6.6.30-rc1 review

Signed-off-by Sasha Levin <[email protected]>

---
 fs/smb/client/cifsglob.h |  2 ++
 fs/smb/client/fscache.c  | 13 +++++++++++++
 fs/smb/client/misc.c     |  3 +++
 3 files changed, 18 insertions(+)

fs/smb/client/cifsglob.h +2 -0

index 68fd61a564089..12a48e1d80c3f 100644
@@ -1247,7 +1247,9 @@ struct cifs_tcon {
 	__u32 max_cached_dirs;
 #ifdef CONFIG_CIFS_FSCACHE
 	u64 resource_id;		/* server resource id */
+	bool fscache_acquired;		/* T if we've tried acquiring a cookie */
 	struct fscache_volume *fscache;	/* cookie for share */
+	struct mutex fscache_lock;	/* Prevent regetting a cookie */
 #endif
 	struct list_head pending_opens;	/* list of incomplete opens */
 	struct cached_fids *cfids;

fs/smb/client/fscache.c +13 -0

index a4ee801b29394..ecabc4b400535 100644
@@ -43,12 +43,23 @@ int cifs_fscache_get_super_cookie(struct cifs_tcon *tcon)
 	char *key;
 	int ret = -ENOMEM;
 
+	if (tcon->fscache_acquired)
+		return 0;
+
+	mutex_lock(&tcon->fscache_lock);
+	if (tcon->fscache_acquired) {
+		mutex_unlock(&tcon->fscache_lock);
+		return 0;
+	}
+	tcon->fscache_acquired = true;
+
 	tcon->fscache = NULL;
 	switch (sa->sa_family) {
 	case AF_INET:
 	case AF_INET6:
 		break;
 	default:
+		mutex_unlock(&tcon->fscache_lock);
 		cifs_dbg(VFS, "Unknown network family '%d'\n", sa->sa_family);
 		return -EINVAL;
 	}

@@ -57,6 +68,7 @@ int cifs_fscache_get_super_cookie(struct cifs_tcon *tcon)
 
 	sharename = extract_sharename(tcon->tree_name);
 	if (IS_ERR(sharename)) {
+		mutex_unlock(&tcon->fscache_lock);
 		cifs_dbg(FYI, "%s: couldn't extract sharename\n", __func__);
 		return PTR_ERR(sharename);
 	}

@@ -90,6 +102,7 @@ int cifs_fscache_get_super_cookie(struct cifs_tcon *tcon)
 	kfree(key);
 out:
 	kfree(sharename);
+	mutex_unlock(&tcon->fscache_lock);
 	return ret;
 }

fs/smb/client/misc.c +3 -0

index 74627d647818a..0d13db80e67c9 100644
@@ -141,6 +141,9 @@ tcon_info_alloc(bool dir_leases_enabled)
 	atomic_set(&ret_buf->num_local_opens, 0);
 	atomic_set(&ret_buf->num_remote_opens, 0);
 	ret_buf->stats_from_time = ktime_get_real_seconds();
+#ifdef CONFIG_CIFS_FSCACHE
+	mutex_init(&ret_buf->fscache_lock);
+#endif
 
 	return ret_buf;
 }

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:06 UTC #2 lore raw

PATCH 2/186 smb: client: fix rename(2) regression against samba

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Paulo Alcantara <[email protected]>

[ Upstream commit 18d86965e31f9be4d477da0744a7cdc9815858de ]

After commit 2c7d399e551c ("smb: client: reuse file lease key in compound operations") the client started reusing lease keys for rename, unlink and set path size operations to prevent it from breaking its own leases and thus causing unnecessary lease breaks to same connection.

The implementation relies on positive dentries and cifsInodeInfo::lease_granted to decide whether reusing lease keys for the compound requests. cifsInodeInfo::lease_granted was introduced by commit 0ab95c2510b6 ("Defer close only when lease is enabled.") to indicate whether lease caching is granted for a specific file, but that can only happen until file is open, so cifsInodeInfo::lease_granted was left uninitialised in ->alloc_inode and then client started sending random lease keys for files that hadn't any leases.

This fixes the following test case against samba:

mount.cifs //srv/share /mnt/1 -o ...,nosharesock
mount.cifs //srv/share /mnt/2 -o ...,nosharesock
touch /mnt/1/foo; tail -f /mnt/1/foo & pid=$!
mv /mnt/2/foo /mnt/2/bar # fails with -EIO
kill $pid

Fixes 0ab95c2510b6 ("Defer close only when lease is enabled.")

Signed-off-by Paulo Alcantara (Red Hat) <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 fs/smb/client/cifsfs.c | 1 +
 1 file changed, 1 insertion(+)

fs/smb/client/cifsfs.c +1 -0

index fcb93a66e47cb..44e2cc37a8b63 100644
@@ -392,6 +392,7 @@ cifs_alloc_inode(struct super_block *sb)
 	 * server, can not assume caching of file data or metadata.
 	 */
 	cifs_set_oplock_level(cifs_inode, 0);
+	cifs_inode->lease_granted = false;
 	cifs_inode->flags = 0;
 	spin_lock_init(&cifs_inode->writers_lock);
 	cifs_inode->writers = 0;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:06 UTC #3 lore raw

PATCH 3/186 cifs: reinstate original behavior again for forceuid/forcegid

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Takayuki Nagata <[email protected]>

[ Upstream commit 77d8aa79ecfb209308e0644c02f655122b31def7 ]

forceuid/forcegid should be enabled by default when uid=/gid= options are specified, but commit 24e0a1eff9e2 ("cifs: switch to new mount api") changed the behavior. Due to the change, a mounted share does not show intentional uid/gid for files and directories even though uid=/gid= options are specified since forceuid/forcegid are not enabled.

This patch reinstates original behavior that overrides uid/gid with specified uid/gid by the options.

Fixes 24e0a1eff9e2 ("cifs: switch to new mount api")

Signed-off-by Takayuki Nagata <[email protected]>

Acked-by Paulo Alcantara (Red Hat) <[email protected]>

Acked-by Ronnie Sahlberg <[email protected]>

Acked-by Tom Talpey <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 fs/smb/client/fs_context.c | 12 ++++++++++++
 fs/smb/client/fs_context.h |  2 ++
 2 files changed, 14 insertions(+)

fs/smb/client/fs_context.c +12 -0

index 58567ae617b9f..103421791bb5d 100644
@@ -714,6 +714,16 @@ static int smb3_fs_context_validate(struct fs_context *fc)
 	/* set the port that we got earlier */
 	cifs_set_port((struct sockaddr *)&ctx->dstaddr, ctx->port);
 
+	if (ctx->uid_specified && !ctx->forceuid_specified) {
+		ctx->override_uid = 1;
+		pr_notice("enabling forceuid mount option implicitly because uid= option is specified\n");
+	}
+
+	if (ctx->gid_specified && !ctx->forcegid_specified) {
+		ctx->override_gid = 1;
+		pr_notice("enabling forcegid mount option implicitly because gid= option is specified\n");
+	}
+
 	if (ctx->override_uid && !ctx->uid_specified) {
 		ctx->override_uid = 0;
 		pr_notice("ignoring forceuid mount option specified with no uid= option\n");

@@ -983,12 +993,14 @@ static int smb3_fs_context_parse_param(struct fs_context *fc,
 			ctx->override_uid = 0;
 		else
 			ctx->override_uid = 1;
+		ctx->forceuid_specified = true;
 		break;
 	case Opt_forcegid:
 		if (result.negated)
 			ctx->override_gid = 0;
 		else
 			ctx->override_gid = 1;
+		ctx->forcegid_specified = true;
 		break;
 	case Opt_perm:
 		if (result.negated)

fs/smb/client/fs_context.h +2 -0

index 8cfc25b609b6b..4e409238fe8f7 100644
@@ -155,6 +155,8 @@ enum cifs_param {
 };
 
 struct smb3_fs_context {
+	bool forceuid_specified;
+	bool forcegid_specified;
 	bool uid_specified;
 	bool cruid_specified;
 	bool gid_specified;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:07 UTC #4 lore raw

PATCH 4/186 HID: intel-ish-hid: ipc: Fix dev_err usage with uninitialized dev->devc

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Zhang Lixu <[email protected]>

[ Upstream commit 92826905ae340b7f2b25759a06c8c60bfc476b9f ]

The variable dev->devc in ish_dev_init was utilized by dev_err before it was properly assigned. To rectify this, the assignment of dev->devc has been moved to immediately follow memory allocation.

Without this change "(NULL device *)" is printed for device information.

Fixes 8ae2f2b0a284 ("HID: intel-ish-hid: ipc: Fix potential use-after-free in work function")

Fixes ae02e5d40d5f ("HID: intel-ish-hid: ipc layer")

Signed-off-by Zhang Lixu <[email protected]>

Acked-by Srinivas Pandruvada <[email protected]>

Signed-off-by Jiri Kosina <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/hid/intel-ish-hid/ipc/ipc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

drivers/hid/intel-ish-hid/ipc/ipc.c +1 -1

index a49c6affd7c4c..dd5fc60874ba1 100644
@@ -948,6 +948,7 @@ struct ishtp_device *ish_dev_init(struct pci_dev *pdev)
 	if (!dev)
 		return NULL;
 
+	dev->devc = &pdev->dev;
 	ishtp_device_init(dev);
 
 	init_waitqueue_head(&dev->wait_hw_ready);

@@ -983,7 +984,6 @@ struct ishtp_device *ish_dev_init(struct pci_dev *pdev)
 	}
 
 	dev->ops = &ish_hw_ops;
-	dev->devc = &pdev->dev;
 	dev->mtu = IPC_PAYLOAD_SIZE - sizeof(struct ishtp_msg_hdr);
 	return dev;
 }

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:07 UTC #5 lore raw

PATCH 5/186 HID: logitech-dj: allow mice to use all types of reports

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yaraslau Furman <[email protected]>

[ Upstream commit 21f28a7eb78dea6c59be6b0a5e0b47bf3d25fcbb ]

You can bind whatever action you want to the mouse's reprogrammable buttons using Windows application. Allow Linux to receive multimedia keycodes.

Fixes 3ed224e273ac ("HID: logitech-dj: Fix 064d:c52f receiver support")

Signed-off-by Yaraslau Furman <[email protected]>

Reviewed-by Hans de Goede <[email protected]>

Signed-off-by Jiri Kosina <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/hid/hid-logitech-dj.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

drivers/hid/hid-logitech-dj.c +1 -3

index e6a8b6d8eab70..3c3c497b6b911 100644
@@ -965,9 +965,7 @@ static void logi_hidpp_dev_conn_notif_equad(struct hid_device *hdev,
 		}
 		break;
 	case REPORT_TYPE_MOUSE:
-		workitem->reports_supported |= STD_MOUSE | HIDPP;
-		if (djrcv_dev->type == recvr_type_mouse_only)
-			workitem->reports_supported |= MULTIMEDIA;
+		workitem->reports_supported |= STD_MOUSE | HIDPP | MULTIMEDIA;
 		break;
 	}
 }

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:07 UTC #6 lore raw

PATCH 6/186 arm64: dts: rockchip: set PHY address of MT7531 switch to 0x1f

Link https://lore.kernel.org/r/20240314-for-rockchip-mt7531-phy-address-v1-1-743b5873358f@arinc9.com open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Arınç ÜNAL <[email protected]>

[ Upstream commit a2ac2a1b02590a22a236c43c455f421cdede45f5 ]

The MT7531 switch listens on PHY address 0x1f on an MDIO bus. I've got two findings that support this. There's no bootstrapping option to change the PHY address of the switch. The Linux driver hardcodes 0x1f as the PHY address of the switch. So the reg property on the device tree is currently ignored by the Linux driver.

Therefore, describe the correct PHY address on Banana Pi BPI-R2 Pro that has this switch.

Signed-off-by Arınç ÜNAL <[email protected]>

Fixes c1804463e5c6 ("arm64: dts: rockchip: Add mt7531 dsa node to BPI-R2-Pro board")

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/boot/dts/rockchip/rk3568-bpi-r2-pro.dts | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

arch/arm64/boot/dts/rockchip/rk3568-bpi-r2-pro.dts +2 -2

index f9127ddfbb7df..87c45d8be420f 100644
@@ -525,9 +525,9 @@
 	#address-cells = <1>;
 	#size-cells = <0>;
 
-	switch@0 {
+	switch@1f {
 		compatible = "mediatek,mt7531";
-		reg = <0>;
+		reg = <0x1f>;
 
 		ports {
 			#address-cells = <1>;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:07 UTC #7 lore raw

PATCH 7/186 arm64: dts: rockchip: enable internal pull-up on Q7_USB_ID for RK3399 Puma

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Quentin Schulz <[email protected]>

[ Upstream commit e6b1168f37e3f86d9966276c5a3fff9eb0df3e5f ]

The Q7_USB_ID has a diode used as a level-shifter, and is used as an input pin. The SoC default for this pin is a pull-up, which is correct but the pinconf in the introducing commit missed that, so let's fix this oversight.

Fixes ed2c66a95c0c ("arm64: dts: rockchip: fix rk3399-puma-haikou USB OTG mode")

Signed-off-by Quentin Schulz <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/boot/dts/rockchip/rk3399-puma.dtsi | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

arch/arm64/boot/dts/rockchip/rk3399-puma.dtsi +1 -1

index 20e3f41efe97f..9b9d70bf7f0ca 100644
@@ -443,7 +443,7 @@
 	usb3 {
 		usb3_id: usb3-id {
 			rockchip,pins =
-			  <1 RK_PC2 RK_FUNC_GPIO &pcfg_pull_none>;
+			  <1 RK_PC2 RK_FUNC_GPIO &pcfg_pull_up>;
 		};
 	};
 };

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:07 UTC #8 lore raw

PATCH 8/186 arm64: dts: rockchip: fix alphabetical ordering RK3399 puma

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Iskander Amara <[email protected]>

[ Upstream commit f0abb4b2c7acf3c3e4130dc3f54cd90cf2ae62bc ]

Nodes overridden by their reference should be ordered alphabetically to make it easier to read the DTS. pinctrl node is defined in the wrong location so let's reorder it.

Signed-off-by Iskander Amara <[email protected]>

Reviewed-by Quentin Schulz <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Stable-dep-of: 945a7c857091 ("arm64: dts: rockchip: enable internal pull-up on PCIE_WAKE# for RK3399 Puma")

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/boot/dts/rockchip/rk3399-puma.dtsi | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

arch/arm64/boot/dts/rockchip/rk3399-puma.dtsi +9 -9

index 9b9d70bf7f0ca..2979e6f4f96ce 100644
@@ -401,15 +401,6 @@
 	gpio1830-supply = <&vcc_1v8>;
 };
 
-&pmu_io_domains {
-	status = "okay";
-	pmu1830-supply = <&vcc_1v8>;
-};
-
-&pwm2 {
-	status = "okay";
-};
-
 &pinctrl {
 	i2c8 {
 		i2c8_xfer_a: i2c8-xfer {

@@ -448,6 +439,15 @@
 	};
 };
 
+&pmu_io_domains {
+	status = "okay";
+	pmu1830-supply = <&vcc_1v8>;
+};
+
+&pwm2 {
+	status = "okay";
+};
+
 &sdhci {
 	/*
 	 * Signal integrity isn't great at 200MHz but 100MHz has proven stable

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:07 UTC #9 lore raw

PATCH 9/186 arm64: dts: rockchip: enable internal pull-up on PCIE_WAKE# for RK3399 Puma

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Quentin Schulz <[email protected]>

[ Upstream commit 945a7c8570916650a415757d15d83e0fa856a686 ]

The PCIE_WAKE# has a diode used as a level-shifter, and is used as an input pin. While the SoC default is to enable the pull-up, the core rk3399 pinconf for this pin opted for pull-none. So as to not disturb the behaviour of other boards which may rely on pull-none instead of pull-up, set the needed pull-up only for RK3399 Puma.

Fixes 60fd9f72ce8a ("arm64: dts: rockchip: add Haikou baseboard with RK3399-Q7 SoM")

Signed-off-by Quentin Schulz <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/boot/dts/rockchip/rk3399-puma.dtsi | 5 +++++
 1 file changed, 5 insertions(+)

arch/arm64/boot/dts/rockchip/rk3399-puma.dtsi +5 -0

index 2979e6f4f96ce..39ec38ac716ed 100644
@@ -401,6 +401,11 @@
 	gpio1830-supply = <&vcc_1v8>;
 };
 
+&pcie_clkreqn_cpm {
+	rockchip,pins =
+		<2 RK_PD2 RK_FUNC_GPIO &pcfg_pull_up>;
+};
+
 &pinctrl {
 	i2c8 {
 		i2c8_xfer_a: i2c8-xfer {

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:05 UTC #10 lore raw

PATCH 10/186 arm64: dts: rockchip: Remove unsupported node from the Pinebook Pro dts

Link https://lore.kernel.org/r/0f82c3f97cb798d012270d13b34d8d15305ef293.1711923520.git.dsimic@manjaro.org open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Dragan Simic <[email protected]>

[ Upstream commit 43853e843aa6c3d47ff2b0cce898318839483d05 ]

Remove a redundant node from the Pine64 Pinebook Pro dts, which is intended to provide a value for the delay in PCI Express enumeration, but that isn't supported without additional out-of-tree kernel patches.

There were already efforts to upstream those kernel patches, because they reportedly make some PCI Express cards (such as LSI SAS HBAs) usable in Pine64 RockPro64 (which is also based on the RK3399); otherwise, those PCI Express cards fail to enumerate. However, providing the required background and explanations proved to be a tough nut to crack, which is the reason why those patches remain outside of the kernel mainline for now.

If those out-of-tree patches eventually become upstreamed, the resulting device-tree changes will almost surely belong to the RK3399 SoC dtsi. Also, the above-mentioned unusable-without-out-of-tree-patches PCI Express devices are in all fairness not usable in a Pinebook Pro without some extensive hardware modifications, which is another reason to delete this redundant node. When it comes to the Pinebook Pro, only M.2 NVMe SSDs can be installed out of the box (using an additional passive adapter PCB sold separately by Pine64), which reportedly works fine with no additional patches.

Fixes 5a65505a6988 ("arm64: dts: rockchip: Add initial support for Pinebook Pro")

Signed-off-by Dragan Simic <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/boot/dts/rockchip/rk3399-pinebook-pro.dts | 1 -
 1 file changed, 1 deletion(-)

arch/arm64/boot/dts/rockchip/rk3399-pinebook-pro.dts +0 -1

index 054c6a4d1a45f..294eb2de263de 100644
@@ -779,7 +779,6 @@
 };
 
 &pcie0 {
-	bus-scan-delay-ms = <1000>;
 	ep-gpios = <&gpio2 RK_PD4 GPIO_ACTIVE_HIGH>;
 	num-lanes = <4>;
 	pinctrl-names = "default";

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:05 UTC #11 lore raw

PATCH 11/186 arm64: dts: mediatek: mt8183: Add power-domains properity to mfgcfg

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ikjoon Jang <[email protected]>

[ Upstream commit 1781f2c461804c0123f59afc7350e520a88edffb ]

mfgcfg clock is under MFG_ASYNC power domain.

Fixes e526c9bc11f8 ("arm64: dts: Add Mediatek SoC MT8183 and evaluation board dts and Makefile")

Fixes 37fb78b9aeb7 ("arm64: dts: mediatek: Add mt8183 power domains controller")

Signed-off-by Weiyi Lu <[email protected]>

Signed-off-by Ikjoon Jang <[email protected]>

Reviewed-by Enric Balletbo i Serra <[email protected]>

Signed-off-by Chen-Yu Tsai <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by AngeloGioacchino Del Regno <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/boot/dts/mediatek/mt8183.dtsi | 1 +
 1 file changed, 1 insertion(+)

arch/arm64/boot/dts/mediatek/mt8183.dtsi +1 -0

index df6e9990cd5fa..8721a5ffca30a 100644
@@ -1628,6 +1628,7 @@
 			compatible = "mediatek,mt8183-mfgcfg", "syscon";
 			reg = <0 0x13000000 0 0x1000>;
 			#clock-cells = <1>;
+			power-domains = <&spm MT8183_POWER_DOMAIN_MFG_ASYNC>;
 		};
 
 		gpu: gpu@13040000 {

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:05 UTC #12 lore raw

PATCH 12/186 arm64: dts: mediatek: mt8192: Add missing gce-client-reg to mutex

Link https://lore.kernel.org/r/20240229-gce-client-reg-add-missing-mt8192-95-v1-1-b12c233a8a33@collabora.com open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Nícolas F. R. A. Prado <[email protected]>

[ Upstream commit 00bcc8810d9dd69d3899a4189e2f3964f263a600 ]

Add the missing mediatek,gce-client-reg property to the mutex node to allow it to use the GCE. This prevents the "can't parse gce-client-reg property" error from being printed and should result in better performance.

Fixes b4b75bac952b ("arm64: dts: mt8192: Add display nodes")

Signed-off-by Nícolas F. R. A. Prado <[email protected]>

Reviewed-by AngeloGioacchino Del Regno <[email protected]>

Signed-off-by AngeloGioacchino Del Regno <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/boot/dts/mediatek/mt8192.dtsi | 1 +
 1 file changed, 1 insertion(+)

arch/arm64/boot/dts/mediatek/mt8192.dtsi +1 -0

index f1fc14e53f8c7..b1443adc55aab 100644
@@ -1412,6 +1412,7 @@
 			reg = <0 0x14001000 0 0x1000>;
 			interrupts = <GIC_SPI 252 IRQ_TYPE_LEVEL_HIGH 0>;
 			clocks = <&mmsys CLK_MM_DISP_MUTEX0>;
+			mediatek,gce-client-reg = <&gce SUBSYS_1400XXXX 0x1000 0x1000>;
 			mediatek,gce-events = <CMDQ_EVENT_DISP_STREAM_DONE_ENG_EVENT_0>,
 					      <CMDQ_EVENT_DISP_STREAM_DONE_ENG_EVENT_1>;
 			power-domains = <&spm MT8192_POWER_DOMAIN_DISP>;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:06 UTC #13 lore raw

PATCH 13/186 arm64: dts: mediatek: mt8195: Add missing gce-client-reg to vpp/vdosys

Link https://lore.kernel.org/r/20240229-gce-client-reg-add-missing-mt8192-95-v1-2-b12c233a8a33@collabora.com open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Nícolas F. R. A. Prado <[email protected]>

[ Upstream commit 96b0c1528ef41fe754f5d1378b1db6c098a2e33f ]

Add the missing mediatek,gce-client-reg property to the vppsys and vdosys nodes to allow them to use the GCE. This prevents the "can't parse gce-client-reg property" error from being printed and should result in better performance.

Fixes 6aa5b46d1755 ("arm64: dts: mt8195: Add vdosys and vppsys clock nodes")

Signed-off-by Nícolas F. R. A. Prado <[email protected]>

Signed-off-by AngeloGioacchino Del Regno <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/boot/dts/mediatek/mt8195.dtsi | 3 +++
 1 file changed, 3 insertions(+)

arch/arm64/boot/dts/mediatek/mt8195.dtsi +3 -0

index 6708c4d21abf9..8fe8917daed1d 100644
@@ -1963,6 +1963,7 @@
 			compatible = "mediatek,mt8195-vppsys0", "syscon";
 			reg = <0 0x14000000 0 0x1000>;
 			#clock-cells = <1>;
+			mediatek,gce-client-reg = <&gce1 SUBSYS_1400XXXX 0 0x1000>;
 		};
 
 		mutex@1400f000 {

@@ -2077,6 +2078,7 @@
 			compatible = "mediatek,mt8195-vppsys1", "syscon";
 			reg = <0 0x14f00000 0 0x1000>;
 			#clock-cells = <1>;
+			mediatek,gce-client-reg = <&gce1 SUBSYS_14f0XXXX 0 0x1000>;
 		};
 
 		mutex@14f01000 {

@@ -2623,6 +2625,7 @@
 			reg = <0 0x1c01a000 0 0x1000>;
 			mboxes = <&gce0 0 CMDQ_THR_PRIO_4>;
 			#clock-cells = <1>;
+			mediatek,gce-client-reg = <&gce0 SUBSYS_1c01XXXX 0xa000 0x1000>;
 		};

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:06 UTC #14 lore raw

PATCH 14/186 arm64: dts: mediatek: mt8195: Add missing gce-client-reg to mutex

Link https://lore.kernel.org/r/20240229-gce-client-reg-add-missing-mt8192-95-v1-3-b12c233a8a33@collabora.com open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Nícolas F. R. A. Prado <[email protected]>

[ Upstream commit 3b129949184a1251e6a42db714f6d68b75fabedd ]

Fixes b852ee68fd72 ("arm64: dts: mt8195: Add display node for vdosys0")

Signed-off-by Nícolas F. R. A. Prado <[email protected]>

Signed-off-by AngeloGioacchino Del Regno <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/boot/dts/mediatek/mt8195.dtsi | 1 +
 1 file changed, 1 insertion(+)

arch/arm64/boot/dts/mediatek/mt8195.dtsi +1 -0

index 8fe8917daed1d..884e56f4a93bb 100644
@@ -2779,6 +2779,7 @@
 			interrupts = <GIC_SPI 658 IRQ_TYPE_LEVEL_HIGH 0>;
 			power-domains = <&spm MT8195_POWER_DOMAIN_VDOSYS0>;
 			clocks = <&vdosys0 CLK_VDO0_DISP_MUTEX0>;
+			mediatek,gce-client-reg = <&gce0 SUBSYS_1c01XXXX 0x6000 0x1000>;
 			mediatek,gce-events = <CMDQ_EVENT_VDO0_DISP_STREAM_DONE_0>;
 		};

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:06 UTC #15 lore raw

PATCH 15/186 arm64: dts: mediatek: mt8195: Add missing gce-client-reg to mutex1

Link https://lore.kernel.org/r/20240229-gce-client-reg-add-missing-mt8192-95-v1-4-b12c233a8a33@collabora.com open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Nícolas F. R. A. Prado <[email protected]>

[ Upstream commit 58f126296c3c52d02bf3fad1f68c331d718c4a9b ]

Add the missing mediatek,gce-client-reg property to the mutex1 node to allow it to use the GCE. This prevents the "can't parse gce-client-reg property" error from being printed and should result in better performance.

Fixes 92d2c23dc269 ("arm64: dts: mt8195: add display node for vdosys1")

Signed-off-by Nícolas F. R. A. Prado <[email protected]>

Signed-off-by AngeloGioacchino Del Regno <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/boot/dts/mediatek/mt8195.dtsi | 1 +
 1 file changed, 1 insertion(+)

arch/arm64/boot/dts/mediatek/mt8195.dtsi +1 -0

index 884e56f4a93bb..2bb9d9aa65fed 100644
@@ -2850,6 +2850,7 @@
 			power-domains = <&spm MT8195_POWER_DOMAIN_VDOSYS1>;
 			clocks = <&vdosys1 CLK_VDO1_DISP_MUTEX>;
 			clock-names = "vdo1_mutex";
+			mediatek,gce-client-reg = <&gce0 SUBSYS_1c10XXXX 0x1000 0x1000>;
 			mediatek,gce-events = <CMDQ_EVENT_VDO1_STREAM_DONE_ENG_0>;
 		};

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:06 UTC #16 lore raw

PATCH 16/186 arm64: dts: mediatek: cherry: Add platform thermal configuration

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: AngeloGioacchino Del Regno <[email protected]>

[ Upstream commit 729f30eac8bce6783f889cf8390ea869d03407e6 ]

This platform has three auxiliary NTC thermistors, connected to the SoC's ADC pins. Enable the auxadc in order to be able to read the ADC values, add a generic-adc-thermal LUT for each and finally assign them to the SoC's thermal zones.

Tested-by Chen-Yu Tsai <[email protected]>

Reviewed-by Chen-Yu Tsai <[email protected]>

Reviewed-by Alexandre Mergnat <[email protected]>

Signed-off-by AngeloGioacchino Del Regno <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Stable-dep-of: 17b33dd9e4a3 ("arm64: dts: mediatek: cherry: Describe CPU supplies")

Signed-off-by Sasha Levin <[email protected]>

---
 .../boot/dts/mediatek/mt8195-cherry.dtsi      | 105 ++++++++++++++++++
 1 file changed, 105 insertions(+)

arch/arm64/boot/dts/mediatek/mt8195-cherry.dtsi +105 -0

index 3f508e5c18434..d721ddb230747 100644
@@ -114,6 +114,77 @@
 		regulator-boot-on;
 	};
 
+	/* Murata NCP03WF104F05RL */
+	tboard_thermistor1: thermal-sensor-t1 {
+		compatible = "generic-adc-thermal";
+		#thermal-sensor-cells = <0>;
+		io-channels = <&auxadc 0>;
+		io-channel-names = "sensor-channel";
+		temperature-lookup-table = <	(-10000) 1553
+						(-5000) 1485
+						0 1406
+						5000 1317
+						10000 1219
+						15000 1115
+						20000 1007
+						25000 900
+						30000 796
+						35000 697
+						40000 605
+						45000 523
+						50000 449
+						55000 384
+						60000 327
+						65000 279
+						70000 237
+						75000 202
+						80000 172
+						85000 147
+						90000 125
+						95000 107
+						100000 92
+						105000 79
+						110000 68
+						115000 59
+						120000 51
+						125000 44>;
+	};
+
+	tboard_thermistor2: thermal-sensor-t2 {
+		compatible = "generic-adc-thermal";
+		#thermal-sensor-cells = <0>;
+		io-channels = <&auxadc 1>;
+		io-channel-names = "sensor-channel";
+		temperature-lookup-table = <	(-10000) 1553
+						(-5000) 1485
+						0 1406
+						5000 1317
+						10000 1219
+						15000 1115
+						20000 1007
+						25000 900
+						30000 796
+						35000 697
+						40000 605
+						45000 523
+						50000 449
+						55000 384
+						60000 327
+						65000 279
+						70000 237
+						75000 202
+						80000 172
+						85000 147
+						90000 125
+						95000 107
+						100000 92
+						105000 79
+						110000 68
+						115000 59
+						120000 51
+						125000 44>;
+	};
+
 	usb_vbus: regulator-5v0-usb-vbus {
 		compatible = "regulator-fixed";
 		regulator-name = "usb-vbus";

@@ -176,6 +247,10 @@
 	memory-region = <&afe_mem>;
 };
 
+&auxadc {
+	status = "okay";
+};
+
 &dp_intf0 {
 	status = "okay";

@@ -1127,6 +1202,36 @@
 	};
 };
 
+&thermal_zones {
+	soc-area-thermal {
+		polling-delay = <1000>;
+		polling-delay-passive = <250>;
+		thermal-sensors = <&tboard_thermistor1>;
+
+		trips {
+			trip-crit {
+				temperature = <84000>;
+				hysteresis = <1000>;
+				type = "critical";
+			};
+		};
+	};
+
+	pmic-area-thermal {
+		polling-delay = <1000>;
+		polling-delay-passive = <0>;
+		thermal-sensors = <&tboard_thermistor2>;
+
+		trips {
+			trip-crit {
+				temperature = <84000>;
+				hysteresis = <1000>;
+				type = "critical";
+			};
+		};
+	};
+};
+
 &u3phy0 {
 	status = "okay";
 };

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:06 UTC #17 lore raw

PATCH 17/186 arm64: dts: mediatek: cherry: Describe CPU supplies

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Nícolas F. R. A. Prado <[email protected]>

[ Upstream commit 17b33dd9e4a38fbaca87c68e532b52f9d0492ba7 ]

Describe in each CPU node the regulator supplying it.

Fixes 260c04d425eb ("arm64: dts: mediatek: cherry: Enable MT6315 regulators on SPMI bus")

Signed-off-by Nícolas F. R. A. Prado <[email protected]>

Reviewed-by AngeloGioacchino Del Regno <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by AngeloGioacchino Del Regno <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 .../boot/dts/mediatek/mt8195-cherry.dtsi      | 32 +++++++++++++++++++
 1 file changed, 32 insertions(+)

arch/arm64/boot/dts/mediatek/mt8195-cherry.dtsi +32 -0

index d721ddb230747..663c9f414b611 100644
@@ -251,6 +251,38 @@
 	status = "okay";
 };
 
+&cpu0 {
+	cpu-supply = <&mt6359_vcore_buck_reg>;
+};
+
+&cpu1 {
+	cpu-supply = <&mt6359_vcore_buck_reg>;
+};
+
+&cpu2 {
+	cpu-supply = <&mt6359_vcore_buck_reg>;
+};
+
+&cpu3 {
+	cpu-supply = <&mt6359_vcore_buck_reg>;
+};
+
+&cpu4 {
+	cpu-supply = <&mt6315_6_vbuck1>;
+};
+
+&cpu5 {
+	cpu-supply = <&mt6315_6_vbuck1>;
+};
+
+&cpu6 {
+	cpu-supply = <&mt6315_6_vbuck1>;
+};
+
+&cpu7 {
+	cpu-supply = <&mt6315_6_vbuck1>;
+};
+
 &dp_intf0 {
 	status = "okay";

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:06 UTC #18 lore raw

PATCH 18/186 arm64: dts: mediatek: mt8192-asurada: Update min voltage constraint for MT6315

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Pin-yen Lin <[email protected]>

[ Upstream commit 374a7c6400e314458178255a63c37d6347845092 ]

Update the minimum voltage from 300000 uV to 400000 uV so it matches the MT6315 datasheet.

Also update the minimum voltage for Vgpu regulator from 606250 uV to 400000 uV because the requested voltage could be lower than the minimum voltage on the GPU OPP table when the MTK Smart Voltage Scaling (SVS) driver is enabled.

Fixes 3183cb62b033 ("arm64: dts: mediatek: asurada: Add SPMI regulators")

Signed-off-by Pin-yen Lin <[email protected]>

Reviewed-by AngeloGioacchino Del Regno <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by AngeloGioacchino Del Regno <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/boot/dts/mediatek/mt8192-asurada.dtsi | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

arch/arm64/boot/dts/mediatek/mt8192-asurada.dtsi +3 -3

index 4bd1494b354c0..dc39ebd1bbfc8 100644
@@ -1392,7 +1392,7 @@
 			mt6315_6_vbuck1: vbuck1 {
 				regulator-compatible = "vbuck1";
 				regulator-name = "Vbcpu";
-				regulator-min-microvolt = <300000>;
+				regulator-min-microvolt = <400000>;
 				regulator-max-microvolt = <1193750>;
 				regulator-enable-ramp-delay = <256>;
 				regulator-allowed-modes = <0 1 2>;

@@ -1402,7 +1402,7 @@
 			mt6315_6_vbuck3: vbuck3 {
 				regulator-compatible = "vbuck3";
 				regulator-name = "Vlcpu";
-				regulator-min-microvolt = <300000>;
+				regulator-min-microvolt = <400000>;
 				regulator-max-microvolt = <1193750>;
 				regulator-enable-ramp-delay = <256>;
 				regulator-allowed-modes = <0 1 2>;

@@ -1419,7 +1419,7 @@
 			mt6315_7_vbuck1: vbuck1 {
 				regulator-compatible = "vbuck1";
 				regulator-name = "Vgpu";
-				regulator-min-microvolt = <606250>;
+				regulator-min-microvolt = <400000>;
 				regulator-max-microvolt = <800000>;
 				regulator-enable-ramp-delay = <256>;
 				regulator-allowed-modes = <0 1 2>;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:06 UTC #19 lore raw

PATCH 19/186 arm64: dts: mediatek: mt8195-cherry: Update min voltage constraint for MT6315

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Pin-yen Lin <[email protected]>

[ Upstream commit e9a6b8b5c61350535c7eb5ea9b2dde0d5745bd1b ]

Update the minimum voltage from 300000 uV to 400000 uV so it matches the MT6315 datasheet.

Also update the minimum voltage for Vgpu regulator from 625000 uV to 400000 uV because the requested voltage could be lower than the minimum voltage on the GPU OPP table when the MTK Smart Voltage Scaling (SVS) driver is enabled.

Fixes 260c04d425eb ("arm64: dts: mediatek: cherry: Enable MT6315 regulators on SPMI bus")

Signed-off-by Pin-yen Lin <[email protected]>

Reviewed-by AngeloGioacchino Del Regno <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by AngeloGioacchino Del Regno <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/boot/dts/mediatek/mt8195-cherry.dtsi | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

arch/arm64/boot/dts/mediatek/mt8195-cherry.dtsi +2 -2

index 663c9f414b611..b78f408110bf7 100644
@@ -1205,7 +1205,7 @@
 			mt6315_6_vbuck1: vbuck1 {
 				regulator-compatible = "vbuck1";
 				regulator-name = "Vbcpu";
-				regulator-min-microvolt = <300000>;
+				regulator-min-microvolt = <400000>;
 				regulator-max-microvolt = <1193750>;
 				regulator-enable-ramp-delay = <256>;
 				regulator-ramp-delay = <6250>;

@@ -1223,7 +1223,7 @@
 			mt6315_7_vbuck1: vbuck1 {
 				regulator-compatible = "vbuck1";
 				regulator-name = "Vgpu";
-				regulator-min-microvolt = <625000>;
+				regulator-min-microvolt = <400000>;
 				regulator-max-microvolt = <1193750>;
 				regulator-enable-ramp-delay = <256>;
 				regulator-ramp-delay = <6250>;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:06 UTC #20 lore raw

PATCH 20/186 arm64: dts: mediatek: mt8183-kukui: Use default min voltage for MT6358

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Pin-yen Lin <[email protected]>

[ Upstream commit 296118a8dc297de47d9b3a364b9743f8446bd612 ]

The requested voltage could be lower than the minimum voltage on the GPU OPP table when the MTK Smart Voltage Scaling (SVS) driver is enabled, so removing the definition in mt8183-kukui to use the default minimum voltage (500000 uV) defined in mt6358.dtsi.

Fixes 31c6732da9d5 ("arm64: dts: mediatek: mt8183-kukui: Override vgpu/vsram_gpu constraints")

Signed-off-by Pin-yen Lin <[email protected]>

Reviewed-by AngeloGioacchino Del Regno <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by AngeloGioacchino Del Regno <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/boot/dts/mediatek/mt8183-kukui.dtsi | 1 -
 1 file changed, 1 deletion(-)

arch/arm64/boot/dts/mediatek/mt8183-kukui.dtsi +0 -1

index 70becf10cacb8..d846342c1d3b2 100644
@@ -405,7 +405,6 @@
 };
 
 &mt6358_vgpu_reg {
-	regulator-min-microvolt = <625000>;
 	regulator-max-microvolt = <900000>;
 
 	regulator-coupled-with = <&mt6358_vsram_gpu_reg>;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:06 UTC #21 lore raw

PATCH 21/186 arm64: dts: mediatek: mt7622: fix clock controllers

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafał Miłecki <[email protected]>

[ Upstream commit 3ba5a61594347ab46e7c2cff6cd63ea0f1282efb ]

1. Drop unneeded "syscon"s (bindings were updated recently)
2. Use "clock-controller" in nodenames
3. Add missing "#clock-cells"

Fixes d7167881e03e ("arm64: dts: mt7622: add clock controller device nodes")

Fixes e9b65ecb7c30 ("arm64: dts: mediatek: mt7622: introduce nodes for Wireless Ethernet Dispatch")

Reviewed-by AngeloGioacchino Del Regno <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by AngeloGioacchino Del Regno <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/boot/dts/mediatek/mt7622.dtsi | 27 +++++++++++-------------
 1 file changed, 12 insertions(+), 15 deletions(-)

arch/arm64/boot/dts/mediatek/mt7622.dtsi +12 ~~-15~~

index 3ee9266fa8e98..283fdf7d2d8b9 100644
@@ -283,16 +283,14 @@
 		};
 	};
 
-	apmixedsys: apmixedsys@10209000 {
-		compatible = "mediatek,mt7622-apmixedsys",
-			     "syscon";
+	apmixedsys: clock-controller@10209000 {
+		compatible = "mediatek,mt7622-apmixedsys";
 		reg = <0 0x10209000 0 0x1000>;
 		#clock-cells = <1>;
 	};
 
-	topckgen: topckgen@10210000 {
-		compatible = "mediatek,mt7622-topckgen",
-			     "syscon";
+	topckgen: clock-controller@10210000 {
+		compatible = "mediatek,mt7622-topckgen";
 		reg = <0 0x10210000 0 0x1000>;
 		#clock-cells = <1>;
 	};

@@ -734,9 +732,8 @@
 		power-domains = <&scpsys MT7622_POWER_DOMAIN_WB>;
 	};
 
-	ssusbsys: ssusbsys@1a000000 {
-		compatible = "mediatek,mt7622-ssusbsys",
-			     "syscon";
+	ssusbsys: clock-controller@1a000000 {
+		compatible = "mediatek,mt7622-ssusbsys";
 		reg = <0 0x1a000000 0 0x1000>;
 		#clock-cells = <1>;
 		#reset-cells = <1>;

@@ -793,9 +790,8 @@
 		};
 	};
 
-	pciesys: pciesys@1a100800 {
-		compatible = "mediatek,mt7622-pciesys",
-			     "syscon";
+	pciesys: clock-controller@1a100800 {
+		compatible = "mediatek,mt7622-pciesys";
 		reg = <0 0x1a100800 0 0x1000>;
 		#clock-cells = <1>;
 		#reset-cells = <1>;

@@ -921,12 +917,13 @@
 		};
 	};
 
-	hifsys: syscon@1af00000 {
-		compatible = "mediatek,mt7622-hifsys", "syscon";
+	hifsys: clock-controller@1af00000 {
+		compatible = "mediatek,mt7622-hifsys";
 		reg = <0 0x1af00000 0 0x70>;
+		#clock-cells = <1>;
 	};
 
-	ethsys: syscon@1b000000 {
+	ethsys: clock-controller@1b000000 {
 		compatible = "mediatek,mt7622-ethsys",
 			     "syscon";
 		reg = <0 0x1b000000 0 0x1000>;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:06 UTC #22 lore raw

PATCH 22/186 arm64: dts: mediatek: mt7622: fix IR nodename

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafał Miłecki <[email protected]>

[ Upstream commit 800dc93c3941e372c94278bf4059e6e82f60bd66 ]

Fix following validation error:
arch/arm64/boot/dts/mediatek/mt7622-rfb1.dtb: cir@10009000: $nodename:0: 'cir@10009000' does not match '^ir(-receiver)?(@[a-f0-9]+)?$'
from schema $id: http://devicetree.org/schemas/media/mediatek,mt7622-cir.yaml#

Fixes ae457b7679c4 ("arm64: dts: mt7622: add SoC and peripheral related device nodes")

Reviewed-by AngeloGioacchino Del Regno <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by AngeloGioacchino Del Regno <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/boot/dts/mediatek/mt7622.dtsi | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

arch/arm64/boot/dts/mediatek/mt7622.dtsi +1 -1

index 283fdf7d2d8b9..4c8a71c8184b7 100644
@@ -252,7 +252,7 @@
 		clock-names = "hif_sel";
 	};
 
-	cir: cir@10009000 {
+	cir: ir-receiver@10009000 {
 		compatible = "mediatek,mt7622-cir";
 		reg = <0 0x10009000 0 0x1000>;
 		interrupts = <GIC_SPI 175 IRQ_TYPE_LEVEL_LOW>;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:06 UTC #23 lore raw

PATCH 23/186 arm64: dts: mediatek: mt7622: fix ethernet controller "compatible"

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafał Miłecki <[email protected]>

[ Upstream commit 208add29ce5b7291f6c466e4dfd9cbf61c72888e ]

Fix following validation error:
arch/arm64/boot/dts/mediatek/mt7622-rfb1.dtb: ethernet@1b100000: compatible: ['mediatek,mt7622-eth', 'mediatek,mt2701-eth', 'syscon'] is too long
from schema $id: http://devicetree.org/schemas/net/mediatek,net.yaml#
(and other complains about wrong clocks).

Fixes 5f599b3a0bb8 ("arm64: dts: mt7622: add ethernet device nodes")

Reviewed-by AngeloGioacchino Del Regno <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by AngeloGioacchino Del Regno <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/boot/dts/mediatek/mt7622.dtsi | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

arch/arm64/boot/dts/mediatek/mt7622.dtsi +1 -3

index 4c8a71c8184b7..8e46480b5364b 100644
@@ -963,9 +963,7 @@
 	};
 
 	eth: ethernet@1b100000 {
-		compatible = "mediatek,mt7622-eth",
-			     "mediatek,mt2701-eth",
-			     "syscon";
+		compatible = "mediatek,mt7622-eth";
 		reg = <0 0x1b100000 0 0x20000>;
 		interrupts = <GIC_SPI 223 IRQ_TYPE_LEVEL_LOW>,
 			     <GIC_SPI 224 IRQ_TYPE_LEVEL_LOW>,

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:06 UTC #24 lore raw

PATCH 24/186 arm64: dts: mediatek: mt7622: drop "reset-names" from thermal block

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafał Miłecki <[email protected]>

[ Upstream commit ecb5b0034f5bcc35003b4b965cf50c6e98316e79 ]

Binding doesn't specify "reset-names" property and Linux driver also doesn't use it.

Fix following validation error:
arch/arm64/boot/dts/mediatek/mt7622-rfb1.dtb: thermal@1100b000: Unevaluated properties are not allowed ('reset-names' was unexpected)
from schema $id: http://devicetree.org/schemas/thermal/mediatek,thermal.yaml#

Fixes ae457b7679c4 ("arm64: dts: mt7622: add SoC and peripheral related device nodes")

Reviewed-by AngeloGioacchino Del Regno <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by AngeloGioacchino Del Regno <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/boot/dts/mediatek/mt7622.dtsi | 1 -
 1 file changed, 1 deletion(-)

arch/arm64/boot/dts/mediatek/mt7622.dtsi +0 -1

index 8e46480b5364b..917fa39a74f8d 100644
@@ -513,7 +513,6 @@
 			 <&pericfg CLK_PERI_AUXADC_PD>;
 		clock-names = "therm", "auxadc";
 		resets = <&pericfg MT7622_PERI_THERM_SW_RST>;
-		reset-names = "therm";
 		mediatek,auxadc = <&auxadc>;
 		mediatek,apmixedsys = <&apmixedsys>;
 		nvmem-cells = <&thermal_calibration>;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:06 UTC #25 lore raw

PATCH 25/186 arm64: dts: mediatek: mt7986: reorder properties

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafał Miłecki <[email protected]>

[ Upstream commit 7eb133c99fbebc6adb1cbd22c926d42d2bbca648 ]

Use order described as preferred in DTS Coding Style. Mostly just move "compatible", "reg" and "ranges" properties. In two nodes also move vendor-prefixed props down.

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by AngeloGioacchino Del Regno <[email protected]>

Stable-dep-of: 3b449bfd2ff6 ("arm64: dts: mediatek: mt7986: drop invalid properties from ethsys")

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/boot/dts/mediatek/mt7986a.dtsi | 67 ++++++++++++-----------
 1 file changed, 34 insertions(+), 33 deletions(-)

arch/arm64/boot/dts/mediatek/mt7986a.dtsi +34 ~~-33~~

index d974739eae1c9..eba5e27a1bbea 100644
@@ -27,34 +27,34 @@
 		#address-cells = <1>;
 		#size-cells = <0>;
 		cpu0: cpu@0 {
-			device_type = "cpu";
 			compatible = "arm,cortex-a53";
-			enable-method = "psci";
 			reg = <0x0>;
+			device_type = "cpu";
+			enable-method = "psci";
 			#cooling-cells = <2>;
 		};
 
 		cpu1: cpu@1 {
-			device_type = "cpu";
 			compatible = "arm,cortex-a53";
-			enable-method = "psci";
 			reg = <0x1>;
+			device_type = "cpu";
+			enable-method = "psci";
 			#cooling-cells = <2>;
 		};
 
 		cpu2: cpu@2 {
-			device_type = "cpu";
 			compatible = "arm,cortex-a53";
-			enable-method = "psci";
 			reg = <0x2>;
+			device_type = "cpu";
+			enable-method = "psci";
 			#cooling-cells = <2>;
 		};
 
 		cpu3: cpu@3 {
-			device_type = "cpu";
-			enable-method = "psci";
 			compatible = "arm,cortex-a53";
 			reg = <0x3>;
+			device_type = "cpu";
+			enable-method = "psci";
 			#cooling-cells = <2>;
 		};
 	};

@@ -131,22 +131,22 @@
 	};
 
 	soc {
-		#address-cells = <2>;
-		#size-cells = <2>;
 		compatible = "simple-bus";
 		ranges;
+		#address-cells = <2>;
+		#size-cells = <2>;
 
 		gic: interrupt-controller@c000000 {
 			compatible = "arm,gic-v3";
-			#interrupt-cells = <3>;
-			interrupt-parent = <&gic>;
-			interrupt-controller;
 			reg = <0 0x0c000000 0 0x10000>,  /* GICD */
 			      <0 0x0c080000 0 0x80000>,  /* GICR */
 			      <0 0x0c400000 0 0x2000>,   /* GICC */
 			      <0 0x0c410000 0 0x1000>,   /* GICH */
 			      <0 0x0c420000 0 0x2000>;   /* GICV */
+			interrupt-parent = <&gic>;
 			interrupts = <GIC_PPI 9 IRQ_TYPE_LEVEL_HIGH>;
+			interrupt-controller;
+			#interrupt-cells = <3>;
 		};
 
 		infracfg: infracfg@10001000 {

@@ -310,9 +310,9 @@
 
 		spi0: spi@1100a000 {
 			compatible = "mediatek,mt7986-spi-ipm", "mediatek,spi-ipm";
+			reg = <0 0x1100a000 0 0x100>;
 			#address-cells = <1>;
 			#size-cells = <0>;
-			reg = <0 0x1100a000 0 0x100>;
 			interrupts = <GIC_SPI 140 IRQ_TYPE_LEVEL_HIGH>;
 			clocks = <&topckgen CLK_TOP_MPLL_D2>,
 				 <&topckgen CLK_TOP_SPI_SEL>,

@@ -324,9 +324,9 @@
 
 		spi1: spi@1100b000 {
 			compatible = "mediatek,mt7986-spi-ipm", "mediatek,spi-ipm";
+			reg = <0 0x1100b000 0 0x100>;
 			#address-cells = <1>;
 			#size-cells = <0>;
-			reg = <0 0x1100b000 0 0x100>;
 			interrupts = <GIC_SPI 141 IRQ_TYPE_LEVEL_HIGH>;
 			clocks = <&topckgen CLK_TOP_MPLL_D2>,
 				 <&topckgen CLK_TOP_SPIM_MST_SEL>,

@@ -388,7 +388,6 @@
 		};
 
 		thermal: thermal@1100c800 {
-			#thermal-sensor-cells = <1>;
 			compatible = "mediatek,mt7986-thermal";
 			reg = <0 0x1100c800 0 0x800>;
 			interrupts = <GIC_SPI 138 IRQ_TYPE_LEVEL_HIGH>;

@@ -396,30 +395,30 @@
 				 <&infracfg CLK_INFRA_ADC_26M_CK>,
 				 <&infracfg CLK_INFRA_ADC_FRC_CK>;
 			clock-names = "therm", "auxadc", "adc_32k";
-			mediatek,auxadc = <&auxadc>;
-			mediatek,apmixedsys = <&apmixedsys>;
 			nvmem-cells = <&thermal_calibration>;
 			nvmem-cell-names = "calibration-data";
+			#thermal-sensor-cells = <1>;
+			mediatek,auxadc = <&auxadc>;
+			mediatek,apmixedsys = <&apmixedsys>;
 		};
 
 		pcie: pcie@11280000 {
 			compatible = "mediatek,mt7986-pcie",
 				     "mediatek,mt8192-pcie";
+			reg = <0x00 0x11280000 0x00 0x4000>;
+			reg-names = "pcie-mac";
+			ranges = <0x82000000 0x00 0x20000000 0x00
+				  0x20000000 0x00 0x10000000>;
 			device_type = "pci";
 			#address-cells = <3>;
 			#size-cells = <2>;
-			reg = <0x00 0x11280000 0x00 0x4000>;
-			reg-names = "pcie-mac";
 			interrupts = <GIC_SPI 168 IRQ_TYPE_LEVEL_HIGH>;
 			bus-range = <0x00 0xff>;
-			ranges = <0x82000000 0x00 0x20000000 0x00
-				  0x20000000 0x00 0x10000000>;
 			clocks = <&infracfg CLK_INFRA_IPCIE_PIPE_CK>,
 				 <&infracfg CLK_INFRA_IPCIE_CK>,
 				 <&infracfg CLK_INFRA_IPCIER_CK>,
 				 <&infracfg CLK_INFRA_IPCIEB_CK>;
 			clock-names = "pl_250m", "tl_26m", "peri_26m", "top_133m";
-			status = "disabled";
 
 			phys = <&pcie_port PHY_TYPE_PCIE>;
 			phy-names = "pcie-phy";

@@ -430,6 +429,8 @@
 					<0 0 0 2 &pcie_intc 1>,
 					<0 0 0 3 &pcie_intc 2>,
 					<0 0 0 4 &pcie_intc 3>;
+			status = "disabled";
+
 			pcie_intc: interrupt-controller {
 				#address-cells = <0>;
 				#interrupt-cells = <1>;

@@ -440,9 +441,9 @@
 		pcie_phy: t-phy {
 			compatible = "mediatek,mt7986-tphy",
 				     "mediatek,generic-tphy-v2";
+			ranges;
 			#address-cells = <2>;
 			#size-cells = <2>;
-			ranges;
 			status = "disabled";
 
 			pcie_port: pcie-phy@11c00000 {

@@ -467,9 +468,9 @@
 		usb_phy: t-phy@11e10000 {
 			compatible = "mediatek,mt7986-tphy",
 				     "mediatek,generic-tphy-v2";
+			ranges = <0 0 0x11e10000 0x1700>;
 			#address-cells = <1>;
 			#size-cells = <1>;
-			ranges = <0 0 0x11e10000 0x1700>;
 			status = "disabled";
 
 			u2port0: usb-phy@0 {

@@ -497,11 +498,11 @@
 		};
 
 		ethsys: syscon@15000000 {
-			 #address-cells = <1>;
-			 #size-cells = <1>;
 			 compatible = "mediatek,mt7986-ethsys",
 				      "syscon";
 			 reg = <0 0x15000000 0 0x1000>;
+			 #address-cells = <1>;
+			 #size-cells = <1>;
 			 #clock-cells = <1>;
 			 #reset-cells = <1>;
 		};

@@ -578,26 +579,26 @@
 					  <&topckgen CLK_TOP_SGM_325M_SEL>;
 			assigned-clock-parents = <&apmixedsys CLK_APMIXED_NET2PLL>,
 						 <&apmixedsys CLK_APMIXED_SGMPLL>;
+			#reset-cells = <1>;
+			#address-cells = <1>;
+			#size-cells = <0>;
 			mediatek,ethsys = <&ethsys>;
 			mediatek,sgmiisys = <&sgmiisys0>, <&sgmiisys1>;
 			mediatek,wed-pcie = <&wed_pcie>;
 			mediatek,wed = <&wed0>, <&wed1>;
-			#reset-cells = <1>;
-			#address-cells = <1>;
-			#size-cells = <0>;
 			status = "disabled";
 		};
 
 		wifi: wifi@18000000 {
 			compatible = "mediatek,mt7986-wmac";
+			reg = <0 0x18000000 0 0x1000000>,
+			      <0 0x10003000 0 0x1000>,
+			      <0 0x11d10000 0 0x1000>;
 			resets = <&watchdog MT7986_TOPRGU_CONSYS_SW_RST>;
 			reset-names = "consys";
 			clocks = <&topckgen CLK_TOP_CONN_MCUSYS_SEL>,
 				 <&topckgen CLK_TOP_AP2CNN_HOST_SEL>;
 			clock-names = "mcu", "ap2conn";
-			reg = <0 0x18000000 0 0x1000000>,
-			      <0 0x10003000 0 0x1000>,
-			      <0 0x11d10000 0 0x1000>;
 			interrupts = <GIC_SPI 213 IRQ_TYPE_LEVEL_HIGH>,
 				     <GIC_SPI 214 IRQ_TYPE_LEVEL_HIGH>,
 				     <GIC_SPI 215 IRQ_TYPE_LEVEL_HIGH>,

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:06 UTC #26 lore raw

PATCH 26/186 arm64: dts: mediatek: mt7986: drop invalid properties from ethsys

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafał Miłecki <[email protected]>

[ Upstream commit 3b449bfd2ff6c5d3ceecfcb18528ff8e1b4ac2fd ]

Mediatek ethsys controller / syscon binding doesn't allow any subnodes so "#address-cells" and "#size-cells" are redundant (actually:
disallowed).

