Skip to contentSkip to conversation

5.16.3-rc1 review

PATCH In review Greg Kroah-Hartman 12 participants 1,058 messages Jan 24, 2022 – Feb 3, 2022 Page 4 of 8

Subsystems (9)
lore
Share & tools
lore
author maintainer
#387 lore raw

PATCH 387/1039 Bluetooth: hci_qca: Fix NULL vs IS_ERR_OR_NULL check in qca_serdev_probe

In reply to Greg Kroah-Hartman

From: Miaoqian Lin <[email protected]>

[ Upstream commit 6845667146a28c09b5dfc401c1ad112374087944 ]

The function devm_gpiod_get_index() return error pointers on error. Thus devm_gpiod_get_index_optional() could return NULL and error pointers. The same as devm_gpiod_get_optional() function. Using IS_ERR_OR_NULL() check to catch error pointers.

Fixes 77131dfe ("Bluetooth: hci_qca: Replace devm_gpiod_get() with devm_gpiod_get_optional()")
---
 drivers/bluetooth/hci_qca.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/bluetooth/hci_qca.c +3 -3
index 9e99311038ae8..f6e91fb432a3b 100644
@@ -2059,14 +2059,14 @@ static int qca_serdev_probe(struct serdev_device *serdev)
 
 		qcadev->bt_en = devm_gpiod_get_optional(&serdev->dev, "enable",
 					       GPIOD_OUT_LOW);
-		if (!qcadev->bt_en && data->soc_type == QCA_WCN6750) {
+		if (IS_ERR_OR_NULL(qcadev->bt_en) && data->soc_type == QCA_WCN6750) {
 			dev_err(&serdev->dev, "failed to acquire BT_EN gpio\n");
 			power_ctrl_enabled = false;
 		}
 
 		qcadev->sw_ctrl = devm_gpiod_get_optional(&serdev->dev, "swctrl",
 					       GPIOD_IN);
-		if (!qcadev->sw_ctrl && data->soc_type == QCA_WCN6750)
+		if (IS_ERR_OR_NULL(qcadev->sw_ctrl) && data->soc_type == QCA_WCN6750)
 			dev_warn(&serdev->dev, "failed to acquire SW_CTRL gpio\n");
 
 		qcadev->susclk = devm_clk_get_optional(&serdev->dev, NULL);
@@ -2088,7 +2088,7 @@ static int qca_serdev_probe(struct serdev_device *serdev)
 
 		qcadev->bt_en = devm_gpiod_get_optional(&serdev->dev, "enable",
 					       GPIOD_OUT_LOW);
-		if (!qcadev->bt_en) {
+		if (IS_ERR_OR_NULL(qcadev->bt_en)) {
 			dev_warn(&serdev->dev, "failed to acquire enable gpio\n");
 			power_ctrl_enabled = false;
 		}
Signature · 1 line
2.34.1
author maintainer
#388 lore raw

PATCH 388/1039 net/smc: Reset conn->lgr when link group registration fails

In reply to Greg Kroah-Hartman

From: Wen Gu <[email protected]>

[ Upstream commit 36595d8ad46d9e4c41cc7c48c4405b7c3322deac ]

SMC connections might fail to be registered in a link group due to unable to find a usable link during its creation. As a result, smc_conn_create() will return a failure and most resources related to the connection won't be applied or initialized, such as conn->abort_work or conn->lnk.

If smc_conn_free() is invoked later, it will try to access the uninitialized resources related to the connection, thus causing a warning or crash.

This patch tries to fix this by resetting conn->lgr to NULL if an abnormal exit occurs in smc_lgr_register_conn(), thus avoiding the access to uninitialized resources in smc_conn_free().

Meanwhile, the new created link group should be terminated if smc connections can't be registered in it. So smc_lgr_cleanup_early() is modified to take care of link group only and invoked to terminate unusable link group by smc_conn_create(). The call to smc_conn_free() is moved out from smc_lgr_cleanup_early() to smc_conn_abort().

Fixes 56bc3b2094b4 ("net/smc: assign link to a new connection")
---
 net/smc/af_smc.c   |  8 +++++---
 net/smc/smc_core.c | 12 +++++++-----
 net/smc/smc_core.h |  2 +-
 3 files changed, 13 insertions(+), 9 deletions(-)
net/smc/af_smc.c +5 -3
index 1c9289f56dc47..211cd91b6c408 100644
@@ -632,10 +632,12 @@ static int smc_connect_decline_fallback(struct smc_sock *smc, int reason_code,
 
 static void smc_conn_abort(struct smc_sock *smc, int local_first)
 {
+	struct smc_connection *conn = &smc->conn;
+	struct smc_link_group *lgr = conn->lgr;
+
+	smc_conn_free(conn);
 	if (local_first)
-		smc_lgr_cleanup_early(&smc->conn);
-	else
-		smc_conn_free(&smc->conn);
+		smc_lgr_cleanup_early(lgr);
 }
 
 /* check if there is a rdma device available for this connection. */
net/smc/smc_core.c +7 -5
index a6849362f4ddd..c9a8092c4ac96 100644
@@ -171,8 +171,10 @@ static int smc_lgr_register_conn(struct smc_connection *conn, bool first)
 
 	if (!conn->lgr->is_smcd) {
 		rc = smcr_lgr_conn_assign_link(conn, first);
-		if (rc)
+		if (rc) {
+			conn->lgr = NULL;
 			return rc;
+		}
 	}
 	/* find a new alert_token_local value not yet used by some connection
 	 * in this link group
@@ -622,15 +624,13 @@ int smcd_nl_get_lgr(struct sk_buff *skb, struct netlink_callback *cb)
 	return skb->len;
 }
 
-void smc_lgr_cleanup_early(struct smc_connection *conn)
+void smc_lgr_cleanup_early(struct smc_link_group *lgr)
 {
-	struct smc_link_group *lgr = conn->lgr;
 	spinlock_t *lgr_lock;
 
 	if (!lgr)
 		return;
 
-	smc_conn_free(conn);
 	smc_lgr_list_head(lgr, &lgr_lock);
 	spin_lock_bh(lgr_lock);
 	/* do not use this link group for new connections */
@@ -1832,8 +1832,10 @@ create:
 		write_lock_bh(&lgr->conns_lock);
 		rc = smc_lgr_register_conn(conn, true);
 		write_unlock_bh(&lgr->conns_lock);
-		if (rc)
+		if (rc) {
+			smc_lgr_cleanup_early(lgr);
 			goto out;
+		}
 	}
 	conn->local_tx_ctrl.common.type = SMC_CDC_MSG_TYPE;
 	conn->local_tx_ctrl.len = SMC_WR_TX_SIZE;
net/smc/smc_core.h +1 -1
index d63b08274197e..73d0c35d3eb77 100644
@@ -468,7 +468,7 @@ static inline void smc_set_pci_values(struct pci_dev *pci_dev,
 struct smc_sock;
 struct smc_clc_msg_accept_confirm;
 
-void smc_lgr_cleanup_early(struct smc_connection *conn);
+void smc_lgr_cleanup_early(struct smc_link_group *lgr);
 void smc_lgr_terminate_sched(struct smc_link_group *lgr);
 void smcr_port_add(struct smc_ib_device *smcibdev, u8 ibport);
 void smcr_port_err(struct smc_ib_device *smcibdev, u8 ibport);
Signature · 1 line
2.34.1
author maintainer
#389 lore raw

PATCH 389/1039 usb: dwc3: qcom: Fix NULL vs IS_ERR checking in dwc3_qcom_probe

In reply to Greg Kroah-Hartman

From: Miaoqian Lin <[email protected]>

[ Upstream commit b52fe2dbb3e655eb1483000adfab68a219549e13 ]

Since the acpi_create_platform_device() function may return error pointers, dwc3_qcom_create_urs_usb_platdev() function may return error pointers too. Using IS_ERR_OR_NULL() to check the return value to fix this.

Fixes c25c210f590e ("usb: dwc3: qcom: add URS Host support for sdm845 ACPI boot")
---
 drivers/usb/dwc3/dwc3-qcom.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)
drivers/usb/dwc3/dwc3-qcom.c +5 -2
index 3cb01cdd02c29..b81a9e1c13153 100644
@@ -769,9 +769,12 @@ static int dwc3_qcom_probe(struct platform_device *pdev)
 
 		if (qcom->acpi_pdata->is_urs) {
 			qcom->urs_usb = dwc3_qcom_create_urs_usb_platdev(dev);
-			if (!qcom->urs_usb) {
+			if (IS_ERR_OR_NULL(qcom->urs_usb)) {
 				dev_err(dev, "failed to create URS USB platdev\n");
-				return -ENODEV;
+				if (!qcom->urs_usb)
+					return -ENODEV;
+				else
+					return PTR_ERR(qcom->urs_usb);
 			}
 		}
 	}
Signature · 1 line
2.34.1
author maintainer
#390 lore raw

PATCH 390/1039 usb: dwc2: do not gate off the hardware if it does not support clock gating

In reply to Greg Kroah-Hartman

From: Dinh Nguyen <[email protected]>

[ Upstream commit 34146c68083f1aef6709196b3dc888c1ceffd357 ]

We should not be clearing the HCD_FLAG_HW_ACCESSIBLE bit if the hardware does not support clock gating.

Fixes 50fb0c128b6e ("usb: dwc2: Add clock gating entering flow by system suspend")
---
 drivers/usb/dwc2/hcd.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
drivers/usb/dwc2/hcd.c +4 -3
index 13c779a28e94f..f63a27d11fac8 100644
@@ -4399,11 +4399,12 @@ static int _dwc2_hcd_suspend(struct usb_hcd *hcd)
 		 * If not hibernation nor partial power down are supported,
 		 * clock gating is used to save power.
 		 */
-		if (!hsotg->params.no_clock_gating)
+		if (!hsotg->params.no_clock_gating) {
 			dwc2_host_enter_clock_gating(hsotg);
 
-		/* After entering suspend, hardware is not accessible */
-		clear_bit(HCD_FLAG_HW_ACCESSIBLE, &hcd->flags);
+			/* After entering suspend, hardware is not accessible */
+			clear_bit(HCD_FLAG_HW_ACCESSIBLE, &hcd->flags);
+		}
 		break;
 	default:
 		goto skip_power_saving;
Signature · 1 line
2.34.1
author maintainer
#391 lore raw

PATCH 391/1039 usb: dwc2: gadget: initialize max_speed from params

In reply to Greg Kroah-Hartman

From: John Keeping <[email protected]>

[ Upstream commit 92ef98a4caacad6d4a1490dda45d81ae5ccf5bc9 ]

DWC2 may be paired with a full-speed PHY which is not capable of high-speed operation. Report this correctly to the gadget core by setting max_speed from the core parameters.

Prior to commit 5324bad66f09f ("usb: dwc2: gadget: implement udc_set_speed()") this didn't cause the hardware to be configured incorrectly, although the speed may have been reported incorrectly. But after that commit params.speed is updated based on a value passed in by the gadget core which may set it to a faster speed than is supported by the hardware. Initialising the max_speed parameter ensures the speed passed to dwc2_gadget_set_speed() will be one supported by the hardware.

---
 drivers/usb/dwc2/gadget.c | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)
drivers/usb/dwc2/gadget.c +12 -1
index ab8d7dad9f567..43cf49c4e5e59 100644
@@ -4974,7 +4974,18 @@ int dwc2_gadget_init(struct dwc2_hsotg *hsotg)
 		hsotg->params.g_np_tx_fifo_size);
 	dev_dbg(dev, "RXFIFO size: %d\n", hsotg->params.g_rx_fifo_size);
 
-	hsotg->gadget.max_speed = USB_SPEED_HIGH;
+	switch (hsotg->params.speed) {
+	case DWC2_SPEED_PARAM_LOW:
+		hsotg->gadget.max_speed = USB_SPEED_LOW;
+		break;
+	case DWC2_SPEED_PARAM_FULL:
+		hsotg->gadget.max_speed = USB_SPEED_FULL;
+		break;
+	default:
+		hsotg->gadget.max_speed = USB_SPEED_HIGH;
+		break;
+	}
+
 	hsotg->gadget.ops = &dwc2_hsotg_gadget_ops;
 	hsotg->gadget.name = dev_name(dev);
 	hsotg->gadget.otg_caps = &hsotg->params.otg_caps;
Signature · 1 line
2.34.1
author maintainer
#392 lore raw

PATCH 392/1039 usb: gadget: u_audio: fix calculations for small bInterval

In reply to Greg Kroah-Hartman

From: John Keeping <[email protected]>

[ Upstream commit f2f69bf65df12176843ca11eab99949ba69e128b ]

If bInterval is 1, then p_interval is 8000 and p_interval_mil is 8E9, which is too big for a 32-bit value. While the storage is indeed 64-bit, this value is used as the divisor in do_div() which will truncate it into a uint32_t leading to incorrect calculated values.

Switch back to keeping the base value in struct snd_uac_chip which fits easily into an int, meaning that the division can be done in two steps with the divisor fitting safely into a uint32_t on both steps.

Fixes 6fec018a7e70 ("usb: gadget: u_audio.c: Adding Playback Pitch ctl for sync playback")
---
 drivers/usb/gadget/function/u_audio.c | 24 +++++++++++++-----------
 1 file changed, 13 insertions(+), 11 deletions(-)
drivers/usb/gadget/function/u_audio.c +13 -11
index c46400be54641..4fb05f9576a67 100644
@@ -76,8 +76,8 @@ struct snd_uac_chip {
 	struct snd_pcm *pcm;
 
 	/* pre-calculated values for playback iso completion */
-	unsigned long long p_interval_mil;
 	unsigned long long p_residue_mil;
+	unsigned int p_interval;
 	unsigned int p_framesize;
 };
 
@@ -194,21 +194,24 @@ static void u_audio_iso_complete(struct usb_ep *ep, struct usb_request *req)
 		 * If there is a residue from this division, add it to the
 		 * residue accumulator.
 		 */
+		unsigned long long p_interval_mil = uac->p_interval * 1000000ULL;
+
 		pitched_rate_mil = (unsigned long long)
 				params->p_srate * prm->pitch;
 		div_result = pitched_rate_mil;
-		do_div(div_result, uac->p_interval_mil);
+		do_div(div_result, uac->p_interval);
+		do_div(div_result, 1000000);
 		frames = (unsigned int) div_result;
 
 		pr_debug("p_srate %d, pitch %d, interval_mil %llu, frames %d\n",
-				params->p_srate, prm->pitch, uac->p_interval_mil, frames);
+				params->p_srate, prm->pitch, p_interval_mil, frames);
 
 		p_pktsize = min_t(unsigned int,
 					uac->p_framesize * frames,
 					ep->maxpacket);
 
 		if (p_pktsize < ep->maxpacket) {
-			residue_frames_mil = pitched_rate_mil - frames * uac->p_interval_mil;
+			residue_frames_mil = pitched_rate_mil - frames * p_interval_mil;
 			p_pktsize_residue_mil = uac->p_framesize * residue_frames_mil;
 		} else
 			p_pktsize_residue_mil = 0;
@@ -222,11 +225,11 @@ static void u_audio_iso_complete(struct usb_ep *ep, struct usb_request *req)
 		 * size and decrease the accumulator.
 		 */
 		div_result = uac->p_residue_mil;
-		do_div(div_result, uac->p_interval_mil);
+		do_div(div_result, uac->p_interval);
+		do_div(div_result, 1000000);
 		if ((unsigned int) div_result >= uac->p_framesize) {
 			req->length += uac->p_framesize;
-			uac->p_residue_mil -= uac->p_framesize *
-					   uac->p_interval_mil;
+			uac->p_residue_mil -= uac->p_framesize * p_interval_mil;
 			pr_debug("increased req length to %d\n", req->length);
 		}
 		pr_debug("remains uac->p_residue_mil %llu\n", uac->p_residue_mil);
@@ -591,7 +594,7 @@ int u_audio_start_playback(struct g_audio *audio_dev)
 	unsigned int factor;
 	const struct usb_endpoint_descriptor *ep_desc;
 	int req_len, i;
-	unsigned int p_interval, p_pktsize;
+	unsigned int p_pktsize;
 
 	ep = audio_dev->in_ep;
 	prm = &uac->p_prm;
@@ -612,11 +615,10 @@ int u_audio_start_playback(struct g_audio *audio_dev)
 	/* pre-compute some values for iso_complete() */
 	uac->p_framesize = params->p_ssize *
 			    num_channels(params->p_chmask);
-	p_interval = factor / (1 << (ep_desc->bInterval - 1));
-	uac->p_interval_mil = (unsigned long long) p_interval * 1000000;
+	uac->p_interval = factor / (1 << (ep_desc->bInterval - 1));
 	p_pktsize = min_t(unsigned int,
 				uac->p_framesize *
-					(params->p_srate / p_interval),
+					(params->p_srate / uac->p_interval),
 				ep->maxpacket);
 
 	req_len = p_pktsize;
Signature · 1 line
2.34.1
author maintainer
#393 lore raw

PATCH 393/1039 usb: gadget: u_audio: Subdevice 0 for capture ctls

In reply to Greg Kroah-Hartman

From: Pavel Hofman <[email protected]>

[ Upstream commit 601a5bc1aeef772ab1f47582fd322957799f5ab5 ]

Both capture and playback alsa devices use subdevice 0. Yet capture-side ctls are defined for subdevice 1. The patch sets subdevice 0 for them.

Fixes 02de698ca812 ("usb: gadget: u_audio: add bi-directional volume and mute support")
---
 drivers/usb/gadget/function/u_audio.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/usb/gadget/function/u_audio.c +2 -2
index 4fb05f9576a67..4561d7a183ff4 100644
@@ -1147,7 +1147,7 @@ int g_audio_setup(struct g_audio *g_audio, const char *pcm_name,
 			}
 
 			kctl->id.device = pcm->device;
-			kctl->id.subdevice = i;
+			kctl->id.subdevice = 0;
 
 			err = snd_ctl_add(card, kctl);
 			if (err < 0)
@@ -1170,7 +1170,7 @@ int g_audio_setup(struct g_audio *g_audio, const char *pcm_name,
 			}
 
 			kctl->id.device = pcm->device;
-			kctl->id.subdevice = i;
+			kctl->id.subdevice = 0;
 
 
 			kctl->tlv.c = u_audio_volume_tlv;
Signature · 1 line
2.34.1
author maintainer
#394 lore raw

PATCH 394/1039 HID: hid-uclogic-params: Invalid parameter check in uclogic_params_init

In reply to Greg Kroah-Hartman

From: José Expósito <[email protected]>

[ Upstream commit f364c571a5c77e96de2d32062ff019d6b8d2e2bc ]

The function performs a check on its input parameters, however, the hdev parameter is used before the check.

Initialize the stack variables after checking the input parameters to avoid a possible NULL pointer dereference.

Fixes 9614219e9310e ("HID: uclogic: Extract tablet parameter discovery into a module")

Addresses-Coverity-ID: 1443831 ("Null pointer dereference")

---
 drivers/hid/hid-uclogic-params.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)
drivers/hid/hid-uclogic-params.c +9 -4
index adff1bd68d9f8..3c10b858cf74c 100644
@@ -834,10 +834,10 @@ int uclogic_params_init(struct uclogic_params *params,
 			struct hid_device *hdev)
 {
 	int rc;
-	struct usb_device *udev = hid_to_usb_dev(hdev);
-	__u8  bNumInterfaces = udev->config->desc.bNumInterfaces;
-	struct usb_interface *iface = to_usb_interface(hdev->dev.parent);
-	__u8 bInterfaceNumber = iface->cur_altsetting->desc.bInterfaceNumber;
+	struct usb_device *udev;
+	__u8  bNumInterfaces;
+	struct usb_interface *iface;
+	__u8 bInterfaceNumber;
 	bool found;
 	/* The resulting parameters (noop) */
 	struct uclogic_params p = {0, };
@@ -848,6 +848,11 @@ int uclogic_params_init(struct uclogic_params *params,
 		goto cleanup;
 	}
 
+	udev = hid_to_usb_dev(hdev);
+	bNumInterfaces = udev->config->desc.bNumInterfaces;
+	iface = to_usb_interface(hdev->dev.parent);
+	bInterfaceNumber = iface->cur_altsetting->desc.bInterfaceNumber;
+
 	/*
 	 * Set replacement report descriptor if the original matches the
 	 * specified size. Otherwise keep interface unchanged.
Signature · 1 line
2.34.1
author maintainer
#395 lore raw

PATCH 395/1039 HID: hid-uclogic-params: Invalid parameter check in uclogic_params_get_str_desc

In reply to Greg Kroah-Hartman

From: José Expósito <[email protected]>

[ Upstream commit 0a94131d6920916ccb6a357037c535533af08819 ]

The function performs a check on the hdev input parameters, however, it is used before the check.

Initialize the udev variable after the sanity check to avoid a possible NULL pointer dereference.

Fixes 9614219e9310e ("HID: uclogic: Extract tablet parameter discovery into a module")

Addresses-Coverity-ID: 1443827 ("Null pointer dereference")

---
 drivers/hid/hid-uclogic-params.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/hid/hid-uclogic-params.c +3 -1
index 3c10b858cf74c..3a83e2c39b4fb 100644
@@ -66,7 +66,7 @@ static int uclogic_params_get_str_desc(__u8 **pbuf, struct hid_device *hdev,
 					__u8 idx, size_t len)
 {
 	int rc;
-	struct usb_device *udev = hid_to_usb_dev(hdev);
+	struct usb_device *udev;
 	__u8 *buf = NULL;
 
 	/* Check arguments */
@@ -75,6 +75,8 @@ static int uclogic_params_get_str_desc(__u8 **pbuf, struct hid_device *hdev,
 		goto cleanup;
 	}
 
+	udev = hid_to_usb_dev(hdev);
+
 	buf = kmalloc(len, GFP_KERNEL);
 	if (buf == NULL) {
 		rc = -ENOMEM;
Signature · 1 line
2.34.1
author maintainer
#396 lore raw

PATCH 396/1039 HID: hid-uclogic-params: Invalid parameter check in uclogic_params_huion_init

In reply to Greg Kroah-Hartman

From: José Expósito <[email protected]>

[ Upstream commit ff6b548afe4d9d1ff3a0f6ef79e8cbca25d8f905 ]

The function performs a check on its input parameters, however, the hdev parameter is used before the check.

Initialize the stack variables after checking the input parameters to avoid a possible NULL pointer dereference.

