 +1612,6 @@ static int write_partial_message_data(struct ceph_connection *con)
 		struct page *page;
 		size_t page_offset;
 		size_t length;
-		bool last_piece;
 		int ret;
 
 		if (!cursor->resid) {
@@ -1620,10 +1619,9 @@ static int write_partial_message_data(struct ceph_connection *con)
 			continue;
 		}
 
-		page = ceph_msg_data_next(cursor, &page_offset, &length,
-					  &last_piece);
-		ret = ceph_tcp_sendpage(con->sock, page, page_offset,
-					length, !last_piece);
+		page = ceph_msg_data_next(cursor, &page_offset, &length, NULL);
+		ret = ceph_tcp_sendpage(con->sock, page, page_offset, length,
+					true);
 		if (ret <= 0) {
 			if (do_datacrc)
 				msg->footer.data_crc = cpu_to_le32(crc);
Signature · 1 line
2.20.1
#140 lore raw

Re: [PATCH AUTOSEL 4.19 140/219] libceph: drop last_piece logic from write_partial_message_data()

Quoted 45 lines · Sasha Levin

From: Ilya Dryomov <[email protected]>

[ Upstream commit 1f6b821aef78e3d79e8d598ae59fc7e23fb6c563 ]

last_piece is for the last piece in the current data item, not in the entire data payload of the message. This is harmful for messages with multiple data items. On top of that, we don't need to signal the end of a data payload either because it is always followed by a footer.

We used to signal "more" unconditionally, until commit fe38a2b67bc6 ("libceph: start defining message data cursor"). Part of a large series, it introduced cursor->last_piece and also mistakenly inverted the hint by passing last_piece for "more". This was corrected with commit c2cfa1940097 ("libceph: Fix ceph_tcp_sendpage()'s more boolean usage").

As it is, last_piece is not helping at all: because Nagle algorithm is disabled, for a simple message with two 512-byte data items we end up emitting three packets: front + first data item, second data item and footer. Go back to the original pre-fe38a2b67bc6 behavior -- a single packet in most cases.

---
 net/ceph/messenger.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)
net/ceph/messenger.c +0 -1
index f7d7f32ac673c..6514816947fbe 100644
@@ -1612,7 +1612,6 @@ static int write_partial_message_data(struct ceph_connection *con)
                struct page *page;
                size_t page_offset;
                size_t length;
-               bool last_piece;
                int ret;
 
                if (!cursor->resid) {
@@ -1620,10 +1619,9 @@ static int write_partial_message_data(struct ceph_connection *con)
                        continue;
                }
 

net/ceph/messenger.c static int write_partial_message_data(struct ceph_connection *con) · line 1627

Quoted 10 lines
net/ceph/messenger.c +3 -4
-               page = ceph_msg_data_next(cursor, &page_offset, &length,
-                                         &last_piece);
-               ret = ceph_tcp_sendpage(con->sock, page, page_offset,
-                                       length, !last_piece);
+               page = ceph_msg_data_next(cursor, &page_offset, &length, NULL);
+               ret = ceph_tcp_sendpage(con->sock, page, page_offset, length,
+                                       true);
                if (ret <= 0) {
                        if (do_datacrc)
                                msg->footer.data_crc = cpu_to_le32(crc);

Hi Sasha,

This commit was part of a larger series and shouldn't be backported on its own. Please drop it.

Thanks,

Ilya

author maintainer
#141 lore raw

Re: [PATCH AUTOSEL 4.19 140/219] libceph: drop last_piece logic from write_partial_message_data()

Quoted 59 lines · Ilya Dryomov
Quoted 45 lines · Sasha Levin

From: Ilya Dryomov <[email protected]>

[ Upstream commit 1f6b821aef78e3d79e8d598ae59fc7e23fb6c563 ]

last_piece is for the last piece in the current data item, not in the entire data payload of the message. This is harmful for messages with multiple data items. On top of that, we don't need to signal the end of a data payload either because it is always followed by a footer.

We used to signal "more" unconditionally, until commit fe38a2b67bc6 ("libceph: start defining message data cursor"). Part of a large series, it introduced cursor->last_piece and also mistakenly inverted the hint by passing last_piece for "more". This was corrected with commit c2cfa1940097 ("libceph: Fix ceph_tcp_sendpage()'s more boolean usage").

As it is, last_piece is not helping at all: because Nagle algorithm is disabled, for a simple message with two 512-byte data items we end up emitting three packets: front + first data item, second data item and footer. Go back to the original pre-fe38a2b67bc6 behavior -- a single packet in most cases.

---
 net/ceph/messenger.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)
net/ceph/messenger.c +0 -1
index f7d7f32ac673c..6514816947fbe 100644
@@ -1612,7 +1612,6 @@ static int write_partial_message_data(struct ceph_connection *con)
                struct page *page;
                size_t page_offset;
                size_t length;
-               bool last_piece;
                int ret;
 
                if (!cursor->resid) {
@@ -1620,10 +1619,9 @@ static int write_partial_message_data(struct ceph_connection *con)
                        continue;
                }
 

net/ceph/messenger.c static int write_partial_message_data(struct ceph_connection *con) · line 1627

Quoted 10 lines
net/ceph/messenger.c +3 -4
-               page = ceph_msg_data_next(cursor, &page_offset, &length,
-                                         &last_piece);
-               ret = ceph_tcp_sendpage(con->sock, page, page_offset,
-                                       length, !last_piece);
+               page = ceph_msg_data_next(cursor, &page_offset, &length, NULL);
+               ret = ceph_tcp_sendpage(con->sock, page, page_offset, length,
+                                       true);
                if (ret <= 0) {
                        if (do_datacrc)
                                msg->footer.data_crc = cpu_to_le32(crc);

Hi Sasha,

Quoted 2 lines

This commit was part of a larger series and shouldn't be backported on its own. Please drop it.

Now dropped, thanks!

Signature · 2 lines
Thanks,
Sasha
author maintainer
#142 lore raw

PATCH 141/219 f2fs: fix block address for __check_sit_bitmap

In reply to Sasha Levin

From: Qiuyang Sun <[email protected]>

[ Upstream commit 9249dded7b5cb539a8c8698b25d08a3c15261470 ]

Should use lstart (logical start address) instead of start (in dev) here. This fixes a bug in multi-device scenarios.

---
 fs/f2fs/segment.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/f2fs/segment.c +1 -1
index 10d5dcdb34be6..01cb87b12d40b 100644
@@ -1101,7 +1101,7 @@ static int __submit_discard_cmd(struct f2fs_sb_info *sbi,
 		list_move_tail(&dc->list, wait_list);
 
 		/* sanity check on discard range */
-		__check_sit_bitmap(sbi, start, start + len);
+		__check_sit_bitmap(sbi, lstart, lstart + len);
 
 		bio->bi_private = dc;
 		bio->bi_end_io = f2fs_submit_discard_endio;
Signature · 8 lines
2.20.1



_______________________________________________
Linux-f2fs-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
author maintainer
#143 lore raw

PATCH 142/219 f2fs: fix to dirty inode synchronously

In reply to Sasha Levin

From: Chao Yu <[email protected]>

[ Upstream commit b32e019049e959ee10ec359893c9dd5d057dad55 ]

If user change inode's i_flags via ioctl, let's add it into global dirty list, so that checkpoint can guarantee its persistence before fsync, it can make checkpoint keeping strong consistency.

---
 fs/f2fs/file.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/f2fs/file.c +1 -1
index 8d1eb8dec6058..7a5e84cfccf5e 100644
@@ -1668,7 +1668,7 @@ static int __f2fs_ioc_setflags(struct inode *inode, unsigned int flags)
 
 	inode->i_ctime = current_time(inode);
 	f2fs_set_inode_flags(inode);
-	f2fs_mark_inode_dirty_sync(inode, false);
+	f2fs_mark_inode_dirty_sync(inode, true);
 	return 0;
 }
 
Signature · 1 line
2.20.1
author maintainer
#144 lore raw

PATCH 143/219 um: Include sys/uio.h to have writev()

In reply to Sasha Levin

From: Richard Weinberger <[email protected]>

[ Upstream commit 0053102a869f1b909904b1b85ac282e2744deaab ]

sys/uio.h gives us writev(), otherwise the build might fail on some systems.