This fixes:
arch/arm64/boot/dts/mediatek/mt7986a-bananapi-bpi-r3.dtb: syscon@15000000: '#address-cells', '#size-cells' do not match any of the regexes: 'pinctrl-[0-9]+'
from schema $id: http://devicetree.org/schemas/clock/mediatek,ethsys.yaml#

Fixes 1f9986b258c2 ("arm64: dts: mediatek: add clock support for mt7986a")

Cc Sam Shih <[email protected]>

Reviewed-by AngeloGioacchino Del Regno <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by AngeloGioacchino Del Regno <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/boot/dts/mediatek/mt7986a.dtsi | 2 --
 1 file changed, 2 deletions(-)

arch/arm64/boot/dts/mediatek/mt7986a.dtsi +0 -2

index eba5e27a1bbea..850b664dfa13d 100644
@@ -501,8 +501,6 @@
 			 compatible = "mediatek,mt7986-ethsys",
 				      "syscon";
 			 reg = <0 0x15000000 0 0x1000>;
-			 #address-cells = <1>;
-			 #size-cells = <1>;
 			 #clock-cells = <1>;
 			 #reset-cells = <1>;
 		};

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:06 UTC #27 lore raw

PATCH 27/186 arm64: dts: mediatek: mt7986: drop "#reset-cells" from Ethernet controller

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafał Miłecki <[email protected]>

[ Upstream commit 9bd88afc94c3570289a0f1c696578b3e1f4e3169 ]

Ethernet block doesn't include or act as a reset controller. Documentation also doesn't document "#reset-cells" for it.

This fixes:
arch/arm64/boot/dts/mediatek/mt7986a-bananapi-bpi-r3.dtb: ethernet@15100000: Unevaluated properties are not allowed ('#reset-cells' was unexpected)
from schema $id: http://devicetree.org/schemas/net/mediatek,net.yaml#

Fixes 082ff36bd5c0 ("arm64: dts: mediatek: mt7986: introduce ethernet nodes")

Cc Lorenzo Bianconi <[email protected]>

Reviewed-by AngeloGioacchino Del Regno <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by AngeloGioacchino Del Regno <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/boot/dts/mediatek/mt7986a.dtsi | 1 -
 1 file changed, 1 deletion(-)

arch/arm64/boot/dts/mediatek/mt7986a.dtsi +0 -1

index 850b664dfa13d..5c2fe2f43a142 100644
@@ -577,7 +577,6 @@
 					  <&topckgen CLK_TOP_SGM_325M_SEL>;
 			assigned-clock-parents = <&apmixedsys CLK_APMIXED_NET2PLL>,
 						 <&apmixedsys CLK_APMIXED_SGMPLL>;
-			#reset-cells = <1>;
 			#address-cells = <1>;
 			#size-cells = <0>;
 			mediatek,ethsys = <&ethsys>;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:06 UTC #28 lore raw

PATCH 28/186 arm64: dts: mediatek: mt7986: reorder nodes

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafał Miłecki <[email protected]>

[ Upstream commit 3f79e8f3364499750d7442767b101b7bc5864ddf ]

Use order described as preferred in DTS Coding Style:
1. Sort bus nodes by unit address
2. Use alpha-numerical order for the rest

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by AngeloGioacchino Del Regno <[email protected]>

Stable-dep-of: 970f8b01bd77 ("arm64: dts: mediatek: mt7986: drop invalid thermal block clock")

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/boot/dts/mediatek/mt7986a.dtsi | 116 +++++++++++-----------
 1 file changed, 58 insertions(+), 58 deletions(-)

arch/arm64/boot/dts/mediatek/mt7986a.dtsi +58 ~~-58~~

index 5c2fe2f43a142..f3a2a89fada41 100644
@@ -16,13 +16,6 @@
 	#address-cells = <2>;
 	#size-cells = <2>;
 
-	clk40m: oscillator-40m {
-		compatible = "fixed-clock";
-		clock-frequency = <40000000>;
-		#clock-cells = <0>;
-		clock-output-names = "clkxtal";
-	};
-
 	cpus {
 		#address-cells = <1>;
 		#size-cells = <0>;

@@ -59,6 +52,13 @@
 		};
 	};
 
+	clk40m: oscillator-40m {
+		compatible = "fixed-clock";
+		clock-frequency = <40000000>;
+		#clock-cells = <0>;
+		clock-output-names = "clkxtal";
+	};
+
 	psci {
 		compatible = "arm,psci-0.2";
 		method = "smc";

@@ -121,15 +121,6 @@
 
 	};
 
-	timer {
-		compatible = "arm,armv8-timer";
-		interrupt-parent = <&gic>;
-		interrupts = <GIC_PPI 13 IRQ_TYPE_LEVEL_LOW>,
-			     <GIC_PPI 14 IRQ_TYPE_LEVEL_LOW>,
-			     <GIC_PPI 11 IRQ_TYPE_LEVEL_LOW>,
-			     <GIC_PPI 10 IRQ_TYPE_LEVEL_LOW>;
-	};
-
 	soc {
 		compatible = "simple-bus";
 		ranges;

@@ -203,6 +194,19 @@
 			#interrupt-cells = <2>;
 		};
 
+		pwm: pwm@10048000 {
+			compatible = "mediatek,mt7986-pwm";
+			reg = <0 0x10048000 0 0x1000>;
+			#pwm-cells = <2>;
+			interrupts = <GIC_SPI 137 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&topckgen CLK_TOP_PWM_SEL>,
+				 <&infracfg CLK_INFRA_PWM_STA>,
+				 <&infracfg CLK_INFRA_PWM1_CK>,
+				 <&infracfg CLK_INFRA_PWM2_CK>;
+			clock-names = "top", "main", "pwm1", "pwm2";
+			status = "disabled";
+		};
+
 		sgmiisys0: syscon@10060000 {
 			compatible = "mediatek,mt7986-sgmiisys_0",
 				     "syscon";

@@ -240,19 +244,6 @@
 			status = "disabled";
 		};
 
-		pwm: pwm@10048000 {
-			compatible = "mediatek,mt7986-pwm";
-			reg = <0 0x10048000 0 0x1000>;
-			#pwm-cells = <2>;
-			interrupts = <GIC_SPI 137 IRQ_TYPE_LEVEL_HIGH>;
-			clocks = <&topckgen CLK_TOP_PWM_SEL>,
-				 <&infracfg CLK_INFRA_PWM_STA>,
-				 <&infracfg CLK_INFRA_PWM1_CK>,
-				 <&infracfg CLK_INFRA_PWM2_CK>;
-			clock-names = "top", "main", "pwm1", "pwm2";
-			status = "disabled";
-		};
-
 		uart0: serial@11002000 {
 			compatible = "mediatek,mt7986-uart",
 				     "mediatek,mt6577-uart";

@@ -336,6 +327,21 @@
 			status = "disabled";
 		};
 
+		thermal: thermal@1100c800 {
+			compatible = "mediatek,mt7986-thermal";
+			reg = <0 0x1100c800 0 0x800>;
+			interrupts = <GIC_SPI 138 IRQ_TYPE_LEVEL_HIGH>;
+			clocks = <&infracfg CLK_INFRA_THERM_CK>,
+				 <&infracfg CLK_INFRA_ADC_26M_CK>,
+				 <&infracfg CLK_INFRA_ADC_FRC_CK>;
+			clock-names = "therm", "auxadc", "adc_32k";
+			nvmem-cells = <&thermal_calibration>;
+			nvmem-cell-names = "calibration-data";
+			#thermal-sensor-cells = <1>;
+			mediatek,auxadc = <&auxadc>;
+			mediatek,apmixedsys = <&apmixedsys>;
+		};
+
 		auxadc: adc@1100d000 {
 			compatible = "mediatek,mt7986-auxadc";
 			reg = <0 0x1100d000 0 0x1000>;

@@ -387,21 +393,6 @@
 			status = "disabled";
 		};
 
-		thermal: thermal@1100c800 {
-			compatible = "mediatek,mt7986-thermal";
-			reg = <0 0x1100c800 0 0x800>;
-			interrupts = <GIC_SPI 138 IRQ_TYPE_LEVEL_HIGH>;
-			clocks = <&infracfg CLK_INFRA_THERM_CK>,
-				 <&infracfg CLK_INFRA_ADC_26M_CK>,
-				 <&infracfg CLK_INFRA_ADC_FRC_CK>;
-			clock-names = "therm", "auxadc", "adc_32k";
-			nvmem-cells = <&thermal_calibration>;
-			nvmem-cell-names = "calibration-data";
-			#thermal-sensor-cells = <1>;
-			mediatek,auxadc = <&auxadc>;
-			mediatek,apmixedsys = <&apmixedsys>;
-		};
-
 		pcie: pcie@11280000 {
 			compatible = "mediatek,mt7986-pcie",
 				     "mediatek,mt8192-pcie";

@@ -531,20 +522,6 @@
 			mediatek,wo-ccif = <&wo_ccif1>;
 		};
 
-		wo_ccif0: syscon@151a5000 {
-			compatible = "mediatek,mt7986-wo-ccif", "syscon";
-			reg = <0 0x151a5000 0 0x1000>;
-			interrupt-parent = <&gic>;
-			interrupts = <GIC_SPI 211 IRQ_TYPE_LEVEL_HIGH>;
-		};
-
-		wo_ccif1: syscon@151ad000 {
-			compatible = "mediatek,mt7986-wo-ccif", "syscon";
-			reg = <0 0x151ad000 0 0x1000>;
-			interrupt-parent = <&gic>;
-			interrupts = <GIC_SPI 212 IRQ_TYPE_LEVEL_HIGH>;
-		};
-
 		eth: ethernet@15100000 {
 			compatible = "mediatek,mt7986-eth";
 			reg = <0 0x15100000 0 0x80000>;

@@ -586,6 +563,20 @@
 			status = "disabled";
 		};
 
+		wo_ccif0: syscon@151a5000 {
+			compatible = "mediatek,mt7986-wo-ccif", "syscon";
+			reg = <0 0x151a5000 0 0x1000>;
+			interrupt-parent = <&gic>;
+			interrupts = <GIC_SPI 211 IRQ_TYPE_LEVEL_HIGH>;
+		};
+
+		wo_ccif1: syscon@151ad000 {
+			compatible = "mediatek,mt7986-wo-ccif", "syscon";
+			reg = <0 0x151ad000 0 0x1000>;
+			interrupt-parent = <&gic>;
+			interrupts = <GIC_SPI 212 IRQ_TYPE_LEVEL_HIGH>;
+		};
+
 		wifi: wifi@18000000 {
 			compatible = "mediatek,mt7986-wmac";
 			reg = <0 0x18000000 0 0x1000000>,

@@ -643,4 +634,13 @@
 			};
 		};
 	};
+
+	timer {
+		compatible = "arm,armv8-timer";
+		interrupt-parent = <&gic>;
+		interrupts = <GIC_PPI 13 IRQ_TYPE_LEVEL_LOW>,
+			     <GIC_PPI 14 IRQ_TYPE_LEVEL_LOW>,
+			     <GIC_PPI 11 IRQ_TYPE_LEVEL_LOW>,
+			     <GIC_PPI 10 IRQ_TYPE_LEVEL_LOW>;
+	};
 };

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:06 UTC #29 lore raw

PATCH 29/186 arm64: dts: mediatek: mt7986: drop invalid thermal block clock

Link https://lore.kernel.org/linux-devicetree/[email protected]/ open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafał Miłecki <[email protected]>

[ Upstream commit 970f8b01bd7719a22e577ba6c78e27f9ccf22783 ]

Thermal block uses only two clocks. Its binding doesn't document or allow "adc_32k". Also Linux driver doesn't support it.

It has been additionally verified by Angelo by his detailed research on MT7981 / MT7986 clocks (thanks!).

This fixes:
arch/arm64/boot/dts/mediatek/mt7986a-bananapi-bpi-r3.dtb: thermal@1100c800: clocks: [[4, 27], [4, 44], [4, 45]] is too long
from schema $id: http://devicetree.org/schemas/thermal/mediatek,thermal.yaml#
arch/arm64/boot/dts/mediatek/mt7986a-bananapi-bpi-r3.dtb: thermal@1100c800: clock-names: ['therm', 'auxadc', 'adc_32k'] is too long
from schema $id: http://devicetree.org/schemas/thermal/mediatek,thermal.yaml#

Fixes 0a9615d58d04 ("arm64: dts: mt7986: add thermal and efuse")

Cc Daniel Golle <[email protected]>

Reviewed-by AngeloGioacchino Del Regno <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by AngeloGioacchino Del Regno <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/boot/dts/mediatek/mt7986a.dtsi | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

arch/arm64/boot/dts/mediatek/mt7986a.dtsi +2 -3

index f3a2a89fada41..559990dcd1d17 100644
@@ -332,9 +332,8 @@
 			reg = <0 0x1100c800 0 0x800>;
 			interrupts = <GIC_SPI 138 IRQ_TYPE_LEVEL_HIGH>;
 			clocks = <&infracfg CLK_INFRA_THERM_CK>,
-				 <&infracfg CLK_INFRA_ADC_26M_CK>,
-				 <&infracfg CLK_INFRA_ADC_FRC_CK>;
-			clock-names = "therm", "auxadc", "adc_32k";
+				 <&infracfg CLK_INFRA_ADC_26M_CK>;
+			clock-names = "therm", "auxadc";
 			nvmem-cells = <&thermal_calibration>;
 			nvmem-cell-names = "calibration-data";
 			#thermal-sensor-cells = <1>;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:07 UTC #30 lore raw

PATCH 30/186 arm64: dts: mediatek: mt7986: prefix BPI-R3 cooling maps with "map-"

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafał Miłecki <[email protected]>

[ Upstream commit f8c65a5e4560781f2ea175d8f26cd75ac98e8d78 ]

This fixes:
arch/arm64/boot/dts/mediatek/mt7986a-bananapi-bpi-r3.dtb: thermal-zones: cpu-thermal:cooling-maps: 'cpu-active-high', 'cpu-active-low', 'cpu-active-med' do not match any of the regexes: '^map[-a-zA-Z0-9]*$', 'pinctrl-[0-9]+'
from schema $id: http://devicetree.org/schemas/thermal/thermal-zones.yaml#

Fixes c26f779a2295 ("arm64: dts: mt7986: add pwm-fan and cooling-maps to BPI-R3 dts")

Cc Daniel Golle <[email protected]>

Reviewed-by AngeloGioacchino Del Regno <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by AngeloGioacchino Del Regno <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/boot/dts/mediatek/mt7986a-bananapi-bpi-r3.dts | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

arch/arm64/boot/dts/mediatek/mt7986a-bananapi-bpi-r3.dts +3 -3

index e1ec2cccf4444..aba6686eb34a3 100644
@@ -146,19 +146,19 @@
 
 &cpu_thermal {
 	cooling-maps {
-		cpu-active-high {
+		map-cpu-active-high {
 			/* active: set fan to cooling level 2 */
 			cooling-device = <&fan 2 2>;
 			trip = <&cpu_trip_active_high>;
 		};
 
-		cpu-active-med {
+		map-cpu-active-med {
 			/* active: set fan to cooling level 1 */
 			cooling-device = <&fan 1 1>;
 			trip = <&cpu_trip_active_med>;
 		};
 
-		cpu-active-low {
+		map-cpu-active-low {
 			/* active: set fan to cooling level 0 */
 			cooling-device = <&fan 0 0>;
 			trip = <&cpu_trip_active_low>;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:07 UTC #31 lore raw

PATCH 31/186 arm64: dts: mediatek: mt2712: fix validation errors

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafał Miłecki <[email protected]>

[ Upstream commit 3baac7291effb501c4d52df7019ebf52011e5772 ]

1. Fixup infracfg clock controller binding
It also acts as reset controller so #reset-cells is required.
2. Use -pins suffix for pinctrl

This fixes:
arch/arm64/boot/dts/mediatek/mt2712-evb.dtb: syscon@10001000: '#reset-cells' is a required property
from schema $id: http://devicetree.org/schemas/arm/mediatek/mediatek,infracfg.yaml#
arch/arm64/boot/dts/mediatek/mt2712-evb.dtb: pinctrl@1000b000: 'eth_default', 'eth_sleep', 'usb0_iddig', 'usb1_iddig' do not match any of the regexes: 'pinctrl-[0-9]+', 'pins$'
from schema $id: http://devicetree.org/schemas/pinctrl/mediatek,mt65xx-pinctrl.yaml#

Reviewed-by AngeloGioacchino Del Regno <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

[Angelo: Added Fixes tags]

Fixes 5d4839709c8e ("arm64: dts: mt2712: Add clock controller device nodes")

Fixes 1724f4cc5133 ("arm64: dts: Add USB3 related nodes for MT2712")

Signed-off-by AngeloGioacchino Del Regno <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/boot/dts/mediatek/mt2712-evb.dts | 8 ++++----
 arch/arm64/boot/dts/mediatek/mt2712e.dtsi   | 3 ++-
 2 files changed, 6 insertions(+), 5 deletions(-)

arch/arm64/boot/dts/mediatek/mt2712-evb.dts +4 -4

index fffdb7bbf889e..2d0ef6f23b3a9 100644
@@ -129,7 +129,7 @@
 };
 
 &pio {
-	eth_default: eth_default {
+	eth_default: eth-default-pins {
 		tx_pins {
 			pinmux = <MT2712_PIN_71_GBE_TXD3__FUNC_GBE_TXD3>,
 				 <MT2712_PIN_72_GBE_TXD2__FUNC_GBE_TXD2>,

@@ -156,7 +156,7 @@
 		};
 	};
 
-	eth_sleep: eth_sleep {
+	eth_sleep: eth-sleep-pins {
 		tx_pins {
 			pinmux = <MT2712_PIN_71_GBE_TXD3__FUNC_GPIO71>,
 				 <MT2712_PIN_72_GBE_TXD2__FUNC_GPIO72>,

@@ -182,14 +182,14 @@
 		};
 	};
 
-	usb0_id_pins_float: usb0_iddig {
+	usb0_id_pins_float: usb0-iddig-pins {
 		pins_iddig {
 			pinmux = <MT2712_PIN_12_IDDIG_P0__FUNC_IDDIG_A>;
 			bias-pull-up;
 		};
 	};
 
-	usb1_id_pins_float: usb1_iddig {
+	usb1_id_pins_float: usb1-iddig-pins {
 		pins_iddig {
 			pinmux = <MT2712_PIN_14_IDDIG_P1__FUNC_IDDIG_B>;
 			bias-pull-up;

arch/arm64/boot/dts/mediatek/mt2712e.dtsi +2 -1

index ed1a9d3194153..f767f921bdee1 100644
@@ -249,10 +249,11 @@
 		#clock-cells = <1>;
 	};
 
-	infracfg: syscon@10001000 {
+	infracfg: clock-controller@10001000 {
 		compatible = "mediatek,mt2712-infracfg", "syscon";
 		reg = <0 0x10001000 0 0x1000>;
 		#clock-cells = <1>;
+		#reset-cells = <1>;
 	};
 
 	pericfg: syscon@10003000 {

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:07 UTC #32 lore raw

PATCH 32/186 arm64: dts: rockchip: regulator for sd needs to be always on for BPI-R2Pro

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jose Ignacio Tornos Martinez <[email protected]>

[ Upstream commit 433d54818f64a2fe0562f8c04c7a81f562368515 ]

With default dts configuration for BPI-R2Pro, the regulator for sd card is powered off when reboot is commanded, and the only solution to detect the sd card again, and therefore, allow rebooting from there, is to do a hardware reset.

Configure the regulator for sd to be always on for BPI-R2Pro in order to avoid this issue.

Fixes f901aaadaa2a ("arm64: dts: rockchip: Add Bananapi R2 Pro")

Signed-off-by Jose Ignacio Tornos Martinez <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/boot/dts/rockchip/rk3568-bpi-r2-pro.dts | 2 ++
 1 file changed, 2 insertions(+)

arch/arm64/boot/dts/rockchip/rk3568-bpi-r2-pro.dts +2 -0

index 87c45d8be420f..dc5892d25c100 100644
@@ -416,6 +416,8 @@
 
 			vccio_sd: LDO_REG5 {
 				regulator-name = "vccio_sd";
+				regulator-always-on;
+				regulator-boot-on;
 				regulator-min-microvolt = <1800000>;
 				regulator-max-microvolt = <3300000>;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:09 UTC #33 lore raw

PATCH 33/186 arm64: dts: qcom: sc8180x: Fix ss_phy_irq for secondary USB controller

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Maximilian Luz <[email protected]>

[ Upstream commit ecda8309098402f878c96184f29a1b7ec682d772 ]

The ACPI DSDT of the Surface Pro X (SQ2) specifies the interrupts for the secondary UBS controller as

    Name (_CRS, ResourceTemplate ()
    {
        Interrupt (ResourceConsumer, Level, ActiveHigh, Shared, ,, )
        {
            0x000000AA,
        }
        Interrupt (ResourceConsumer, Level, ActiveHigh, SharedAndWake, ,, )
        {
            0x000000A7,     // hs_phy_irq: &intc GIC_SPI 136
        }
        Interrupt (ResourceConsumer, Level, ActiveHigh, SharedAndWake, ,, )
        {
            0x00000228,     // ss_phy_irq: &pdc 40
        }
        Interrupt (ResourceConsumer, Edge, ActiveHigh, SharedAndWake, ,, )
        {
            0x0000020A,     // dm_hs_phy_irq: &pdc 10
        }
        Interrupt (ResourceConsumer, Edge, ActiveHigh, SharedAndWake, ,, )
        {
            0x0000020B,     // dp_hs_phy_irq: &pdc 11
        }
    })

Generally, the interrupts above 0x200 map to the PDC interrupts (as used in the devicetree) as ACPI_NUMBER - 0x200. Note that this lines up with dm_hs_phy_irq and dp_hs_phy_irq (as well as the interrupts for the primary USB controller).

Based on the snippet above, ss_phy_irq should therefore be PDC 40 (= 0x28) and not PDC 7. The latter is according to ACPI instead used as ss_phy_irq for port 0 of the multiport USB controller). Fix this by setting ss_phy_irq to '&pdc 40'.

Fixes b080f53a8f44 ("arm64: dts: qcom: sc8180x: Add remoteprocs, wifi and usb nodes")

Signed-off-by Maximilian Luz <[email protected]>

Reviewed-by Bjorn Andersson <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Bjorn Andersson <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm64/boot/dts/qcom/sc8180x.dtsi | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

arch/arm64/boot/dts/qcom/sc8180x.dtsi +1 -1

index 6eb4c5eb6bb8c..fbb9bf09078a0 100644
@@ -2644,7 +2644,7 @@
 			resets = <&gcc GCC_USB30_SEC_BCR>;
 			power-domains = <&gcc USB30_SEC_GDSC>;
 			interrupts-extended = <&intc GIC_SPI 136 IRQ_TYPE_LEVEL_HIGH>,
-					      <&pdc 7 IRQ_TYPE_LEVEL_HIGH>,
+					      <&pdc 40 IRQ_TYPE_LEVEL_HIGH>,
 					      <&pdc 10 IRQ_TYPE_EDGE_BOTH>,
 					      <&pdc 11 IRQ_TYPE_EDGE_BOTH>;
 			interrupt-names = "hs_phy_irq", "ss_phy_irq",

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:07 UTC #34 lore raw

PATCH 34/186 gpio: tangier: Use correct type for the IRQ chip data

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Andy Shevchenko <[email protected]>

[ Upstream commit 7d045025a24b6336d444d359bd4312f351d017f9 ]

IRQ chip data contains a pointer to the GPIO chip. Luckily we have the pointers the same, but strictly speaking it's not guaranteed. Even though, still better to fix this.

Fixes ccf6fd6dcc86 ("gpio: merrifield: Introduce GPIO driver to support Merrifield")

Signed-off-by Andy Shevchenko <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/gpio/gpio-tangier.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

drivers/gpio/gpio-tangier.c +6 -3

index 7ce3eddaed257..1ce40b7673b11 100644
@@ -205,7 +205,8 @@ static int tng_gpio_set_config(struct gpio_chip *chip, unsigned int offset,
 
 static void tng_irq_ack(struct irq_data *d)
 {
-	struct tng_gpio *priv = irq_data_get_irq_chip_data(d);
+	struct gpio_chip *gc = irq_data_get_irq_chip_data(d);
+	struct tng_gpio *priv = gpiochip_get_data(gc);
 	irq_hw_number_t gpio = irqd_to_hwirq(d);
 	unsigned long flags;
 	void __iomem *gisr;

@@ -241,7 +242,8 @@ static void tng_irq_unmask_mask(struct tng_gpio *priv, u32 gpio, bool unmask)
 
 static void tng_irq_mask(struct irq_data *d)
 {
-	struct tng_gpio *priv = irq_data_get_irq_chip_data(d);
+	struct gpio_chip *gc = irq_data_get_irq_chip_data(d);
+	struct tng_gpio *priv = gpiochip_get_data(gc);
 	irq_hw_number_t gpio = irqd_to_hwirq(d);
 
 	tng_irq_unmask_mask(priv, gpio, false);

@@ -250,7 +252,8 @@ static void tng_irq_mask(struct irq_data *d)
 
 static void tng_irq_unmask(struct irq_data *d)
 {
-	struct tng_gpio *priv = irq_data_get_irq_chip_data(d);
+	struct gpio_chip *gc = irq_data_get_irq_chip_data(d);
+	struct tng_gpio *priv = gpiochip_get_data(gc);
 	irq_hw_number_t gpio = irqd_to_hwirq(d);
 
 	gpiochip_enable_irq(&priv->chip, gpio);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:08 UTC #35 lore raw

PATCH 35/186 ARC: [plat-hsdk]: Remove misplaced interrupt-cells property

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alexey Brodkin <[email protected]>

[ Upstream commit 61231eb8113ce47991f35024f9c20810b37996bf ]

"gmac" node stands for just an ordinary Ethernet controller, which is by no means a provider of interrupts, i.e. it doesn't serve as an interrupt controller, thus "#interrupt-cells" property doesn't belong to it and so we remove it.

Fixes:
------------>8------------
DTC arch/arc/boot/dts/hsdk.dtb
arch/arc/boot/dts/hsdk.dts:207.23-235.5: Warning (interrupt_provider): /soc/ethernet@8000: '#interrupt-cells' found, but node is not an interrupt provider arch/arc/boot/dts/hsdk.dtb: Warning (interrupt_map): Failed prerequisite 'interrupt_provider'
------------>8------------

Reported-by Vineet Gupta <[email protected]>

Signed-off-by Alexey Brodkin <[email protected]>

Signed-off-by Vineet Gupta <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arc/boot/dts/hsdk.dts | 1 -
 1 file changed, 1 deletion(-)

arch/arc/boot/dts/hsdk.dts +0 -1

index 6691f42550778..41b980df862b1 100644
@@ -205,7 +205,6 @@
 		};
 
 		gmac: ethernet@8000 {
-			#interrupt-cells = <1>;
 			compatible = "snps,dwmac";
 			reg = <0x8000 0x2000>;
 			interrupts = <10>;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:08 UTC #36 lore raw

PATCH 36/186 wifi: mac80211: clean up assignments to pointer cache.

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Colin Ian King <[email protected]>

[ Upstream commit ba4b1fa3128b2fbf14e167230315cbd9074b629b ]

The assignment to pointer cache in function mesh_fast_tx_gc can be made at the declaration time rather than a later assignment. There are also 3 functions where pointer cache is being initialized at declaration time and later re-assigned again with the same value, these are redundant and can be removed.

Cleans up code and three clang scan build warnings:
warning: Value stored to 'cache' during its initialization is never read [deadcode.DeadStores]

Signed-off-by Colin Ian King <[email protected]>

Link https://msgid.link/[email protected] open here

Stable-dep-of: 8c75cdcdf869 ("wifi: mac80211: split mesh fast tx cache into local/proxied/forwarded")

Signed-off-by Sasha Levin <[email protected]>

---
 net/mac80211/mesh_pathtbl.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)

net/mac80211/mesh_pathtbl.c +1 -5

index 3e52aaa57b1fc..c7173190f9b93 100644
@@ -595,11 +595,10 @@ void mesh_fast_tx_cache(struct ieee80211_sub_if_data *sdata,
 void mesh_fast_tx_gc(struct ieee80211_sub_if_data *sdata)
 {
 	unsigned long timeout = msecs_to_jiffies(MESH_FAST_TX_CACHE_TIMEOUT);
-	struct mesh_tx_cache *cache;
+	struct mesh_tx_cache *cache = &sdata->u.mesh.tx_cache;
 	struct ieee80211_mesh_fast_tx *entry;
 	struct hlist_node *n;
 
-	cache = &sdata->u.mesh.tx_cache;
 	if (atomic_read(&cache->rht.nelems) < MESH_FAST_TX_CACHE_THRESHOLD_SIZE)
 		return;

@@ -617,7 +616,6 @@ void mesh_fast_tx_flush_mpath(struct mesh_path *mpath)
 	struct ieee80211_mesh_fast_tx *entry;
 	struct hlist_node *n;
 
-	cache = &sdata->u.mesh.tx_cache;
 	spin_lock_bh(&cache->walk_lock);
 	hlist_for_each_entry_safe(entry, n, &cache->walk_head, walk_list)
 		if (entry->mpath == mpath)

@@ -632,7 +630,6 @@ void mesh_fast_tx_flush_sta(struct ieee80211_sub_if_data *sdata,
 	struct ieee80211_mesh_fast_tx *entry;
 	struct hlist_node *n;
 
-	cache = &sdata->u.mesh.tx_cache;
 	spin_lock_bh(&cache->walk_lock);
 	hlist_for_each_entry_safe(entry, n, &cache->walk_head, walk_list)
 		if (rcu_access_pointer(entry->mpath->next_hop) == sta)

@@ -646,7 +643,6 @@ void mesh_fast_tx_flush_addr(struct ieee80211_sub_if_data *sdata,
 	struct mesh_tx_cache *cache = &sdata->u.mesh.tx_cache;
 	struct ieee80211_mesh_fast_tx *entry;
 
-	cache = &sdata->u.mesh.tx_cache;
 	spin_lock_bh(&cache->walk_lock);
 	entry = rhashtable_lookup_fast(&cache->rht, addr, fast_tx_rht_params);
 	if (entry)

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:09 UTC #37 lore raw

PATCH 37/186 wifi: mac80211: split mesh fast tx cache into local/proxied/forwarded

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Felix Fietkau <[email protected]>

[ Upstream commit 8c75cdcdf869acabfdc7858827099dcde9f24e6c ]

Depending on the origin of the packets (and their SA), 802.11 + mesh headers could be filled in differently. In order to properly deal with that, add a new field to the lookup key, indicating the type (local, proxied or forwarded). This can fix spurious packet drop issues that depend on the order in which nodes/hosts communicate with each other.

Fixes d5edb9ae8d56 ("wifi: mac80211: mesh fast xmit support")

Signed-off-by Felix Fietkau <[email protected]>

Link https://msgid.link/[email protected] open here

[use sizeof_field() for key_len]

Signed-off-by Sasha Levin <[email protected]>

---
 net/mac80211/mesh.c         |  8 +++++++-
 net/mac80211/mesh.h         | 36 +++++++++++++++++++++++++++++++++---
 net/mac80211/mesh_pathtbl.c | 31 ++++++++++++++++++++++---------
 net/mac80211/rx.c           | 13 ++++++++++---
 4 files changed, 72 insertions(+), 16 deletions(-)

net/mac80211/mesh.c +7 -1

index e31c312c124a1..7b3ecc288f09d 100644
@@ -765,6 +765,9 @@ bool ieee80211_mesh_xmit_fast(struct ieee80211_sub_if_data *sdata,
 			      struct sk_buff *skb, u32 ctrl_flags)
 {
 	struct ieee80211_if_mesh *ifmsh = &sdata->u.mesh;
+	struct ieee80211_mesh_fast_tx_key key = {
+		.type = MESH_FAST_TX_TYPE_LOCAL
+	};
 	struct ieee80211_mesh_fast_tx *entry;
 	struct ieee80211s_hdr *meshhdr;
 	u8 sa[ETH_ALEN] __aligned(2);

@@ -800,7 +803,10 @@ bool ieee80211_mesh_xmit_fast(struct ieee80211_sub_if_data *sdata,
 			return false;
 	}
 
-	entry = mesh_fast_tx_get(sdata, skb->data);
+	ether_addr_copy(key.addr, skb->data);
+	if (!ether_addr_equal(skb->data + ETH_ALEN, sdata->vif.addr))
+		key.type = MESH_FAST_TX_TYPE_PROXIED;
+	entry = mesh_fast_tx_get(sdata, &key);
 	if (!entry)
 		return false;

net/mac80211/mesh.h +33 -3

index ad8469293d712..58c619874ca6a 100644
@@ -133,10 +133,39 @@ struct mesh_path {
 #define MESH_FAST_TX_CACHE_THRESHOLD_SIZE	384
 #define MESH_FAST_TX_CACHE_TIMEOUT		8000 /* msecs */
 
+/**
+ * enum ieee80211_mesh_fast_tx_type - cached mesh fast tx entry type
+ *
+ * @MESH_FAST_TX_TYPE_LOCAL: tx from the local vif address as SA
+ * @MESH_FAST_TX_TYPE_PROXIED: local tx with a different SA (e.g. bridged)
+ * @MESH_FAST_TX_TYPE_FORWARDED: forwarded from a different mesh point
+ * @NUM_MESH_FAST_TX_TYPE: number of entry types
+ */
+enum ieee80211_mesh_fast_tx_type {
+	MESH_FAST_TX_TYPE_LOCAL,
+	MESH_FAST_TX_TYPE_PROXIED,
+	MESH_FAST_TX_TYPE_FORWARDED,
+
+	/* must be last */
+	NUM_MESH_FAST_TX_TYPE
+};
+
+
+/**
+ * struct ieee80211_mesh_fast_tx_key - cached mesh fast tx entry key
+ *
+ * @addr: The Ethernet DA for this entry
+ * @type: cache entry type
+ */
+struct ieee80211_mesh_fast_tx_key {
+	u8 addr[ETH_ALEN] __aligned(2);
+	u16 type;
+};
+
 /**
  * struct ieee80211_mesh_fast_tx - cached mesh fast tx entry
  * @rhash: rhashtable pointer
- * @addr_key: The Ethernet DA which is the key for this entry
+ * @key: the lookup key for this cache entry
  * @fast_tx: base fast_tx data
  * @hdr: cached mesh and rfc1042 headers
  * @hdrlen: length of mesh + rfc1042

@@ -147,7 +176,7 @@ struct mesh_path {
  */
 struct ieee80211_mesh_fast_tx {
 	struct rhash_head rhash;
-	u8 addr_key[ETH_ALEN] __aligned(2);
+	struct ieee80211_mesh_fast_tx_key key;
 
 	struct ieee80211_fast_tx fast_tx;
 	u8 hdr[sizeof(struct ieee80211s_hdr) + sizeof(rfc1042_header)];

@@ -333,7 +362,8 @@ void mesh_path_tx_root_frame(struct ieee80211_sub_if_data *sdata);
 
 bool mesh_action_is_path_sel(struct ieee80211_mgmt *mgmt);
 struct ieee80211_mesh_fast_tx *
-mesh_fast_tx_get(struct ieee80211_sub_if_data *sdata, const u8 *addr);
+mesh_fast_tx_get(struct ieee80211_sub_if_data *sdata,
+		 struct ieee80211_mesh_fast_tx_key *key);
 bool ieee80211_mesh_xmit_fast(struct ieee80211_sub_if_data *sdata,
 			      struct sk_buff *skb, u32 ctrl_flags);
 void mesh_fast_tx_cache(struct ieee80211_sub_if_data *sdata,

net/mac80211/mesh_pathtbl.c +22 -9

index c7173190f9b93..59f7264194ce3 100644
@@ -36,8 +36,8 @@ static const struct rhashtable_params mesh_rht_params = {
 static const struct rhashtable_params fast_tx_rht_params = {
 	.nelem_hint = 10,
 	.automatic_shrinking = true,
-	.key_len = ETH_ALEN,
-	.key_offset = offsetof(struct ieee80211_mesh_fast_tx, addr_key),
+	.key_len = sizeof_field(struct ieee80211_mesh_fast_tx, key),
+	.key_offset = offsetof(struct ieee80211_mesh_fast_tx, key),
 	.head_offset = offsetof(struct ieee80211_mesh_fast_tx, rhash),
 	.hashfn = mesh_table_hash,
 };

@@ -426,20 +426,21 @@ static void mesh_fast_tx_entry_free(struct mesh_tx_cache *cache,
 }
 
 struct ieee80211_mesh_fast_tx *
-mesh_fast_tx_get(struct ieee80211_sub_if_data *sdata, const u8 *addr)
+mesh_fast_tx_get(struct ieee80211_sub_if_data *sdata,
+		 struct ieee80211_mesh_fast_tx_key *key)
 {
 	struct ieee80211_mesh_fast_tx *entry;
 	struct mesh_tx_cache *cache;
 
 	cache = &sdata->u.mesh.tx_cache;
-	entry = rhashtable_lookup(&cache->rht, addr, fast_tx_rht_params);
+	entry = rhashtable_lookup(&cache->rht, key, fast_tx_rht_params);
 	if (!entry)
 		return NULL;
 
 	if (!(entry->mpath->flags & MESH_PATH_ACTIVE) ||
 	    mpath_expired(entry->mpath)) {
 		spin_lock_bh(&cache->walk_lock);
-		entry = rhashtable_lookup(&cache->rht, addr, fast_tx_rht_params);
+		entry = rhashtable_lookup(&cache->rht, key, fast_tx_rht_params);
 		if (entry)
 		    mesh_fast_tx_entry_free(cache, entry);
 		spin_unlock_bh(&cache->walk_lock);

@@ -484,18 +485,24 @@ void mesh_fast_tx_cache(struct ieee80211_sub_if_data *sdata,
 	if (!sta)
 		return;
 
+	build.key.type = MESH_FAST_TX_TYPE_LOCAL;
 	if ((meshhdr->flags & MESH_FLAGS_AE) == MESH_FLAGS_AE_A5_A6) {
 		/* This is required to keep the mppath alive */
 		mppath = mpp_path_lookup(sdata, meshhdr->eaddr1);
 		if (!mppath)
 			return;
 		build.mppath = mppath;
+		if (!ether_addr_equal(meshhdr->eaddr2, sdata->vif.addr))
+			build.key.type = MESH_FAST_TX_TYPE_PROXIED;
 	} else if (ieee80211_has_a4(hdr->frame_control)) {
 		mppath = mpath;
 	} else {
 		return;
 	}
 
+	if (!ether_addr_equal(hdr->addr4, sdata->vif.addr))
+		build.key.type = MESH_FAST_TX_TYPE_FORWARDED;
+
 	/* rate limit, in case fast xmit can't be enabled */
 	if (mppath->fast_tx_check == jiffies)
 		return;

@@ -542,7 +549,7 @@ void mesh_fast_tx_cache(struct ieee80211_sub_if_data *sdata,
 		}
 	}
 
-	memcpy(build.addr_key, mppath->dst, ETH_ALEN);
+	memcpy(build.key.addr, mppath->dst, ETH_ALEN);
 	build.timestamp = jiffies;
 	build.fast_tx.band = info->band;
 	build.fast_tx.da_offs = offsetof(struct ieee80211_hdr, addr3);

@@ -641,12 +648,18 @@ void mesh_fast_tx_flush_addr(struct ieee80211_sub_if_data *sdata,
 			     const u8 *addr)
 {
 	struct mesh_tx_cache *cache = &sdata->u.mesh.tx_cache;
+	struct ieee80211_mesh_fast_tx_key key = {};
 	struct ieee80211_mesh_fast_tx *entry;
+	int i;
 
+	ether_addr_copy(key.addr, addr);
 	spin_lock_bh(&cache->walk_lock);
-	entry = rhashtable_lookup_fast(&cache->rht, addr, fast_tx_rht_params);
-	if (entry)
-		mesh_fast_tx_entry_free(cache, entry);
+	for (i = 0; i < NUM_MESH_FAST_TX_TYPE; i++) {
+		key.type = i;
+		entry = rhashtable_lookup_fast(&cache->rht, &key, fast_tx_rht_params);
+		if (entry)
+			mesh_fast_tx_entry_free(cache, entry);
+	}
 	spin_unlock_bh(&cache->walk_lock);
 }

net/mac80211/rx.c +10 -3

index 26ca2f5dc52b2..604863cebc198 100644
@@ -2726,7 +2726,10 @@ ieee80211_rx_mesh_fast_forward(struct ieee80211_sub_if_data *sdata,
 			       struct sk_buff *skb, int hdrlen)
 {
 	struct ieee80211_if_mesh *ifmsh = &sdata->u.mesh;
-	struct ieee80211_mesh_fast_tx *entry = NULL;
+	struct ieee80211_mesh_fast_tx_key key = {
+		.type = MESH_FAST_TX_TYPE_FORWARDED
+	};
+	struct ieee80211_mesh_fast_tx *entry;
 	struct ieee80211s_hdr *mesh_hdr;
 	struct tid_ampdu_tx *tid_tx;
 	struct sta_info *sta;

@@ -2735,9 +2738,13 @@ ieee80211_rx_mesh_fast_forward(struct ieee80211_sub_if_data *sdata,
 
 	mesh_hdr = (struct ieee80211s_hdr *)(skb->data + sizeof(eth));
 	if ((mesh_hdr->flags & MESH_FLAGS_AE) == MESH_FLAGS_AE_A5_A6)
-		entry = mesh_fast_tx_get(sdata, mesh_hdr->eaddr1);
+		ether_addr_copy(key.addr, mesh_hdr->eaddr1);
 	else if (!(mesh_hdr->flags & MESH_FLAGS_AE))
-		entry = mesh_fast_tx_get(sdata, skb->data);
+		ether_addr_copy(key.addr, skb->data);
+	else
+		return false;
+
+	entry = mesh_fast_tx_get(sdata, &key);
 	if (!entry)
 		return false;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:09 UTC #38 lore raw

PATCH 38/186 wifi: iwlwifi: mvm: remove old PASN station when adding a new one

Link https://msgid.link/20240415114847.ef3544a416f2.I4e8c7c8ca22737f4f908ae5cd4fc0b920c703dd3@changeid open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Avraham Stern <[email protected]>

[ Upstream commit dbfff5bf9292714f02ace002fea8ce6599ea1145 ]

If a PASN station is added, and an old PASN station already exists for the same mac address, remove the old station before adding the new one. Keeping the old station caueses old security context to be used in measurements.

Fixes 0739a7d70e00 ("iwlwifi: mvm: initiator: add option for adding a PASN responder")

Signed-off-by Avraham Stern <[email protected]>

Signed-off-by Miri Korenblit <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c | 2 ++
 1 file changed, 2 insertions(+)

drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c +2 -0

index 233ae81884a0e..ae0eb585b61ee 100644
@@ -53,6 +53,8 @@ int iwl_mvm_ftm_add_pasn_sta(struct iwl_mvm *mvm, struct ieee80211_vif *vif,
 	if (!pasn)
 		return -ENOBUFS;
 
+	iwl_mvm_ftm_remove_pasn_sta(mvm, addr);
+
 	pasn->cipher = iwl_mvm_cipher_to_location_cipher(cipher);
 
 	switch (pasn->cipher) {

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:09 UTC #39 lore raw

PATCH 39/186 wifi: iwlwifi: mvm: return uid from iwl_mvm_build_scan_cmd

Link https://msgid.link/20240415114847.5e2d602b3190.I4c4931021be74a67a869384c8f8ee7463e0c7857@changeid open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Miri Korenblit <[email protected]>

[ Upstream commit bada85a3f584763deadd201147778c3e791d279c ]

This function is supposed to return a uid on success, and an errno in failure.
But it currently returns the return value of the specific cmd version handler, which in turn returns 0 on success and errno otherwise. This means that on success, iwl_mvm_build_scan_cmd will return 0 regardless if the actual uid.
Fix this by returning the uid if the handler succeeded.

Fixes 687db6ff5b70 ("iwlwifi: scan: make new scan req versioning flow")

Signed-off-by Miri Korenblit <[email protected]>

Reviewed-by Ilan Peer <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/wireless/intel/iwlwifi/mvm/scan.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

drivers/net/wireless/intel/iwlwifi/mvm/scan.c +2 -1

index 3cbe2c0b8d6bc..03ec900a33433 100644
@@ -2819,7 +2819,8 @@ static int iwl_mvm_build_scan_cmd(struct iwl_mvm *mvm,
 		if (ver_handler->version != scan_ver)
 			continue;
 
-		return ver_handler->handler(mvm, vif, params, type, uid);
+		err = ver_handler->handler(mvm, vif, params, type, uid);
+		return err ? : uid;
 	}
 
 	err = iwl_mvm_scan_umac(mvm, vif, params, type, uid);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:09 UTC #40 lore raw

PATCH 40/186 drm/gma500: Remove lid code

Link https://patchwork.freedesktop.org/patch/msgid/[email protected]

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Patrik Jakobsson <[email protected]>

[ Upstream commit 6aff4c26ed677b1f464f721fbd3e7767f24a684d ]

Due to a change in the order of initialization, the lid timer got started before proper setup was made. This resulted in a crash during boot.

The lid switch is handled by gma500 through a timer that periodically polls the opregion for changes. These types of ACPI events shouldn't be handled by the graphics driver so let's get rid of the lid code. This fixes the crash during boot.

Reported-by Enrico Bartky <[email protected]>

Fixes 8f1aaccb04b7 ("drm/gma500: Implement client-based fbdev emulation")

Tested-by Enrico Bartky <[email protected]>

Signed-off-by Patrik Jakobsson <[email protected]>

Reviewed-by Thomas Zimmermann <[email protected]>

Signed-off-by Thomas Zimmermann <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/gpu/drm/gma500/Makefile     |  1 -
 drivers/gpu/drm/gma500/psb_device.c |  5 +-
 drivers/gpu/drm/gma500/psb_drv.h    |  9 ----
 drivers/gpu/drm/gma500/psb_lid.c    | 80 -----------------------------
 4 files changed, 1 insertion(+), 94 deletions(-)
 delete mode 100644 drivers/gpu/drm/gma500/psb_lid.c

drivers/gpu/drm/gma500/Makefile +0 -1

index 4f302cd5e1a6c..58fed80c7392a 100644
@@ -34,7 +34,6 @@ gma500_gfx-y += \
 	  psb_intel_lvds.o \
 	  psb_intel_modes.o \
 	  psb_intel_sdvo.o \
-	  psb_lid.o \
 	  psb_irq.o
 
 gma500_gfx-$(CONFIG_ACPI) +=  opregion.o

drivers/gpu/drm/gma500/psb_device.c +1 -4

index dcfcd7b89d4a1..6dece8f0e380f 100644
@@ -73,8 +73,7 @@ static int psb_backlight_setup(struct drm_device *dev)
 	}
 
 	psb_intel_lvds_set_brightness(dev, PSB_MAX_BRIGHTNESS);
-	/* This must occur after the backlight is properly initialised */
-	psb_lid_timer_init(dev_priv);
+
 	return 0;
 }

@@ -259,8 +258,6 @@ static int psb_chip_setup(struct drm_device *dev)
 
 static void psb_chip_teardown(struct drm_device *dev)
 {
-	struct drm_psb_private *dev_priv = to_drm_psb_private(dev);
-	psb_lid_timer_takedown(dev_priv);
 	gma_intel_teardown_gmbus(dev);
 }

drivers/gpu/drm/gma500/psb_drv.h +0 -9

index 70d9adafa2333..bb1cd45c085cd 100644
@@ -170,7 +170,6 @@
 
 #define PSB_NUM_VBLANKS 2
 #define PSB_WATCHDOG_DELAY (HZ * 2)
-#define PSB_LID_DELAY (HZ / 10)
 
 #define PSB_MAX_BRIGHTNESS		100

@@ -499,11 +498,7 @@ struct drm_psb_private {
 	/* Hotplug handling */
 	struct work_struct hotplug_work;
 
-	/* LID-Switch */
-	spinlock_t lid_lock;
-	struct timer_list lid_timer;
 	struct psb_intel_opregion opregion;
-	u32 lid_last_state;
 
 	/* Watchdog */
 	uint32_t apm_reg;

@@ -599,10 +594,6 @@ struct psb_ops {
 	int i2c_bus;		/* I2C bus identifier for Moorestown */
 };
 
-/* psb_lid.c */
-extern void psb_lid_timer_init(struct drm_psb_private *dev_priv);
-extern void psb_lid_timer_takedown(struct drm_psb_private *dev_priv);
-
 /* modesetting */
 extern void psb_modeset_init(struct drm_device *dev);
 extern void psb_modeset_cleanup(struct drm_device *dev);

drivers/gpu/drm/gma500/psb_lid.c deleted +0 ~~-80~~

80 lines not on this page · raw message

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:09 UTC #41 lore raw

PATCH 41/186 wifi: mac80211_hwsim: init peer measurement result

Link https://msgid.link/20240418105220.e1317621c1f9.If7dd447de24d7493d133284db5e9e482e4e299f8@changeid open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Johannes Berg <[email protected]>

[ Upstream commit 2a4e01e5270b9fa9f6e6e0a4c24ac51a758636f9 ]

If we don't get all the values here, we might pass them to cfg80211 uninitialized. Fix that, even if the input might then not make much sense.

Fixes 2af3b2a631b1 ("mac80211_hwsim: add PMSR report support via virtio")

Reviewed-by Miriam Rachel Korenblit <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/wireless/virtual/mac80211_hwsim.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

drivers/net/wireless/virtual/mac80211_hwsim.c +1 -1

index f5a0880da3fcc..07be0adc13ec5 100644
@@ -3795,7 +3795,7 @@ static int hwsim_pmsr_report_nl(struct sk_buff *msg, struct genl_info *info)
 	}
 
 	nla_for_each_nested(peer, peers, rem) {
-		struct cfg80211_pmsr_result result;
+		struct cfg80211_pmsr_result result = {};
 
 		err = mac80211_hwsim_parse_pmsr_result(peer, &result, info);
 		if (err)

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:09 UTC #42 lore raw

PATCH 42/186 wifi: mac80211: remove link before AP

Link https://msgid.link/20240418105220.03ac4a09fa74.Ifb8c8d38e3402721a81ce5981568f47b5c5889cb@changeid open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Johannes Berg <[email protected]>

[ Upstream commit cb55e08dba3526796e35d24a6d5db4ed6dcb8a4b ]

If the AP removal timer is long, we don't really want to remove the link immediately. However, we really should do it _before_ the AP removes it (which happens at or after count reaches 0), so subtract 1 from the countdown when scheduling the timer. This causes the link removal work to run just after the beacon with value 1 is received. If the counter is already zero, do it immediately.

This fixes an issue where we do the removal too late and receive a beacon from the AP that's no longer associated with the MLD, but thus removed EHT and ML elements, and then we disconnect instead from the whole MLD, since one of the associated APs changed mode from EHT to HE.

Fixes 8eb8dd2ffbbb ("wifi: mac80211: Support link removal using Reconfiguration ML element")

Reviewed-by Ilan Peer <[email protected]>

Reviewed-by Miriam Rachel Korenblit <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/mac80211/mlme.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

net/mac80211/mlme.c +5 -2

index c6044ab4e7fc1..e3e769b2f2ef1 100644
@@ -5884,8 +5884,11 @@ static void ieee80211_ml_reconfiguration(struct ieee80211_sub_if_data *sdata,
 			continue;
 		}
 
-		link_delay = link_conf->beacon_int *
-			link_removal_timeout[link_id];
+		if (link_removal_timeout[link_id] < 1)
+			link_delay = 0;
+		else
+			link_delay = link_conf->beacon_int *
+				(link_removal_timeout[link_id] - 1);
 
 		if (!delay)
 			delay = link_delay;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:07 UTC #43 lore raw

PATCH 43/186 wifi: mac80211: fix unaligned le16 access

Link https://msgid.link/20240418105220.356788ba0045.I2b3cdb3644e205d5bb10322c345c0499171cf5d2@changeid open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Johannes Berg <[email protected]>

[ Upstream commit c53d8a59351e4347452e263e2e5d7446ec93da83 ]

The AP removal timer field need not be aligned, so the code shouldn't access it directly, but use unaligned loads. Use get_unaligned_le16(), which even is shorter than the current code since it doesn't need a cast.

Fixes 8eb8dd2ffbbb ("wifi: mac80211: Support link removal using Reconfiguration ML element")

Reviewed-by Ilan Peer <[email protected]>

Reviewed-by Miriam Rachel Korenblit <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/mac80211/mlme.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

net/mac80211/mlme.c +1 -1

index e3e769b2f2ef1..6e574e2adc22e 100644
@@ -5859,7 +5859,7 @@ static void ieee80211_ml_reconfiguration(struct ieee80211_sub_if_data *sdata,
 		 */
 		if (control &
 		    IEEE80211_MLE_STA_RECONF_CONTROL_AP_REM_TIMER_PRESENT)
-			link_removal_timeout[link_id] = le16_to_cpu(*(__le16 *)pos);
+			link_removal_timeout[link_id] = get_unaligned_le16(pos);
 	}
 
 	removed_links &= sdata->vif.valid_links;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:07 UTC #44 lore raw

PATCH 44/186 net: libwx: fix alloc msix vectors failed

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Duanqiang Wen <[email protected]>

[ Upstream commit 69197dfc64007b5292cc960581548f41ccd44828 ]

driver needs queue msix vectors and one misc irq vector, but only queue vectors need irq affinity.
when num_online_cpus is less than chip max msix vectors, driver will acquire (num_online_cpus + 1) vecotrs, and call pci_alloc_irq_vectors_affinity functions with affinity params without setting pre_vectors or post_vectors, it will cause return error code -ENOSPC.
Misc irq vector is vector 0, driver need to set affinity params .pre_vectors = 1.

Fixes 3f703186113f ("net: libwx: Add irq flow functions")

Signed-off-by Duanqiang Wen <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/wangxun/libwx/wx_lib.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

drivers/net/ethernet/wangxun/libwx/wx_lib.c +1 -1

index e078f4071dc23..be434c833c69c 100644
@@ -1585,7 +1585,7 @@ static void wx_set_num_queues(struct wx *wx)
  */
 static int wx_acquire_msix_vectors(struct wx *wx)
 {
-	struct irq_affinity affd = {0, };
+	struct irq_affinity affd = { .pre_vectors = 1 };
 	int nvecs, i;
 
 	nvecs = min_t(int, num_online_cpus(), wx->mac.max_msix_vectors);

Signature · 1 line

2.43.0

[email protected]

2024-09-27 01:55 UTC #45 lore raw

答复: [PATCH 6.6 044/186] net: libwx: fix alloc msix vectors failed

Quoted 10 lines

-----邮件原件-----
发件人: Greg Kroah-Hartman <[email protected]>
发送时间: 2024年4月30日 18:38
收件人: [email protected]
抄送: Greg Kroah-Hartman <[email protected]>; [email protected]; Duanqiang Wen <[email protected]>; David S. Miller <[email protected]>; Sasha Levin <[email protected]> 主题: [PATCH 6.6 044/186] net: libwx: fix alloc msix vectors failed
6.6-stable review patch. If anyone has any objections, please let me

know.

Quoted 8 lines

------------------
From: Duanqiang Wen <[email protected]>
[ Upstream commit 69197dfc64007b5292cc960581548f41ccd44828 ]
driver needs queue msix vectors and one misc irq vector, but only queue

vectors

Quoted 2 lines

need irq affinity.
when num_online_cpus is less than chip max msix vectors, driver will

acquire

Quoted 1 line

(num_online_cpus + 1) vecotrs, and call pci_alloc_irq_vectors_affinity

functions

Quoted 1 line

with affinity params without setting pre_vectors or post_vectors, it will

cause

Quoted 2 lines

return error code -ENOSPC.
Misc irq vector is vector 0, driver need to set affinity params

.pre_vectors = 1.

Quoted 23 lines

Fixes 3f703186113f ("net: libwx: Add irq flow functions")
Signed-off-by Duanqiang Wen <[email protected]>
Signed-off-by David S. Miller <[email protected]>
Signed-off-by Sasha Levin <[email protected]>
---
 drivers/net/ethernet/wangxun/libwx/wx_lib.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/wangxun/libwx/wx_lib.c +1 -2
index e078f4071dc23..be434c833c69c 100644
@@ -1585,7 +1585,7 @@ static void wx_set_num_queues(struct wx *wx)
  */
 static int wx_acquire_msix_vectors(struct wx *wx)  {
-	struct irq_affinity affd = {0, };
+	struct irq_affinity affd = { .pre_vectors = 1 };
 	int nvecs, i;
 
 	nvecs = min_t(int, num_online_cpus(), wx->mac.max_msix_vectors);
--

Quoted 3 lines

2.43.0

This patch in kernel-6.6 and kernel 6.7 will cause problems. In kernel-6.6 and kernel 6.7,
Wangxun txgbe & ngbe driver adjust misc irq to vector 0 not yet. How to revert it in
kernel-6.6 and kernel-6.7 stable?

Quoted 1 line

[email protected]

2024-09-27 02:02 UTC #46 lore raw

RE: [PATCH 6.6 044/186] net: libwx: fix alloc msix vectors failed

Quoted 10 lines

-----Original Message-----
From: Greg Kroah-Hartman <[email protected]>
Sent: 2024年4月30日 18:38
To: [email protected]
Cc: Greg Kroah-Hartman <[email protected]>; [email protected]; Duanqiang Wen <[email protected]>; David S. Miller <[email protected]>; Sasha Levin <[email protected]>
Subject: [PATCH 6.6 044/186] net: libwx: fix alloc msix vectors failed
6.6-stable review patch. If anyone has any objections, please let me

know.