Fixes 9614219e9310e ("HID: uclogic: Extract tablet parameter discovery into a module")

Addresses-Coverity-ID: 1443804 ("Null pointer dereference")

---
 drivers/hid/hid-uclogic-params.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)
drivers/hid/hid-uclogic-params.c +7 -3
index 3a83e2c39b4fb..4136837e4d158 100644
@@ -709,9 +709,9 @@ static int uclogic_params_huion_init(struct uclogic_params *params,
 				     struct hid_device *hdev)
 {
 	int rc;
-	struct usb_device *udev = hid_to_usb_dev(hdev);
-	struct usb_interface *iface = to_usb_interface(hdev->dev.parent);
-	__u8 bInterfaceNumber = iface->cur_altsetting->desc.bInterfaceNumber;
+	struct usb_device *udev;
+	struct usb_interface *iface;
+	__u8 bInterfaceNumber;
 	bool found;
 	/* The resulting parameters (noop) */
 	struct uclogic_params p = {0, };
@@ -725,6 +725,10 @@ static int uclogic_params_huion_init(struct uclogic_params *params,
 		goto cleanup;
 	}
 
+	udev = hid_to_usb_dev(hdev);
+	iface = to_usb_interface(hdev->dev.parent);
+	bInterfaceNumber = iface->cur_altsetting->desc.bInterfaceNumber;
+
 	/* If it's not a pen interface */
 	if (bInterfaceNumber != 0) {
 		/* TODO: Consider marking the interface invalid */
Signature · 1 line
2.34.1
author maintainer
#397 lore raw

PATCH 397/1039 HID: hid-uclogic-params: Invalid parameter check in uclogic_params_frame_init_v1_buttonpad

In reply to Greg Kroah-Hartman

From: José Expósito <[email protected]>

[ Upstream commit aa320fdbbbb482c19100f51461bd0069753ce3d7 ]

The function performs a check on the hdev input parameters, however, it is used before the check.

Initialize the udev variable after the sanity check to avoid a possible NULL pointer dereference.

Fixes 9614219e9310e ("HID: uclogic: Extract tablet parameter discovery into a module")

Addresses-Coverity-ID: 1443763 ("Null pointer dereference")

---
 drivers/hid/hid-uclogic-params.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/hid/hid-uclogic-params.c +3 -1
index 4136837e4d158..3e70f969fb849 100644
@@ -452,7 +452,7 @@ static int uclogic_params_frame_init_v1_buttonpad(
 {
 	int rc;
 	bool found = false;
-	struct usb_device *usb_dev = hid_to_usb_dev(hdev);
+	struct usb_device *usb_dev;
 	char *str_buf = NULL;
 	const size_t str_len = 16;
 
@@ -462,6 +462,8 @@ static int uclogic_params_frame_init_v1_buttonpad(
 		goto cleanup;
 	}
 
+	usb_dev = hid_to_usb_dev(hdev);
+
 	/*
 	 * Enable generic button mode
 	 */
Signature · 1 line
2.34.1
author maintainer
#398 lore raw

PATCH 398/1039 debugfs: lockdown: Allow reading debugfs files that are not world readable

In reply to Greg Kroah-Hartman

From: Michal Suchanek <[email protected]>

[ Upstream commit 358fcf5ddbec4e6706405847d6a666f5933a6c25 ]

When the kernel is locked down the kernel allows reading only debugfs files with mode 444. Mode 400 is also valid but is not allowed.

Make the 444 into a mask.

Fixes 5496197f9b08 ("debugfs: Restrict debugfs when the kernel is locked down")
---
 fs/debugfs/file.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/debugfs/file.c +1 -1
index 7d162b0efbf03..950c63fa4d0b2 100644
@@ -147,7 +147,7 @@ static int debugfs_locked_down(struct inode *inode,
 			       struct file *filp,
 			       const struct file_operations *real_fops)
 {
-	if ((inode->i_mode & 07777) == 0444 &&
+	if ((inode->i_mode & 07777 & ~0444) == 0 &&
 	    !(filp->f_mode & FMODE_WRITE) &&
 	    !real_fops->unlocked_ioctl &&
 	    !real_fops->compat_ioctl &&
Signature · 1 line
2.34.1
author maintainer
#399 lore raw

PATCH 399/1039 drivers/firmware: Add missing platform_device_put() in sysfb_create_simplefb

In reply to Greg Kroah-Hartman

From: Miaoqian Lin <[email protected]>

[ Upstream commit 0589e8889dce8e0f0ea5bbf757f38865e2a469c1 ]

Add the missing platform_device_put() before return from sysfb_create_simplefb() in the error handling case.

Fixes 8633ef82f101 ("drivers/firmware: consolidate EFI framebuffer setup for all arches")
---
 drivers/firmware/sysfb_simplefb.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)
drivers/firmware/sysfb_simplefb.c +6 -2
index b86761904949c..303a491e520d1 100644
@@ -113,12 +113,16 @@ __init int sysfb_create_simplefb(const struct screen_info *si,
 	sysfb_apply_efi_quirks(pd);
 
 	ret = platform_device_add_resources(pd, &res, 1);
-	if (ret)
+	if (ret) {
+		platform_device_put(pd);
 		return ret;
+	}
 
 	ret = platform_device_add_data(pd, mode, sizeof(*mode));
-	if (ret)
+	if (ret) {
+		platform_device_put(pd);
 		return ret;
+	}
 
 	return platform_device_add(pd);
 }
Signature · 1 line
2.34.1
author maintainer
#400 lore raw

PATCH 400/1039 serial: liteuart: fix MODULE_ALIAS

In reply to Greg Kroah-Hartman

From: Alyssa Ross <[email protected]>

[ Upstream commit 556172fabd226ba14b70c1740d0826a4717473dc ]

modprobe can't handle spaces in aliases.

Fixes 1da81e5562fa ("drivers/tty/serial: add LiteUART driver")
---
 drivers/tty/serial/liteuart.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/tty/serial/liteuart.c +1 -1
index 2941659e52747..7f74bf7bdcff8 100644
@@ -436,4 +436,4 @@ module_exit(liteuart_exit);
 MODULE_AUTHOR("Antmicro <www.antmicro.com>");
 MODULE_DESCRIPTION("LiteUART serial driver");
 MODULE_LICENSE("GPL v2");
-MODULE_ALIAS("platform: liteuart");
+MODULE_ALIAS("platform:liteuart");
Signature · 1 line
2.34.1
author maintainer
#401 lore raw

PATCH 401/1039 serial: stm32: move tx dma terminate DMA to shutdown

In reply to Greg Kroah-Hartman

From: Valentin Caron <[email protected]>

[ Upstream commit 56a23f9319e86e1d62a109896e2c7e52c414e67d ]

Terminate DMA transaction and clear CR3_DMAT when shutdown is requested, instead of when remove is requested. If DMA transfer is not stopped in shutdown ops, driver will fail to start a new DMA transfer after next startup ops.

---
 drivers/tty/serial/stm32-usart.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
drivers/tty/serial/stm32-usart.c +5 -1
index 3244e7f6818ca..6cfc3bec67492 100644
@@ -883,6 +883,11 @@ static void stm32_usart_shutdown(struct uart_port *port)
 	u32 val, isr;
 	int ret;
 
+	if (stm32_port->tx_dma_busy) {
+		dmaengine_terminate_async(stm32_port->tx_ch);
+		stm32_usart_clr_bits(port, ofs->cr3, USART_CR3_DMAT);
+	}
+
 	/* Disable modem control interrupts */
 	stm32_usart_disable_ms(port);
 
@@ -1570,7 +1575,6 @@ static int stm32_usart_serial_remove(struct platform_device *pdev)
 	writel_relaxed(cr3, port->membase + ofs->cr3);
 
 	if (stm32_port->tx_ch) {
-		dmaengine_terminate_async(stm32_port->tx_ch);
 		stm32_usart_of_dma_tx_remove(stm32_port, pdev);
 		dma_release_channel(stm32_port->tx_ch);
 	}
Signature · 1 line
2.34.1
author maintainer
#402 lore raw

PATCH 402/1039 spi: qcom: geni: set the error code for gpi transfer

In reply to Greg Kroah-Hartman

From: Vinod Koul <[email protected]>

[ Upstream commit 74b86d6af81be73bb74995ebeba74417e84b6b6f ]

Before we invoke spi_finalize_current_transfer() in spi_gsi_callback_result() we should set the spi->cur_msg->status as appropriate (0 for success, error otherwise).

The helps to return error on transfer and not wait till it timesout on error

Fixes b59c122484ec ("spi: spi-geni-qcom: Add support for GPI dma")
---
 drivers/spi/spi-geni-qcom.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
drivers/spi/spi-geni-qcom.c +5 -1
index e2affaee4e769..69e71aac85129 100644
@@ -350,17 +350,21 @@ spi_gsi_callback_result(void *cb, const struct dmaengine_result *result)
 {
 	struct spi_master *spi = cb;
 
+	spi->cur_msg->status = -EIO;
 	if (result->result != DMA_TRANS_NOERROR) {
 		dev_err(&spi->dev, "DMA txn failed: %d\n", result->result);
+		spi_finalize_current_transfer(spi);
 		return;
 	}
 
 	if (!result->residue) {
+		spi->cur_msg->status = 0;
 		dev_dbg(&spi->dev, "DMA txn completed\n");
-		spi_finalize_current_transfer(spi);
 	} else {
 		dev_err(&spi->dev, "DMA xfer has pending: %d\n", result->residue);
 	}
+
+	spi_finalize_current_transfer(spi);
 }
 
 static int setup_gsi_xfer(struct spi_transfer *xfer, struct spi_geni_master *mas,
Signature · 1 line
2.34.1
author maintainer
#403 lore raw

PATCH 403/1039 spi: qcom: geni: handle timeout for gpi mode

In reply to Greg Kroah-Hartman

From: Vinod Koul <[email protected]>

[ Upstream commit f8039ea55d4ccac2238a247a574f0acb3bc1dc4b ]

We missed adding handle_err for gpi mode, so add a new function spi_geni_handle_err() which would call handle_fifo_timeout() or newly added handle_gpi_timeout() based on mode

---
 drivers/spi/spi-geni-qcom.c | 26 +++++++++++++++++++++++++-
 1 file changed, 25 insertions(+), 1 deletion(-)
drivers/spi/spi-geni-qcom.c +25 -1
index 69e71aac85129..079d0cb783ee3 100644
@@ -168,6 +168,30 @@ static void handle_fifo_timeout(struct spi_master *spi,
 	}
 }
 
+static void handle_gpi_timeout(struct spi_master *spi, struct spi_message *msg)
+{
+	struct spi_geni_master *mas = spi_master_get_devdata(spi);
+
+	dmaengine_terminate_sync(mas->tx);
+	dmaengine_terminate_sync(mas->rx);
+}
+
+static void spi_geni_handle_err(struct spi_master *spi, struct spi_message *msg)
+{
+	struct spi_geni_master *mas = spi_master_get_devdata(spi);
+
+	switch (mas->cur_xfer_mode) {
+	case GENI_SE_FIFO:
+		handle_fifo_timeout(spi, msg);
+		break;
+	case GENI_GPI_DMA:
+		handle_gpi_timeout(spi, msg);
+		break;
+	default:
+		dev_err(mas->dev, "Abort on Mode:%d not supported", mas->cur_xfer_mode);
+	}
+}
+
 static bool spi_geni_is_abort_still_pending(struct spi_geni_master *mas)
 {
 	struct geni_se *se = &mas->se;
@@ -926,7 +950,7 @@ static int spi_geni_probe(struct platform_device *pdev)
 	spi->can_dma = geni_can_dma;
 	spi->dma_map_dev = dev->parent;
 	spi->auto_runtime_pm = true;
-	spi->handle_err = handle_fifo_timeout;
+	spi->handle_err = spi_geni_handle_err;
 	spi->use_gpio_descriptors = true;
 
 	init_completion(&mas->cs_done);
Signature · 1 line
2.34.1
author maintainer
#404 lore raw

PATCH 404/1039 x86, sched: Fix undefined reference to init_freq_invariance_cppc() build error

In reply to Greg Kroah-Hartman

From: Huang Rui <[email protected]>

[ Upstream commit 6c4ab1b86dac3954d15c00c1a6396d60a1023fab ]

The init_freq_invariance_cppc function is implemented in smpboot and depends on CONFIG_SMP.

MODPOST vmlinux.symvers
MODINFO modules.builtin.modinfo
GEN modules.builtin
LD .tmp_vmlinux.kallsyms1
ld: drivers/acpi/cppc_acpi.o: in function `acpi_cppc_processor_probe':
/home/ray/brahma3/linux/drivers/acpi/cppc_acpi.c:819: undefined reference to `init_freq_invariance_cppc'
make: *** [Makefile:1161: vmlinux] Error 1

See https://lore.kernel.org/lkml/[email protected]/.

Fixes 41ea667227ba ("x86, sched: Calculate frequency invariance for AMD systems")
[ rjw: Subject edits ]
---
 arch/x86/include/asm/topology.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/include/asm/topology.h +1 -1
index cc164777e6619..2f0b6be8eaabc 100644
@@ -221,7 +221,7 @@ static inline void arch_set_max_freq_ratio(bool turbo_disabled)
 }
 #endif
 
-#ifdef CONFIG_ACPI_CPPC_LIB
+#if defined(CONFIG_ACPI_CPPC_LIB) && defined(CONFIG_SMP)
 void init_freq_invariance_cppc(void);
 #define init_freq_invariance_cppc init_freq_invariance_cppc
 #endif
Signature · 1 line
2.34.1
author maintainer
#405 lore raw

PATCH 405/1039 net/mlx5e: Fix page DMA map/unmap attributes

In reply to Greg Kroah-Hartman

From: Aya Levin <[email protected]>

[ Upstream commit 0b7cfa4082fbf550595bc0e40f05614bd83bf0cd ]

Driver initiates DMA sync, hence it may skip CPU sync. Add DMA_ATTR_SKIP_CPU_SYNC as input attribute both to dma_map_page and dma_unmap_page to avoid redundant sync with the CPU.
When forcing the device to work with SWIOTLB, the extra sync might cause data corruption. The driver unmaps the whole page while the hardware used just a part of the bounce buffer. So syncing overrides the entire page with bounce buffer that only partially contains real data.

Fixes bc77b240b3c5 ("net/mlx5e: Add fragmented memory support for RX multi packet WQE")
Fixes db05815b36cb ("net/mlx5e: Add XSK zero-copy support")
---
 drivers/net/ethernet/mellanox/mlx5/core/en/xsk/pool.c | 4 ++--
 drivers/net/ethernet/mellanox/mlx5/core/en_rx.c       | 7 ++++---
 2 files changed, 6 insertions(+), 5 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/en/xsk/pool.c +2 -2
index 7b562d2c8a196..279cd8f4e79f7 100644
@@ -11,13 +11,13 @@ static int mlx5e_xsk_map_pool(struct mlx5e_priv *priv,
 {
 	struct device *dev = mlx5_core_dma_dev(priv->mdev);
 
-	return xsk_pool_dma_map(pool, dev, 0);
+	return xsk_pool_dma_map(pool, dev, DMA_ATTR_SKIP_CPU_SYNC);
 }
 
 static void mlx5e_xsk_unmap_pool(struct mlx5e_priv *priv,
 				 struct xsk_buff_pool *pool)
 {
-	return xsk_pool_dma_unmap(pool, 0);
+	return xsk_pool_dma_unmap(pool, DMA_ATTR_SKIP_CPU_SYNC);
 }
 
 static int mlx5e_xsk_get_pools(struct mlx5e_xsk *xsk)
drivers/net/ethernet/mellanox/mlx5/core/en_rx.c +4 -3
index 793511d5ee4cd..dfc6604b9538b 100644
@@ -278,8 +278,8 @@ static inline int mlx5e_page_alloc_pool(struct mlx5e_rq *rq,
 	if (unlikely(!dma_info->page))
 		return -ENOMEM;
 
-	dma_info->addr = dma_map_page(rq->pdev, dma_info->page, 0,
-				      PAGE_SIZE, rq->buff.map_dir);
+	dma_info->addr = dma_map_page_attrs(rq->pdev, dma_info->page, 0, PAGE_SIZE,
+					    rq->buff.map_dir, DMA_ATTR_SKIP_CPU_SYNC);
 	if (unlikely(dma_mapping_error(rq->pdev, dma_info->addr))) {
 		page_pool_recycle_direct(rq->page_pool, dma_info->page);
 		dma_info->page = NULL;
@@ -300,7 +300,8 @@ static inline int mlx5e_page_alloc(struct mlx5e_rq *rq,
 
 void mlx5e_page_dma_unmap(struct mlx5e_rq *rq, struct mlx5e_dma_info *dma_info)
 {
-	dma_unmap_page(rq->pdev, dma_info->addr, PAGE_SIZE, rq->buff.map_dir);
+	dma_unmap_page_attrs(rq->pdev, dma_info->addr, PAGE_SIZE, rq->buff.map_dir,
+			     DMA_ATTR_SKIP_CPU_SYNC);
 }
 
 void mlx5e_page_release_dynamic(struct mlx5e_rq *rq,
Signature · 1 line
2.34.1
author maintainer
#406 lore raw

PATCH 406/1039 net/mlx5e: Fix nullptr on deleting mirroring rule

In reply to Greg Kroah-Hartman

From: Dima Chumak <[email protected]>

[ Upstream commit de31854ece175e12ff3c35d07f340988823aed34 ]

Deleting a Tc rule with multiple outputs, one of which is internal port, like this one:

  tc filter del dev enp8s0f0_0 ingress protocol ip pref 5 flower \
      dst_mac 0c:42:a1:d1:d0:88 \
      src_mac e4:ea:09:08:00:02 \
      action tunnel_key  set \
          src_ip 0.0.0.0 \
          dst_ip 7.7.7.8 \
          id 8 \
          dst_port 4789 \
      action mirred egress mirror dev vxlan_sys_4789 pipe \
      action mirred egress redirect dev enp8s0f0_1

Triggers a call trace:

  BUG: kernel NULL pointer dereference, address: 0000000000000230
  RIP: 0010:del_sw_hw_rule+0x2b/0x1f0 [mlx5_core]
  Call Trace:
   tree_remove_node+0x16/0x30 [mlx5_core]
   mlx5_del_flow_rules+0x51/0x160 [mlx5_core]
   __mlx5_eswitch_del_rule+0x4b/0x170 [mlx5_core]
   mlx5e_tc_del_fdb_flow+0x295/0x550 [mlx5_core]
   mlx5e_flow_put+0x1f/0x70 [mlx5_core]
   mlx5e_delete_flower+0x286/0x390 [mlx5_core]
   tc_setup_cb_destroy+0xac/0x170
   fl_hw_destroy_filter+0x94/0xc0 [cls_flower]
   __fl_delete+0x15e/0x170 [cls_flower]
   fl_delete+0x36/0x80 [cls_flower]
   tc_del_tfilter+0x3a6/0x6e0
   rtnetlink_rcv_msg+0xe5/0x360
   ? rtnl_calcit.isra.0+0x110/0x110
   netlink_rcv_skb+0x46/0x110
   netlink_unicast+0x16b/0x200
   netlink_sendmsg+0x202/0x3d0
   sock_sendmsg+0x33/0x40
   ____sys_sendmsg+0x1c3/0x200
   ? copy_msghdr_from_user+0xd6/0x150
   ___sys_sendmsg+0x88/0xd0
   ? ___sys_recvmsg+0x88/0xc0
   ? do_futex+0x10c/0x460
   __sys_sendmsg+0x59/0xa0
   do_syscall_64+0x48/0x140
   entry_SYSCALL_64_after_hwframe+0x44/0xa9

Fix by disabling offloading for flows matching esw_is_chain_src_port_rewrite() which have more than one output.

Fixes 10742efc20a4 ("net/mlx5e: VF tunnel TX traffic offloading")
---
 .../mellanox/mlx5/core/eswitch_offloads.c     | 28 ++++++++++---------
 1 file changed, 15 insertions(+), 13 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +15 -13
index 32bc08a399256..ccb66428aeb5b 100644
@@ -295,26 +295,28 @@ esw_setup_chain_src_port_rewrite(struct mlx5_flow_destination *dest,
 				 int *i)
 {
 	struct mlx5_esw_flow_attr *esw_attr = attr->esw_attr;
-	int j, err;
+	int err;
 
 	if (!(attr->flags & MLX5_ESW_ATTR_FLAG_SRC_REWRITE))
 		return -EOPNOTSUPP;
 
-	for (j = esw_attr->split_count; j < esw_attr->out_count; j++, (*i)++) {
-		err = esw_setup_chain_dest(dest, flow_act, chains, attr->dest_chain, 1, 0, *i);
-		if (err)
-			goto err_setup_chain;
+	/* flow steering cannot handle more than one dest with the same ft
+	 * in a single flow
+	 */
+	if (esw_attr->out_count - esw_attr->split_count > 1)
+		return -EOPNOTSUPP;
 
-		if (esw_attr->dests[j].pkt_reformat) {
-			flow_act->action |= MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT;
-			flow_act->pkt_reformat = esw_attr->dests[j].pkt_reformat;
-		}
+	err = esw_setup_chain_dest(dest, flow_act, chains, attr->dest_chain, 1, 0, *i);
+	if (err)
+		return err;
+
+	if (esw_attr->dests[esw_attr->split_count].pkt_reformat) {
+		flow_act->action |= MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT;
+		flow_act->pkt_reformat = esw_attr->dests[esw_attr->split_count].pkt_reformat;
 	}
-	return 0;
+	(*i)++;
 
-err_setup_chain:
-	esw_put_dest_tables_loop(esw, attr, esw_attr->split_count, j);
-	return err;
+	return 0;
 }
 
 static void esw_cleanup_chain_src_port_rewrite(struct mlx5_eswitch *esw,
Signature · 1 line
2.34.1
author maintainer
#407 lore raw

PATCH 407/1039 net/mlx5e: Fix wrong usage of fib_info_nh when routes with nexthop objects are used

In reply to Greg Kroah-Hartman

From: Maor Dickman <[email protected]>

[ Upstream commit 885751eb1b01d276e38f57d78c583e4ce006c5ed ]

Creating routes with nexthop objects while in switchdev mode leads to access to un-allocated memory and trigger bellow call trace due to hitting WARN_ON. This is caused due to illegal usage of fib_info_nh in TC tunnel FIB event handling to resolve the FIB device while fib_info built in with nexthop.

Fixed by ignoring attempts to use nexthop objects with routes until support can be properly added.