Fixes 49da7e64f33e ("High Performance UML Vector Network Driver")
---
 arch/um/drivers/vector_user.c | 1 +
 1 file changed, 1 insertion(+)
arch/um/drivers/vector_user.c +1 -0
index 4d6a78e31089f..00c4c2735a5f7 100644
@@ -30,6 +30,7 @@
 #include <stdlib.h>
 #include <os.h>
 #include <um_malloc.h>
+#include <sys/uio.h>
 #include "vector_user.h"
 
 #define ID_GRE 0
Signature · 1 line
2.20.1
author maintainer
#145 lore raw

PATCH 144/219 um: Make GCOV depend on !KCOV

In reply to Sasha Levin

From: Richard Weinberger <[email protected]>

[ Upstream commit 550ed0e2036663b35cec12374b835444f9c60454 ]

Both do more or less the same thing and are mutually exclusive. If both are enabled the build will fail.
Sooner or later we can kill UML's GCOV.

---
 arch/um/Kconfig.debug | 1 +
 1 file changed, 1 insertion(+)
arch/um/Kconfig.debug +1 -0
index 2014597605ea9..85726eeec3451 100644
@@ -16,6 +16,7 @@ config GPROF
 config GCOV
 	bool "Enable gcov support"
 	depends on DEBUG_INFO
+	depends on !KCOV
 	help
 	  This option allows developers to retrieve coverage data from a UML
 	  session.
Signature · 7 lines
2.20.1


_______________________________________________
linux-um mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-um
author maintainer
#146 lore raw

PATCH 145/219 net: (cpts) fix a missing check of clk_prepare

In reply to Sasha Levin

From: Kangjie Lu <[email protected]>

[ Upstream commit 2d822f2dbab7f4c820f72eb8570aacf3f35855bd ]

clk_prepare() could fail, so let's check its status, and if it fails, return its error code upstream.

---
 drivers/net/ethernet/ti/cpts.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/net/ethernet/ti/cpts.c +3 -1
index b96b93c686bf1..4f644ac314fe8 100644
@@ -572,7 +572,9 @@ struct cpts *cpts_create(struct device *dev, void __iomem *regs,
 		return ERR_CAST(cpts->refclk);
 	}
 
-	clk_prepare(cpts->refclk);
+	ret = clk_prepare(cpts->refclk);
+	if (ret)
+		return ERR_PTR(ret);
 
 	cpts->cc.read = cpts_systim_read;
 	cpts->cc.mask = CLOCKSOURCE_MASK(32);
Signature · 1 line
2.20.1
author maintainer
#147 lore raw

PATCH 146/219 net: stmicro: fix a missing check of clk_prepare

In reply to Sasha Levin

From: Kangjie Lu <[email protected]>

[ Upstream commit f86a3b83833e7cfe558ca4d70b64ebc48903efec ]

clk_prepare() could fail, so let's check its status, and if it fails, return its error code upstream.

---
 drivers/net/ethernet/stmicro/stmmac/dwmac-sunxi.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/net/ethernet/stmicro/stmmac/dwmac-sunxi.c +3 -1
index d07520fb969e6..62ccbd47c1db2 100644
@@ -59,7 +59,9 @@ static int sun7i_gmac_init(struct platform_device *pdev, void *priv)
 		gmac->clk_enabled = 1;
 	} else {
 		clk_set_rate(gmac->tx_clk, SUN7I_GMAC_MII_RATE);
-		clk_prepare(gmac->tx_clk);
+		ret = clk_prepare(gmac->tx_clk);
+		if (ret)
+			return ret;
 	}
 
 	return 0;
Signature · 7 lines
2.20.1


_______________________________________________
linux-arm-kernel mailing list
[email protected]
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
author maintainer
#148 lore raw

PATCH 147/219 net: dsa: bcm_sf2: Propagate error value from mdio_write

In reply to Sasha Levin

From: Kangjie Lu <[email protected]>

[ Upstream commit e49505f7255be8ced695919c08a29bf2c3d79616 ]

Both bcm_sf2_sw_indir_rw and mdiobus_write_nested could fail, so let's return their error codes upstream.

---
 drivers/net/dsa/bcm_sf2.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)
drivers/net/dsa/bcm_sf2.c +3 -4
index ca3655d28e00f..1ffdad278899f 100644
@@ -309,11 +309,10 @@ static int bcm_sf2_sw_mdio_write(struct mii_bus *bus, int addr, int regnum,
 	 * send them to our master MDIO bus controller
 	 */
 	if (addr == BRCM_PSEUDO_PHY_ADDR && priv->indir_phy_mask & BIT(addr))
-		bcm_sf2_sw_indir_rw(priv, 0, addr, regnum, val);
+		return bcm_sf2_sw_indir_rw(priv, 0, addr, regnum, val);
 	else
-		mdiobus_write_nested(priv->master_mii_bus, addr, regnum, val);
-
-	return 0;
+		return mdiobus_write_nested(priv->master_mii_bus, addr,
+				regnum, val);
 }
 
 static irqreturn_t bcm_sf2_switch_0_isr(int irq, void *dev_id)
Signature · 1 line
2.20.1
author maintainer
#149 lore raw

PATCH 148/219 atl1e: checking the status of atl1e_write_phy_reg

In reply to Sasha Levin

From: Kangjie Lu <[email protected]>

[ Upstream commit ff07d48d7bc0974d4f96a85a4df14564fb09f1ef ]

atl1e_write_phy_reg() could fail. The fix issues an error message when it fails.

---
 drivers/net/ethernet/atheros/atl1e/atl1e_main.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/net/ethernet/atheros/atl1e/atl1e_main.c +3 -1
index 9dc6da039a6d9..3164aad29bcf8 100644
@@ -473,7 +473,9 @@ static void atl1e_mdio_write(struct net_device *netdev, int phy_id,
 {
 	struct atl1e_adapter *adapter = netdev_priv(netdev);
 
-	atl1e_write_phy_reg(&adapter->hw, reg_num & MDIO_REG_ADDR_MASK, val);
+	if (atl1e_write_phy_reg(&adapter->hw,
+				reg_num & MDIO_REG_ADDR_MASK, val))
+		netdev_err(netdev, "write phy register failed\n");
 }
 
 static int atl1e_mii_ioctl(struct net_device *netdev,
Signature · 1 line
2.20.1
author maintainer
#150 lore raw

PATCH 149/219 tipc: fix a missing check of genlmsg_put

In reply to Sasha Levin

From: Kangjie Lu <[email protected]>

[ Upstream commit 46273cf7e009231d2b6bc10a926e82b8928a9fb2 ]

genlmsg_put could fail. The fix inserts a check of its return value, and if it fails, returns -EMSGSIZE.

---
 net/tipc/netlink_compat.c | 2 ++
 1 file changed, 2 insertions(+)
net/tipc/netlink_compat.c +2 -0
index 6494d6b5e1b24..fab0384d2b4b3 100644
@@ -983,6 +983,8 @@ static int tipc_nl_compat_publ_dump(struct tipc_nl_compat_msg *msg, u32 sock)
 
 	hdr = genlmsg_put(args, 0, 0, &tipc_genl_family, NLM_F_MULTI,
 			  TIPC_NL_PUBL_GET);
+	if (!hdr)
+		return -EMSGSIZE;
 
 	nest = nla_nest_start(args, TIPC_NLA_SOCK);
 	if (!nest) {
Signature · 1 line
2.20.1
author maintainer
#151 lore raw

PATCH 150/219 net: marvell: fix a missing check of acpi_match_device

In reply to Sasha Levin

From: Kangjie Lu <[email protected]>

[ Upstream commit 92ee77d148bf06d8c52664be4d1b862583fd5c0e ]

When acpi_match_device fails, its return value is NULL. Directly using the return value without a check may result in a NULL-pointer dereference. The fix checks if acpi_match_device fails, and if so, returns -EINVAL.

---
 drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c +2 -0
index 9b608d23ff7ee..9d95add05e3e6 100644
@@ -5131,6 +5131,8 @@ static int mvpp2_probe(struct platform_device *pdev)
 	if (has_acpi_companion(&pdev->dev)) {
 		acpi_id = acpi_match_device(pdev->dev.driver->acpi_match_table,
 					    &pdev->dev);
+		if (!acpi_id)
+			return -EINVAL;
 		priv->hw_version = (unsigned long)acpi_id->driver_data;
 	} else {
 		priv->hw_version =
Signature · 1 line
2.20.1
author maintainer
#152 lore raw

PATCH 151/219 net/wan/fsl_ucc_hdlc: Avoid double free in ucc_hdlc_probe()

In reply to Sasha Levin

From: Wen Yang <[email protected]>

[ Upstream commit 40752b3eae29f8ca2378e978a02bd6dbeeb06d16 ]

This patch fixes potential double frees if register_hdlc_device() fails.