Quoted 10 lines

------------------
From: Duanqiang Wen <[email protected]>
[ Upstream commit 69197dfc64007b5292cc960581548f41ccd44828 ]
driver needs queue msix vectors and one misc irq vector, but only queue vectors need irq affinity.
when num_online_cpus is less than chip max msix vectors, driver will

acquire

Quoted 2 lines

(num_online_cpus + 1) vecotrs, and call pci_alloc_irq_vectors_affinity functions with affinity params without setting pre_vectors or

post_vectors, it

Quoted 2 lines

will cause return error code -ENOSPC.
Misc irq vector is vector 0, driver need to set affinity params

.pre_vectors = 1.

Quoted 23 lines

Fixes 3f703186113f ("net: libwx: Add irq flow functions")
Signed-off-by Duanqiang Wen <[email protected]>
Signed-off-by David S. Miller <[email protected]>
Signed-off-by Sasha Levin <[email protected]>
---
 drivers/net/ethernet/wangxun/libwx/wx_lib.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/wangxun/libwx/wx_lib.c +1 -2
index e078f4071dc23..be434c833c69c 100644
@@ -1585,7 +1585,7 @@ static void wx_set_num_queues(struct wx *wx)
  */
 static int wx_acquire_msix_vectors(struct wx *wx)  {
-	struct irq_affinity affd = {0, };
+	struct irq_affinity affd = { .pre_vectors = 1 };
 	int nvecs, i;
 
 	nvecs = min_t(int, num_online_cpus(), wx->mac.max_msix_vectors);
--

Quoted 2 lines

2.43.0

Quoted 1 line

'Greg Kroah-Hartman'

author maintainer

2024-09-27 06:53 UTC #47 lore raw

Re: [PATCH 6.6 044/186] net: libwx: fix alloc msix vectors failed

Quoted 49 lines · On Fri, Sep 27, 2024 at 10:02:14AM +0800, [email protected] wrote:

Quoted 10 lines
-----Original Message-----
From: Greg Kroah-Hartman <[email protected]>
Sent: 2024年4月30日 18:38
To: [email protected]
Cc: Greg Kroah-Hartman <[email protected]>; [email protected]; Duanqiang Wen <[email protected]>; David S. Miller <[email protected]>; Sasha Levin <[email protected]>
Subject: [PATCH 6.6 044/186] net: libwx: fix alloc msix vectors failed
6.6-stable review patch. If anyone has any objections, please let me
know.
Quoted 10 lines
------------------
From: Duanqiang Wen <[email protected]>
[ Upstream commit 69197dfc64007b5292cc960581548f41ccd44828 ]
driver needs queue msix vectors and one misc irq vector, but only queue vectors need irq affinity.
when num_online_cpus is less than chip max msix vectors, driver will
acquire
Quoted 2 lines
(num_online_cpus + 1) vecotrs, and call pci_alloc_irq_vectors_affinity functions with affinity params without setting pre_vectors or
post_vectors, it
Quoted 2 lines
will cause return error code -ENOSPC.
Misc irq vector is vector 0, driver need to set affinity params
.pre_vectors = 1.
Quoted 21 lines
Fixes 3f703186113f ("net: libwx: Add irq flow functions")
Signed-off-by Duanqiang Wen <[email protected]>
Signed-off-by David S. Miller <[email protected]>
Signed-off-by Sasha Levin <[email protected]>
---
 drivers/net/ethernet/wangxun/libwx/wx_lib.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/wangxun/libwx/wx_lib.c +1 -1
index e078f4071dc23..be434c833c69c 100644
@@ -1585,7 +1585,7 @@ static void wx_set_num_queues(struct wx *wx)
  */
 static int wx_acquire_msix_vectors(struct wx *wx)  {
-	struct irq_affinity affd = {0, };
+	struct irq_affinity affd = { .pre_vectors = 1 };
 	int nvecs, i;
 

Quoted 9 lines

Quoted 4 lines
drivers/net/ethernet/wangxun/libwx/wx_lib.c +0 -1
 	nvecs = min_t(int, num_online_cpus(), wx->mac.max_msix_vectors);
--
2.43.0
This patch in kernel-6.6 and kernel 6.7 will cause problems. In kernel-6.6 and kernel 6.7,
Wangxun txgbe & ngbe driver adjust misc irq to vector 0 not yet. How to revert it in
kernel-6.6 and kernel-6.7 stable?>

Please send a revert.

thanks,

greg k-h

[email protected]

2024-09-27 07:00 UTC #48 lore raw

RE: [PATCH 6.6 044/186] net: libwx: fix alloc msix vectors failed

Quoted 73 lines

-----Original Message-----
From: 'Greg Kroah-Hartman' <[email protected]>
Sent: 2024年9月27日 14:53
To: [email protected]
Cc: [email protected]; [email protected]; 'David S. Miller' <[email protected]>; 'Sasha Levin' <[email protected]>
Subject: Re: [PATCH 6.6 044/186] net: libwx: fix alloc msix vectors failed
Quoted 6 lines · On Fri, Sep 27, 2024 at 10:02:14AM +0800, [email protected] wrote:
Quoted 6 lines
-----Original Message-----
From: Greg Kroah-Hartman <[email protected]>
Sent: 2024年4月30日 18:38
To: [email protected]
Cc: Greg Kroah-Hartman <[email protected]>; [email protected]; Duanqiang Wen <duanqiangwen@net-
swift.com>;
Quoted 47 lines
Quoted 7 lines
David S. Miller <[email protected]>; Sasha Levin <[email protected]>
Subject: [PATCH 6.6 044/186] net: libwx: fix alloc msix vectors failed
6.6-stable review patch. If anyone has any objections, please let me
know.
Quoted 10 lines
------------------
From: Duanqiang Wen <[email protected]>
[ Upstream commit 69197dfc64007b5292cc960581548f41ccd44828 ]
driver needs queue msix vectors and one misc irq vector, but only queue vectors need irq affinity.
when num_online_cpus is less than chip max msix vectors, driver will
acquire
Quoted 3 lines
(num_online_cpus + 1) vecotrs, and call pci_alloc_irq_vectors_affinity functions with affinity params without setting pre_vectors or
post_vectors, it
Quoted 2 lines
will cause return error code -ENOSPC.
Misc irq vector is vector 0, driver need to set affinity params
.pre_vectors = 1.
Quoted 21 lines
Fixes 3f703186113f ("net: libwx: Add irq flow functions")
Signed-off-by Duanqiang Wen <[email protected]>
Signed-off-by David S. Miller <[email protected]>
Signed-off-by Sasha Levin <[email protected]>
---
 drivers/net/ethernet/wangxun/libwx/wx_lib.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/wangxun/libwx/wx_lib.c +1 -1
index e078f4071dc23..be434c833c69c 100644
@@ -1585,7 +1585,7 @@ static void wx_set_num_queues(struct wx *wx)
  */
 static int wx_acquire_msix_vectors(struct wx *wx)  {
-	struct irq_affinity affd = {0, };
+	struct irq_affinity affd = { .pre_vectors = 1 };
 	int nvecs, i;
 
Quoted 8 lines
Quoted 4 lines
drivers/net/ethernet/wangxun/libwx/wx_lib.c +0 -1
 	nvecs = min_t(int, num_online_cpus(), wx->mac.max_msix_vectors);
--
2.43.0
This patch in kernel-6.6 and kernel 6.7 will cause problems. In kernel-6.6 and kernel 6.7, Wangxun txgbe & ngbe driver adjust misc irq to vector 0 not yet. How to revert it in
kernel-6.6 and kernel-6.7 stable?>

Quoted 6 lines

Please send a revert.
thanks,
greg k-h

Should I send revert to stable/master repo? I only wan't to revert it in 6.6.y and 6.7.y. How maintainer recognize this revert should be applied in different tags?

'Greg Kroah-Hartman'

author maintainer

2024-09-27 07:07 UTC #49 lore raw

Re: [PATCH 6.6 044/186] net: libwx: fix alloc msix vectors failed

Quoted 73 lines · On Fri, Sep 27, 2024 at 02:59:42PM +0800, [email protected] wrote:

Quoted 73 lines
-----Original Message-----
From: 'Greg Kroah-Hartman' <[email protected]>
Sent: 2024年9月27日 14:53
To: [email protected]
Cc: [email protected]; [email protected]; 'David S. Miller' <[email protected]>; 'Sasha Levin' <[email protected]>
Subject: Re: [PATCH 6.6 044/186] net: libwx: fix alloc msix vectors failed
Quoted 6 lines · On Fri, Sep 27, 2024 at 10:02:14AM +0800, [email protected] wrote:
Quoted 6 lines
-----Original Message-----
From: Greg Kroah-Hartman <[email protected]>
Sent: 2024年4月30日 18:38
To: [email protected]
Cc: Greg Kroah-Hartman <[email protected]>; [email protected]; Duanqiang Wen <duanqiangwen@net-
swift.com>;
Quoted 55 lines
Quoted 7 lines
David S. Miller <[email protected]>; Sasha Levin <[email protected]>
Subject: [PATCH 6.6 044/186] net: libwx: fix alloc msix vectors failed
6.6-stable review patch. If anyone has any objections, please let me
know.
Quoted 10 lines
------------------
From: Duanqiang Wen <[email protected]>
[ Upstream commit 69197dfc64007b5292cc960581548f41ccd44828 ]
driver needs queue msix vectors and one misc irq vector, but only queue vectors need irq affinity.
when num_online_cpus is less than chip max msix vectors, driver will
acquire
Quoted 3 lines
(num_online_cpus + 1) vecotrs, and call pci_alloc_irq_vectors_affinity functions with affinity params without setting pre_vectors or
post_vectors, it
Quoted 2 lines
will cause return error code -ENOSPC.
Misc irq vector is vector 0, driver need to set affinity params
.pre_vectors = 1.
Quoted 23 lines
Fixes 3f703186113f ("net: libwx: Add irq flow functions")
Signed-off-by Duanqiang Wen <[email protected]>
Signed-off-by David S. Miller <[email protected]>
Signed-off-by Sasha Levin <[email protected]>
---
 drivers/net/ethernet/wangxun/libwx/wx_lib.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/wangxun/libwx/wx_lib.c +1 -2
index e078f4071dc23..be434c833c69c 100644
@@ -1585,7 +1585,7 @@ static void wx_set_num_queues(struct wx *wx)
  */
 static int wx_acquire_msix_vectors(struct wx *wx)  {
-	struct irq_affinity affd = {0, };
+	struct irq_affinity affd = { .pre_vectors = 1 };
 	int nvecs, i;
 
 	nvecs = min_t(int, num_online_cpus(), wx->mac.max_msix_vectors);
--
Quoted 2 lines
2.43.0
This patch in kernel-6.6 and kernel 6.7 will cause problems. In kernel-6.6 and kernel 6.7, Wangxun txgbe & ngbe driver adjust misc irq to vector 0 not yet. How to revert it in
kernel-6.6 and kernel-6.7 stable?>

Quoted 7 lines

Quoted 6 lines
Please send a revert.
thanks,
greg k-h
Should I send revert to stable/master repo? I only wan't to revert it in 6.6.y and 6.7.y.

6.7 is no longer being maintained, please see the front page of kernel.org for the listed versions that are.

Quoted 1 line

How maintainer recognize this revert should be applied in different tags?

Explain in the changelog why this is only relevant for one branch and not all the others.

thanks,

greg k-h

Greg Kroah-Hartman

author maintainer

2024-04-30 11:07 UTC #50 lore raw

PATCH 45/186 vxlan: drop packets from invalid src-address

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: David Bauer <[email protected]>

[ Upstream commit f58f45c1e5b92975e91754f5407250085a6ae7cf ]

The VXLAN driver currently does not check if the inner layer2 source-address is valid.

In case source-address snooping/learning is enabled, a entry in the FDB for the invalid address is created with the layer3 address of the tunnel endpoint.

If the frame happens to have a non-unicast address set, all this non-unicast traffic is subsequently not flooded to the tunnel network but sent to the learnt host in the FDB. To make matters worse, this FDB entry does not expire.

Apply the same filtering for packets as it is done for bridges. This not only drops these invalid packets but avoids them from being learnt into the FDB.

Fixes d342894c5d2f ("vxlan: virtual extensible lan")

Signed-off-by David Bauer <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/vxlan/vxlan_core.c | 4 ++++
 1 file changed, 4 insertions(+)

drivers/net/vxlan/vxlan_core.c +4 -0

index 99ede13124194..ecdf0276004f9 100644
@@ -1615,6 +1615,10 @@ static bool vxlan_set_mac(struct vxlan_dev *vxlan,
 	if (ether_addr_equal(eth_hdr(skb)->h_source, vxlan->dev->dev_addr))
 		return false;
 
+	/* Ignore packets from invalid src-address */
+	if (!is_valid_ether_addr(eth_hdr(skb)->h_source))
+		return false;
+
 	/* Get address from the outer IP header */
 	if (vxlan_get_sk_family(vs) == AF_INET) {
 		saddr.sin.sin_addr.s_addr = ip_hdr(skb)->saddr;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:07 UTC #51 lore raw

PATCH 46/186 net: bcmasp: fix memory leak when bringing down interface

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Justin Chen <[email protected]>

[ Upstream commit 9f898fc2c31fbf0ac5ecd289f528a716464cb005 ]

When bringing down the TX rings we flush the rings but forget to reclaimed the flushed packets. This leads to a memory leak since we do not free the dma mapped buffers. This also leads to tx control block corruption when bringing down the interface for power management.

Fixes 490cb412007d ("net: bcmasp: Add support for ASP2.0 Ethernet controller")

Signed-off-by Justin Chen <[email protected]>

Acked-by Florian Fainelli <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 .../net/ethernet/broadcom/asp2/bcmasp_intf.c  | 21 ++++++++++++-------
 1 file changed, 14 insertions(+), 7 deletions(-)

drivers/net/ethernet/broadcom/asp2/bcmasp_intf.c +14 -7

index b3d04f49f77e9..6bf149d645941 100644
@@ -435,10 +435,8 @@ static void umac_init(struct bcmasp_intf *intf)
 	umac_wl(intf, 0x800, UMC_RX_MAX_PKT_SZ);
 }
 
-static int bcmasp_tx_poll(struct napi_struct *napi, int budget)
+static int bcmasp_tx_reclaim(struct bcmasp_intf *intf)
 {
-	struct bcmasp_intf *intf =
-		container_of(napi, struct bcmasp_intf, tx_napi);
 	struct bcmasp_intf_stats64 *stats = &intf->stats64;
 	struct device *kdev = &intf->parent->pdev->dev;
 	unsigned long read, released = 0;

@@ -481,10 +479,16 @@ static int bcmasp_tx_poll(struct napi_struct *napi, int budget)
 							DESC_RING_COUNT);
 	}
 
-	/* Ensure all descriptors have been written to DRAM for the hardware
-	 * to see updated contents.
-	 */
-	wmb();
+	return released;
+}
+
+static int bcmasp_tx_poll(struct napi_struct *napi, int budget)
+{
+	struct bcmasp_intf *intf =
+		container_of(napi, struct bcmasp_intf, tx_napi);
+	int released = 0;
+
+	released = bcmasp_tx_reclaim(intf);
 
 	napi_complete(&intf->tx_napi);

@@ -794,6 +798,7 @@ static int bcmasp_init_tx(struct bcmasp_intf *intf)
 
 	intf->tx_spb_index = 0;
 	intf->tx_spb_clean_index = 0;
+	memset(intf->tx_cbs, 0, sizeof(struct bcmasp_tx_cb) * DESC_RING_COUNT);
 
 	netif_napi_add_tx(intf->ndev, &intf->tx_napi, bcmasp_tx_poll);

@@ -904,6 +909,8 @@ static void bcmasp_netif_deinit(struct net_device *dev)
 	} while (timeout-- > 0);
 	tx_spb_dma_wl(intf, 0x0, TX_SPB_DMA_FIFO_CTRL);
 
+	bcmasp_tx_reclaim(intf);
+
 	umac_enable_set(intf, UMC_CMD_TX_EN, 0);
 
 	phy_stop(dev->phydev);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:07 UTC #52 lore raw

PATCH 47/186 mlxsw: core: Unregister EMAD trap using FORWARD action

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ido Schimmel <[email protected]>

[ Upstream commit 976c44af48141cd8595601c0af2a19a43c5b228b ]

The device's manual (PRM - Programmer's Reference Manual) classifies the trap that is used to deliver EMAD responses as an "event trap". Among other things, it means that the only actions that can be associated with the trap are TRAP and FORWARD (NOP).

Currently, during driver de-initialization the driver unregisters the trap by setting its action to DISCARD, which violates the above guideline. Future firmware versions will prevent such misuses by returning an error. This does not prevent the driver from working, but an error will be printed to the kernel log during module removal / devlink reload:

mlxsw_spectrum 0000:03:00.0: Reg cmd access status failed (status=7(bad parameter)) mlxsw_spectrum 0000:03:00.0: Reg cmd access failed (reg_id=7003(hpkt),type=write)

Suppress the error message by aligning the driver to the manual and use a FORWARD (NOP) action when unregistering the trap.

Fixes 4ec14b7634b2 ("mlxsw: Add interface to access registers and process events")

Cc Jiri Pirko <[email protected]>

Cc Amit Cohen <[email protected]>

Link https://lore.kernel.org/r/753a89e14008fde08cb4a2c1e5f537b81d8eb2d6.1713446092.git.petrm@nvidia.com open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/mellanox/mlxsw/core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

drivers/net/ethernet/mellanox/mlxsw/core.c +1 -1

index 1ccf3b73ed724..85507d01fd457 100644
@@ -835,7 +835,7 @@ static void mlxsw_emad_rx_listener_func(struct sk_buff *skb, u16 local_port,
 
 static const struct mlxsw_listener mlxsw_emad_rx_listener =
 	MLXSW_RXL(mlxsw_emad_rx_listener_func, ETHEMAD, TRAP_TO_CPU, false,
-		  EMAD, DISCARD);
+		  EMAD, FORWARD);
 
 static int mlxsw_emad_tlv_enable(struct mlxsw_core *mlxsw_core)
 {

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:07 UTC #53 lore raw

PATCH 48/186 mlxsw: core_env: Fix driver initialization with old firmware

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ido Schimmel <[email protected]>

[ Upstream commit 7e2050a8366315aeaf0316b3d362e67cf58f3ea8 ]

The driver queries the Management Capabilities Mask (MCAM) register during initialization to understand if it can read up to 128 bytes from transceiver modules.

However, not all firmware versions support this register, leading to the driver failing to load.

Fix by treating an error in the register query as an indication that the feature is not supported.

Fixes 1f4aea1f72da ("mlxsw: core_env: Read transceiver module EEPROM in 128 bytes chunks")

Reported-by Tim 'mithro' Ansell <[email protected]>

Link https://lore.kernel.org/r/0afa8b2e8bac178f5f88211344429176dcc72281.1713446092.git.petrm@nvidia.com open here

Signed-off-by Sasha Levin <[email protected]>

---
 .../net/ethernet/mellanox/mlxsw/core_env.c    | 20 ++++++-------------
 1 file changed, 6 insertions(+), 14 deletions(-)

drivers/net/ethernet/mellanox/mlxsw/core_env.c +6 ~~-14~~

index d637c0348fa15..b71bc23245fe2 100644
@@ -1357,24 +1357,20 @@ static struct mlxsw_linecards_event_ops mlxsw_env_event_ops = {
 	.got_inactive = mlxsw_env_got_inactive,
 };
 
-static int mlxsw_env_max_module_eeprom_len_query(struct mlxsw_env *mlxsw_env)
+static void mlxsw_env_max_module_eeprom_len_query(struct mlxsw_env *mlxsw_env)
 {
 	char mcam_pl[MLXSW_REG_MCAM_LEN];
-	bool mcia_128b_supported;
+	bool mcia_128b_supported = false;
 	int err;
 
 	mlxsw_reg_mcam_pack(mcam_pl,
 			    MLXSW_REG_MCAM_FEATURE_GROUP_ENHANCED_FEATURES);
 	err = mlxsw_reg_query(mlxsw_env->core, MLXSW_REG(mcam), mcam_pl);
-	if (err)
-		return err;
-
-	mlxsw_reg_mcam_unpack(mcam_pl, MLXSW_REG_MCAM_MCIA_128B,
-			      &mcia_128b_supported);
+	if (!err)
+		mlxsw_reg_mcam_unpack(mcam_pl, MLXSW_REG_MCAM_MCIA_128B,
+				      &mcia_128b_supported);
 
 	mlxsw_env->max_eeprom_len = mcia_128b_supported ? 128 : 48;
-
-	return 0;
 }
 
 int mlxsw_env_init(struct mlxsw_core *mlxsw_core,

@@ -1445,15 +1441,11 @@ int mlxsw_env_init(struct mlxsw_core *mlxsw_core,
 	if (err)
 		goto err_type_set;
 
-	err = mlxsw_env_max_module_eeprom_len_query(env);
-	if (err)
-		goto err_eeprom_len_query;
-
+	mlxsw_env_max_module_eeprom_len_query(env);
 	env->line_cards[0]->active = true;
 
 	return 0;
 
-err_eeprom_len_query:
 err_type_set:
 	mlxsw_env_module_event_disable(env, 0);
 err_mlxsw_env_module_event_enable:

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:07 UTC #54 lore raw

PATCH 49/186 ARM: dts: microchip: at91-sama7g5ek: Replace regulator-suspend-voltage with the valid property

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Andrei Simion <[email protected]>

[ Upstream commit e027b71762e84ee9d4ba9ad5401b956b9e83ed2a ]

By checking the pmic node with microchip,mcp16502.yaml# 'regulator-suspend-voltage' does not match any of the regexes 'pinctrl-[0-9]+' from schema microchip,mcp16502.yaml# which inherits regulator.yaml#. So replace regulator-suspend-voltage with regulator-suspend-microvolt to avoid the inconsitency.

Fixes 85b1304b9daa ("ARM: dts: at91: sama7g5ek: set regulator voltages for standby state")

Signed-off-by Andrei Simion <[email protected]>

Acked-by Nicolas Ferre <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

[claudiu.beznea: added a dot before starting the last sentence in commit
description]

Signed-off-by Claudiu Beznea <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm/boot/dts/microchip/at91-sama7g5ek.dts | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

arch/arm/boot/dts/microchip/at91-sama7g5ek.dts +4 -4

index 217e9b96c61e5..20b2497657ae4 100644
@@ -293,7 +293,7 @@
 
 					regulator-state-standby {
 						regulator-on-in-suspend;
-						regulator-suspend-voltage = <1150000>;
+						regulator-suspend-microvolt = <1150000>;
 						regulator-mode = <4>;
 					};

@@ -314,7 +314,7 @@
 
 					regulator-state-standby {
 						regulator-on-in-suspend;
-						regulator-suspend-voltage = <1050000>;
+						regulator-suspend-microvolt = <1050000>;
 						regulator-mode = <4>;
 					};

@@ -331,7 +331,7 @@
 					regulator-always-on;
 
 					regulator-state-standby {
-						regulator-suspend-voltage = <1800000>;
+						regulator-suspend-microvolt = <1800000>;
 						regulator-on-in-suspend;
 					};

@@ -346,7 +346,7 @@
 					regulator-max-microvolt = <3700000>;
 
 					regulator-state-standby {
-						regulator-suspend-voltage = <1800000>;
+						regulator-suspend-microvolt = <1800000>;
 						regulator-on-in-suspend;
 					};

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:08 UTC #55 lore raw

PATCH 50/186 icmp: prevent possible NULL dereferences from icmp_build_probe()

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eric Dumazet <[email protected]>

[ Upstream commit c58e88d49097bd12dfcfef4f075b43f5d5830941 ]

First problem is a double call to __in_dev_get_rcu(), because the second one could return NULL.

if (__in_dev_get_rcu(dev) && __in_dev_get_rcu(dev)->ifa_list)

Second problem is a read from dev->ip6_ptr with no NULL check:

if (!list_empty(&rcu_dereference(dev->ip6_ptr)->addr_list))

Use the correct RCU API to fix these.

v2: add missing include <net/addrconf.h>

Fixes d329ea5bd884 ("icmp: add response to RFC 8335 PROBE messages")

Cc Andreas Roeseler <[email protected]>

Reviewed-by David Ahern <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/ipv4/icmp.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)

net/ipv4/icmp.c +10 -2

index b8607763d113a..3b221643206de 100644
@@ -92,6 +92,7 @@
 #include <net/inet_common.h>
 #include <net/ip_fib.h>
 #include <net/l3mdev.h>
+#include <net/addrconf.h>
 
 /*
  *	Build xmit assembly blocks

@@ -1032,6 +1033,8 @@ bool icmp_build_probe(struct sk_buff *skb, struct icmphdr *icmphdr)
 	struct icmp_ext_hdr *ext_hdr, _ext_hdr;
 	struct icmp_ext_echo_iio *iio, _iio;
 	struct net *net = dev_net(skb->dev);
+	struct inet6_dev *in6_dev;
+	struct in_device *in_dev;
 	struct net_device *dev;
 	char buff[IFNAMSIZ];
 	u16 ident_len;

@@ -1115,10 +1118,15 @@ bool icmp_build_probe(struct sk_buff *skb, struct icmphdr *icmphdr)
 	/* Fill bits in reply message */
 	if (dev->flags & IFF_UP)
 		status |= ICMP_EXT_ECHOREPLY_ACTIVE;
-	if (__in_dev_get_rcu(dev) && __in_dev_get_rcu(dev)->ifa_list)
+
+	in_dev = __in_dev_get_rcu(dev);
+	if (in_dev && rcu_access_pointer(in_dev->ifa_list))
 		status |= ICMP_EXT_ECHOREPLY_IPV4;
-	if (!list_empty(&rcu_dereference(dev->ip6_ptr)->addr_list))
+
+	in6_dev = __in6_dev_get(dev);
+	if (in6_dev && !list_empty(&in6_dev->addr_list))
 		status |= ICMP_EXT_ECHOREPLY_IPV6;
+
 	dev_put(dev);
 	icmphdr->un.echo.sequence |= htons(status);
 	return true;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:08 UTC #56 lore raw

PATCH 51/186 bridge/br_netlink.c: no need to return void function

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hangbin Liu <[email protected]>

[ Upstream commit 4fd1edcdf13c0d234543ecf502092be65c5177db ]

br_info_notify is a void function. There is no need to return.

Fixes b6d0425b816e ("bridge: cfm: Netlink Notifications.")

Signed-off-by Hangbin Liu <[email protected]>

Acked-by Nikolay Aleksandrov <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/bridge/br_netlink.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

net/bridge/br_netlink.c +1 -1

index 10f0d33d8ccf2..65e9ed3851425 100644
@@ -666,7 +666,7 @@ void br_ifinfo_notify(int event, const struct net_bridge *br,
 {
 	u32 filter = RTEXT_FILTER_BRVLAN_COMPRESSED;
 
-	return br_info_notify(event, br, port, filter);
+	br_info_notify(event, br, port, filter);
 }
 
 /*

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:08 UTC #57 lore raw

PATCH 52/186 bnxt_en: refactor reset close code

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vikas Gupta <[email protected]>

[ Upstream commit 7474b1c82be3780692d537d331f9aa7fc1e5a368 ]

Introduce bnxt_fw_fatal_close() API which can be used to stop data path and disable device when firmware is in fatal state.

Signed-off-by Vikas Gupta <[email protected]>

Signed-off-by Michael Chan <[email protected]>

Stable-dep-of: a1acdc226bae ("bnxt_en: Fix the PCI-AER routines")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)

drivers/net/ethernet/broadcom/bnxt/bnxt.c +11 -6

index 38e3b2225ff1c..1cf4aae8702b4 100644
@@ -11807,6 +11807,16 @@ static void bnxt_rx_ring_reset(struct bnxt *bp)
 	bnxt_rtnl_unlock_sp(bp);
 }
 
+static void bnxt_fw_fatal_close(struct bnxt *bp)
+{
+	bnxt_tx_disable(bp);
+	bnxt_disable_napi(bp);
+	bnxt_disable_int_sync(bp);
+	bnxt_free_irq(bp);
+	bnxt_clear_int_mode(bp);
+	pci_disable_device(bp->pdev);
+}
+
 static void bnxt_fw_reset_close(struct bnxt *bp)
 {
 	bnxt_ulp_stop(bp);

@@ -11820,12 +11830,7 @@ static void bnxt_fw_reset_close(struct bnxt *bp)
 		pci_read_config_word(bp->pdev, PCI_SUBSYSTEM_ID, &val);
 		if (val == 0xffff)
 			bp->fw_reset_min_dsecs = 0;
-		bnxt_tx_disable(bp);
-		bnxt_disable_napi(bp);
-		bnxt_disable_int_sync(bp);
-		bnxt_free_irq(bp);
-		bnxt_clear_int_mode(bp);
-		pci_disable_device(bp->pdev);
+		bnxt_fw_fatal_close(bp);
 	}
 	__bnxt_close_nic(bp, true, false);
 	bnxt_vf_reps_free(bp);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:08 UTC #58 lore raw

PATCH 53/186 bnxt_en: Fix the PCI-AER routines

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vikas Gupta <[email protected]>

[ Upstream commit a1acdc226baec331512f815d6ac9dd6f8435cc7f ]

We do not support two simultaneous recoveries so check for reset flag, BNXT_STATE_IN_FW_RESET, and do not proceed with AER further. When the pci channel state is pci_channel_io_frozen, the PCIe link can not be trusted so we disable the traffic immediately and stop BAR access by calling bnxt_fw_fatal_close(). BAR access after AER fatal error can cause an NMI.

Fixes f75d9a0aa967 ("bnxt_en: Re-write PCI BARs after PCI fatal error.")

Signed-off-by Vikas Gupta <[email protected]>

Signed-off-by Michael Chan <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 19 ++++++++++++++++---
 1 file changed, 16 insertions(+), 3 deletions(-)

drivers/net/ethernet/broadcom/bnxt/bnxt.c +16 -3

index 1cf4aae8702b4..d40b91719b79b 100644
@@ -13972,6 +13972,7 @@ static pci_ers_result_t bnxt_io_error_detected(struct pci_dev *pdev,
 {
 	struct net_device *netdev = pci_get_drvdata(pdev);
 	struct bnxt *bp = netdev_priv(netdev);
+	bool abort = false;
 
 	netdev_info(netdev, "PCI I/O error detected\n");

@@ -13980,16 +13981,27 @@ static pci_ers_result_t bnxt_io_error_detected(struct pci_dev *pdev,
 
 	bnxt_ulp_stop(bp);
 
-	if (state == pci_channel_io_perm_failure) {
+	if (test_and_set_bit(BNXT_STATE_IN_FW_RESET, &bp->state)) {
+		netdev_err(bp->dev, "Firmware reset already in progress\n");
+		abort = true;
+	}
+
+	if (abort || state == pci_channel_io_perm_failure) {
 		rtnl_unlock();
 		return PCI_ERS_RESULT_DISCONNECT;
 	}
 
-	if (state == pci_channel_io_frozen)
+	/* Link is not reliable anymore if state is pci_channel_io_frozen
+	 * so we disable bus master to prevent any potential bad DMAs before
+	 * freeing kernel memory.
+	 */
+	if (state == pci_channel_io_frozen) {
 		set_bit(BNXT_STATE_PCI_CHANNEL_IO_FROZEN, &bp->state);
+		bnxt_fw_fatal_close(bp);
+	}
 
 	if (netif_running(netdev))
-		bnxt_close(netdev);
+		__bnxt_close_nic(bp, true, true);
 
 	if (pci_is_enabled(pdev))
 		pci_disable_device(pdev);

@@ -14075,6 +14087,7 @@ static pci_ers_result_t bnxt_io_slot_reset(struct pci_dev *pdev)
 	}
 
 reset_exit:
+	clear_bit(BNXT_STATE_IN_FW_RESET, &bp->state);
 	bnxt_clear_reservations(bp, true);
 	rtnl_unlock();

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:08 UTC #59 lore raw

PATCH 54/186 cxl/core: Fix potential payload size confusion in cxl_mem_get_poison()

Link https://lore.kernel.org/r/171235441633.2716581.12330082428680958635.stgit@dwillia2-xfh.jf.intel.com open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Dan Williams <[email protected]>

[ Upstream commit 4b759dd5765503bd466defac7d93aca14c23a15d ]

A recent change to cxl_mem_get_records_log() [1] highlighted a subtle nuance of looping calls to cxl_internal_send_cmd(), i.e. that cxl_internal_send_cmd() modifies the 'size_out' member of the @mbox_cmd argument. That mechanism is useful for communicating underflow, but it is unwanted when reusing @mbox_cmd for a subsequent submission. It turns out that cxl_xfer_log() avoids this scenario by always redefining @mbox_cmd each iteration.

Update cxl_mem_get_records_log() and cxl_mem_get_poison() to follow the same style as cxl_xfer_log(), i.e. re-define @mbox_cmd each iteration. The cxl_mem_get_records_log() change is just a style fixup, but the cxl_mem_get_poison() change is a potential fix, per Alison [2]:

Poison list retrieval can hit this case if the MORE flag is set and a follow on read of the list delivers more records than the previous read. ie. device gives one record, sets the _MORE flag, then gives 5.

Not an urgent fix since this behavior has not been seen in the wild, but worth tracking as a fix.

Cc Kwangjin Ko <[email protected]>

Cc Alison Schofield <[email protected]>

Fixes ed83f7ca398b ("cxl/mbox: Add GET_POISON_LIST mailbox command")

Link http://lore.kernel.org/r/[email protected] [1] open here

Link http://lore.kernel.org/r/ZhAhAL/GOaWFrauw@aschofie-mobl2 [2] open here

Signed-off-by Dan Williams <[email protected]>

Reviewed-by Ira Weiny <[email protected]>

Reviewed-by Alison Schofield <[email protected]>

Signed-off-by Dave Jiang <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/cxl/core/mbox.c | 38 +++++++++++++++++---------------------
 1 file changed, 17 insertions(+), 21 deletions(-)

drivers/cxl/core/mbox.c +17 ~~-21~~

index 4b4c15e943380..fecaa18f4dd20 100644
@@ -959,25 +959,22 @@ static void cxl_mem_get_records_log(struct cxl_memdev_state *mds,
 	struct cxl_memdev *cxlmd = mds->cxlds.cxlmd;
 	struct device *dev = mds->cxlds.dev;
 	struct cxl_get_event_payload *payload;
-	struct cxl_mbox_cmd mbox_cmd;
 	u8 log_type = type;
 	u16 nr_rec;
 
 	mutex_lock(&mds->event.log_lock);
 	payload = mds->event.buf;
 
-	mbox_cmd = (struct cxl_mbox_cmd) {
-		.opcode = CXL_MBOX_OP_GET_EVENT_RECORD,
-		.payload_in = &log_type,
-		.size_in = sizeof(log_type),
-		.payload_out = payload,
-		.min_out = struct_size(payload, records, 0),
-	};
-
 	do {
 		int rc, i;
-
-		mbox_cmd.size_out = mds->payload_size;
+		struct cxl_mbox_cmd mbox_cmd = (struct cxl_mbox_cmd) {
+			.opcode = CXL_MBOX_OP_GET_EVENT_RECORD,
+			.payload_in = &log_type,
+			.size_in = sizeof(log_type),
+			.payload_out = payload,
+			.size_out = mds->payload_size,
+			.min_out = struct_size(payload, records, 0),
+		};
 
 		rc = cxl_internal_send_cmd(mds, &mbox_cmd);
 		if (rc) {

@@ -1311,7 +1308,6 @@ int cxl_mem_get_poison(struct cxl_memdev *cxlmd, u64 offset, u64 len,
 	struct cxl_memdev_state *mds = to_cxl_memdev_state(cxlmd->cxlds);
 	struct cxl_mbox_poison_out *po;
 	struct cxl_mbox_poison_in pi;
-	struct cxl_mbox_cmd mbox_cmd;
 	int nr_records = 0;
 	int rc;

@@ -1323,16 +1319,16 @@ int cxl_mem_get_poison(struct cxl_memdev *cxlmd, u64 offset, u64 len,
 	pi.offset = cpu_to_le64(offset);
 	pi.length = cpu_to_le64(len / CXL_POISON_LEN_MULT);
 
-	mbox_cmd = (struct cxl_mbox_cmd) {
-		.opcode = CXL_MBOX_OP_GET_POISON,
-		.size_in = sizeof(pi),
-		.payload_in = &pi,
-		.size_out = mds->payload_size,
-		.payload_out = po,
-		.min_out = struct_size(po, record, 0),
-	};
-
 	do {
+		struct cxl_mbox_cmd mbox_cmd = (struct cxl_mbox_cmd){
+			.opcode = CXL_MBOX_OP_GET_POISON,
+			.size_in = sizeof(pi),
+			.payload_in = &pi,
+			.size_out = mds->payload_size,
+			.payload_out = po,
+			.min_out = struct_size(po, record, 0),
+		};
+
 		rc = cxl_internal_send_cmd(mds, &mbox_cmd);
 		if (rc)
 			break;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:08 UTC #60 lore raw

PATCH 55/186 net: dsa: mv88e6xx: fix supported_interfaces setup in mv88e6250_phylink_get_caps()

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Matthias Schiffer <[email protected]>

[ Upstream commit a4e3899065ffa87d49dc20e8c17501edbc189692 ]

With the recent PHYLINK changes requiring supported_interfaces to be set, MV88E6250 family switches like the 88E6020 fail to probe - cmode is never initialized on these devices, so mv88e6250_phylink_get_caps() does not set any supported_interfaces flags.

Instead of a cmode, on 88E6250 we have a read-only port mode value that encodes similar information. There is no reason to bother mapping port mode to the cmodes of other switch models; instead we introduce a mv88e6250_setup_supported_interfaces() that is called directly from mv88e6250_phylink_get_caps().

Fixes de5c9bf40c45 ("net: phylink: require supported_interfaces to be filled")

Signed-off-by Matthias Schiffer <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/dsa/mv88e6xxx/chip.c | 56 +++++++++++++++++++++++++++++---
 drivers/net/dsa/mv88e6xxx/port.h | 23 ++++++++++---
 2 files changed, 71 insertions(+), 8 deletions(-)

drivers/net/dsa/mv88e6xxx/chip.c +52 -4

index b8fde22aebf93..8556502f06721 100644
@@ -566,13 +566,61 @@ static void mv88e6xxx_translate_cmode(u8 cmode, unsigned long *supported)
 		phy_interface_set_rgmii(supported);
 }
 
-static void mv88e6250_phylink_get_caps(struct mv88e6xxx_chip *chip, int port,
-				       struct phylink_config *config)
+static void
+mv88e6250_setup_supported_interfaces(struct mv88e6xxx_chip *chip, int port,
+				     struct phylink_config *config)
 {
 	unsigned long *supported = config->supported_interfaces;
+	int err;
+	u16 reg;
 
-	/* Translate the default cmode */
-	mv88e6xxx_translate_cmode(chip->ports[port].cmode, supported);
+	err = mv88e6xxx_port_read(chip, port, MV88E6XXX_PORT_STS, &reg);
+	if (err) {
+		dev_err(chip->dev, "p%d: failed to read port status\n", port);
+		return;
+	}
+
+	switch (reg & MV88E6250_PORT_STS_PORTMODE_MASK) {
+	case MV88E6250_PORT_STS_PORTMODE_MII_10_HALF_PHY:
+	case MV88E6250_PORT_STS_PORTMODE_MII_100_HALF_PHY:
+	case MV88E6250_PORT_STS_PORTMODE_MII_10_FULL_PHY:
+	case MV88E6250_PORT_STS_PORTMODE_MII_100_FULL_PHY:
+		__set_bit(PHY_INTERFACE_MODE_REVMII, supported);
+		break;
+
+	case MV88E6250_PORT_STS_PORTMODE_MII_HALF:
+	case MV88E6250_PORT_STS_PORTMODE_MII_FULL:
+		__set_bit(PHY_INTERFACE_MODE_MII, supported);
+		break;
+
+	case MV88E6250_PORT_STS_PORTMODE_MII_DUAL_100_RMII_FULL_PHY:
+	case MV88E6250_PORT_STS_PORTMODE_MII_200_RMII_FULL_PHY:
+	case MV88E6250_PORT_STS_PORTMODE_MII_10_100_RMII_HALF_PHY:
+	case MV88E6250_PORT_STS_PORTMODE_MII_10_100_RMII_FULL_PHY:
+		__set_bit(PHY_INTERFACE_MODE_REVRMII, supported);
+		break;
+
+	case MV88E6250_PORT_STS_PORTMODE_MII_DUAL_100_RMII_FULL:
+	case MV88E6250_PORT_STS_PORTMODE_MII_10_100_RMII_FULL:
+		__set_bit(PHY_INTERFACE_MODE_RMII, supported);
+		break;
+
+	case MV88E6250_PORT_STS_PORTMODE_MII_100_RGMII:
+		__set_bit(PHY_INTERFACE_MODE_RGMII, supported);
+		break;
+
+	default:
+		dev_err(chip->dev,
+			"p%d: invalid port mode in status register: %04x\n",
+			port, reg);
+	}
+}
+
+static void mv88e6250_phylink_get_caps(struct mv88e6xxx_chip *chip, int port,
+				       struct phylink_config *config)
+{
+	if (!mv88e6xxx_phy_is_internal(chip, port))
+		mv88e6250_setup_supported_interfaces(chip, port, config);
 
 	config->mac_capabilities = MAC_SYM_PAUSE | MAC_10 | MAC_100;
 }

drivers/net/dsa/mv88e6xxx/port.h +19 -4

index 86deeb347cbc1..ddadeb9bfdaee 100644
@@ -25,10 +25,25 @@
 #define MV88E6250_PORT_STS_PORTMODE_PHY_100_HALF	0x0900
 #define MV88E6250_PORT_STS_PORTMODE_PHY_10_FULL		0x0a00
 #define MV88E6250_PORT_STS_PORTMODE_PHY_100_FULL	0x0b00
-#define MV88E6250_PORT_STS_PORTMODE_MII_10_HALF		0x0c00
-#define MV88E6250_PORT_STS_PORTMODE_MII_100_HALF	0x0d00
-#define MV88E6250_PORT_STS_PORTMODE_MII_10_FULL		0x0e00
-#define MV88E6250_PORT_STS_PORTMODE_MII_100_FULL	0x0f00
+/* - Modes with PHY suffix use output instead of input clock
+ * - Modes without RMII or RGMII use MII
+ * - Modes without speed do not have a fixed speed specified in the manual
+ *   ("DC to x MHz" - variable clock support?)
+ */
+#define MV88E6250_PORT_STS_PORTMODE_MII_DISABLED		0x0000
+#define MV88E6250_PORT_STS_PORTMODE_MII_100_RGMII		0x0100
+#define MV88E6250_PORT_STS_PORTMODE_MII_DUAL_100_RMII_FULL_PHY	0x0200
+#define MV88E6250_PORT_STS_PORTMODE_MII_200_RMII_FULL_PHY	0x0400
+#define MV88E6250_PORT_STS_PORTMODE_MII_DUAL_100_RMII_FULL	0x0600
+#define MV88E6250_PORT_STS_PORTMODE_MII_10_100_RMII_FULL	0x0700
+#define MV88E6250_PORT_STS_PORTMODE_MII_HALF			0x0800
+#define MV88E6250_PORT_STS_PORTMODE_MII_10_100_RMII_HALF_PHY	0x0900
+#define MV88E6250_PORT_STS_PORTMODE_MII_FULL			0x0a00
+#define MV88E6250_PORT_STS_PORTMODE_MII_10_100_RMII_FULL_PHY	0x0b00
+#define MV88E6250_PORT_STS_PORTMODE_MII_10_HALF_PHY		0x0c00
+#define MV88E6250_PORT_STS_PORTMODE_MII_100_HALF_PHY		0x0d00
+#define MV88E6250_PORT_STS_PORTMODE_MII_10_FULL_PHY		0x0e00
+#define MV88E6250_PORT_STS_PORTMODE_MII_100_FULL_PHY		0x0f00
 #define MV88E6XXX_PORT_STS_LINK			0x0800
 #define MV88E6XXX_PORT_STS_DUPLEX		0x0400
 #define MV88E6XXX_PORT_STS_SPEED_MASK		0x0300

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:08 UTC #61 lore raw

PATCH 56/186 NFC: trf7970a: disable all regulators on removal

Link https://lore.kernel.org/r/DB7PR09MB26847A4EBF88D9EDFEB1DA0F950E2@DB7PR09MB2684.eurprd09.prod.outlook.com open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Paul Geurts <[email protected]>

[ Upstream commit 6bea4f03c6a4e973ef369e15aac88f37981db49e ]

During module probe, regulator 'vin' and 'vdd-io' are used and enabled, but the vdd-io regulator overwrites the 'vin' regulator pointer. During remove, only the vdd-io is disabled, as the vin regulator pointer is not available anymore. When regulator_put() is called during resource cleanup a kernel warning is given, as the regulator is still enabled.

Store the two regulators in separate pointers and disable both the regulators on module remove.

Fixes 49d22c70aaf0 ("NFC: trf7970a: Add device tree option of 1.8 Volt IO voltage")

Signed-off-by Paul Geurts <[email protected]>

Reviewed-by Krzysztof Kozlowski <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/nfc/trf7970a.c | 42 +++++++++++++++++++++++-------------------
 1 file changed, 23 insertions(+), 19 deletions(-)

drivers/nfc/trf7970a.c +23 ~~-19~~

index 7eb17f46a8153..9e1a34e23af26 100644
@@ -424,7 +424,8 @@ struct trf7970a {
 	enum trf7970a_state		state;
 	struct device			*dev;
 	struct spi_device		*spi;
-	struct regulator		*regulator;
+	struct regulator		*vin_regulator;
+	struct regulator		*vddio_regulator;
 	struct nfc_digital_dev		*ddev;
 	u32				quirks;
 	bool				is_initiator;

@@ -1883,7 +1884,7 @@ static int trf7970a_power_up(struct trf7970a *trf)
 	if (trf->state != TRF7970A_ST_PWR_OFF)
 		return 0;
 
-	ret = regulator_enable(trf->regulator);
+	ret = regulator_enable(trf->vin_regulator);
 	if (ret) {
 		dev_err(trf->dev, "%s - Can't enable VIN: %d\n", __func__, ret);
 		return ret;

@@ -1926,7 +1927,7 @@ static int trf7970a_power_down(struct trf7970a *trf)
 	if (trf->en2_gpiod && !(trf->quirks & TRF7970A_QUIRK_EN2_MUST_STAY_LOW))
 		gpiod_set_value_cansleep(trf->en2_gpiod, 0);
 
-	ret = regulator_disable(trf->regulator);
+	ret = regulator_disable(trf->vin_regulator);
 	if (ret)
 		dev_err(trf->dev, "%s - Can't disable VIN: %d\n", __func__,
 			ret);

@@ -2065,37 +2066,37 @@ static int trf7970a_probe(struct spi_device *spi)
 	mutex_init(&trf->lock);
 	INIT_DELAYED_WORK(&trf->timeout_work, trf7970a_timeout_work_handler);
 
-	trf->regulator = devm_regulator_get(&spi->dev, "vin");
-	if (IS_ERR(trf->regulator)) {
-		ret = PTR_ERR(trf->regulator);
+	trf->vin_regulator = devm_regulator_get(&spi->dev, "vin");
+	if (IS_ERR(trf->vin_regulator)) {
+		ret = PTR_ERR(trf->vin_regulator);
 		dev_err(trf->dev, "Can't get VIN regulator: %d\n", ret);
 		goto err_destroy_lock;
 	}
 
-	ret = regulator_enable(trf->regulator);
+	ret = regulator_enable(trf->vin_regulator);
 	if (ret) {
 		dev_err(trf->dev, "Can't enable VIN: %d\n", ret);
 		goto err_destroy_lock;
 	}
 
-	uvolts = regulator_get_voltage(trf->regulator);
+	uvolts = regulator_get_voltage(trf->vin_regulator);
 	if (uvolts > 4000000)
 		trf->chip_status_ctrl = TRF7970A_CHIP_STATUS_VRS5_3;
 
-	trf->regulator = devm_regulator_get(&spi->dev, "vdd-io");
-	if (IS_ERR(trf->regulator)) {
-		ret = PTR_ERR(trf->regulator);
+	trf->vddio_regulator = devm_regulator_get(&spi->dev, "vdd-io");
+	if (IS_ERR(trf->vddio_regulator)) {
+		ret = PTR_ERR(trf->vddio_regulator);
 		dev_err(trf->dev, "Can't get VDD_IO regulator: %d\n", ret);
-		goto err_destroy_lock;
+		goto err_disable_vin_regulator;
 	}
 
-	ret = regulator_enable(trf->regulator);
+	ret = regulator_enable(trf->vddio_regulator);
 	if (ret) {
 		dev_err(trf->dev, "Can't enable VDD_IO: %d\n", ret);
-		goto err_destroy_lock;
+		goto err_disable_vin_regulator;
 	}
 
-	if (regulator_get_voltage(trf->regulator) == 1800000) {
+	if (regulator_get_voltage(trf->vddio_regulator) == 1800000) {
 		trf->io_ctrl = TRF7970A_REG_IO_CTRL_IO_LOW;
 		dev_dbg(trf->dev, "trf7970a config vdd_io to 1.8V\n");
 	}

@@ -2108,7 +2109,7 @@ static int trf7970a_probe(struct spi_device *spi)
 	if (!trf->ddev) {
 		dev_err(trf->dev, "Can't allocate NFC digital device\n");
 		ret = -ENOMEM;
-		goto err_disable_regulator;
+		goto err_disable_vddio_regulator;
 	}
 
 	nfc_digital_set_parent_dev(trf->ddev, trf->dev);

@@ -2137,8 +2138,10 @@ static int trf7970a_probe(struct spi_device *spi)
 	trf7970a_shutdown(trf);
 err_free_ddev:
 	nfc_digital_free_device(trf->ddev);
-err_disable_regulator:
-	regulator_disable(trf->regulator);
+err_disable_vddio_regulator:
+	regulator_disable(trf->vddio_regulator);
+err_disable_vin_regulator:
+	regulator_disable(trf->vin_regulator);
 err_destroy_lock:
 	mutex_destroy(&trf->lock);
 	return ret;

@@ -2157,7 +2160,8 @@ static void trf7970a_remove(struct spi_device *spi)
 	nfc_digital_unregister_device(trf->ddev);
 	nfc_digital_free_device(trf->ddev);
 
-	regulator_disable(trf->regulator);
+	regulator_disable(trf->vddio_regulator);
+	regulator_disable(trf->vin_regulator);
 
 	mutex_destroy(&trf->lock);
 }

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:08 UTC #62 lore raw

PATCH 57/186 ax25: Fix netdev refcount issue

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Duoming Zhou <[email protected]>

[ Upstream commit 467324bcfe1a31ec65d0cf4aa59421d6b7a7d52b ]

The dev_tracker is added to ax25_cb in ax25_bind(). When the ax25 device is detaching, the dev_tracker of ax25_cb should be deallocated in ax25_kill_by_device() instead of the dev_tracker of ax25_dev. The log reported by ref_tracker is shown below:

[   80.884935] ref_tracker: reference already released.
[   80.885150] ref_tracker: allocated in:
[   80.885349]  ax25_dev_device_up+0x105/0x540
[   80.885730]  ax25_device_event+0xa4/0x420
[   80.885730]  notifier_call_chain+0xc9/0x1e0
[   80.885730]  __dev_notify_flags+0x138/0x280
[   80.885730]  dev_change_flags+0xd7/0x180
[   80.885730]  dev_ifsioc+0x6a9/0xa30
[   80.885730]  dev_ioctl+0x4d8/0xd90
[   80.885730]  sock_do_ioctl+0x1c2/0x2d0
[   80.885730]  sock_ioctl+0x38b/0x4f0
[   80.885730]  __se_sys_ioctl+0xad/0xf0
[   80.885730]  do_syscall_64+0xc4/0x1b0
[   80.885730]  entry_SYSCALL_64_after_hwframe+0x67/0x6f
[   80.885730] ref_tracker: freed in:
[   80.885730]  ax25_device_event+0x272/0x420
[   80.885730]  notifier_call_chain+0xc9/0x1e0
[   80.885730]  dev_close_many+0x272/0x370
[   80.885730]  unregister_netdevice_many_notify+0x3b5/0x1180
[   80.885730]  unregister_netdev+0xcf/0x120
[   80.885730]  sixpack_close+0x11f/0x1b0
[   80.885730]  tty_ldisc_kill+0xcb/0x190
[   80.885730]  tty_ldisc_hangup+0x338/0x3d0
[   80.885730]  __tty_hangup+0x504/0x740
[   80.885730]  tty_release+0x46e/0xd80
[   80.885730]  __fput+0x37f/0x770
[   80.885730]  __x64_sys_close+0x7b/0xb0
[   80.885730]  do_syscall_64+0xc4/0x1b0
[   80.885730]  entry_SYSCALL_64_after_hwframe+0x67/0x6f
[   80.893739] ------------[ cut here ]------------
[   80.894030] WARNING: CPU: 2 PID: 140 at lib/ref_tracker.c:255 ref_tracker_free+0x47b/0x6b0
[   80.894297] Modules linked in:
[   80.894929] CPU: 2 PID: 140 Comm: ax25_conn_rel_6 Not tainted 6.9.0-rc4-g8cd26fd90c1a #11
[   80.895190] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.14.0-0-g155821a1990b-prebuilt.qem4
[   80.895514] RIP: 0010:ref_tracker_free+0x47b/0x6b0
[   80.895808] Code: 83 c5 18 4c 89 eb 48 c1 eb 03 8a 04 13 84 c0 0f 85 df 01 00 00 41 83 7d 00 00 75 4b 4c 89 ff 9
[   80.896171] RSP: 0018:ffff888009edf8c0 EFLAGS: 00000286
[   80.896339] RAX: 1ffff1100141ac00 RBX: 1ffff1100149463b RCX: dffffc0000000000
[   80.896502] RDX: 0000000000000001 RSI: 0000000000000246 RDI: ffff88800a0d6518
[   80.896925] RBP: ffff888009edf9b0 R08: ffff88806d3288d3 R09: 1ffff1100da6511a
[   80.897212] R10: dffffc0000000000 R11: ffffed100da6511b R12: ffff88800a4a31d4
[   80.897859] R13: ffff88800a4a31d8 R14: dffffc0000000000 R15: ffff88800a0d6518
[   80.898279] FS:  00007fd88b7fe700(0000) GS:ffff88806d300000(0000) knlGS:0000000000000000
[   80.899436] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   80.900181] CR2: 00007fd88c001d48 CR3: 000000000993e000 CR4: 00000000000006f0
...
[   80.935774] ref_tracker: sp%d@000000000bb9df3d has 1/1 users at
[   80.935774]      ax25_bind+0x424/0x4e0
[   80.935774]      __sys_bind+0x1d9/0x270
[   80.935774]      __x64_sys_bind+0x75/0x80
[   80.935774]      do_syscall_64+0xc4/0x1b0
[   80.935774]      entry_SYSCALL_64_after_hwframe+0x67/0x6f

Change ax25_dev->dev_tracker to the dev_tracker of ax25_cb in order to mitigate the bug.

Fixes feef318c855a ("ax25: fix UAF bugs of net_device caused by rebinding operation")

Signed-off-by Duoming Zhou <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Paolo Abeni <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/ax25/af_ax25.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

net/ax25/af_ax25.c +1 -1

index 5db805d5f74d7..9d11d26e46c0e 100644
@@ -103,7 +103,7 @@ static void ax25_kill_by_device(struct net_device *dev)
 			s->ax25_dev = NULL;
 			if (sk->sk_socket) {
 				netdev_put(ax25_dev->dev,
-					   &ax25_dev->dev_tracker);
+					   &s->dev_tracker);
 				ax25_dev_put(ax25_dev);
 			}
 			ax25_cb_del(s);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:08 UTC #63 lore raw

PATCH 58/186 tools: ynl: dont ignore errors in NLMSG_DONE messages

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jakub Kicinski <[email protected]>

[ Upstream commit a44f2eb106a46f2275a79de54ce0ea63e4f3d8c8 ]

NLMSG_DONE contains an error code, it has to be extracted. Prior to this change all dumps will end in success, and in case of failure the result is silently truncated.

Fixes e4b48ed460d3 ("tools: ynl: add a completely generic client")

Reviewed-by Donald Hunter <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Paolo Abeni <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 tools/net/ynl/lib/ynl.py | 1 +
 1 file changed, 1 insertion(+)

tools/net/ynl/lib/ynl.py +1 -0

index 13c4b019a881f..44ea0965c9d9c 100644
@@ -201,6 +201,7 @@ class NlMsg:
             self.done = 1
             extack_off = 20
         elif self.nl_type == Netlink.NLMSG_DONE:
+            self.error = struct.unpack("i", self.raw[0:4])[0]
             self.done = 1
             extack_off = 4

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:08 UTC #64 lore raw

PATCH 59/186 net: make SK_MEMORY_PCPU_RESERV tunable

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Adam Li <[email protected]>

[ Upstream commit 12a686c2e761f1f1f6e6e2117a9ab9c6de2ac8a7 ]

This patch adds /proc/sys/net/core/mem_pcpu_rsv sysctl file, to make SK_MEMORY_PCPU_RESERV tunable.