WARNING: CPU: 1 PID: 1724 at include/net/nexthop.h:468 mlx5e_tc_tun_fib_event+0x448/0x570 [mlx5_core]
CPU: 1 PID: 1724 Comm: ip Not tainted 5.15.0_for_upstream_min_debug_2021_11_09_02_04 #1
RIP: 0010:mlx5e_tc_tun_fib_event+0x448/0x570 [mlx5_core]
RSP: 0018:ffff8881349f7910 EFLAGS: 00010202
RAX: ffff8881492f1980 RBX: ffff8881349f79e8 RCX: 0000000000000000
RDX: ffff8881349f79e8 RSI: 0000000000000000 RDI: 0000000000000000
RBP: ffff8881349f7950 R08: 00000000000000fe R09: 0000000000000001
R10: 0000000000000000 R11: 0000000000000000 R12: ffff88811e9d0000
R13: ffff88810eb62000 R14: ffff888106710268 R15: 0000000000000018
FS:  00007f1d5ca6e800(0000) GS:ffff88852c880000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007ffedba44ff8 CR3: 0000000129808004 CR4: 0000000000370ea0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 <TASK>
 atomic_notifier_call_chain+0x42/0x60
 call_fib_notifiers+0x21/0x40
 fib_table_insert+0x479/0x6d0
 ? try_charge_memcg+0x480/0x6d0
 inet_rtm_newroute+0x65/0xb0
 rtnetlink_rcv_msg+0x2af/0x360
 ? page_add_file_rmap+0x13/0x130
 ? do_set_pte+0xcd/0x120
 ? rtnl_calcit.isra.0+0x120/0x120
 netlink_rcv_skb+0x4e/0xf0
 netlink_unicast+0x1ee/0x2b0
 netlink_sendmsg+0x22e/0x460
 sock_sendmsg+0x33/0x40
 ____sys_sendmsg+0x1d1/0x1f0
 ___sys_sendmsg+0xab/0xf0
 ? __mod_memcg_lruvec_state+0x40/0x60
 ? __mod_lruvec_page_state+0x95/0xd0
 ? page_add_new_anon_rmap+0x4e/0xf0
 ? __handle_mm_fault+0xec6/0x1470
 __sys_sendmsg+0x51/0x90
 ? internal_get_user_pages_fast+0x480/0xa10
 do_syscall_64+0x3d/0x90
 entry_SYSCALL_64_after_hwframe+0x44/0xae
Fixes 8914add2c9e5 ("net/mlx5e: Handle FIB events to update tunnel endpoint device")
---
 drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c +2 -0
index 042b1abe1437f..62cbd15ffc341 100644
@@ -1579,6 +1579,8 @@ mlx5e_init_fib_work_ipv4(struct mlx5e_priv *priv,
 	struct net_device *fib_dev;
 
 	fen_info = container_of(info, struct fib_entry_notifier_info, info);
+	if (fen_info->fi->nh)
+		return NULL;
 	fib_dev = fib_info_nh(fen_info->fi, 0)->fib_nh_dev;
 	if (!fib_dev || fib_dev->netdev_ops != &mlx5e_netdev_ops ||
 	    fen_info->dst_len != 32)
Signature · 1 line
2.34.1
author maintainer
#408 lore raw

PATCH 408/1039 net/mlx5e: Dont block routes with nexthop objects in SW

In reply to Greg Kroah-Hartman

From: Maor Dickman <[email protected]>

[ Upstream commit 9e72a55a3c9d54b38a704bb7292d984574a81d9d ]

Routes with nexthop objects is currently not supported by multipath offload and any attempts to use it is blocked, however this also block adding SW routes with nexthop.

Resolve this by returning NOTIFY_DONE instead of an error which will allow such a route to be created in SW but not offloaded.

This fix also solve an issue which block adding such routes on different devices due to missing check if the route FIB device is one of multipath devices.

Fixes 6a87afc072c3 ("mlx5: Fail attempts to use routes with nexthop objects")
---
 drivers/net/ethernet/mellanox/mlx5/core/lag/mp.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/lag/mp.c +2 -4
index bf4d3cbefa633..1ca01a5b6cdd8 100644
@@ -268,10 +268,8 @@ static int mlx5_lag_fib_event(struct notifier_block *nb,
 		fen_info = container_of(info, struct fib_entry_notifier_info,
 					info);
 		fi = fen_info->fi;
-		if (fi->nh) {
-			NL_SET_ERR_MSG_MOD(info->extack, "IPv4 route with nexthop objects is not supported");
-			return notifier_from_errno(-EINVAL);
-		}
+		if (fi->nh)
+			return NOTIFY_DONE;
 		fib_dev = fib_info_nh(fen_info->fi, 0)->fib_nh_dev;
 		if (fib_dev != ldev->pf[MLX5_LAG_P1].netdev &&
 		    fib_dev != ldev->pf[MLX5_LAG_P2].netdev) {
Signature · 1 line
2.34.1
author maintainer
#409 lore raw

PATCH 409/1039 Revert "net/mlx5e: Block offload of outer header csum for UDP tunnels"

In reply to Greg Kroah-Hartman

From: Aya Levin <[email protected]>

[ Upstream commit 64050cdad0983ad8060e33c3f4b5aee2366bcebd ]

This reverts commit 6d6727dddc7f93fcc155cb8d0c49c29ae0e71122.

Although the NIC doesn't support offload of outer header CSUM, using gso_partial_features allows offloading the tunnel's segmentation. The driver relies on the stack CSUM calculation of the outer header. For this, NETIF_F_GSO_UDP_TUNNEL_CSUM must be a member of the device's features.

Fixes 6d6727dddc7f ("net/mlx5e: Block offload of outer header csum for UDP tunnels")
---
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/en_main.c +7 -3
index 41379844eee1f..de8acd3217c18 100644
@@ -4789,9 +4789,13 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev)
 	}
 
 	if (mlx5_vxlan_allowed(mdev->vxlan) || mlx5_geneve_tx_allowed(mdev)) {
-		netdev->hw_features     |= NETIF_F_GSO_UDP_TUNNEL;
-		netdev->hw_enc_features |= NETIF_F_GSO_UDP_TUNNEL;
-		netdev->vlan_features |= NETIF_F_GSO_UDP_TUNNEL;
+		netdev->hw_features     |= NETIF_F_GSO_UDP_TUNNEL |
+					   NETIF_F_GSO_UDP_TUNNEL_CSUM;
+		netdev->hw_enc_features |= NETIF_F_GSO_UDP_TUNNEL |
+					   NETIF_F_GSO_UDP_TUNNEL_CSUM;
+		netdev->gso_partial_features = NETIF_F_GSO_UDP_TUNNEL_CSUM;
+		netdev->vlan_features |= NETIF_F_GSO_UDP_TUNNEL |
+					 NETIF_F_GSO_UDP_TUNNEL_CSUM;
 	}
 
 	if (mlx5e_tunnel_proto_supported_tx(mdev, IPPROTO_GRE)) {
Signature · 1 line
2.34.1
author maintainer
#410 lore raw

PATCH 410/1039 Revert "net/mlx5e: Block offload of outer header csum for GRE tunnel"

In reply to Greg Kroah-Hartman

From: Aya Levin <[email protected]>

[ Upstream commit 01c3fd113ef50490ffd43f78f347ef6bb008510b ]

This reverts commit 54e1217b90486c94b26f24dcee1ee5ef5372f832.

Although the NIC doesn't support offload of outer header CSUM, using gso_partial_features allows offloading the tunnel's segmentation. The driver relies on the stack CSUM calculation of the outer header. For this, NETIF_F_GSO_GRE_CSUM must be a member of the device's features.

Fixes 54e1217b9048 ("net/mlx5e: Block offload of outer header csum for GRE tunnel")
---
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/en_main.c +6 -3
index de8acd3217c18..d92b82cdfd4e1 100644
@@ -4799,9 +4799,12 @@ static void mlx5e_build_nic_netdev(struct net_device *netdev)
 	}
 
 	if (mlx5e_tunnel_proto_supported_tx(mdev, IPPROTO_GRE)) {
-		netdev->hw_features     |= NETIF_F_GSO_GRE;
-		netdev->hw_enc_features |= NETIF_F_GSO_GRE;
-		netdev->gso_partial_features |= NETIF_F_GSO_GRE;
+		netdev->hw_features     |= NETIF_F_GSO_GRE |
+					   NETIF_F_GSO_GRE_CSUM;
+		netdev->hw_enc_features |= NETIF_F_GSO_GRE |
+					   NETIF_F_GSO_GRE_CSUM;
+		netdev->gso_partial_features |= NETIF_F_GSO_GRE |
+						NETIF_F_GSO_GRE_CSUM;
 	}
 
 	if (mlx5e_tunnel_proto_supported_tx(mdev, IPPROTO_IPIP)) {
Signature · 1 line
2.34.1
author maintainer
#411 lore raw

PATCH 411/1039 net/mlx5e: Fix matching on modified inner ip_ecn bits

In reply to Greg Kroah-Hartman

From: Paul Blakey <[email protected]>

[ Upstream commit b6dfff21a170af5c695ebaa153b7f5e297ddca03 ]

Tunnel device follows RFC 6040, and during decapsulation inner ip_ecn might change depending on inner and outer ip_ecn as follows:

 +---------+----------------------------------------+
 |Arriving |         Arriving Outer Header          |
 |   Inner +---------+---------+---------+----------+
 |  Header | Not-ECT | ECT(0)  | ECT(1)  |   CE     |
 +---------+---------+---------+---------+----------+
 | Not-ECT | Not-ECT | Not-ECT | Not-ECT | <drop>   |
 |  ECT(0) |  ECT(0) | ECT(0)  | ECT(1)  |   CE*    |
 |  ECT(1) |  ECT(1) | ECT(1)  | ECT(1)* |   CE*    |
 |    CE   |   CE    |  CE     | CE      |   CE     |
 +---------+---------+---------+---------+----------+

Cells marked above are changed from original inner packet ip_ecn value.

Tc then matches on the modified inner ip_ecn, but hw offload which matches the inner ip_ecn value before decap, will fail.

Fix that by mapping all the cases of outer and inner ip_ecn matching, and only supporting cases where we know inner wouldn't be changed by decap, or in the outer ip_ecn=CE case, inner ip_ecn didn't matter.

Fixes bcef735c59f2 ("net/mlx5e: Offload TC matching on tos/ttl for ip tunnels")
---
 .../net/ethernet/mellanox/mlx5/core/en_tc.c   | 120 +++++++++++++++++-
 1 file changed, 116 insertions(+), 4 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +116 -4
index 5e454a14428f2..9b3adaccc9beb 100644
@@ -1949,6 +1949,111 @@ u8 mlx5e_tc_get_ip_version(struct mlx5_flow_spec *spec, bool outer)
 	return ip_version;
 }
 
+/* Tunnel device follows RFC 6040, see include/net/inet_ecn.h.
+ * And changes inner ip_ecn depending on inner and outer ip_ecn as follows:
+ *      +---------+----------------------------------------+
+ *      |Arriving |         Arriving Outer Header          |
+ *      |   Inner +---------+---------+---------+----------+
+ *      |  Header | Not-ECT | ECT(0)  | ECT(1)  |   CE     |
+ *      +---------+---------+---------+---------+----------+
+ *      | Not-ECT | Not-ECT | Not-ECT | Not-ECT | <drop>   |
+ *      |  ECT(0) |  ECT(0) | ECT(0)  | ECT(1)  |   CE*    |
+ *      |  ECT(1) |  ECT(1) | ECT(1)  | ECT(1)* |   CE*    |
+ *      |    CE   |   CE    |  CE     | CE      |   CE     |
+ *      +---------+---------+---------+---------+----------+
+ *
+ * Tc matches on inner after decapsulation on tunnel device, but hw offload matches
+ * the inner ip_ecn value before hardware decap action.
+ *
+ * Cells marked are changed from original inner packet ip_ecn value during decap, and
+ * so matching those values on inner ip_ecn before decap will fail.
+ *
+ * The following helper allows offload when inner ip_ecn won't be changed by outer ip_ecn,
+ * except for the outer ip_ecn = CE, where in all cases inner ip_ecn will be changed to CE,
+ * and such we can drop the inner ip_ecn=CE match.
+ */
+
+static int mlx5e_tc_verify_tunnel_ecn(struct mlx5e_priv *priv,
+				      struct flow_cls_offload *f,
+				      bool *match_inner_ecn)
+{
+	u8 outer_ecn_mask = 0, outer_ecn_key = 0, inner_ecn_mask = 0, inner_ecn_key = 0;
+	struct flow_rule *rule = flow_cls_offload_flow_rule(f);
+	struct netlink_ext_ack *extack = f->common.extack;
+	struct flow_match_ip match;
+
+	*match_inner_ecn = true;
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ENC_IP)) {
+		flow_rule_match_enc_ip(rule, &match);
+		outer_ecn_key = match.key->tos & INET_ECN_MASK;
+		outer_ecn_mask = match.mask->tos & INET_ECN_MASK;
+	}
+
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_IP)) {
+		flow_rule_match_ip(rule, &match);
+		inner_ecn_key = match.key->tos & INET_ECN_MASK;
+		inner_ecn_mask = match.mask->tos & INET_ECN_MASK;
+	}
+
+	if (outer_ecn_mask != 0 && outer_ecn_mask != INET_ECN_MASK) {
+		NL_SET_ERR_MSG_MOD(extack, "Partial match on enc_tos ecn bits isn't supported");
+		netdev_warn(priv->netdev, "Partial match on enc_tos ecn bits isn't supported");
+		return -EOPNOTSUPP;
+	}
+
+	if (!outer_ecn_mask) {
+		if (!inner_ecn_mask)
+			return 0;
+
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Matching on tos ecn bits without also matching enc_tos ecn bits isn't supported");
+		netdev_warn(priv->netdev,
+			    "Matching on tos ecn bits without also matching enc_tos ecn bits isn't supported");
+		return -EOPNOTSUPP;
+	}
+
+	if (inner_ecn_mask && inner_ecn_mask != INET_ECN_MASK) {
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Partial match on tos ecn bits with match on enc_tos ecn bits isn't supported");
+		netdev_warn(priv->netdev,
+			    "Partial match on tos ecn bits with match on enc_tos ecn bits isn't supported");
+		return -EOPNOTSUPP;
+	}
+
+	if (!inner_ecn_mask)
+		return 0;
+
+	/* Both inner and outer have full mask on ecn */
+
+	if (outer_ecn_key == INET_ECN_ECT_1) {
+		/* inner ecn might change by DECAP action */
+
+		NL_SET_ERR_MSG_MOD(extack, "Match on enc_tos ecn = ECT(1) isn't supported");
+		netdev_warn(priv->netdev, "Match on enc_tos ecn = ECT(1) isn't supported");
+		return -EOPNOTSUPP;
+	}
+
+	if (outer_ecn_key != INET_ECN_CE)
+		return 0;
+
+	if (inner_ecn_key != INET_ECN_CE) {
+		/* Can't happen in software, as packet ecn will be changed to CE after decap */
+		NL_SET_ERR_MSG_MOD(extack,
+				   "Match on tos enc_tos ecn = CE while match on tos ecn != CE isn't supported");
+		netdev_warn(priv->netdev,
+			    "Match on tos enc_tos ecn = CE while match on tos ecn != CE isn't supported");
+		return -EOPNOTSUPP;
+	}
+
+	/* outer ecn = CE, inner ecn = CE, as decap will change inner ecn to CE in anycase,
+	 * drop match on inner ecn
+	 */
+	*match_inner_ecn = false;
+
+	return 0;
+}
+
 static int parse_tunnel_attr(struct mlx5e_priv *priv,
 			     struct mlx5e_tc_flow *flow,
 			     struct mlx5_flow_spec *spec,
@@ -2144,6 +2249,7 @@ static int __parse_cls_flower(struct mlx5e_priv *priv,
 	struct flow_rule *rule = flow_cls_offload_flow_rule(f);
 	struct flow_dissector *dissector = rule->match.dissector;
 	enum fs_flow_table_type fs_type;
+	bool match_inner_ecn = true;
 	u16 addr_type = 0;
 	u8 ip_proto = 0;
 	u8 *match_level;
@@ -2197,6 +2303,10 @@ static int __parse_cls_flower(struct mlx5e_priv *priv,
 			headers_c = get_match_inner_headers_criteria(spec);
 			headers_v = get_match_inner_headers_value(spec);
 		}
+
+		err = mlx5e_tc_verify_tunnel_ecn(priv, f, &match_inner_ecn);
+		if (err)
+			return err;
 	}
 
 	err = mlx5e_flower_parse_meta(filter_dev, f);
@@ -2420,10 +2530,12 @@ static int __parse_cls_flower(struct mlx5e_priv *priv,
 		struct flow_match_ip match;
 
 		flow_rule_match_ip(rule, &match);
-		MLX5_SET(fte_match_set_lyr_2_4, headers_c, ip_ecn,
-			 match.mask->tos & 0x3);
-		MLX5_SET(fte_match_set_lyr_2_4, headers_v, ip_ecn,
-			 match.key->tos & 0x3);
+		if (match_inner_ecn) {
+			MLX5_SET(fte_match_set_lyr_2_4, headers_c, ip_ecn,
+				 match.mask->tos & 0x3);
+			MLX5_SET(fte_match_set_lyr_2_4, headers_v, ip_ecn,
+				 match.key->tos & 0x3);
+		}
 
 		MLX5_SET(fte_match_set_lyr_2_4, headers_c, ip_dscp,
 			 match.mask->tos >> 2);
Signature · 1 line
2.34.1
author maintainer
#412 lore raw

PATCH 412/1039 net/mlx5: Fix access to sf_dev_table on allocation failure

In reply to Greg Kroah-Hartman

From: Shay Drory <[email protected]>

[ Upstream commit a1c7c49c2091926962f8c1c866d386febffec5d8 ]

Even when SF devices are supported, the SF device table allocation can still fail.
In such case mlx5_sf_dev_supported still reports true, but SF device table is invalid. This can result in NULL table access.

Hence, fix it by adding NULL table check.

Fixes 1958fc2f0712 ("net/mlx5: SF, Add auxiliary device driver")
---
 drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.c +1 -4
index f37db7cc32a65..7da012ff0d419 100644
@@ -30,10 +30,7 @@ bool mlx5_sf_dev_allocated(const struct mlx5_core_dev *dev)
 {
 	struct mlx5_sf_dev_table *table = dev->priv.sf_dev_table;
 
-	if (!mlx5_sf_dev_supported(dev))
-		return false;
-
-	return !xa_empty(&table->devices);
+	return table && !xa_empty(&table->devices);
 }
 
 static ssize_t sfnum_show(struct device *dev, struct device_attribute *attr, char *buf)
Signature · 1 line
2.34.1
author maintainer
#413 lore raw

PATCH 413/1039 net/mlx5e: Sync VXLAN udp ports during uplink representor profile change

In reply to Greg Kroah-Hartman

From: Maor Dickman <[email protected]>

[ Upstream commit 07f6dc4024ea1d2314b9c8b81fd4e492864fcca1 ]

Currently during NIC profile disablement all VXLAN udp ports offloaded to the HW are flushed and during its enablement the driver send notification to the stack to inform the core that the entire UDP tunnel port state has been lost, uplink representor doesn't have the same behavior which can cause VXLAN udp ports offload to be in bad state while moving between modes while VXLAN interface exist.

Fixed by aligning the uplink representor profile behavior to the NIC behavior.

Fixes 84db66124714 ("net/mlx5e: Move set vxlan nic info to profile init")
---
 drivers/net/ethernet/mellanox/mlx5/core/en_rep.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/net/ethernet/mellanox/mlx5/core/en_rep.c +3 -0
index 48895d79796a8..c0df4b1115b72 100644
@@ -50,6 +50,7 @@
 #include "fs_core.h"
 #include "lib/mlx5.h"
 #include "lib/devcom.h"
+#include "lib/vxlan.h"
 #define CREATE_TRACE_POINTS
 #include "diag/en_rep_tracepoint.h"
 #include "en_accel/ipsec.h"
@@ -1027,6 +1028,7 @@ static void mlx5e_uplink_rep_enable(struct mlx5e_priv *priv)
 	rtnl_lock();
 	if (netif_running(netdev))
 		mlx5e_open(netdev);
+	udp_tunnel_nic_reset_ntf(priv->netdev);
 	netif_device_attach(netdev);
 	rtnl_unlock();
 }
@@ -1048,6 +1050,7 @@ static void mlx5e_uplink_rep_disable(struct mlx5e_priv *priv)
 	mlx5_notifier_unregister(mdev, &priv->events_nb);
 	mlx5e_rep_tc_disable(priv);
 	mlx5_lag_remove_netdev(mdev, priv->netdev);
+	mlx5_vxlan_reset_to_default(mdev->vxlan);
 }
 
 static MLX5E_DEFINE_STATS_GRP(sw_rep, 0);
Signature · 1 line
2.34.1
author maintainer
#414 lore raw

PATCH 414/1039 net/mlx5: Set command entry semaphore up once got index free

In reply to Greg Kroah-Hartman

From: Moshe Shemesh <[email protected]>

[ Upstream commit 8e715cd613a1e872b9d918e912d90b399785761a ]

Avoid a race where command work handler may fail to allocate command entry index, by holding the command semaphore down till command entry index is being freed.

Fixes 410bd754cd73 ("net/mlx5: Add retry mechanism to the command entry index allocation")
---
 drivers/net/ethernet/mellanox/mlx5/core/cmd.c | 15 ++++++---------
 1 file changed, 6 insertions(+), 9 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/cmd.c +6 -9
index a46284ca51720..f588503157d04 100644
@@ -148,8 +148,12 @@ static void cmd_ent_put(struct mlx5_cmd_work_ent *ent)
 	if (!refcount_dec_and_test(&ent->refcnt))
 		return;
 
-	if (ent->idx >= 0)
-		cmd_free_index(ent->cmd, ent->idx);
+	if (ent->idx >= 0) {
+		struct mlx5_cmd *cmd = ent->cmd;
+
+		cmd_free_index(cmd, ent->idx);
+		up(ent->page_queue ? &cmd->pages_sem : &cmd->sem);
+	}
 
 	cmd_free_ent(ent);
 }
@@ -1602,8 +1606,6 @@ static void mlx5_cmd_comp_handler(struct mlx5_core_dev *dev, u64 vec, bool force
 	vector = vec & 0xffffffff;
 	for (i = 0; i < (1 << cmd->log_sz); i++) {
 		if (test_bit(i, &vector)) {
-			struct semaphore *sem;
-
 			ent = cmd->ent_arr[i];
 
 			/* if we already completed the command, ignore it */
@@ -1626,10 +1628,6 @@ static void mlx5_cmd_comp_handler(struct mlx5_core_dev *dev, u64 vec, bool force
 			    dev->state == MLX5_DEVICE_STATE_INTERNAL_ERROR)
 				cmd_ent_put(ent);
 
-			if (ent->page_queue)
-				sem = &cmd->pages_sem;
-			else
-				sem = &cmd->sem;
 			ent->ts2 = ktime_get_ns();
 			memcpy(ent->out->first.data, ent->lay->out, sizeof(ent->lay->out));
 			dump_command(dev, ent, 0);
@@ -1683,7 +1681,6 @@ static void mlx5_cmd_comp_handler(struct mlx5_core_dev *dev, u64 vec, bool force
 				 */
 				complete(&ent->done);
 			}
-			up(sem);
 		}
 	}
 }
Signature · 1 line
2.34.1
author maintainer
#415 lore raw

PATCH 415/1039 lib/mpi: Add the return value check of kcalloc()

In reply to Greg Kroah-Hartman

From: Zizhuang Deng <[email protected]>

[ Upstream commit dd827abe296fe4249b2f8c9b95f72f814ea8348c ]

Add the return value check of kcalloc() to avoid potential NULL ptr dereference.