---
 drivers/net/wan/fsl_ucc_hdlc.c | 1 -
 1 file changed, 1 deletion(-)
drivers/net/wan/fsl_ucc_hdlc.c +0 -1
index 5f0366a125e26..0212f576a838c 100644
@@ -1113,7 +1113,6 @@ static int ucc_hdlc_probe(struct platform_device *pdev)
 	if (register_hdlc_device(dev)) {
 		ret = -ENOBUFS;
 		pr_err("ucc_hdlc: unable to register hdlc device\n");
-		free_netdev(dev);
 		goto free_dev;
 	}
 
Signature · 1 line
2.20.1
author maintainer
#153 lore raw

PATCH 152/219 ocfs2: clear journal dirty flag after shutdown journal

In reply to Sasha Levin

From: Junxiao Bi <[email protected]>

[ Upstream commit d85400af790dba2aa294f0a77e712f166681f977 ]

Dirty flag of the journal should be cleared at the last stage of umount, if do it before jbd2_journal_destroy(), then some metadata in uncommitted transaction could be lost due to io error, but as dirty flag of journal was already cleared, we can't find that until run a full fsck. This may cause system panic or other corruption.

---
 fs/ocfs2/journal.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)
fs/ocfs2/journal.c +2 -4
index c492cbb2410f6..4c7e9b8dca316 100644
@@ -1018,7 +1018,8 @@ void ocfs2_journal_shutdown(struct ocfs2_super *osb)
 			mlog_errno(status);
 	}
 
-	if (status == 0) {
+	/* Shutdown the kernel journal system */
+	if (!jbd2_journal_destroy(journal->j_journal) && !status) {
 		/*
 		 * Do not toggle if flush was unsuccessful otherwise
 		 * will leave dirty metadata in a "clean" journal
@@ -1027,9 +1028,6 @@ void ocfs2_journal_shutdown(struct ocfs2_super *osb)
 		if (status < 0)
 			mlog_errno(status);
 	}
-
-	/* Shutdown the kernel journal system */
-	jbd2_journal_destroy(journal->j_journal);
 	journal->j_journal = NULL;
 
 	OCFS2_I(inode)->ip_open_count--;
Signature · 1 line
2.20.1
author maintainer
#154 lore raw

PATCH 153/219 vmscan: return NODE_RECLAIM_NOSCAN in node_reclaim() when CONFIG_NUMA is n

In reply to Sasha Levin

From: Wei Yang <[email protected]>

[ Upstream commit 8b09549c2bfd9f3f8f4cdad74107ef4f4ff9cdd7 ]

Commit fa5e084e43eb ("vmscan: do not unconditionally treat zones that fail zone_reclaim() as full") changed the return value of node_reclaim(). The original return value 0 means NODE_RECLAIM_SOME after this commit.

While the return value of node_reclaim() when CONFIG_NUMA is n is not changed. This will leads to call zone_watermark_ok() again.

This patch fixes the return value by adjusting to NODE_RECLAIM_NOSCAN. Since node_reclaim() is only called in page_alloc.c, move it to mm/internal.h.

---
 include/linux/swap.h |  6 ------
 mm/internal.h        | 10 ++++++++++
 2 files changed, 10 insertions(+), 6 deletions(-)
include/linux/swap.h +0 -6
index 7bd0a6f2ac2b0..ee8f9f554a9e1 100644
@@ -371,14 +371,8 @@ extern unsigned long vm_total_pages;
 extern int node_reclaim_mode;
 extern int sysctl_min_unmapped_ratio;
 extern int sysctl_min_slab_ratio;
-extern int node_reclaim(struct pglist_data *, gfp_t, unsigned int);
 #else
 #define node_reclaim_mode 0
-static inline int node_reclaim(struct pglist_data *pgdat, gfp_t mask,
-				unsigned int order)
-{
-	return 0;
-}
 #endif
 
 extern int page_evictable(struct page *page);
mm/internal.h +10 -0
index 87256ae1bef86..397183c8fe47b 100644
@@ -444,6 +444,16 @@ static inline void mminit_validate_memmodel_limits(unsigned long *start_pfn,
 #define NODE_RECLAIM_SOME	0
 #define NODE_RECLAIM_SUCCESS	1
 
+#ifdef CONFIG_NUMA
+extern int node_reclaim(struct pglist_data *, gfp_t, unsigned int);
+#else
+static inline int node_reclaim(struct pglist_data *pgdat, gfp_t mask,
+				unsigned int order)
+{
+	return NODE_RECLAIM_NOSCAN;
+}
+#endif
+
 extern int hwpoison_filter(struct page *p);
 
 extern u32 hwpoison_filter_dev_major;
Signature · 1 line
2.20.1
author maintainer
#155 lore raw

PATCH 154/219 mm/page_alloc.c: free order-0 pages through PCP in page_frag_free()

In reply to Sasha Levin

From: Aaron Lu <[email protected]>

[ Upstream commit 65895b67ad27df0f62bfaf82dd5622f95ea29196 ]

page_frag_free() calls __free_pages_ok() to free the page back to Buddy. This is OK for high order page, but for order-0 pages, it misses the optimization opportunity of using Per-Cpu-Pages and can cause zone lock contention when called frequently.

Pawel Staszewski recently shared his result of 'how Linux kernel handles normal traffic'[1] and from perf data, Jesper Dangaard Brouer found the lock contention comes from page allocator:

mlx5e_poll_tx_cq
|
--16.34%--napi_consume_skb

             |
             |--12.65%--__free_pages_ok
             |          |
             |           --11.86%--free_one_page
             |                     |
             |                     |--10.10%--queued_spin_lock_slowpath
             |                     |
             |                      --0.65%--_raw_spin_lock
             |
             |--1.55%--page_frag_free
             |
              --1.44%--skb_release_data

Jesper explained how it happened: mlx5 driver RX-page recycle mechanism is not effective in this workload and pages have to go through the page allocator. The lock contention happens during mlx5 DMA TX completion cycle. And the page allocator cannot keep up at these speeds.[2]

I thought that __free_pages_ok() are mostly freeing high order pages and thought this is an lock contention for high order pages but Jesper explained in detail that __free_pages_ok() here are actually freeing order-0 pages because mlx5 is using order-0 pages to satisfy its page pool allocation request.[3]

The free path as pointed out by Jesper is:
skb_free_head()
-> skb_free_frag()

    -> page_frag_free()

And the pages being freed on this path are order-0 pages.

Fix this by doing similar things as in __page_frag_cache_drain() - send the being freed page to PCP if it's an order-0 page, or directly to Buddy if it is a high order page.

With this change, Paweł hasn't noticed lock contention yet in his workload and Jesper has noticed a 7% performance improvement using a micro benchmark and lock contention is gone. Ilias' test on a 'low' speed 1Gbit interface on an cortex-a53 shows ~11% performance boost testing with 64byte packets and __free_pages_ok() disappeared from perf top.

[1]: https://www.spinics.net/lists/netdev/msg531362.html
[2]: https://www.spinics.net/lists/netdev/msg531421.html
[3]: https://www.spinics.net/lists/netdev/msg531556.html

---
 mm/page_alloc.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)
mm/page_alloc.c +8 -2
index b34348a41bfed..dcc46d955df2e 100644
@@ -4581,8 +4581,14 @@ void page_frag_free(void *addr)
 {
 	struct page *page = virt_to_head_page(addr);
 
-	if (unlikely(put_page_testzero(page)))
-		__free_pages_ok(page, compound_order(page));
+	if (unlikely(put_page_testzero(page))) {
+		unsigned int order = compound_order(page);
+
+		if (order == 0)		/* Via pcp? */
+			free_unref_page(page);
+		else
+			__free_pages_ok(page, order);
+	}
 }
 EXPORT_SYMBOL(page_frag_free);
 
Signature · 1 line
2.20.1
author maintainer
#156 lore raw

PATCH 155/219 mm/page_alloc.c: use a single function to free page

In reply to Sasha Levin

From: Aaron Lu <[email protected]>

[ Upstream commit 742aa7fb52c56fb3b307e704f93e67b698959cc2 ]

There are multiple places of freeing a page, they all do the same things so a common function can be used to reduce code duplicate.

It also avoids bug fixed in one function but left in another.