Commit 3cd3399dd7a8 ("net: implement per-cpu reserves for memory_allocated") introduced per-cpu forward alloc cache:

"Implement a per-cpu cache of +1/-1 MB, to reduce number of changes to sk->sk_prot->memory_allocated, which would otherwise be cause of false sharing."

sk_prot->memory_allocated points to global atomic variable:
atomic_long_t tcp_memory_allocated ____cacheline_aligned_in_smp;

If increasing the per-cpu cache size from 1MB to e.g. 16MB, changes to sk->sk_prot->memory_allocated can be further reduced. Performance may be improved on system with many cores.

Signed-off-by Adam Li <[email protected]>

Reviewed-by Christoph Lameter (Ampere) <[email protected]>

Stable-dep-of: 3584718cf2ec ("net: fix sk_memory_allocated_{add|sub} vs softirqs")

Signed-off-by Sasha Levin <[email protected]>

---
 Documentation/admin-guide/sysctl/net.rst | 5 +++++
 include/net/sock.h                       | 5 +++--
 net/core/sock.c                          | 1 +
 net/core/sysctl_net_core.c               | 9 +++++++++
 4 files changed, 18 insertions(+), 2 deletions(-)

Documentation/admin-guide/sysctl/net.rst +5 -0

index 4877563241f3b..5f1748f33d9a2 100644
@@ -205,6 +205,11 @@ Will increase power usage.
 
 Default: 0 (off)
 
+mem_pcpu_rsv
+------------
+
+Per-cpu reserved forward alloc cache size in page units. Default 1MB per CPU.
+
 rmem_default
 ------------

include/net/sock.h +3 -2

index 25780942ec8bf..4372a7225ae6a 100644
@@ -1458,6 +1458,7 @@ sk_memory_allocated(const struct sock *sk)
 
 /* 1 MB per cpu, in page units */
 #define SK_MEMORY_PCPU_RESERVE (1 << (20 - PAGE_SHIFT))
+extern int sysctl_mem_pcpu_rsv;
 
 static inline void
 sk_memory_allocated_add(struct sock *sk, int amt)

@@ -1466,7 +1467,7 @@ sk_memory_allocated_add(struct sock *sk, int amt)
 
 	preempt_disable();
 	local_reserve = __this_cpu_add_return(*sk->sk_prot->per_cpu_fw_alloc, amt);
-	if (local_reserve >= SK_MEMORY_PCPU_RESERVE) {
+	if (local_reserve >= READ_ONCE(sysctl_mem_pcpu_rsv)) {
 		__this_cpu_sub(*sk->sk_prot->per_cpu_fw_alloc, local_reserve);
 		atomic_long_add(local_reserve, sk->sk_prot->memory_allocated);
 	}

@@ -1480,7 +1481,7 @@ sk_memory_allocated_sub(struct sock *sk, int amt)
 
 	preempt_disable();
 	local_reserve = __this_cpu_sub_return(*sk->sk_prot->per_cpu_fw_alloc, amt);
-	if (local_reserve <= -SK_MEMORY_PCPU_RESERVE) {
+	if (local_reserve <= -READ_ONCE(sysctl_mem_pcpu_rsv)) {
 		__this_cpu_sub(*sk->sk_prot->per_cpu_fw_alloc, local_reserve);
 		atomic_long_add(local_reserve, sk->sk_prot->memory_allocated);
 	}

net/core/sock.c +1 -0

index 383e30fe79f41..1471c0a862b36 100644
@@ -283,6 +283,7 @@ __u32 sysctl_rmem_max __read_mostly = SK_RMEM_MAX;
 EXPORT_SYMBOL(sysctl_rmem_max);
 __u32 sysctl_wmem_default __read_mostly = SK_WMEM_MAX;
 __u32 sysctl_rmem_default __read_mostly = SK_RMEM_MAX;
+int sysctl_mem_pcpu_rsv __read_mostly = SK_MEMORY_PCPU_RESERVE;
 
 /* Maximal space eaten by iovec or ancillary data plus some space */
 int sysctl_optmem_max __read_mostly = sizeof(unsigned long)*(2*UIO_MAXIOV+512);

net/core/sysctl_net_core.c +9 -0

index 03f1edb948d7d..373b5b2231c49 100644
@@ -30,6 +30,7 @@ static int int_3600 = 3600;
 static int min_sndbuf = SOCK_MIN_SNDBUF;
 static int min_rcvbuf = SOCK_MIN_RCVBUF;
 static int max_skb_frags = MAX_SKB_FRAGS;
+static int min_mem_pcpu_rsv = SK_MEMORY_PCPU_RESERVE;
 
 static int net_msg_warn;	/* Unused, but still a sysctl */

@@ -407,6 +408,14 @@ static struct ctl_table net_core_table[] = {
 		.proc_handler	= proc_dointvec_minmax,
 		.extra1		= &min_rcvbuf,
 	},
+	{
+		.procname	= "mem_pcpu_rsv",
+		.data		= &sysctl_mem_pcpu_rsv,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= &min_mem_pcpu_rsv,
+	},
 	{
 		.procname	= "dev_weight",
 		.data		= &weight_p,

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:08 UTC #65 lore raw

PATCH 60/186 net: fix sk_memory_allocated_{add|sub} vs softirqs

Closes https://lore.kernel.org/netdev/VI1PR01MB42407D7947B2EA448F1E04EFD10D2@VI1PR01MB4240.eurprd01.prod.exchangelabs.com/ open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eric Dumazet <[email protected]>

[ Upstream commit 3584718cf2ec7e79b6814f2596dcf398c5fb2eca ]

Jonathan Heathcote reported a regression caused by blamed commit on aarch64 architecture.

x86 happens to have irq-safe __this_cpu_add_return() and __this_cpu_sub(), but this is not generic.

I think my confusion came from "struct sock" argument, because these helpers are called with a locked socket. But the memory accounting is per-proto (and per-cpu after the blamed commit). We might cleanup these helpers later to directly accept a "struct proto *proto" argument.

Switch to this_cpu_add_return() and this_cpu_xchg() operations, and get rid of preempt_disable()/preempt_enable() pairs.

Fast path becomes a bit faster as a result :)

Many thanks to Jonathan Heathcote for his awesome report and investigations.

Fixes 3cd3399dd7a8 ("net: implement per-cpu reserves for memory_allocated")

Reported-by Jonathan Heathcote <[email protected]>

Acked-by Soheil Hassas Yeganeh <[email protected]>

Reviewed-by Shakeel Butt <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 include/net/sock.h | 38 ++++++++++++++++++++------------------
 1 file changed, 20 insertions(+), 18 deletions(-)

include/net/sock.h +20 ~~-18~~

index 4372a7225ae6a..53b81e0a89810 100644
@@ -1460,32 +1460,34 @@ sk_memory_allocated(const struct sock *sk)
 #define SK_MEMORY_PCPU_RESERVE (1 << (20 - PAGE_SHIFT))
 extern int sysctl_mem_pcpu_rsv;
 
+static inline void proto_memory_pcpu_drain(struct proto *proto)
+{
+	int val = this_cpu_xchg(*proto->per_cpu_fw_alloc, 0);
+
+	if (val)
+		atomic_long_add(val, proto->memory_allocated);
+}
+
 static inline void
-sk_memory_allocated_add(struct sock *sk, int amt)
+sk_memory_allocated_add(const struct sock *sk, int val)
 {
-	int local_reserve;
+	struct proto *proto = sk->sk_prot;
 
-	preempt_disable();
-	local_reserve = __this_cpu_add_return(*sk->sk_prot->per_cpu_fw_alloc, amt);
-	if (local_reserve >= READ_ONCE(sysctl_mem_pcpu_rsv)) {
-		__this_cpu_sub(*sk->sk_prot->per_cpu_fw_alloc, local_reserve);
-		atomic_long_add(local_reserve, sk->sk_prot->memory_allocated);
-	}
-	preempt_enable();
+	val = this_cpu_add_return(*proto->per_cpu_fw_alloc, val);
+
+	if (unlikely(val >= READ_ONCE(sysctl_mem_pcpu_rsv)))
+		proto_memory_pcpu_drain(proto);
 }
 
 static inline void
-sk_memory_allocated_sub(struct sock *sk, int amt)
+sk_memory_allocated_sub(const struct sock *sk, int val)
 {
-	int local_reserve;
+	struct proto *proto = sk->sk_prot;
 
-	preempt_disable();
-	local_reserve = __this_cpu_sub_return(*sk->sk_prot->per_cpu_fw_alloc, amt);
-	if (local_reserve <= -READ_ONCE(sysctl_mem_pcpu_rsv)) {
-		__this_cpu_sub(*sk->sk_prot->per_cpu_fw_alloc, local_reserve);
-		atomic_long_add(local_reserve, sk->sk_prot->memory_allocated);
-	}
-	preempt_enable();
+	val = this_cpu_sub_return(*proto->per_cpu_fw_alloc, val);
+
+	if (unlikely(val <= -READ_ONCE(sysctl_mem_pcpu_rsv)))
+		proto_memory_pcpu_drain(proto);
 }
 
 #define SK_ALLOC_PERCPU_COUNTER_BATCH 16

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:08 UTC #66 lore raw

PATCH 61/186 ipv4: check for NULL idev in ip_route_use_hint()

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eric Dumazet <[email protected]>

[ Upstream commit 58a4c9b1e5a3e53c9148e80b90e1e43897ce77d1 ]

syzbot was able to trigger a NULL deref in fib_validate_source() in an old tree [1].

It appears the bug exists in latest trees.

All calls to __in_dev_get_rcu() must be checked for a NULL result.

[1]

general protection fault, probably for non-canonical address 0xdffffc0000000000: 0000 [#1] SMP KASAN
KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007]
CPU: 2 PID: 3257 Comm: syz-executor.3 Not tainted 5.10.0-syzkaller #0
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2~bpo12+1 04/01/2014
 RIP: 0010:fib_validate_source+0xbf/0x15a0 net/ipv4/fib_frontend.c:425
Code: 18 f2 f2 f2 f2 42 c7 44 20 23 f3 f3 f3 f3 48 89 44 24 78 42 c6 44 20 27 f3 e8 5d 88 48 fc 4c 89 e8 48 c1 e8 03 48 89 44 24 18 <42> 80 3c 20 00 74 08 4c 89 ef e8 d2 15 98 fc 48 89 5c 24 10 41 bf
RSP: 0018:ffffc900015fee40 EFLAGS: 00010246
RAX: 0000000000000000 RBX: ffff88800f7a4000 RCX: ffff88800f4f90c0
RDX: 0000000000000000 RSI: 0000000004001eac RDI: ffff8880160c64c0
RBP: ffffc900015ff060 R08: 0000000000000000 R09: ffff88800f7a4000
R10: 0000000000000002 R11: ffff88800f4f90c0 R12: dffffc0000000000
R13: 0000000000000000 R14: 0000000000000000 R15: ffff88800f7a4000
FS:  00007f938acfe6c0(0000) GS:ffff888058c00000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f938acddd58 CR3: 000000001248e000 CR4: 0000000000352ef0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
  ip_route_use_hint+0x410/0x9b0 net/ipv4/route.c:2231
  ip_rcv_finish_core+0x2c4/0x1a30 net/ipv4/ip_input.c:327
  ip_list_rcv_finish net/ipv4/ip_input.c:612 [inline]
  ip_sublist_rcv+0x3ed/0xe50 net/ipv4/ip_input.c:638
  ip_list_rcv+0x422/0x470 net/ipv4/ip_input.c:673
  __netif_receive_skb_list_ptype net/core/dev.c:5572 [inline]
  __netif_receive_skb_list_core+0x6b1/0x890 net/core/dev.c:5620
  __netif_receive_skb_list net/core/dev.c:5672 [inline]
  netif_receive_skb_list_internal+0x9f9/0xdc0 net/core/dev.c:5764
  netif_receive_skb_list+0x55/0x3e0 net/core/dev.c:5816
  xdp_recv_frames net/bpf/test_run.c:257 [inline]
  xdp_test_run_batch net/bpf/test_run.c:335 [inline]
  bpf_test_run_xdp_live+0x1818/0x1d00 net/bpf/test_run.c:363
  bpf_prog_test_run_xdp+0x81f/0x1170 net/bpf/test_run.c:1376
  bpf_prog_test_run+0x349/0x3c0 kernel/bpf/syscall.c:3736
  __sys_bpf+0x45c/0x710 kernel/bpf/syscall.c:5115
  __do_sys_bpf kernel/bpf/syscall.c:5201 [inline]
  __se_sys_bpf kernel/bpf/syscall.c:5199 [inline]
  __x64_sys_bpf+0x7c/0x90 kernel/bpf/syscall.c:5199

Fixes 02b24941619f ("ipv4: use dst hint for ipv4 list receive")

Reported-by syzbot <[email protected]>

Acked-by Paolo Abeni <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 net/ipv4/route.c | 3 +++
 1 file changed, 3 insertions(+)

net/ipv4/route.c +3 -0

index e1e30c09a1753..7c05cbcd39d33 100644
@@ -2166,6 +2166,9 @@ int ip_route_use_hint(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 	int err = -EINVAL;
 	u32 tag = 0;
 
+	if (!in_dev)
+		return -EINVAL;
+
 	if (ipv4_is_multicast(saddr) || ipv4_is_lbcast(saddr))
 		goto martian_source;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:08 UTC #67 lore raw

PATCH 62/186 net: usb: ax88179_178a: stop lying about skb->truesize

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eric Dumazet <[email protected]>

[ Upstream commit 4ce62d5b2f7aecd4900e7d6115588ad7f9acccca ]

Some usb drivers try to set small skb->truesize and break core networking stacks.

In this patch, I removed one of the skb->truesize overide.

I also replaced one skb_clone() by an allocation of a fresh and small skb, to get minimally sized skbs, like we did in commit 1e2c61172342 ("net: cdc_ncm: reduce skb truesize in rx path")

Fixes f8ebb3ac881b ("net: usb: ax88179_178a: Fix packet receiving")

Reported-by shironeko <[email protected]>

Closes https://lore.kernel.org/netdev/[email protected]/ open here

Cc Jose Alonso <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/usb/ax88179_178a.c | 11 +++--------
 1 file changed, 3 insertions(+), 8 deletions(-)

drivers/net/usb/ax88179_178a.c +3 -8

index 3078511f76083..21b6c4d94a632 100644
@@ -1456,21 +1456,16 @@ static int ax88179_rx_fixup(struct usbnet *dev, struct sk_buff *skb)
 			/* Skip IP alignment pseudo header */
 			skb_pull(skb, 2);
 
-			skb->truesize = SKB_TRUESIZE(pkt_len_plus_padd);
 			ax88179_rx_checksum(skb, pkt_hdr);
 			return 1;
 		}
 
-		ax_skb = skb_clone(skb, GFP_ATOMIC);
+		ax_skb = netdev_alloc_skb_ip_align(dev->net, pkt_len);
 		if (!ax_skb)
 			return 0;
-		skb_trim(ax_skb, pkt_len);
+		skb_put(ax_skb, pkt_len);
+		memcpy(ax_skb->data, skb->data + 2, pkt_len);
 
-		/* Skip IP alignment pseudo header */
-		skb_pull(ax_skb, 2);
-
-		skb->truesize = pkt_len_plus_padd +
-				SKB_DATA_ALIGN(sizeof(struct sk_buff));
 		ax88179_rx_checksum(ax_skb, pkt_hdr);
 		usbnet_skb_return(dev, ax_skb);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:08 UTC #68 lore raw

PATCH 63/186 net: gtp: Fix Use-After-Free in gtp_dellink

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hyunwoo Kim <[email protected]>

[ Upstream commit f2a904107ee2b647bb7794a1a82b67740d7c8a64 ]

Since call_rcu, which is called in the hlist_for_each_entry_rcu traversal of gtp_dellink, is not part of the RCU read critical section, it is possible that the RCU grace period will pass during the traversal and the key will be free.

To prevent this, it should be changed to hlist_for_each_entry_safe.

Fixes 94dc550a5062 ("gtp: fix an use-after-free in ipv4_pdp_find()")

Signed-off-by Hyunwoo Kim <[email protected]>

Reviewed-by Eric Dumazet <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/gtp.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

drivers/net/gtp.c +2 -1

index 2b5357d94ff56..63f932256c9f5 100644
@@ -1111,11 +1111,12 @@ static int gtp_newlink(struct net *src_net, struct net_device *dev,
 static void gtp_dellink(struct net_device *dev, struct list_head *head)
 {
 	struct gtp_dev *gtp = netdev_priv(dev);
+	struct hlist_node *next;
 	struct pdp_ctx *pctx;
 	int i;
 
 	for (i = 0; i < gtp->hash_size; i++)
-		hlist_for_each_entry_rcu(pctx, &gtp->tid_hash[i], hlist_tid)
+		hlist_for_each_entry_safe(pctx, next, &gtp->tid_hash[i], hlist_tid)
 			pdp_context_delete(pctx);
 
 	list_del_rcu(&gtp->list);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:08 UTC #69 lore raw

PATCH 64/186 net: phy: mediatek-ge-soc: follow netdev LED trigger semantics

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Daniel Golle <[email protected]>

[ Upstream commit 5b5f724b05c550e10693a53a81cadca901aefd16 ]

Only blink if the link is up on a LED which is programmed to also indicate link-status.

Otherwise, if both LEDs are in use to indicate different speeds, the resulting blinking being inverted on LEDs which aren't switched on at a specific speed is quite counter-intuitive.

Also make sure that state left behind by reset or the bootloader is recognized correctly including the half-duplex and full-duplex bits as well as the (unsupported by Linux netdev trigger semantics) link-down bit.

Fixes c66937b0f8db ("net: phy: mediatek-ge-soc: support PHY LEDs")

Signed-off-by Daniel Golle <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/phy/mediatek-ge-soc.c | 43 +++++++++++++++++++------------
 1 file changed, 26 insertions(+), 17 deletions(-)

drivers/net/phy/mediatek-ge-soc.c +26 ~~-17~~

index 0f3a1538a8b8e..f4f9412d0cd7e 100644
@@ -216,6 +216,9 @@
 #define   MTK_PHY_LED_ON_LINK1000		BIT(0)
 #define   MTK_PHY_LED_ON_LINK100		BIT(1)
 #define   MTK_PHY_LED_ON_LINK10			BIT(2)
+#define   MTK_PHY_LED_ON_LINK			(MTK_PHY_LED_ON_LINK10 |\
+						 MTK_PHY_LED_ON_LINK100 |\
+						 MTK_PHY_LED_ON_LINK1000)
 #define   MTK_PHY_LED_ON_LINKDOWN		BIT(3)
 #define   MTK_PHY_LED_ON_FDX			BIT(4) /* Full duplex */
 #define   MTK_PHY_LED_ON_HDX			BIT(5) /* Half duplex */

@@ -231,6 +234,12 @@
 #define   MTK_PHY_LED_BLINK_100RX		BIT(3)
 #define   MTK_PHY_LED_BLINK_10TX		BIT(4)
 #define   MTK_PHY_LED_BLINK_10RX		BIT(5)
+#define   MTK_PHY_LED_BLINK_RX			(MTK_PHY_LED_BLINK_10RX |\
+						 MTK_PHY_LED_BLINK_100RX |\
+						 MTK_PHY_LED_BLINK_1000RX)
+#define   MTK_PHY_LED_BLINK_TX			(MTK_PHY_LED_BLINK_10TX |\
+						 MTK_PHY_LED_BLINK_100TX |\
+						 MTK_PHY_LED_BLINK_1000TX)
 #define   MTK_PHY_LED_BLINK_COLLISION		BIT(6)
 #define   MTK_PHY_LED_BLINK_RX_CRC_ERR		BIT(7)
 #define   MTK_PHY_LED_BLINK_RX_IDLE_ERR		BIT(8)

@@ -1247,11 +1256,9 @@ static int mt798x_phy_led_hw_control_get(struct phy_device *phydev, u8 index,
 	if (blink < 0)
 		return -EIO;
 
-	if ((on & (MTK_PHY_LED_ON_LINK1000 | MTK_PHY_LED_ON_LINK100 |
-		   MTK_PHY_LED_ON_LINK10)) ||
-	    (blink & (MTK_PHY_LED_BLINK_1000RX | MTK_PHY_LED_BLINK_100RX |
-		      MTK_PHY_LED_BLINK_10RX | MTK_PHY_LED_BLINK_1000TX |
-		      MTK_PHY_LED_BLINK_100TX | MTK_PHY_LED_BLINK_10TX)))
+	if ((on & (MTK_PHY_LED_ON_LINK | MTK_PHY_LED_ON_FDX | MTK_PHY_LED_ON_HDX |
+		   MTK_PHY_LED_ON_LINKDOWN)) ||
+	    (blink & (MTK_PHY_LED_BLINK_RX | MTK_PHY_LED_BLINK_TX)))
 		set_bit(bit_netdev, &priv->led_state);
 	else
 		clear_bit(bit_netdev, &priv->led_state);

@@ -1269,7 +1276,7 @@ static int mt798x_phy_led_hw_control_get(struct phy_device *phydev, u8 index,
 	if (!rules)
 		return 0;
 
-	if (on & (MTK_PHY_LED_ON_LINK1000 | MTK_PHY_LED_ON_LINK100 | MTK_PHY_LED_ON_LINK10))
+	if (on & MTK_PHY_LED_ON_LINK)
 		*rules |= BIT(TRIGGER_NETDEV_LINK);
 
 	if (on & MTK_PHY_LED_ON_LINK10)

@@ -1287,10 +1294,10 @@ static int mt798x_phy_led_hw_control_get(struct phy_device *phydev, u8 index,
 	if (on & MTK_PHY_LED_ON_HDX)
 		*rules |= BIT(TRIGGER_NETDEV_HALF_DUPLEX);
 
-	if (blink & (MTK_PHY_LED_BLINK_1000RX | MTK_PHY_LED_BLINK_100RX | MTK_PHY_LED_BLINK_10RX))
+	if (blink & MTK_PHY_LED_BLINK_RX)
 		*rules |= BIT(TRIGGER_NETDEV_RX);
 
-	if (blink & (MTK_PHY_LED_BLINK_1000TX | MTK_PHY_LED_BLINK_100TX | MTK_PHY_LED_BLINK_10TX))
+	if (blink & MTK_PHY_LED_BLINK_TX)
 		*rules |= BIT(TRIGGER_NETDEV_TX);
 
 	return 0;

@@ -1323,15 +1330,19 @@ static int mt798x_phy_led_hw_control_set(struct phy_device *phydev, u8 index,
 		on |= MTK_PHY_LED_ON_LINK1000;
 
 	if (rules & BIT(TRIGGER_NETDEV_RX)) {
-		blink |= MTK_PHY_LED_BLINK_10RX  |
-			 MTK_PHY_LED_BLINK_100RX |
-			 MTK_PHY_LED_BLINK_1000RX;
+		blink |= (on & MTK_PHY_LED_ON_LINK) ?
+			  (((on & MTK_PHY_LED_ON_LINK10) ? MTK_PHY_LED_BLINK_10RX : 0) |
+			   ((on & MTK_PHY_LED_ON_LINK100) ? MTK_PHY_LED_BLINK_100RX : 0) |
+			   ((on & MTK_PHY_LED_ON_LINK1000) ? MTK_PHY_LED_BLINK_1000RX : 0)) :
+			  MTK_PHY_LED_BLINK_RX;
 	}
 
 	if (rules & BIT(TRIGGER_NETDEV_TX)) {
-		blink |= MTK_PHY_LED_BLINK_10TX  |
-			 MTK_PHY_LED_BLINK_100TX |
-			 MTK_PHY_LED_BLINK_1000TX;
+		blink |= (on & MTK_PHY_LED_ON_LINK) ?
+			  (((on & MTK_PHY_LED_ON_LINK10) ? MTK_PHY_LED_BLINK_10TX : 0) |
+			   ((on & MTK_PHY_LED_ON_LINK100) ? MTK_PHY_LED_BLINK_100TX : 0) |
+			   ((on & MTK_PHY_LED_ON_LINK1000) ? MTK_PHY_LED_BLINK_1000TX : 0)) :
+			  MTK_PHY_LED_BLINK_TX;
 	}
 
 	if (blink || on)

@@ -1344,9 +1355,7 @@ static int mt798x_phy_led_hw_control_set(struct phy_device *phydev, u8 index,
 				MTK_PHY_LED0_ON_CTRL,
 			     MTK_PHY_LED_ON_FDX     |
 			     MTK_PHY_LED_ON_HDX     |
-			     MTK_PHY_LED_ON_LINK10  |
-			     MTK_PHY_LED_ON_LINK100 |
-			     MTK_PHY_LED_ON_LINK1000,
+			     MTK_PHY_LED_ON_LINK,
 			     on);
 
 	if (ret)

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:08 UTC #70 lore raw

PATCH 65/186 gpio: tegra186: Fix tegra186_gpio_is_accessible() check

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Prathamesh Shete <[email protected]>

[ Upstream commit d806f474a9a7993648a2c70642ee129316d8deff ]

The controller has several register bits describing access control information for a given GPIO pin. When SCR_SEC_[R|W]EN is unset, it means we have full read/write access to all the registers for given GPIO pin. When SCR_SEC[R|W]EN is set, it means we need to further check the accompanying SCR_SEC_G1[R|W] bit to determine read/write access to all the registers for given GPIO pin.

This check was previously declaring that a GPIO pin was accessible only if either of the following conditions were met:

- SCR_SEC_REN + SCR_SEC_WEN both set

- SCR_SEC_REN + SCR_SEC_WEN both set and
SCR_SEC_G1R + SCR_SEC_G1W both set

Update the check to properly handle cases where only one of SCR_SEC_REN or SCR_SEC_WEN is set.

Fixes b2b56a163230 ("gpio: tegra186: Check GPIO pin permission before access.")

Signed-off-by Prathamesh Shete <[email protected]>

Acked-by Thierry Reding <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Bartosz Golaszewski <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/gpio/gpio-tegra186.c | 20 +++++++++++---------
 1 file changed, 11 insertions(+), 9 deletions(-)

drivers/gpio/gpio-tegra186.c +11 -9

index d87dd06db40d0..9130c691a2dd3 100644
@@ -36,12 +36,6 @@
 #define  TEGRA186_GPIO_SCR_SEC_REN		BIT(27)
 #define  TEGRA186_GPIO_SCR_SEC_G1W		BIT(9)
 #define  TEGRA186_GPIO_SCR_SEC_G1R		BIT(1)
-#define  TEGRA186_GPIO_FULL_ACCESS		(TEGRA186_GPIO_SCR_SEC_WEN | \
-						 TEGRA186_GPIO_SCR_SEC_REN | \
-						 TEGRA186_GPIO_SCR_SEC_G1R | \
-						 TEGRA186_GPIO_SCR_SEC_G1W)
-#define  TEGRA186_GPIO_SCR_SEC_ENABLE		(TEGRA186_GPIO_SCR_SEC_WEN | \
-						 TEGRA186_GPIO_SCR_SEC_REN)
 
 /* control registers */
 #define TEGRA186_GPIO_ENABLE_CONFIG 0x00

@@ -177,10 +171,18 @@ static inline bool tegra186_gpio_is_accessible(struct tegra_gpio *gpio, unsigned
 
 	value = __raw_readl(secure + TEGRA186_GPIO_SCR);
 
-	if ((value & TEGRA186_GPIO_SCR_SEC_ENABLE) == 0)
-		return true;
+	/*
+	 * When SCR_SEC_[R|W]EN is unset, then we have full read/write access to all the
+	 * registers for given GPIO pin.
+	 * When SCR_SEC[R|W]EN is set, then there is need to further check the accompanying
+	 * SCR_SEC_G1[R|W] bit to determine read/write access to all the registers for given
+	 * GPIO pin.
+	 */
 
-	if ((value & TEGRA186_GPIO_FULL_ACCESS) == TEGRA186_GPIO_FULL_ACCESS)
+	if (((value & TEGRA186_GPIO_SCR_SEC_REN) == 0 ||
+	     ((value & TEGRA186_GPIO_SCR_SEC_REN) && (value & TEGRA186_GPIO_SCR_SEC_G1R))) &&
+	     ((value & TEGRA186_GPIO_SCR_SEC_WEN) == 0 ||
+	     ((value & TEGRA186_GPIO_SCR_SEC_WEN) && (value & TEGRA186_GPIO_SCR_SEC_G1W))))
 		return true;
 
 	return false;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:08 UTC #71 lore raw

PATCH 66/186 Bluetooth: btusb: Fix triggering coredump implementation for QCA

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Zijun Hu <[email protected]>

[ Upstream commit b23d98d46d2858dcc0fd016caff165cbdc24e70a ]

btusb_coredump_qca() uses __hci_cmd_sync() to send a vendor-specific command to trigger firmware coredump, but the command does not have any event as its sync response, so it is not suitable to use __hci_cmd_sync(), fixed by using __hci_cmd_send().

Fixes 20981ce2d5a5 ("Bluetooth: btusb: Add WCN6855 devcoredump support")

Signed-off-by Zijun Hu <[email protected]>

Signed-off-by Luiz Augusto von Dentz <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/bluetooth/btusb.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

drivers/bluetooth/btusb.c +4 -5

index 1976593bc804e..02f38417aae3d 100644
@@ -3457,13 +3457,12 @@ static void btusb_dump_hdr_qca(struct hci_dev *hdev, struct sk_buff *skb)
 
 static void btusb_coredump_qca(struct hci_dev *hdev)
 {
+	int err;
 	static const u8 param[] = { 0x26 };
-	struct sk_buff *skb;
 
-	skb = __hci_cmd_sync(hdev, 0xfc0c, 1, param, HCI_CMD_TIMEOUT);
-	if (IS_ERR(skb))
-		bt_dev_err(hdev, "%s: triggle crash failed (%ld)", __func__, PTR_ERR(skb));
-	kfree_skb(skb);
+	err = __hci_cmd_send(hdev, 0xfc0c, 1, param);
+	if (err < 0)
+		bt_dev_err(hdev, "%s: triggle crash failed (%d)", __func__, err);
 }
 
 /*

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:09 UTC #72 lore raw

PATCH 67/186 Bluetooth: hci_event: Fix sending HCI_OP_READ_ENC_KEY_SIZE

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Luiz Augusto von Dentz <[email protected]>

[ Upstream commit a9a830a676a9a93c5020f5c61236166931fa4266 ]

The code shall always check if HCI_QUIRK_BROKEN_READ_ENC_KEY_SIZE has been set before attempting to use HCI_OP_READ_ENC_KEY_SIZE.

Fixes c569242cd492 ("Bluetooth: hci_event: set the conn encrypted before conn establishes")

Signed-off-by Luiz Augusto von Dentz <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 include/net/bluetooth/hci_core.h | 4 ++++
 net/bluetooth/hci_event.c        | 5 ++---
 2 files changed, 6 insertions(+), 3 deletions(-)

include/net/bluetooth/hci_core.h +4 -0

index 103b290d6efb0..e6f659ce534e6 100644
@@ -1865,6 +1865,10 @@ void hci_conn_del_sysfs(struct hci_conn *conn);
 #define privacy_mode_capable(dev) (use_ll_privacy(dev) && \
 				   (hdev->commands[39] & 0x04))
 
+#define read_key_size_capable(dev) \
+	((dev)->commands[20] & 0x10 && \
+	 !test_bit(HCI_QUIRK_BROKEN_READ_ENC_KEY_SIZE, &hdev->quirks))
+
 /* Use enhanced synchronous connection if command is supported and its quirk
  * has not been set.
  */

net/bluetooth/hci_event.c +2 -3

index 80e71ce32f09f..1b4abf8e90f6b 100644
@@ -3229,7 +3229,7 @@ static void hci_conn_complete_evt(struct hci_dev *hdev, void *data,
 			if (key) {
 				set_bit(HCI_CONN_ENCRYPT, &conn->flags);
 
-				if (!(hdev->commands[20] & 0x10)) {
+				if (!read_key_size_capable(hdev)) {
 					conn->enc_key_size = HCI_LINK_KEY_SIZE;
 				} else {
 					cp.handle = cpu_to_le16(conn->handle);

@@ -3679,8 +3679,7 @@ static void hci_encrypt_change_evt(struct hci_dev *hdev, void *data,
 		 * controller really supports it. If it doesn't, assume
 		 * the default size (16).
 		 */
-		if (!(hdev->commands[20] & 0x10) ||
-		    test_bit(HCI_QUIRK_BROKEN_READ_ENC_KEY_SIZE, &hdev->quirks)) {
+		if (!read_key_size_capable(hdev)) {
 			conn->enc_key_size = HCI_LINK_KEY_SIZE;
 			goto notify;
 		}

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:09 UTC #73 lore raw

PATCH 68/186 Bluetooth: MGMT: Fix failing to MGMT_OP_ADD_UUID/MGMT_OP_REMOVE_UUID

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Luiz Augusto von Dentz <[email protected]>

[ Upstream commit 6eb5fcc416f127f220b9177a5c9ae751cac1cda8 ]

These commands don't require the adapter to be up and running so don't use hci_cmd_sync_queue which would check that flag, instead use hci_cmd_sync_submit which would ensure mgmt_class_complete is set properly regardless if any command was actually run or not.

Link https://github.com/bluez/bluez/issues/809

Fixes d883a4669a1d ("Bluetooth: hci_sync: Only allow hci_cmd_sync_queue if running")

Signed-off-by Luiz Augusto von Dentz <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/bluetooth/mgmt.c | 20 +++++++++++++++-----
 1 file changed, 15 insertions(+), 5 deletions(-)

net/bluetooth/mgmt.c +15 -5

index 92fd3786bbdff..6d41bc0e7896a 100644
@@ -2616,7 +2616,11 @@ static int add_uuid(struct sock *sk, struct hci_dev *hdev, void *data, u16 len)
 		goto failed;
 	}
 
-	err = hci_cmd_sync_queue(hdev, add_uuid_sync, cmd, mgmt_class_complete);
+	/* MGMT_OP_ADD_UUID don't require adapter the UP/Running so use
+	 * hci_cmd_sync_submit instead of hci_cmd_sync_queue.
+	 */
+	err = hci_cmd_sync_submit(hdev, add_uuid_sync, cmd,
+				  mgmt_class_complete);
 	if (err < 0) {
 		mgmt_pending_free(cmd);
 		goto failed;

@@ -2710,8 +2714,11 @@ static int remove_uuid(struct sock *sk, struct hci_dev *hdev, void *data,
 		goto unlock;
 	}
 
-	err = hci_cmd_sync_queue(hdev, remove_uuid_sync, cmd,
-				 mgmt_class_complete);
+	/* MGMT_OP_REMOVE_UUID don't require adapter the UP/Running so use
+	 * hci_cmd_sync_submit instead of hci_cmd_sync_queue.
+	 */
+	err = hci_cmd_sync_submit(hdev, remove_uuid_sync, cmd,
+				  mgmt_class_complete);
 	if (err < 0)
 		mgmt_pending_free(cmd);

@@ -2777,8 +2784,11 @@ static int set_dev_class(struct sock *sk, struct hci_dev *hdev, void *data,
 		goto unlock;
 	}
 
-	err = hci_cmd_sync_queue(hdev, set_class_sync, cmd,
-				 mgmt_class_complete);
+	/* MGMT_OP_SET_DEV_CLASS don't require adapter the UP/Running so use
+	 * hci_cmd_sync_submit instead of hci_cmd_sync_queue.
+	 */
+	err = hci_cmd_sync_submit(hdev, set_class_sync, cmd,
+				  mgmt_class_complete);
 	if (err < 0)
 		mgmt_pending_free(cmd);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:09 UTC #74 lore raw

PATCH 69/186 Bluetooth: btusb: mediatek: Fix double free of skb in coredump

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sean Wang <[email protected]>

[ Upstream commit 18bdb386a1a30e7a3d7732a98e45e69cf6b5710d ]

hci_devcd_append() would free the skb on error so the caller don't have to free it again otherwise it would cause the double free of skb.

Fixes 0b7015132878 ("Bluetooth: btusb: mediatek: add MediaTek devcoredump support")

Reported-by Dan Carpenter <[email protected]>

Signed-off-by Sean Wang <[email protected]>

Signed-off-by Luiz Augusto von Dentz <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/bluetooth/btmtk.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

drivers/bluetooth/btmtk.c +3 -4

index ac8ebccd35075..812fd2a8f853e 100644
@@ -380,8 +380,10 @@ int btmtk_process_coredump(struct hci_dev *hdev, struct sk_buff *skb)
 	switch (data->cd_info.state) {
 	case HCI_DEVCOREDUMP_IDLE:
 		err = hci_devcd_init(hdev, MTK_COREDUMP_SIZE);
-		if (err < 0)
+		if (err < 0) {
+			kfree_skb(skb);
 			break;
+		}
 		data->cd_info.cnt = 0;
 
 		/* It is supposed coredump can be done within 5 seconds */

@@ -407,9 +409,6 @@ int btmtk_process_coredump(struct hci_dev *hdev, struct sk_buff *skb)
 		break;
 	}
 
-	if (err < 0)
-		kfree_skb(skb);
-
 	return err;
 }
 EXPORT_SYMBOL_GPL(btmtk_process_coredump);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:09 UTC #75 lore raw

PATCH 70/186 Bluetooth: hci_sync: Using hci_cmd_sync_submit when removing Adv Monitor

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Chun-Yi Lee <[email protected]>

[ Upstream commit 88cd6e6b2d327faa13e4505b07f1e380e51b21ff ]

Since the d883a4669a1de be introduced in v6.4, bluetooth daemon got the following failed message of MGMT_OP_REMOVE_ADV_MONITOR command when controller is power-off:

bluetoothd[20976]:
src/adapter.c:reset_adv_monitors_complete() Failed to reset Adv
Monitors: Failed>

Normally this situation is happened when the bluetoothd deamon be started manually after system booting. Which means that bluetoothd received MGMT_EV_INDEX_ADDED event after kernel runs hci_power_off().

Base on doc/mgmt-api.txt, the MGMT_OP_REMOVE_ADV_MONITOR command can be used when the controller is not powered. This patch changes the code in remove_adv_monitor() to use hci_cmd_sync_submit() instead of hci_cmd_sync_queue().

Fixes d883a4669a1de ("Bluetooth: hci_sync: Only allow hci_cmd_sync_queue if running")

Cc Luiz Augusto von Dentz <[email protected]>

Cc Manish Mandlik <[email protected]>

Cc Archie Pusaka <[email protected]>

Cc Miao-chen Chou <[email protected]>

Signed-off-by Chun-Yi Lee <[email protected]>

Signed-off-by Luiz Augusto von Dentz <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/bluetooth/mgmt.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

net/bluetooth/mgmt.c +2 -2

index 6d41bc0e7896a..ac693e64f1f9f 100644
@@ -5477,8 +5477,8 @@ static int remove_adv_monitor(struct sock *sk, struct hci_dev *hdev,
 		goto unlock;
 	}
 
-	err = hci_cmd_sync_queue(hdev, mgmt_remove_adv_monitor_sync, cmd,
-				 mgmt_remove_adv_monitor_complete);
+	err = hci_cmd_sync_submit(hdev, mgmt_remove_adv_monitor_sync, cmd,
+				  mgmt_remove_adv_monitor_complete);
 
 	if (err) {
 		mgmt_pending_remove(cmd);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:09 UTC #76 lore raw

PATCH 71/186 Bluetooth: qca: set power_ctrl_enabled on NULL returned by gpiod_get_optional()

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Bartosz Golaszewski <[email protected]>

[ Upstream commit 3d05fc82237aa97162d0d7dc300b55bb34e91d02 ]

Any return value from gpiod_get_optional() other than a pointer to a GPIO descriptor or a NULL-pointer is an error and the driver should abort probing. That being said: commit 56d074d26c58 ("Bluetooth: hci_qca:
don't use IS_ERR_OR_NULL() with gpiod_get_optional()") no longer sets power_ctrl_enabled on NULL-pointer returned by devm_gpiod_get_optional(). Restore this behavior but bail-out on errors. While at it: also bail-out on error returned when trying to get the "swctrl" GPIO.

Reported-by Wren Turkal <[email protected]>

Reported-by Zijun Hu <[email protected]>

Closes https://lore.kernel.org/linux-bluetooth/[email protected]/ open here

Fixes 56d074d26c58 ("Bluetooth: hci_qca: don't use IS_ERR_OR_NULL() with gpiod_get_optional()")

Reviewed-by Krzysztof Kozlowski <[email protected]>

Signed-off-by Bartosz Golaszewski <[email protected]>

Tested-by Wren Turkal <[email protected]>

Reported-by Wren Turkal <[email protected]>

Reported-by Zijun Hu <[email protected]>

Reviewed-by Krzysztof Kozlowski<[email protected]>

Reviewed-by Krzysztof Kozlowski <[email protected]>

Signed-off-by Luiz Augusto von Dentz <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/bluetooth/hci_qca.c | 18 +++++++++++++-----
 1 file changed, 13 insertions(+), 5 deletions(-)

drivers/bluetooth/hci_qca.c +13 -5

index 8861b8017fbdf..b491dd9acf315 100644
@@ -2298,16 +2298,21 @@ static int qca_serdev_probe(struct serdev_device *serdev)
 		    (data->soc_type == QCA_WCN6750 ||
 		     data->soc_type == QCA_WCN6855)) {
 			dev_err(&serdev->dev, "failed to acquire BT_EN gpio\n");
-			power_ctrl_enabled = false;
+			return PTR_ERR(qcadev->bt_en);
 		}
 
+		if (!qcadev->bt_en)
+			power_ctrl_enabled = false;
+
 		qcadev->sw_ctrl = devm_gpiod_get_optional(&serdev->dev, "swctrl",
 					       GPIOD_IN);
 		if (IS_ERR(qcadev->sw_ctrl) &&
 		    (data->soc_type == QCA_WCN6750 ||
 		     data->soc_type == QCA_WCN6855 ||
-		     data->soc_type == QCA_WCN7850))
-			dev_warn(&serdev->dev, "failed to acquire SW_CTRL gpio\n");
+		     data->soc_type == QCA_WCN7850)) {
+			dev_err(&serdev->dev, "failed to acquire SW_CTRL gpio\n");
+			return PTR_ERR(qcadev->sw_ctrl);
+		}
 
 		qcadev->susclk = devm_clk_get_optional(&serdev->dev, NULL);
 		if (IS_ERR(qcadev->susclk)) {

@@ -2326,10 +2331,13 @@ static int qca_serdev_probe(struct serdev_device *serdev)
 		qcadev->bt_en = devm_gpiod_get_optional(&serdev->dev, "enable",
 					       GPIOD_OUT_LOW);
 		if (IS_ERR(qcadev->bt_en)) {
-			dev_warn(&serdev->dev, "failed to acquire enable gpio\n");
-			power_ctrl_enabled = false;
+			dev_err(&serdev->dev, "failed to acquire enable gpio\n");
+			return PTR_ERR(qcadev->bt_en);
 		}
 
+		if (!qcadev->bt_en)
+			power_ctrl_enabled = false;
+
 		qcadev->susclk = devm_clk_get_optional(&serdev->dev, NULL);
 		if (IS_ERR(qcadev->susclk)) {
 			dev_warn(&serdev->dev, "failed to acquire clk\n");

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:09 UTC #77 lore raw

PATCH 72/186 ipvs: Fix checksumming on GSO of SCTP packets

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ismael Luceno <[email protected]>

[ Upstream commit e10d3ba4d434ed172914617ed8d74bd411421193 ]

It was observed in the wild that pairs of consecutive packets would leave the IPVS with the same wrong checksum, and the issue only went away when disabling GSO.

IPVS needs to avoid computing the SCTP checksum when using GSO.

Fixes 90017accff61 ("sctp: Add GSO support")

Co-developed-by Firo Yang <[email protected]>

Signed-off-by Ismael Luceno <[email protected]>

Tested-by Andreas Taschner <[email protected]>

Acked-by Julian Anastasov <[email protected]>

Signed-off-by Pablo Neira Ayuso <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/netfilter/ipvs/ip_vs_proto_sctp.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

net/netfilter/ipvs/ip_vs_proto_sctp.c +4 -2

index a0921adc31a9f..1e689c7141271 100644
@@ -126,7 +126,8 @@ sctp_snat_handler(struct sk_buff *skb, struct ip_vs_protocol *pp,
 	if (sctph->source != cp->vport || payload_csum ||
 	    skb->ip_summed == CHECKSUM_PARTIAL) {
 		sctph->source = cp->vport;
-		sctp_nat_csum(skb, sctph, sctphoff);
+		if (!skb_is_gso(skb) || !skb_is_gso_sctp(skb))
+			sctp_nat_csum(skb, sctph, sctphoff);
 	} else {
 		skb->ip_summed = CHECKSUM_UNNECESSARY;
 	}

@@ -174,7 +175,8 @@ sctp_dnat_handler(struct sk_buff *skb, struct ip_vs_protocol *pp,
 	    (skb->ip_summed == CHECKSUM_PARTIAL &&
 	     !(skb_dst(skb)->dev->features & NETIF_F_SCTP_CRC))) {
 		sctph->dest = cp->dport;
-		sctp_nat_csum(skb, sctph, sctphoff);
+		if (!skb_is_gso(skb) || !skb_is_gso_sctp(skb))
+			sctp_nat_csum(skb, sctph, sctphoff);
 	} else if (skb->ip_summed != CHECKSUM_PARTIAL) {
 		skb->ip_summed = CHECKSUM_UNNECESSARY;
 	}

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:09 UTC #78 lore raw

PATCH 73/186 net: openvswitch: Fix Use-After-Free in ovs_ct_exit

Link https://lore.kernel.org/r/ZiYvzQN/Ry5oeFQW@v4bel-B760M-AORUS-ELITE-AX open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hyunwoo Kim <[email protected]>

[ Upstream commit 5ea7b72d4fac2fdbc0425cd8f2ea33abe95235b2 ]

Since kfree_rcu, which is called in the hlist_for_each_entry_rcu traversal of ovs_ct_limit_exit, is not part of the RCU read critical section, it is possible that the RCU grace period will pass during the traversal and the key will be free.

To prevent this, it should be changed to hlist_for_each_entry_safe.

Fixes 11efd5cb04a1 ("openvswitch: Support conntrack zone limit")

Signed-off-by Hyunwoo Kim <[email protected]>

Reviewed-by Eric Dumazet <[email protected]>

Reviewed-by Aaron Conole <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/openvswitch/conntrack.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

net/openvswitch/conntrack.c +2 -2

index 74b63cdb59923..2928c142a2ddb 100644
@@ -1593,9 +1593,9 @@ static void ovs_ct_limit_exit(struct net *net, struct ovs_net *ovs_net)
 	for (i = 0; i < CT_LIMIT_HASH_BUCKETS; ++i) {
 		struct hlist_head *head = &info->limits[i];
 		struct ovs_ct_limit *ct_limit;
+		struct hlist_node *next;
 
-		hlist_for_each_entry_rcu(ct_limit, head, hlist_node,
-					 lockdep_ovsl_is_held())
+		hlist_for_each_entry_safe(ct_limit, next, head, hlist_node)
 			kfree_rcu(ct_limit, rcu);
 	}
 	kfree(info->limits);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:09 UTC #79 lore raw

PATCH 74/186 mlxsw: Use refcount_t for reference counting

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Amit Cohen <[email protected]>

[ Upstream commit 1267f7223bec186dc26ef4e6075496c6217355de ]

mlxsw driver uses 'unsigned int' for reference counters in several structures. Instead, use refcount_t type which allows us to catch overflow and underflow issues. Change the type of the counters and use the appropriate API.

Signed-off-by Amit Cohen <[email protected]>

Reviewed-by Ido Schimmel <[email protected]>

Signed-off-by Paolo Abeni <[email protected]>

Stable-dep-of: 627f9c1bb882 ("mlxsw: spectrum_acl_tcam: Fix race in region ID allocation")

Signed-off-by Sasha Levin <[email protected]>

---
 .../mellanox/mlxsw/core_acl_flex_actions.c      | 16 ++++++++--------
 .../mellanox/mlxsw/core_acl_flex_keys.c         |  9 +++++----
 .../net/ethernet/mellanox/mlxsw/spectrum_acl.c  | 11 ++++++-----
 .../ethernet/mellanox/mlxsw/spectrum_acl_tcam.c | 17 +++++++++--------
 .../ethernet/mellanox/mlxsw/spectrum_router.c   | 15 ++++++++-------
 .../mellanox/mlxsw/spectrum_switchdev.c         |  8 ++++----
 6 files changed, 40 insertions(+), 36 deletions(-)

drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_actions.c +8 -8

index faa63ea9b83e1..1915fa41c6224 100644
@@ -95,7 +95,7 @@ struct mlxsw_afa_set {
 		      */
 	   has_trap:1,
 	   has_police:1;
-	unsigned int ref_count;
+	refcount_t ref_count;
 	struct mlxsw_afa_set *next; /* Pointer to the next set. */
 	struct mlxsw_afa_set *prev; /* Pointer to the previous set,
 				     * note that set may have multiple

@@ -120,7 +120,7 @@ struct mlxsw_afa_fwd_entry {
 	struct rhash_head ht_node;
 	struct mlxsw_afa_fwd_entry_ht_key ht_key;
 	u32 kvdl_index;
-	unsigned int ref_count;
+	refcount_t ref_count;
 };
 
 static const struct rhashtable_params mlxsw_afa_fwd_entry_ht_params = {

@@ -282,7 +282,7 @@ static struct mlxsw_afa_set *mlxsw_afa_set_create(bool is_first)
 	/* Need to initialize the set to pass by default */
 	mlxsw_afa_set_goto_set(set, MLXSW_AFA_SET_GOTO_BINDING_CMD_TERM, 0);
 	set->ht_key.is_first = is_first;
-	set->ref_count = 1;
+	refcount_set(&set->ref_count, 1);
 	return set;
 }

@@ -330,7 +330,7 @@ static void mlxsw_afa_set_unshare(struct mlxsw_afa *mlxsw_afa,
 static void mlxsw_afa_set_put(struct mlxsw_afa *mlxsw_afa,
 			      struct mlxsw_afa_set *set)
 {
-	if (--set->ref_count)
+	if (!refcount_dec_and_test(&set->ref_count))
 		return;
 	if (set->shared)
 		mlxsw_afa_set_unshare(mlxsw_afa, set);

@@ -350,7 +350,7 @@ static struct mlxsw_afa_set *mlxsw_afa_set_get(struct mlxsw_afa *mlxsw_afa,
 	set = rhashtable_lookup_fast(&mlxsw_afa->set_ht, &orig_set->ht_key,
 				     mlxsw_afa_set_ht_params);
 	if (set) {
-		set->ref_count++;
+		refcount_inc(&set->ref_count);
 		mlxsw_afa_set_put(mlxsw_afa, orig_set);
 	} else {
 		set = orig_set;

@@ -564,7 +564,7 @@ mlxsw_afa_fwd_entry_create(struct mlxsw_afa *mlxsw_afa, u16 local_port)
 	if (!fwd_entry)
 		return ERR_PTR(-ENOMEM);
 	fwd_entry->ht_key.local_port = local_port;
-	fwd_entry->ref_count = 1;
+	refcount_set(&fwd_entry->ref_count, 1);
 
 	err = rhashtable_insert_fast(&mlxsw_afa->fwd_entry_ht,
 				     &fwd_entry->ht_node,

@@ -607,7 +607,7 @@ mlxsw_afa_fwd_entry_get(struct mlxsw_afa *mlxsw_afa, u16 local_port)
 	fwd_entry = rhashtable_lookup_fast(&mlxsw_afa->fwd_entry_ht, &ht_key,
 					   mlxsw_afa_fwd_entry_ht_params);
 	if (fwd_entry) {
-		fwd_entry->ref_count++;
+		refcount_inc(&fwd_entry->ref_count);
 		return fwd_entry;
 	}
 	return mlxsw_afa_fwd_entry_create(mlxsw_afa, local_port);

@@ -616,7 +616,7 @@ mlxsw_afa_fwd_entry_get(struct mlxsw_afa *mlxsw_afa, u16 local_port)
 static void mlxsw_afa_fwd_entry_put(struct mlxsw_afa *mlxsw_afa,
 				    struct mlxsw_afa_fwd_entry *fwd_entry)
 {
-	if (--fwd_entry->ref_count)
+	if (!refcount_dec_and_test(&fwd_entry->ref_count))
 		return;
 	mlxsw_afa_fwd_entry_destroy(mlxsw_afa, fwd_entry);
 }

drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.c +5 -4

index 70f9b5e85a26f..bf140e7416e19 100644
@@ -5,6 +5,7 @@
 #include <linux/slab.h>
 #include <linux/list.h>
 #include <linux/errno.h>
+#include <linux/refcount.h>
 
 #include "item.h"
 #include "core_acl_flex_keys.h"

@@ -105,7 +106,7 @@ EXPORT_SYMBOL(mlxsw_afk_destroy);
 
 struct mlxsw_afk_key_info {
 	struct list_head list;
-	unsigned int ref_count;
+	refcount_t ref_count;
 	unsigned int blocks_count;
 	int element_to_block[MLXSW_AFK_ELEMENT_MAX]; /* index is element, value
 						      * is index inside "blocks"

@@ -282,7 +283,7 @@ mlxsw_afk_key_info_create(struct mlxsw_afk *mlxsw_afk,
 	if (err)
 		goto err_picker;
 	list_add(&key_info->list, &mlxsw_afk->key_info_list);
-	key_info->ref_count = 1;
+	refcount_set(&key_info->ref_count, 1);
 	return key_info;
 
 err_picker:

@@ -304,7 +305,7 @@ mlxsw_afk_key_info_get(struct mlxsw_afk *mlxsw_afk,
 
 	key_info = mlxsw_afk_key_info_find(mlxsw_afk, elusage);
 	if (key_info) {
-		key_info->ref_count++;
+		refcount_inc(&key_info->ref_count);
 		return key_info;
 	}
 	return mlxsw_afk_key_info_create(mlxsw_afk, elusage);

@@ -313,7 +314,7 @@ EXPORT_SYMBOL(mlxsw_afk_key_info_get);
 
 void mlxsw_afk_key_info_put(struct mlxsw_afk_key_info *key_info)
 {
-	if (--key_info->ref_count)
+	if (!refcount_dec_and_test(&key_info->ref_count))
 		return;
 	mlxsw_afk_key_info_destroy(key_info);
 }

drivers/net/ethernet/mellanox/mlxsw/spectrum_acl.c +6 -5

index 7c59c8a135840..b01b000bc71c1 100644
@@ -9,6 +9,7 @@
 #include <linux/rhashtable.h>
 #include <linux/netdevice.h>
 #include <linux/mutex.h>
+#include <linux/refcount.h>
 #include <net/net_namespace.h>
 #include <net/tc_act/tc_vlan.h>

@@ -55,7 +56,7 @@ struct mlxsw_sp_acl_ruleset {
 	struct rhash_head ht_node; /* Member of acl HT */
 	struct mlxsw_sp_acl_ruleset_ht_key ht_key;
 	struct rhashtable rule_ht;
-	unsigned int ref_count;
+	refcount_t ref_count;
 	unsigned int min_prio;
 	unsigned int max_prio;
 	unsigned long priv[];

@@ -99,7 +100,7 @@ static bool
 mlxsw_sp_acl_ruleset_is_singular(const struct mlxsw_sp_acl_ruleset *ruleset)
 {
 	/* We hold a reference on ruleset ourselves */
-	return ruleset->ref_count == 2;
+	return refcount_read(&ruleset->ref_count) == 2;
 }
 
 int mlxsw_sp_acl_ruleset_bind(struct mlxsw_sp *mlxsw_sp,

@@ -176,7 +177,7 @@ mlxsw_sp_acl_ruleset_create(struct mlxsw_sp *mlxsw_sp,
 	ruleset = kzalloc(alloc_size, GFP_KERNEL);
 	if (!ruleset)
 		return ERR_PTR(-ENOMEM);
-	ruleset->ref_count = 1;
+	refcount_set(&ruleset->ref_count, 1);
 	ruleset->ht_key.block = block;
 	ruleset->ht_key.chain_index = chain_index;
 	ruleset->ht_key.ops = ops;