Fixes a8ea8bdd9df9 ("lib/mpi: Extend the MPI library")
---
 lib/mpi/mpi-mod.c | 2 ++
 1 file changed, 2 insertions(+)
lib/mpi/mpi-mod.c +2 -0
index 47bc59edd4ff9..54fcc01564d9d 100644
@@ -40,6 +40,8 @@ mpi_barrett_t mpi_barrett_init(MPI m, int copy)
 
 	mpi_normalize(m);
 	ctx = kcalloc(1, sizeof(*ctx), GFP_KERNEL);
+	if (!ctx)
+		return NULL;
 
 	if (copy) {
 		ctx->m = mpi_copy(m);
Signature · 1 line
2.34.1
author maintainer
#416 lore raw

PATCH 416/1039 Bluetooth: L2CAP: uninitialized variables in l2cap_sock_setsockopt()

In reply to Greg Kroah-Hartman

From: Dan Carpenter <[email protected]>

[ Upstream commit 2b70d4f9b20635ac328836e50d183632e1930f94 ]

The "opt" variable is a u32, but on some paths only the top bytes were initialized and the others contained random stack data.

Fixes a7b75c5a8c41 ("net: pass a sockptr_t into ->setsockopt")
---
 net/bluetooth/l2cap_sock.c | 14 ++++++++------
 1 file changed, 8 insertions(+), 6 deletions(-)
net/bluetooth/l2cap_sock.c +8 -6
index 251017c69ab7f..d2c6785205992 100644
@@ -903,6 +903,8 @@ static int l2cap_sock_setsockopt(struct socket *sock, int level, int optname,
 	struct l2cap_conn *conn;
 	int len, err = 0;
 	u32 opt;
+	u16 mtu;
+	u8 mode;
 
 	BT_DBG("sk %p", sk);
 
@@ -1085,16 +1087,16 @@ static int l2cap_sock_setsockopt(struct socket *sock, int level, int optname,
 			break;
 		}
 
-		if (copy_from_sockptr(&opt, optval, sizeof(u16))) {
+		if (copy_from_sockptr(&mtu, optval, sizeof(u16))) {
 			err = -EFAULT;
 			break;
 		}
 
 		if (chan->mode == L2CAP_MODE_EXT_FLOWCTL &&
 		    sk->sk_state == BT_CONNECTED)
-			err = l2cap_chan_reconfigure(chan, opt);
+			err = l2cap_chan_reconfigure(chan, mtu);
 		else
-			chan->imtu = opt;
+			chan->imtu = mtu;
 
 		break;
 
@@ -1116,14 +1118,14 @@ static int l2cap_sock_setsockopt(struct socket *sock, int level, int optname,
 			break;
 		}
 
-		if (copy_from_sockptr(&opt, optval, sizeof(u8))) {
+		if (copy_from_sockptr(&mode, optval, sizeof(u8))) {
 			err = -EFAULT;
 			break;
 		}
 
-		BT_DBG("opt %u", opt);
+		BT_DBG("mode %u", mode);
 
-		err = l2cap_set_mode(chan, opt);
+		err = l2cap_set_mode(chan, mode);
 		if (err)
 			break;
 
Signature · 1 line
2.34.1
author maintainer
#417 lore raw

PATCH 417/1039 Bluetooth: hci_sock: fix endian bug in hci_sock_setsockopt()

In reply to Greg Kroah-Hartman

From: Dan Carpenter <[email protected]>

[ Upstream commit b9f9dbad0bd1c302d357fdd327c398f51f5fc2b1 ]

This copies a u16 into the high bits of an int, which works on a big endian system but not on a little endian system.

Fixes 09572fca7223 ("Bluetooth: hci_sock: Add support for BT_{SND,RCV}BUF")
---
 net/bluetooth/hci_sock.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
net/bluetooth/hci_sock.c +3 -2
index d0dad1fafe079..f2506e656f3e4 100644
@@ -1915,7 +1915,8 @@ static int hci_sock_setsockopt(struct socket *sock, int level, int optname,
 			       sockptr_t optval, unsigned int len)
 {
 	struct sock *sk = sock->sk;
-	int err = 0, opt = 0;
+	int err = 0;
+	u16 opt;
 
 	BT_DBG("sk %p, opt %d", sk, optname);
 
@@ -1941,7 +1942,7 @@ static int hci_sock_setsockopt(struct socket *sock, int level, int optname,
 			goto done;
 		}
 
-		if (copy_from_sockptr(&opt, optval, sizeof(u16))) {
+		if (copy_from_sockptr(&opt, optval, sizeof(opt))) {
 			err = -EFAULT;
 			break;
 		}
Signature · 1 line
2.34.1
author maintainer
#418 lore raw

PATCH 418/1039 mptcp: fix per socket endpoint accounting

In reply to Greg Kroah-Hartman

From: Paolo Abeni <[email protected]>

[ Upstream commit f7d6a237d7422809d458d754016de2844017cb4d ]

Since full-mesh endpoint support, the reception of a single ADD_ADDR option can cause multiple subflows creation. When such option is accepted we increment 'add_addr_accepted' by one. When we received a paired RM_ADDR option, we deleted all the relevant subflows, decrementing 'add_addr_accepted' by one for each of them.

We have a similar issue for 'local_addr_used'

Fix them moving the pm endpoint accounting outside the subflow traversal.

Fixes 1a0d6136c5f0 ("mptcp: local addresses fullmesh")
---
 net/mptcp/pm_netlink.c | 18 ++++++++++++------
 1 file changed, 12 insertions(+), 6 deletions(-)
net/mptcp/pm_netlink.c +12 -6
index f523051f5aef3..65764c8171b37 100644
@@ -710,6 +710,8 @@ static void mptcp_pm_nl_rm_addr_or_subflow(struct mptcp_sock *msk,
 		return;
 
 	for (i = 0; i < rm_list->nr; i++) {
+		bool removed = false;
+
 		list_for_each_entry_safe(subflow, tmp, &msk->conn_list, node) {
 			struct sock *ssk = mptcp_subflow_tcp_sock(subflow);
 			int how = RCV_SHUTDOWN | SEND_SHUTDOWN;
@@ -729,15 +731,19 @@ static void mptcp_pm_nl_rm_addr_or_subflow(struct mptcp_sock *msk,
 			mptcp_close_ssk(sk, ssk, subflow);
 			spin_lock_bh(&msk->pm.lock);
 
-			if (rm_type == MPTCP_MIB_RMADDR) {
-				msk->pm.add_addr_accepted--;
-				WRITE_ONCE(msk->pm.accept_addr, true);
-			} else if (rm_type == MPTCP_MIB_RMSUBFLOW) {
-				msk->pm.local_addr_used--;
-			}
+			removed = true;
 			msk->pm.subflows--;
 			__MPTCP_INC_STATS(sock_net(sk), rm_type);
 		}
+		if (!removed)
+			continue;
+
+		if (rm_type == MPTCP_MIB_RMADDR) {
+			msk->pm.add_addr_accepted--;
+			WRITE_ONCE(msk->pm.accept_addr, true);
+		} else if (rm_type == MPTCP_MIB_RMSUBFLOW) {
+			msk->pm.local_addr_used--;
+		}
 	}
 }
 
Signature · 1 line
2.34.1
author maintainer
#419 lore raw

PATCH 419/1039 mptcp: fix opt size when sending DSS + MP_FAIL

In reply to Greg Kroah-Hartman

From: Matthieu Baerts <[email protected]>

[ Upstream commit 04fac2cae9422a3401c172571afbcfdd58fa5c7e ]

When these two options had to be sent -- which is not common -- the DSS size was not being taken into account in the remaining size.

Additionally in this situation, the reported size was only the one of the MP_FAIL which can cause issue if at the end, we need to write more in the TCP options than previously said.

Here we use a dedicated variable for MP_FAIL size to keep the WARN_ON_ONCE() just after.

Fixes c25aeb4e0953 ("mptcp: MP_FAIL suboption sending")
---
 net/mptcp/options.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)
net/mptcp/options.c +6 -3
index fe98e4f475baa..96c6efdd48bcc 100644
@@ -821,10 +821,13 @@ bool mptcp_established_options(struct sock *sk, struct sk_buff *skb,
 	if (mptcp_established_options_mp(sk, skb, snd_data_fin, &opt_size, remaining, opts))
 		ret = true;
 	else if (mptcp_established_options_dss(sk, skb, snd_data_fin, &opt_size, remaining, opts)) {
+		unsigned int mp_fail_size;
+
 		ret = true;
-		if (mptcp_established_options_mp_fail(sk, &opt_size, remaining, opts)) {
-			*size += opt_size;
-			remaining -= opt_size;
+		if (mptcp_established_options_mp_fail(sk, &mp_fail_size,
+						      remaining - opt_size, opts)) {
+			*size += opt_size + mp_fail_size;
+			remaining -= opt_size - mp_fail_size;
 			return true;
 		}
 	}
Signature · 1 line
2.34.1
author maintainer
#420 lore raw

PATCH 420/1039 mptcp: fix a DSS option writing error

In reply to Greg Kroah-Hartman

From: Geliang Tang <[email protected]>

[ Upstream commit 110b6d1fe98fd7af9893992459b651594d789293 ]

'ptr += 1;' was omitted in the original code.

If the DSS is the last option -- which is what we have most of the time -- that's not an issue. But it is if we need to send something else after like a RM_ADDR or an MP_PRIO.

Fixes 1bff1e43a30e ("mptcp: optimize out option generation")
---
 net/mptcp/options.c | 1 +
 1 file changed, 1 insertion(+)
net/mptcp/options.c +1 -0
index 96c6efdd48bcc..6661b1d6520f1 100644
@@ -1319,6 +1319,7 @@ void mptcp_write_options(__be32 *ptr, const struct tcp_sock *tp,
 				put_unaligned_be32(mpext->data_len << 16 |
 						   TCPOPT_NOP << 8 | TCPOPT_NOP, ptr);
 			}
+			ptr += 1;
 		}
 	} else if (OPTIONS_MPTCP_MPC & opts->suboptions) {
 		u8 len, flag = MPTCP_CAP_HMAC_SHA256;
Signature · 1 line
2.34.1
author maintainer
#421 lore raw

PATCH 421/1039 mptcp: Check reclaim amount before reducing allocation

In reply to Greg Kroah-Hartman

From: Mat Martineau <[email protected]>

[ Upstream commit 269bda9e7da48eafb599d01c96199caa2f7547e5 ]

syzbot found a page counter underflow that was triggered by MPTCP's reclaim code:

page_counter underflow: -4294964789 nr_pages=4294967295

WARNING: CPU: 2 PID: 3785 at mm/page_counter.c:56 page_counter_cancel+0xcf/0xe0 mm/page_counter.c:56
Modules linked in:
CPU: 2 PID: 3785 Comm: kworker/2:6 Not tainted 5.16.0-rc1-syzkaller #0
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.14.0-2 04/01/2014
Workqueue: events mptcp_worker

RIP: 0010:page_counter_cancel+0xcf/0xe0 mm/page_counter.c:56
Code: c7 04 24 00 00 00 00 45 31 f6 eb 97 e8 2a 2b b5 ff 4c 89 ea 48 89 ee 48 c7 c7 00 9e b8 89 c6 05 a0 c1 ba 0b 01 e8 95 e4 4b 07 <0f> 0b eb a8 4c 89 e7 e8 25 5a fb ff eb c7 0f 1f 00 41 56 41 55 49
RSP: 0018:ffffc90002d4f918 EFLAGS: 00010082
RAX: 0000000000000000 RBX: ffff88806a494120 RCX: 0000000000000000
RDX: ffff8880688c41c0 RSI: ffffffff815e8f28 RDI: fffff520005a9f15
RBP: ffffffff000009cb R08: 0000000000000000 R09: 0000000000000000
R10: ffffffff815e2cfe R11: 0000000000000000 R12: ffff88806a494120
R13: 00000000ffffffff R14: 0000000000000000 R15: 0000000000000001
FS:  0000000000000000(0000) GS:ffff88802cc00000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000001b2de21000 CR3: 000000005ad59000 CR4: 0000000000150ee0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 <TASK>
 page_counter_uncharge+0x2e/0x60 mm/page_counter.c:160
 drain_stock+0xc1/0x180 mm/memcontrol.c:2219
 refill_stock+0x139/0x2f0 mm/memcontrol.c:2271
 __sk_mem_reduce_allocated+0x24d/0x550 net/core/sock.c:2945
 __mptcp_rmem_reclaim net/mptcp/protocol.c:167 [inline]
 __mptcp_mem_reclaim_partial+0x124/0x410 net/mptcp/protocol.c:975
 mptcp_mem_reclaim_partial net/mptcp/protocol.c:982 [inline]
 mptcp_alloc_tx_skb net/mptcp/protocol.c:1212 [inline]
 mptcp_sendmsg_frag+0x18c6/0x2190 net/mptcp/protocol.c:1279
 __mptcp_push_pending+0x232/0x720 net/mptcp/protocol.c:1545
 mptcp_release_cb+0xfe/0x200 net/mptcp/protocol.c:2975
 release_sock+0xb4/0x1b0 net/core/sock.c:3306
 mptcp_worker+0x51e/0xc10 net/mptcp/protocol.c:2443
 process_one_work+0x9b2/0x1690 kernel/workqueue.c:2298
 worker_thread+0x658/0x11f0 kernel/workqueue.c:2445
 kthread+0x405/0x4f0 kernel/kthread.c:327
 ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
 </TASK>

__mptcp_mem_reclaim_partial() could call __mptcp_rmem_reclaim() with a negative value, which passed that negative value to __sk_mem_reduce_allocated() and triggered the splat above.

Check for a reclaim amount that is positive and large enough for __mptcp_rmem_reclaim() to actually adjust rmem_fwd_alloc (much like the sk_mem_reclaim_partial() code the function is based on).

v2: Use '>' instead of '>=', since SK_MEM_QUANTUM - 1 would get right-shifted into nothing by __mptcp_rmem_reclaim.

Fixes 6511882cdd82 ("mptcp: allocate fwd memory separately on the rx and tx path")
Reported-and-tested-by [email protected]
Cc Andrew Morton <[email protected]>
Cc Michal Hocko <[email protected]>
---
 net/mptcp/protocol.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
net/mptcp/protocol.c +3 -1
index 54613f5b75217..0cd55e4c30fab 100644
@@ -972,7 +972,9 @@ static void __mptcp_mem_reclaim_partial(struct sock *sk)
 
 	lockdep_assert_held_once(&sk->sk_lock.slock);
 
-	__mptcp_rmem_reclaim(sk, reclaimable - 1);
+	if (reclaimable > SK_MEM_QUANTUM)
+		__mptcp_rmem_reclaim(sk, reclaimable - 1);
+
 	sk_mem_reclaim_partial(sk);
 }
 
Signature · 1 line
2.34.1
author maintainer
#422 lore raw

PATCH 422/1039 spi: spi-meson-spifc: Add missing pm_runtime_disable() in meson_spifc_probe

In reply to Greg Kroah-Hartman

From: Miaoqian Lin <[email protected]>

[ Upstream commit 69c1b87516e327a60b39f96b778fe683259408bf ]

If the probe fails, we should use pm_runtime_disable() to balance pm_runtime_enable().
Add missing pm_runtime_disable() for meson_spifc_probe.

Fixes c3e4bc5434d2 ("spi: meson: Add support for Amlogic Meson SPIFC")
---
 drivers/spi/spi-meson-spifc.c | 1 +
 1 file changed, 1 insertion(+)
drivers/spi/spi-meson-spifc.c +1 -0
index 8eca6f24cb799..c8ed7815c4ba6 100644
@@ -349,6 +349,7 @@ static int meson_spifc_probe(struct platform_device *pdev)
 	return 0;
 out_clk:
 	clk_disable_unprepare(spifc->clk);
+	pm_runtime_disable(spifc->dev);
 out_err:
 	spi_master_put(master);
 	return ret;
Signature · 1 line
2.34.1
author maintainer
#423 lore raw

PATCH 423/1039 octeontx2-af: Increment ptp refcount before use

In reply to Greg Kroah-Hartman

From: Subbaraya Sundeep <[email protected]>

[ Upstream commit 93440f4888cf049dbd22b41aaf94d2e2153b3eb8 ]

Before using the ptp pci device by AF driver increment the reference count of it.

Fixes a8b90c9d26d6 ("octeontx2-af: Add PTP device id for CN10K and 95O silcons")
---
 drivers/net/ethernet/marvell/octeontx2/af/ptp.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/net/ethernet/marvell/octeontx2/af/ptp.c +2 -0
index d6321de3cc171..e682b7bfde640 100644
@@ -60,6 +60,8 @@ struct ptp *ptp_get(void)
 	/* Check driver is bound to PTP block */
 	if (!ptp)
 		ptp = ERR_PTR(-EPROBE_DEFER);
+	else
+		pci_dev_get(ptp->pdev);
 
 	return ptp;
 }
Signature · 1 line
2.34.1
author maintainer
#424 lore raw

PATCH 424/1039 octeontx2-nicvf: Free VF PTP resources.

In reply to Greg Kroah-Hartman

From: Rakesh Babu Saladi <[email protected]>

[ Upstream commit eabd0f88b0d2d433c5dfe88218d4ce1c11ef04b8 ]

When a VF is removed respective PTP resources are not being freed currently. This patch fixes it.

Fixes 43510ef4ddad ("octeontx2-nicvf: Add PTP hardware clock support to NIX VF")
---
 drivers/net/ethernet/marvell/octeontx2/nic/otx2_vf.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/net/ethernet/marvell/octeontx2/nic/otx2_vf.c +4 -1
index 78944ad3492ff..d75f3a78fabf1 100644
@@ -684,7 +684,7 @@ static int otx2vf_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	err = register_netdev(netdev);
 	if (err) {
 		dev_err(dev, "Failed to register netdevice\n");
-		goto err_detach_rsrc;
+		goto err_ptp_destroy;
 	}
 
 	err = otx2_wq_init(vf);
@@ -709,6 +709,8 @@ static int otx2vf_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 
 err_unreg_netdev:
 	unregister_netdev(netdev);
+err_ptp_destroy:
+	otx2_ptp_destroy(vf);
 err_detach_rsrc:
 	if (test_bit(CN10K_LMTST, &vf->hw.cap_flag))
 		qmem_free(vf->dev, vf->dync_lmt);
@@ -742,6 +744,7 @@ static void otx2vf_remove(struct pci_dev *pdev)
 	unregister_netdev(netdev);
 	if (vf->otx2_wq)
 		destroy_workqueue(vf->otx2_wq);
+	otx2_ptp_destroy(vf);
 	otx2vf_disable_mbox_intr(vf);
 	otx2_detach_resources(&vf->mbox);
 	if (test_bit(CN10K_LMTST, &vf->hw.cap_flag))
Signature · 1 line
2.34.1
author maintainer
#425 lore raw

PATCH 425/1039 ax25: uninitialized variable in ax25_setsockopt()

In reply to Greg Kroah-Hartman

From: Dan Carpenter <[email protected]>

[ Upstream commit 9371937092d5fd502032c1bb4475b36b39b1f1b3 ]

The "opt" variable is unsigned long but we only copy 4 bytes from the user so the lower 4 bytes are uninitialized.

I have changed the integer overflow checks from ULONG to UINT as well. This is a slight API change but I don't expect it to break anything.

Fixes a7b75c5a8c41 ("net: pass a sockptr_t into ->setsockopt")
---
 net/ax25/af_ax25.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)
net/ax25/af_ax25.c +5 -5
index cfca99e295b80..02f43f3e2c564 100644
@@ -536,7 +536,7 @@ static int ax25_setsockopt(struct socket *sock, int level, int optname,
 	ax25_cb *ax25;
 	struct net_device *dev;
 	char devname[IFNAMSIZ];
-	unsigned long opt;
+	unsigned int opt;
 	int res = 0;
 
 	if (level != SOL_AX25)
@@ -568,7 +568,7 @@ static int ax25_setsockopt(struct socket *sock, int level, int optname,
 		break;
 
 	case AX25_T1:
-		if (opt < 1 || opt > ULONG_MAX / HZ) {
+		if (opt < 1 || opt > UINT_MAX / HZ) {
 			res = -EINVAL;
 			break;
 		}
@@ -577,7 +577,7 @@ static int ax25_setsockopt(struct socket *sock, int level, int optname,
 		break;
 
 	case AX25_T2:
-		if (opt < 1 || opt > ULONG_MAX / HZ) {
+		if (opt < 1 || opt > UINT_MAX / HZ) {
 			res = -EINVAL;
 			break;
 		}
@@ -593,7 +593,7 @@ static int ax25_setsockopt(struct socket *sock, int level, int optname,
 		break;
 
 	case AX25_T3:
-		if (opt < 1 || opt > ULONG_MAX / HZ) {
+		if (opt < 1 || opt > UINT_MAX / HZ) {
 			res = -EINVAL;
 			break;
 		}
@@ -601,7 +601,7 @@ static int ax25_setsockopt(struct socket *sock, int level, int optname,
 		break;
 
 	case AX25_IDLE:
-		if (opt > ULONG_MAX / (60 * HZ)) {
+		if (opt > UINT_MAX / (60 * HZ)) {
 			res = -EINVAL;
 			break;
 		}
Signature · 1 line
2.34.1
author maintainer
#426 lore raw

PATCH 426/1039 netrom: fix api breakage in nr_setsockopt()

In reply to Greg Kroah-Hartman

From: Dan Carpenter <[email protected]>

[ Upstream commit dc35616e6c2907b0c0c391a205802d8880f7fd85 ]

This needs to copy an unsigned int from user space instead of a long to avoid breaking user space with an API change.

I have updated all the integer overflow checks from ULONG to UINT as well. This is a slight API change but I do not expect it to affect anything in real life.