---
 mm/page_alloc.c | 37 ++++++++++++++-----------------------
 1 file changed, 14 insertions(+), 23 deletions(-)
mm/page_alloc.c +14 -23
index dcc46d955df2e..74fb5c338e8fb 100644
@@ -4451,16 +4451,19 @@ unsigned long get_zeroed_page(gfp_t gfp_mask)
 }
 EXPORT_SYMBOL(get_zeroed_page);
 
-void __free_pages(struct page *page, unsigned int order)
+static inline void free_the_page(struct page *page, unsigned int order)
 {
-	if (put_page_testzero(page)) {
-		if (order == 0)
-			free_unref_page(page);
-		else
-			__free_pages_ok(page, order);
-	}
+	if (order == 0)		/* Via pcp? */
+		free_unref_page(page);
+	else
+		__free_pages_ok(page, order);
 }
 
+void __free_pages(struct page *page, unsigned int order)
+{
+	if (put_page_testzero(page))
+		free_the_page(page, order);
+}
 EXPORT_SYMBOL(__free_pages);
 
 void free_pages(unsigned long addr, unsigned int order)
@@ -4509,14 +4512,8 @@ void __page_frag_cache_drain(struct page *page, unsigned int count)
 {
 	VM_BUG_ON_PAGE(page_ref_count(page) == 0, page);
 
-	if (page_ref_sub_and_test(page, count)) {
-		unsigned int order = compound_order(page);
-
-		if (order == 0)
-			free_unref_page(page);
-		else
-			__free_pages_ok(page, order);
-	}
+	if (page_ref_sub_and_test(page, count))
+		free_the_page(page, compound_order(page));
 }
 EXPORT_SYMBOL(__page_frag_cache_drain);
 
@@ -4581,14 +4578,8 @@ void page_frag_free(void *addr)
 {
 	struct page *page = virt_to_head_page(addr);
 
-	if (unlikely(put_page_testzero(page))) {
-		unsigned int order = compound_order(page);
-
-		if (order == 0)		/* Via pcp? */
-			free_unref_page(page);
-		else
-			__free_pages_ok(page, order);
-	}
+	if (unlikely(put_page_testzero(page)))
+		free_the_page(page, compound_order(page));
 }
 EXPORT_SYMBOL(page_frag_free);
 
Signature · 1 line
2.20.1
author maintainer
#157 lore raw

PATCH 156/219 mm/page_alloc.c: deduplicate __memblock_free_early() and memblock_free()

In reply to Sasha Levin

From: Wentao Wang <[email protected]>

[ Upstream commit d31cfe7bff9109476da92c245b56083e9b48d60a ]

---
 mm/memblock.c | 7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)
mm/memblock.c +1 -6
index 237944479d25a..bb4e32c6b19e9 100644
@@ -1537,12 +1537,7 @@ void * __init memblock_virt_alloc_try_nid(
  */
 void __init __memblock_free_early(phys_addr_t base, phys_addr_t size)
 {
-	phys_addr_t end = base + size - 1;
-
-	memblock_dbg("%s: [%pa-%pa] %pF\n",
-		     __func__, &base, &end, (void *)_RET_IP_);
-	kmemleak_free_part_phys(base, size);
-	memblock_remove_range(&memblock.reserved, base, size);
+	memblock_free(base, size);
 }
 
 /**
Signature · 1 line
2.20.1
author maintainer
#158 lore raw

PATCH 157/219 tools/vm/page-types.c: fix "kpagecount returned fewer pages than expected" failures

In reply to Sasha Levin

From: Anthony Yznaga <[email protected]>

[ Upstream commit b6fb87b8e3ff1ef6bcf68470f24a97c984554d5a ]

Because kpagecount_read() fakes success if map counts are not being collected, clamp the page count passed to it by walk_pfn() to the pages value returned by the preceding call to kpageflags_read().

Fixes 7f1d23e60718 ("tools/vm/page-types.c: include shared map counts")
Cc Vlastimil Babka <[email protected]>
Cc Matthew Wilcox <[email protected]>
Cc David Rientjes <[email protected]>
---
 tools/vm/page-types.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
tools/vm/page-types.c +1 -1
index 37908a83ddc27..1ff3a6c0367b0 100644
@@ -701,7 +701,7 @@ static void walk_pfn(unsigned long voffset,
 		if (kpagecgroup_read(cgi, index, pages) != pages)
 			fatal("kpagecgroup returned fewer pages than expected");
 
-		if (kpagecount_read(cnt, index, batch) != pages)
+		if (kpagecount_read(cnt, index, pages) != pages)
 			fatal("kpagecount returned fewer pages than expected");
 
 		for (i = 0; i < pages; i++)
Signature · 1 line
2.20.1
author maintainer
#159 lore raw

PATCH 158/219 netfilter: nf_tables: fix a missing check of nla_put_failure

In reply to Sasha Levin

From: Kangjie Lu <[email protected]>

[ Upstream commit eb8950861c1bfd3eecc8f6faad213e3bca0dc395 ]

If nla_nest_start() may fail. The fix checks its return value and goes to nla_put_failure if it fails.

---
 net/netfilter/nf_tables_api.c | 2 ++
 1 file changed, 2 insertions(+)
net/netfilter/nf_tables_api.c +2 -0
index 24fddf0322790..4c1a88b7ba285 100644
@@ -5735,6 +5735,8 @@ static int nf_tables_fill_flowtable_info(struct sk_buff *skb, struct net *net,
 		goto nla_put_failure;
 
 	nest = nla_nest_start(skb, NFTA_FLOWTABLE_HOOK);
+	if (!nest)
+		goto nla_put_failure;
 	if (nla_put_be32(skb, NFTA_FLOWTABLE_HOOK_NUM, htonl(flowtable->hooknum)) ||
 	    nla_put_be32(skb, NFTA_FLOWTABLE_HOOK_PRIORITY, htonl(flowtable->priority)))
 		goto nla_put_failure;
Signature · 1 line
2.20.1
author maintainer
#160 lore raw

PATCH 159/219 xprtrdma: Prevent leak of rpcrdma_rep objects

In reply to Sasha Levin

From: Chuck Lever <[email protected]>

[ Upstream commit 07e10308ee5da8e6132e0b737ece1c99dd651fb6 ]

If a reply has been processed but the RPC is later retransmitted anyway, the req->rl_reply field still contains the only pointer to the old rpcrdma rep. When the next reply comes in, the reply handler will stomp on the rl_reply field, leaking the old rep.

A trace event is added to capture such leaks.

This problem seems to be worsened by the restructuring of the RPC Call path in v4.20. Fully addressing this issue will require at least a re-architecture of the disconnect logic, which is not appropriate during -rc.

---
 include/trace/events/rpcrdma.h | 28 ++++++++++++++++++++++++++++
 net/sunrpc/xprtrdma/rpc_rdma.c |  4 ++++
 2 files changed, 32 insertions(+)
include/trace/events/rpcrdma.h +28 -0
index 53df203b8057a..4c91cadd1871d 100644
@@ -917,6 +917,34 @@ TRACE_EVENT(xprtrdma_cb_setup,
 DEFINE_CB_EVENT(xprtrdma_cb_call);
 DEFINE_CB_EVENT(xprtrdma_cb_reply);
 
+TRACE_EVENT(xprtrdma_leaked_rep,
+	TP_PROTO(
+		const struct rpc_rqst *rqst,
+		const struct rpcrdma_rep *rep
+	),
+
+	TP_ARGS(rqst, rep),
+
+	TP_STRUCT__entry(
+		__field(unsigned int, task_id)
+		__field(unsigned int, client_id)
+		__field(u32, xid)
+		__field(const void *, rep)
+	),
+
+	TP_fast_assign(
+		__entry->task_id = rqst->rq_task->tk_pid;
+		__entry->client_id = rqst->rq_task->tk_client->cl_clid;
+		__entry->xid = be32_to_cpu(rqst->rq_xid);
+		__entry->rep = rep;
+	),
+
+	TP_printk("task:%u@%u xid=0x%08x rep=%p",
+		__entry->task_id, __entry->client_id, __entry->xid,
+		__entry->rep
+	)
+);
+
 /**
  ** Server-side RPC/RDMA events
  **/
net/sunrpc/xprtrdma/rpc_rdma.c +4 -0
index c8ae983c6cc01..f2eaf264726be 100644
@@ -1360,6 +1360,10 @@ void rpcrdma_reply_handler(struct rpcrdma_rep *rep)
 	spin_unlock(&xprt->recv_lock);
 
 	req = rpcr_to_rdmar(rqst);
+	if (req->rl_reply) {
+		trace_xprtrdma_leaked_rep(rqst, req->rl_reply);
+		rpcrdma_recv_buffer_put(req->rl_reply);
+	}
 	req->rl_reply = rep;
 	rep->rr_rqst = rqst;
 	clear_bit(RPCRDMA_REQ_F_PENDING, &req->rl_flags);
Signature · 1 line
2.20.1
author maintainer
#161 lore raw

PATCH 160/219 infiniband: bnxt_re: qplib: Check the return value of send_message

In reply to Sasha Levin

From: Aditya Pakki <[email protected]>

[ Upstream commit 94edd87a1c59f3efa6fdf4e98d6d492e6cec6173 ]

In bnxt_qplib_map_tc2cos(), bnxt_qplib_rcfw_send_message() can return an error value but it is lost. Propagate this error to the callers.