@@ -222,13 +223,13 @@ static void mlxsw_sp_acl_ruleset_destroy(struct mlxsw_sp *mlxsw_sp,
 
 static void mlxsw_sp_acl_ruleset_ref_inc(struct mlxsw_sp_acl_ruleset *ruleset)
 {
-	ruleset->ref_count++;
+	refcount_inc(&ruleset->ref_count);
 }
 
 static void mlxsw_sp_acl_ruleset_ref_dec(struct mlxsw_sp *mlxsw_sp,
 					 struct mlxsw_sp_acl_ruleset *ruleset)
 {
-	if (--ruleset->ref_count)
+	if (!refcount_dec_and_test(&ruleset->ref_count))
 		return;
 	mlxsw_sp_acl_ruleset_destroy(mlxsw_sp, ruleset);
 }

drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c +9 -8

index 50ea1eff02b2f..f20052776b3f2 100644
@@ -9,6 +9,7 @@
 #include <linux/rhashtable.h>
 #include <linux/netdevice.h>
 #include <linux/mutex.h>
+#include <linux/refcount.h>
 #include <net/devlink.h>
 #include <trace/events/mlxsw.h>

@@ -155,7 +156,7 @@ struct mlxsw_sp_acl_tcam_vregion {
 		struct mlxsw_sp_acl_tcam_rehash_ctx ctx;
 	} rehash;
 	struct mlxsw_sp *mlxsw_sp;
-	unsigned int ref_count;
+	refcount_t ref_count;
 };
 
 struct mlxsw_sp_acl_tcam_vchunk;

@@ -176,7 +177,7 @@ struct mlxsw_sp_acl_tcam_vchunk {
 	unsigned int priority; /* Priority within the vregion and group */
 	struct mlxsw_sp_acl_tcam_vgroup *vgroup;
 	struct mlxsw_sp_acl_tcam_vregion *vregion;
-	unsigned int ref_count;
+	refcount_t ref_count;
 };
 
 struct mlxsw_sp_acl_tcam_entry {

@@ -769,7 +770,7 @@ mlxsw_sp_acl_tcam_vregion_create(struct mlxsw_sp *mlxsw_sp,
 	vregion->tcam = tcam;
 	vregion->mlxsw_sp = mlxsw_sp;
 	vregion->vgroup = vgroup;
-	vregion->ref_count = 1;
+	refcount_set(&vregion->ref_count, 1);
 
 	vregion->key_info = mlxsw_afk_key_info_get(afk, elusage);
 	if (IS_ERR(vregion->key_info)) {

@@ -856,7 +857,7 @@ mlxsw_sp_acl_tcam_vregion_get(struct mlxsw_sp *mlxsw_sp,
 			 */
 			return ERR_PTR(-EOPNOTSUPP);
 		}
-		vregion->ref_count++;
+		refcount_inc(&vregion->ref_count);
 		return vregion;
 	}

@@ -871,7 +872,7 @@ static void
 mlxsw_sp_acl_tcam_vregion_put(struct mlxsw_sp *mlxsw_sp,
 			      struct mlxsw_sp_acl_tcam_vregion *vregion)
 {
-	if (--vregion->ref_count)
+	if (!refcount_dec_and_test(&vregion->ref_count))
 		return;
 	mlxsw_sp_acl_tcam_vregion_destroy(mlxsw_sp, vregion);
 }

@@ -924,7 +925,7 @@ mlxsw_sp_acl_tcam_vchunk_create(struct mlxsw_sp *mlxsw_sp,
 	INIT_LIST_HEAD(&vchunk->ventry_list);
 	vchunk->priority = priority;
 	vchunk->vgroup = vgroup;
-	vchunk->ref_count = 1;
+	refcount_set(&vchunk->ref_count, 1);
 
 	vregion = mlxsw_sp_acl_tcam_vregion_get(mlxsw_sp, vgroup,
 						priority, elusage);

@@ -1008,7 +1009,7 @@ mlxsw_sp_acl_tcam_vchunk_get(struct mlxsw_sp *mlxsw_sp,
 		if (WARN_ON(!mlxsw_afk_key_info_subset(vchunk->vregion->key_info,
 						       elusage)))
 			return ERR_PTR(-EINVAL);
-		vchunk->ref_count++;
+		refcount_inc(&vchunk->ref_count);
 		return vchunk;
 	}
 	return mlxsw_sp_acl_tcam_vchunk_create(mlxsw_sp, vgroup,

@@ -1019,7 +1020,7 @@ static void
 mlxsw_sp_acl_tcam_vchunk_put(struct mlxsw_sp *mlxsw_sp,
 			     struct mlxsw_sp_acl_tcam_vchunk *vchunk)
 {
-	if (--vchunk->ref_count)
+	if (!refcount_dec_and_test(&vchunk->ref_count))
 		return;
 	mlxsw_sp_acl_tcam_vchunk_destroy(mlxsw_sp, vchunk);
 }

drivers/net/ethernet/mellanox/mlxsw/spectrum_router.c +8 -7

index ae2fb9efbc509..d15aa6b25a888 100644
@@ -501,7 +501,7 @@ struct mlxsw_sp_rt6 {
 
 struct mlxsw_sp_lpm_tree {
 	u8 id; /* tree ID */
-	unsigned int ref_count;
+	refcount_t ref_count;
 	enum mlxsw_sp_l3proto proto;
 	unsigned long prefix_ref_count[MLXSW_SP_PREFIX_COUNT];
 	struct mlxsw_sp_prefix_usage prefix_usage;

@@ -578,7 +578,7 @@ mlxsw_sp_lpm_tree_find_unused(struct mlxsw_sp *mlxsw_sp)
 
 	for (i = 0; i < mlxsw_sp->router->lpm.tree_count; i++) {
 		lpm_tree = &mlxsw_sp->router->lpm.trees[i];
-		if (lpm_tree->ref_count == 0)
+		if (refcount_read(&lpm_tree->ref_count) == 0)
 			return lpm_tree;
 	}
 	return NULL;

@@ -654,7 +654,7 @@ mlxsw_sp_lpm_tree_create(struct mlxsw_sp *mlxsw_sp,
 	       sizeof(lpm_tree->prefix_usage));
 	memset(&lpm_tree->prefix_ref_count, 0,
 	       sizeof(lpm_tree->prefix_ref_count));
-	lpm_tree->ref_count = 1;
+	refcount_set(&lpm_tree->ref_count, 1);
 	return lpm_tree;
 
 err_left_struct_set:

@@ -678,7 +678,7 @@ mlxsw_sp_lpm_tree_get(struct mlxsw_sp *mlxsw_sp,
 
 	for (i = 0; i < mlxsw_sp->router->lpm.tree_count; i++) {
 		lpm_tree = &mlxsw_sp->router->lpm.trees[i];
-		if (lpm_tree->ref_count != 0 &&
+		if (refcount_read(&lpm_tree->ref_count) &&
 		    lpm_tree->proto == proto &&
 		    mlxsw_sp_prefix_usage_eq(&lpm_tree->prefix_usage,
 					     prefix_usage)) {

@@ -691,14 +691,15 @@ mlxsw_sp_lpm_tree_get(struct mlxsw_sp *mlxsw_sp,
 
 static void mlxsw_sp_lpm_tree_hold(struct mlxsw_sp_lpm_tree *lpm_tree)
 {
-	lpm_tree->ref_count++;
+	refcount_inc(&lpm_tree->ref_count);
 }
 
 static void mlxsw_sp_lpm_tree_put(struct mlxsw_sp *mlxsw_sp,
 				  struct mlxsw_sp_lpm_tree *lpm_tree)
 {
-	if (--lpm_tree->ref_count == 0)
-		mlxsw_sp_lpm_tree_destroy(mlxsw_sp, lpm_tree);
+	if (!refcount_dec_and_test(&lpm_tree->ref_count))
+		return;
+	mlxsw_sp_lpm_tree_destroy(mlxsw_sp, lpm_tree);
 }
 
 #define MLXSW_SP_LPM_TREE_MIN 1 /* tree 0 is reserved */

drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c +4 -4

index 6c749c148148d..6397ff0dc951c 100644
@@ -61,7 +61,7 @@ struct mlxsw_sp_bridge_port {
 	struct mlxsw_sp_bridge_device *bridge_device;
 	struct list_head list;
 	struct list_head vlans_list;
-	unsigned int ref_count;
+	refcount_t ref_count;
 	u8 stp_state;
 	unsigned long flags;
 	bool mrouter;

@@ -495,7 +495,7 @@ mlxsw_sp_bridge_port_create(struct mlxsw_sp_bridge_device *bridge_device,
 			     BR_MCAST_FLOOD;
 	INIT_LIST_HEAD(&bridge_port->vlans_list);
 	list_add(&bridge_port->list, &bridge_device->ports_list);
-	bridge_port->ref_count = 1;
+	refcount_set(&bridge_port->ref_count, 1);
 
 	err = switchdev_bridge_port_offload(brport_dev, mlxsw_sp_port->dev,
 					    NULL, NULL, NULL, false, extack);

@@ -531,7 +531,7 @@ mlxsw_sp_bridge_port_get(struct mlxsw_sp_bridge *bridge,
 
 	bridge_port = mlxsw_sp_bridge_port_find(bridge, brport_dev);
 	if (bridge_port) {
-		bridge_port->ref_count++;
+		refcount_inc(&bridge_port->ref_count);
 		return bridge_port;
 	}

@@ -558,7 +558,7 @@ static void mlxsw_sp_bridge_port_put(struct mlxsw_sp_bridge *bridge,
 {
 	struct mlxsw_sp_bridge_device *bridge_device;
 
-	if (--bridge_port->ref_count != 0)
+	if (!refcount_dec_and_test(&bridge_port->ref_count))
 		return;
 	bridge_device = bridge_port->bridge_device;
 	mlxsw_sp_bridge_port_destroy(bridge_port);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:12 UTC #80 lore raw

PATCH 75/186 mlxsw: spectrum_acl_tcam: Fix race in region ID allocation

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ido Schimmel <[email protected]>

[ Upstream commit 627f9c1bb882765a84aa78015abbacd783d429be ]

Region identifiers can be allocated both when user space tries to insert a new tc filter and when filters are migrated from one region to another as part of the rehash delayed work.

There is no lock protecting the bitmap from which these identifiers are allocated from, which is racy and leads to bad parameter errors from the device's firmware.

Fix by converting the bitmap to IDA which handles its own locking. For consistency, do the same for the group identifiers that are part of the same structure.

Fixes 2bffc5322fd8 ("mlxsw: spectrum_acl: Don't take mutex in mlxsw_sp_acl_tcam_vregion_rehash_work()")

Reported-by Amit Cohen <[email protected]>

Link https://lore.kernel.org/r/ce494b7940cadfe84f3e18da7785b51ef5f776e3.1713797103.git.petrm@nvidia.com open here

Signed-off-by Sasha Levin <[email protected]>

---
 .../mellanox/mlxsw/spectrum_acl_tcam.c        | 61 ++++++++-----------
 .../mellanox/mlxsw/spectrum_acl_tcam.h        |  5 +-
 2 files changed, 30 insertions(+), 36 deletions(-)

drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c +27 ~~-34~~

index f20052776b3f2..b6a4652a6475a 100644
@@ -10,6 +10,7 @@
 #include <linux/netdevice.h>
 #include <linux/mutex.h>
 #include <linux/refcount.h>
+#include <linux/idr.h>
 #include <net/devlink.h>
 #include <trace/events/mlxsw.h>

@@ -58,41 +59,43 @@ int mlxsw_sp_acl_tcam_priority_get(struct mlxsw_sp *mlxsw_sp,
 static int mlxsw_sp_acl_tcam_region_id_get(struct mlxsw_sp_acl_tcam *tcam,
 					   u16 *p_id)
 {
-	u16 id;
+	int id;
 
-	id = find_first_zero_bit(tcam->used_regions, tcam->max_regions);
-	if (id < tcam->max_regions) {
-		__set_bit(id, tcam->used_regions);
-		*p_id = id;
-		return 0;
-	}
-	return -ENOBUFS;
+	id = ida_alloc_max(&tcam->used_regions, tcam->max_regions - 1,
+			   GFP_KERNEL);
+	if (id < 0)
+		return id;
+
+	*p_id = id;
+
+	return 0;
 }
 
 static void mlxsw_sp_acl_tcam_region_id_put(struct mlxsw_sp_acl_tcam *tcam,
 					    u16 id)
 {
-	__clear_bit(id, tcam->used_regions);
+	ida_free(&tcam->used_regions, id);
 }
 
 static int mlxsw_sp_acl_tcam_group_id_get(struct mlxsw_sp_acl_tcam *tcam,
 					  u16 *p_id)
 {
-	u16 id;
+	int id;
 
-	id = find_first_zero_bit(tcam->used_groups, tcam->max_groups);
-	if (id < tcam->max_groups) {
-		__set_bit(id, tcam->used_groups);
-		*p_id = id;
-		return 0;
-	}
-	return -ENOBUFS;
+	id = ida_alloc_max(&tcam->used_groups, tcam->max_groups - 1,
+			   GFP_KERNEL);
+	if (id < 0)
+		return id;
+
+	*p_id = id;
+
+	return 0;
 }
 
 static void mlxsw_sp_acl_tcam_group_id_put(struct mlxsw_sp_acl_tcam *tcam,
 					   u16 id)
 {
-	__clear_bit(id, tcam->used_groups);
+	ida_free(&tcam->used_groups, id);
 }
 
 struct mlxsw_sp_acl_tcam_pattern {

@@ -1549,19 +1552,11 @@ int mlxsw_sp_acl_tcam_init(struct mlxsw_sp *mlxsw_sp,
 	if (max_tcam_regions < max_regions)
 		max_regions = max_tcam_regions;
 
-	tcam->used_regions = bitmap_zalloc(max_regions, GFP_KERNEL);
-	if (!tcam->used_regions) {
-		err = -ENOMEM;
-		goto err_alloc_used_regions;
-	}
+	ida_init(&tcam->used_regions);
 	tcam->max_regions = max_regions;
 
 	max_groups = MLXSW_CORE_RES_GET(mlxsw_sp->core, ACL_MAX_GROUPS);
-	tcam->used_groups = bitmap_zalloc(max_groups, GFP_KERNEL);
-	if (!tcam->used_groups) {
-		err = -ENOMEM;
-		goto err_alloc_used_groups;
-	}
+	ida_init(&tcam->used_groups);
 	tcam->max_groups = max_groups;
 	tcam->max_group_size = MLXSW_CORE_RES_GET(mlxsw_sp->core,
 						  ACL_MAX_GROUP_SIZE);

@@ -1575,10 +1570,8 @@ int mlxsw_sp_acl_tcam_init(struct mlxsw_sp *mlxsw_sp,
 	return 0;
 
 err_tcam_init:
-	bitmap_free(tcam->used_groups);
-err_alloc_used_groups:
-	bitmap_free(tcam->used_regions);
-err_alloc_used_regions:
+	ida_destroy(&tcam->used_groups);
+	ida_destroy(&tcam->used_regions);
 	mlxsw_sp_acl_tcam_rehash_params_unregister(mlxsw_sp);
 err_rehash_params_register:
 	mutex_destroy(&tcam->lock);

@@ -1591,8 +1584,8 @@ void mlxsw_sp_acl_tcam_fini(struct mlxsw_sp *mlxsw_sp,
 	const struct mlxsw_sp_acl_tcam_ops *ops = mlxsw_sp->acl_tcam_ops;
 
 	ops->fini(mlxsw_sp, tcam->priv);
-	bitmap_free(tcam->used_groups);
-	bitmap_free(tcam->used_regions);
+	ida_destroy(&tcam->used_groups);
+	ida_destroy(&tcam->used_regions);
 	mlxsw_sp_acl_tcam_rehash_params_unregister(mlxsw_sp);
 	mutex_destroy(&tcam->lock);
 }

drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.h +3 -2

index 462bf448497d3..79a1d86065125 100644
@@ -6,15 +6,16 @@
 
 #include <linux/list.h>
 #include <linux/parman.h>
+#include <linux/idr.h>
 
 #include "reg.h"
 #include "spectrum.h"
 #include "core_acl_flex_keys.h"
 
 struct mlxsw_sp_acl_tcam {
-	unsigned long *used_regions; /* bit array */
+	struct ida used_regions;
 	unsigned int max_regions;
-	unsigned long *used_groups;  /* bit array */
+	struct ida used_groups;
 	unsigned int max_groups;
 	unsigned int max_group_size;
 	struct mutex lock; /* guards vregion list */

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:10 UTC #81 lore raw

PATCH 76/186 mlxsw: spectrum_acl_tcam: Fix race during rehash delayed work

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ido Schimmel <[email protected]>

[ Upstream commit d90cfe20562407d9f080d24123078d666d730707 ]

The purpose of the rehash delayed work is to reduce the number of masks (eRPs) used by an ACL region as the eRP bank is a global and limited resource.

This is done in three steps:

1. Creating a new set of masks and a new ACL region which will use the new masks and to which the existing filters will be migrated to. The new region is assigned to 'vregion->region' and the region from which the filters are migrated from is assigned to 'vregion->region2'.

2. Migrating all the filters from the old region to the new region.

3. Destroying the old region and setting 'vregion->region2' to NULL.

Only the second steps is performed under the 'vregion->lock' mutex although its comments says that among other things it "Protects consistency of region, region2 pointers".

This is problematic as the first step can race with filter insertion from user space that uses 'vregion->region', but under the mutex.

Fix by holding the mutex across the entirety of the delayed work and not only during the second step.

Fixes 2bffc5322fd8 ("mlxsw: spectrum_acl: Don't take mutex in mlxsw_sp_acl_tcam_vregion_rehash_work()")

Link https://lore.kernel.org/r/1ec1d54edf2bad0a369e6b4fa030aba64e1f124b.1713797103.git.petrm@nvidia.com open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c +2 -2

index b6a4652a6475a..9c0c728bb42dc 100644
@@ -718,7 +718,9 @@ static void mlxsw_sp_acl_tcam_vregion_rehash_work(struct work_struct *work)
 			     rehash.dw.work);
 	int credits = MLXSW_SP_ACL_TCAM_VREGION_REHASH_CREDITS;
 
+	mutex_lock(&vregion->lock);
 	mlxsw_sp_acl_tcam_vregion_rehash(vregion->mlxsw_sp, vregion, &credits);
+	mutex_unlock(&vregion->lock);
 	if (credits < 0)
 		/* Rehash gone out of credits so it was interrupted.
 		 * Schedule the work as soon as possible to continue.

@@ -1323,7 +1325,6 @@ mlxsw_sp_acl_tcam_vregion_migrate(struct mlxsw_sp *mlxsw_sp,
 	int err, err2;
 
 	trace_mlxsw_sp_acl_tcam_vregion_migrate(mlxsw_sp, vregion);
-	mutex_lock(&vregion->lock);
 	err = mlxsw_sp_acl_tcam_vchunk_migrate_all(mlxsw_sp, vregion,
 						   ctx, credits);
 	if (err) {

@@ -1343,7 +1344,6 @@ mlxsw_sp_acl_tcam_vregion_migrate(struct mlxsw_sp *mlxsw_sp,
 			/* Let the rollback to be continued later on. */
 		}
 	}
-	mutex_unlock(&vregion->lock);
 	trace_mlxsw_sp_acl_tcam_vregion_migrate_end(mlxsw_sp, vregion);
 	return err;
 }

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:10 UTC #82 lore raw

PATCH 77/186 mlxsw: spectrum_acl_tcam: Fix possible use-after-free during activity update

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ido Schimmel <[email protected]>

[ Upstream commit 79b5b4b18bc85b19d3a518483f9abbbe6d7b3ba4 ]

The rule activity update delayed work periodically traverses the list of configured rules and queries their activity from the device.

As part of this task it accesses the entry pointed by 'ventry->entry', but this entry can be changed concurrently by the rehash delayed work, leading to a use-after-free [1].

Fix by closing the race and perform the activity query under the 'vregion->lock' mutex.

[1]

BUG: KASAN: slab-use-after-free in mlxsw_sp_acl_tcam_flower_rule_activity_get+0x121/0x140
Read of size 8 at addr ffff8881054ed808 by task kworker/0:18/181

CPU: 0 PID: 181 Comm: kworker/0:18 Not tainted 6.9.0-rc2-custom-00781-gd5ab772d32f7 #2
Hardware name: Mellanox Technologies Ltd. MSN3700/VMOD0005, BIOS 5.11 01/06/2019
Workqueue: mlxsw_core mlxsw_sp_acl_rule_activity_update_work
Call Trace:
 <TASK>
 dump_stack_lvl+0xc6/0x120
 print_report+0xce/0x670
 kasan_report+0xd7/0x110
 mlxsw_sp_acl_tcam_flower_rule_activity_get+0x121/0x140
 mlxsw_sp_acl_rule_activity_update_work+0x219/0x400
 process_one_work+0x8eb/0x19b0
 worker_thread+0x6c9/0xf70
 kthread+0x2c9/0x3b0
 ret_from_fork+0x4d/0x80
 ret_from_fork_asm+0x1a/0x30
 </TASK>

Allocated by task 1039:
 kasan_save_stack+0x33/0x60
 kasan_save_track+0x14/0x30
 __kasan_kmalloc+0x8f/0xa0
 __kmalloc+0x19c/0x360
 mlxsw_sp_acl_tcam_entry_create+0x7b/0x1f0
 mlxsw_sp_acl_tcam_vchunk_migrate_all+0x30d/0xb50
 mlxsw_sp_acl_tcam_vregion_rehash_work+0x157/0x1300
 process_one_work+0x8eb/0x19b0
 worker_thread+0x6c9/0xf70
 kthread+0x2c9/0x3b0
 ret_from_fork+0x4d/0x80
 ret_from_fork_asm+0x1a/0x30

Freed by task 1039:
 kasan_save_stack+0x33/0x60
 kasan_save_track+0x14/0x30
 kasan_save_free_info+0x3b/0x60
 poison_slab_object+0x102/0x170
 __kasan_slab_free+0x14/0x30
 kfree+0xc1/0x290
 mlxsw_sp_acl_tcam_vchunk_migrate_all+0x3d7/0xb50
 mlxsw_sp_acl_tcam_vregion_rehash_work+0x157/0x1300
 process_one_work+0x8eb/0x19b0
 worker_thread+0x6c9/0xf70
 kthread+0x2c9/0x3b0
 ret_from_fork+0x4d/0x80
 ret_from_fork_asm+0x1a/0x30

Fixes 2bffc5322fd8 ("mlxsw: spectrum_acl: Don't take mutex in mlxsw_sp_acl_tcam_vregion_rehash_work()")

Link https://lore.kernel.org/r/1fcce0a60b231ebeb2515d91022284ba7b4ffe7a.1713797103.git.petrm@nvidia.com open here

Signed-off-by Sasha Levin <[email protected]>

---
 .../net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c    | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c +8 -2

index 9c0c728bb42dc..7e69225c057de 100644
@@ -1159,8 +1159,14 @@ mlxsw_sp_acl_tcam_ventry_activity_get(struct mlxsw_sp *mlxsw_sp,
 				      struct mlxsw_sp_acl_tcam_ventry *ventry,
 				      bool *activity)
 {
-	return mlxsw_sp_acl_tcam_entry_activity_get(mlxsw_sp,
-						    ventry->entry, activity);
+	struct mlxsw_sp_acl_tcam_vregion *vregion = ventry->vchunk->vregion;
+	int err;
+
+	mutex_lock(&vregion->lock);
+	err = mlxsw_sp_acl_tcam_entry_activity_get(mlxsw_sp, ventry->entry,
+						   activity);
+	mutex_unlock(&vregion->lock);
+	return err;
 }
 
 static int

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:11 UTC #83 lore raw

PATCH 78/186 mlxsw: spectrum_acl_tcam: Fix possible use-after-free during rehash

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ido Schimmel <[email protected]>

[ Upstream commit 54225988889931467a9b55fdbef534079b665519 ]

The rehash delayed work migrates filters from one region to another according to the number of available credits.

The migrated from region is destroyed at the end of the work if the number of credits is non-negative as the assumption is that this is indicative of migration being complete. This assumption is incorrect as a non-negative number of credits can also be the result of a failed migration.

The destruction of a region that still has filters referencing it can result in a use-after-free [1].

Fix by not destroying the region if migration failed.

[1]

BUG: KASAN: slab-use-after-free in mlxsw_sp_acl_ctcam_region_entry_remove+0x21d/0x230
Read of size 8 at addr ffff8881735319e8 by task kworker/0:31/3858

CPU: 0 PID: 3858 Comm: kworker/0:31 Tainted: G        W          6.9.0-rc2-custom-00782-gf2275c2157d8 #5
Hardware name: Mellanox Technologies Ltd. MSN3700/VMOD0005, BIOS 5.11 01/06/2019
Workqueue: mlxsw_core mlxsw_sp_acl_tcam_vregion_rehash_work
Call Trace:
 <TASK>
 dump_stack_lvl+0xc6/0x120
 print_report+0xce/0x670
 kasan_report+0xd7/0x110
 mlxsw_sp_acl_ctcam_region_entry_remove+0x21d/0x230
 mlxsw_sp_acl_ctcam_entry_del+0x2e/0x70
 mlxsw_sp_acl_atcam_entry_del+0x81/0x210
 mlxsw_sp_acl_tcam_vchunk_migrate_all+0x3cd/0xb50
 mlxsw_sp_acl_tcam_vregion_rehash_work+0x157/0x1300
 process_one_work+0x8eb/0x19b0
 worker_thread+0x6c9/0xf70
 kthread+0x2c9/0x3b0
 ret_from_fork+0x4d/0x80
 ret_from_fork_asm+0x1a/0x30
 </TASK>

Allocated by task 174:
 kasan_save_stack+0x33/0x60
 kasan_save_track+0x14/0x30
 __kasan_kmalloc+0x8f/0xa0
 __kmalloc+0x19c/0x360
 mlxsw_sp_acl_tcam_region_create+0xdf/0x9c0
 mlxsw_sp_acl_tcam_vregion_rehash_work+0x954/0x1300
 process_one_work+0x8eb/0x19b0
 worker_thread+0x6c9/0xf70
 kthread+0x2c9/0x3b0
 ret_from_fork+0x4d/0x80
 ret_from_fork_asm+0x1a/0x30

Freed by task 7:
 kasan_save_stack+0x33/0x60
 kasan_save_track+0x14/0x30
 kasan_save_free_info+0x3b/0x60
 poison_slab_object+0x102/0x170
 __kasan_slab_free+0x14/0x30
 kfree+0xc1/0x290
 mlxsw_sp_acl_tcam_region_destroy+0x272/0x310
 mlxsw_sp_acl_tcam_vregion_rehash_work+0x731/0x1300
 process_one_work+0x8eb/0x19b0
 worker_thread+0x6c9/0xf70
 kthread+0x2c9/0x3b0
 ret_from_fork+0x4d/0x80
 ret_from_fork_asm+0x1a/0x30

Fixes c9c9af91f1d9 ("mlxsw: spectrum_acl: Allow to interrupt/continue rehash work")

Link https://lore.kernel.org/r/3e412b5659ec2310c5c615760dfe5eac18dd7ebd.1713797103.git.petrm@nvidia.com open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c | 1 +
 1 file changed, 1 insertion(+)

drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c +1 -0

index 7e69225c057de..1ff0b2c7c11de 100644
@@ -1451,6 +1451,7 @@ mlxsw_sp_acl_tcam_vregion_rehash(struct mlxsw_sp *mlxsw_sp,
 						ctx, credits);
 	if (err) {
 		dev_err(mlxsw_sp->bus_info->dev, "Failed to migrate vregion\n");
+		return;
 	}
 
 	if (*credits >= 0)

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:11 UTC #84 lore raw

PATCH 79/186 mlxsw: spectrum_acl_tcam: Rate limit error message

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ido Schimmel <[email protected]>

[ Upstream commit 5bcf925587e9b5d36420d572a0b4d131c90fb306 ]

In the rare cases when the device resources are exhausted it is likely that the rehash delayed work will fail. An error message will be printed whenever this happens which can be overwhelming considering the fact that the work is per-region and that there can be hundreds of regions.

Fix by rate limiting the error message.

Fixes e5e7962ee5c2 ("mlxsw: spectrum_acl: Implement region migration according to hints")

Link https://lore.kernel.org/r/c510763b2ebd25e7990d80183feff91cde593145.1713797103.git.petrm@nvidia.com open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c +1 -1

index 1ff0b2c7c11de..568ae7092fe0e 100644
@@ -1450,7 +1450,7 @@ mlxsw_sp_acl_tcam_vregion_rehash(struct mlxsw_sp *mlxsw_sp,
 	err = mlxsw_sp_acl_tcam_vregion_migrate(mlxsw_sp, vregion,
 						ctx, credits);
 	if (err) {
-		dev_err(mlxsw_sp->bus_info->dev, "Failed to migrate vregion\n");
+		dev_err_ratelimited(mlxsw_sp->bus_info->dev, "Failed to migrate vregion\n");
 		return;
 	}

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:11 UTC #85 lore raw

PATCH 80/186 mlxsw: spectrum_acl_tcam: Fix memory leak during rehash

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ido Schimmel <[email protected]>

[ Upstream commit 8ca3f7a7b61393804c46f170743c3b839df13977 ]

The rehash delayed work migrates filters from one region to another. This is done by iterating over all chunks (all the filters with the same priority) in the region and in each chunk iterating over all the filters.

If the migration fails, the code tries to migrate the filters back to the old region. However, the rollback itself can also fail in which case another migration will be erroneously performed. Besides the fact that this ping pong is not a very good idea, it also creates a problem.

Each virtual chunk references two chunks: The currently used one ('vchunk->chunk') and a backup ('vchunk->chunk2'). During migration the first holds the chunk we want to migrate filters to and the second holds the chunk we are migrating filters from.

The code currently assumes - but does not verify - that the backup chunk does not exist (NULL) if the currently used chunk does not reference the target region. This assumption breaks when we are trying to rollback a rollback, resulting in the backup chunk being overwritten and leaked [1].

Fix by not rolling back a failed rollback and add a warning to avoid future cases.

[1]

WARNING: CPU: 5 PID: 1063 at lib/parman.c:291 parman_destroy+0x17/0x20
Modules linked in:
CPU: 5 PID: 1063 Comm: kworker/5:11 Tainted: G        W          6.9.0-rc2-custom-00784-gc6a05c468a0b #14
Hardware name: Mellanox Technologies Ltd. MSN3700/VMOD0005, BIOS 5.11 01/06/2019
Workqueue: mlxsw_core mlxsw_sp_acl_tcam_vregion_rehash_work
RIP: 0010:parman_destroy+0x17/0x20
[...]
Call Trace:
 <TASK>
 mlxsw_sp_acl_atcam_region_fini+0x19/0x60
 mlxsw_sp_acl_tcam_region_destroy+0x49/0xf0
 mlxsw_sp_acl_tcam_vregion_rehash_work+0x1f1/0x470
 process_one_work+0x151/0x370
 worker_thread+0x2cb/0x3e0
 kthread+0xd0/0x100
 ret_from_fork+0x34/0x50
 ret_from_fork_asm+0x1a/0x30
 </TASK>

Fixes 843500518509 ("mlxsw: spectrum_acl: Do rollback as another call to mlxsw_sp_acl_tcam_vchunk_migrate_all()")

Link https://lore.kernel.org/r/d5edd4f4503934186ae5cfe268503b16345b4e0f.1713797103.git.petrm@nvidia.com open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c | 4 ++++
 1 file changed, 4 insertions(+)

drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c +4 -0

index 568ae7092fe0e..0902eb7651e14 100644
@@ -1200,6 +1200,8 @@ mlxsw_sp_acl_tcam_vchunk_migrate_start(struct mlxsw_sp *mlxsw_sp,
 {
 	struct mlxsw_sp_acl_tcam_chunk *new_chunk;
 
+	WARN_ON(vchunk->chunk2);
+
 	new_chunk = mlxsw_sp_acl_tcam_chunk_create(mlxsw_sp, vchunk, region);
 	if (IS_ERR(new_chunk))
 		return PTR_ERR(new_chunk);

@@ -1334,6 +1336,8 @@ mlxsw_sp_acl_tcam_vregion_migrate(struct mlxsw_sp *mlxsw_sp,
 	err = mlxsw_sp_acl_tcam_vchunk_migrate_all(mlxsw_sp, vregion,
 						   ctx, credits);
 	if (err) {
+		if (ctx->this_is_rollback)
+			return err;
 		/* In case migration was not successful, we need to swap
 		 * so the original region pointer is assigned again
 		 * to vregion->region.

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:11 UTC #86 lore raw

PATCH 81/186 mlxsw: spectrum_acl_tcam: Fix warning during rehash

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ido Schimmel <[email protected]>

[ Upstream commit 743edc8547a92b6192aa1f1b6bb78233fa21dc9b ]

As previously explained, the rehash delayed work migrates filters from one region to another. This is done by iterating over all chunks (all the filters with the same priority) in the region and in each chunk iterating over all the filters.

When the work runs out of credits it stores the current chunk and entry as markers in the per-work context so that it would know where to resume the migration from the next time the work is scheduled.

Upon error, the chunk marker is reset to NULL, but without resetting the entry markers despite being relative to it. This can result in migration being resumed from an entry that does not belong to the chunk being migrated. In turn, this will eventually lead to a chunk being iterated over as if it is an entry. Because of how the two structures happen to be defined, this does not lead to KASAN splats, but to warnings such as [1].

Fix by creating a helper that resets all the markers and call it from all the places the currently only reset the chunk marker. For good measures also call it when starting a completely new rehash. Add a warning to avoid future cases.

[1]

WARNING: CPU: 7 PID: 1076 at drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_keys.c:407 mlxsw_afk_encode+0x242/0x2f0
Modules linked in:
CPU: 7 PID: 1076 Comm: kworker/7:24 Tainted: G        W          6.9.0-rc3-custom-00880-g29e61d91b77b #29
Hardware name: Mellanox Technologies Ltd. MSN3700/VMOD0005, BIOS 5.11 01/06/2019
Workqueue: mlxsw_core mlxsw_sp_acl_tcam_vregion_rehash_work
RIP: 0010:mlxsw_afk_encode+0x242/0x2f0
[...]
Call Trace:
 <TASK>
 mlxsw_sp_acl_atcam_entry_add+0xd9/0x3c0
 mlxsw_sp_acl_tcam_entry_create+0x5e/0xa0
 mlxsw_sp_acl_tcam_vchunk_migrate_all+0x109/0x290
 mlxsw_sp_acl_tcam_vregion_rehash_work+0x6c/0x470
 process_one_work+0x151/0x370
 worker_thread+0x2cb/0x3e0
 kthread+0xd0/0x100
 ret_from_fork+0x34/0x50
 </TASK>

Fixes 6f9579d4e302 ("mlxsw: spectrum_acl: Remember where to continue rehash migration")

Link https://lore.kernel.org/r/cc17eed86b41dd829d39b07906fec074a9ce580e.1713797103.git.petrm@nvidia.com open here

Signed-off-by Sasha Levin <[email protected]>

---
 .../mellanox/mlxsw/spectrum_acl_tcam.c        | 20 ++++++++++++++++---
 1 file changed, 17 insertions(+), 3 deletions(-)

drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c +17 -3

index 0902eb7651e14..e8c6078866213 100644
@@ -730,6 +730,17 @@ static void mlxsw_sp_acl_tcam_vregion_rehash_work(struct work_struct *work)
 		mlxsw_sp_acl_tcam_vregion_rehash_work_schedule(vregion);
 }
 
+static void
+mlxsw_sp_acl_tcam_rehash_ctx_vchunk_reset(struct mlxsw_sp_acl_tcam_rehash_ctx *ctx)
+{
+	/* The entry markers are relative to the current chunk and therefore
+	 * needs to be reset together with the chunk marker.
+	 */
+	ctx->current_vchunk = NULL;
+	ctx->start_ventry = NULL;
+	ctx->stop_ventry = NULL;
+}
+
 static void
 mlxsw_sp_acl_tcam_rehash_ctx_vchunk_changed(struct mlxsw_sp_acl_tcam_vchunk *vchunk)
 {

@@ -752,7 +763,7 @@ mlxsw_sp_acl_tcam_rehash_ctx_vregion_changed(struct mlxsw_sp_acl_tcam_vregion *v
 	 * the current chunk pointer to make sure all chunks
 	 * are properly migrated.
 	 */
-	vregion->rehash.ctx.current_vchunk = NULL;
+	mlxsw_sp_acl_tcam_rehash_ctx_vchunk_reset(&vregion->rehash.ctx);
 }
 
 static struct mlxsw_sp_acl_tcam_vregion *

@@ -1220,7 +1231,7 @@ mlxsw_sp_acl_tcam_vchunk_migrate_end(struct mlxsw_sp *mlxsw_sp,
 {
 	mlxsw_sp_acl_tcam_chunk_destroy(mlxsw_sp, vchunk->chunk2);
 	vchunk->chunk2 = NULL;
-	ctx->current_vchunk = NULL;
+	mlxsw_sp_acl_tcam_rehash_ctx_vchunk_reset(ctx);
 }
 
 static int

@@ -1252,6 +1263,8 @@ mlxsw_sp_acl_tcam_vchunk_migrate_one(struct mlxsw_sp *mlxsw_sp,
 		ventry = list_first_entry(&vchunk->ventry_list,
 					  typeof(*ventry), list);
 
+	WARN_ON(ventry->vchunk != vchunk);
+
 	list_for_each_entry_from(ventry, &vchunk->ventry_list, list) {
 		/* During rollback, once we reach the ventry that failed
 		 * to migrate, we are done.

@@ -1343,7 +1356,7 @@ mlxsw_sp_acl_tcam_vregion_migrate(struct mlxsw_sp *mlxsw_sp,
 		 * to vregion->region.
 		 */
 		swap(vregion->region, vregion->region2);
-		ctx->current_vchunk = NULL;
+		mlxsw_sp_acl_tcam_rehash_ctx_vchunk_reset(ctx);
 		ctx->this_is_rollback = true;
 		err2 = mlxsw_sp_acl_tcam_vchunk_migrate_all(mlxsw_sp, vregion,
 							    ctx, credits);

@@ -1402,6 +1415,7 @@ mlxsw_sp_acl_tcam_vregion_rehash_start(struct mlxsw_sp *mlxsw_sp,
 
 	ctx->hints_priv = hints_priv;
 	ctx->this_is_rollback = false;
+	mlxsw_sp_acl_tcam_rehash_ctx_vchunk_reset(ctx);
 
 	return 0;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:11 UTC #87 lore raw

PATCH 82/186 mlxsw: spectrum_acl_tcam: Fix incorrect list API usage

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ido Schimmel <[email protected]>

[ Upstream commit b377add0f0117409c418ddd6504bd682ebe0bf79 ]

Both the function that migrates all the chunks within a region and the function that migrates all the entries within a chunk call list_first_entry() on the respective lists without checking that the lists are not empty. This is incorrect usage of the API, which leads to the following warning [1].

Fix by returning if the lists are empty as there is nothing to migrate in this case.

[1]

WARNING: CPU: 0 PID: 6437 at drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c:1266 mlxsw_sp_acl_tcam_vchunk_migrate_all+0x1f1/0>
Modules linked in:
CPU: 0 PID: 6437 Comm: kworker/0:37 Not tainted 6.9.0-rc3-custom-00883-g94a65f079ef6 #39
Hardware name: Mellanox Technologies Ltd. MSN3700/VMOD0005, BIOS 5.11 01/06/2019
Workqueue: mlxsw_core mlxsw_sp_acl_tcam_vregion_rehash_work
RIP: 0010:mlxsw_sp_acl_tcam_vchunk_migrate_all+0x1f1/0x2c0
[...]
Call Trace:
 <TASK>
 mlxsw_sp_acl_tcam_vregion_rehash_work+0x6c/0x4a0
 process_one_work+0x151/0x370
 worker_thread+0x2cb/0x3e0
 kthread+0xd0/0x100
 ret_from_fork+0x34/0x50
 ret_from_fork_asm+0x1a/0x30
 </TASK>

Fixes 6f9579d4e302 ("mlxsw: spectrum_acl: Remember where to continue rehash migration")

Link https://lore.kernel.org/r/4628e9a22d1d84818e28310abbbc498e7bc31bc9.1713797103.git.petrm@nvidia.com open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c | 7 +++++++
 1 file changed, 7 insertions(+)

drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c +7 -0

index e8c6078866213..89a5ebc3463ff 100644
@@ -1254,6 +1254,9 @@ mlxsw_sp_acl_tcam_vchunk_migrate_one(struct mlxsw_sp *mlxsw_sp,
 		return 0;
 	}
 
+	if (list_empty(&vchunk->ventry_list))
+		goto out;
+
 	/* If the migration got interrupted, we have the ventry to start from
 	 * stored in context.
 	 */

@@ -1305,6 +1308,7 @@ mlxsw_sp_acl_tcam_vchunk_migrate_one(struct mlxsw_sp *mlxsw_sp,
 		}
 	}
 
+out:
 	mlxsw_sp_acl_tcam_vchunk_migrate_end(mlxsw_sp, vchunk, ctx);
 	return 0;
 }

@@ -1318,6 +1322,9 @@ mlxsw_sp_acl_tcam_vchunk_migrate_all(struct mlxsw_sp *mlxsw_sp,
 	struct mlxsw_sp_acl_tcam_vchunk *vchunk;
 	int err;
 
+	if (list_empty(&vregion->vchunk_list))
+		return 0;
+
 	/* If the migration got interrupted, we have the vchunk
 	 * we are working on stored in context.
 	 */

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:12 UTC #88 lore raw

PATCH 83/186 mlxsw: spectrum_acl_tcam: Fix memory leak when canceling rehash work

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ido Schimmel <[email protected]>

[ Upstream commit fb4e2b70a7194b209fc7320bbf33b375f7114bd5 ]

The rehash delayed work is rescheduled with a delay if the number of credits at end of the work is not negative as supposedly it means that the migration ended. Otherwise, it is rescheduled immediately.

After "mlxsw: spectrum_acl_tcam: Fix possible use-after-free during rehash" the above is no longer accurate as a non-negative number of credits is no longer indicative of the migration being done. It can also happen if the work encountered an error in which case the migration will resume the next time the work is scheduled.

The significance of the above is that it is possible for the work to be pending and associated with hints that were allocated when the migration started. This leads to the hints being leaked [1] when the work is canceled while pending as part of ACL region dismantle.

Fix by freeing the hints if hints are associated with a work that was canceled while pending.

Blame the original commit since the reliance on not having a pending work associated with hints is fragile.

[1]
unreferenced object 0xffff88810e7c3000 (size 256):
comm "kworker/0:16", pid 176, jiffies 4295460353 hex dump (first 32 bytes):

    00 30 95 11 81 88 ff ff 61 00 00 00 00 00 00 80  .0......a.......
    00 00 61 00 40 00 00 00 00 00 00 00 04 00 00 00  ..a.@...........
  backtrace (crc 2544ddb9):
    [<00000000cf8cfab3>] kmalloc_trace+0x23f/0x2a0
    [<000000004d9a1ad9>] objagg_hints_get+0x42/0x390
    [<000000000b143cf3>] mlxsw_sp_acl_erp_rehash_hints_get+0xca/0x400
    [<0000000059bdb60a>] mlxsw_sp_acl_tcam_vregion_rehash_work+0x868/0x1160
    [<00000000e81fd734>] process_one_work+0x59c/0xf20
    [<00000000ceee9e81>] worker_thread+0x799/0x12c0
    [<00000000bda6fe39>] kthread+0x246/0x300
    [<0000000070056d23>] ret_from_fork+0x34/0x70
    [<00000000dea2b93e>] ret_from_fork_asm+0x1a/0x30

Fixes c9c9af91f1d9 ("mlxsw: spectrum_acl: Allow to interrupt/continue rehash work")

Link https://lore.kernel.org/r/0cc12ebb07c4d4c41a1265ee2c28b392ff997a86.1713797103.git.petrm@nvidia.com open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

drivers/net/ethernet/mellanox/mlxsw/spectrum_acl_tcam.c +5 -1

index 89a5ebc3463ff..92a406f02eae7 100644
@@ -836,10 +836,14 @@ mlxsw_sp_acl_tcam_vregion_destroy(struct mlxsw_sp *mlxsw_sp,
 	struct mlxsw_sp_acl_tcam *tcam = vregion->tcam;
 
 	if (vgroup->vregion_rehash_enabled && ops->region_rehash_hints_get) {
+		struct mlxsw_sp_acl_tcam_rehash_ctx *ctx = &vregion->rehash.ctx;
+
 		mutex_lock(&tcam->lock);
 		list_del(&vregion->tlist);
 		mutex_unlock(&tcam->lock);
-		cancel_delayed_work_sync(&vregion->rehash.dw);
+		if (cancel_delayed_work_sync(&vregion->rehash.dw) &&
+		    ctx->hints_priv)
+			ops->region_rehash_hints_put(ctx->hints_priv);
 	}
 	mlxsw_sp_acl_tcam_vgroup_vregion_detach(mlxsw_sp, vregion);
 	if (vregion->region2)

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:12 UTC #89 lore raw

PATCH 84/186 eth: bnxt: fix counting packets discarded due to OOM and netpoll

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jakub Kicinski <[email protected]>

[ Upstream commit 730117730709992c9f6535dd7b47638ee561ec45 ]

I added OOM and netpoll discard counters, naively assuming that the cpr pointer is pointing to a common completion ring. Turns out that is usually *a* completion ring but not *the* completion ring which bnapi->cp_ring points to. bnapi->cp_ring is where the stats are read from, so we end up reporting 0 thru ethtool -S and qstat even though the drop events have happened. Make 100% sure we're recording statistics in the correct structure.

Fixes 907fd4a294db ("bnxt: count discards due to memory allocation errors")

Reviewed-by Michael Chan <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c | 44 ++++++++++-------------
 1 file changed, 18 insertions(+), 26 deletions(-)

drivers/net/ethernet/broadcom/bnxt/bnxt.c +18 ~~-26~~

index d40b91719b79b..724624737d095 100644
@@ -1659,7 +1659,7 @@ static inline struct sk_buff *bnxt_tpa_end(struct bnxt *bp,
 		skb = bnxt_copy_skb(bnapi, data_ptr, len, mapping);
 		if (!skb) {
 			bnxt_abort_tpa(cpr, idx, agg_bufs);
-			cpr->sw_stats.rx.rx_oom_discards += 1;
+			cpr->bnapi->cp_ring.sw_stats.rx.rx_oom_discards += 1;
 			return NULL;
 		}
 	} else {

@@ -1669,7 +1669,7 @@ static inline struct sk_buff *bnxt_tpa_end(struct bnxt *bp,
 		new_data = __bnxt_alloc_rx_frag(bp, &new_mapping, GFP_ATOMIC);
 		if (!new_data) {
 			bnxt_abort_tpa(cpr, idx, agg_bufs);
-			cpr->sw_stats.rx.rx_oom_discards += 1;
+			cpr->bnapi->cp_ring.sw_stats.rx.rx_oom_discards += 1;
 			return NULL;
 		}

@@ -1685,7 +1685,7 @@ static inline struct sk_buff *bnxt_tpa_end(struct bnxt *bp,
 		if (!skb) {
 			skb_free_frag(data);
 			bnxt_abort_tpa(cpr, idx, agg_bufs);
-			cpr->sw_stats.rx.rx_oom_discards += 1;
+			cpr->bnapi->cp_ring.sw_stats.rx.rx_oom_discards += 1;
 			return NULL;
 		}
 		skb_reserve(skb, bp->rx_offset);

@@ -1696,7 +1696,7 @@ static inline struct sk_buff *bnxt_tpa_end(struct bnxt *bp,
 		skb = bnxt_rx_agg_pages_skb(bp, cpr, skb, idx, agg_bufs, true);
 		if (!skb) {
 			/* Page reuse already handled by bnxt_rx_pages(). */
-			cpr->sw_stats.rx.rx_oom_discards += 1;
+			cpr->bnapi->cp_ring.sw_stats.rx.rx_oom_discards += 1;
 			return NULL;
 		}
 	}

@@ -1914,11 +1914,8 @@ static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,
 			u32 frag_len = bnxt_rx_agg_pages_xdp(bp, cpr, &xdp,
 							     cp_cons, agg_bufs,
 							     false);
-			if (!frag_len) {
-				cpr->sw_stats.rx.rx_oom_discards += 1;
-				rc = -ENOMEM;
-				goto next_rx;
-			}
+			if (!frag_len)
+				goto oom_next_rx;
 		}
 		xdp_active = true;
 	}

@@ -1941,9 +1938,7 @@ static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,
 				else
 					bnxt_xdp_buff_frags_free(rxr, &xdp);
 			}
-			cpr->sw_stats.rx.rx_oom_discards += 1;
-			rc = -ENOMEM;
-			goto next_rx;
+			goto oom_next_rx;
 		}
 	} else {
 		u32 payload;

@@ -1954,29 +1949,21 @@ static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,
 			payload = 0;
 		skb = bp->rx_skb_func(bp, rxr, cons, data, data_ptr, dma_addr,
 				      payload | len);
-		if (!skb) {
-			cpr->sw_stats.rx.rx_oom_discards += 1;
-			rc = -ENOMEM;
-			goto next_rx;
-		}
+		if (!skb)
+			goto oom_next_rx;
 	}
 
 	if (agg_bufs) {
 		if (!xdp_active) {
 			skb = bnxt_rx_agg_pages_skb(bp, cpr, skb, cp_cons, agg_bufs, false);
-			if (!skb) {
-				cpr->sw_stats.rx.rx_oom_discards += 1;
-				rc = -ENOMEM;
-				goto next_rx;
-			}
+			if (!skb)
+				goto oom_next_rx;
 		} else {
 			skb = bnxt_xdp_build_skb(bp, skb, agg_bufs, rxr->page_pool, &xdp, rxcmp1);
 			if (!skb) {
 				/* we should be able to free the old skb here */
 				bnxt_xdp_buff_frags_free(rxr, &xdp);
-				cpr->sw_stats.rx.rx_oom_discards += 1;
-				rc = -ENOMEM;
-				goto next_rx;
+				goto oom_next_rx;
 			}
 		}
 	}

@@ -2054,6 +2041,11 @@ static int bnxt_rx_pkt(struct bnxt *bp, struct bnxt_cp_ring_info *cpr,
 	*raw_cons = tmp_raw_cons;
 
 	return rc;
+
+oom_next_rx:
+	cpr->bnapi->cp_ring.sw_stats.rx.rx_oom_discards += 1;
+	rc = -ENOMEM;
+	goto next_rx;
 }
 
 /* In netpoll mode, if we are using a combined completion ring, we need to

@@ -2099,7 +2091,7 @@ static int bnxt_force_rx_discard(struct bnxt *bp,
 	}
 	rc = bnxt_rx_pkt(bp, cpr, raw_cons, event);
 	if (rc && rc != -EBUSY)
-		cpr->sw_stats.rx.rx_netpoll_discards += 1;
+		cpr->bnapi->cp_ring.sw_stats.rx.rx_netpoll_discards += 1;
 	return rc;
 }

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:10 UTC #90 lore raw

PATCH 85/186 ARM: dts: imx6ull-tarragon: fix USB over-current polarity

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Michael Heimpold <[email protected]>

[ Upstream commit d7f3040a565214a30e2f07dc9b91566d316e2d36 ]

Our Tarragon platform uses a active-low signal to inform the i.MX6ULL about the over-current detection.

Fixes 5e4f393ccbf0 ("ARM: dts: imx6ull: Add chargebyte Tarragon support")

Signed-off-by Michael Heimpold <[email protected]>

Signed-off-by Stefan Wahren <[email protected]>

Signed-off-by Shawn Guo <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/arm/boot/dts/nxp/imx/imx6ull-tarragon-common.dtsi | 1 +
 1 file changed, 1 insertion(+)

arch/arm/boot/dts/nxp/imx/imx6ull-tarragon-common.dtsi +1 -0

index 3fdece5bd31f9..5248a058230c8 100644
@@ -805,6 +805,7 @@
 		     &pinctrl_usb_pwr>;
 	dr_mode = "host";
 	power-active-high;
+	over-current-active-low;
 	disable-over-current;
 	status = "okay";
 };

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:10 UTC #91 lore raw

PATCH 86/186 netfilter: nf_tables: honor table dormant flag from netdev release event path

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Pablo Neira Ayuso <[email protected]>

[ Upstream commit 8e30abc9ace4f0add4cd761dfdbfaebae5632dd2 ]

Check for table dormant flag otherwise netdev release event path tries to unregister an already unregistered hook.

[524854.857999] ------------[ cut here ]------------
[524854.858010] WARNING: CPU: 0 PID: 3386599 at net/netfilter/core.c:501 __nf_unregister_net_hook+0x21a/0x260
[...]
[524854.858848] CPU: 0 PID: 3386599 Comm: kworker/u32:2 Not tainted 6.9.0-rc3+ #365
[524854.858869] Workqueue: netns cleanup_net
[524854.858886] RIP: 0010:__nf_unregister_net_hook+0x21a/0x260
[524854.858903] Code: 24 e8 aa 73 83 ff 48 63 43 1c 83 f8 01 0f 85 3d ff ff ff e8 98 d1 f0 ff 48 8b 3c 24 e8 8f 73 83 ff 48 63 43 1c e9 26 ff ff ff <0f> 0b 48 83 c4 18 48 c7 c7 00 68 e9 82 5b 5d 41 5c 41 5d 41 5e 41
[524854.858914] RSP: 0018:ffff8881e36d79e0 EFLAGS: 00010246
[524854.858926] RAX: 0000000000000000 RBX: ffff8881339ae790 RCX: ffffffff81ba524a
[524854.858936] RDX: dffffc0000000000 RSI: 0000000000000008 RDI: ffff8881c8a16438
[524854.858945] RBP: ffff8881c8a16438 R08: 0000000000000001 R09: ffffed103c6daf34
[524854.858954] R10: ffff8881e36d79a7 R11: 0000000000000000 R12: 0000000000000005
[524854.858962] R13: ffff8881c8a16000 R14: 0000000000000000 R15: ffff8881351b5a00
[524854.858971] FS:  0000000000000000(0000) GS:ffff888390800000(0000) knlGS:0000000000000000
[524854.858982] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[524854.858991] CR2: 00007fc9be0f16f4 CR3: 00000001437cc004 CR4: 00000000001706f0
[524854.859000] Call Trace:
[524854.859006]  <TASK>
[524854.859013]  ? __warn+0x9f/0x1a0
[524854.859027]  ? __nf_unregister_net_hook+0x21a/0x260
[524854.859044]  ? report_bug+0x1b1/0x1e0
[524854.859060]  ? handle_bug+0x3c/0x70
[524854.859071]  ? exc_invalid_op+0x17/0x40
[524854.859083]  ? asm_exc_invalid_op+0x1a/0x20
[524854.859100]  ? __nf_unregister_net_hook+0x6a/0x260
[524854.859116]  ? __nf_unregister_net_hook+0x21a/0x260
[524854.859135]  nf_tables_netdev_event+0x337/0x390 [nf_tables]
[524854.859304]  ? __pfx_nf_tables_netdev_event+0x10/0x10 [nf_tables]
[524854.859461]  ? packet_notifier+0xb3/0x360
[524854.859476]  ? _raw_spin_unlock_irqrestore+0x11/0x40
[524854.859489]  ? dcbnl_netdevice_event+0x35/0x140
[524854.859507]  ? __pfx_nf_tables_netdev_event+0x10/0x10 [nf_tables]
[524854.859661]  notifier_call_chain+0x7d/0x140
[524854.859677]  unregister_netdevice_many_notify+0x5e1/0xae0

Fixes d54725cd11a5 ("netfilter: nf_tables: support for multiple devices per netdev hook")

Signed-off-by Pablo Neira Ayuso <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 net/netfilter/nft_chain_filter.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

net/netfilter/nft_chain_filter.c +3 -1

index 274b6f7e6bb57..d170758a1eb5d 100644
@@ -338,7 +338,9 @@ static void nft_netdev_event(unsigned long event, struct net_device *dev,
 		return;
 
 	if (n > 1) {
-		nf_unregister_net_hook(ctx->net, &found->ops);
+		if (!(ctx->chain->table->flags & NFT_TABLE_F_DORMANT))
+			nf_unregister_net_hook(ctx->net, &found->ops);
+
 		list_del_rcu(&found->list);
 		kfree_rcu(found, rcu);
 		return;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:10 UTC #92 lore raw

PATCH 87/186 net: phy: dp83869: Fix MII mode failure

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: MD Danish Anwar <[email protected]>

[ Upstream commit 6c9cd59dbcb09a2122b5ce0dfc07c74e6fc00dc0 ]

The DP83869 driver sets the MII bit (needed for PHY to work in MII mode) only if the op-mode is either DP83869_100M_MEDIA_CONVERT or DP83869_RGMII_100_BASE.

Some drivers i.e. ICSSG support MII mode with op-mode as DP83869_RGMII_COPPER_ETHERNET for which the MII bit is not set in dp83869 driver. As a result MII mode on ICSSG doesn't work and below log is seen.

TI DP83869 300b2400.mdio:0f: selected op-mode is not valid with MII mode icssg-prueth icssg1-eth: couldn't connect to phy ethernet-phy@0 icssg-prueth icssg1-eth: can't phy connect port MII0

Fix this by setting MII bit for DP83869_RGMII_COPPER_ETHERNET op-mode as well.