Fixes 3087a6f36ee0 ("netrom: fix copying in user data in nr_setsockopt")
---
 net/netrom/af_netrom.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)
net/netrom/af_netrom.c +6 -6
index f1ba7dd3d253d..fa9dc2ba39418 100644
@@ -298,7 +298,7 @@ static int nr_setsockopt(struct socket *sock, int level, int optname,
 {
 	struct sock *sk = sock->sk;
 	struct nr_sock *nr = nr_sk(sk);
-	unsigned long opt;
+	unsigned int opt;
 
 	if (level != SOL_NETROM)
 		return -ENOPROTOOPT;
@@ -306,18 +306,18 @@ static int nr_setsockopt(struct socket *sock, int level, int optname,
 	if (optlen < sizeof(unsigned int))
 		return -EINVAL;
 
-	if (copy_from_sockptr(&opt, optval, sizeof(unsigned long)))
+	if (copy_from_sockptr(&opt, optval, sizeof(opt)))
 		return -EFAULT;
 
 	switch (optname) {
 	case NETROM_T1:
-		if (opt < 1 || opt > ULONG_MAX / HZ)
+		if (opt < 1 || opt > UINT_MAX / HZ)
 			return -EINVAL;
 		nr->t1 = opt * HZ;
 		return 0;
 
 	case NETROM_T2:
-		if (opt < 1 || opt > ULONG_MAX / HZ)
+		if (opt < 1 || opt > UINT_MAX / HZ)
 			return -EINVAL;
 		nr->t2 = opt * HZ;
 		return 0;
@@ -329,13 +329,13 @@ static int nr_setsockopt(struct socket *sock, int level, int optname,
 		return 0;
 
 	case NETROM_T4:
-		if (opt < 1 || opt > ULONG_MAX / HZ)
+		if (opt < 1 || opt > UINT_MAX / HZ)
 			return -EINVAL;
 		nr->t4 = opt * HZ;
 		return 0;
 
 	case NETROM_IDLE:
-		if (opt > ULONG_MAX / (60 * HZ))
+		if (opt > UINT_MAX / (60 * HZ))
 			return -EINVAL;
 		nr->idle = opt * 60 * HZ;
 		return 0;
Signature · 1 line
2.34.1
author maintainer
#427 lore raw

PATCH 427/1039 regmap: Call regmap_debugfs_exit() prior to _init()

In reply to Greg Kroah-Hartman

From: Fabio Estevam <[email protected]>

[ Upstream commit 530792efa6cb86f5612ff093333fec735793b582 ]

Since commit cffa4b2122f5 ("regmap: debugfs: Fix a memory leak when calling regmap_attach_dev"), the following debugfs error is seen on i.MX boards:

debugfs: Directory 'dummy-iomuxc-gpr@20e0000' with parent 'regmap' already present!

In the attempt to fix the memory leak, the above commit added a NULL check for map->debugfs_name. For the first debufs entry, map->debugfs_name is NULL and then the new name is allocated via kasprintf().

For the second debugfs entry, map->debugfs_name() is no longer NULL, so it will keep using the old entry name and the duplicate name error is seen.

Quoting Mark Brown:

"That means that if the device gets freed we'll end up with the old debugfs file hanging around pointing at nothing.
...
To be more explicit this means we need a call to regmap_debugfs_exit() which will clean up all the existing debugfs stuff before we loose references to it."

Call regmap_debugfs_exit() prior to regmap_debugfs_init() to fix the problem.

Tested on i.MX6Q and i.MX6SX boards.

Fixes cffa4b2122f5 ("regmap: debugfs: Fix a memory leak when calling regmap_attach_dev")
---
 drivers/base/regmap/regmap.c | 1 +
 1 file changed, 1 insertion(+)
drivers/base/regmap/regmap.c +1 -0
index 21a0c2562ec06..f7811641ed5ae 100644
@@ -647,6 +647,7 @@ int regmap_attach_dev(struct device *dev, struct regmap *map,
 	if (ret)
 		return ret;
 
+	regmap_debugfs_exit(map);
 	regmap_debugfs_init(map);
 
 	/* Add a devres resource for dev_get_regmap() */
Signature · 1 line
2.34.1
author maintainer
#428 lore raw

PATCH 428/1039 net: mscc: ocelot: fix incorrect balancing with down LAG ports

In reply to Greg Kroah-Hartman

From: Vladimir Oltean <[email protected]>

[ Upstream commit a14e6b69f393d651913edcbe4ec0dec27b8b4b40 ]

Assuming the test setup described here:
https://patchwork.kernel.org/project/netdevbpf/cover/[email protected]/
(swp1 and swp2 are in bond0, and bond0 is in a bridge with swp0)

it can be seen that when swp1 goes down (on either board A or B), then traffic that should go through that port isn't forwarded anywhere.

A dump of the PGID table shows the following:

PGID_DST[0] = ports 0
PGID_DST[1] = ports 1
PGID_DST[2] = ports 2
PGID_DST[3] = ports 3
PGID_DST[4] = ports 4
PGID_DST[5] = ports 5
PGID_DST[6] = no ports
PGID_AGGR[0] = ports 0, 1, 2, 3, 4, 5
PGID_AGGR[1] = ports 0, 1, 2, 3, 4, 5
PGID_AGGR[2] = ports 0, 1, 2, 3, 4, 5
PGID_AGGR[3] = ports 0, 1, 2, 3, 4, 5
PGID_AGGR[4] = ports 0, 1, 2, 3, 4, 5
PGID_AGGR[5] = ports 0, 1, 2, 3, 4, 5
PGID_AGGR[6] = ports 0, 1, 2, 3, 4, 5
PGID_AGGR[7] = ports 0, 1, 2, 3, 4, 5
PGID_AGGR[8] = ports 0, 1, 2, 3, 4, 5
PGID_AGGR[9] = ports 0, 1, 2, 3, 4, 5
PGID_AGGR[10] = ports 0, 1, 2, 3, 4, 5
PGID_AGGR[11] = ports 0, 1, 2, 3, 4, 5
PGID_AGGR[12] = ports 0, 1, 2, 3, 4, 5
PGID_AGGR[13] = ports 0, 1, 2, 3, 4, 5
PGID_AGGR[14] = ports 0, 1, 2, 3, 4, 5
PGID_AGGR[15] = ports 0, 1, 2, 3, 4, 5
PGID_SRC[0] = ports 1, 2
PGID_SRC[1] = ports 0
PGID_SRC[2] = ports 0
PGID_SRC[3] = no ports
PGID_SRC[4] = no ports
PGID_SRC[5] = no ports
PGID_SRC[6] = ports 0, 1, 2, 3, 4, 5

Whereas a "good" PGID configuration for that setup should have looked like this:

PGID_DST[0] = ports 0
PGID_DST[1] = ports 1, 2
PGID_DST[2] = ports 1, 2
PGID_DST[3] = ports 3
PGID_DST[4] = ports 4
PGID_DST[5] = ports 5
PGID_DST[6] = no ports
PGID_AGGR[0] = ports 0, 2, 3, 4, 5
PGID_AGGR[1] = ports 0, 2, 3, 4, 5
PGID_AGGR[2] = ports 0, 2, 3, 4, 5
PGID_AGGR[3] = ports 0, 2, 3, 4, 5
PGID_AGGR[4] = ports 0, 2, 3, 4, 5
PGID_AGGR[5] = ports 0, 2, 3, 4, 5
PGID_AGGR[6] = ports 0, 2, 3, 4, 5
PGID_AGGR[7] = ports 0, 2, 3, 4, 5
PGID_AGGR[8] = ports 0, 2, 3, 4, 5
PGID_AGGR[9] = ports 0, 2, 3, 4, 5
PGID_AGGR[10] = ports 0, 2, 3, 4, 5
PGID_AGGR[11] = ports 0, 2, 3, 4, 5
PGID_AGGR[12] = ports 0, 2, 3, 4, 5
PGID_AGGR[13] = ports 0, 2, 3, 4, 5
PGID_AGGR[14] = ports 0, 2, 3, 4, 5
PGID_AGGR[15] = ports 0, 2, 3, 4, 5
PGID_SRC[0] = ports 1, 2
PGID_SRC[1] = ports 0
PGID_SRC[2] = ports 0
PGID_SRC[3] = no ports
PGID_SRC[4] = no ports
PGID_SRC[5] = no ports
PGID_SRC[6] = ports 0, 1, 2, 3, 4, 5

In other words, in the "bad" configuration, the attempt is to remove the inactive swp1 from the destination ports via PGID_DST. But when a MAC table entry is learned, it is learned towards PGID_DST 1, because that is the logical port id of the LAG itself (it is equal to the lowest numbered member port). So when swp1 becomes inactive, if we set PGID_DST[1] to contain just swp1 and not swp2, the packet will not have any chance to reach the destination via swp2.

The "correct" way to remove swp1 as a destination is via PGID_AGGR (remove swp1 from the aggregation port groups for all aggregation codes). This means that PGID_DST[1] and PGID_DST[2] must still contain both swp1 and swp2. This makes the MAC table still treat packets destined towards the single-port LAG as "multicast", and the inactive ports are removed via the aggregation code tables.

The change presented here is a design one: the ocelot_get_bond_mask() function used to take an "only_active_ports" argument. We don't need that. The only call site that specifies only_active_ports=true, ocelot_set_aggr_pgids(), must retrieve the entire bonding mask, because it must program that into PGID_DST. Additionally, it must also clear the inactive ports from the bond mask here, which it can't do if bond_mask just contains the active ports:

	ac = ocelot_read_rix(ocelot, ANA_PGID_PGID, i);
	ac &= ~bond_mask;  <---- here
	/* Don't do division by zero if there was no active
	 * port. Just make all aggregation codes zero.
	 */
	if (num_active_ports)
		ac |= BIT(aggr_idx[i % num_active_ports]);
	ocelot_write_rix(ocelot, ac, ANA_PGID_PGID, i);

So it becomes the responsibility of ocelot_set_aggr_pgids() to take ocelot_port->lag_tx_active into consideration when populating the aggr_idx array.

Fixes 23ca3b727ee6 ("net: mscc: ocelot: rebalance LAGs on link up/down events")
---
 drivers/net/ethernet/mscc/ocelot.c | 26 +++++++++++---------------
 1 file changed, 11 insertions(+), 15 deletions(-)
drivers/net/ethernet/mscc/ocelot.c +11 -15
index 1e4ad953cffbc..d1e883666951c 100644
@@ -1688,8 +1688,7 @@ int ocelot_get_ts_info(struct ocelot *ocelot, int port,
 }
 EXPORT_SYMBOL(ocelot_get_ts_info);
 
-static u32 ocelot_get_bond_mask(struct ocelot *ocelot, struct net_device *bond,
-				bool only_active_ports)
+static u32 ocelot_get_bond_mask(struct ocelot *ocelot, struct net_device *bond)
 {
 	u32 mask = 0;
 	int port;
@@ -1700,12 +1699,8 @@ static u32 ocelot_get_bond_mask(struct ocelot *ocelot, struct net_device *bond,
 		if (!ocelot_port)
 			continue;
 
-		if (ocelot_port->bond == bond) {
-			if (only_active_ports && !ocelot_port->lag_tx_active)
-				continue;
-
+		if (ocelot_port->bond == bond)
 			mask |= BIT(port);
-		}
 	}
 
 	return mask;
@@ -1792,10 +1787,8 @@ void ocelot_apply_bridge_fwd_mask(struct ocelot *ocelot)
 			mask = ocelot_get_bridge_fwd_mask(ocelot, port, bridge);
 			mask |= cpu_fwd_mask;
 			mask &= ~BIT(port);
-			if (bond) {
-				mask &= ~ocelot_get_bond_mask(ocelot, bond,
-							      false);
-			}
+			if (bond)
+				mask &= ~ocelot_get_bond_mask(ocelot, bond);
 		} else {
 			/* Standalone ports forward only to DSA tag_8021q CPU
 			 * ports (if those exist), or to the hardware CPU port
@@ -2112,13 +2105,17 @@ static void ocelot_set_aggr_pgids(struct ocelot *ocelot)
 		if (!bond || (visited & BIT(lag)))
 			continue;
 
-		bond_mask = ocelot_get_bond_mask(ocelot, bond, true);
+		bond_mask = ocelot_get_bond_mask(ocelot, bond);
 
 		for_each_set_bit(port, &bond_mask, ocelot->num_phys_ports) {
+			struct ocelot_port *ocelot_port = ocelot->ports[port];
+
 			// Destination mask
 			ocelot_write_rix(ocelot, bond_mask,
 					 ANA_PGID_PGID, port);
-			aggr_idx[num_active_ports++] = port;
+
+			if (ocelot_port->lag_tx_active)
+				aggr_idx[num_active_ports++] = port;
 		}
 
 		for_each_aggr_pgid(ocelot, i) {
@@ -2167,8 +2164,7 @@ static void ocelot_setup_logical_port_ids(struct ocelot *ocelot)
 
 		bond = ocelot_port->bond;
 		if (bond) {
-			int lag = __ffs(ocelot_get_bond_mask(ocelot, bond,
-							     false));
+			int lag = __ffs(ocelot_get_bond_mask(ocelot, bond));
 
 			ocelot_rmw_gix(ocelot,
 				       ANA_PORT_PORT_CFG_PORTID_VAL(lag),
Signature · 1 line
2.34.1
author maintainer
#429 lore raw

PATCH 429/1039 octeontx2-af: Fix interrupt name strings

In reply to Greg Kroah-Hartman

From: Sunil Goutham <[email protected]>

[ Upstream commit 6dc9a23e29061e50c36523270de60039ccf536fa ]

Fixed interrupt name string logic which currently results in wrong memory location being accessed while dumping /proc/interrupts.

---
 drivers/net/ethernet/marvell/octeontx2/af/rvu_cpt.c     | 5 ++---
 drivers/net/ethernet/marvell/octeontx2/af/rvu_devlink.c | 2 +-
 2 files changed, 3 insertions(+), 4 deletions(-)
drivers/net/ethernet/marvell/octeontx2/af/rvu_cpt.c +2 -3
index 45357deecabbf..a73a8017e0ee9 100644
@@ -172,14 +172,13 @@ static int cpt_10k_register_interrupts(struct rvu_block *block, int off)
 {
 	struct rvu *rvu = block->rvu;
 	int blkaddr = block->addr;
-	char irq_name[16];
 	int i, ret;
 
 	for (i = CPT_10K_AF_INT_VEC_FLT0; i < CPT_10K_AF_INT_VEC_RVU; i++) {
-		snprintf(irq_name, sizeof(irq_name), "CPTAF FLT%d", i);
+		sprintf(&rvu->irq_name[(off + i) * NAME_SIZE], "CPTAF FLT%d", i);
 		ret = rvu_cpt_do_register_interrupt(block, off + i,
 						    rvu_cpt_af_flt_intr_handler,
-						    irq_name);
+						    &rvu->irq_name[(off + i) * NAME_SIZE]);
 		if (ret)
 			goto err;
 		rvu_write64(rvu, blkaddr, CPT_AF_FLTX_INT_ENA_W1S(i), 0x1);
drivers/net/ethernet/marvell/octeontx2/af/rvu_devlink.c +1 -1
index 70bacd38a6d9d..d0ab8f233a029 100644
@@ -41,7 +41,7 @@ static bool rvu_common_request_irq(struct rvu *rvu, int offset,
 	struct rvu_devlink *rvu_dl = rvu->rvu_dl;
 	int rc;
 
-	sprintf(&rvu->irq_name[offset * NAME_SIZE], name);
+	sprintf(&rvu->irq_name[offset * NAME_SIZE], "%s", name);
 	rc = request_irq(pci_irq_vector(rvu->pdev, offset), fn, 0,
 			 &rvu->irq_name[offset * NAME_SIZE], rvu_dl);
 	if (rc)
Signature · 1 line
2.34.1
author maintainer
#430 lore raw

PATCH 430/1039 can: mcp251xfd: add missing newline to printed strings

In reply to Greg Kroah-Hartman

From: Marc Kleine-Budde <[email protected]>

[ Upstream commit 3bd9d8ce6f8c5c43ee2f1106021db0f98882cc75 ]

This patch adds the missing newline to printed strings.

Fixes 55e5b97f003e ("can: mcp25xxfd: add driver for Microchip MCP25xxFD SPI CAN")
---
 drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c +2 -2
index e16dc482f3270..63613c186f173 100644
@@ -2625,7 +2625,7 @@ static int mcp251xfd_register_chip_detect(struct mcp251xfd_priv *priv)
 	if (!mcp251xfd_is_251X(priv) &&
 	    priv->devtype_data.model != devtype_data->model) {
 		netdev_info(ndev,
-			    "Detected %s, but firmware specifies a %s. Fixing up.",
+			    "Detected %s, but firmware specifies a %s. Fixing up.\n",
 			    __mcp251xfd_get_model_str(devtype_data->model),
 			    mcp251xfd_get_model_str(priv));
 	}
@@ -2662,7 +2662,7 @@ static int mcp251xfd_register_check_rx_int(struct mcp251xfd_priv *priv)
 		return 0;
 
 	netdev_info(priv->ndev,
-		    "RX_INT active after softreset, disabling RX_INT support.");
+		    "RX_INT active after softreset, disabling RX_INT support.\n");
 	devm_gpiod_put(&priv->spi->dev, priv->rx_int);
 	priv->rx_int = NULL;
 
Signature · 1 line
2.34.1
author maintainer
#431 lore raw

PATCH 431/1039 tpm: add request_locality before write TPM_INT_ENABLE

In reply to Greg Kroah-Hartman

From: Chen Jun <[email protected]>

[ Upstream commit 0ef333f5ba7f24f5d8478425c163d3097f1c7afd ]

Locality is not appropriately requested before writing the int mask. Add the missing boilerplate.

Fixes e6aef069b6e9 ("tpm_tis: convert to using locality callbacks")
---
 drivers/char/tpm/tpm_tis_core.c | 8 ++++++++
 1 file changed, 8 insertions(+)
drivers/char/tpm/tpm_tis_core.c +8 -0
index b2659a4c40168..e2df1098a812f 100644
@@ -994,7 +994,15 @@ int tpm_tis_core_init(struct device *dev, struct tpm_tis_data *priv, int irq,
 	intmask |= TPM_INTF_CMD_READY_INT | TPM_INTF_LOCALITY_CHANGE_INT |
 		   TPM_INTF_DATA_AVAIL_INT | TPM_INTF_STS_VALID_INT;
 	intmask &= ~TPM_GLOBAL_INT_ENABLE;
+
+	rc = request_locality(chip, 0);
+	if (rc < 0) {
+		rc = -ENODEV;
+		goto out_err;
+	}
+
 	tpm_tis_write32(priv, TPM_INT_ENABLE(priv->locality), intmask);
+	release_locality(chip, 0);
 
 	rc = tpm_chip_start(chip);
 	if (rc)
Signature · 1 line
2.34.1
author maintainer
#432 lore raw

PATCH 432/1039 tpm_tis: Fix an error handling path in tpm_tis_core_init()

In reply to Greg Kroah-Hartman

From: Christophe Jaillet <[email protected]>

[ Upstream commit e96d52822f5ac0a25de78f95cd23421bcbc93584 ]

Commit 79ca6f74dae0 ("tpm: fix Atmel TPM crash caused by too frequent queries") has moved some code around without updating the error handling path.

This is now pointless to 'goto out_err' when neither 'clk_enable()' nor 'ioremap()' have been called yet.

Make a direct return instead to avoid undoing things that have not been done.

Fixes 79ca6f74dae0 ("tpm: fix Atmel TPM crash caused by too frequent queries")
---
 drivers/char/tpm/tpm_tis_core.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/char/tpm/tpm_tis_core.c +1 -1
index e2df1098a812f..36d1ad8f479d7 100644
@@ -952,7 +952,7 @@ int tpm_tis_core_init(struct device *dev, struct tpm_tis_data *priv, int irq,
 
 	rc = tpm_tis_read32(priv, TPM_DID_VID(0), &vendor);
 	if (rc < 0)
-		goto out_err;
+		return rc;
 
 	priv->manufacturer_id = vendor;
 
Signature · 1 line
2.34.1
author maintainer
#433 lore raw

PATCH 433/1039 can: softing: softing_startstop(): fix set but not used variable warning

In reply to Greg Kroah-Hartman

From: Marc Kleine-Budde <[email protected]>

[ Upstream commit 370d988cc529598ebaec6487d4f84c2115dc696b ]

In the function softing_startstop() the variable error_reporting is assigned but not used. The code that uses this variable is commented out. Its stated that the functionality is not finally verified.

To fix the warning:

| drivers/net/can/softing/softing_fw.c:424:9: error: variable 'error_reporting' set but not used [-Werror,-Wunused-but-set-variable]

remove the comment, activate the code, but add a "0 &&" to the if expression and rely on the optimizer rather than the preprocessor to remove the code.

Fixes 03fd3cf5a179 ("can: add driver for Softing card")
Cc Kurt Van Dijck <[email protected]>
---
 drivers/net/can/softing/softing_fw.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)
drivers/net/can/softing/softing_fw.c +6 -5
index 7e15368779931..32286f861a195 100644
@@ -565,18 +565,19 @@ int softing_startstop(struct net_device *dev, int up)
 		if (ret < 0)
 			goto failed;
 	}
-	/* enable_error_frame */
-	/*
+
+	/* enable_error_frame
+	 *
 	 * Error reporting is switched off at the moment since
 	 * the receiving of them is not yet 100% verified
 	 * This should be enabled sooner or later
-	 *
-	if (error_reporting) {
+	 */
+	if (0 && error_reporting) {
 		ret = softing_fct_cmd(card, 51, "enable_error_frame");
 		if (ret < 0)
 			goto failed;
 	}
-	*/
+
 	/* initialize interface */
 	iowrite16(1, &card->dpram[DPRAM_FCT_PARAM + 2]);
 	iowrite16(1, &card->dpram[DPRAM_FCT_PARAM + 4]);
Signature · 1 line
2.34.1
author maintainer
#434 lore raw

PATCH 434/1039 can: xilinx_can: xcan_probe(): check for error irq

In reply to Greg Kroah-Hartman

From: Jiasheng Jiang <[email protected]>

[ Upstream commit c6564c13dae25cd7f8e1de5127b4da4500ee5844 ]

For the possible failure of the platform_get_irq(), the returned irq could be error number and will finally cause the failure of the request_irq().

Consider that platform_get_irq() can now in certain cases return -EPROBE_DEFER, and the consequences of letting request_irq() effectively convert that into -EINVAL, even at probe time rather than later on. So it might be better to check just now.

---
 drivers/net/can/xilinx_can.c | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
drivers/net/can/xilinx_can.c +6 -1
index e2b15d29d15eb..af4a2adc85983 100644
@@ -1761,7 +1761,12 @@ static int xcan_probe(struct platform_device *pdev)
 	spin_lock_init(&priv->tx_lock);
 
 	/* Get IRQ for the device */
-	ndev->irq = platform_get_irq(pdev, 0);
+	ret = platform_get_irq(pdev, 0);
+	if (ret < 0)
+		goto err_free;
+
+	ndev->irq = ret;
+
 	ndev->flags |= IFF_ECHO;	/* We support local echo */
 
 	platform_set_drvdata(pdev, ndev);
Signature · 1 line
2.34.1
author maintainer
#435 lore raw

PATCH 435/1039 can: rcar_canfd: rcar_canfd_channel_probe(): make sure we free CAN network device

In reply to Greg Kroah-Hartman

From: Lad Prabhakar <[email protected]>

[ Upstream commit 72b1e360572f9fa7d08ee554f1da29abce23f288 ]

Make sure we free CAN network device in the error path. There are several jumps to fail label after allocating the CAN network device successfully. This patch places the free_candev() under fail label so that in failure path a jump to fail label frees the CAN network device.