---
 drivers/infiniband/hw/bnxt_re/qplib_sp.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
drivers/infiniband/hw/bnxt_re/qplib_sp.c +2 -3
index 4097f3fa25c5f..09e7d3dd30553 100644
@@ -775,9 +775,8 @@ int bnxt_qplib_map_tc2cos(struct bnxt_qplib_res *res, u16 *cids)
 	req.cos0 = cpu_to_le16(cids[0]);
 	req.cos1 = cpu_to_le16(cids[1]);
 
-	bnxt_qplib_rcfw_send_message(rcfw, (void *)&req, (void *)&resp, NULL,
-				     0);
-	return 0;
+	return bnxt_qplib_rcfw_send_message(rcfw, (void *)&req, (void *)&resp,
+						NULL, 0);
 }
 
 int bnxt_qplib_get_roce_stats(struct bnxt_qplib_rcfw *rcfw,
Signature · 1 line
2.20.1
author maintainer
#162 lore raw

PATCH 161/219 infiniband/qedr: Potential null ptr dereference of qp

In reply to Sasha Levin

From: Aditya Pakki <[email protected]>

[ Upstream commit 9c6260de505b63638dd86fcc33849b17f6146d94 ]

idr_find() may fail and return a NULL pointer. The fix checks the return value of the function and returns an error in case of NULL.

---
 drivers/infiniband/hw/qedr/qedr_iw_cm.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/infiniband/hw/qedr/qedr_iw_cm.c +2 -0
index 505fa36487629..93b16237b7677 100644
@@ -492,6 +492,8 @@ int qedr_iw_connect(struct iw_cm_id *cm_id, struct iw_cm_conn_param *conn_param)
 	int i;
 
 	qp = idr_find(&dev->qpidr.idr, conn_param->qpn);
+	if (unlikely(!qp))
+		return -EINVAL;
 
 	laddr = (struct sockaddr_in *)&cm_id->m_local_addr;
 	raddr = (struct sockaddr_in *)&cm_id->m_remote_addr;
Signature · 1 line
2.20.1
author maintainer
#163 lore raw

PATCH 162/219 firmware: arm_sdei: fix wrong of_node_put() in init function

In reply to Sasha Levin

From: Nicolas Saenz Julienne <[email protected]>

[ Upstream commit c3790b3799f8d75d93d26f6fd7bb569fc8c8b0cb ]

After finding a "firmware" dt node arm_sdei tries to match it's compatible string with it. To do so it's calling of_find_matching_node() which already takes care of decreasing the refcount on the "firmware" node. We are then incorrectly decreasing the refcount on that node again.

This patch removes the unwarranted call to of_node_put().

---
 drivers/firmware/arm_sdei.c | 1 -
 1 file changed, 1 deletion(-)
drivers/firmware/arm_sdei.c +0 -1
index 1ea71640fdc21..dffb47c6b4801 100644
@@ -1017,7 +1017,6 @@ static bool __init sdei_present_dt(void)
 		return false;
 
 	np = of_find_matching_node(fw_np, sdei_of_match);
-	of_node_put(fw_np);
 	if (!np)
 		return false;
 
Signature · 1 line
2.20.1
author maintainer
#164 lore raw

PATCH 163/219 firmware: arm_sdei: Fix DT platform device creation

In reply to Sasha Levin

From: James Morse <[email protected]>

[ Upstream commit acafce48b07bf5f9994a38e7fe237193d43d092e ]

It turns out the dt-probing part of this wasn't tested properly after it was merged. commit 3aa0582fdb82 ("of: platform: populate /firmware/ node from of_platform_default_populate_init()") changed the core-code to generate the platform devices, meaning the driver's attempt fails, and it bails out.

Fix this by removing the manual platform-device creation for DT systems, core code has always done this for us.

---
 drivers/firmware/arm_sdei.c | 5 -----
 1 file changed, 5 deletions(-)
drivers/firmware/arm_sdei.c +0 -5
index dffb47c6b4801..c64c7da738297 100644
@@ -1009,7 +1009,6 @@ static struct platform_driver sdei_driver = {
 
 static bool __init sdei_present_dt(void)
 {
-	struct platform_device *pdev;
 	struct device_node *np, *fw_np;
 
 	fw_np = of_find_node_by_name(NULL, "firmware");
@@ -1019,11 +1018,7 @@ static bool __init sdei_present_dt(void)
 	np = of_find_matching_node(fw_np, sdei_of_match);
 	if (!np)
 		return false;
-
-	pdev = of_platform_device_create(np, sdei_driver.driver.name, NULL);
 	of_node_put(np);
-	if (!pdev)
-		return false;
 
 	return true;
 }
Signature · 1 line
2.20.1
author maintainer
#165 lore raw

PATCH 164/219 lib/genalloc.c: fix allocation of aligned buffer from non-aligned chunk

In reply to Sasha Levin

From: Alexey Skidanov <[email protected]>

[ Upstream commit 52fbf1134d479234d7e64ba9dcbaea23405f229e ]

gen_pool_alloc_algo() uses different allocation functions implementing different allocation algorithms. With gen_pool_first_fit_align() allocation function, the returned address should be aligned on the requested boundary.

If chunk start address isn't aligned on the requested boundary, the returned address isn't aligned too. The only way to get properly aligned address is to initialize the pool with chunks aligned on the requested boundary. If want to have an ability to allocate buffers aligned on different boundaries (for example, 4K, 1MB, ...), the chunk start address should be aligned on the max possible alignment.

This happens because gen_pool_first_fit_align() looks for properly aligned memory block without taking into account the chunk start address alignment.

To fix this, we provide chunk start address to gen_pool_first_fit_align() and change its implementation such that it starts looking for properly aligned block with appropriate offset (exactly as is done in CMA).

---
 include/linux/genalloc.h | 13 +++++++------
 lib/genalloc.c           | 20 ++++++++++++--------
 2 files changed, 19 insertions(+), 14 deletions(-)
include/linux/genalloc.h +7 -6
index 872f930f1b06d..dd0a452373e71 100644
@@ -51,7 +51,8 @@ typedef unsigned long (*genpool_algo_t)(unsigned long *map,
 			unsigned long size,
 			unsigned long start,
 			unsigned int nr,
-			void *data, struct gen_pool *pool);
+			void *data, struct gen_pool *pool,
+			unsigned long start_addr);
 
 /*
  *  General purpose special memory pool descriptor.
@@ -131,24 +132,24 @@ extern void gen_pool_set_algo(struct gen_pool *pool, genpool_algo_t algo,
 
 extern unsigned long gen_pool_first_fit(unsigned long *map, unsigned long size,
 		unsigned long start, unsigned int nr, void *data,
-		struct gen_pool *pool);
+		struct gen_pool *pool, unsigned long start_addr);
 
 extern unsigned long gen_pool_fixed_alloc(unsigned long *map,
 		unsigned long size, unsigned long start, unsigned int nr,
-		void *data, struct gen_pool *pool);
+		void *data, struct gen_pool *pool, unsigned long start_addr);
 
 extern unsigned long gen_pool_first_fit_align(unsigned long *map,
 		unsigned long size, unsigned long start, unsigned int nr,
-		void *data, struct gen_pool *pool);
+		void *data, struct gen_pool *pool, unsigned long start_addr);
 
 
 extern unsigned long gen_pool_first_fit_order_align(unsigned long *map,
 		unsigned long size, unsigned long start, unsigned int nr,
-		void *data, struct gen_pool *pool);
+		void *data, struct gen_pool *pool, unsigned long start_addr);
 
 extern unsigned long gen_pool_best_fit(unsigned long *map, unsigned long size,
 		unsigned long start, unsigned int nr, void *data,
-		struct gen_pool *pool);
+		struct gen_pool *pool, unsigned long start_addr);
 