Fixes 94e86ef1b801 ("net: phy: dp83869: support mii mode when rgmii strap cfg is used")

Signed-off-by MD Danish Anwar <[email protected]>

Reviewed-by Ravi Gunasekaran <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/phy/dp83869.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

drivers/net/phy/dp83869.c +2 -1

index fa8c6fdcf3018..d7aaefb5226b6 100644
@@ -695,7 +695,8 @@ static int dp83869_configure_mode(struct phy_device *phydev,
 	phy_ctrl_val = dp83869->mode;
 	if (phydev->interface == PHY_INTERFACE_MODE_MII) {
 		if (dp83869->mode == DP83869_100M_MEDIA_CONVERT ||
-		    dp83869->mode == DP83869_RGMII_100_BASE) {
+		    dp83869->mode == DP83869_RGMII_100_BASE ||
+		    dp83869->mode == DP83869_RGMII_COPPER_ETHERNET) {
 			phy_ctrl_val |= DP83869_OP_MODE_MII;
 		} else {
 			phydev_err(phydev, "selected op-mode is not valid with MII mode\n");

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:10 UTC #93 lore raw

PATCH 88/186 net: ti: icssg-prueth: Fix signedness bug in prueth_init_rx_chns()

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Dan Carpenter <[email protected]>

[ Upstream commit 4dcd0e83ea1d1df9b2e0174a6d3e795b3477d64e ]

The rx_chn->irq[] array is unsigned int but it should be signed for the error handling to work. Also if k3_udma_glue_rx_get_irq() returns zero then we should return -ENXIO instead of success.

Fixes 128d5874c082 ("net: ti: icssg-prueth: Add ICSSG ethernet driver")

Signed-off-by Dan Carpenter <[email protected]>

Reviewed-by Roger Quadros <[email protected]>

Reviewed-by MD Danish Anwar <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/ti/icssg/icssg_prueth.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

drivers/net/ethernet/ti/icssg/icssg_prueth.c +5 -3

index c09ecb3da7723..925044c16c6ae 100644
@@ -421,12 +421,14 @@ static int prueth_init_rx_chns(struct prueth_emac *emac,
 		if (!i)
 			fdqring_id = k3_udma_glue_rx_flow_get_fdq_id(rx_chn->rx_chn,
 								     i);
-		rx_chn->irq[i] = k3_udma_glue_rx_get_irq(rx_chn->rx_chn, i);
-		if (rx_chn->irq[i] <= 0) {
-			ret = rx_chn->irq[i];
+		ret = k3_udma_glue_rx_get_irq(rx_chn->rx_chn, i);
+		if (ret <= 0) {
+			if (!ret)
+				ret = -ENXIO;
 			netdev_err(ndev, "Failed to get rx dma irq");
 			goto fail;
 		}
+		rx_chn->irq[i] = ret;
 	}
 
 	return 0;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:10 UTC #94 lore raw

PATCH 89/186 i40e: Do not use WQ_MEM_RECLAIM flag for workqueue

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sindhu Devale <[email protected]>

[ Upstream commit 2cc7d150550cc981aceedf008f5459193282425c ]

Issue reported by customer during SRIOV testing, call trace:
When both i40e and the i40iw driver are loaded, a warning in check_flush_dependency is being triggered. This seems to be because of the i40e driver workqueue is allocated with the WQ_MEM_RECLAIM flag, and the i40iw one is not.

Similar error was encountered on ice too and it was fixed by removing the flag. Do the same for i40e too.

[Feb 9 09:08] ------------[ cut here ]------------
[ +0.000004] workqueue: WQ_MEM_RECLAIM i40e:i40e_service_task [i40e] is flushing !WQ_MEM_RECLAIM infiniband:0x0
[ +0.000060] WARNING: CPU: 0 PID: 937 at kernel/workqueue.c:2966 check_flush_dependency+0x10b/0x120
[ +0.000007] Modules linked in: snd_seq_dummy snd_hrtimer snd_seq snd_timer snd_seq_device snd soundcore nls_utf8 cifs cifs_arc4 nls_ucs2_utils rdma_cm iw_cm ib_cm cifs_md4 dns_resolver netfs qrtr rfkill sunrpc vfat fat intel_rapl_msr intel_rapl_common irdma intel_uncore_frequency intel_uncore_frequency_common ice ipmi_ssif isst_if_common skx_edac nfit libnvdimm x86_pkg_temp_thermal intel_powerclamp gnss coretemp ib_uverbs rapl intel_cstate ib_core iTCO_wdt iTCO_vendor_support acpi_ipmi mei_me ipmi_si intel_uncore ioatdma i2c_i801 joydev pcspkr mei ipmi_devintf lpc_ich intel_pch_thermal i2c_smbus ipmi_msghandler acpi_power_meter acpi_pad xfs libcrc32c ast sd_mod drm_shmem_helper t10_pi drm_kms_helper sg ixgbe drm i40e ahci crct10dif_pclmul libahci crc32_pclmul igb crc32c_intel libata ghash_clmulni_intel i2c_algo_bit mdio dca wmi dm_mirror dm_region_hash dm_log dm_mod fuse
[ +0.000050] CPU: 0 PID: 937 Comm: kworker/0:3 Kdump: loaded Not tainted 6.8.0-rc2-Feb-net_dev-Qiueue-00279-gbd43c5687e05 #1 [ +0.000003] Hardware name: Intel Corporation S2600BPB/S2600BPB, BIOS SE5C620.86B.02.01.0013.121520200651 12/15/2020

[  +0.000001] Workqueue: i40e i40e_service_task [i40e]
[  +0.000024] RIP: 0010:check_flush_dependency+0x10b/0x120
[  +0.000003] Code: ff 49 8b 54 24 18 48 8d 8b b0 00 00 00 49 89 e8 48

81 c6 b0 00 00 00 48 c7 c7 b0 97 fa 9f c6 05 8a cc 1f 02 01 e8 35 b3 fd ff <0f> 0b e9 10 ff ff ff 80 3d 78 cc 1f 02 00 75 94 e9 46 ff ff ff 90 [ +0.000002] RSP: 0018:ffffbd294976bcf8 EFLAGS: 00010282
[ +0.000002] RAX: 0000000000000000 RBX: ffff94d4c483c000 RCX:
0000000000000027
[ +0.000001] RDX: ffff94d47f620bc8 RSI: 0000000000000001 RDI:
ffff94d47f620bc0
[ +0.000001] RBP: 0000000000000000 R08: 0000000000000000 R09:
00000000ffff7fff
[ +0.000001] R10: ffffbd294976bb98 R11: ffffffffa0be65e8 R12:
ffff94c5451ea180
[ +0.000001] R13: ffff94c5ab5e8000 R14: ffff94c5c20b6e05 R15:
ffff94c5f1330ab0
[ +0.000001] FS: 0000000000000000(0000) GS:ffff94d47f600000(0000) knlGS:0000000000000000
[ +0.000002] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ +0.000001] CR2: 00007f9e6f1fca70 CR3: 0000000038e20004 CR4:
00000000007706f0
[ +0.000000] DR0: 0000000000000000 DR1: 0000000000000000 DR2:
0000000000000000
[ +0.000001] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7:
0000000000000400

[  +0.000001] PKRU: 55555554
[  +0.000001] Call Trace:
[  +0.000001]  <TASK>
[  +0.000002]  ? __warn+0x80/0x130
[  +0.000003]  ? check_flush_dependency+0x10b/0x120
[  +0.000002]  ? report_bug+0x195/0x1a0
[  +0.000005]  ? handle_bug+0x3c/0x70
[  +0.000003]  ? exc_invalid_op+0x14/0x70
[  +0.000002]  ? asm_exc_invalid_op+0x16/0x20
[  +0.000006]  ? check_flush_dependency+0x10b/0x120
[  +0.000002]  ? check_flush_dependency+0x10b/0x120
[  +0.000002]  __flush_workqueue+0x126/0x3f0
[  +0.000015]  ib_cache_cleanup_one+0x1c/0xe0 [ib_core]
[  +0.000056]  __ib_unregister_device+0x6a/0xb0 [ib_core]
[  +0.000023]  ib_unregister_device_and_put+0x34/0x50 [ib_core]
[  +0.000020]  i40iw_close+0x4b/0x90 [irdma]
[  +0.000022]  i40e_notify_client_of_netdev_close+0x54/0xc0 [i40e]
[  +0.000035]  i40e_service_task+0x126/0x190 [i40e]
[  +0.000024]  process_one_work+0x174/0x340
[  +0.000003]  worker_thread+0x27e/0x390
[  +0.000001]  ? __pfx_worker_thread+0x10/0x10
[  +0.000002]  kthread+0xdf/0x110
[  +0.000002]  ? __pfx_kthread+0x10/0x10
[  +0.000002]  ret_from_fork+0x2d/0x50
[  +0.000003]  ? __pfx_kthread+0x10/0x10
[  +0.000001]  ret_from_fork_asm+0x1b/0x30
[  +0.000004]  </TASK>
[  +0.000001] ---[ end trace 0000000000000000 ]---

Fixes 4d5957cbdecd ("i40e: remove WQ_UNBOUND and the task limit of our workqueue")

Signed-off-by Sindhu Devale <[email protected]>

Reviewed-by Arkadiusz Kubalewski <[email protected]>

Reviewed-by Mateusz Polchlopek <[email protected]>

Signed-off-by Aleksandr Loktionov <[email protected]>

Tested-by Robert Ganzynkowicz <[email protected]>

Signed-off-by Tony Nguyen <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/intel/i40e/i40e_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

drivers/net/ethernet/intel/i40e/i40e_main.c +1 -1

index a21fc92aa2725..54d3bd8ec5ea3 100644
@@ -16778,7 +16778,7 @@ static int __init i40e_init_module(void)
 	 * since we need to be able to guarantee forward progress even under
 	 * memory pressure.
 	 */
-	i40e_wq = alloc_workqueue("%s", WQ_MEM_RECLAIM, 0, i40e_driver_name);
+	i40e_wq = alloc_workqueue("%s", 0, 0, i40e_driver_name);
 	if (!i40e_wq) {
 		pr_err("%s: Failed to create workqueue\n", i40e_driver_name);
 		return -ENOMEM;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:10 UTC #95 lore raw

PATCH 90/186 i40e: Report MFS in decimal base instead of hex

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Erwan Velu <[email protected]>

[ Upstream commit ef3c313119ea448c22da10366faa26b5b4b1a18e ]

If the MFS is set below the default (0x2600), a warning message is reported like the following :

MFS for port 1 has been set below the default: 600

This message is a bit confusing as the number shown here (600) is in fact an hexa number: 0x600 = 1536

Without any explicit "0x" prefix, this message is read like the MFS is set to 600 bytes.

MFS, as per MTUs, are usually expressed in decimal base.

This commit reports both current and default MFS values in decimal so it's less confusing for end-users.

A typical warning message looks like the following :

MFS for port 1 (1536) has been set below the default (9728)

Signed-off-by Erwan Velu <[email protected]>

Tested-by Tony Brelinski <[email protected]>

Signed-off-by Tony Nguyen <[email protected]>

Fixes 3a2c6ced90e1 ("i40e: Add a check to see if MFS is set")

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/intel/i40e/i40e_main.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

drivers/net/ethernet/intel/i40e/i40e_main.c +2 -2

index 54d3bd8ec5ea3..f8d1a994c2f65 100644
@@ -16237,8 +16237,8 @@ static int i40e_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 	val = (rd32(&pf->hw, I40E_PRTGL_SAH) &
 	       I40E_PRTGL_SAH_MFS_MASK) >> I40E_PRTGL_SAH_MFS_SHIFT;
 	if (val < MAX_FRAME_SIZE_DEFAULT)
-		dev_warn(&pdev->dev, "MFS for port %x has been set below the default: %x\n",
-			 pf->hw.port, val);
+		dev_warn(&pdev->dev, "MFS for port %x (%d) has been set below the default (%d)\n",
+			 pf->hw.port, val, MAX_FRAME_SIZE_DEFAULT);
 
 	/* Add a filter to drop all Flow control frames from any VSI from being
 	 * transmitted. By doing so we stop a malicious VF from sending out

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:10 UTC #96 lore raw

PATCH 91/186 iavf: Fix TC config comparison with existing adapter TC config

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sudheer Mogilappagari <[email protected]>

[ Upstream commit 54976cf58d6168b8d15cebb395069f23b2f34b31 ]

Same number of TCs doesn't imply that underlying TC configs are same. The config could be different due to difference in number of queues in each TC. Add utility function to determine if TC configs are same.

Fixes d5b33d024496 ("i40evf: add ndo_setup_tc callback to i40evf")

Signed-off-by Sudheer Mogilappagari <[email protected]>

Tested-by Mineri Bhange <[email protected]> (A Contingent Worker at Intel)

Signed-off-by Tony Nguyen <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/intel/iavf/iavf_main.c | 30 ++++++++++++++++++++-
 1 file changed, 29 insertions(+), 1 deletion(-)

drivers/net/ethernet/intel/iavf/iavf_main.c +29 -1

index 257865647c865..ce0b919995264 100644
@@ -3574,6 +3574,34 @@ static void iavf_del_all_cloud_filters(struct iavf_adapter *adapter)
 	spin_unlock_bh(&adapter->cloud_filter_list_lock);
 }
 
+/**
+ * iavf_is_tc_config_same - Compare the mqprio TC config with the
+ * TC config already configured on this adapter.
+ * @adapter: board private structure
+ * @mqprio_qopt: TC config received from kernel.
+ *
+ * This function compares the TC config received from the kernel
+ * with the config already configured on the adapter.
+ *
+ * Return: True if configuration is same, false otherwise.
+ **/
+static bool iavf_is_tc_config_same(struct iavf_adapter *adapter,
+				   struct tc_mqprio_qopt *mqprio_qopt)
+{
+	struct virtchnl_channel_info *ch = &adapter->ch_config.ch_info[0];
+	int i;
+
+	if (adapter->num_tc != mqprio_qopt->num_tc)
+		return false;
+
+	for (i = 0; i < adapter->num_tc; i++) {
+		if (ch[i].count != mqprio_qopt->count[i] ||
+		    ch[i].offset != mqprio_qopt->offset[i])
+			return false;
+	}
+	return true;
+}
+
 /**
  * __iavf_setup_tc - configure multiple traffic classes
  * @netdev: network interface device structure

@@ -3631,7 +3659,7 @@ static int __iavf_setup_tc(struct net_device *netdev, void *type_data)
 		if (ret)
 			return ret;
 		/* Return if same TC config is requested */
-		if (adapter->num_tc == num_tc)
+		if (iavf_is_tc_config_same(adapter, &mqprio_qopt->qopt))
 			return 0;
 		adapter->num_tc = num_tc;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:10 UTC #97 lore raw

PATCH 92/186 ice: fix LAG and VF lock dependency in ice_reset_vf()

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jacob Keller <[email protected]>

[ Upstream commit 96fdd1f6b4ed72a741fb0eb705c0e13049b8721f ]

9f74a3dfcf83 ("ice: Fix VF Reset paths when interface in a failed over aggregate"), the ice driver has acquired the LAG mutex in ice_reset_vf(). The commit placed this lock acquisition just prior to the acquisition of the VF configuration lock.

If ice_reset_vf() acquires the configuration lock via the ICE_VF_RESET_LOCK flag, this could deadlock with ice_vc_cfg_qs_msg() because it always acquires the locks in the order of the VF configuration lock and then the LAG mutex.

Lockdep reports this violation almost immediately on creating and then removing 2 VF:

======================================================
WARNING: possible circular locking dependency detected
6.8.0-rc6 #54 Tainted: G W O
------------------------------------------------------
kworker/60:3/6771 is trying to acquire lock:
ff40d43e099380a0 (&vf->cfg_lock){+.+.}-{3:3}, at: ice_reset_vf+0x22f/0x4d0 [ice]

but task is already holding lock:
ff40d43ea1961210 (&pf->lag_mutex){+.+.}-{3:3}, at: ice_reset_vf+0xb7/0x4d0 [ice]

which lock already depends on the new lock.

the existing dependency chain (in reverse order) is:

-> #1 (&pf->lag_mutex){+.+.}-{3:3}:

       __lock_acquire+0x4f8/0xb40
       lock_acquire+0xd4/0x2d0
       __mutex_lock+0x9b/0xbf0
       ice_vc_cfg_qs_msg+0x45/0x690 [ice]
       ice_vc_process_vf_msg+0x4f5/0x870 [ice]
       __ice_clean_ctrlq+0x2b5/0x600 [ice]
       ice_service_task+0x2c9/0x480 [ice]
       process_one_work+0x1e9/0x4d0
       worker_thread+0x1e1/0x3d0
       kthread+0x104/0x140
       ret_from_fork+0x31/0x50
       ret_from_fork_asm+0x1b/0x30

-> #0 (&vf->cfg_lock){+.+.}-{3:3}:

       check_prev_add+0xe2/0xc50
       validate_chain+0x558/0x800
       __lock_acquire+0x4f8/0xb40
       lock_acquire+0xd4/0x2d0
       __mutex_lock+0x9b/0xbf0
       ice_reset_vf+0x22f/0x4d0 [ice]
       ice_process_vflr_event+0x98/0xd0 [ice]
       ice_service_task+0x1cc/0x480 [ice]
       process_one_work+0x1e9/0x4d0
       worker_thread+0x1e1/0x3d0
       kthread+0x104/0x140
       ret_from_fork+0x31/0x50
       ret_from_fork_asm+0x1b/0x30

other info that might help us debug this:

 Possible unsafe locking scenario:
       CPU0                    CPU1
       ----                    ----
  lock(&pf->lag_mutex);
                               lock(&vf->cfg_lock);
                               lock(&pf->lag_mutex);
  lock(&vf->cfg_lock);

*** DEADLOCK ***
4 locks held by kworker/60:3/6771:
#0: ff40d43e05428b38 ((wq_completion)ice){+.+.}-{0:0}, at: process_one_work+0x176/0x4d0 #1: ff50d06e05197e58 ((work_completion)(&pf->serv_task)){+.+.}-{0:0}, at: process_one_work+0x176/0x4d0 #2: ff40d43ea1960e50 (&pf->vfs.table_lock){+.+.}-{3:3}, at: ice_process_vflr_event+0x48/0xd0 [ice] #3: ff40d43ea1961210 (&pf->lag_mutex){+.+.}-{3:3}, at: ice_reset_vf+0xb7/0x4d0 [ice]

stack backtrace:
CPU: 60 PID: 6771 Comm: kworker/60:3 Tainted: G        W  O       6.8.0-rc6 #54
Hardware name:
Workqueue: ice ice_service_task [ice]
Call Trace:
 <TASK>
 dump_stack_lvl+0x4a/0x80
 check_noncircular+0x12d/0x150
 check_prev_add+0xe2/0xc50
 ? save_trace+0x59/0x230
 ? add_chain_cache+0x109/0x450
 validate_chain+0x558/0x800
 __lock_acquire+0x4f8/0xb40
 ? lockdep_hardirqs_on+0x7d/0x100
 lock_acquire+0xd4/0x2d0
 ? ice_reset_vf+0x22f/0x4d0 [ice]
 ? lock_is_held_type+0xc7/0x120
 __mutex_lock+0x9b/0xbf0
 ? ice_reset_vf+0x22f/0x4d0 [ice]
 ? ice_reset_vf+0x22f/0x4d0 [ice]
 ? rcu_is_watching+0x11/0x50
 ? ice_reset_vf+0x22f/0x4d0 [ice]
 ice_reset_vf+0x22f/0x4d0 [ice]
 ? process_one_work+0x176/0x4d0
 ice_process_vflr_event+0x98/0xd0 [ice]
 ice_service_task+0x1cc/0x480 [ice]
 process_one_work+0x1e9/0x4d0
 worker_thread+0x1e1/0x3d0
 ? __pfx_worker_thread+0x10/0x10
 kthread+0x104/0x140
 ? __pfx_kthread+0x10/0x10
 ret_from_fork+0x31/0x50
 ? __pfx_kthread+0x10/0x10
 ret_from_fork_asm+0x1b/0x30
 </TASK>

To avoid deadlock, we must acquire the LAG mutex only after acquiring the VF configuration lock. Fix the ice_reset_vf() to acquire the LAG mutex only after we either acquire or check that the VF configuration lock is held.

Fixes 9f74a3dfcf83 ("ice: Fix VF Reset paths when interface in a failed over aggregate")

Signed-off-by Jacob Keller <[email protected]>

Reviewed-by Dave Ertman <[email protected]>

Reviewed-by Mateusz Polchlopek <[email protected]>

Tested-by Przemek Kitszel <[email protected]>

Tested-by Rafal Romanowski <[email protected]>

Signed-off-by Tony Nguyen <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/intel/ice/ice_vf_lib.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

drivers/net/ethernet/intel/ice/ice_vf_lib.c +8 -8

index d488c7156d093..03b9d7d748518 100644
@@ -847,6 +847,11 @@ int ice_reset_vf(struct ice_vf *vf, u32 flags)
 		return 0;
 	}
 
+	if (flags & ICE_VF_RESET_LOCK)
+		mutex_lock(&vf->cfg_lock);
+	else
+		lockdep_assert_held(&vf->cfg_lock);
+
 	lag = pf->lag;
 	mutex_lock(&pf->lag_mutex);
 	if (lag && lag->bonded && lag->primary) {

@@ -858,11 +863,6 @@ int ice_reset_vf(struct ice_vf *vf, u32 flags)
 			act_prt = ICE_LAG_INVALID_PORT;
 	}
 
-	if (flags & ICE_VF_RESET_LOCK)
-		mutex_lock(&vf->cfg_lock);
-	else
-		lockdep_assert_held(&vf->cfg_lock);
-
 	if (ice_is_vf_disabled(vf)) {
 		vsi = ice_get_vf_vsi(vf);
 		if (!vsi) {

@@ -947,14 +947,14 @@ int ice_reset_vf(struct ice_vf *vf, u32 flags)
 	ice_mbx_clear_malvf(&vf->mbx_info);
 
 out_unlock:
-	if (flags & ICE_VF_RESET_LOCK)
-		mutex_unlock(&vf->cfg_lock);
-
 	if (lag && lag->bonded && lag->primary &&
 	    act_prt != ICE_LAG_INVALID_PORT)
 		ice_lag_move_vf_nodes_cfg(lag, pri_prt, act_prt);
 	mutex_unlock(&pf->lag_mutex);
 
+	if (flags & ICE_VF_RESET_LOCK)
+		mutex_unlock(&vf->cfg_lock);
+
 	return err;
 }

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:10 UTC #98 lore raw

PATCH 93/186 net: ethernet: ti: am65-cpts: Fix PTPv1 message type on TX packets

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jason Reeder <[email protected]>

[ Upstream commit 1b9e743e923b256e353a9a644195372285e5a6c0 ]

The CPTS, by design, captures the messageType (Sync, Delay_Req, etc.) field from the second nibble of the PTP header which is defined in the PTPv2 (1588-2008) specification. In the PTPv1 (1588-2002) specification the first two bytes of the PTP header are defined as the versionType which is always 0x0001. This means that any PTPv1 packets that are tagged for TX timestamping by the CPTS will have their messageType set to 0x0 which corresponds to a Sync message type. This causes issues when a PTPv1 stack is expecting a Delay_Req (messageType: 0x1) timestamp that never appears.

Fix this by checking if the ptp_class of the timestamped TX packet is PTP_CLASS_V1 and then matching the PTP sequence ID to the stored sequence ID in the skb->cb data structure. If the sequence IDs match and the packet is of type PTPv1 then there is a chance that the messageType has been incorrectly stored by the CPTS so overwrite the messageType stored by the CPTS with the messageType from the skb->cb data structure. This allows the PTPv1 stack to receive TX timestamps for Delay_Req packets which are necessary to lock onto a PTP Leader.

Signed-off-by Jason Reeder <[email protected]>

Signed-off-by Ravi Gunasekaran <[email protected]>

Tested-by Ed Trexel <[email protected]>

Fixes f6bd59526ca5 ("net: ethernet: ti: introduce am654 common platform time sync driver")

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/net/ethernet/ti/am65-cpts.c | 5 +++++
 1 file changed, 5 insertions(+)

drivers/net/ethernet/ti/am65-cpts.c +5 -0

index c66618d91c28f..f89716b1cfb64 100644
@@ -784,6 +784,11 @@ static bool am65_cpts_match_tx_ts(struct am65_cpts *cpts,
 		struct am65_cpts_skb_cb_data *skb_cb =
 					(struct am65_cpts_skb_cb_data *)skb->cb;
 
+		if ((ptp_classify_raw(skb) & PTP_CLASS_V1) &&
+		    ((mtype_seqid & AM65_CPTS_EVENT_1_SEQUENCE_ID_MASK) ==
+		     (skb_cb->skb_mtype_seqid & AM65_CPTS_EVENT_1_SEQUENCE_ID_MASK)))
+			mtype_seqid = skb_cb->skb_mtype_seqid;
+
 		if (mtype_seqid == skb_cb->skb_mtype_seqid) {
 			u64 ns = event->timestamp;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:10 UTC #99 lore raw

PATCH 94/186 tls: fix lockless read of strp->msg_ready in ->poll

Link https://lore.kernel.org/r/0b7ee062319037cf86af6b317b3d72f7bfcd2e97.1713797701.git.sd@queasysnail.net open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sabrina Dubroca <[email protected]>

[ Upstream commit 0844370f8945086eb9335739d10205dcea8d707b ]

tls_sk_poll is called without locking the socket, and needs to read strp->msg_ready (via tls_strp_msg_ready). Convert msg_ready to a bool and use READ_ONCE/WRITE_ONCE where needed. The remaining reads are only performed when the socket is locked.

Fixes 121dca784fc0 ("tls: suppress wakeups unless we have a full record")

Signed-off-by Sabrina Dubroca <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 include/net/tls.h  | 3 ++-
 net/tls/tls.h      | 2 +-
 net/tls/tls_strp.c | 6 +++---
 3 files changed, 6 insertions(+), 5 deletions(-)

include/net/tls.h +2 -1

index 5fdd5dd251df2..2ad28545b15f0 100644
@@ -110,7 +110,8 @@ struct tls_strparser {
 	u32 stopped : 1;
 	u32 copy_mode : 1;
 	u32 mixed_decrypted : 1;
-	u32 msg_ready : 1;
+
+	bool msg_ready;
 
 	struct strp_msg stm;

net/tls/tls.h +1 -1

index 28a8c0e80e3c5..02038d0381b75 100644
@@ -212,7 +212,7 @@ static inline struct sk_buff *tls_strp_msg(struct tls_sw_context_rx *ctx)
 
 static inline bool tls_strp_msg_ready(struct tls_sw_context_rx *ctx)
 {
-	return ctx->strp.msg_ready;
+	return READ_ONCE(ctx->strp.msg_ready);
 }
 
 static inline bool tls_strp_msg_mixed_decrypted(struct tls_sw_context_rx *ctx)

net/tls/tls_strp.c +3 -3

index ca1e0e198ceb4..5df08d848b5c9 100644
@@ -360,7 +360,7 @@ static int tls_strp_copyin(read_descriptor_t *desc, struct sk_buff *in_skb,
 	if (strp->stm.full_len && strp->stm.full_len == skb->len) {
 		desc->count = 0;
 
-		strp->msg_ready = 1;
+		WRITE_ONCE(strp->msg_ready, 1);
 		tls_rx_msg_ready(strp);
 	}

@@ -528,7 +528,7 @@ static int tls_strp_read_sock(struct tls_strparser *strp)
 	if (!tls_strp_check_queue_ok(strp))
 		return tls_strp_read_copy(strp, false);
 
-	strp->msg_ready = 1;
+	WRITE_ONCE(strp->msg_ready, 1);
 	tls_rx_msg_ready(strp);
 
 	return 0;

@@ -580,7 +580,7 @@ void tls_strp_msg_done(struct tls_strparser *strp)
 	else
 		tls_strp_flush_anchor_copy(strp);
 
-	strp->msg_ready = 0;
+	WRITE_ONCE(strp->msg_ready, 0);
 	memset(&strp->stm, 0, sizeof(strp->stm));
 
 	tls_strp_check_rcv(strp);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:10 UTC #100 lore raw

PATCH 95/186 af_unix: Suppress false-positive lockdep splat for spin_lock() in __unix_gc().

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Kuniyuki Iwashima <[email protected]>

[ Upstream commit 1971d13ffa84a551d29a81fdf5b5ec5be166ac83 ]

syzbot reported a lockdep splat regarding unix_gc_lock and unix_state_lock().

One is called from recvmsg() for a connected socket, and another is called from GC for TCP_LISTEN socket.

So, the splat is false-positive.

Let's add a dedicated lock class for the latter to suppress the splat.

Note that this change is not necessary for net-next.git as the issue is only applied to the old GC impl.

[0]:
WARNING: possible circular locking dependency detected 6.9.0-rc5-syzkaller-00007-g4d2008430ce8 #0 Not tainted
-----------------------------------------------------
kworker/u8:1/11 is trying to acquire lock:
ffff88807cea4e70 (&u->lock){+.+.}-{2:2}, at: spin_lock include/linux/spinlock.h:351 [inline] ffff88807cea4e70 (&u->lock){+.+.}-{2:2}, at: __unix_gc+0x40e/0xf70 net/unix/garbage.c:302

but task is already holding lock:
ffffffff8f6ab638 (unix_gc_lock){+.+.}-{2:2}, at: spin_lock include/linux/spinlock.h:351 [inline] ffffffff8f6ab638 (unix_gc_lock){+.+.}-{2:2}, at: __unix_gc+0x117/0xf70 net/unix/garbage.c:261

which lock already depends on the new lock.

the existing dependency chain (in reverse order) is:

-> #1 (unix_gc_lock){+.+.}-{2:2}:

       lock_acquire+0x1ed/0x550 kernel/locking/lockdep.c:5754
       __raw_spin_lock include/linux/spinlock_api_smp.h:133 [inline]
       _raw_spin_lock+0x2e/0x40 kernel/locking/spinlock.c:154
       spin_lock include/linux/spinlock.h:351 [inline]
       unix_notinflight+0x13d/0x390 net/unix/garbage.c:140
       unix_detach_fds net/unix/af_unix.c:1819 [inline]
       unix_destruct_scm+0x221/0x350 net/unix/af_unix.c:1876
       skb_release_head_state+0x100/0x250 net/core/skbuff.c:1188
       skb_release_all net/core/skbuff.c:1200 [inline]
       __kfree_skb net/core/skbuff.c:1216 [inline]
       kfree_skb_reason+0x16d/0x3b0 net/core/skbuff.c:1252
       kfree_skb include/linux/skbuff.h:1262 [inline]
       manage_oob net/unix/af_unix.c:2672 [inline]
       unix_stream_read_generic+0x1125/0x2700 net/unix/af_unix.c:2749
       unix_stream_splice_read+0x239/0x320 net/unix/af_unix.c:2981
       do_splice_read fs/splice.c:985 [inline]
       splice_file_to_pipe+0x299/0x500 fs/splice.c:1295
       do_splice+0xf2d/0x1880 fs/splice.c:1379
       __do_splice fs/splice.c:1436 [inline]
       __do_sys_splice fs/splice.c:1652 [inline]
       __se_sys_splice+0x331/0x4a0 fs/splice.c:1634
       do_syscall_x64 arch/x86/entry/common.c:52 [inline]
       do_syscall_64+0xf5/0x240 arch/x86/entry/common.c:83
       entry_SYSCALL_64_after_hwframe+0x77/0x7f

-> #0 (&u->lock){+.+.}-{2:2}:

       check_prev_add kernel/locking/lockdep.c:3134 [inline]
       check_prevs_add kernel/locking/lockdep.c:3253 [inline]
       validate_chain+0x18cb/0x58e0 kernel/locking/lockdep.c:3869
       __lock_acquire+0x1346/0x1fd0 kernel/locking/lockdep.c:5137
       lock_acquire+0x1ed/0x550 kernel/locking/lockdep.c:5754
       __raw_spin_lock include/linux/spinlock_api_smp.h:133 [inline]
       _raw_spin_lock+0x2e/0x40 kernel/locking/spinlock.c:154
       spin_lock include/linux/spinlock.h:351 [inline]
       __unix_gc+0x40e/0xf70 net/unix/garbage.c:302
       process_one_work kernel/workqueue.c:3254 [inline]
       process_scheduled_works+0xa10/0x17c0 kernel/workqueue.c:3335
       worker_thread+0x86d/0xd70 kernel/workqueue.c:3416
       kthread+0x2f0/0x390 kernel/kthread.c:388
       ret_from_fork+0x4b/0x80 arch/x86/kernel/process.c:147
       ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244

other info that might help us debug this:

 Possible unsafe locking scenario:

       CPU0                    CPU1
       ----                    ----
  lock(unix_gc_lock);
                               lock(&u->lock);
                               lock(unix_gc_lock);

lock(&u->lock);

*** DEADLOCK ***

3 locks held by kworker/u8:1/11:
#0: ffff888015089148 ((wq_completion)events_unbound){+.+.}-{0:0}, at: process_one_work kernel/workqueue.c:3229 [inline] #0: ffff888015089148 ((wq_completion)events_unbound){+.+.}-{0:0}, at: process_scheduled_works+0x8e0/0x17c0 kernel/workqueue.c:3335 #1: ffffc90000107d00 (unix_gc_work){+.+.}-{0:0}, at: process_one_work kernel/workqueue.c:3230 [inline]
#1: ffffc90000107d00 (unix_gc_work){+.+.}-{0:0}, at: process_scheduled_works+0x91b/0x17c0 kernel/workqueue.c:3335 #2: ffffffff8f6ab638 (unix_gc_lock){+.+.}-{2:2}, at: spin_lock include/linux/spinlock.h:351 [inline]
#2: ffffffff8f6ab638 (unix_gc_lock){+.+.}-{2:2}, at: __unix_gc+0x117/0xf70 net/unix/garbage.c:261

stack backtrace:
CPU: 0 PID: 11 Comm: kworker/u8:1 Not tainted 6.9.0-rc5-syzkaller-00007-g4d2008430ce8 #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 03/27/2024
Workqueue: events_unbound __unix_gc
Call Trace:
 <TASK>
 __dump_stack lib/dump_stack.c:88 [inline]
 dump_stack_lvl+0x241/0x360 lib/dump_stack.c:114
 check_noncircular+0x36a/0x4a0 kernel/locking/lockdep.c:2187
 check_prev_add kernel/locking/lockdep.c:3134 [inline]
 check_prevs_add kernel/locking/lockdep.c:3253 [inline]
 validate_chain+0x18cb/0x58e0 kernel/locking/lockdep.c:3869
 __lock_acquire+0x1346/0x1fd0 kernel/locking/lockdep.c:5137
 lock_acquire+0x1ed/0x550 kernel/locking/lockdep.c:5754
 __raw_spin_lock include/linux/spinlock_api_smp.h:133 [inline]
 _raw_spin_lock+0x2e/0x40 kernel/locking/spinlock.c:154
 spin_lock include/linux/spinlock.h:351 [inline]
 __unix_gc+0x40e/0xf70 net/unix/garbage.c:302
 process_one_work kernel/workqueue.c:3254 [inline]
 process_scheduled_works+0xa10/0x17c0 kernel/workqueue.c:3335
 worker_thread+0x86d/0xd70 kernel/workqueue.c:3416
 kthread+0x2f0/0x390 kernel/kthread.c:388
 ret_from_fork+0x4b/0x80 arch/x86/kernel/process.c:147
 ret_from_fork_asm+0x1a/0x30 arch/x86/entry/entry_64.S:244
 </TASK>

Fixes 47d8ac011fe1 ("af_unix: Fix garbage collector racing against connect()")

Reported-and-tested-by [email protected]

Closes https://syzkaller.appspot.com/bug?extid=fa379358c28cc87cc307

Signed-off-by Kuniyuki Iwashima <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sasha Levin <[email protected]>

---
 include/net/af_unix.h | 3 +++
 net/unix/garbage.c    | 2 +-
 2 files changed, 4 insertions(+), 1 deletion(-)

include/net/af_unix.h +3 -0

index d1b07ddbe677e..77bf30203d3cf 100644
@@ -77,6 +77,9 @@ enum unix_socket_lock_class {
 	U_LOCK_NORMAL,
 	U_LOCK_SECOND,	/* for double locking, see unix_state_double_lock(). */
 	U_LOCK_DIAG, /* used while dumping icons, see sk_diag_dump_icons(). */
+	U_LOCK_GC_LISTENER, /* used for listening socket while determining gc
+			     * candidates to close a small race window.
+			     */
 };
 
 static inline void unix_state_lock_nested(struct sock *sk,

net/unix/garbage.c +1 -1

index 8734c0c1fc197..2a758531e1027 100644
@@ -260,7 +260,7 @@ void unix_gc(void)
 			__set_bit(UNIX_GC_MAYBE_CYCLE, &u->gc_flags);
 
 			if (sk->sk_state == TCP_LISTEN) {
-				unix_state_lock(sk);
+				unix_state_lock_nested(sk, U_LOCK_GC_LISTENER);
 				unix_state_unlock(sk);
 			}
 		}

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:10 UTC #101 lore raw

PATCH 96/186 KVM: x86/pmu: Zero out PMU metadata on AMD if PMU is disabled

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sean Christopherson <[email protected]>

[ Upstream commit f933b88e20150f15787390e2a1754a7e412754ed ]

Move the purging of common PMU metadata from intel_pmu_refresh() to kvm_pmu_refresh(), and invoke the vendor refresh() hook if and only if the VM is supposed to have a vPMU.

KVM already denies access to the PMU based on kvm->arch.enable_pmu, as get_gp_pmc_amd() returns NULL for all PMCs in that case, i.e. KVM already violates AMD's architecture by not virtualizing a PMU (kernels have long since learned to not panic when the PMU is unavailable). But configuring the PMU as if it were enabled causes unwanted side effects, e.g. calls to kvm_pmu_trigger_event() waste an absurd number of cycles due to the all_valid_pmc_idx bitmap being non-zero.

Fixes b1d66dad65dc ("KVM: x86/svm: Add module param to control PMU virtualization")

Reported-by Konstantin Khorenko <[email protected]>

Closes https://lore.kernel.org/all/[email protected] open here

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sean Christopherson <[email protected]>

Stable-dep-of: de120e1d692d ("KVM: x86/pmu: Set enable bits for GP counters in PERF_GLOBAL_CTRL at "RESET"")

Signed-off-by Sasha Levin <[email protected]>

---
 arch/x86/kvm/pmu.c           | 20 ++++++++++++++++++--
 arch/x86/kvm/vmx/pmu_intel.c | 16 ++--------------
 2 files changed, 20 insertions(+), 16 deletions(-)

arch/x86/kvm/pmu.c +18 -2

index dc8e8e907cfbf..fa6f5cd70d4c8 100644
@@ -691,6 +691,8 @@ void kvm_pmu_reset(struct kvm_vcpu *vcpu)
  */
 void kvm_pmu_refresh(struct kvm_vcpu *vcpu)
 {
+	struct kvm_pmu *pmu = vcpu_to_pmu(vcpu);
+
 	if (KVM_BUG_ON(kvm_vcpu_has_run(vcpu), vcpu->kvm))
 		return;

@@ -700,8 +702,22 @@ void kvm_pmu_refresh(struct kvm_vcpu *vcpu)
 	 */
 	kvm_pmu_reset(vcpu);
 
-	bitmap_zero(vcpu_to_pmu(vcpu)->all_valid_pmc_idx, X86_PMC_IDX_MAX);
-	static_call(kvm_x86_pmu_refresh)(vcpu);
+	pmu->version = 0;
+	pmu->nr_arch_gp_counters = 0;
+	pmu->nr_arch_fixed_counters = 0;
+	pmu->counter_bitmask[KVM_PMC_GP] = 0;
+	pmu->counter_bitmask[KVM_PMC_FIXED] = 0;
+	pmu->reserved_bits = 0xffffffff00200000ull;
+	pmu->raw_event_mask = X86_RAW_EVENT_MASK;
+	pmu->global_ctrl_mask = ~0ull;
+	pmu->global_status_mask = ~0ull;
+	pmu->fixed_ctr_ctrl_mask = ~0ull;
+	pmu->pebs_enable_mask = ~0ull;
+	pmu->pebs_data_cfg_mask = ~0ull;
+	bitmap_zero(pmu->all_valid_pmc_idx, X86_PMC_IDX_MAX);
+
+	if (vcpu->kvm->arch.enable_pmu)
+		static_call(kvm_x86_pmu_refresh)(vcpu);
 }
 
 void kvm_pmu_init(struct kvm_vcpu *vcpu)

arch/x86/kvm/vmx/pmu_intel.c +2 ~~-14~~

index 1549461fa42b7..48a2f77f62ef3 100644
@@ -493,19 +493,6 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)
 	u64 counter_mask;
 	int i;
 
-	pmu->nr_arch_gp_counters = 0;
-	pmu->nr_arch_fixed_counters = 0;
-	pmu->counter_bitmask[KVM_PMC_GP] = 0;
-	pmu->counter_bitmask[KVM_PMC_FIXED] = 0;
-	pmu->version = 0;
-	pmu->reserved_bits = 0xffffffff00200000ull;
-	pmu->raw_event_mask = X86_RAW_EVENT_MASK;
-	pmu->global_ctrl_mask = ~0ull;
-	pmu->global_status_mask = ~0ull;
-	pmu->fixed_ctr_ctrl_mask = ~0ull;
-	pmu->pebs_enable_mask = ~0ull;
-	pmu->pebs_data_cfg_mask = ~0ull;
-
 	memset(&lbr_desc->records, 0, sizeof(lbr_desc->records));
 
 	/*

@@ -517,8 +504,9 @@ static void intel_pmu_refresh(struct kvm_vcpu *vcpu)
 		return;
 
 	entry = kvm_find_cpuid_entry(vcpu, 0xa);
-	if (!entry || !vcpu->kvm->arch.enable_pmu)
+	if (!entry)
 		return;
+
 	eax.full = entry->eax;
 	edx.full = entry->edx;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:10 UTC #102 lore raw

PATCH 97/186 KVM: x86/pmu: Set enable bits for GP counters in PERF_GLOBAL_CTRL at "RESET"

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sean Christopherson <[email protected]>

[ Upstream commit de120e1d692d73c7eefa3278837b1eb68f90728a ]

Set the enable bits for general purpose counters in IA32_PERF_GLOBAL_CTRL when refreshing the PMU to emulate the MSR's architecturally defined post-RESET behavior. Per Intel's SDM:

IA32_PERF_GLOBAL_CTRL: Sets bits n-1:0 and clears the upper bits.

and

Where "n" is the number of general-purpose counters available in the processor.

AMD also documents this behavior for PerfMonV2 CPUs in one of AMD's many PPRs.

Do not set any PERF_GLOBAL_CTRL bits if there are no general purpose counters, although a literal reading of the SDM would require the CPU to set either bits 63:0 or 31:0. The intent of the behavior is to globally enable all GP counters; honor the intent, if not the letter of the law.

Leaving PERF_GLOBAL_CTRL '0' effectively breaks PMU usage in guests that haven't been updated to work with PMUs that support PERF_GLOBAL_CTRL. This bug was recently exposed when KVM added supported for AMD's PerfMonV2, i.e. when KVM started exposing a vPMU with PERF_GLOBAL_CTRL to guest software that only knew how to program v1 PMUs (that don't support PERF_GLOBAL_CTRL).

Failure to emulate the post-RESET behavior results in such guests unknowingly leaving all general purpose counters globally disabled (the entire reason the post-RESET value sets the GP counter enable bits is to maintain backwards compatibility).

The bug has likely gone unnoticed because PERF_GLOBAL_CTRL has been supported on Intel CPUs for as long as KVM has existed, i.e. hardly anyone is running guest software that isn't aware of PERF_GLOBAL_CTRL on Intel PMUs. And because up until v6.0, KVM _did_ emulate the behavior for Intel CPUs, although the old behavior was likely dumb luck.

Because (a) that old code was also broken in its own way (the history of this code is a comedy of errors), and (b) PERF_GLOBAL_CTRL was documented as having a value of '0' post-RESET in all SDMs before March 2023.

Initial vPMU support in commit f5132b01386b ("KVM: Expose a version 2 architectural PMU to a guests") *almost* got it right (again likely by dumb luck), but for some reason only set the bits if the guest PMU was advertised as v1:

        if (pmu->version == 1) {
                pmu->global_ctrl = (1 << pmu->nr_arch_gp_counters) - 1;
                return;
        }

Commit f19a0c2c2e6a ("KVM: PMU emulation: GLOBAL_CTRL MSR should be enabled on reset") then tried to remedy that goof, presumably because guest PMUs were leaving PERF_GLOBAL_CTRL '0', i.e. weren't enabling counters.

        pmu->global_ctrl = ((1 << pmu->nr_arch_gp_counters) - 1) |
                (((1ull << pmu->nr_arch_fixed_counters) - 1) << X86_PMC_IDX_FIXED);
        pmu->global_ctrl_mask = ~pmu->global_ctrl;

That was KVM's behavior up until commit c49467a45fe0 ("KVM: x86/pmu:
Don't overwrite the pmu->global_ctrl when refreshing") removed *everything*. However, it did so based on the behavior defined by the SDM , which at the time stated that "Global Perf Counter Controls" is '0' at Power-Up and RESET.

But then the March 2023 SDM (325462-079US), stealthily changed its "IA-32 and Intel 64 Processor States Following Power-up, Reset, or INIT" table to say:

IA32_PERF_GLOBAL_CTRL: Sets bits n-1:0 and clears the upper bits.

Note, kvm_pmu_refresh() can be invoked multiple times, i.e. it's not a "pure" RESET flow. But it can only be called prior to the first KVM_RUN, i.e. the guest will only ever observe the final value.

Note #2, KVM has always cleared global_ctrl during refresh (see commit f5132b01386b ("KVM: Expose a version 2 architectural PMU to a guests")), i.e. there is no danger of breaking existing setups by clobbering a value set by userspace.

Reported-by Babu Moger <[email protected]>

Cc Sandipan Das <[email protected]>

Cc Like Xu <[email protected]>

Cc Mingwei Zhang <[email protected]>

Cc Dapeng Mi <[email protected]>

Reviewed-by Dapeng Mi <[email protected]>

Tested-by Dapeng Mi <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Sean Christopherson <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 arch/x86/kvm/pmu.c | 16 ++++++++++++++--
 1 file changed, 14 insertions(+), 2 deletions(-)

arch/x86/kvm/pmu.c +14 -2

index fa6f5cd70d4c8..da2d82e3a8735 100644
@@ -716,8 +716,20 @@ void kvm_pmu_refresh(struct kvm_vcpu *vcpu)
 	pmu->pebs_data_cfg_mask = ~0ull;
 	bitmap_zero(pmu->all_valid_pmc_idx, X86_PMC_IDX_MAX);
 
-	if (vcpu->kvm->arch.enable_pmu)
-		static_call(kvm_x86_pmu_refresh)(vcpu);
+	if (!vcpu->kvm->arch.enable_pmu)
+		return;
+
+	static_call(kvm_x86_pmu_refresh)(vcpu);
+
+	/*
+	 * At RESET, both Intel and AMD CPUs set all enable bits for general
+	 * purpose counters in IA32_PERF_GLOBAL_CTRL (so that software that
+	 * was written for v1 PMUs don't unknowingly leave GP counters disabled
+	 * in the global controls).  Emulate that behavior when refreshing the
+	 * PMU so that userspace doesn't need to manually set PERF_GLOBAL_CTRL.
+	 */
+	if (kvm_pmu_has_perf_global_ctrl(pmu) && pmu->nr_arch_gp_counters)
+		pmu->global_ctrl = GENMASK_ULL(pmu->nr_arch_gp_counters - 1, 0);
 }
 
 void kvm_pmu_init(struct kvm_vcpu *vcpu)

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:10 UTC #103 lore raw

PATCH 98/186 mm/gup: explicitly define and check internal GUP flags, disallow FOLL_TOUCH

Link https://lkml.kernel.org/r/971e013dfe20915612ea8b704e801d7aef9a66b6.1696288092.git.lstoakes@gmail.com open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Lorenzo Stoakes <[email protected]>

[ Upstream commit 0f20bba1688bdf3b32df0162511a67d4eda15790 ]

Rather than open-coding a list of internal GUP flags in is_valid_gup_args(), define which ones are internal.

In addition, explicitly check to see if the user passed in FOLL_TOUCH somehow, as this appears to have been accidentally excluded.

Signed-off-by Lorenzo Stoakes <[email protected]>

Reviewed-by Arnd Bergmann <[email protected]>

Reviewed-by David Hildenbrand <[email protected]>

Reviewed-by Jason Gunthorpe <[email protected]>

Cc Adrian Hunter <[email protected]>

Cc Alexander Shishkin <[email protected]>

Cc Arnaldo Carvalho de Melo <[email protected]>

Cc Catalin Marinas <[email protected]>

Cc Ian Rogers <[email protected]>

Cc Ingo Molnar <[email protected]>

Cc Jiri Olsa <[email protected]>

Cc John Hubbard <[email protected]>

Cc Mark Rutland <[email protected]>

Cc Namhyung Kim <[email protected]>

Cc Oleg Nesterov <[email protected]>

Cc Peter Zijlstra <[email protected]>

Cc Richard Cochran <[email protected]>

Cc Will Deacon <[email protected]>

Stable-dep-of: 631426ba1d45 ("mm/madvise: make MADV_POPULATE_(READ|WRITE) handle VM_FAULT_RETRY properly")

Signed-off-by Sasha Levin <[email protected]>

---
 mm/gup.c      | 5 ++---
 mm/internal.h | 3 +++
 2 files changed, 5 insertions(+), 3 deletions(-)

mm/gup.c +2 -3

index 2f8a2d89fde19..b21b33d1787e1 100644
@@ -2227,12 +2227,11 @@ static bool is_valid_gup_args(struct page **pages, int *locked,
 	/*
 	 * These flags not allowed to be specified externally to the gup
 	 * interfaces:
-	 * - FOLL_PIN/FOLL_TRIED/FOLL_FAST_ONLY are internal only
+	 * - FOLL_TOUCH/FOLL_PIN/FOLL_TRIED/FOLL_FAST_ONLY are internal only
 	 * - FOLL_REMOTE is internal only and used on follow_page()
 	 * - FOLL_UNLOCKABLE is internal only and used if locked is !NULL
 	 */
-	if (WARN_ON_ONCE(gup_flags & (FOLL_PIN | FOLL_TRIED | FOLL_UNLOCKABLE |
-				      FOLL_REMOTE | FOLL_FAST_ONLY)))
+	if (WARN_ON_ONCE(gup_flags & INTERNAL_GUP_FLAGS))
 		return false;
 
 	gup_flags |= to_set;

mm/internal.h +3 -0

index 30cf724ddbce3..50cf76d30a88f 100644
@@ -964,6 +964,9 @@ enum {
 	FOLL_UNLOCKABLE = 1 << 21,
 };
 
+#define INTERNAL_GUP_FLAGS (FOLL_TOUCH | FOLL_TRIED | FOLL_REMOTE | FOLL_PIN | \
+			    FOLL_FAST_ONLY | FOLL_UNLOCKABLE)
+
 /*
  * Indicates for which pages that are write-protected in the page table,
  * whether GUP has to trigger unsharing via FAULT_FLAG_UNSHARE such that the

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:10 UTC #104 lore raw

PATCH 99/186 mm/madvise: make MADV_POPULATE_(READ|WRITE) handle VM_FAULT_RETRY properly

[1] https://lore.kernel.org/all/20240313171936.GN1927156@frogsfrogsfrogs/

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: David Hildenbrand <[email protected]>

[ Upstream commit 631426ba1d45a8672b177ee85ad4cabe760dd131 ]

Darrick reports that in some cases where pread() would fail with -EIO and mmap()+access would generate a SIGBUS signal, MADV_POPULATE_READ / MADV_POPULATE_WRITE will keep retrying forever and not fail with -EFAULT.

While the madvise() call can be interrupted by a signal, this is not the desired behavior. MADV_POPULATE_READ / MADV_POPULATE_WRITE should behave like page faults in that case: fail and not retry forever.

A reproducer can be found at [1].

The reason is that __get_user_pages(), as called by faultin_vma_page_range(), will not handle VM_FAULT_RETRY in a proper way:
it will simply return 0 when VM_FAULT_RETRY happened, making madvise_populate()->faultin_vma_page_range() retry again and again, never setting FOLL_TRIED->FAULT_FLAG_TRIED for __get_user_pages().

__get_user_pages_locked() does what we want, but duplicating that logic in faultin_vma_page_range() feels wrong.

So let's use __get_user_pages_locked() instead, that will detect VM_FAULT_RETRY and set FOLL_TRIED when retrying, making the fault handler return VM_FAULT_SIGBUS (VM_FAULT_ERROR) at some point, propagating -EFAULT from faultin_page() to __get_user_pages(), all the way to madvise_populate().

But, there is an issue: __get_user_pages_locked() will end up re-taking the MM lock and then __get_user_pages() will do another VMA lookup. In the meantime, the VMA layout could have changed and we'd fail with different error codes than we'd want to.

As __get_user_pages() will currently do a new VMA lookup either way, let it do the VMA handling in a different way, controlled by a new FOLL_MADV_POPULATE flag, effectively moving these checks from madvise_populate() + faultin_page_range() in there.

With this change, Darricks reproducer properly fails with -EFAULT, as documented for MADV_POPULATE_READ / MADV_POPULATE_WRITE.