---
 drivers/net/can/rcar/rcar_canfd.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
drivers/net/can/rcar/rcar_canfd.c +2 -3
index ff9d0f5ae0dd2..388521e70837f 100644
@@ -1640,8 +1640,7 @@ static int rcar_canfd_channel_probe(struct rcar_canfd_global *gpriv, u32 ch,
 	ndev = alloc_candev(sizeof(*priv), RCANFD_FIFO_DEPTH);
 	if (!ndev) {
 		dev_err(&pdev->dev, "alloc_candev() failed\n");
-		err = -ENOMEM;
-		goto fail;
+		return -ENOMEM;
 	}
 	priv = netdev_priv(ndev);
 
@@ -1735,8 +1734,8 @@ static int rcar_canfd_channel_probe(struct rcar_canfd_global *gpriv, u32 ch,
 
 fail_candev:
 	netif_napi_del(&priv->napi);
-	free_candev(ndev);
 fail:
+	free_candev(ndev);
 	return err;
 }
 
Signature · 1 line
2.34.1
author maintainer
#436 lore raw

PATCH 436/1039 pcmcia: fix setting of kthread task states

In reply to Greg Kroah-Hartman

From: Dominik Brodowski <[email protected]>

[ Upstream commit fbb3485f1f931102d8ba606f1c28123f5b48afa3 ]

We need to set TASK_INTERRUPTIBLE before calling kthread_should_stop(). Otherwise, kthread_stop() might see that the pccardd thread is still in TASK_RUNNING state and fail to wake it up.

Additionally, we only need to set the state back to TASK_RUNNING if kthread_should_stop() breaks the loop.

Cc Greg Kroah-Hartman <[email protected]>
Fixes d3046ba809ce ("pcmcia: fix a boot time warning in pcmcia cs code")
---
 drivers/pcmcia/cs.c | 8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)
drivers/pcmcia/cs.c +3 -5
index e211e2619680c..f70197154a362 100644
@@ -666,18 +666,16 @@ static int pccardd(void *__skt)
 		if (events || sysfs_events)
 			continue;
 
+		set_current_state(TASK_INTERRUPTIBLE);
 		if (kthread_should_stop())
 			break;
 
-		set_current_state(TASK_INTERRUPTIBLE);
-
 		schedule();
 
-		/* make sure we are running */
-		__set_current_state(TASK_RUNNING);
-
 		try_to_freeze();
 	}
+	/* make sure we are running before we exit */
+	__set_current_state(TASK_RUNNING);
 
 	/* shut down socket, if a device is still present */
 	if (skt->state & SOCKET_PRESENT) {
Signature · 1 line
2.34.1
author maintainer
#437 lore raw

PATCH 437/1039 netfilter: egress: avoid a lockdep splat

In reply to Greg Kroah-Hartman

From: Florian Westphal <[email protected]>

[ Upstream commit 6316136ec6e3dd1c302f7e7289a9ee46ecc610ae ]

include/linux/netfilter_netdev.h:97 suspicious rcu_dereference_check() usage! 2 locks held by sd-resolve/1100:
0: ..(rcu_read_lock_bh){1:3}, at: ip_finish_output2
1: ..(rcu_read_lock_bh){1:3}, at: __dev_queue_xmit __dev_queue_xmit+0 ..

The helper has two callers, one uses rcu_read_lock, the other rcu_read_lock_bh(). Annotate the dereference to reflect this.

Fixes 42df6e1d221dd ("netfilter: Introduce egress hook")
---
 include/linux/netfilter_netdev.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
include/linux/netfilter_netdev.h +1 -1
index b71b57a83bb4f..b4dd96e4dc8dc 100644
@@ -94,7 +94,7 @@ static inline struct sk_buff *nf_hook_egress(struct sk_buff *skb, int *rc,
 		return skb;
 #endif
 
-	e = rcu_dereference(dev->nf_hooks_egress);
+	e = rcu_dereference_check(dev->nf_hooks_egress, rcu_read_lock_bh_held());
 	if (!e)
 		return skb;
 
Signature · 1 line
2.34.1
author maintainer
#438 lore raw

PATCH 438/1039 net: openvswitch: Fix ct_state nat flags for conns arriving from tc

In reply to Greg Kroah-Hartman

From: Paul Blakey <[email protected]>

[ Upstream commit 6f022c2ddbcefaee79502ce5386dfe351d457070 ]

Netfilter conntrack maintains NAT flags per connection indicating whether NAT was configured for the connection. Openvswitch maintains NAT flags on the per packet flow key ct_state field, indicating whether NAT was actually executed on the packet.

When a packet misses from tc to ovs the conntrack NAT flags are set. However, NAT was not necessarily executed on the packet because the connection's state might still be in NEW state. As such, openvswitch wrongly assumes that NAT was executed and sets an incorrect flow key NAT flags.

Fix this, by flagging to openvswitch which NAT was actually done in act_ct via tc_skb_ext and tc_skb_cb to the openvswitch module, so the packet flow key NAT flags will be correctly set.

---
 include/linux/skbuff.h  |  4 +++-
 include/net/pkt_sched.h |  4 +++-
 net/openvswitch/flow.c  | 16 +++++++++++++---
 net/sched/act_ct.c      |  6 ++++++
 net/sched/cls_api.c     |  2 ++
 5 files changed, 27 insertions(+), 5 deletions(-)
include/linux/skbuff.h +3 -1
index 4507d77d6941f..60ab0c2fe5674 100644
@@ -287,7 +287,9 @@ struct tc_skb_ext {
 	__u32 chain;
 	__u16 mru;
 	__u16 zone;
-	bool post_ct;
+	u8 post_ct:1;
+	u8 post_ct_snat:1;
+	u8 post_ct_dnat:1;
 };
 #endif
 
include/net/pkt_sched.h +3 -1
index 9e71691c491b7..9e7b21c0b3a6d 100644
@@ -197,7 +197,9 @@ struct tc_skb_cb {
 	struct qdisc_skb_cb qdisc_cb;
 
 	u16 mru;
-	bool post_ct;
+	u8 post_ct:1;
+	u8 post_ct_snat:1;
+	u8 post_ct_dnat:1;
 	u16 zone; /* Only valid if post_ct = true */
 };
 
net/openvswitch/flow.c +13 -3
index 6d262d9aa10ea..02096f2ec6784 100644
@@ -859,7 +859,7 @@ int ovs_flow_key_extract(const struct ip_tunnel_info *tun_info,
 #if IS_ENABLED(CONFIG_NET_TC_SKB_EXT)
 	struct tc_skb_ext *tc_ext;
 #endif
-	bool post_ct = false;
+	bool post_ct = false, post_ct_snat = false, post_ct_dnat = false;
 	int res, err;
 	u16 zone = 0;
 
@@ -900,6 +900,8 @@ int ovs_flow_key_extract(const struct ip_tunnel_info *tun_info,
 		key->recirc_id = tc_ext ? tc_ext->chain : 0;
 		OVS_CB(skb)->mru = tc_ext ? tc_ext->mru : 0;
 		post_ct = tc_ext ? tc_ext->post_ct : false;
+		post_ct_snat = post_ct ? tc_ext->post_ct_snat : false;
+		post_ct_dnat = post_ct ? tc_ext->post_ct_dnat : false;
 		zone = post_ct ? tc_ext->zone : 0;
 	} else {
 		key->recirc_id = 0;
@@ -911,8 +913,16 @@ int ovs_flow_key_extract(const struct ip_tunnel_info *tun_info,
 	err = key_extract(skb, key);
 	if (!err) {
 		ovs_ct_fill_key(skb, key, post_ct);   /* Must be after key_extract(). */
-		if (post_ct && !skb_get_nfct(skb))
-			key->ct_zone = zone;
+		if (post_ct) {
+			if (!skb_get_nfct(skb)) {
+				key->ct_zone = zone;
+			} else {
+				if (!post_ct_dnat)
+					key->ct_state &= ~OVS_CS_F_DST_NAT;
+				if (!post_ct_snat)
+					key->ct_state &= ~OVS_CS_F_SRC_NAT;
+			}
+		}
 	}
 	return err;
 }
net/sched/act_ct.c +6 -0
index ab3591408419f..2a17eb77c9049 100644
@@ -839,6 +839,12 @@ static int ct_nat_execute(struct sk_buff *skb, struct nf_conn *ct,
 	}
 
 	err = nf_nat_packet(ct, ctinfo, hooknum, skb);
+	if (err == NF_ACCEPT) {
+		if (maniptype == NF_NAT_MANIP_SRC)
+			tc_skb_cb(skb)->post_ct_snat = 1;
+		if (maniptype == NF_NAT_MANIP_DST)
+			tc_skb_cb(skb)->post_ct_dnat = 1;
+	}
 out:
 	return err;
 }
net/sched/cls_api.c +2 -0
index 35c74bdde848e..cc9409aa755eb 100644
@@ -1625,6 +1625,8 @@ int tcf_classify(struct sk_buff *skb,
 		ext->chain = last_executed_chain;
 		ext->mru = cb->mru;
 		ext->post_ct = cb->post_ct;
+		ext->post_ct_snat = cb->post_ct_snat;
+		ext->post_ct_dnat = cb->post_ct_dnat;
 		ext->zone = cb->zone;
 	}
 
Signature · 1 line
2.34.1
author maintainer
#439 lore raw

PATCH 439/1039 iwlwifi: mvm: Use div_s64 instead of do_div in iwl_mvm_ftm_rtt_smoothing()

In reply to Greg Kroah-Hartman

From: Nathan Chancellor <[email protected]>

[ Upstream commit 4ccdcc8ffd955490feec05380223db6a48961eb5 ]

When building ARCH=arm allmodconfig:

drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c: In function ‘iwl_mvm_ftm_rtt_smoothing’:
./include/asm-generic/div64.h:222:35: error: comparison of distinct pointer types lacks a cast [-Werror]
222 | (void)(((typeof((n)) *)0) == ((uint64_t *)0)); \
| ^~
drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c:1070:9: note: in expansion of macro ‘do_div’

 1070 |         do_div(rtt_avg, 100);
      |         ^~~~~~

do_div() has to be used with an unsigned 64-bit integer dividend but rtt_avg is a signed 64-bit integer.

div_s64() expects a signed 64-bit integer dividend and signed 32-bit divisor, which fits this scenario, so use that function here to fix the warning.

Fixes 8b0f92549f2c ("iwlwifi: mvm: fix 32-bit build in FTM")
---
 drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/net/wireless/intel/iwlwifi/mvm/ftm-initiator.c +1 -2
index 9449d1af3c11a..628aee634b2ad 100644
@@ -1066,8 +1066,7 @@ static void iwl_mvm_ftm_rtt_smoothing(struct iwl_mvm *mvm,
 	overshoot = IWL_MVM_FTM_INITIATOR_SMOOTH_OVERSHOOT;
 	alpha = IWL_MVM_FTM_INITIATOR_SMOOTH_ALPHA;
 
-	rtt_avg = alpha * rtt + (100 - alpha) * resp->rtt_avg;
-	do_div(rtt_avg, 100);
+	rtt_avg = div_s64(alpha * rtt + (100 - alpha) * resp->rtt_avg, 100);
 
 	IWL_DEBUG_INFO(mvm,
 		       "%pM: prev rtt_avg=%lld, new rtt_avg=%lld, rtt=%lld\n",
Signature · 1 line
2.34.1
author maintainer
#440 lore raw

PATCH 440/1039 bnxt_en: use firmware provided max timeout for messages

In reply to Greg Kroah-Hartman

From: Edwin Peer <[email protected]>

[ Upstream commit bce9a0b7900836df223ab638090df0cb8430d9e8 ]

Some older devices cannot accommodate the 40 seconds timeout cap for long running commands (such as NVRAM commands) due to hardware limitations. Allow these devices to request more time for these long running commands, but print a warning, since the longer timeout may cause the hung task watchdog to trigger. In the case of a firmware update operation, this is preferable to failing outright.

v2: Use bp->hwrm_cmd_max_timeout directly without the constants.

Fixes 881d8353b05e ("bnxt_en: Add an upper bound for all firmware command timeouts.")
---
 drivers/net/ethernet/broadcom/bnxt/bnxt.c          | 6 ++++++
 drivers/net/ethernet/broadcom/bnxt/bnxt.h          | 3 ++-
 drivers/net/ethernet/broadcom/bnxt/bnxt_coredump.c | 4 ++--
 drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c  | 9 +++------
 drivers/net/ethernet/broadcom/bnxt/bnxt_hwrm.c     | 2 +-
 drivers/net/ethernet/broadcom/bnxt/bnxt_hwrm.h     | 3 +--
 6 files changed, 15 insertions(+), 12 deletions(-)
drivers/net/ethernet/broadcom/bnxt/bnxt.c +6 -0
index c04ea83188e22..7eaf74e5b2929 100644
@@ -8008,6 +8008,12 @@ static int bnxt_hwrm_ver_get(struct bnxt *bp)
 	bp->hwrm_cmd_timeout = le16_to_cpu(resp->def_req_timeout);
 	if (!bp->hwrm_cmd_timeout)
 		bp->hwrm_cmd_timeout = DFLT_HWRM_CMD_TIMEOUT;
+	bp->hwrm_cmd_max_timeout = le16_to_cpu(resp->max_req_timeout) * 1000;
+	if (!bp->hwrm_cmd_max_timeout)
+		bp->hwrm_cmd_max_timeout = HWRM_CMD_MAX_TIMEOUT;
+	else if (bp->hwrm_cmd_max_timeout > HWRM_CMD_MAX_TIMEOUT)
+		netdev_warn(bp->dev, "Device requests max timeout of %d seconds, may trigger hung task watchdog\n",
+			    bp->hwrm_cmd_max_timeout / 1000);
 
 	if (resp->hwrm_intf_maj_8b >= 1) {
 		bp->hwrm_max_req_len = le16_to_cpu(resp->max_req_win_len);
drivers/net/ethernet/broadcom/bnxt/bnxt.h +2 -1
index 4c9507d82fd0d..6bacd5fae6ba5 100644
@@ -1985,7 +1985,8 @@ struct bnxt {
 
 	u16			hwrm_max_req_len;
 	u16			hwrm_max_ext_req_len;
-	int			hwrm_cmd_timeout;
+	unsigned int		hwrm_cmd_timeout;
+	unsigned int		hwrm_cmd_max_timeout;
 	struct mutex		hwrm_cmd_lock;	/* serialize hwrm messages */
 	struct hwrm_ver_get_output	ver_resp;
 #define FW_VER_STR_LEN		32
drivers/net/ethernet/broadcom/bnxt/bnxt_coredump.c +2 -2
index d3cb2f21946da..c067898820360 100644
@@ -32,7 +32,7 @@ static int bnxt_hwrm_dbg_dma_data(struct bnxt *bp, void *msg,
 		return -ENOMEM;
 	}
 
-	hwrm_req_timeout(bp, msg, HWRM_COREDUMP_TIMEOUT);
+	hwrm_req_timeout(bp, msg, bp->hwrm_cmd_max_timeout);
 	cmn_resp = hwrm_req_hold(bp, msg);
 	resp = cmn_resp;
 
@@ -125,7 +125,7 @@ static int bnxt_hwrm_dbg_coredump_initiate(struct bnxt *bp, u16 component_id,
 	if (rc)
 		return rc;
 
-	hwrm_req_timeout(bp, req, HWRM_COREDUMP_TIMEOUT);
+	hwrm_req_timeout(bp, req, bp->hwrm_cmd_max_timeout);
 	req->component_id = cpu_to_le16(component_id);
 	req->segment_id = cpu_to_le16(segment_id);
 
drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c +3 -6
index 8188d55722e4b..7307df49c1313 100644
@@ -31,9 +31,6 @@
 #include "bnxt_nvm_defs.h"	/* NVRAM content constant and structure defs */
 #include "bnxt_fw_hdr.h"	/* Firmware hdr constant and structure defs */
 #include "bnxt_coredump.h"
-#define FLASH_NVRAM_TIMEOUT	((HWRM_CMD_TIMEOUT) * 100)
-#define FLASH_PACKAGE_TIMEOUT	((HWRM_CMD_TIMEOUT) * 200)
-#define INSTALL_PACKAGE_TIMEOUT	((HWRM_CMD_TIMEOUT) * 200)
 
 static u32 bnxt_get_msglevel(struct net_device *dev)
 {
@@ -2169,7 +2166,7 @@ static int bnxt_flash_nvram(struct net_device *dev, u16 dir_type,
 		req->host_src_addr = cpu_to_le64(dma_handle);
 	}
 
-	hwrm_req_timeout(bp, req, FLASH_NVRAM_TIMEOUT);
+	hwrm_req_timeout(bp, req, bp->hwrm_cmd_max_timeout);
 	req->dir_type = cpu_to_le16(dir_type);
 	req->dir_ordinal = cpu_to_le16(dir_ordinal);
 	req->dir_ext = cpu_to_le16(dir_ext);
@@ -2515,8 +2512,8 @@ int bnxt_flash_package_from_fw_obj(struct net_device *dev, const struct firmware
 		return rc;
 	}
 
-	hwrm_req_timeout(bp, modify, FLASH_PACKAGE_TIMEOUT);
-	hwrm_req_timeout(bp, install, INSTALL_PACKAGE_TIMEOUT);
+	hwrm_req_timeout(bp, modify, bp->hwrm_cmd_max_timeout);
+	hwrm_req_timeout(bp, install, bp->hwrm_cmd_max_timeout);
 
 	hwrm_req_hold(bp, modify);
 	modify->host_src_addr = cpu_to_le64(dma_handle);
drivers/net/ethernet/broadcom/bnxt/bnxt_hwrm.c +1 -1
index bb7327b82d0b2..8171f4912fa01 100644
@@ -496,7 +496,7 @@ static int __hwrm_send(struct bnxt *bp, struct bnxt_hwrm_ctx *ctx)
 	}
 
 	/* Limit timeout to an upper limit */
-	timeout = min_t(uint, ctx->timeout, HWRM_CMD_MAX_TIMEOUT);
+	timeout = min(ctx->timeout, bp->hwrm_cmd_max_timeout ?: HWRM_CMD_MAX_TIMEOUT);
 	/* convert timeout to usec */
 	timeout *= 1000;
 
drivers/net/ethernet/broadcom/bnxt/bnxt_hwrm.h +1 -2
index 4d17f0d5363bb..9a9fc4e8041b6 100644
@@ -58,11 +58,10 @@ void hwrm_update_token(struct bnxt *bp, u16 seq, enum bnxt_hwrm_wait_state s);
 
 #define BNXT_HWRM_MAX_REQ_LEN		(bp->hwrm_max_req_len)
 #define BNXT_HWRM_SHORT_REQ_LEN		sizeof(struct hwrm_short_input)
-#define HWRM_CMD_MAX_TIMEOUT		40000
+#define HWRM_CMD_MAX_TIMEOUT		40000U
 #define SHORT_HWRM_CMD_TIMEOUT		20
 #define HWRM_CMD_TIMEOUT		(bp->hwrm_cmd_timeout)
 #define HWRM_RESET_TIMEOUT		((HWRM_CMD_TIMEOUT) * 4)
-#define HWRM_COREDUMP_TIMEOUT		((HWRM_CMD_TIMEOUT) * 12)
 #define BNXT_HWRM_TARGET		0xffff
 #define BNXT_HWRM_NO_CMPL_RING		-1
 #define BNXT_HWRM_REQ_MAX_SIZE		128
Signature · 1 line
2.34.1
author maintainer
#441 lore raw

PATCH 441/1039 net: mcs7830: handle usb read errors properly

In reply to Greg Kroah-Hartman

From: Pavel Skripkin <[email protected]>

[ Upstream commit d668769eb9c52b150753f1653f7f5a0aeb8239d2 ]

Syzbot reported uninit value in mcs7830_bind(). The problem was in missing validation check for bytes read via usbnet_read_cmd().

usbnet_read_cmd() internally calls usb_control_msg(), that returns number of bytes read. Code should validate that requested number of bytes was actually read.

So, this patch adds missing size validation check inside mcs7830_get_reg() to prevent uninit value bugs

Reported-and-tested-by [email protected]
Fixes 2a36d7083438 ("USB: driver for mcs7830 (aka DeLOCK) USB ethernet adapter")
---
 drivers/net/usb/mcs7830.c | 12 ++++++++++--
 1 file changed, 10 insertions(+), 2 deletions(-)
drivers/net/usb/mcs7830.c +10 -2
index 326cc4e749d80..fdda0616704ea 100644
@@ -108,8 +108,16 @@ static const char driver_name[] = "MOSCHIP usb-ethernet driver";
 
 static int mcs7830_get_reg(struct usbnet *dev, u16 index, u16 size, void *data)
 {
-	return usbnet_read_cmd(dev, MCS7830_RD_BREQ, MCS7830_RD_BMREQ,
-				0x0000, index, data, size);
+	int ret;
+
+	ret = usbnet_read_cmd(dev, MCS7830_RD_BREQ, MCS7830_RD_BMREQ,
+			      0x0000, index, data, size);
+	if (ret < 0)
+		return ret;
+	else if (ret < size)
+		return -ENODATA;
+
+	return ret;
 }
 
 static int mcs7830_set_reg(struct usbnet *dev, u16 index, u16 size, const void *data)
Signature · 1 line
2.34.1
author maintainer
#442 lore raw

PATCH 442/1039 amt: fix wrong return type of amt_send_membership_update()

In reply to Greg Kroah-Hartman

From: Taehee Yoo <[email protected]>

[ Upstream commit dd3ca4c5184ea98e40acb8eb293d85b88ea04ee2 ]

amt_send_membership_update() would return -1 but it's return type is bool. So, it should be used TRUE instead of -1.

---
 drivers/net/amt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/amt.c +1 -1
index b732ee9a50ef9..d3a9dda6c7286 100644
@@ -1106,7 +1106,7 @@ static bool amt_send_membership_query(struct amt_dev *amt,
 	rt = ip_route_output_key(amt->net, &fl4);
 	if (IS_ERR(rt)) {
 		netdev_dbg(amt->dev, "no route to %pI4\n", &tunnel->ip4);
-		return -1;
+		return true;
 	}
 
 	amtmq		= skb_push(skb, sizeof(*amtmq));
Signature · 1 line
2.34.1
author maintainer
#443 lore raw

PATCH 443/1039 ext4: avoid trim error on fs with small groups

In reply to Greg Kroah-Hartman

From: Jan Kara <[email protected]>

[ Upstream commit 173b6e383d2a204c9921ffc1eca3b87aa2106c33 ]

A user reported FITRIM ioctl failing for him on ext4 on some devices without apparent reason. After some debugging we've found out that these devices (being LVM volumes) report rather large discard granularity of 42MB and the filesystem had 1k blocksize and thus group size of 8MB. Because ext4 FITRIM implementation puts discard granularity into minlen, ext4_trim_fs() declared the trim request as invalid. However just silently doing nothing seems to be a more appropriate reaction to such combination of parameters since user did not specify anything wrong.