 
 extern struct gen_pool *devm_gen_pool_create(struct device *dev,
lib/genalloc.c +12 -8
index ca06adc4f4451..5deb25c40a5a1 100644
@@ -311,7 +311,7 @@ unsigned long gen_pool_alloc_algo(struct gen_pool *pool, size_t size,
 		end_bit = chunk_size(chunk) >> order;
 retry:
 		start_bit = algo(chunk->bits, end_bit, start_bit,
-				 nbits, data, pool);
+				 nbits, data, pool, chunk->start_addr);
 		if (start_bit >= end_bit)
 			continue;
 		remain = bitmap_set_ll(chunk->bits, start_bit, nbits);
@@ -525,7 +525,7 @@ EXPORT_SYMBOL(gen_pool_set_algo);
  */
 unsigned long gen_pool_first_fit(unsigned long *map, unsigned long size,
 		unsigned long start, unsigned int nr, void *data,
-		struct gen_pool *pool)
+		struct gen_pool *pool, unsigned long start_addr)
 {
 	return bitmap_find_next_zero_area(map, size, start, nr, 0);
 }
@@ -543,16 +543,19 @@ EXPORT_SYMBOL(gen_pool_first_fit);
  */
 unsigned long gen_pool_first_fit_align(unsigned long *map, unsigned long size,
 		unsigned long start, unsigned int nr, void *data,
-		struct gen_pool *pool)
+		struct gen_pool *pool, unsigned long start_addr)
 {
 	struct genpool_data_align *alignment;
-	unsigned long align_mask;
+	unsigned long align_mask, align_off;
 	int order;
 
 	alignment = data;
 	order = pool->min_alloc_order;
 	align_mask = ((alignment->align + (1UL << order) - 1) >> order) - 1;
-	return bitmap_find_next_zero_area(map, size, start, nr, align_mask);
+	align_off = (start_addr & (alignment->align - 1)) >> order;
+
+	return bitmap_find_next_zero_area_off(map, size, start, nr,
+					      align_mask, align_off);
 }
 EXPORT_SYMBOL(gen_pool_first_fit_align);
 
@@ -567,7 +570,7 @@ EXPORT_SYMBOL(gen_pool_first_fit_align);
  */
 unsigned long gen_pool_fixed_alloc(unsigned long *map, unsigned long size,
 		unsigned long start, unsigned int nr, void *data,
-		struct gen_pool *pool)
+		struct gen_pool *pool, unsigned long start_addr)
 {
 	struct genpool_data_fixed *fixed_data;
 	int order;
@@ -601,7 +604,8 @@ EXPORT_SYMBOL(gen_pool_fixed_alloc);
  */
 unsigned long gen_pool_first_fit_order_align(unsigned long *map,
 		unsigned long size, unsigned long start,
-		unsigned int nr, void *data, struct gen_pool *pool)
+		unsigned int nr, void *data, struct gen_pool *pool,
+		unsigned long start_addr)
 {
 	unsigned long align_mask = roundup_pow_of_two(nr) - 1;
 
@@ -624,7 +628,7 @@ EXPORT_SYMBOL(gen_pool_first_fit_order_align);
  */
 unsigned long gen_pool_best_fit(unsigned long *map, unsigned long size,
 		unsigned long start, unsigned int nr, void *data,
-		struct gen_pool *pool)
+		struct gen_pool *pool, unsigned long start_addr)
 {
 	unsigned long start_bit = size;
 	unsigned long len = size + 1;
Signature · 1 line
2.20.1
author maintainer
#166 lore raw

PATCH 165/219 lib/genalloc.c: use vzalloc_node() to allocate the bitmap

In reply to Sasha Levin

From: Huang Shijie <[email protected]>

[ Upstream commit 6862d2fc81859f88c1f3f660886427893f2b4f3f ]

Some devices may have big memory on chip, such as over 1G. In some cases, the nbytes maybe bigger then 4M which is the bounday of the memory buddy system (4K default).

So use vzalloc_node() to allocate the bitmap. Also use vfree to free it.

---
 lib/genalloc.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
lib/genalloc.c +2 -2
index 5deb25c40a5a1..f365d71cdc774 100644
@@ -187,7 +187,7 @@ int gen_pool_add_virt(struct gen_pool *pool, unsigned long virt, phys_addr_t phy
 	int nbytes = sizeof(struct gen_pool_chunk) +
 				BITS_TO_LONGS(nbits) * sizeof(long);
 
-	chunk = kzalloc_node(nbytes, GFP_KERNEL, nid);
+	chunk = vzalloc_node(nbytes, nid);
 	if (unlikely(chunk == NULL))
 		return -ENOMEM;
 
@@ -251,7 +251,7 @@ void gen_pool_destroy(struct gen_pool *pool)
 		bit = find_next_bit(chunk->bits, end_bit, 0);
 		BUG_ON(bit < end_bit);
 
-		kfree(chunk);
+		vfree(chunk);
 	}
 	kfree_const(pool->name);
 	kfree(pool);
Signature · 1 line
2.20.1
author maintainer
#167 lore raw

PATCH 166/219 fork: fix some -Wmissing-prototypes warnings

In reply to Sasha Levin

From: Yi Wang <[email protected]>

[ Upstream commit fb5bf31722d0805a3f394f7d59f2e8cd07acccb7 ]

We get a warning when building kernel with W=1:

kernel/fork.c:167:13: warning: no previous prototype for arch_release_thread_stack' [-Wmissing-prototypes] kernel/fork.c:779:13: warning: no previous prototype for fork_init' [-Wmissing-prototypes]

Add the missing declaration in head file to fix this.

Also, remove arch_release_thread_stack() completely because no arch seems to implement it since bb9d81264 (arch: remove tile port).

---
 include/linux/sched/task.h | 2 ++
 init/main.c                | 1 -
 kernel/fork.c              | 5 -----
 3 files changed, 2 insertions(+), 6 deletions(-)
include/linux/sched/task.h +2 -0
index 108ede99e5335..44c6f15800ff5 100644
@@ -39,6 +39,8 @@ void __noreturn do_task_dead(void);
 
 extern void proc_caches_init(void);
 
+extern void fork_init(void);
+
 extern void release_task(struct task_struct * p);
 
 #ifdef CONFIG_HAVE_COPY_THREAD_TLS
init/main.c +0 -1
index 020972fed1171..38a603f62b7bb 100644
@@ -105,7 +105,6 @@
 static int kernel_init(void *);
 
 extern void init_IRQ(void);
-extern void fork_init(void);
 extern void radix_tree_init(void);
 
 /*
kernel/fork.c +0 -5
index aef1430bdce0b..8cb5cd7c97e19 100644
@@ -163,10 +163,6 @@ static inline void free_task_struct(struct task_struct *tsk)
 }
 #endif
 
-void __weak arch_release_thread_stack(unsigned long *stack)
-{
-}
-
 #ifndef CONFIG_ARCH_THREAD_STACK_ALLOCATOR
 
 /*
@@ -376,7 +372,6 @@ static void release_task_stack(struct task_struct *tsk)
 		return;  /* Better to leak the stack than to free prematurely */
 
 	account_kernel_stack(tsk, -1);
-	arch_release_thread_stack(tsk->stack);
 	free_thread_stack(tsk);
 	tsk->stack = NULL;
 #ifdef CONFIG_VMAP_STACK
Signature · 1 line
2.20.1
author maintainer
#168 lore raw

PATCH 167/219 drivers/base/platform.c: kmemleak ignore a known leak

In reply to Sasha Levin

From: Qian Cai <[email protected]>

[ Upstream commit 967d3010df8b6f6f9aa95c198edc5fe3646ebf36 ]

unreferenced object 0xffff808ec6dc5a80 (size 128):
comm "swapper/0", pid 1, jiffies 4294938063 (age 2560.530s) hex dump (first 32 bytes):
ff ff ff ff 00 00 00 00 6b 6b 6b 6b 6b 6b 6b 6b ........kkkkkkkk 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b 6b kkkkkkkkkkkkkkkk
backtrace:

    [<00000000476dcf8c>] kmem_cache_alloc_trace+0x430/0x500
    [<000000004f708d37>] platform_device_register_full+0xbc/0x1e8
    [<000000006c2a7ec7>] acpi_create_platform_device+0x370/0x450
    [<00000000ef135642>] acpi_default_enumeration+0x34/0x78
    [<000000003bd9a052>] acpi_bus_attach+0x2dc/0x3e0
    [<000000003cf4f7f2>] acpi_bus_attach+0x108/0x3e0
    [<000000003cf4f7f2>] acpi_bus_attach+0x108/0x3e0
    [<000000002968643e>] acpi_bus_scan+0xb0/0x110
    [<0000000010dd0bd7>] acpi_scan_init+0x1a8/0x410
    [<00000000965b3c5a>] acpi_init+0x408/0x49c
    [<00000000ed4b9fe2>] do_one_initcall+0x178/0x7f4
    [<00000000a5ac5a74>] kernel_init_freeable+0x9d4/0xa9c
    [<0000000070ea6c15>] kernel_init+0x18/0x138
    [<00000000fb8fff06>] ret_from_fork+0x10/0x1c
    [<0000000041273a0d>] 0xffffffffffffffff

Then, faddr2line pointed out this line,

/*
* This memory isn't freed when the device is put,
* I don't have a nice idea for that though. Conceptually
* dma_mask in struct device should not be a pointer.
* See http://thread.gmane.org/gmane.linux.kernel.pci/9081
*/
pdev->dev.dma_mask =

	kmalloc(sizeof(*pdev->dev.dma_mask), GFP_KERNEL);

Since this leak has existed for more than 8 years and it does not reference other parts of the memory, let kmemleak ignore it, so users don't need to waste time reporting this in the future.