Link https://lkml.kernel.org/r/[email protected] open here

Fixes 4ca9b3859dac ("mm/madvise: introduce MADV_POPULATE_(READ|WRITE) to prefault page tables")

Signed-off-by David Hildenbrand <[email protected]>

Reported-by Darrick J. Wong <[email protected]>

Closes https://lore.kernel.org/all/20240311223815.GW1927156@frogsfrogsfrogs/ open here

Cc Darrick J. Wong <[email protected]>

Cc Hugh Dickins <[email protected]>

Cc Jason Gunthorpe <[email protected]>

Cc John Hubbard <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 mm/gup.c      | 54 ++++++++++++++++++++++++++++++---------------------
 mm/internal.h | 10 ++++++----
 mm/madvise.c  | 17 ++--------------
 3 files changed, 40 insertions(+), 41 deletions(-)

mm/gup.c +32 ~~-22~~

index b21b33d1787e1..cfc0a66d951b9 100644
@@ -1204,6 +1204,22 @@ static long __get_user_pages(struct mm_struct *mm,
 
 		/* first iteration or cross vma bound */
 		if (!vma || start >= vma->vm_end) {
+			/*
+			 * MADV_POPULATE_(READ|WRITE) wants to handle VMA
+			 * lookups+error reporting differently.
+			 */
+			if (gup_flags & FOLL_MADV_POPULATE) {
+				vma = vma_lookup(mm, start);
+				if (!vma) {
+					ret = -ENOMEM;
+					goto out;
+				}
+				if (check_vma_flags(vma, gup_flags)) {
+					ret = -EINVAL;
+					goto out;
+				}
+				goto retry;
+			}
 			vma = gup_vma_lookup(mm, start);
 			if (!vma && in_gate_area(mm, start)) {
 				ret = get_gate_page(mm, start & PAGE_MASK,

@@ -1670,35 +1686,35 @@ long populate_vma_page_range(struct vm_area_struct *vma,
 }
 
 /*
- * faultin_vma_page_range() - populate (prefault) page tables inside the
- *			      given VMA range readable/writable
+ * faultin_page_range() - populate (prefault) page tables inside the
+ *			  given range readable/writable
  *
  * This takes care of mlocking the pages, too, if VM_LOCKED is set.
  *
- * @vma: target vma
+ * @mm: the mm to populate page tables in
  * @start: start address
  * @end: end address
  * @write: whether to prefault readable or writable
  * @locked: whether the mmap_lock is still held
  *
- * Returns either number of processed pages in the vma, or a negative error
- * code on error (see __get_user_pages()).
+ * Returns either number of processed pages in the MM, or a negative error
+ * code on error (see __get_user_pages()). Note that this function reports
+ * errors related to VMAs, such as incompatible mappings, as expected by
+ * MADV_POPULATE_(READ|WRITE).
  *
- * vma->vm_mm->mmap_lock must be held. The range must be page-aligned and
- * covered by the VMA. If it's released, *@locked will be set to 0.
+ * The range must be page-aligned.
+ *
+ * mm->mmap_lock must be held. If it's released, *@locked will be set to 0.
  */
-long faultin_vma_page_range(struct vm_area_struct *vma, unsigned long start,
-			    unsigned long end, bool write, int *locked)
+long faultin_page_range(struct mm_struct *mm, unsigned long start,
+			unsigned long end, bool write, int *locked)
 {
-	struct mm_struct *mm = vma->vm_mm;
 	unsigned long nr_pages = (end - start) / PAGE_SIZE;
 	int gup_flags;
 	long ret;
 
 	VM_BUG_ON(!PAGE_ALIGNED(start));
 	VM_BUG_ON(!PAGE_ALIGNED(end));
-	VM_BUG_ON_VMA(start < vma->vm_start, vma);
-	VM_BUG_ON_VMA(end > vma->vm_end, vma);
 	mmap_assert_locked(mm);
 
 	/*

@@ -1710,19 +1726,13 @@ long faultin_vma_page_range(struct vm_area_struct *vma, unsigned long start,
 	 *		  a poisoned page.
 	 * !FOLL_FORCE: Require proper access permissions.
 	 */
-	gup_flags = FOLL_TOUCH | FOLL_HWPOISON | FOLL_UNLOCKABLE;
+	gup_flags = FOLL_TOUCH | FOLL_HWPOISON | FOLL_UNLOCKABLE |
+		    FOLL_MADV_POPULATE;
 	if (write)
 		gup_flags |= FOLL_WRITE;
 
-	/*
-	 * We want to report -EINVAL instead of -EFAULT for any permission
-	 * problems or incompatible mappings.
-	 */
-	if (check_vma_flags(vma, gup_flags))
-		return -EINVAL;
-
-	ret = __get_user_pages(mm, start, nr_pages, gup_flags,
-			       NULL, locked);
+	ret = __get_user_pages_locked(mm, start, nr_pages, NULL, locked,
+				      gup_flags);
 	lru_add_drain();
 	return ret;
 }

mm/internal.h +6 -4

index 50cf76d30a88f..abed947f784b7 100644
@@ -581,9 +581,8 @@ struct anon_vma *folio_anon_vma(struct folio *folio);
 void unmap_mapping_folio(struct folio *folio);
 extern long populate_vma_page_range(struct vm_area_struct *vma,
 		unsigned long start, unsigned long end, int *locked);
-extern long faultin_vma_page_range(struct vm_area_struct *vma,
-				   unsigned long start, unsigned long end,
-				   bool write, int *locked);
+extern long faultin_page_range(struct mm_struct *mm, unsigned long start,
+		unsigned long end, bool write, int *locked);
 extern bool mlock_future_ok(struct mm_struct *mm, unsigned long flags,
 			       unsigned long bytes);
 /*

@@ -962,10 +961,13 @@ enum {
 	FOLL_FAST_ONLY = 1 << 20,
 	/* allow unlocking the mmap lock */
 	FOLL_UNLOCKABLE = 1 << 21,
+	/* VMA lookup+checks compatible with MADV_POPULATE_(READ|WRITE) */
+	FOLL_MADV_POPULATE = 1 << 22,
 };
 
 #define INTERNAL_GUP_FLAGS (FOLL_TOUCH | FOLL_TRIED | FOLL_REMOTE | FOLL_PIN | \
-			    FOLL_FAST_ONLY | FOLL_UNLOCKABLE)
+			    FOLL_FAST_ONLY | FOLL_UNLOCKABLE | \
+			    FOLL_MADV_POPULATE)
 
 /*
  * Indicates for which pages that are write-protected in the page table,

mm/madvise.c +2 ~~-15~~

index 4dded5d27e7ea..98fdb9288a68a 100644
@@ -917,27 +917,14 @@ static long madvise_populate(struct vm_area_struct *vma,
 {
 	const bool write = behavior == MADV_POPULATE_WRITE;
 	struct mm_struct *mm = vma->vm_mm;
-	unsigned long tmp_end;
 	int locked = 1;
 	long pages;
 
 	*prev = vma;
 
 	while (start < end) {
-		/*
-		 * We might have temporarily dropped the lock. For example,
-		 * our VMA might have been split.
-		 */
-		if (!vma || start >= vma->vm_end) {
-			vma = vma_lookup(mm, start);
-			if (!vma)
-				return -ENOMEM;
-		}
-
-		tmp_end = min_t(unsigned long, end, vma->vm_end);
 		/* Populate (prefault) page tables readable/writable. */
-		pages = faultin_vma_page_range(vma, start, tmp_end, write,
-					       &locked);
+		pages = faultin_page_range(mm, start, end, write, &locked);
 		if (!locked) {
 			mmap_read_lock(mm);
 			locked = 1;

@@ -958,7 +945,7 @@ static long madvise_populate(struct vm_area_struct *vma,
 				pr_warn_once("%s: unhandled return value: %ld\n",
 					     __func__, pages);
 				fallthrough;
-			case -ENOMEM:
+			case -ENOMEM: /* No VMA or out of memory. */
 				return -ENOMEM;
 			}
 		}

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:10 UTC #105 lore raw

PATCH 100/186 drm: add drm_gem_object_is_shared_for_memory_stats() helper

Link https://lore.kernel.org/all/[email protected]/ open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alex Deucher <[email protected]>

[ Upstream commit b31f5eba32ae8cc28e7cfa5a55ec8670d8c718e2 ]

Add a helper so that drm drivers can consistently report shared status via the fdinfo shared memory stats interface.

In addition to handle count, show buffers as shared if they are shared via dma-buf as well (e.g., shared with v4l or some other subsystem).

v2: switch to inline function

Reviewed-by Tvrtko Ursulin <[email protected]> (v1)

Signed-off-by Alex Deucher <[email protected]>

Reviewed-by Christian König <[email protected]>

Stable-dep-of: a6ff969fe9cb ("drm/amdgpu: fix visible VRAM handling during faults")

Signed-off-by Sasha Levin <[email protected]>

---
 include/drm/drm_gem.h | 13 +++++++++++++
 1 file changed, 13 insertions(+)

include/drm/drm_gem.h +13 -0

index bc9f6aa2f3fec..7c2ec139c464a 100644
@@ -544,6 +544,19 @@ unsigned long drm_gem_lru_scan(struct drm_gem_lru *lru,
 
 int drm_gem_evict(struct drm_gem_object *obj);
 
+/**
+ * drm_gem_object_is_shared_for_memory_stats - helper for shared memory stats
+ *
+ * This helper should only be used for fdinfo shared memory stats to determine
+ * if a GEM object is shared.
+ *
+ * @obj: obj in question
+ */
+static inline bool drm_gem_object_is_shared_for_memory_stats(struct drm_gem_object *obj)
+{
+	return (obj->handle_count > 1) || obj->dma_buf;
+}
+
 #ifdef CONFIG_LOCKDEP
 /**
  * drm_gem_gpuva_set_lock() - Set the lock protecting accesses to the gpuva list.

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:10 UTC #106 lore raw

PATCH 101/186 drm/amdgpu: add shared fdinfo stats

Link https://lore.kernel.org/all/[email protected]/ open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alex Deucher <[email protected]>

[ Upstream commit ba1a58d5b907bdf1814f8f57434aebc86233430f ]

Add shared stats. Useful for seeing shared memory.

v2: take dma-buf into account as well
v3: use the new gem helper

Signed-off-by Alex Deucher <[email protected]>

Cc Rob Clark <[email protected]>

Reviewed-by Christian König <[email protected]>

Stable-dep-of: a6ff969fe9cb ("drm/amdgpu: fix visible VRAM handling during faults")

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.c |  4 ++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 11 +++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.h |  6 ++++++
 3 files changed, 21 insertions(+)

drivers/gpu/drm/amd/amdgpu/amdgpu_fdinfo.c +4 -0

index 6038b5021b27b..792c059ff7b35 100644
@@ -105,6 +105,10 @@ void amdgpu_show_fdinfo(struct drm_printer *p, struct drm_file *file)
 		   stats.requested_visible_vram/1024UL);
 	drm_printf(p, "amd-requested-gtt:\t%llu KiB\n",
 		   stats.requested_gtt/1024UL);
+	drm_printf(p, "drm-shared-vram:\t%llu KiB\n", stats.vram_shared/1024UL);
+	drm_printf(p, "drm-shared-gtt:\t%llu KiB\n", stats.gtt_shared/1024UL);
+	drm_printf(p, "drm-shared-cpu:\t%llu KiB\n", stats.cpu_shared/1024UL);
+
 	for (hw_ip = 0; hw_ip < AMDGPU_HW_IP_NUM; ++hw_ip) {
 		if (!usage[hw_ip])
 			continue;

drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +11 -0

index 173b43a5aa13b..394f475877e3b 100644
@@ -1281,25 +1281,36 @@ void amdgpu_bo_get_memory(struct amdgpu_bo *bo,
 			  struct amdgpu_mem_stats *stats)
 {
 	uint64_t size = amdgpu_bo_size(bo);
+	struct drm_gem_object *obj;
 	unsigned int domain;
+	bool shared;
 
 	/* Abort if the BO doesn't currently have a backing store */
 	if (!bo->tbo.resource)
 		return;
 
+	obj = &bo->tbo.base;
+	shared = drm_gem_object_is_shared_for_memory_stats(obj);
+
 	domain = amdgpu_mem_type_to_domain(bo->tbo.resource->mem_type);
 	switch (domain) {
 	case AMDGPU_GEM_DOMAIN_VRAM:
 		stats->vram += size;
 		if (amdgpu_bo_in_cpu_visible_vram(bo))
 			stats->visible_vram += size;
+		if (shared)
+			stats->vram_shared += size;
 		break;
 	case AMDGPU_GEM_DOMAIN_GTT:
 		stats->gtt += size;
+		if (shared)
+			stats->gtt_shared += size;
 		break;
 	case AMDGPU_GEM_DOMAIN_CPU:
 	default:
 		stats->cpu += size;
+		if (shared)
+			stats->cpu_shared += size;
 		break;
 	}

drivers/gpu/drm/amd/amdgpu/amdgpu_object.h +6 -0

index a3ea8a82db23a..be679c42b0b8c 100644
@@ -138,12 +138,18 @@ struct amdgpu_bo_vm {
 struct amdgpu_mem_stats {
 	/* current VRAM usage, includes visible VRAM */
 	uint64_t vram;
+	/* current shared VRAM usage, includes visible VRAM */
+	uint64_t vram_shared;
 	/* current visible VRAM usage */
 	uint64_t visible_vram;
 	/* current GTT usage */
 	uint64_t gtt;
+	/* current shared GTT usage */
+	uint64_t gtt_shared;
 	/* current system memory usage */
 	uint64_t cpu;
+	/* current shared system memory usage */
+	uint64_t cpu_shared;
 	/* sum of evicted buffers, includes visible VRAM */
 	uint64_t evicted_vram;
 	/* sum of evicted buffers due to CPU access */

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:11 UTC #107 lore raw

PATCH 102/186 drm/amdgpu: fix visible VRAM handling during faults

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Christian König <[email protected]>

[ Upstream commit a6ff969fe9cbf369e3cd0ac54261fec1122682ec ]

When we removed the hacky start code check we actually didn't took into account that *all* VRAM pages needs to be CPU accessible.

Clean up the code and unify the handling into a single helper which checks if the whole resource is CPU accessible.

The only place where a partial check would make sense is during eviction, but that is neglitible.

Fixes aed01a68047b ("drm/amdgpu: Remove TTM resource->start visible VRAM condition v2")

Reviewed-by Alex Deucher <[email protected]>

Signed-off-by Alex Deucher <[email protected]>

CC [email protected]

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c     |  2 +-
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 22 ++++----
 drivers/gpu/drm/amd/amdgpu/amdgpu_object.h | 22 --------
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c    | 61 ++++++++++++++--------
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h    |  3 ++
 5 files changed, 53 insertions(+), 57 deletions(-)

drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c +1 -1

index c0a3afe81bb1a..4294f5e7bff9a 100644
@@ -819,7 +819,7 @@ static int amdgpu_cs_bo_validate(void *param, struct amdgpu_bo *bo)
 
 	p->bytes_moved += ctx.bytes_moved;
 	if (!amdgpu_gmc_vram_full_visible(&adev->gmc) &&
-	    amdgpu_bo_in_cpu_visible_vram(bo))
+	    amdgpu_res_cpu_visible(adev, bo->tbo.resource))
 		p->bytes_moved_vis += ctx.bytes_moved;
 
 	if (unlikely(r == -ENOMEM) && domain != bo->allowed_domains) {

drivers/gpu/drm/amd/amdgpu/amdgpu_object.c +11 ~~-11~~

index 394f475877e3b..361f2cc94e8e5 100644
@@ -625,8 +625,7 @@ int amdgpu_bo_create(struct amdgpu_device *adev,
 		return r;
 
 	if (!amdgpu_gmc_vram_full_visible(&adev->gmc) &&
-	    bo->tbo.resource->mem_type == TTM_PL_VRAM &&
-	    amdgpu_bo_in_cpu_visible_vram(bo))
+	    amdgpu_res_cpu_visible(adev, bo->tbo.resource))
 		amdgpu_cs_report_moved_bytes(adev, ctx.bytes_moved,
 					     ctx.bytes_moved);
 	else

@@ -1280,23 +1279,25 @@ void amdgpu_bo_move_notify(struct ttm_buffer_object *bo, bool evict)
 void amdgpu_bo_get_memory(struct amdgpu_bo *bo,
 			  struct amdgpu_mem_stats *stats)
 {
+	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
+	struct ttm_resource *res = bo->tbo.resource;
 	uint64_t size = amdgpu_bo_size(bo);
 	struct drm_gem_object *obj;
 	unsigned int domain;
 	bool shared;
 
 	/* Abort if the BO doesn't currently have a backing store */
-	if (!bo->tbo.resource)
+	if (!res)
 		return;
 
 	obj = &bo->tbo.base;
 	shared = drm_gem_object_is_shared_for_memory_stats(obj);
 
-	domain = amdgpu_mem_type_to_domain(bo->tbo.resource->mem_type);
+	domain = amdgpu_mem_type_to_domain(res->mem_type);
 	switch (domain) {
 	case AMDGPU_GEM_DOMAIN_VRAM:
 		stats->vram += size;
-		if (amdgpu_bo_in_cpu_visible_vram(bo))
+		if (amdgpu_res_cpu_visible(adev, bo->tbo.resource))
 			stats->visible_vram += size;
 		if (shared)
 			stats->vram_shared += size;

@@ -1395,10 +1396,7 @@ vm_fault_t amdgpu_bo_fault_reserve_notify(struct ttm_buffer_object *bo)
 	/* Remember that this BO was accessed by the CPU */
 	abo->flags |= AMDGPU_GEM_CREATE_CPU_ACCESS_REQUIRED;
 
-	if (bo->resource->mem_type != TTM_PL_VRAM)
-		return 0;
-
-	if (amdgpu_bo_in_cpu_visible_vram(abo))
+	if (amdgpu_res_cpu_visible(adev, bo->resource))
 		return 0;
 
 	/* Can't move a pinned BO to visible VRAM */

@@ -1422,7 +1420,7 @@ vm_fault_t amdgpu_bo_fault_reserve_notify(struct ttm_buffer_object *bo)
 
 	/* this should never happen */
 	if (bo->resource->mem_type == TTM_PL_VRAM &&
-	    !amdgpu_bo_in_cpu_visible_vram(abo))
+	    !amdgpu_res_cpu_visible(adev, bo->resource))
 		return VM_FAULT_SIGBUS;
 
 	ttm_bo_move_to_lru_tail_unlocked(bo);

@@ -1582,6 +1580,7 @@ uint32_t amdgpu_bo_get_preferred_domain(struct amdgpu_device *adev,
  */
 u64 amdgpu_bo_print_info(int id, struct amdgpu_bo *bo, struct seq_file *m)
 {
+	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
 	struct dma_buf_attachment *attachment;
 	struct dma_buf *dma_buf;
 	const char *placement;

@@ -1590,10 +1589,11 @@ u64 amdgpu_bo_print_info(int id, struct amdgpu_bo *bo, struct seq_file *m)
 
 	if (dma_resv_trylock(bo->tbo.base.resv)) {
 		unsigned int domain;
+
 		domain = amdgpu_mem_type_to_domain(bo->tbo.resource->mem_type);
 		switch (domain) {
 		case AMDGPU_GEM_DOMAIN_VRAM:
-			if (amdgpu_bo_in_cpu_visible_vram(bo))
+			if (amdgpu_res_cpu_visible(adev, bo->tbo.resource))
 				placement = "VRAM VISIBLE";
 			else
 				placement = "VRAM";

drivers/gpu/drm/amd/amdgpu/amdgpu_object.h +0 ~~-22~~

index be679c42b0b8c..fa03d9e4874cc 100644
@@ -250,28 +250,6 @@ static inline u64 amdgpu_bo_mmap_offset(struct amdgpu_bo *bo)
 	return drm_vma_node_offset_addr(&bo->tbo.base.vma_node);
 }
 
-/**
- * amdgpu_bo_in_cpu_visible_vram - check if BO is (partly) in visible VRAM
- */
-static inline bool amdgpu_bo_in_cpu_visible_vram(struct amdgpu_bo *bo)
-{
-	struct amdgpu_device *adev = amdgpu_ttm_adev(bo->tbo.bdev);
-	struct amdgpu_res_cursor cursor;
-
-	if (!bo->tbo.resource || bo->tbo.resource->mem_type != TTM_PL_VRAM)
-		return false;
-
-	amdgpu_res_first(bo->tbo.resource, 0, amdgpu_bo_size(bo), &cursor);
-	while (cursor.remaining) {
-		if (cursor.start < adev->gmc.visible_vram_size)
-			return true;
-
-		amdgpu_res_next(&cursor, cursor.size);
-	}
-
-	return false;
-}
-
 /**
  * amdgpu_bo_explicit_sync - return whether the bo is explicitly synced
  */

drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c +38 ~~-23~~

93 lines not on this page · raw message

drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h +3 -0

index 65ec82141a8e0..32cf6b6f6efd9 100644
@@ -139,6 +139,9 @@ int amdgpu_vram_mgr_reserve_range(struct amdgpu_vram_mgr *mgr,
 int amdgpu_vram_mgr_query_page_status(struct amdgpu_vram_mgr *mgr,
 				      uint64_t start);
 
+bool amdgpu_res_cpu_visible(struct amdgpu_device *adev,
+			    struct ttm_resource *res);
+
 int amdgpu_ttm_init(struct amdgpu_device *adev);
 void amdgpu_ttm_fini(struct amdgpu_device *adev);
 void amdgpu_ttm_set_buffer_funcs_status(struct amdgpu_device *adev,

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:11 UTC #108 lore raw

PATCH 103/186 mm, treewide: introduce NR_PAGE_ORDERS

Link https://lkml.kernel.org/r/20240101111512.7empzyifq7kxtzk3@box open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Kirill A. Shutemov <[email protected]>

[ Upstream commit fd37721803c6e73619108f76ad2e12a9aa5fafaf ]

NR_PAGE_ORDERS defines the number of page orders supported by the page allocator, ranging from 0 to MAX_ORDER, MAX_ORDER + 1 in total.

NR_PAGE_ORDERS assists in defining arrays of page orders and allows for more natural iteration over them.

[[email protected]: fixup for kerneldoc warning]

Link https://lkml.kernel.org/r/[email protected] open here

Signed-off-by Kirill A. Shutemov <[email protected]>

Reviewed-by Zi Yan <[email protected]>

Cc Linus Torvalds <[email protected]>

Stable-dep-of: b6976f323a86 ("drm/ttm: stop pooling cached NUMA pages v2")

Signed-off-by Sasha Levin <[email protected]>

---
 .../admin-guide/kdump/vmcoreinfo.rst          |  6 +++---
 arch/arm64/kvm/hyp/include/nvhe/gfp.h         |  2 +-
 arch/sparc/kernel/traps_64.c                  |  2 +-
 drivers/gpu/drm/ttm/tests/ttm_device_test.c   |  2 +-
 drivers/gpu/drm/ttm/ttm_pool.c                | 20 +++++++++----------
 include/drm/ttm/ttm_pool.h                    |  2 +-
 include/linux/mmzone.h                        |  6 ++++--
 kernel/crash_core.c                           |  2 +-
 lib/test_meminit.c                            |  2 +-
 mm/compaction.c                               |  2 +-
 mm/kmsan/init.c                               |  2 +-
 mm/page_alloc.c                               | 13 ++++++------
 mm/page_reporting.c                           |  2 +-
 mm/show_mem.c                                 |  8 ++++----
 mm/vmstat.c                                   | 12 +++++------
 15 files changed, 42 insertions(+), 41 deletions(-)

Documentation/admin-guide/kdump/vmcoreinfo.rst +3 -3

index 599e8d3bcbc31..9235cf4fbabff 100644
@@ -172,7 +172,7 @@ variables.
 Offset of the free_list's member. This value is used to compute the number
 of free pages.
 
-Each zone has a free_area structure array called free_area[MAX_ORDER + 1].
+Each zone has a free_area structure array called free_area[NR_PAGE_ORDERS].
 The free_list represents a linked list of free page blocks.
 
 (list_head, next|prev)

@@ -189,8 +189,8 @@ Offsets of the vmap_area's members. They carry vmalloc-specific
 information. Makedumpfile gets the start address of the vmalloc region
 from this.
 
-(zone.free_area, MAX_ORDER + 1)
--------------------------------
+(zone.free_area, NR_PAGE_ORDERS)
+--------------------------------
 
 Free areas descriptor. User-space tools use this value to iterate the
 free_area ranges. MAX_ORDER is used by the zone buddy allocator.

arch/arm64/kvm/hyp/include/nvhe/gfp.h +1 -1

index fe5472a184a37..97c527ef53c2a 100644
@@ -16,7 +16,7 @@ struct hyp_pool {
 	 * API at EL2.
 	 */
 	hyp_spinlock_t lock;
-	struct list_head free_area[MAX_ORDER + 1];
+	struct list_head free_area[NR_PAGE_ORDERS];
 	phys_addr_t range_start;
 	phys_addr_t range_end;
 	unsigned short max_order;

arch/sparc/kernel/traps_64.c +1 -1

index 08ffd17d5ec34..523a6e5ee9251 100644
@@ -897,7 +897,7 @@ void __init cheetah_ecache_flush_init(void)
 
 	/* Now allocate error trap reporting scoreboard. */
 	sz = NR_CPUS * (2 * sizeof(struct cheetah_err_info));
-	for (order = 0; order <= MAX_ORDER; order++) {
+	for (order = 0; order < NR_PAGE_ORDERS; order++) {
 		if ((PAGE_SIZE << order) >= sz)
 			break;
 	}

drivers/gpu/drm/ttm/tests/ttm_device_test.c +1 -1

index b1b423b68cdf1..19eaff22e6ae0 100644
@@ -175,7 +175,7 @@ static void ttm_device_init_pools(struct kunit *test)
 
 	if (params->pools_init_expected) {
 		for (int i = 0; i < TTM_NUM_CACHING_TYPES; ++i) {
-			for (int j = 0; j <= MAX_ORDER; ++j) {
+			for (int j = 0; j < NR_PAGE_ORDERS; ++j) {
 				pt = pool->caching[i].orders[j];
 				KUNIT_EXPECT_PTR_EQ(test, pt.pool, pool);
 				KUNIT_EXPECT_EQ(test, pt.caching, i);

drivers/gpu/drm/ttm/ttm_pool.c +10 ~~-10~~

index 9b60222511d65..c8ec6a2cac5d4 100644
@@ -65,11 +65,11 @@ module_param(page_pool_size, ulong, 0644);
 
 static atomic_long_t allocated_pages;
 
-static struct ttm_pool_type global_write_combined[MAX_ORDER + 1];
-static struct ttm_pool_type global_uncached[MAX_ORDER + 1];
+static struct ttm_pool_type global_write_combined[NR_PAGE_ORDERS];
+static struct ttm_pool_type global_uncached[NR_PAGE_ORDERS];
 
-static struct ttm_pool_type global_dma32_write_combined[MAX_ORDER + 1];
-static struct ttm_pool_type global_dma32_uncached[MAX_ORDER + 1];
+static struct ttm_pool_type global_dma32_write_combined[NR_PAGE_ORDERS];
+static struct ttm_pool_type global_dma32_uncached[NR_PAGE_ORDERS];
 
 static spinlock_t shrinker_lock;
 static struct list_head shrinker_list;

@@ -565,7 +565,7 @@ void ttm_pool_init(struct ttm_pool *pool, struct device *dev,
 
 	if (use_dma_alloc || nid != NUMA_NO_NODE) {
 		for (i = 0; i < TTM_NUM_CACHING_TYPES; ++i)
-			for (j = 0; j <= MAX_ORDER; ++j)
+			for (j = 0; j < NR_PAGE_ORDERS; ++j)
 				ttm_pool_type_init(&pool->caching[i].orders[j],
 						   pool, i, j);
 	}

@@ -586,7 +586,7 @@ void ttm_pool_fini(struct ttm_pool *pool)
 
 	if (pool->use_dma_alloc || pool->nid != NUMA_NO_NODE) {
 		for (i = 0; i < TTM_NUM_CACHING_TYPES; ++i)
-			for (j = 0; j <= MAX_ORDER; ++j)
+			for (j = 0; j < NR_PAGE_ORDERS; ++j)
 				ttm_pool_type_fini(&pool->caching[i].orders[j]);
 	}

@@ -641,7 +641,7 @@ static void ttm_pool_debugfs_header(struct seq_file *m)
 	unsigned int i;
 
 	seq_puts(m, "\t ");
-	for (i = 0; i <= MAX_ORDER; ++i)
+	for (i = 0; i < NR_PAGE_ORDERS; ++i)
 		seq_printf(m, " ---%2u---", i);
 	seq_puts(m, "\n");
 }

@@ -652,7 +652,7 @@ static void ttm_pool_debugfs_orders(struct ttm_pool_type *pt,
 {
 	unsigned int i;
 
-	for (i = 0; i <= MAX_ORDER; ++i)
+	for (i = 0; i < NR_PAGE_ORDERS; ++i)
 		seq_printf(m, " %8u", ttm_pool_type_count(&pt[i]));
 	seq_puts(m, "\n");
 }

@@ -761,7 +761,7 @@ int ttm_pool_mgr_init(unsigned long num_pages)
 	spin_lock_init(&shrinker_lock);
 	INIT_LIST_HEAD(&shrinker_list);
 
-	for (i = 0; i <= MAX_ORDER; ++i) {
+	for (i = 0; i < NR_PAGE_ORDERS; ++i) {
 		ttm_pool_type_init(&global_write_combined[i], NULL,
 				   ttm_write_combined, i);
 		ttm_pool_type_init(&global_uncached[i], NULL, ttm_uncached, i);

@@ -794,7 +794,7 @@ void ttm_pool_mgr_fini(void)
 {
 	unsigned int i;
 
-	for (i = 0; i <= MAX_ORDER; ++i) {
+	for (i = 0; i < NR_PAGE_ORDERS; ++i) {
 		ttm_pool_type_fini(&global_write_combined[i]);
 		ttm_pool_type_fini(&global_uncached[i]);

include/drm/ttm/ttm_pool.h +1 -1

index 30a347e5aa114..4490d43c63e33 100644
@@ -74,7 +74,7 @@ struct ttm_pool {
 	bool use_dma32;
 
 	struct {
-		struct ttm_pool_type orders[MAX_ORDER + 1];
+		struct ttm_pool_type orders[NR_PAGE_ORDERS];
 	} caching[TTM_NUM_CACHING_TYPES];
 };

include/linux/mmzone.h +4 -2

index 0f62786269d0c..1acbc6ce1fe43 100644
@@ -34,6 +34,8 @@
 
 #define IS_MAX_ORDER_ALIGNED(pfn) IS_ALIGNED(pfn, MAX_ORDER_NR_PAGES)
 
+#define NR_PAGE_ORDERS (MAX_ORDER + 1)
+
 /*
  * PAGE_ALLOC_COSTLY_ORDER is the order at which allocations are deemed
  * costly to service.  That is between allocation orders which should

@@ -95,7 +97,7 @@ static inline bool migratetype_is_mergeable(int mt)
 }
 
 #define for_each_migratetype_order(order, type) \
-	for (order = 0; order <= MAX_ORDER; order++) \
+	for (order = 0; order < NR_PAGE_ORDERS; order++) \
 		for (type = 0; type < MIGRATE_TYPES; type++)
 
 extern int page_group_by_mobility_disabled;

@@ -929,7 +931,7 @@ struct zone {
 	CACHELINE_PADDING(_pad1_);
 
 	/* free areas of different sizes */
-	struct free_area	free_area[MAX_ORDER + 1];
+	struct free_area	free_area[NR_PAGE_ORDERS];
 
 #ifdef CONFIG_UNACCEPTED_MEMORY
 	/* Pages to be accepted. All pages on the list are MAX_ORDER */

kernel/crash_core.c +1 -1

index 2f675ef045d40..b685e94605841 100644
@@ -660,7 +660,7 @@ static int __init crash_save_vmcoreinfo_init(void)
 	VMCOREINFO_OFFSET(list_head, prev);
 	VMCOREINFO_OFFSET(vmap_area, va_start);
 	VMCOREINFO_OFFSET(vmap_area, list);
-	VMCOREINFO_LENGTH(zone.free_area, MAX_ORDER + 1);
+	VMCOREINFO_LENGTH(zone.free_area, NR_PAGE_ORDERS);
 	log_buf_vmcoreinfo_setup();
 	VMCOREINFO_LENGTH(free_area.free_list, MIGRATE_TYPES);
 	VMCOREINFO_NUMBER(NR_FREE_PAGES);

lib/test_meminit.c +1 -1

index 0ae35223d7733..0dc173849a542 100644
@@ -93,7 +93,7 @@ static int __init test_pages(int *total_failures)
 	int failures = 0, num_tests = 0;
 	int i;
 
-	for (i = 0; i <= MAX_ORDER; i++)
+	for (i = 0; i < NR_PAGE_ORDERS; i++)
 		num_tests += do_alloc_pages_order(i, &failures);
 
 	REPORT_FAILURES_IN_FN();

mm/compaction.c +1 -1

index 5a3c644c978e2..61c741f11e9bb 100644
@@ -2225,7 +2225,7 @@ static enum compact_result __compact_finished(struct compact_control *cc)
 
 	/* Direct compactor: Is a suitable page free? */
 	ret = COMPACT_NO_SUITABLE_PAGE;
-	for (order = cc->order; order <= MAX_ORDER; order++) {
+	for (order = cc->order; order < NR_PAGE_ORDERS; order++) {
 		struct free_area *area = &cc->zone->free_area[order];
 		bool can_steal;

mm/kmsan/init.c +1 -1

index ffedf4dbc49d7..103e2e88ea033 100644
@@ -96,7 +96,7 @@ void __init kmsan_init_shadow(void)
 struct metadata_page_pair {
 	struct page *shadow, *origin;
 };
-static struct metadata_page_pair held_back[MAX_ORDER + 1] __initdata;
+static struct metadata_page_pair held_back[NR_PAGE_ORDERS] __initdata;
 
 /*
  * Eager metadata allocation. When the memblock allocator is freeing pages to

mm/page_alloc.c +6 -7

index ab71417350127..6b4c30fcae1c9 100644
@@ -1570,7 +1570,7 @@ struct page *__rmqueue_smallest(struct zone *zone, unsigned int order,
 	struct page *page;
 
 	/* Find a page of the appropriate size in the preferred list */
-	for (current_order = order; current_order <= MAX_ORDER; ++current_order) {
+	for (current_order = order; current_order < NR_PAGE_ORDERS; ++current_order) {
 		area = &(zone->free_area[current_order]);
 		page = get_page_from_free_area(area, migratetype);
 		if (!page)

@@ -1940,7 +1940,7 @@ static bool unreserve_highatomic_pageblock(const struct alloc_context *ac,
 			continue;
 
 		spin_lock_irqsave(&zone->lock, flags);
-		for (order = 0; order <= MAX_ORDER; order++) {
+		for (order = 0; order < NR_PAGE_ORDERS; order++) {
 			struct free_area *area = &(zone->free_area[order]);
 
 			page = get_page_from_free_area(area, MIGRATE_HIGHATOMIC);

@@ -2050,8 +2050,7 @@ __rmqueue_fallback(struct zone *zone, int order, int start_migratetype,
 	return false;
 
 find_smallest:
-	for (current_order = order; current_order <= MAX_ORDER;
-							current_order++) {
+	for (current_order = order; current_order < NR_PAGE_ORDERS; current_order++) {
 		area = &(zone->free_area[current_order]);
 		fallback_mt = find_suitable_fallback(area, current_order,
 				start_migratetype, false, &can_steal);

@@ -2884,7 +2883,7 @@ bool __zone_watermark_ok(struct zone *z, unsigned int order, unsigned long mark,
 		return true;
 
 	/* For a high-order request, check at least one suitable page is free */
-	for (o = order; o <= MAX_ORDER; o++) {
+	for (o = order; o < NR_PAGE_ORDERS; o++) {
 		struct free_area *area = &z->free_area[o];
 		int mt;

@@ -6442,7 +6441,7 @@ bool is_free_buddy_page(struct page *page)
 	unsigned long pfn = page_to_pfn(page);
 	unsigned int order;
 
-	for (order = 0; order <= MAX_ORDER; order++) {
+	for (order = 0; order < NR_PAGE_ORDERS; order++) {
 		struct page *page_head = page - (pfn & ((1 << order) - 1));
 
 		if (PageBuddy(page_head) &&

@@ -6501,7 +6500,7 @@ bool take_page_off_buddy(struct page *page)
 	bool ret = false;
 
 	spin_lock_irqsave(&zone->lock, flags);
-	for (order = 0; order <= MAX_ORDER; order++) {
+	for (order = 0; order < NR_PAGE_ORDERS; order++) {
 		struct page *page_head = page - (pfn & ((1 << order) - 1));
 		int page_order = buddy_order(page_head);

mm/page_reporting.c +1 -1

index b021f482a4cb3..66369cc5279bf 100644
@@ -276,7 +276,7 @@ page_reporting_process_zone(struct page_reporting_dev_info *prdev,
 		return err;
 
 	/* Process each free list starting from lowest order/mt */
-	for (order = page_reporting_order; order <= MAX_ORDER; order++) {
+	for (order = page_reporting_order; order < NR_PAGE_ORDERS; order++) {
 		for (mt = 0; mt < MIGRATE_TYPES; mt++) {
 			/* We do not pull pages from the isolate free list */
 			if (is_migrate_isolate(mt))

mm/show_mem.c +4 -4

index 4b888b18bddea..b896e54e3a26c 100644
@@ -355,8 +355,8 @@ static void show_free_areas(unsigned int filter, nodemask_t *nodemask, int max_z
 
 	for_each_populated_zone(zone) {
 		unsigned int order;
-		unsigned long nr[MAX_ORDER + 1], flags, total = 0;
-		unsigned char types[MAX_ORDER + 1];
+		unsigned long nr[NR_PAGE_ORDERS], flags, total = 0;
+		unsigned char types[NR_PAGE_ORDERS];
 
 		if (zone_idx(zone) > max_zone_idx)
 			continue;

@@ -366,7 +366,7 @@ static void show_free_areas(unsigned int filter, nodemask_t *nodemask, int max_z
 		printk(KERN_CONT "%s: ", zone->name);
 
 		spin_lock_irqsave(&zone->lock, flags);
-		for (order = 0; order <= MAX_ORDER; order++) {
+		for (order = 0; order < NR_PAGE_ORDERS; order++) {
 			struct free_area *area = &zone->free_area[order];
 			int type;

@@ -380,7 +380,7 @@ static void show_free_areas(unsigned int filter, nodemask_t *nodemask, int max_z
 			}
 		}
 		spin_unlock_irqrestore(&zone->lock, flags);
-		for (order = 0; order <= MAX_ORDER; order++) {
+		for (order = 0; order < NR_PAGE_ORDERS; order++) {
 			printk(KERN_CONT "%lu*%lukB ",
 			       nr[order], K(1UL) << order);
 			if (nr[order])

mm/vmstat.c +6 -6

index 00e81e99c6ee2..e9616c4ca12db 100644
@@ -1055,7 +1055,7 @@ static void fill_contig_page_info(struct zone *zone,
 	info->free_blocks_total = 0;
 	info->free_blocks_suitable = 0;
 
-	for (order = 0; order <= MAX_ORDER; order++) {
+	for (order = 0; order < NR_PAGE_ORDERS; order++) {
 		unsigned long blocks;
 
 		/*

@@ -1471,7 +1471,7 @@ static void frag_show_print(struct seq_file *m, pg_data_t *pgdat,
 	int order;
 
 	seq_printf(m, "Node %d, zone %8s ", pgdat->node_id, zone->name);
-	for (order = 0; order <= MAX_ORDER; ++order)
+	for (order = 0; order < NR_PAGE_ORDERS; ++order)
 		/*
 		 * Access to nr_free is lockless as nr_free is used only for
 		 * printing purposes. Use data_race to avoid KCSAN warning.

@@ -1500,7 +1500,7 @@ static void pagetypeinfo_showfree_print(struct seq_file *m,
 					pgdat->node_id,
 					zone->name,
 					migratetype_names[mtype]);
-		for (order = 0; order <= MAX_ORDER; ++order) {
+		for (order = 0; order < NR_PAGE_ORDERS; ++order) {
 			unsigned long freecount = 0;
 			struct free_area *area;
 			struct list_head *curr;

@@ -1540,7 +1540,7 @@ static void pagetypeinfo_showfree(struct seq_file *m, void *arg)
 
 	/* Print header */
 	seq_printf(m, "%-43s ", "Free pages count per migrate type at order");
-	for (order = 0; order <= MAX_ORDER; ++order)
+	for (order = 0; order < NR_PAGE_ORDERS; ++order)
 		seq_printf(m, "%6d ", order);
 	seq_putc(m, '\n');

@@ -2176,7 +2176,7 @@ static void unusable_show_print(struct seq_file *m,
 	seq_printf(m, "Node %d, zone %8s ",
 				pgdat->node_id,
 				zone->name);
-	for (order = 0; order <= MAX_ORDER; ++order) {
+	for (order = 0; order < NR_PAGE_ORDERS; ++order) {
 		fill_contig_page_info(zone, order, &info);
 		index = unusable_free_index(order, &info);
 		seq_printf(m, "%d.%03d ", index / 1000, index % 1000);

@@ -2228,7 +2228,7 @@ static void extfrag_show_print(struct seq_file *m,
 	seq_printf(m, "Node %d, zone %8s ",
 				pgdat->node_id,
 				zone->name);
-	for (order = 0; order <= MAX_ORDER; ++order) {
+	for (order = 0; order < NR_PAGE_ORDERS; ++order) {
 		fill_contig_page_info(zone, order, &info);
 		index = __fragmentation_index(order, &info);
 		seq_printf(m, "%2d.%03d ", index / 1000, index % 1000);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:11 UTC #109 lore raw

PATCH 104/186 drm/ttm: stop pooling cached NUMA pages v2

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Christian König <[email protected]>

[ Upstream commit b6976f323a8687cc0d55bc92c2086fd934324ed5 ]

We only pool write combined and uncached allocations because they require extra overhead on allocation and release.

If we also pool cached NUMA it not only means some extra unnecessary overhead, but also that under memory pressure it can happen that pages from the wrong NUMA node enters the pool and are re-used over and over again.

This can lead to performance reduction after running into memory pressure.

v2: restructure and cleanup the code a bit from the internal hack to
test this.

Link https://patchwork.freedesktop.org/patch/msgid/[email protected]

Fixes 4482d3c94d7f ("drm/ttm: add NUMA node id to the pool")

CC [email protected]

Reviewed-by Felix Kuehling <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 drivers/gpu/drm/ttm/ttm_pool.c | 38 +++++++++++++++++++++++++---------
 1 file changed, 28 insertions(+), 10 deletions(-)

drivers/gpu/drm/ttm/ttm_pool.c +28 ~~-10~~

index c8ec6a2cac5d4..37c08fac7e7d0 100644
@@ -287,17 +287,23 @@ static struct ttm_pool_type *ttm_pool_select_type(struct ttm_pool *pool,
 						  enum ttm_caching caching,
 						  unsigned int order)
 {
-	if (pool->use_dma_alloc || pool->nid != NUMA_NO_NODE)
+	if (pool->use_dma_alloc)
 		return &pool->caching[caching].orders[order];
 
 #ifdef CONFIG_X86
 	switch (caching) {
 	case ttm_write_combined:
+		if (pool->nid != NUMA_NO_NODE)
+			return &pool->caching[caching].orders[order];
+
 		if (pool->use_dma32)
 			return &global_dma32_write_combined[order];
 
 		return &global_write_combined[order];
 	case ttm_uncached:
+		if (pool->nid != NUMA_NO_NODE)
+			return &pool->caching[caching].orders[order];
+
 		if (pool->use_dma32)
 			return &global_dma32_uncached[order];

@@ -563,11 +569,17 @@ void ttm_pool_init(struct ttm_pool *pool, struct device *dev,
 	pool->use_dma_alloc = use_dma_alloc;
 	pool->use_dma32 = use_dma32;
 
-	if (use_dma_alloc || nid != NUMA_NO_NODE) {
-		for (i = 0; i < TTM_NUM_CACHING_TYPES; ++i)
-			for (j = 0; j < NR_PAGE_ORDERS; ++j)
-				ttm_pool_type_init(&pool->caching[i].orders[j],
-						   pool, i, j);
+	for (i = 0; i < TTM_NUM_CACHING_TYPES; ++i) {
+		for (j = 0; j < NR_PAGE_ORDERS; ++j) {
+			struct ttm_pool_type *pt;
+
+			/* Initialize only pool types which are actually used */
+			pt = ttm_pool_select_type(pool, i, j);
+			if (pt != &pool->caching[i].orders[j])
+				continue;
+
+			ttm_pool_type_init(pt, pool, i, j);
+		}
 	}
 }
 EXPORT_SYMBOL(ttm_pool_init);

@@ -584,10 +596,16 @@ void ttm_pool_fini(struct ttm_pool *pool)
 {
 	unsigned int i, j;
 
-	if (pool->use_dma_alloc || pool->nid != NUMA_NO_NODE) {
-		for (i = 0; i < TTM_NUM_CACHING_TYPES; ++i)
-			for (j = 0; j < NR_PAGE_ORDERS; ++j)
-				ttm_pool_type_fini(&pool->caching[i].orders[j]);
+	for (i = 0; i < TTM_NUM_CACHING_TYPES; ++i) {
+		for (j = 0; j < NR_PAGE_ORDERS; ++j) {
+			struct ttm_pool_type *pt;
+
+			pt = ttm_pool_select_type(pool, i, j);
+			if (pt != &pool->caching[i].orders[j])
+				continue;
+
+			ttm_pool_type_fini(pt);
+		}
 	}
 
 	/* We removed the pool types from the LRU, but we need to also make sure

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:11 UTC #110 lore raw

PATCH 105/186 squashfs: convert to new timestamp accessors

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jeff Layton <[email protected]>

[ Upstream commit a1f13ed8c74893ed31d41c5bca156a623b0e9a86 ]

Convert to using the new inode timestamp accessor functions.

Signed-off-by Jeff Layton <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Christian Brauner <[email protected]>

Stable-dep-of: 9253c54e01b6 ("Squashfs: check the inode number is not the invalid value of zero")

Signed-off-by Sasha Levin <[email protected]>

---
 fs/squashfs/inode.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

fs/squashfs/inode.c +3 -3

index c6e626b00546b..aa3411354e66d 100644
@@ -59,9 +59,9 @@ static int squashfs_new_inode(struct super_block *sb, struct inode *inode,
 	i_uid_write(inode, i_uid);
 	i_gid_write(inode, i_gid);
 	inode->i_ino = le32_to_cpu(sqsh_ino->inode_number);
-	inode->i_mtime.tv_sec = le32_to_cpu(sqsh_ino->mtime);
-	inode->i_atime.tv_sec = inode->i_mtime.tv_sec;
-	inode_set_ctime(inode, inode->i_mtime.tv_sec, 0);
+	inode_set_mtime(inode, le32_to_cpu(sqsh_ino->mtime), 0);
+	inode_set_atime(inode, inode_get_mtime_sec(inode), 0);
+	inode_set_ctime(inode, inode_get_mtime_sec(inode), 0);
 	inode->i_mode = le16_to_cpu(sqsh_ino->mode);
 	inode->i_size = 0;

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:11 UTC #111 lore raw

PATCH 106/186 Squashfs: check the inode number is not the invalid value of zero

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Phillip Lougher <[email protected]>

[ Upstream commit 9253c54e01b6505d348afbc02abaa4d9f8a01395 ]

Syskiller has produced an out of bounds access in fill_meta_index().

That out of bounds access is ultimately caused because the inode has an inode number with the invalid value of zero, which was not checked.

The reason this causes the out of bounds access is due to following sequence of events:

1. Fill_meta_index() is called to allocate (via empty_meta_index()) and fill a metadata index. It however suffers a data read error and aborts, invalidating the newly returned empty metadata index. It does this by setting the inode number of the index to zero, which means unused (zero is not a valid inode number).

2. When fill_meta_index() is subsequently called again on another read operation, locate_meta_index() returns the previous index because it matches the inode number of 0. Because this index has been returned it is expected to have been filled, and because it hasn't been, an out of bounds access is performed.

This patch adds a sanity check which checks that the inode number is not zero when the inode is created and returns -EINVAL if it is.

[[email protected]: whitespace fix]

Link https://lkml.kernel.org/r/[email protected] open here

Signed-off-by Phillip Lougher <[email protected]>

Reported-by "Ubisectech Sirius" <[email protected]>

Closes https://lore.kernel.org/lkml/[email protected]/ open here

Cc Christian Brauner <[email protected]>

Signed-off-by Sasha Levin <[email protected]>

---
 fs/squashfs/inode.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

fs/squashfs/inode.c +4 -1

index aa3411354e66d..16bd693d0b3aa 100644
@@ -48,6 +48,10 @@ static int squashfs_new_inode(struct super_block *sb, struct inode *inode,
 	gid_t i_gid;
 	int err;
 
+	inode->i_ino = le32_to_cpu(sqsh_ino->inode_number);
+	if (inode->i_ino == 0)
+		return -EINVAL;
+
 	err = squashfs_get_id(sb, le16_to_cpu(sqsh_ino->uid), &i_uid);
 	if (err)
 		return err;

@@ -58,7 +62,6 @@ static int squashfs_new_inode(struct super_block *sb, struct inode *inode,
 
 	i_uid_write(inode, i_uid);
 	i_gid_write(inode, i_gid);
-	inode->i_ino = le32_to_cpu(sqsh_ino->inode_number);
 	inode_set_mtime(inode, le32_to_cpu(sqsh_ino->mtime), 0);
 	inode_set_atime(inode, inode_get_mtime_sec(inode), 0);
 	inode_set_ctime(inode, inode_get_mtime_sec(inode), 0);

Signature · 1 line

2.43.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:11 UTC #112 lore raw

PATCH 107/186 selftests/seccomp: user_notification_addfd check nextfd is available

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Terry Tritton <[email protected]>

commit 8e3c9f9f3a0742cd12b682a1766674253b33fcf0 upstream.

Currently the user_notification_addfd test checks what the next expected file descriptor will be by incrementing a variable nextfd. This does not account for file descriptors that may already be open before the test is started and will cause the test to fail if any exist.

Replace nextfd++ with a function get_next_fd which will check and return the next available file descriptor.

Signed-off-by Terry Tritton <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Kees Cook <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 tools/testing/selftests/seccomp/seccomp_bpf.c |   24 +++++++++++++++++++-----
 1 file changed, 19 insertions(+), 5 deletions(-)

tools/testing/selftests/seccomp/seccomp_bpf.c +19 -5

@@ -4037,6 +4037,16 @@ TEST(user_notification_filter_empty_thre
 	EXPECT_GT((pollfd.revents & POLLHUP) ?: 0, 0);
 }
 
+
+int get_next_fd(int prev_fd)
+{
+	for (int i = prev_fd + 1; i < FD_SETSIZE; ++i) {
+		if (fcntl(i, F_GETFD) == -1)
+			return i;
+	}
+	_exit(EXIT_FAILURE);
+}
+
 TEST(user_notification_addfd)
 {
 	pid_t pid;

@@ -4053,7 +4063,7 @@ TEST(user_notification_addfd)
 	/* There may be arbitrary already-open fds at test start. */
 	memfd = memfd_create("test", 0);
 	ASSERT_GE(memfd, 0);
-	nextfd = memfd + 1;
+	nextfd = get_next_fd(memfd);
 
 	ret = prctl(PR_SET_NO_NEW_PRIVS, 1, 0, 0, 0);
 	ASSERT_EQ(0, ret) {

@@ -4064,7 +4074,8 @@ TEST(user_notification_addfd)
 	/* Check that the basic notification machinery works */
 	listener = user_notif_syscall(__NR_getppid,
 				      SECCOMP_FILTER_FLAG_NEW_LISTENER);
-	ASSERT_EQ(listener, nextfd++);
+	ASSERT_EQ(listener, nextfd);
+	nextfd = get_next_fd(nextfd);
 
 	pid = fork();
 	ASSERT_GE(pid, 0);

@@ -4119,14 +4130,16 @@ TEST(user_notification_addfd)
 
 	/* Verify we can set an arbitrary remote fd */
 	fd = ioctl(listener, SECCOMP_IOCTL_NOTIF_ADDFD, &addfd);
-	EXPECT_EQ(fd, nextfd++);
+	EXPECT_EQ(fd, nextfd);
+	nextfd = get_next_fd(nextfd);
 	EXPECT_EQ(filecmp(getpid(), pid, memfd, fd), 0);
 
 	/* Verify we can set an arbitrary remote fd with large size */
 	memset(&big, 0x0, sizeof(big));
 	big.addfd = addfd;
 	fd = ioctl(listener, SECCOMP_IOCTL_NOTIF_ADDFD_BIG, &big);
-	EXPECT_EQ(fd, nextfd++);
+	EXPECT_EQ(fd, nextfd);
+	nextfd = get_next_fd(nextfd);
 
 	/* Verify we can set a specific remote fd */
 	addfd.newfd = 42;

@@ -4164,7 +4177,8 @@ TEST(user_notification_addfd)
 	 * Child has earlier "low" fds and now 42, so we expect the next
 	 * lowest available fd to be assigned here.
 	 */
-	EXPECT_EQ(fd, nextfd++);
+	EXPECT_EQ(fd, nextfd);
+	nextfd = get_next_fd(nextfd);
 	ASSERT_EQ(filecmp(getpid(), pid, memfd, fd), 0);
 
 	/*

Greg Kroah-Hartman

author maintainer

2024-04-30 11:11 UTC #113 lore raw

PATCH 108/186 selftests/seccomp: Change the syscall used in KILL_THREAD test

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Terry Tritton <[email protected]>

commit 471dbc547612adeaa769e48498ef591c6c95a57a upstream.

The Bionic version of pthread_create used on Android calls the prctl function to give the stack and thread local storage a useful name. This will cause the KILL_THREAD test to fail as it will kill the thread as soon as it is created.

change the test to use getpid instead of prctl.

Signed-off-by Terry Tritton <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Kees Cook <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 tools/testing/selftests/seccomp/seccomp_bpf.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

tools/testing/selftests/seccomp/seccomp_bpf.c +4 -4

@@ -784,7 +784,7 @@ void *kill_thread(void *data)
 	bool die = (bool)data;
 
 	if (die) {
-		prctl(PR_GET_SECCOMP, 0, 0, 0, 0);
+		syscall(__NR_getpid);
 		return (void *)SIBLING_EXIT_FAILURE;
 	}

@@ -803,11 +803,11 @@ void kill_thread_or_group(struct __test_
 {
 	pthread_t thread;
 	void *status;
-	/* Kill only when calling __NR_prctl. */
+	/* Kill only when calling __NR_getpid. */
 	struct sock_filter filter_thread[] = {
 		BPF_STMT(BPF_LD|BPF_W|BPF_ABS,
 			offsetof(struct seccomp_data, nr)),
-		BPF_JUMP(BPF_JMP|BPF_JEQ|BPF_K, __NR_prctl, 0, 1),
+		BPF_JUMP(BPF_JMP|BPF_JEQ|BPF_K, __NR_getpid, 0, 1),
 		BPF_STMT(BPF_RET|BPF_K, SECCOMP_RET_KILL_THREAD),
 		BPF_STMT(BPF_RET|BPF_K, SECCOMP_RET_ALLOW),
 	};

@@ -819,7 +819,7 @@ void kill_thread_or_group(struct __test_
 	struct sock_filter filter_process[] = {
 		BPF_STMT(BPF_LD|BPF_W|BPF_ABS,
 			offsetof(struct seccomp_data, nr)),
-		BPF_JUMP(BPF_JMP|BPF_JEQ|BPF_K, __NR_prctl, 0, 1),
+		BPF_JUMP(BPF_JMP|BPF_JEQ|BPF_K, __NR_getpid, 0, 1),
 		BPF_STMT(BPF_RET|BPF_K, kill),
 		BPF_STMT(BPF_RET|BPF_K, SECCOMP_RET_ALLOW),
 	};

Greg Kroah-Hartman

author maintainer

2024-04-30 11:11 UTC #114 lore raw

PATCH 109/186 selftests/seccomp: Handle EINVAL on unshare(CLONE_NEWPID)

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Terry Tritton <[email protected]>

commit ecaaa55c9fa5e8058445a8b891070b12208cdb6d upstream.

unshare(CLONE_NEWPID) can return EINVAL if the kernel does not have the CONFIG_PID_NS option enabled.

Add a check on these calls to skip the test if we receive EINVAL.

Signed-off-by Terry Tritton <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Kees Cook <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 tools/testing/selftests/seccomp/seccomp_bpf.c |    9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

tools/testing/selftests/seccomp/seccomp_bpf.c +8 -1

@@ -3709,7 +3709,12 @@ TEST(user_notification_sibling_pid_ns)
 	ASSERT_GE(pid, 0);
 
 	if (pid == 0) {
-		ASSERT_EQ(unshare(CLONE_NEWPID), 0);
+		ASSERT_EQ(unshare(CLONE_NEWPID), 0) {
+			if (errno == EPERM)
+				SKIP(return, "CLONE_NEWPID requires CAP_SYS_ADMIN");
+			else if (errno == EINVAL)
+				SKIP(return, "CLONE_NEWPID is invalid (missing CONFIG_PID_NS?)");
+		}
 
 		pid2 = fork();
 		ASSERT_GE(pid2, 0);

@@ -3727,6 +3732,8 @@ TEST(user_notification_sibling_pid_ns)
 	ASSERT_EQ(unshare(CLONE_NEWPID), 0) {
 		if (errno == EPERM)
 			SKIP(return, "CLONE_NEWPID requires CAP_SYS_ADMIN");
+		else if (errno == EINVAL)
+			SKIP(return, "CLONE_NEWPID is invalid (missing CONFIG_PID_NS?)");
 	}
 	ASSERT_EQ(errno, 0);

Greg Kroah-Hartman

author maintainer

2024-04-30 11:11 UTC #115 lore raw

PATCH 110/186 fork: defer linking file vma until vma is fully initialized

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Miaohe Lin <[email protected]>

commit 35e351780fa9d8240dd6f7e4f245f9ea37e96c19 upstream.

Thorvald reported a WARNING [1]. And the root cause is below race:

CPU 1 CPU 2
fork hugetlbfs_fallocate
dup_mmap hugetlbfs_punch_hole
i_mmap_lock_write(mapping);
vma_interval_tree_insert_after -- Child vma is visible through i_mmap tree. i_mmap_unlock_write(mapping);
hugetlb_dup_vma_private -- Clear vma_lock outside i_mmap_rwsem!

					 i_mmap_lock_write(mapping);
   					 hugetlb_vmdelete_list
					  vma_interval_tree_foreach
					   hugetlb_vma_trylock_write -- Vma_lock is cleared.
   tmp->vm_ops->open -- Alloc new vma_lock outside i_mmap_rwsem!
					   hugetlb_vma_unlock_write -- Vma_lock is assigned!!!
					 i_mmap_unlock_write(mapping);

hugetlb_dup_vma_private() and hugetlb_vm_op_open() are called outside i_mmap_rwsem lock while vma lock can be used in the same time. Fix this by deferring linking file vma until vma is fully initialized. Those vmas should be initialized first before they can be used.