CC Lukas Czerner <[email protected]>
Fixes 5c2ed62fd447 ("ext4: Adjust minlen with discard_granularity in the FITRIM ioctl")
---
 fs/ext4/ioctl.c   | 2 --
 fs/ext4/mballoc.c | 8 ++++++++
 2 files changed, 8 insertions(+), 2 deletions(-)
fs/ext4/ioctl.c +0 -2
index 606dee9e08a32..220a4c8178b5e 100644
@@ -1117,8 +1117,6 @@ resizefs_out:
 		    sizeof(range)))
 			return -EFAULT;
 
-		range.minlen = max((unsigned int)range.minlen,
-				   q->limits.discard_granularity);
 		ret = ext4_trim_fs(sb, &range);
 		if (ret < 0)
 			return ret;
fs/ext4/mballoc.c +8 -0
index 215b7068f548a..9af83b6242ddf 100644
@@ -6404,6 +6404,7 @@ ext4_trim_all_free(struct super_block *sb, ext4_group_t group,
  */
 int ext4_trim_fs(struct super_block *sb, struct fstrim_range *range)
 {
+	struct request_queue *q = bdev_get_queue(sb->s_bdev);
 	struct ext4_group_info *grp;
 	ext4_group_t group, first_group, last_group;
 	ext4_grpblk_t cnt = 0, first_cluster, last_cluster;
@@ -6422,6 +6423,13 @@ int ext4_trim_fs(struct super_block *sb, struct fstrim_range *range)
 	    start >= max_blks ||
 	    range->len < sb->s_blocksize)
 		return -EINVAL;
+	/* No point to try to trim less than discard granularity */
+	if (range->minlen < q->limits.discard_granularity) {
+		minlen = EXT4_NUM_B2C(EXT4_SB(sb),
+			q->limits.discard_granularity >> sb->s_blocksize_bits);
+		if (minlen > EXT4_CLUSTERS_PER_GROUP(sb))
+			goto out;
+	}
 	if (end >= max_blks)
 		end = max_blks - 1;
 	if (end <= first_data_blk)
Signature · 1 line
2.34.1
author maintainer
#444 lore raw

PATCH 444/1039 ASoC: Intel: sof_sdw: fix jack detection on HP Spectre x360 convertible

In reply to Greg Kroah-Hartman

From: Pierre-Louis Bossart <[email protected]>

[ Upstream commit 0527b19fa4f390a6054612e1fa1dd4f8efc96739 ]

Tests on device show the JD2 mode does not work at all, the 'Headphone Jack' and 'Headset Mic Jack' are shown as 'on' always.

JD1 seems to be the better option, with at least a change between the two cases.

Jack not plugged-in:
[root@fedora ~]# amixer -Dhw:0 cget numid=12
numid=12,iface=CARD,name='Headphone Jack'
; type=BOOLEAN,access=r-------,values=1
: values=off
[root@fedora ~]# amixer -Dhw:0 cget numid=13
numid=13,iface=CARD,name='Headset Mic Jack'
; type=BOOLEAN,access=r-------,values=1
: values=off

Jack plugged-in:
[root@fedora ~]# amixer -Dhw:0 cget numid=13
numid=13,iface=CARD,name='Headset Mic Jack'
; type=BOOLEAN,access=r-------,values=1
: values=on
[root@fedora ~]# amixer -Dhw:0 cget numid=13
numid=13,iface=CARD,name='Headset Mic Jack'
; type=BOOLEAN,access=r-------,values=1
: values=on

The 'Headset Mic Jack' is updated with a delay which seems normal with additional calibration needed.

Fixes d92e279dee56 ('ASoC: Intel: sof_sdw: add quirk for HP Spectre x360 convertible')
---
 sound/soc/intel/boards/sof_sdw.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
sound/soc/intel/boards/sof_sdw.c +1 -1
index 77219c3f8766c..54eefaff62a7e 100644
@@ -188,7 +188,7 @@ static const struct dmi_system_id sof_sdw_quirk_table[] = {
 		},
 		.driver_data = (void *)(SOF_SDW_TGL_HDMI |
 					SOF_SDW_PCH_DMIC |
-					RT711_JD2),
+					RT711_JD1),
 	},
 	{
 		/* NUC15 'Bishop County' LAPBC510 and LAPBC710 skews */
Signature · 1 line
2.34.1
author maintainer
#445 lore raw

PATCH 445/1039 ALSA: jack: Add missing rwsem around snd_ctl_remove() calls

In reply to Greg Kroah-Hartman

From: Takashi Iwai <[email protected]>

[ Upstream commit 06764dc931848c3a9bc01a63bbf76a605408bb54 ]

snd_ctl_remove() has to be called with card->controls_rwsem held (when called after the card instantiation). This patch add the missing rwsem calls around it.

Fixes 9058cbe1eed2 ("ALSA: jack: implement kctl creating for jack devices")
---
 sound/core/jack.c | 3 +++
 1 file changed, 3 insertions(+)
sound/core/jack.c +3 -0
index 537df1e98f8ac..d1e3055f2b6a5 100644
@@ -62,10 +62,13 @@ static int snd_jack_dev_free(struct snd_device *device)
 	struct snd_card *card = device->card;
 	struct snd_jack_kctl *jack_kctl, *tmp_jack_kctl;
 
+	down_write(&card->controls_rwsem);
 	list_for_each_entry_safe(jack_kctl, tmp_jack_kctl, &jack->kctl_list, list) {
 		list_del_init(&jack_kctl->list);
 		snd_ctl_remove(card, jack_kctl->kctl);
 	}
+	up_write(&card->controls_rwsem);
+
 	if (jack->private_free)
 		jack->private_free(jack);
 
Signature · 1 line
2.34.1
author maintainer
#446 lore raw

PATCH 446/1039 ALSA: PCM: Add missing rwsem around snd_ctl_remove() calls

In reply to Greg Kroah-Hartman

From: Takashi Iwai <[email protected]>

[ Upstream commit 5471e9762e1af4b7df057a96bfd46cc250979b88 ]

snd_ctl_remove() has to be called with card->controls_rwsem held (when called after the card instantiation). This patch add the missing rwsem calls around it.

Fixes a8ff48cb7083 ("ALSA: pcm: Free chmap at PCM free callback, too")
---
 sound/core/pcm.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
sound/core/pcm.c +5 -1
index 6fd3677685d70..ba4a987ed1c62 100644
@@ -810,7 +810,11 @@ EXPORT_SYMBOL(snd_pcm_new_internal);
 static void free_chmap(struct snd_pcm_str *pstr)
 {
 	if (pstr->chmap_kctl) {
-		snd_ctl_remove(pstr->pcm->card, pstr->chmap_kctl);
+		struct snd_card *card = pstr->pcm->card;
+
+		down_write(&card->controls_rwsem);
+		snd_ctl_remove(card, pstr->chmap_kctl);
+		up_write(&card->controls_rwsem);
 		pstr->chmap_kctl = NULL;
 	}
 }
Signature · 1 line
2.34.1
author maintainer
#447 lore raw

PATCH 447/1039 ALSA: hda: Add missing rwsem around snd_ctl_remove() calls

In reply to Greg Kroah-Hartman

From: Takashi Iwai <[email protected]>

[ Upstream commit 80bd64af75b4bb11c0329bc66c35da2ddfb66d88 ]

snd_ctl_remove() has to be called with card->controls_rwsem held (when called after the card instantiation). This patch add the missing rwsem calls around it.

Fixes d13bd412dce2 ("ALSA: hda - Manage kcontrol lists")
---
 sound/pci/hda/hda_codec.c | 3 +++
 1 file changed, 3 insertions(+)
sound/pci/hda/hda_codec.c +3 -0
index 0c4a337c9fc0d..eda70814369bd 100644
@@ -1727,8 +1727,11 @@ void snd_hda_ctls_clear(struct hda_codec *codec)
 {
 	int i;
 	struct hda_nid_item *items = codec->mixers.list;
+
+	down_write(&codec->card->controls_rwsem);
 	for (i = 0; i < codec->mixers.used; i++)
 		snd_ctl_remove(codec->card, items[i].kctl);
+	up_write(&codec->card->controls_rwsem);
 	snd_array_free(&codec->mixers);
 	snd_array_free(&codec->nids);
 }
Signature · 1 line
2.34.1
author maintainer
#448 lore raw

PATCH 448/1039 ALSA: hda: Fix potential deadlock at codec unbinding

In reply to Greg Kroah-Hartman

From: Takashi Iwai <[email protected]>

[ Upstream commit 7206998f578d5553989bc01ea2e544b622e79539 ]

When a codec is unbound dynamically via sysfs while its stream is in use, we may face a potential deadlock at the proc remove or a UAF. This happens since the hda_pcm is managed by a linked list, as it handles the hda_pcm object release via kref.

When a PCM is opened at the unbinding time, the release of hda_pcm gets delayed and it ends up with the close of the PCM stream releasing the associated hda_pcm object of its own. The hda_pcm destructor contains the PCM device release that includes the removal of procfs entries. And, this removal has the sync of the close of all in-use files -- which would never finish because it's called from the PCM file descriptor itself, i.e. it's trying to shoot its foot.

For addressing the deadlock above, this patch changes the way to manage and release the hda_pcm object. The kref of hda_pcm is dropped, and instead a simple refcount is introduced in hda_codec for keeping the track of the active PCM streams, and at each PCM open and close, this refcount is adjusted accordingly. At unbinding, the driver calls snd_device_disconnect() for each PCM stream, then synchronizes with the refcount finish, and finally releases the object resources.

Fixes bbbc7e8502c9 ("ALSA: hda - Allocate hda_pcm objects dynamically")
---
 include/sound/hda_codec.h |  8 +++++---
 sound/pci/hda/hda_bind.c  |  5 +++++
 sound/pci/hda/hda_codec.c | 42 ++++++++++++++++++++++++---------------
 sound/pci/hda/hda_local.h |  1 +
 4 files changed, 37 insertions(+), 19 deletions(-)
include/sound/hda_codec.h +5 -3
index 0e45963bb767f..82d9daa178517 100644
@@ -8,7 +8,7 @@
 #ifndef __SOUND_HDA_CODEC_H
 #define __SOUND_HDA_CODEC_H
 
-#include <linux/kref.h>
+#include <linux/refcount.h>
 #include <linux/mod_devicetable.h>
 #include <sound/info.h>
 #include <sound/control.h>
@@ -166,8 +166,8 @@ struct hda_pcm {
 	bool own_chmap;		/* codec driver provides own channel maps */
 	/* private: */
 	struct hda_codec *codec;
-	struct kref kref;
 	struct list_head list;
+	unsigned int disconnected:1;
 };
 
 /* codec information */
@@ -187,6 +187,8 @@ struct hda_codec {
 
 	/* PCM to create, set by patch_ops.build_pcms callback */
 	struct list_head pcm_list_head;
+	refcount_t pcm_ref;
+	wait_queue_head_t remove_sleep;
 
 	/* codec specific info */
 	void *spec;
@@ -420,7 +422,7 @@ void snd_hda_codec_cleanup_for_unbind(struct hda_codec *codec);
 
 static inline void snd_hda_codec_pcm_get(struct hda_pcm *pcm)
 {
-	kref_get(&pcm->kref);
+	refcount_inc(&pcm->codec->pcm_ref);
 }
 void snd_hda_codec_pcm_put(struct hda_pcm *pcm);
 
sound/pci/hda/hda_bind.c +5 -0
index 1c8bffc3eec6e..7153bd53e1893 100644
@@ -156,6 +156,11 @@ static int hda_codec_driver_remove(struct device *dev)
 		return codec->bus->core.ext_ops->hdev_detach(&codec->core);
 	}
 
+	refcount_dec(&codec->pcm_ref);
+	snd_hda_codec_disconnect_pcms(codec);
+	wait_event(codec->remove_sleep, !refcount_read(&codec->pcm_ref));
+	snd_power_sync_ref(codec->bus->card);
+
 	if (codec->patch_ops.free)
 		codec->patch_ops.free(codec);
 	snd_hda_codec_cleanup_for_unbind(codec);
sound/pci/hda/hda_codec.c +26 -16

92 lines not on this page · raw message

sound/pci/hda/hda_local.h +1 -0
index d22c96eb2f8fb..8621f576446b8 100644
@@ -137,6 +137,7 @@ int __snd_hda_add_vmaster(struct hda_codec *codec, char *name,
 int snd_hda_codec_reset(struct hda_codec *codec);
 void snd_hda_codec_register(struct hda_codec *codec);
 void snd_hda_codec_cleanup_for_unbind(struct hda_codec *codec);
+void snd_hda_codec_disconnect_pcms(struct hda_codec *codec);
 
 #define snd_hda_regmap_sync(codec)	snd_hdac_regmap_sync(&(codec)->core)
 
Signature · 1 line
2.34.1
author maintainer
#449 lore raw

PATCH 449/1039 RDMA/bnxt_re: Scan the whole bitmap when checking if "disabling RCFW with pending cmd-bit"

In reply to Greg Kroah-Hartman

From: Christophe JAILLET <[email protected]>

[ Upstream commit a917dfb66c0a1fa1caacf3d71edcafcab48e6ff0 ]

The 'cmdq->cmdq_bitmap' bitmap is 'rcfw->cmdq_depth' bits long. The size stored in 'cmdq->bmap_size' is the size of the bitmap in bytes.

Remove this erroneous 'bmap_size' and use 'rcfw->cmdq_depth' directly in 'bnxt_qplib_disable_rcfw_channel()'. Otherwise some error messages may be missing.

Other uses of 'cmdq_bitmap' already take into account 'rcfw->cmdq_depth' directly.

---
 drivers/infiniband/hw/bnxt_re/qplib_rcfw.c | 6 ++----
 drivers/infiniband/hw/bnxt_re/qplib_rcfw.h | 1 -
 2 files changed, 2 insertions(+), 5 deletions(-)
drivers/infiniband/hw/bnxt_re/qplib_rcfw.c +2 -4
index 3de854727460e..19a0778d38a2d 100644
@@ -618,8 +618,6 @@ int bnxt_qplib_alloc_rcfw_channel(struct bnxt_qplib_res *res,
 	if (!cmdq->cmdq_bitmap)
 		goto fail;
 
-	cmdq->bmap_size = bmap_size;
-
 	/* Allocate one extra to hold the QP1 entries */
 	rcfw->qp_tbl_size = qp_tbl_sz + 1;
 	rcfw->qp_tbl = kcalloc(rcfw->qp_tbl_size, sizeof(struct bnxt_qplib_qp_node),
@@ -667,8 +665,8 @@ void bnxt_qplib_disable_rcfw_channel(struct bnxt_qplib_rcfw *rcfw)
 	iounmap(cmdq->cmdq_mbox.reg.bar_reg);
 	iounmap(creq->creq_db.reg.bar_reg);
 
-	indx = find_first_bit(cmdq->cmdq_bitmap, cmdq->bmap_size);
-	if (indx != cmdq->bmap_size)
+	indx = find_first_bit(cmdq->cmdq_bitmap, rcfw->cmdq_depth);
+	if (indx != rcfw->cmdq_depth)
 		dev_err(&rcfw->pdev->dev,
 			"disabling RCFW with pending cmd-bit %lx\n", indx);
 
drivers/infiniband/hw/bnxt_re/qplib_rcfw.h +0 -1
index 82faa4e4cda84..0a3d8e7da3d42 100644
@@ -152,7 +152,6 @@ struct bnxt_qplib_cmdq_ctx {
 	wait_queue_head_t		waitq;
 	unsigned long			flags;
 	unsigned long			*cmdq_bitmap;
-	u32				bmap_size;
 	u32				seq_num;
 };
 
Signature · 1 line
2.34.1
author maintainer
#450 lore raw

PATCH 450/1039 RDMA/hns: Validate the pkey index

In reply to Greg Kroah-Hartman

From: Kamal Heib <[email protected]>

[ Upstream commit 2a67fcfa0db6b4075515bd23497750849b88850f ]

Before query pkey, make sure that the queried index is valid.

Fixes 9a4435375cd1 ("IB/hns: Add driver files for hns RoCE driver")
---
 drivers/infiniband/hw/hns/hns_roce_main.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/infiniband/hw/hns/hns_roce_main.c +3 -0
index 4194b626f3c65..8233bec053ee3 100644
@@ -270,6 +270,9 @@ static enum rdma_link_layer hns_roce_get_link_layer(struct ib_device *device,
 static int hns_roce_query_pkey(struct ib_device *ib_dev, u32 port, u16 index,
 			       u16 *pkey)
 {
+	if (index > 0)
+		return -EINVAL;
+
 	*pkey = PKEY_ID;
 
 	return 0;
Signature · 1 line
2.34.1
author maintainer
#451 lore raw

PATCH 451/1039 scsi: pm80xx: Update WARN_ON check in pm8001_mpi_build_cmd()

In reply to Greg Kroah-Hartman

From: Igor Pylypiv <[email protected]>

[ Upstream commit 606c54ae975ad3af540b505b46b55a687501711f ]

Starting from commit 05c6c029a44d ("scsi: pm80xx: Increase number of supported queues") driver initializes only max_q_num queues. Do not use an invalid queue if the WARN_ON condition is true.

---
 drivers/scsi/pm8001/pm8001_hwi.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/scsi/pm8001/pm8001_hwi.c +3 -1
index 124cb69740c67..4390c8b9170cd 100644
@@ -1325,7 +1325,9 @@ int pm8001_mpi_build_cmd(struct pm8001_hba_info *pm8001_ha,
 	int q_index = circularQ - pm8001_ha->inbnd_q_tbl;
 	int rv;
 
-	WARN_ON(q_index >= PM8001_MAX_INB_NUM);
+	if (WARN_ON(q_index >= pm8001_ha->max_q_num))
+		return -EINVAL;
+
 	spin_lock_irqsave(&circularQ->iq_lock, flags);
 	rv = pm8001_mpi_msg_free_get(circularQ, pm8001_ha->iomb_size,
 			&pMessage);
Signature · 1 line
2.34.1
author maintainer
#452 lore raw

PATCH 452/1039 clk: renesas: rzg2l: Check return value of pm_genpd_init()

In reply to Greg Kroah-Hartman

From: Lad Prabhakar <[email protected]>

[ Upstream commit 27527a3d3b162e4512798c058c0e8a216c721187 ]

Make sure we check the return value of pm_genpd_init() which might fail. Also add a devres action to remove the power-domain in-case the probe callback fails further down in the code flow.

Fixes ef3c613ccd68a ("clk: renesas: Add CPG core wrapper for RZ/G2L SoC")
---
 drivers/clk/renesas/rzg2l-cpg.c | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)
drivers/clk/renesas/rzg2l-cpg.c +13 -1
index 4021f6cabda4b..4f04ff57d468a 100644
@@ -850,10 +850,16 @@ static void rzg2l_cpg_detach_dev(struct generic_pm_domain *unused, struct device
 		pm_clk_destroy(dev);
 }
 
+static void rzg2l_cpg_genpd_remove(void *data)
+{
+	pm_genpd_remove(data);
+}
+
 static int __init rzg2l_cpg_add_clk_domain(struct device *dev)
 {
 	struct device_node *np = dev->of_node;
 	struct generic_pm_domain *genpd;
+	int ret;
 
 	genpd = devm_kzalloc(dev, sizeof(*genpd), GFP_KERNEL);
 	if (!genpd)
@@ -864,7 +870,13 @@ static int __init rzg2l_cpg_add_clk_domain(struct device *dev)
 		       GENPD_FLAG_ACTIVE_WAKEUP;
 	genpd->attach_dev = rzg2l_cpg_attach_dev;
 	genpd->detach_dev = rzg2l_cpg_detach_dev;
-	pm_genpd_init(genpd, &pm_domain_always_on_gov, false);
+	ret = pm_genpd_init(genpd, &pm_domain_always_on_gov, false);
+	if (ret)
+		return ret;
+
+	ret = devm_add_action_or_reset(dev, rzg2l_cpg_genpd_remove, genpd);
+	if (ret)
+		return ret;
 
 	of_genpd_add_provider_simple(np, genpd);
 	return 0;
Signature · 1 line
2.34.1
author maintainer
#453 lore raw

PATCH 453/1039 clk: renesas: rzg2l: propagate return value of_genpd_add_provider_simple()

In reply to Greg Kroah-Hartman

From: Lad Prabhakar <[email protected]>

[ Upstream commit 33748744f15a110a233b6ae0380f476006e770f0 ]

of_genpd_add_provider_simple() might fail, this patch makes sure we check the return value of of_genpd_add_provider_simple() by propagating the return value to the caller of rzg2l_cpg_add_clk_domain().

Fixes ef3c613ccd68a ("clk: renesas: Add CPG core wrapper for RZ/G2L SoC")
---
 drivers/clk/renesas/rzg2l-cpg.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/clk/renesas/rzg2l-cpg.c +1 -2
index 4f04ff57d468a..aafd1879ff562 100644
@@ -878,8 +878,7 @@ static int __init rzg2l_cpg_add_clk_domain(struct device *dev)
 	if (ret)
 		return ret;
 
-	of_genpd_add_provider_simple(np, genpd);
-	return 0;
+	return of_genpd_add_provider_simple(np, genpd);
 }
 
 static int __init rzg2l_cpg_probe(struct platform_device *pdev)
Signature · 1 line
2.34.1
author maintainer
#454 lore raw

PATCH 454/1039 clk: imx8mn: Fix imx8mn_clko1_sels

In reply to Greg Kroah-Hartman

From: Adam Ford <[email protected]>

[ Upstream commit 570727e9acfac1c2330a01dd5e1272e9c3acec08 ]

When attempting to use sys_pll1_80m as the parent for clko1, the system hangs. This is due to the fact that the source select for sys_pll1_80m was incorrectly pointing to m7_alt_pll_clk, which doesn't yet exist.

According to Rev 3 of the TRM, The imx8mn_clko1_sels also incorrectly references an osc_27m which does not exist, nor does an entry for source select bits 010b. Fix both by inserting a dummy clock into the missing space in the table and renaming the incorrectly name clock with dummy.

Fixes 96d6392b54db ("clk: imx: Add support for i.MX8MN clock driver")
---
 drivers/clk/imx/clk-imx8mn.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/clk/imx/clk-imx8mn.c +3 -3
index c55577604e16a..021355a247081 100644
@@ -277,9 +277,9 @@ static const char * const imx8mn_pdm_sels[] = {"osc_24m", "sys_pll2_100m", "audi
 
 static const char * const imx8mn_dram_core_sels[] = {"dram_pll_out", "dram_alt_root", };
 
-static const char * const imx8mn_clko1_sels[] = {"osc_24m", "sys_pll1_800m", "osc_27m",
-						 "sys_pll1_200m", "audio_pll2_out", "vpu_pll",
-						 "sys_pll1_80m", };
+static const char * const imx8mn_clko1_sels[] = {"osc_24m", "sys_pll1_800m", "dummy",
+						 "sys_pll1_200m", "audio_pll2_out", "sys_pll2_500m",
+						 "dummy", "sys_pll1_80m", };
 static const char * const imx8mn_clko2_sels[] = {"osc_24m", "sys_pll2_200m", "sys_pll1_400m",
 						 "sys_pll2_166m", "sys_pll3_out", "audio_pll1_out",
 						 "video_pll1_out", "osc_32k", };
Signature · 1 line
2.34.1
author maintainer
#455 lore raw

PATCH 455/1039 ASoC: cs42l42: Report initial jack state

In reply to Greg Kroah-Hartman

From: Richard Fitzgerald <[email protected]>

[ Upstream commit fdd535283779ec9f9c35fda352585c629121214f ]

When a jack handler is registered in cs42l42_set_jack() the initial state should be reported if an attached headphone/headset has already been detected.