---
 drivers/base/platform.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/base/platform.c +3 -0
index dff82a3c2caa9..e9be1f56929af 100644
@@ -26,6 +26,7 @@
 #include <linux/clk/clk-conf.h>
 #include <linux/limits.h>
 #include <linux/property.h>
+#include <linux/kmemleak.h>
 
 #include "base.h"
 #include "power/power.h"
@@ -525,6 +526,8 @@ struct platform_device *platform_device_register_full(
 		if (!pdev->dev.dma_mask)
 			goto err;
 
+		kmemleak_ignore(pdev->dev.dma_mask);
+
 		*pdev->dev.dma_mask = pdevinfo->dma_mask;
 		pdev->dev.coherent_dma_mask = pdevinfo->dma_mask;
 	}
Signature · 1 line
2.20.1
author maintainer
#169 lore raw

PATCH 168/219 lib/genalloc.c: include vmalloc.h

In reply to Sasha Levin

From: Olof Johansson <[email protected]>

[ Upstream commit 35004f2e55807a1a1491db24ab512dd2f770a130 ]

Fixes build break on most ARM/ARM64 defconfigs:

lib/genalloc.c: In function 'gen_pool_add_virt':
lib/genalloc.c:190:10: error: implicit declaration of function 'vzalloc_node'; did you mean 'kzalloc_node'? lib/genalloc.c:190:8: warning: assignment to 'struct gen_pool_chunk *' from 'int' makes pointer from integer without a cast [-Wint-conversion] lib/genalloc.c: In function 'gen_pool_destroy':
lib/genalloc.c:254:3: error: implicit declaration of function 'vfree'; did you mean 'kfree'?

Fixes 6862d2fc8185 ('lib/genalloc.c: use vzalloc_node() to allocate the bitmap')
Cc Huang Shijie <[email protected]>
Cc Andrew Morton <[email protected]>
Cc Alexey Skidanov <[email protected]>
---
 lib/genalloc.c | 1 +
 1 file changed, 1 insertion(+)
lib/genalloc.c +1 -0
index f365d71cdc774..7e85d1e37a6ea 100644
@@ -35,6 +35,7 @@
 #include <linux/interrupt.h>
 #include <linux/genalloc.h>
 #include <linux/of_device.h>
+#include <linux/vmalloc.h>
 
 static inline size_t chunk_size(const struct gen_pool_chunk *chunk)
 {
Signature · 1 line
2.20.1
author maintainer
#170 lore raw

PATCH 169/219 mtd: Check add_mtd_device() ret code

In reply to Sasha Levin

From: Boris Brezillon <[email protected]>

[ Upstream commit 2b6f0090a3335b7bdd03ca520c35591159463041 ]

add_mtd_device() can fail. We should always check its return value and gracefully handle the failure case. Fix the call sites where this not done (in mtdpart.c) and add a __must_check attribute to the prototype to avoid this kind of mistakes.

---
 drivers/mtd/mtdcore.h |  2 +-
 drivers/mtd/mtdpart.c | 36 +++++++++++++++++++++++++++++++-----
 2 files changed, 32 insertions(+), 6 deletions(-)
drivers/mtd/mtdcore.h +1 -1
index 9887bda317cd9..b31c868019ada 100644
@@ -7,7 +7,7 @@
 extern struct mutex mtd_table_mutex;
 
 struct mtd_info *__mtd_next_device(int i);
-int add_mtd_device(struct mtd_info *mtd);
+int __must_check add_mtd_device(struct mtd_info *mtd);
 int del_mtd_device(struct mtd_info *mtd);
 int add_mtd_partitions(struct mtd_info *, const struct mtd_partition *, int);
 int del_mtd_partitions(struct mtd_info *);
drivers/mtd/mtdpart.c +31 -5
index 0bbb23b014f1b..b9b308cde0a5d 100644
@@ -612,10 +612,22 @@ int mtd_add_partition(struct mtd_info *parent, const char *name,
 	list_add(&new->list, &mtd_partitions);
 	mutex_unlock(&mtd_partitions_mutex);
 
-	add_mtd_device(&new->mtd);
+	ret = add_mtd_device(&new->mtd);
+	if (ret)
+		goto err_remove_part;
 
 	mtd_add_partition_attrs(new);
 
+	return 0;
+
+err_remove_part:
+	mutex_lock(&mtd_partitions_mutex);
+	list_del(&new->list);
+	mutex_unlock(&mtd_partitions_mutex);
+
+	free_partition(new);
+	pr_info("%s:%i\n", __func__, __LINE__);
+
 	return ret;
 }
 EXPORT_SYMBOL_GPL(mtd_add_partition);
@@ -706,22 +718,31 @@ int add_mtd_partitions(struct mtd_info *master,
 {
 	struct mtd_part *slave;
 	uint64_t cur_offset = 0;
-	int i;
+	int i, ret;
 
 	printk(KERN_NOTICE "Creating %d MTD partitions on \"%s\":\n", nbparts, master->name);
 
 	for (i = 0; i < nbparts; i++) {
 		slave = allocate_partition(master, parts + i, i, cur_offset);
 		if (IS_ERR(slave)) {
-			del_mtd_partitions(master);
-			return PTR_ERR(slave);
+			ret = PTR_ERR(slave);
+			goto err_del_partitions;
 		}
 
 		mutex_lock(&mtd_partitions_mutex);
 		list_add(&slave->list, &mtd_partitions);
 		mutex_unlock(&mtd_partitions_mutex);
 
-		add_mtd_device(&slave->mtd);
+		ret = add_mtd_device(&slave->mtd);
+		if (ret) {
+			mutex_lock(&mtd_partitions_mutex);
+			list_del(&slave->list);
+			mutex_unlock(&mtd_partitions_mutex);
+
+			free_partition(slave);
+			goto err_del_partitions;
+		}
+
 		mtd_add_partition_attrs(slave);
 		/* Look for subpartitions */
 		parse_mtd_partitions(&slave->mtd, parts[i].types, NULL);
@@ -730,6 +751,11 @@ int add_mtd_partitions(struct mtd_info *master,
 	}
 
 	return 0;
+
+err_del_partitions:
+	del_mtd_partitions(master);
+
+	return ret;
 }
 
 static DEFINE_SPINLOCK(part_parser_lock);
Signature · 6 lines
2.20.1


______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/
author maintainer
#171 lore raw

PATCH 170/219 tipc: fix memory leak in tipc_nl_compat_publ_dump

In reply to Sasha Levin

From: "Gustavo A. R. Silva" <[email protected]>

[ Upstream commit f87d8ad9233f115db92c6c087d58403b0009ed36 ]

There is a memory leak in case genlmsg_put fails.

Fix this by freeing *args* before return.

Addresses-Coverity-ID: 1476406 ("Resource leak")

Fixes 46273cf7e009 ("tipc: fix a missing check of genlmsg_put")
---
 net/tipc/netlink_compat.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
net/tipc/netlink_compat.c +3 -1
index fab0384d2b4b3..b4860084a8f56 100644
@@ -983,8 +983,10 @@ static int tipc_nl_compat_publ_dump(struct tipc_nl_compat_msg *msg, u32 sock)
 
 	hdr = genlmsg_put(args, 0, 0, &tipc_genl_family, NLM_F_MULTI,
 			  TIPC_NL_PUBL_GET);
-	if (!hdr)
+	if (!hdr) {
+		kfree_skb(args);
 		return -EMSGSIZE;
+	}
 
 	nest = nla_nest_start(args, TIPC_NLA_SOCK);
 	if (!nest) {
Signature · 1 line
2.20.1
author maintainer
#172 lore raw

PATCH 171/219 net/core/neighbour: tell kmemleak about hash tables

In reply to Sasha Levin

From: Konstantin Khlebnikov <[email protected]>

[ Upstream commit 85704cb8dcfd88d351bfc87faaeba1c8214f3177 ]

This fixes false-positive kmemleak reports about leaked neighbour entries:

unreferenced object 0xffff8885c6e4d0a8 (size 1024):
comm "softirq", pid 0, jiffies 4294922664 (age 167640.804s) hex dump (first 32 bytes):