Link https://lkml.kernel.org/r/[email protected] open here

Fixes 8d9bfb260814 ("hugetlb: add vma based lock for pmd sharing")

Signed-off-by Miaohe Lin <[email protected]>

Reported-by Thorvald Natvig <[email protected]>

Closes https://lore.kernel.org/linux-mm/20240129161735.6gmjsswx62o4pbja@revolver/T/ [1] open here

Reviewed-by Jane Chu <[email protected]>

Cc Christian Brauner <[email protected]>

Cc Heiko Carstens <[email protected]>

Cc Kent Overstreet <[email protected]>

Cc Liam R. Howlett <[email protected]>

Cc Mateusz Guzik <[email protected]>

Cc Matthew Wilcox (Oracle) <[email protected]>

Cc Miaohe Lin <[email protected]>

Cc Muchun Song <[email protected]>

Cc Oleg Nesterov <[email protected]>

Cc Peng Zhang <[email protected]>

Cc Tycho Andersen <[email protected]>

Signed-off-by Miaohe Lin <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 kernel/fork.c |   18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

kernel/fork.c +9 -9

@@ -727,6 +727,15 @@ static __latent_entropy int dup_mmap(str
 		} else if (anon_vma_fork(tmp, mpnt))
 			goto fail_nomem_anon_vma_fork;
 		vm_flags_clear(tmp, VM_LOCKED_MASK);
+		/*
+		 * Copy/update hugetlb private vma information.
+		 */
+		if (is_vm_hugetlb_page(tmp))
+			hugetlb_dup_vma_private(tmp);
+
+		if (tmp->vm_ops && tmp->vm_ops->open)
+			tmp->vm_ops->open(tmp);
+
 		file = tmp->vm_file;
 		if (file) {
 			struct address_space *mapping = file->f_mapping;

@@ -743,12 +752,6 @@ static __latent_entropy int dup_mmap(str
 			i_mmap_unlock_write(mapping);
 		}
 
-		/*
-		 * Copy/update hugetlb private vma information.
-		 */
-		if (is_vm_hugetlb_page(tmp))
-			hugetlb_dup_vma_private(tmp);
-
 		/* Link the vma into the MT */
 		if (vma_iter_bulk_store(&vmi, tmp))
 			goto fail_nomem_vmi_store;

@@ -757,9 +760,6 @@ static __latent_entropy int dup_mmap(str
 		if (!(tmp->vm_flags & VM_WIPEONFORK))
 			retval = copy_page_range(tmp, mpnt);
 
-		if (tmp->vm_ops && tmp->vm_ops->open)
-			tmp->vm_ops->open(tmp);
-
 		if (retval)
 			goto loop_out;
 	}

Greg Kroah-Hartman

author maintainer

2024-04-30 11:11 UTC #116 lore raw

PATCH 111/186 x86/cpu: Fix check for RDPKRU in __show_regs()

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: David Kaplan <[email protected]>

commit b53c6bd5d271d023857174b8fd3e32f98ae51372 upstream.

cpu_feature_enabled(X86_FEATURE_OSPKE) does not necessarily reflect whether CR4.PKE is set on the CPU. In particular, they may differ on non-BSP CPUs before setup_pku() is executed. In this scenario, RDPKRU will #UD causing the system to hang.

Fix by checking CR4 for PKE enablement which is always correct for the current CPU.

The scenario happens by inserting a WARN* before setup_pku() in identiy_cpu() or some other diagnostic which would lead to calling __show_regs().

[ bp: Massage commit message. ]

Signed-off-by David Kaplan <[email protected]>

Signed-off-by Borislav Petkov (AMD) <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 arch/x86/kernel/process_64.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

arch/x86/kernel/process_64.c +1 -1

@@ -138,7 +138,7 @@ void __show_regs(struct pt_regs *regs, e
 		       log_lvl, d3, d6, d7);
 	}
 
-	if (cpu_feature_enabled(X86_FEATURE_OSPKE))
+	if (cr4 & X86_CR4_PKE)
 		printk("%sPKRU: %08x\n", log_lvl, read_pkru());
 }

Greg Kroah-Hartman

author maintainer

2024-04-30 11:11 UTC #117 lore raw

PATCH 112/186 rust: dont select CONSTRUCTORS

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alice Ryhl <[email protected]>

commit 7d49f53af4b988b188d3932deac2c9c80fd7d9ce upstream.

This was originally part of commit 4b9a68f2e59a0 ("rust: add support for static synchronisation primitives") from the old Rust branch, which used module constructors to initialize globals containing various synchronisation primitives with pin-init. That commit has never been upstreamed, but the select CONSTRUCTORS statement ended up being included in the patch that initially added Rust support to the Linux Kernel.

We are not using module constructors, so let's remove the select.

Signed-off-by Alice Ryhl <[email protected]>

Reviewed-by Benno Lossin <[email protected]>

Fixes 2f7ab1267dc9 ("Kbuild: add Rust support")

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 init/Kconfig |    1 -
 1 file changed, 1 deletion(-)

init/Kconfig +0 -1

@@ -1898,7 +1898,6 @@ config RUST
 	depends on !GCC_PLUGINS
 	depends on !RANDSTRUCT
 	depends on !DEBUG_INFO_BTF || PAHOLE_HAS_LANG_EXCLUDE
-	select CONSTRUCTORS
 	help
 	  Enables Rust support in the kernel.

Greg Kroah-Hartman

author maintainer

2024-04-30 11:11 UTC #118 lore raw

PATCH 113/186 rust: init: remove impl Zeroable for Infallible

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Laine Taffin Altman <[email protected]>

commit 49ceae68a0df9a92617a61e9ce8a0efcf6419585 upstream.

In Rust, producing an invalid value of any type is immediate undefined behavior (UB); this includes via zeroing memory. Therefore, since an uninhabited type has no valid values, producing any values at all for it is UB.

The Rust standard library type core::convert::Infallible is uninhabited, by virtue of having been declared as an enum with no cases, which always produces uninhabited types in Rust.

The current kernel code allows this UB to be triggered, for example by code like Box::<core::convert::Infallible>::init(kernel::init::zeroed()).

Thus, remove the implementation of Zeroable for Infallible, thereby avoiding the unsoundness (potential for future UB).

Fixes 38cde0bd7b67 ("rust: init: add `Zeroable` trait and `init::zeroed` function")

Closes https://github.com/Rust-for-Linux/pinned-init/pull/13

Signed-off-by Laine Taffin Altman <[email protected]>

Reviewed-by Boqun Feng <[email protected]>

Reviewed-by Benno Lossin <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

[ Reformatted the comment slightly. ]

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 rust/kernel/init.rs |   11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

rust/kernel/init.rs +9 -2

@@ -1292,8 +1292,15 @@ impl_zeroable! {
     i8, i16, i32, i64, i128, isize,
     f32, f64,
 
-    // SAFETY: These are ZSTs, there is nothing to zero.
-    {<T: ?Sized>} PhantomData<T>, core::marker::PhantomPinned, Infallible, (),
+    // Note: do not add uninhabited types (such as `!` or `core::convert::Infallible`) to this list;
+    // creating an instance of an uninhabited type is immediate undefined behavior. For more on
+    // uninhabited/empty types, consult The Rustonomicon:
+    // <https://doc.rust-lang.org/stable/nomicon/exotic-sizes.html#empty-types>. The Rust Reference
+    // also has information on undefined behavior:
+    // <https://doc.rust-lang.org/stable/reference/behavior-considered-undefined.html>.
+    //
+    // SAFETY: These are inhabited ZSTs; there is nothing to zero and a valid value exists.
+    {<T: ?Sized>} PhantomData<T>, core::marker::PhantomPinned, (),
 
     // SAFETY: Type is allowed to take any value, including all zeros.
     {<T>} MaybeUninit<T>,

Greg Kroah-Hartman

author maintainer

2024-04-30 11:11 UTC #119 lore raw

PATCH 114/186 rust: make mutually exclusive with CFI_CLANG

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Conor Dooley <[email protected]>

commit 8933cf4651e02853ca679be7b2d978dfcdcc5e0c upstream.

On RISC-V and arm64, and presumably x86, if CFI_CLANG is enabled, loading a rust module will trigger a kernel panic. Support for sanitisers, including kcfi (CFI_CLANG), is in the works, but for now they're nightly-only options in rustc. Make RUST depend on !CFI_CLANG to prevent configuring a kernel without symmetrical support for kfi.

[ Matthew Maurer writes [1]:

This patch is fine by me - the last patch needed for KCFI to be functional in Rust just landed upstream last night, so we should revisit this (in the form of enabling it) once we move to rustc-1.79.0 or later.

Ramon de C Valle also gave feedback [2] on the status of KCFI for
Rust and created a tracking issue [3] in upstream Rust. - Miguel ]

Fixes 2f7ab1267dc9 ("Kbuild: add Rust support")

Link https://lore.kernel.org/rust-for-linux/CAGSQo024u1gHJgzsO38Xg3c4or+JupoPABQx_+0BLEpPg0cOEA@mail.gmail.com/ [1] open here

Signed-off-by Conor Dooley <[email protected]>

Acked-by Nathan Chancellor <[email protected]>

Link https://lore.kernel.org/rust-for-linux/CAOcBZOS2kPyH0Dm7Fuh4GC3=v7nZhyzBj_-dKu3PfAnrHZvaxg@mail.gmail.com/ [2] open here

Link https://github.com/rust-lang/rust/issues/123479 [3]

Link https://lore.kernel.org/r/20240404-providing-emporium-e652e359c711@spud open here

[ Added feedback from the list, links, and used Cc for the tag. ]

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 init/Kconfig |    1 +
 1 file changed, 1 insertion(+)

init/Kconfig +1 -0

@@ -1894,6 +1894,7 @@ config RUST
 	bool "Rust support"
 	depends on HAVE_RUST
 	depends on RUST_IS_AVAILABLE
+	depends on !CFI_CLANG
 	depends on !MODVERSIONS
 	depends on !GCC_PLUGINS
 	depends on !RANDSTRUCT

Greg Kroah-Hartman

author maintainer

2024-04-30 11:11 UTC #120 lore raw

PATCH 115/186 kbuild: rust: remove unneeded `@rustc_cfg` to avoid ICE

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Miguel Ojeda <[email protected]>

commit 50cfe93b01475ba36878b65d35d812e1bb48ac71 upstream.

When KUnit tests are enabled, under very big kernel configurations (e.g. allyesconfig), we can trigger a rustdoc ICE [1]:

RUSTDOC TK rust/kernel/lib.rs
error: the compiler unexpectedly panicked. this is a bug.

The reason is that this build step has a duplicated @rustc_cfg argument, which contains the kernel configuration, and thus a lot of arguments. The factor 2 happens to be enough to reach the ICE.

Thus remove the unneeded @rustc_cfg. By doing so, we clean up the command and workaround the ICE.

The ICE has been fixed in the upcoming Rust 1.79 [2].

Fixes a66d733da801 ("rust: support running Rust documentation tests as KUnit ones")

Link https://github.com/rust-lang/rust/issues/122722 [1]

Link https://github.com/rust-lang/rust/pull/122840 [2]

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 rust/Makefile |    1 -
 1 file changed, 1 deletion(-)

rust/Makefile +0 -1

@@ -173,7 +173,6 @@ quiet_cmd_rustdoc_test_kernel = RUSTDOC
 	mkdir -p $(objtree)/$(obj)/test/doctests/kernel; \
 	OBJTREE=$(abspath $(objtree)) \
 	$(RUSTDOC) --test $(rust_flags) \
-		@$(objtree)/include/generated/rustc_cfg \
 		-L$(objtree)/$(obj) --extern alloc --extern kernel \
 		--extern build_error --extern macros \
 		--extern bindings --extern uapi \

Greg Kroah-Hartman

author maintainer

2024-04-30 11:14 UTC #121 lore raw

PATCH 116/186 kbuild: rust: force `alloc` extern to allow "empty" Rust files

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Miguel Ojeda <[email protected]>

commit ded103c7eb23753f22597afa500a7c1ad34116ba upstream.

If one attempts to build an essentially empty file somewhere in the kernel tree, it leads to a build error because the compiler does not recognize the new_uninit unstable feature:

    error[E0635]: unknown feature `new_uninit`
     --> <crate attribute>:1:9
      |
    1 | feature(new_uninit)
      |         ^^^^^^^^^^

The reason is that we pass -Zcrate-attr='feature(new_uninit)' (together with -Zallow-features=new_uninit) to let non-rust/ code use that unstable feature.

However, the compiler only recognizes the feature if the alloc crate is resolved (the feature is an alloc one). --extern alloc, which we pass, is not enough to resolve the crate.

Introducing a reference like use alloc; or extern crate alloc; solves the issue, thus this is not seen in normal files. For instance, useing the kernel prelude introduces such a reference, since alloc is used inside.

While normal use of the build system is not impacted by this, it can still be fairly confusing for kernel developers [1], thus use the unstable force option of --extern [2] (added in Rust 1.71 [3]) to force the compiler to resolve alloc.

This new unstable feature is only needed meanwhile we use the other unstable feature, since then we will not need -Zcrate-attr.

Cc [email protected] # v6.6+

Reported-by Daniel Almeida <[email protected]>

Reported-by Julian Stecklina <[email protected]>

Closes https://rust-for-linux.zulipchat.com/#narrow/stream/288089-General/topic/x/near/424096982 [1]

Fixes 2f7ab1267dc9 ("Kbuild: add Rust support")

Link https://github.com/rust-lang/rust/issues/111302 [2]

Link https://github.com/rust-lang/rust/pull/109421 [3]

Reviewed-by Gary Guo <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 scripts/Makefile.build |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

scripts/Makefile.build +1 -1

@@ -272,7 +272,7 @@ rust_common_cmd = \
 	-Zallow-features=$(rust_allowed_features) \
 	-Zcrate-attr=no_std \
 	-Zcrate-attr='feature($(rust_allowed_features))' \
-	--extern alloc --extern kernel \
+	-Zunstable-options --extern force:alloc --extern kernel \
 	--crate-type rlib -L $(objtree)/rust/ \
 	--crate-name $(basename $(notdir $@)) \
 	--out-dir $(dir $@) --emit=dep-info=$(depfile)

Greg Kroah-Hartman

author maintainer

2024-04-30 11:12 UTC #122 lore raw

PATCH 117/186 rust: remove `params` from `module` macro example

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Aswin Unnikrishnan <[email protected]>

commit 19843452dca40e28d6d3f4793d998b681d505c7f upstream.

Remove argument params from the module macro example, because the macro does not currently support module parameters since it was not sent with the initial merge.

Signed-off-by Aswin Unnikrishnan <[email protected]>

Fixes 1fbde52bde73 ("rust: add `macros` crate")

Link https://lore.kernel.org/r/[email protected] open here

[ Reworded slightly. ]

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 rust/macros/lib.rs |   12 ------------
 1 file changed, 12 deletions(-)

rust/macros/lib.rs +0 ~~-12~~

@@ -35,18 +35,6 @@ use proc_macro::TokenStream;
 ///     author: "Rust for Linux Contributors",
 ///     description: "My very own kernel module!",
 ///     license: "GPL",
-///     params: {
-///        my_i32: i32 {
-///            default: 42,
-///            permissions: 0o000,
-///            description: "Example of i32",
-///        },
-///        writeable_i32: i32 {
-///            default: 42,
-///            permissions: 0o644,
-///            description: "Example of i32",
-///        },
-///    },
 /// }
 ///
 /// struct MyModule;

Greg Kroah-Hartman

author maintainer

2024-04-30 11:12 UTC #123 lore raw

PATCH 118/186 Bluetooth: Fix type of len in {l2cap,sco}_sock_getsockopt_old()

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Nathan Chancellor <[email protected]>

commit 9bf4e919ccad613b3596eebf1ff37b05b6405307 upstream.

After an innocuous optimization change in LLVM main (19.0.0), x86_64 allmodconfig (which enables CONFIG_KCSAN / -fsanitize=thread) fails to build due to the checks in check_copy_size():

In file included from net/bluetooth/sco.c:27:
In file included from include/linux/module.h:13:
In file included from include/linux/stat.h:19:
In file included from include/linux/time.h:60:
In file included from include/linux/time32.h:13:
In file included from include/linux/timex.h:67:
In file included from arch/x86/include/asm/timex.h:6:
In file included from arch/x86/include/asm/tsc.h:10:
In file included from arch/x86/include/asm/msr.h:15:
In file included from include/linux/percpu.h:7:
In file included from include/linux/smp.h:118:
include/linux/thread_info.h:244:4: error: call to '__bad_copy_from' declared with 'error' attribute: copy source size is too small

    244 |                         __bad_copy_from();
        |                         ^

The same exact error occurs in l2cap_sock.c. The copy_to_user() statements that are failing come from l2cap_sock_getsockopt_old() and sco_sock_getsockopt_old(). This does not occur with GCC with or without KCSAN or Clang without KCSAN enabled.

len is defined as an 'int' because it is assigned from '__user int *optlen'. However, it is clamped against the result of sizeof(), which has a type of 'size_t' ('unsigned long' for 64-bit platforms). This is done with min_t() because min() requires compatible types, which results in both len and the result of sizeof() being casted to 'unsigned int', meaning len changes signs and the result of sizeof() is truncated. From there, len is passed to copy_to_user(), which has a third parameter type of 'unsigned long', so it is widened and changes signs again. This excessive casting in combination with the KCSAN instrumentation causes LLVM to fail to eliminate the __bad_copy_from() call, failing the build.

The official recommendation from LLVM developers is to consistently use long types for all size variables to avoid the unnecessary casting in the first place. Change the type of len to size_t in both l2cap_sock_getsockopt_old() and sco_sock_getsockopt_old(). This clears up the error while allowing min_t() to be replaced with min(), resulting in simpler code with no casts and fewer implicit conversions. While len is a different type than optlen now, it should result in no functional change because the result of sizeof() will clamp all values of optlen in the same manner as before.

Link https://github.com/llvm/llvm-project/issues/85647

Closes https://github.com/ClangBuiltLinux/linux/issues/2007

Signed-off-by Nathan Chancellor <[email protected]>

Reviewed-by Justin Stitt <[email protected]>

Signed-off-by Luiz Augusto von Dentz <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 net/bluetooth/l2cap_sock.c |    7 ++++---
 net/bluetooth/sco.c        |    7 ++++---
 2 files changed, 8 insertions(+), 6 deletions(-)

net/bluetooth/l2cap_sock.c +4 -3

@@ -438,7 +438,8 @@ static int l2cap_sock_getsockopt_old(str
 	struct l2cap_chan *chan = l2cap_pi(sk)->chan;
 	struct l2cap_options opts;
 	struct l2cap_conninfo cinfo;
-	int len, err = 0;
+	int err = 0;
+	size_t len;
 	u32 opt;
 
 	BT_DBG("sk %p", sk);

@@ -485,7 +486,7 @@ static int l2cap_sock_getsockopt_old(str
 
 		BT_DBG("mode 0x%2.2x", chan->mode);
 
-		len = min_t(unsigned int, len, sizeof(opts));
+		len = min(len, sizeof(opts));
 		if (copy_to_user(optval, (char *) &opts, len))
 			err = -EFAULT;

@@ -535,7 +536,7 @@ static int l2cap_sock_getsockopt_old(str
 		cinfo.hci_handle = chan->conn->hcon->handle;
 		memcpy(cinfo.dev_class, chan->conn->hcon->dev_class, 3);
 
-		len = min_t(unsigned int, len, sizeof(cinfo));
+		len = min(len, sizeof(cinfo));
 		if (copy_to_user(optval, (char *) &cinfo, len))
 			err = -EFAULT;

net/bluetooth/sco.c +4 -3

@@ -963,7 +963,8 @@ static int sco_sock_getsockopt_old(struc
 	struct sock *sk = sock->sk;
 	struct sco_options opts;
 	struct sco_conninfo cinfo;
-	int len, err = 0;
+	int err = 0;
+	size_t len;
 
 	BT_DBG("sk %p", sk);

@@ -985,7 +986,7 @@ static int sco_sock_getsockopt_old(struc
 
 		BT_DBG("mtu %u", opts.mtu);
 
-		len = min_t(unsigned int, len, sizeof(opts));
+		len = min(len, sizeof(opts));
 		if (copy_to_user(optval, (char *)&opts, len))
 			err = -EFAULT;

@@ -1003,7 +1004,7 @@ static int sco_sock_getsockopt_old(struc
 		cinfo.hci_handle = sco_pi(sk)->conn->hcon->handle;
 		memcpy(cinfo.dev_class, sco_pi(sk)->conn->hcon->dev_class, 3);
 
-		len = min_t(unsigned int, len, sizeof(cinfo));
+		len = min(len, sizeof(cinfo));
 		if (copy_to_user(optval, (char *)&cinfo, len))
 			err = -EFAULT;

Greg Kroah-Hartman

author maintainer

2024-04-30 11:13 UTC #124 lore raw

PATCH 119/186 Bluetooth: btusb: Add Realtek RTL8852BE support ID 0x0bda:0x4853

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: WangYuli <[email protected]>

commit d1a5a7eede2977da3d2002d5ea3b519019cc1a98 upstream.

Add the support ID(0x0bda, 0x4853) to usb_device_id table for Realtek RTL8852BE.

Without this change the device utilizes an obsolete version of the firmware that is encoded in it rather than the updated Realtek firmware and config files from the firmware directory. The latter files implement many new features.

The device table is as follows:

T: Bus=03 Lev=01 Prnt=01 Port=09 Cnt=03 Dev#= 4 Spd=12 MxCh= 0
D: Ver= 1.00 Cls=e0(wlcon) Sub=01 Prot=01 MxPS=64 #Cfgs= 1
P: Vendor=0bda ProdID=4853 Rev= 0.00
S: Manufacturer=Realtek
S: Product=Bluetooth Radio
S: SerialNumber=00e04c000001
C:* #Ifs= 2 Cfg#= 1 Atr=e0 MxPwr=500mA
I:* If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E: Ad=81(I) Atr=03(Int.) MxPS= 16 Ivl=1ms
E: Ad=02(O) Atr=02(Bulk) MxPS= 64 Ivl=0ms
E: Ad=82(I) Atr=02(Bulk) MxPS= 64 Ivl=0ms
I:* If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E: Ad=03(O) Atr=01(Isoc) MxPS= 0 Ivl=1ms
E: Ad=83(I) Atr=01(Isoc) MxPS= 0 Ivl=1ms
I: If#= 1 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E: Ad=03(O) Atr=01(Isoc) MxPS= 9 Ivl=1ms
E: Ad=83(I) Atr=01(Isoc) MxPS= 9 Ivl=1ms
I: If#= 1 Alt= 2 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E: Ad=03(O) Atr=01(Isoc) MxPS= 17 Ivl=1ms
E: Ad=83(I) Atr=01(Isoc) MxPS= 17 Ivl=1ms
I: If#= 1 Alt= 3 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E: Ad=03(O) Atr=01(Isoc) MxPS= 25 Ivl=1ms
E: Ad=83(I) Atr=01(Isoc) MxPS= 25 Ivl=1ms
I: If#= 1 Alt= 4 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E: Ad=03(O) Atr=01(Isoc) MxPS= 33 Ivl=1ms
E: Ad=83(I) Atr=01(Isoc) MxPS= 33 Ivl=1ms
I: If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E: Ad=03(O) Atr=01(Isoc) MxPS= 49 Ivl=1ms
E: Ad=83(I) Atr=01(Isoc) MxPS= 49 Ivl=1ms

Signed-off-by Larry Finger <[email protected]>

Signed-off-by WangYuli <[email protected]>

Signed-off-by Luiz Augusto von Dentz <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/bluetooth/btusb.c |    2 ++
 1 file changed, 2 insertions(+)

drivers/bluetooth/btusb.c +2 -0

@@ -541,6 +541,8 @@ static const struct usb_device_id quirks
 	/* Realtek 8852BE Bluetooth devices */
 	{ USB_DEVICE(0x0cb8, 0xc559), .driver_info = BTUSB_REALTEK |
 						     BTUSB_WIDEBAND_SPEECH },
+	{ USB_DEVICE(0x0bda, 0x4853), .driver_info = BTUSB_REALTEK |
+						     BTUSB_WIDEBAND_SPEECH },
 	{ USB_DEVICE(0x0bda, 0x887b), .driver_info = BTUSB_REALTEK |
 						     BTUSB_WIDEBAND_SPEECH },
 	{ USB_DEVICE(0x0bda, 0xb85b), .driver_info = BTUSB_REALTEK |

Greg Kroah-Hartman

author maintainer

2024-04-30 11:13 UTC #125 lore raw

PATCH 120/186 Bluetooth: qca: fix NULL-deref on non-serdev suspend

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Johan Hovold <[email protected]>

commit 73e87c0a49fda31d7b589edccf4c72e924411371 upstream.

Qualcomm ROME controllers can be registered from the Bluetooth line discipline and in this case the HCI UART serdev pointer is NULL.

Add the missing sanity check to prevent a NULL-pointer dereference when wakeup() is called for a non-serdev controller during suspend.

Just return true for now to restore the original behaviour and address the crash with pre-6.2 kernels, which do not have commit e9b3e5b8c657 ("Bluetooth: hci_qca: only assign wakeup with serial port support") that causes the crash to happen already at setup() time.

Fixes c1a74160eaf1 ("Bluetooth: hci_qca: Add device_may_wakeup support")

Cc [email protected] # 5.13

Signed-off-by Johan Hovold <[email protected]>

Signed-off-by Luiz Augusto von Dentz <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/bluetooth/hci_qca.c |    3 +++
 1 file changed, 3 insertions(+)

drivers/bluetooth/hci_qca.c +3 -0

@@ -1672,6 +1672,9 @@ static bool qca_wakeup(struct hci_dev *h
 	struct hci_uart *hu = hci_get_drvdata(hdev);
 	bool wakeup;
 
+	if (!hu->serdev)
+		return true;
+
 	/* BT SoC attached through the serial bus is handled by the serdev driver.
 	 * So we need to use the device handle of the serdev driver to get the
 	 * status of device may wakeup.

Greg Kroah-Hartman

author maintainer

2024-04-30 11:14 UTC #126 lore raw

PATCH 121/186 Bluetooth: qca: fix NULL-deref on non-serdev setup

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Johan Hovold <[email protected]>

commit 7ddb9de6af0f1c71147785b12fd7c8ec3f06cc86 upstream.

Qualcomm ROME controllers can be registered from the Bluetooth line discipline and in this case the HCI UART serdev pointer is NULL.

Add the missing sanity check to prevent a NULL-pointer dereference when setup() is called for a non-serdev controller.

Fixes e9b3e5b8c657 ("Bluetooth: hci_qca: only assign wakeup with serial port support")

Cc [email protected] # 6.2

Cc Zhengping Jiang <[email protected]>

Signed-off-by Johan Hovold <[email protected]>

Signed-off-by Luiz Augusto von Dentz <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/bluetooth/hci_qca.c |    6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

drivers/bluetooth/hci_qca.c +4 -2

@@ -1938,8 +1938,10 @@ retry:
 		qca_debugfs_init(hdev);
 		hu->hdev->hw_error = qca_hw_error;
 		hu->hdev->cmd_timeout = qca_cmd_timeout;
-		if (device_can_wakeup(hu->serdev->ctrl->dev.parent))
-			hu->hdev->wakeup = qca_wakeup;
+		if (hu->serdev) {
+			if (device_can_wakeup(hu->serdev->ctrl->dev.parent))
+				hu->hdev->wakeup = qca_wakeup;
+		}
 	} else if (ret == -ENOENT) {
 		/* No patch/nvm-config found, run with original fw/config */
 		set_bit(QCA_ROM_FW, &qca->flags);

Greg Kroah-Hartman

author maintainer

2024-04-30 11:14 UTC #127 lore raw

PATCH 122/186 mtd: rawnand: qcom: Fix broken OP_RESET_DEVICE command in qcom_misc_cmd_type_exec()

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Christian Marangi <[email protected]>

commit b61bb5bc2c1cd00bb53db42f705735db6e8700f0 upstream.

While migrating to exec_ops in commit a82990c8a409 ("mtd: rawnand: qcom:
Add read/read_start ops in exec_op path"), OP_RESET_DEVICE command handling got broken unintentionally. Right now for the OP_RESET_DEVICE command, qcom_misc_cmd_type_exec() will simply return 0 without handling it. Even, if that gets fixed, an unnecessary FLASH_STATUS read descriptor command is being added in the middle and that seems to be causing the command to fail on IPQ806x devices.

So let's fix the above two issues to make OP_RESET_DEVICE command working again.

Fixes a82990c8a409 ("mtd: rawnand: qcom: Add read/read_start ops in exec_op path")

Link https://lore.kernel.org/linux-mtd/[email protected] open here

Reviewed-by Manivannan Sadhasivam <[email protected]>

Signed-off-by Christian Marangi <[email protected]>

Signed-off-by Miquel Raynal <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/mtd/nand/raw/qcom_nandc.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

drivers/mtd/nand/raw/qcom_nandc.c +3 -4

index b079605c84d3..b8cff9240b28 100644
@@ -2815,7 +2815,7 @@ static int qcom_misc_cmd_type_exec(struct nand_chip *chip, const struct nand_sub
 			      host->cfg0_raw & ~(7 << CW_PER_PAGE));
 		nandc_set_reg(chip, NAND_DEV0_CFG1, host->cfg1_raw);
 		instrs = 3;
-	} else {
+	} else if (q_op.cmd_reg != OP_RESET_DEVICE) {
 		return 0;
 	}

@@ -2830,9 +2830,8 @@ static int qcom_misc_cmd_type_exec(struct nand_chip *chip, const struct nand_sub
 	nandc_set_reg(chip, NAND_EXEC_CMD, 1);
 
 	write_reg_dma(nandc, NAND_FLASH_CMD, instrs, NAND_BAM_NEXT_SGL);
-	(q_op.cmd_reg == OP_BLOCK_ERASE) ? write_reg_dma(nandc, NAND_DEV0_CFG0,
-	2, NAND_BAM_NEXT_SGL) : read_reg_dma(nandc,
-	NAND_FLASH_STATUS, 1, NAND_BAM_NEXT_SGL);
+	if (q_op.cmd_reg == OP_BLOCK_ERASE)
+		write_reg_dma(nandc, NAND_DEV0_CFG0, 2, NAND_BAM_NEXT_SGL);
 
 	write_reg_dma(nandc, NAND_EXEC_CMD, 1, NAND_BAM_NEXT_SGL);
 	read_reg_dma(nandc, NAND_FLASH_STATUS, 1, NAND_BAM_NEXT_SGL);

Signature · 1 line

2.44.0

Greg Kroah-Hartman

author maintainer

2024-04-30 11:14 UTC #128 lore raw

PATCH 123/186 mm/hugetlb: fix missing hugetlb_lock for resv uncharge

https://lore.kernel.org/all/[email protected]/

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Peter Xu <[email protected]>

commit b76b46902c2d0395488c8412e1116c2486cdfcb2 upstream.

There is a recent report on UFFDIO_COPY over hugetlb:

350: lockdep_assert_held(&hugetlb_lock);

Should be an issue in hugetlb but triggered in an userfault context, where it goes into the unlikely path where two threads modifying the resv map together. Mike has a fix in that path for resv uncharge but it looks like the locking criteria was overlooked: hugetlb_cgroup_uncharge_folio_rsvd() will update the cgroup pointer, so it requires to be called with the lock held.

Link https://lkml.kernel.org/r/[email protected] open here

Fixes 79aa925bf239 ("hugetlb_cgroup: fix reservation accounting")

Signed-off-by Peter Xu <[email protected]>

Reported-by [email protected]

Reviewed-by Mina Almasry <[email protected]>

Cc David Hildenbrand <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 mm/hugetlb.c |    5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

mm/hugetlb.c +4 -1

@@ -3153,9 +3153,12 @@ struct folio *alloc_hugetlb_folio(struct
 
 		rsv_adjust = hugepage_subpool_put_pages(spool, 1);
 		hugetlb_acct_memory(h, -rsv_adjust);
-		if (deferred_reserve)
+		if (deferred_reserve) {
+			spin_lock_irq(&hugetlb_lock);
 			hugetlb_cgroup_uncharge_folio_rsvd(hstate_index(h),
 					pages_per_huge_page(h), folio);
+			spin_unlock_irq(&hugetlb_lock);
+		}
 	}
 	return folio;

Greg Kroah-Hartman

author maintainer

2024-04-30 11:14 UTC #129 lore raw

PATCH 124/186 mmc: sdhci-msm: pervent access to suspended controller

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mantas Pucka <[email protected]>

commit f8def10f73a516b771051a2f70f2f0446902cb4f upstream.

Generic sdhci code registers LED device and uses host->runtime_suspended flag to protect access to it. The sdhci-msm driver doesn't set this flag, which causes a crash when LED is accessed while controller is runtime suspended. Fix this by setting the flag correctly.

Fixes 67e6db113c90 ("mmc: sdhci-msm: Add pm_runtime and system PM support")

Signed-off-by Mantas Pucka <[email protected]>

Acked-by Adrian Hunter <[email protected]>

Link https://lore.kernel.org/r/[email protected] open here

Signed-off-by Ulf Hansson <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/mmc/host/sdhci-msm.c |   16 +++++++++++++++-
 1 file changed, 15 insertions(+), 1 deletion(-)

drivers/mmc/host/sdhci-msm.c +15 -1

@@ -2694,6 +2694,11 @@ static __maybe_unused int sdhci_msm_runt
 	struct sdhci_host *host = dev_get_drvdata(dev);
 	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
 	struct sdhci_msm_host *msm_host = sdhci_pltfm_priv(pltfm_host);
+	unsigned long flags;
+
+	spin_lock_irqsave(&host->lock, flags);
+	host->runtime_suspended = true;
+	spin_unlock_irqrestore(&host->lock, flags);
 
 	/* Drop the performance vote */
 	dev_pm_opp_set_rate(dev, 0);

@@ -2708,6 +2713,7 @@ static __maybe_unused int sdhci_msm_runt
 	struct sdhci_host *host = dev_get_drvdata(dev);
 	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
 	struct sdhci_msm_host *msm_host = sdhci_pltfm_priv(pltfm_host);
+	unsigned long flags;
 	int ret;
 
 	ret = clk_bulk_prepare_enable(ARRAY_SIZE(msm_host->bulk_clks),

@@ -2726,7 +2732,15 @@ static __maybe_unused int sdhci_msm_runt
 
 	dev_pm_opp_set_rate(dev, msm_host->clk_rate);
 
-	return sdhci_msm_ice_resume(msm_host);
+	ret = sdhci_msm_ice_resume(msm_host);
+	if (ret)
+		return ret;
+
+	spin_lock_irqsave(&host->lock, flags);
+	host->runtime_suspended = false;
+	spin_unlock_irqrestore(&host->lock, flags);
+
+	return ret;
 }
 
 static const struct dev_pm_ops sdhci_msm_pm_ops = {

Greg Kroah-Hartman

author maintainer

2024-04-30 11:14 UTC #130 lore raw

PATCH 125/186 mm: create FOLIO_FLAG_FALSE and FOLIO_TYPE_OPS macros

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Matthew Wilcox (Oracle) <[email protected]>

commit 12bbaae7635a56049779db3bef6e7140d9aa5f67 upstream.

Following the separation of FOLIO_FLAGS from PAGEFLAGS, separate FOLIO_FLAG_FALSE from PAGEFLAG_FALSE and FOLIO_TYPE_OPS from PAGE_TYPE_OPS.

Link https://lkml.kernel.org/r/[email protected] open here

Fixes 9c5ccf2db04b ("mm: remove HUGETLB_PAGE_DTOR")

Signed-off-by Matthew Wilcox (Oracle) <[email protected]>

Reviewed-by David Hildenbrand <[email protected]>

Acked-by Vlastimil Babka <[email protected]>

Cc Miaohe Lin <[email protected]>

Cc Muchun Song <[email protected]>

Cc Oscar Salvador <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 include/linux/page-flags.h |   70 ++++++++++++++++++++++++++++++---------------
 1 file changed, 47 insertions(+), 23 deletions(-)

include/linux/page-flags.h +47 ~~-23~~

@@ -432,30 +432,51 @@ static __always_inline int TestClearPage
 	TESTSETFLAG(uname, lname, policy)				\
 	TESTCLEARFLAG(uname, lname, policy)
 
+#define FOLIO_TEST_FLAG_FALSE(name)					\
+static inline bool folio_test_##name(const struct folio *folio)		\
+{ return false; }
+#define FOLIO_SET_FLAG_NOOP(name)					\
+static inline void folio_set_##name(struct folio *folio) { }
+#define FOLIO_CLEAR_FLAG_NOOP(name)					\
+static inline void folio_clear_##name(struct folio *folio) { }
+#define __FOLIO_SET_FLAG_NOOP(name)					\
+static inline void __folio_set_##name(struct folio *folio) { }
+#define __FOLIO_CLEAR_FLAG_NOOP(name)					\
+static inline void __folio_clear_##name(struct folio *folio) { }
+#define FOLIO_TEST_SET_FLAG_FALSE(name)					\
+static inline bool folio_test_set_##name(struct folio *folio)		\
+{ return false; }
+#define FOLIO_TEST_CLEAR_FLAG_FALSE(name)				\
+static inline bool folio_test_clear_##name(struct folio *folio)		\
+{ return false; }
+
+#define FOLIO_FLAG_FALSE(name)						\
+FOLIO_TEST_FLAG_FALSE(name)						\
+FOLIO_SET_FLAG_NOOP(name)						\
+FOLIO_CLEAR_FLAG_NOOP(name)
+
 #define TESTPAGEFLAG_FALSE(uname, lname)				\
-static inline bool folio_test_##lname(const struct folio *folio) { return false; } \
+FOLIO_TEST_FLAG_FALSE(lname)						\
 static inline int Page##uname(const struct page *page) { return 0; }
 
 #define SETPAGEFLAG_NOOP(uname, lname)					\
-static inline void folio_set_##lname(struct folio *folio) { }		\
+FOLIO_SET_FLAG_NOOP(lname)						\
 static inline void SetPage##uname(struct page *page) {  }
 
 #define CLEARPAGEFLAG_NOOP(uname, lname)				\
-static inline void folio_clear_##lname(struct folio *folio) { }		\
+FOLIO_CLEAR_FLAG_NOOP(lname)						\
 static inline void ClearPage##uname(struct page *page) {  }
 
 #define __CLEARPAGEFLAG_NOOP(uname, lname)				\
-static inline void __folio_clear_##lname(struct folio *folio) { }	\
+__FOLIO_CLEAR_FLAG_NOOP(lname)						\
 static inline void __ClearPage##uname(struct page *page) {  }
 
 #define TESTSETFLAG_FALSE(uname, lname)					\
-static inline bool folio_test_set_##lname(struct folio *folio)		\
-{ return 0; }								\
+FOLIO_TEST_SET_FLAG_FALSE(lname)					\
 static inline int TestSetPage##uname(struct page *page) { return 0; }
 
 #define TESTCLEARFLAG_FALSE(uname, lname)				\
-static inline bool folio_test_clear_##lname(struct folio *folio)	\
-{ return 0; }								\
+FOLIO_TEST_CLEAR_FLAG_FALSE(lname)					\
 static inline int TestClearPage##uname(struct page *page) { return 0; }
 
 #define PAGEFLAG_FALSE(uname, lname) TESTPAGEFLAG_FALSE(uname, lname)	\

@@ -937,35 +958,38 @@ static inline int page_has_type(struct p
 	return page_type_has_type(page->page_type);
 }
 
+#define FOLIO_TYPE_OPS(lname, fname)					\
+static __always_inline bool folio_test_##fname(const struct folio *folio)\
+{									\
+	return folio_test_type(folio, PG_##lname);			\
+}									\
+static __always_inline void __folio_set_##fname(struct folio *folio)	\
+{									\
+	VM_BUG_ON_FOLIO(!folio_test_type(folio, 0), folio);		\
+	folio->page.page_type &= ~PG_##lname;				\
+}									\
+static __always_inline void __folio_clear_##fname(struct folio *folio)	\
+{									\
+	VM_BUG_ON_FOLIO(!folio_test_##fname(folio), folio);		\
+	folio->page.page_type |= PG_##lname;				\
+}
+
 #define PAGE_TYPE_OPS(uname, lname, fname)				\
+FOLIO_TYPE_OPS(lname, fname)						\
 static __always_inline int Page##uname(const struct page *page)		\
 {									\
 	return PageType(page, PG_##lname);				\
 }									\
-static __always_inline int folio_test_##fname(const struct folio *folio)\
-{									\
-	return folio_test_type(folio, PG_##lname);			\
-}									\
 static __always_inline void __SetPage##uname(struct page *page)		\
 {									\
 	VM_BUG_ON_PAGE(!PageType(page, 0), page);			\
 	page->page_type &= ~PG_##lname;					\
 }									\
-static __always_inline void __folio_set_##fname(struct folio *folio)	\
-{									\
-	VM_BUG_ON_FOLIO(!folio_test_type(folio, 0), folio);		\
-	folio->page.page_type &= ~PG_##lname;				\
-}									\
 static __always_inline void __ClearPage##uname(struct page *page)	\
 {									\
 	VM_BUG_ON_PAGE(!Page##uname(page), page);			\
 	page->page_type |= PG_##lname;					\
-}									\
-static __always_inline void __folio_clear_##fname(struct folio *folio)	\
-{									\
-	VM_BUG_ON_FOLIO(!folio_test_##fname(folio), folio);		\
-	folio->page.page_type |= PG_##lname;				\
-}									\
+}
 
 /*
  * PageBuddy() indicates that the page is free and in the buddy system

Greg Kroah-Hartman

author maintainer

2024-04-30 11:12 UTC #131 lore raw

PATCH 126/186 mm: support page_mapcount() on page_has_type() pages

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Matthew Wilcox (Oracle) <[email protected]>

commit fd1a745ce03e37945674c14833870a9af0882e2d upstream.

Return 0 for pages which can't be mapped. This matches how page_mapped() works. It is more convenient for users to not have to filter out these pages.

Link https://lkml.kernel.org/r/[email protected] open here

Fixes 9c5ccf2db04b ("mm: remove HUGETLB_PAGE_DTOR")

Signed-off-by Matthew Wilcox (Oracle) <[email protected]>

Reviewed-by David Hildenbrand <[email protected]>

Acked-by Vlastimil Babka <[email protected]>

Cc Miaohe Lin <[email protected]>

Cc Muchun Song <[email protected]>

Cc Oscar Salvador <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 fs/proc/page.c             |    7 ++-----
 include/linux/mm.h         |    8 +++++---
 include/linux/page-flags.h |    4 ++--
 3 files changed, 9 insertions(+), 10 deletions(-)

fs/proc/page.c +2 -5

@@ -67,7 +67,7 @@ static ssize_t kpagecount_read(struct fi
 		 */
 		ppage = pfn_to_online_page(pfn);
 
-		if (!ppage || PageSlab(ppage) || page_has_type(ppage))
+		if (!ppage)
 			pcount = 0;
 		else
 			pcount = page_mapcount(ppage);

@@ -124,11 +124,8 @@ u64 stable_page_flags(struct page *page)
 
 	/*
 	 * pseudo flags for the well known (anonymous) memory mapped pages
-	 *
-	 * Note that page->_mapcount is overloaded in SLAB, so the
-	 * simple test in page_mapped() is not enough.
 	 */
-	if (!PageSlab(page) && page_mapped(page))
+	if (page_mapped(page))
 		u |= 1 << KPF_MMAP;
 	if (PageAnon(page))
 		u |= 1 << KPF_ANON;

include/linux/mm.h +5 -3

@@ -1184,14 +1184,16 @@ static inline void page_mapcount_reset(s
  * a large folio, it includes the number of times this page is mapped
  * as part of that folio.
  *
- * The result is undefined for pages which cannot be mapped into userspace.
- * For example SLAB or special types of pages. See function page_has_type().
- * They use this field in struct page differently.
+ * Will report 0 for pages which cannot be mapped into userspace, eg
+ * slab, page tables and similar.
  */
 static inline int page_mapcount(struct page *page)
 {
 	int mapcount = atomic_read(&page->_mapcount) + 1;
 
+	/* Handle page_has_type() pages */
+	if (mapcount < 0)
+		mapcount = 0;
 	if (unlikely(PageCompound(page)))
 		mapcount += folio_entire_mapcount(page_folio(page));

include/linux/page-flags.h +2 -2

@@ -931,12 +931,12 @@ static inline bool is_page_hwpoison(stru
  * page_type may be used.  Because it is initialised to -1, we invert the
  * sense of the bit, so __SetPageFoo *clears* the bit used for PageFoo, and
  * __ClearPageFoo *sets* the bit used for PageFoo.  We reserve a few high and
- * low bits so that an underflow or overflow of page_mapcount() won't be
+ * low bits so that an underflow or overflow of _mapcount won't be
  * mistaken for a page type value.
  */
 
 #define PAGE_TYPE_BASE	0xf0000000
-/* Reserve		0x0000007f to catch underflows of page_mapcount */
+/* Reserve		0x0000007f to catch underflows of _mapcount */
 #define PAGE_MAPCOUNT_RESERVE	-128
 #define PG_buddy	0x00000080
 #define PG_offline	0x00000100

Greg Kroah-Hartman

author maintainer

2024-04-30 11:12 UTC #132 lore raw

PATCH 127/186 smb: client: Fix struct_group() usage in __packed structs

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

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

commit 9a1f1d04f63c59550a5364858b46eeffdf03e8d6 upstream.

Use struct_group_attr() in __packed structs, instead of struct_group().

Below you can see the pahole output before/after changes:

pahole -C smb2_file_network_open_info fs/smb/client/smb2ops.o struct smb2_file_network_open_info {
union {
struct {
__le64 CreationTime; /* 0 8 */
__le64 LastAccessTime; /* 8 8 */
__le64 LastWriteTime; /* 16 8 */
__le64 ChangeTime; /* 24 8 */
__le64 AllocationSize; /* 32 8 */
__le64 EndOfFile; /* 40 8 */
__le32 Attributes; /* 48 4 */
}; /* 0 56 */
struct {
__le64 CreationTime; /* 0 8 */
__le64 LastAccessTime; /* 8 8 */
__le64 LastWriteTime; /* 16 8 */
__le64 ChangeTime; /* 24 8 */
__le64 AllocationSize; /* 32 8 */
__le64 EndOfFile; /* 40 8 */
__le32 Attributes; /* 48 4 */
} network_open_info; /* 0 56 */
}; /* 0 56 */
__le32 Reserved; /* 56 4 */

/* size: 60, cachelines: 1, members: 2 */ /* last cacheline: 60 bytes */
} __attribute__((__packed__));

pahole -C smb2_file_network_open_info fs/smb/client/smb2ops.o struct smb2_file_network_open_info {
union {
struct {
__le64 CreationTime; /* 0 8 */
__le64 LastAccessTime; /* 8 8 */
__le64 LastWriteTime; /* 16 8 */
__le64 ChangeTime; /* 24 8 */
__le64 AllocationSize; /* 32 8 */
__le64 EndOfFile; /* 40 8 */
__le32 Attributes; /* 48 4 */
} __attribute__((__packed__)); /* 0 52 */
struct {
__le64 CreationTime; /* 0 8 */
__le64 LastAccessTime; /* 8 8 */
__le64 LastWriteTime; /* 16 8 */
__le64 ChangeTime; /* 24 8 */
__le64 AllocationSize; /* 32 8 */
__le64 EndOfFile; /* 40 8 */
__le32 Attributes; /* 48 4 */
} __attribute__((__packed__)) network_open_info; /* 0 52 */
}; /* 0 52 */
__le32 Reserved; /* 52 4 */

/* size: 56, cachelines: 1, members: 2 */ /* last cacheline: 56 bytes */
};

pahole -C smb_com_open_rsp fs/smb/client/cifssmb.o

struct smb_com_open_rsp {
	...

	union {
		struct {
			__le64     CreationTime;         /*    48     8 */
			__le64     LastAccessTime;       /*    56     8 */
			/* --- cacheline 1 boundary (64 bytes) --- */
			__le64     LastWriteTime;        /*    64     8 */
			__le64     ChangeTime;           /*    72     8 */
			__le32     FileAttributes;       /*    80     4 */
		};                                       /*    48    40 */
		struct {
			__le64     CreationTime;         /*    48     8 */
			__le64     LastAccessTime;       /*    56     8 */
			/* --- cacheline 1 boundary (64 bytes) --- */
			__le64     LastWriteTime;        /*    64     8 */
			__le64     ChangeTime;           /*    72     8 */
			__le32     FileAttributes;       /*    80     4 */
		} common_attributes;                     /*    48    40 */
	};                                               /*    48    40 */

	...

	/* size: 111, cachelines: 2, members: 14 */
	/* last cacheline: 47 bytes */
} __attribute__((__packed__));

pahole -C smb_com_open_rsp fs/smb/client/cifssmb.o

struct smb_com_open_rsp {
	...

	union {
		struct {
			__le64     CreationTime;         /*    48     8 */
			__le64     LastAccessTime;       /*    56     8 */
			/* --- cacheline 1 boundary (64 bytes) --- */
			__le64     LastWriteTime;        /*    64     8 */
			__le64     ChangeTime;           /*    72     8 */
			__le32     FileAttributes;       /*    80     4 */
		} __attribute__((__packed__));           /*    48    36 */
		struct {
			__le64     CreationTime;         /*    48     8 */
			__le64     LastAccessTime;       /*    56     8 */
			/* --- cacheline 1 boundary (64 bytes) --- */
			__le64     LastWriteTime;        /*    64     8 */
			__le64     ChangeTime;           /*    72     8 */
			__le32     FileAttributes;       /*    80     4 */
		} __attribute__((__packed__)) common_attributes;       /*    48    36 */
	};                                               /*    48    36 */

	...

	/* size: 107, cachelines: 2, members: 14 */
	/* last cacheline: 43 bytes */
} __attribute__((__packed__));

pahole -C FILE_ALL_INFO fs/smb/client/cifssmb.o typedef struct {
union {
struct {
__le64 CreationTime; /* 0 8 */
__le64 LastAccessTime; /* 8 8 */
__le64 LastWriteTime; /* 16 8 */
__le64 ChangeTime; /* 24 8 */
__le32 Attributes; /* 32 4 */
}; /* 0 40 */
struct {
__le64 CreationTime; /* 0 8 */
__le64 LastAccessTime; /* 8 8 */
__le64 LastWriteTime; /* 16 8 */
__le64 ChangeTime; /* 24 8 */
__le32 Attributes; /* 32 4 */
} common_attributes; /* 0 40 */
}; /* 0 40 */

...

/* size: 113, cachelines: 2, members: 17 */ /* last cacheline: 49 bytes */
} __attribute__((__packed__)) FILE_ALL_INFO;

pahole -C FILE_ALL_INFO fs/smb/client/cifssmb.o typedef struct {
union {
struct {
__le64 CreationTime; /* 0 8 */
__le64 LastAccessTime; /* 8 8 */
__le64 LastWriteTime; /* 16 8 */
__le64 ChangeTime; /* 24 8 */
__le32 Attributes; /* 32 4 */
} __attribute__((__packed__)); /* 0 36 */
struct {
__le64 CreationTime; /* 0 8 */
__le64 LastAccessTime; /* 8 8 */
__le64 LastWriteTime; /* 16 8 */
__le64 ChangeTime; /* 24 8 */
__le32 Attributes; /* 32 4 */
} __attribute__((__packed__)) common_attributes; /* 0 36 */
}; /* 0 36 */

...

/* size: 109, cachelines: 2, members: 17 */ /* last cacheline: 45 bytes */
} __attribute__((__packed__)) FILE_ALL_INFO;

Fixes 0015eb6e1238 ("smb: client, common: fix fortify warnings")

Reviewed-by Namjae Jeon <[email protected]>

Signed-off-by Gustavo A. R. Silva <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 fs/smb/client/cifspdu.h |    4 ++--
 fs/smb/client/smb2pdu.h |    2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

fs/smb/client/cifspdu.h +2 -2

@@ -882,7 +882,7 @@ typedef struct smb_com_open_rsp {
 	__u8 OplockLevel;
 	__u16 Fid;
 	__le32 CreateAction;
-	struct_group(common_attributes,
+	struct_group_attr(common_attributes, __packed,
 		__le64 CreationTime;
 		__le64 LastAccessTime;
 		__le64 LastWriteTime;

@@ -2266,7 +2266,7 @@ typedef struct {
 /* QueryFileInfo/QueryPathinfo (also for SetPath/SetFile) data buffer formats */
 /******************************************************************************/
 typedef struct { /* data block encoding of response to level 263 QPathInfo */
-	struct_group(common_attributes,
+	struct_group_attr(common_attributes, __packed,
 		__le64 CreationTime;
 		__le64 LastAccessTime;
 		__le64 LastWriteTime;

fs/smb/client/smb2pdu.h +1 -1

@@ -319,7 +319,7 @@ struct smb2_file_reparse_point_info {
 } __packed;
 
 struct smb2_file_network_open_info {
-	struct_group(network_open_info,
+	struct_group_attr(network_open_info, __packed,
 		__le64 CreationTime;
 		__le64 LastAccessTime;
 		__le64 LastWriteTime;

Greg Kroah-Hartman

author maintainer

2024-04-30 11:12 UTC #133 lore raw

PATCH 128/186 smb3: missing lock when picking channel

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Steve French <[email protected]>

commit 8094a600245e9b28eb36a13036f202ad67c1f887 upstream.

Coverity spotted a place where we should have been holding the channel lock when accessing the ses channel index.

Addresses-Coverity: 1582039 ("Data race condition (MISSING_LOCK)")

Reviewed-by Shyam Prasad N <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 fs/smb/client/transport.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

fs/smb/client/transport.c +3 -1

@@ -1057,9 +1057,11 @@ struct TCP_Server_Info *cifs_pick_channe
 		index = (uint)atomic_inc_return(&ses->chan_seq);
 		index %= ses->chan_count;
 	}
+
+	server = ses->chans[index].server;
 	spin_unlock(&ses->chan_lock);
 
-	return ses->chans[index].server;
+	return server;
 }
 
 int

Greg Kroah-Hartman

author maintainer

2024-04-30 11:12 UTC #134 lore raw

PATCH 129/186 smb3: fix lock ordering potential deadlock in cifs_sync_mid_result

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Steve French <[email protected]>

commit 8861fd5180476f45f9e8853db154600469a0284f upstream.

Coverity spotted that the cifs_sync_mid_result function could deadlock

"Thread deadlock (ORDER_REVERSAL) lock_order: Calling spin_lock acquires lock TCP_Server_Info.srv_lock while holding lock TCP_Server_Info.mid_lock"

Addresses-Coverity: 1590401 ("Thread deadlock (ORDER_REVERSAL)")

Reviewed-by Shyam Prasad N <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 fs/smb/client/transport.c |    3 +++
 1 file changed, 3 insertions(+)

fs/smb/client/transport.c +3 -0

@@ -909,12 +909,15 @@ cifs_sync_mid_result(struct mid_q_entry
 			list_del_init(&mid->qhead);
 			mid->mid_flags |= MID_DELETED;
 		}
+		spin_unlock(&server->mid_lock);
 		cifs_server_dbg(VFS, "%s: invalid mid state mid=%llu state=%d\n",
 			 __func__, mid->mid, mid->mid_state);
 		rc = -EIO;
+		goto sync_mid_done;
 	}
 	spin_unlock(&server->mid_lock);
 
+sync_mid_done:
 	release_mid(mid);
 	return rc;
 }

Greg Kroah-Hartman

author maintainer

2024-04-30 11:12 UTC #135 lore raw

PATCH 130/186 HID: i2c-hid: remove I2C_HID_READ_PENDING flag to prevent lock-up

Closes https://lore.kernel.org/r/CA+eeCSPUDpUg76ZO8dszSbAGn+UHjcyv8F1J-CUPVARAzEtW9w@mail.gmail.com open here

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Nam Cao <[email protected]>

commit 9c0f59e47a90c54d0153f8ddc0f80d7a36207d0e upstream.

The flag I2C_HID_READ_PENDING is used to serialize I2C operations. However, this is not necessary, because I2C core already has its own locking for that.

More importantly, this flag can cause a lock-up: if the flag is set in i2c_hid_xfer() and an interrupt happens, the interrupt handler (i2c_hid_irq) will check this flag and return immediately without doing anything, then the interrupt handler will be invoked again in an infinite loop.

Since interrupt handler is an RT task, it takes over the CPU and the flag-clearing task never gets scheduled, thus we have a lock-up.

Delete this unnecessary flag.

Reported-and-tested-by Eva Kurchatova <[email protected]>

Fixes 4a200c3b9a40 ("HID: i2c-hid: introduce HID over i2c specification implementation")

Signed-off-by Nam Cao <[email protected]>

Signed-off-by Jiri Kosina <[email protected]>

Signed-off-by Greg Kroah-Hartman <[email protected]>

---
 drivers/hid/i2c-hid/i2c-hid-core.c |    9 ---------
 1 file changed, 9 deletions(-)

drivers/hid/i2c-hid/i2c-hid-core.c +0 -9

@@ -64,7 +64,6 @@
 /* flags */
 #define I2C_HID_STARTED		0
 #define I2C_HID_RESET_PENDING	1
-#define I2C_HID_READ_PENDING	2
 
 #define I2C_HID_PWR_ON		0x00
 #define I2C_HID_PWR_SLEEP	0x01

@@ -190,15 +189,10 @@ static int i2c_hid_xfer(struct i2c_hid *
 		msgs[n].len = recv_len;
 		msgs[n].buf = recv_buf;
 		n++;
-
-		set_bit(I2C_HID_READ_PENDING, &ihid->flags);
 	}
 
 	ret = i2c_transfer(client->adapter, msgs, n);
 
-	if (recv_len)
-		clear_bit(I2C_HID_READ_PENDING, &ihid->flags);
-
 	if (ret != n)
 		return ret < 0 ? ret : -EIO;

@@ -566,9 +560,6 @@ static irqreturn_t i2c_hid_irq(int irq,
 {
 	struct i2c_hid *ihid = dev_id;
 
-	if (test_bit(I2C_HID_READ_PENDING, &ihid->flags))
-		return IRQ_HANDLED;
-
 	i2c_hid_get_input(ihid);
 
 	return IRQ_HANDLED;

Greg Kroah-Hartman

author maintainer

2024-04-30 11:12 UTC #136 lore raw

PATCH 131/186 btrfs: fallback if compressed IO fails for ENOSPC