    00 00 00 00 00 00 00 00 20 2c f3 83 ff ff ff ff  ........ ,......
    08 c0 ef 5f 84 88 ff ff 01 8c 7d 02 01 00 00 00  ..._......}.....
  backtrace:
    [<00000000748509fe>] ip6_finish_output2+0x887/0x1e40
    [<0000000036d7a0d8>] ip6_output+0x1ba/0x600
    [<0000000027ea7dba>] ip6_send_skb+0x92/0x2f0
    [<00000000d6e2111d>] udp_v6_send_skb.isra.24+0x680/0x15e0
    [<000000000668a8be>] udpv6_sendmsg+0x18c9/0x27a0
    [<000000004bd5fa90>] sock_sendmsg+0xb3/0xf0
    [<000000008227b29f>] ___sys_sendmsg+0x745/0x8f0
    [<000000008698009d>] __sys_sendmsg+0xde/0x170
    [<00000000889dacf1>] do_syscall_64+0x9b/0x400
    [<0000000081cdb353>] entry_SYSCALL_64_after_hwframe+0x49/0xbe
    [<000000005767ed39>] 0xffffffffffffffff
---
 net/core/neighbour.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)
net/core/neighbour.c +9 -4
index c52d6e6b341cf..4721793babed5 100644
@@ -18,6 +18,7 @@
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #include <linux/slab.h>
+#include <linux/kmemleak.h>
 #include <linux/types.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
@@ -363,12 +364,14 @@ static struct neigh_hash_table *neigh_hash_alloc(unsigned int shift)
 	ret = kmalloc(sizeof(*ret), GFP_ATOMIC);
 	if (!ret)
 		return NULL;
-	if (size <= PAGE_SIZE)
+	if (size <= PAGE_SIZE) {
 		buckets = kzalloc(size, GFP_ATOMIC);
-	else
+	} else {
 		buckets = (struct neighbour __rcu **)
 			  __get_free_pages(GFP_ATOMIC | __GFP_ZERO,
 					   get_order(size));
+		kmemleak_alloc(buckets, size, 0, GFP_ATOMIC);
+	}
 	if (!buckets) {
 		kfree(ret);
 		return NULL;
@@ -388,10 +391,12 @@ static void neigh_hash_free_rcu(struct rcu_head *head)
 	size_t size = (1 << nht->hash_shift) * sizeof(struct neighbour *);
 	struct neighbour __rcu **buckets = nht->hash_buckets;
 
-	if (size <= PAGE_SIZE)
+	if (size <= PAGE_SIZE) {
 		kfree(buckets);
-	else
+	} else {
+		kmemleak_free(buckets);
 		free_pages((unsigned long)buckets, get_order(size));
+	}
 	kfree(nht);
 }
 
Signature · 1 line
2.20.1
author maintainer
#173 lore raw

PATCH 172/219 ata: ahci: mvebu: do Armada 38x configuration only on relevant SoCs

In reply to Sasha Levin

From: Miquel Raynal <[email protected]>

[ Upstream commit 96dbcb40e4b1a387cdb9b21f43638c759aebb5a4 ]

At the beginning, only Armada 38x SoCs where supported by the ahci_mvebu.c driver. Commit 15d3ce7b63bd ("ata: ahci_mvebu: add support for Armada 3700 variant") introduced Armada 3700 support. As opposed to Armada 38x SoCs, the 3700 variants do not have to configure mbus and the regret option. This patch took care of avoiding such configuration when not needed in the probe function, but failed to do the same in the resume path. While doing so looks harmless by experience, let's clean the driver logic and avoid doing this useless configuration with Armada 3700 SoCs.

Because the logic is very similar between these two places, it has been decided to factorize this code and put it in a "Armada 38x configuration function". This function is part of a new (per-compatible) platform data structure, so that the addition of such configuration function for Armada 3700 will be eased.

Fixes 15d3ce7b63bd ("ata: ahci_mvebu: add support for Armada 3700 variant")
---
 drivers/ata/ahci_mvebu.c | 68 ++++++++++++++++++++++++++++++----------
 1 file changed, 51 insertions(+), 17 deletions(-)
drivers/ata/ahci_mvebu.c +51 -17
index f9cb51be38ebf..a54214291481f 100644
@@ -28,6 +28,10 @@
 #define AHCI_WINDOW_BASE(win)	(0x64 + ((win) << 4))
 #define AHCI_WINDOW_SIZE(win)	(0x68 + ((win) << 4))
 
+struct ahci_mvebu_plat_data {
+	int (*plat_config)(struct ahci_host_priv *hpriv);
+};
+
 static void ahci_mvebu_mbus_config(struct ahci_host_priv *hpriv,
 				   const struct mbus_dram_target_info *dram)
 {
@@ -62,6 +66,22 @@ static void ahci_mvebu_regret_option(struct ahci_host_priv *hpriv)
 	writel(0x80, hpriv->mmio + AHCI_VENDOR_SPECIFIC_0_DATA);
 }
 
+static int ahci_mvebu_armada_380_config(struct ahci_host_priv *hpriv)
+{
+	const struct mbus_dram_target_info *dram;
+	int rc = 0;
+
+	dram = mv_mbus_dram_info();
+	if (dram)
+		ahci_mvebu_mbus_config(hpriv, dram);
+	else
+		rc = -ENODEV;
+
+	ahci_mvebu_regret_option(hpriv);
+
+	return rc;
+}
+
 /**
  * ahci_mvebu_stop_engine
  *
@@ -126,13 +146,10 @@ static int ahci_mvebu_resume(struct platform_device *pdev)
 {
 	struct ata_host *host = platform_get_drvdata(pdev);
 	struct ahci_host_priv *hpriv = host->private_data;
-	const struct mbus_dram_target_info *dram;
+	const struct ahci_mvebu_plat_data *pdata = hpriv->plat_data;
 
-	dram = mv_mbus_dram_info();
-	if (dram)
-		ahci_mvebu_mbus_config(hpriv, dram);
-
-	ahci_mvebu_regret_option(hpriv);
+	if (pdata->plat_config)
+		pdata->plat_config(hpriv);
 
 	return ahci_platform_resume_host(&pdev->dev);
 }
@@ -154,28 +171,31 @@ static struct scsi_host_template ahci_platform_sht = {
 
 static int ahci_mvebu_probe(struct platform_device *pdev)
 {
+	const struct ahci_mvebu_plat_data *pdata;
 	struct ahci_host_priv *hpriv;
-	const struct mbus_dram_target_info *dram;
 	int rc;
 
+	pdata = of_device_get_match_data(&pdev->dev);
+	if (!pdata)
+		return -EINVAL;
+
 	hpriv = ahci_platform_get_resources(pdev, 0);
 	if (IS_ERR(hpriv))
 		return PTR_ERR(hpriv);
 
+	hpriv->plat_data = (void *)pdata;
+
 	rc = ahci_platform_enable_resources(hpriv);
 	if (rc)
 		return rc;
 
 	hpriv->stop_engine = ahci_mvebu_stop_engine;
 
-	if (of_device_is_compatible(pdev->dev.of_node,
-				    "marvell,armada-380-ahci")) {
-		dram = mv_mbus_dram_info();
-		if (!dram)
-			return -ENODEV;
-
-		ahci_mvebu_mbus_config(hpriv, dram);
-		ahci_mvebu_regret_option(hpriv);
+	pdata = hpriv->plat_data;
+	if (pdata->plat_config) {
+		rc = pdata->plat_config(hpriv);
+		if (rc)
+			goto disable_resources;
 	}
 
 	rc = ahci_platform_init_host(pdev, hpriv, &ahci_mvebu_port_info,
@@ -190,9 +210,23 @@ static int ahci_mvebu_probe(struct platform_device *pdev)
 	return rc;
 }
 
+static const struct ahci_mvebu_plat_data ahci_mvebu_armada_380_plat_data = {
+	.plat_config = ahci_mvebu_armada_380_config,
+};
+
+static const struct ahci_mvebu_plat_data ahci_mvebu_armada_3700_plat_data = {
+	.plat_config = NULL,
+};
+
 static const struct of_device_id ahci_mvebu_of_match[] = {
-	{ .compatible = "marvell,armada-380-ahci", },
-	{ .compatible = "marvell,armada-3700-ahci", },
+	{
+		.compatible = "marvell,armada-380-ahci",
+		.data = &ahci_mvebu_armada_380_plat_data,
+	},
+	{
+		.compatible = "marvell,armada-3700-ahci",
+		.data = &ahci_mvebu_armada_3700_plat_data,
+	},
 	{ },
 };
 MODULE_DEVICE_TABLE(of, ahci_mvebu_of_match);
Signature · 1 line
2.20.1