The jack detect sequence takes around 1 second: typically long enough for the machine driver to probe and register the jack handler in time to receive the first report from the interrupt handler. So it is possible on some systems that the correct initial state was seen simply because of lucky timing. Modular builds were more likely to miss the reporting of the initial state.

Fixes 4ca239f33737 ("ASoC: cs42l42: Always enable TS_PLUG and TS_UNPLUG interrupts")
---
 sound/soc/codecs/cs42l42.c | 22 ++++++++++++++++++++++
 sound/soc/codecs/cs42l42.h |  2 ++
 2 files changed, 24 insertions(+)
sound/soc/codecs/cs42l42.c +22 -0
index 27a1c4c73074f..a63fba4e6c9c2 100644
@@ -521,8 +521,25 @@ static int cs42l42_set_jack(struct snd_soc_component *component, struct snd_soc_
 {
 	struct cs42l42_private *cs42l42 = snd_soc_component_get_drvdata(component);
 
+	/* Prevent race with interrupt handler */
+	mutex_lock(&cs42l42->jack_detect_mutex);
 	cs42l42->jack = jk;
 
+	if (jk) {
+		switch (cs42l42->hs_type) {
+		case CS42L42_PLUG_CTIA:
+		case CS42L42_PLUG_OMTP:
+			snd_soc_jack_report(jk, SND_JACK_HEADSET, SND_JACK_HEADSET);
+			break;
+		case CS42L42_PLUG_HEADPHONE:
+			snd_soc_jack_report(jk, SND_JACK_HEADPHONE, SND_JACK_HEADPHONE);
+			break;
+		default:
+			break;
+		}
+	}
+	mutex_unlock(&cs42l42->jack_detect_mutex);
+
 	return 0;
 }
 
@@ -1611,6 +1628,8 @@ static irqreturn_t cs42l42_irq_thread(int irq, void *data)
 		CS42L42_M_DETECT_FT_MASK |
 		CS42L42_M_HSBIAS_HIZ_MASK);
 
+	mutex_lock(&cs42l42->jack_detect_mutex);
+
 	/* Check auto-detect status */
 	if ((~masks[5]) & irq_params_table[5].mask) {
 		if (stickies[5] & CS42L42_HSDET_AUTO_DONE_MASK) {
@@ -1689,6 +1708,8 @@ static irqreturn_t cs42l42_irq_thread(int irq, void *data)
 		}
 	}
 
+	mutex_unlock(&cs42l42->jack_detect_mutex);
+
 	return IRQ_HANDLED;
 }
 
@@ -2033,6 +2054,7 @@ static int cs42l42_i2c_probe(struct i2c_client *i2c_client,
 
 	cs42l42->dev = &i2c_client->dev;
 	i2c_set_clientdata(i2c_client, cs42l42);
+	mutex_init(&cs42l42->jack_detect_mutex);
 
 	cs42l42->regmap = devm_regmap_init_i2c(i2c_client, &cs42l42_regmap);
 	if (IS_ERR(cs42l42->regmap)) {
sound/soc/codecs/cs42l42.h +2 -0
index f45bcc9a3a62f..02128ebf8989a 100644
@@ -12,6 +12,7 @@
 #ifndef __CS42L42_H__
 #define __CS42L42_H__
 
+#include <linux/mutex.h>
 #include <sound/jack.h>
 
 #define CS42L42_PAGE_REGISTER	0x00	/* Page Select Register */
@@ -838,6 +839,7 @@ struct  cs42l42_private {
 	struct gpio_desc *reset_gpio;
 	struct completion pdn_done;
 	struct snd_soc_jack *jack;
+	struct mutex jack_detect_mutex;
 	int pll_config;
 	int bclk;
 	u32 sclk;
Signature · 1 line
2.34.1
author maintainer
#456 lore raw

PATCH 456/1039 powerpc/prom_init: Fix improper check of prom_getprop()

In reply to Greg Kroah-Hartman

From: Peiwei Hu <[email protected]>

[ Upstream commit 869fb7e5aecbc163003f93f36dcc26d0554319f6 ]

prom_getprop() can return PROM_ERROR. Binary operator can not identify it.

Fixes 94d2dde738a5 ("[POWERPC] Efika: prune fixups and make them more carefull")
---
 arch/powerpc/kernel/prom_init.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/powerpc/kernel/prom_init.c +1 -1
index 18b04b08b9833..f845065c860e3 100644
@@ -2991,7 +2991,7 @@ static void __init fixup_device_tree_efika_add_phy(void)
 
 	/* Check if the phy-handle property exists - bail if it does */
 	rv = prom_getprop(node, "phy-handle", prop, sizeof(prop));
-	if (!rv)
+	if (rv <= 0)
 		return;
 
 	/*
Signature · 1 line
2.34.1
author maintainer
#457 lore raw

PATCH 457/1039 ASoC: uniphier: drop selecting non-existing SND_SOC_UNIPHIER_AIO_DMA

In reply to Greg Kroah-Hartman

From: Lukas Bulwahn <[email protected]>

[ Upstream commit 49f893253ab43566e34332a969324531fea463f6 ]

Commit f37fe2f9987b ("ASoC: uniphier: add support for UniPhier AIO common driver") adds configs SND_SOC_UNIPHIER_{LD11,PXS2}, which select the non-existing config SND_SOC_UNIPHIER_AIO_DMA.

Hence, ./scripts/checkkconfigsymbols.py warns:

SND_SOC_UNIPHIER_AIO_DMA
Referencing files: sound/soc/uniphier/Kconfig

Probably, there is actually no further config intended to be selected here. So, just drop selecting the non-existing config.

Fixes f37fe2f9987b ("ASoC: uniphier: add support for UniPhier AIO common driver")
---
 sound/soc/uniphier/Kconfig | 2 --
 1 file changed, 2 deletions(-)
sound/soc/uniphier/Kconfig +0 -2
index aa3592ee1358b..ddfa6424c656b 100644
@@ -23,7 +23,6 @@ config SND_SOC_UNIPHIER_LD11
 	tristate "UniPhier LD11/LD20 Device Driver"
 	depends on SND_SOC_UNIPHIER
 	select SND_SOC_UNIPHIER_AIO
-	select SND_SOC_UNIPHIER_AIO_DMA
 	help
 	  This adds ASoC driver for Socionext UniPhier LD11/LD20
 	  input and output that can be used with other codecs.
@@ -34,7 +33,6 @@ config SND_SOC_UNIPHIER_PXS2
 	tristate "UniPhier PXs2 Device Driver"
 	depends on SND_SOC_UNIPHIER
 	select SND_SOC_UNIPHIER_AIO
-	select SND_SOC_UNIPHIER_AIO_DMA
 	help
 	  This adds ASoC driver for Socionext UniPhier PXs2
 	  input and output that can be used with other codecs.
Signature · 1 line
2.34.1
author maintainer
#458 lore raw

PATCH 458/1039 ASoC: codecs: wcd938x: add SND_SOC_WCD938_SDW to codec list instead

In reply to Greg Kroah-Hartman

From: Lukas Bulwahn <[email protected]>

[ Upstream commit 2039cc1da4bee1fd0df644e26b28ed769cd32a81 ]

Commit 045442228868 ("ASoC: codecs: wcd938x: add audio routing and Kconfig") adds SND_SOC_WCD937X, which does not exist, and SND_SOC_WCD938X, which seems not really to be the intended config to be selected, but only a supporting config symbol to the actual config SND_SOC_WCD938X_SDW for the codec.

Add SND_SOC_WCD938_SDW to the list instead of SND_SOC_WCD93{7,8}X.

The issue was identified with ./scripts/checkkconfigsymbols.py.

Fixes 045442228868 ("ASoC: codecs: wcd938x: add audio routing and Kconfig")
---
 sound/soc/codecs/Kconfig | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
sound/soc/codecs/Kconfig +1 -2
index 326f2d611ad4e..3a610ba183ffb 100644
@@ -241,8 +241,7 @@ config SND_SOC_ALL_CODECS
 	imply SND_SOC_UDA1380
 	imply SND_SOC_WCD9335
 	imply SND_SOC_WCD934X
-	imply SND_SOC_WCD937X
-	imply SND_SOC_WCD938X
+	imply SND_SOC_WCD938X_SDW
 	imply SND_SOC_LPASS_RX_MACRO
 	imply SND_SOC_LPASS_TX_MACRO
 	imply SND_SOC_WL1273
Signature · 1 line
2.34.1
author maintainer
#459 lore raw

PATCH 459/1039 RDMA/rtrs-clt: Fix the initial value of min_latency

In reply to Greg Kroah-Hartman

From: Jack Wang <[email protected]>

[ Upstream commit 925cac6358677d3d64f9b25f205eeb3d31c9f7f8 ]

The type of min_latency is ktime_t, so use KTIME_MAX to initialize the initial value.

Fixes dc3b66a0ce70 ("RDMA/rtrs-clt: Add a minimum latency multipath policy")
---
 drivers/infiniband/ulp/rtrs/rtrs-clt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/infiniband/ulp/rtrs/rtrs-clt.c +1 -1
index 15c0077dd27eb..e39709dee179d 100644
@@ -867,7 +867,7 @@ static struct rtrs_clt_sess *get_next_path_min_latency(struct path_it *it)
 	struct rtrs_clt_sess *min_path = NULL;
 	struct rtrs_clt *clt = it->clt;
 	struct rtrs_clt_sess *sess;
-	ktime_t min_latency = INT_MAX;
+	ktime_t min_latency = KTIME_MAX;
 	ktime_t latency;
 
 	list_for_each_entry_rcu(sess, &clt->paths_list, s.entry) {
Signature · 1 line
2.34.1
author maintainer
#460 lore raw

PATCH 460/1039 ALSA: hda: Make proper use of timecounter

In reply to Greg Kroah-Hartman

From: Thomas Gleixner <[email protected]>

[ Upstream commit 6dd21ad81bf96478db3403b1bbe251c0612d0431 ]

HDA uses a timecounter to read a hardware clock running at 24 MHz. The conversion factor is set with a mult value of 125 and a shift value of 0, which is not converting the hardware clock to nanoseconds, it is converting to 1/3 nanoseconds because the conversion factor from 24Mhz to nanoseconds is 125/3. The usage sites divide the "nanoseconds" value returned by timecounter_read() by 3 to get a real nanoseconds value.

There is a lengthy comment in azx_timecounter_init() explaining this choice. That comment makes blatantly wrong assumptions about how timecounters work and what can overflow.

The comment says:

* Applying the 1/3 factor as part of the multiplication
* requires at least 20 bits for a decent precision, however
* overflows occur after about 4 hours or less, not a option.

timecounters operate on time deltas between two readouts of a clock and use the mult/shift pair to calculate a precise nanoseconds value:

    delta_nsec = (delta_clock * mult) >> shift;

The fractional part is also taken into account and preserved to prevent accumulated rounding errors. For details see cyclecounter_cyc2ns().

The mult/shift pair has to be chosen so that the multiplication of the maximum expected delta value does not result in a 64bit overflow. As the counter wraps around on 32bit, the maximum observable delta between two reads is (1 << 32) - 1 which is about 178.9 seconds.

That in turn means the maximum multiplication factor which fits into an u32 will not cause a 64bit overflow ever because it's guaranteed that:

     ((1 << 32) - 1) ^ 2 < (1 << 64)

The resulting correct multiplication factor is 2796202667 and the shift value is 26, i.e. 26 bit precision. The overflow of the multiplication would happen exactly at a clock readout delta of 6597069765 which is way after the wrap around of the hardware clock at around 274.8 seconds which is off from the claimed 4 hours by more than an order of magnitude.

If the counter ever wraps around the last read value then the calculation is off by the number of wrap arounds times 178.9 seconds because the overflow cannot be observed.

Use clocks_calc_mult_shift(), which calculates the most accurate mult/shift pair based on the given clock frequency, and remove the bogus comment along with the divisions at the readout sites.

---
 sound/hda/hdac_stream.c           | 14 ++++----------
 sound/pci/hda/hda_controller.c    |  1 -
 sound/soc/intel/skylake/skl-pcm.c |  1 -
 3 files changed, 4 insertions(+), 12 deletions(-)
sound/hda/hdac_stream.c +4 -10
index 9867555883c34..aa7955fdf68a0 100644
@@ -534,17 +534,11 @@ static void azx_timecounter_init(struct hdac_stream *azx_dev,
 	cc->mask = CLOCKSOURCE_MASK(32);
 
 	/*
-	 * Converting from 24 MHz to ns means applying a 125/3 factor.
-	 * To avoid any saturation issues in intermediate operations,
-	 * the 125 factor is applied first. The division is applied
-	 * last after reading the timecounter value.
-	 * Applying the 1/3 factor as part of the multiplication
-	 * requires at least 20 bits for a decent precision, however
-	 * overflows occur after about 4 hours or less, not a option.
+	 * Calculate the optimal mult/shift values. The counter wraps
+	 * around after ~178.9 seconds.
 	 */
-
-	cc->mult = 125; /* saturation after 195 years */
-	cc->shift = 0;
+	clocks_calc_mult_shift(&cc->mult, &cc->shift, 24000000,
+			       NSEC_PER_SEC, 178);
 
 	nsec = 0; /* audio time is elapsed time since trigger */
 	timecounter_init(tc, cc, nsec);
sound/pci/hda/hda_controller.c +0 -1
index 930ae4002a818..75dcb14ff20ad 100644
@@ -504,7 +504,6 @@ static int azx_get_time_info(struct snd_pcm_substream *substream,
 		snd_pcm_gettime(substream->runtime, system_ts);
 
 		nsec = timecounter_read(&azx_dev->core.tc);
-		nsec = div_u64(nsec, 3); /* can be optimized */
 		if (audio_tstamp_config->report_delay)
 			nsec = azx_adjust_codec_delay(substream, nsec);
 
sound/soc/intel/skylake/skl-pcm.c +0 -1
index 9ecaf6a1e8475..e4aa366d356eb 100644
@@ -1251,7 +1251,6 @@ static int skl_platform_soc_get_time_info(
 		snd_pcm_gettime(substream->runtime, system_ts);
 
 		nsec = timecounter_read(&hstr->tc);
-		nsec = div_u64(nsec, 3); /* can be optimized */
 		if (audio_tstamp_config->report_delay)
 			nsec = skl_adjust_codec_delay(substream, nsec);
 
Signature · 1 line
2.34.1
author maintainer
#461 lore raw

PATCH 461/1039 dt-bindings: thermal: Fix definition of cooling-maps contribution property

In reply to Greg Kroah-Hartman

From: Niklas Söderlund <[email protected]>

[ Upstream commit 49bcb1506f2e095262c01bda7fd1c0db524c91e2 ]

When converting the thermal-zones bindings to yaml the definition of the contribution property changed. The intention is the same, an integer value expressing a ratio of a sum on how much cooling is provided by the device to the zone. But after the conversion the integer value is limited to the range 0 to 100 and expressed as a percentage.

This is problematic for two reasons.

- This do not match how the binding is used. Out of the 18 files that make use of the property only two (ste-dbx5x0.dtsi and ste-hrefv60plus.dtsi) sets it at a value that satisfy the binding,
100. The remaining 16 files set the value higher and fail to validate.

- Expressing the value as a percentage instead of a ratio of the sum is confusing as there is nothing to enforce the sum in the zone is not greater then 100.

This patch restore the pre yaml conversion description and removes the value limitation allowing the usage of the bindings to validate.

Fixes 1202a442a31fd2e5 ("dt-bindings: thermal: Add yaml bindings for thermal zones")
---
 .../devicetree/bindings/thermal/thermal-zones.yaml       | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)
Documentation/devicetree/bindings/thermal/thermal-zones.yaml +4 -5
index a07de5ed0ca6a..2d34f3ccb2572 100644
@@ -199,12 +199,11 @@ patternProperties:
 
               contribution:
                 $ref: /schemas/types.yaml#/definitions/uint32
-                minimum: 0
-                maximum: 100
                 description:
-                  The percentage contribution of the cooling devices at the
-                  specific trip temperature referenced in this map
-                  to this thermal zone
+                  The cooling contribution to the thermal zone of the referred
+                  cooling device at the referred trip point. The contribution is
+                  a ratio of the sum of all cooling contributions within a
+                  thermal zone.
 
             required:
               - trip
Signature · 1 line
2.34.1
author maintainer
#462 lore raw

PATCH 462/1039 powerpc/perf: Fix PMU callbacks to clear pending PMI before resetting an overflown PMC

In reply to Greg Kroah-Hartman

From: Athira Rajeev <[email protected]>

[ Upstream commit 2c9ac51b850d84ee496b0a5d832ce66d411ae552 ]

Running perf fuzzer showed below in dmesg logs:
"Can't find PMC that caused IRQ"

This means a PMU exception happened, but none of the PMC's (Performance Monitor Counter) were found to be overflown. There are some corner cases that clears the PMCs after PMI gets masked. In such cases, the perf interrupt handler will not find the active PMC values that had caused the overflow and thus leads to this message while replaying.

Case 1: PMU Interrupt happens during replay of other interrupts and counter values gets cleared by PMU callbacks before replay:

During replay of interrupts like timer, __do_irq() and doorbell exception, we conditionally enable interrupts via may_hard_irq_enable(). This could potentially create a window to generate a PMI. Since irq soft mask is set to ALL_DISABLED, the PMI will get masked here. We could get IPIs run before perf interrupt is replayed and the PMU events could be deleted or stopped. This will change the PMU SPR values and resets the counters. Snippet of ftrace log showing PMU callbacks invoked in __do_irq():

<idle>-0 [051] dns. 132025441306354: __do_irq <-call_do_irq
<idle>-0 [051] dns. 132025441306430: irq_enter <-__do_irq
<idle>-0 [051] dns. 132025441306503: irq_enter_rcu <-__do_irq
<idle>-0 [051] dnH. 132025441306599: xive_get_irq <-__do_irq
<<>>
<idle>-0 [051] dnH. 132025441307770: generic_smp_call_function_single_interrupt <-smp_ipi_demux_relaxed <idle>-0 [051] dnH. 132025441307839: flush_smp_call_function_queue <-smp_ipi_demux_relaxed <idle>-0 [051] dnH. 132025441308057: _raw_spin_lock <-event_function
<idle>-0 [051] dnH. 132025441308206: power_pmu_disable <-perf_pmu_disable
<idle>-0 [051] dnH. 132025441308337: power_pmu_del <-event_sched_out
<idle>-0 [051] dnH. 132025441308407: power_pmu_read <-power_pmu_del
<idle>-0 [051] dnH. 132025441308477: read_pmc <-power_pmu_read
<idle>-0 [051] dnH. 132025441308590: isa207_disable_pmc <-power_pmu_del
<idle>-0 [051] dnH. 132025441308663: write_pmc <-power_pmu_del
<idle>-0 [051] dnH. 132025441308787: power_pmu_event_idx <-perf_event_update_userpage <idle>-0 [051] dnH. 132025441308859: rcu_read_unlock_strict <-perf_event_update_userpage <idle>-0 [051] dnH. 132025441308975: power_pmu_enable <-perf_pmu_enable
<<>>
<idle>-0 [051] dnH. 132025441311108: irq_exit <-__do_irq
<idle>-0 [051] dns. 132025441311319: performance_monitor_exception <-replay_soft_interrupts

Case 2: PMI's masked during local_* operations, example local_add(). If the local_add() operation happens within a local_irq_save(), replay of PMI will be during local_irq_restore(). Similar to case 1, this could also create a window before replay where PMU events gets deleted or stopped.

Fix it by updating the PMU callback function power_pmu_disable() to check for pending perf interrupt. If there is an overflown PMC and pending perf interrupt indicated in paca, clear the PMI bit in paca to drop that sample. Clearing of PMI bit is done in power_pmu_disable() since disable is invoked before any event gets deleted/stopped. With this fix, if there are more than one event running in the PMU, there is a chance that we clear the PMI bit for the event which is not getting deleted/stopped. The other events may still remain active. Hence to make sure we don't drop valid sample in such cases, another check is added in power_pmu_enable. This checks if there is an overflown PMC found among the active events and if so enable back the PMI bit. Two new helper functions are introduced to clear/set the PMI, ie clear_pmi_irq_pending() and set_pmi_irq_pending(). Helper function pmi_irq_pending() is introduced to give a warning if there is pending PMI bit in paca, but no PMC is overflown.

Also there are corner cases which result in performance monitor interrupts being triggered during power_pmu_disable(). This happens since PMXE bit is not cleared along with disabling of other MMCR0 bits in the pmu_disable. Such PMI's could leave the PMU running and could trigger PMI again which will set MMCR0 PMAO bit. This could lead to spurious interrupts in some corner cases. Example, a timer after power_pmu_del() which will re-enable interrupts and triggers a PMI again since PMAO bit is still set. But fails to find valid overflow since PMC was cleared in power_pmu_del(). Fix that by disabling PMXE along with disabling of other MMCR0 bits in power_pmu_disable().

We can't just replay PMI any time. Hence this approach is preferred rather than replaying PMI before resetting overflown PMC. Patch also documents core-book3s on a race condition which can trigger these PMC messages during idle path in PowerNV.

Fixes f442d004806e ("powerpc/64s: Add support to mask perf interrupts and replay them")
[mpe: Make pmi_irq_pending() return bool, reflow/reword some comments]
---
 arch/powerpc/include/asm/hw_irq.h | 40 +++++++++++++++++++++
 arch/powerpc/perf/core-book3s.c   | 58 ++++++++++++++++++++++++++++++-
 2 files changed, 97 insertions(+), 1 deletion(-)
arch/powerpc/include/asm/hw_irq.h +40 -0
index 21cc571ea9c2d..5c98a950eca0d 100644
@@ -224,6 +224,42 @@ static inline bool arch_irqs_disabled(void)
 	return arch_irqs_disabled_flags(arch_local_save_flags());
 }
 
+static inline void set_pmi_irq_pending(void)
+{
+	/*
+	 * Invoked from PMU callback functions to set PMI bit in the paca.
+	 * This has to be called with irq's disabled (via hard_irq_disable()).
+	 */
+	if (IS_ENABLED(CONFIG_PPC_IRQ_SOFT_MASK_DEBUG))
+		WARN_ON_ONCE(mfmsr() & MSR_EE);
+
+	get_paca()->irq_happened |= PACA_IRQ_PMI;
+}
+
+static inline void clear_pmi_irq_pending(void)
+{
+	/*
+	 * Invoked from PMU callback functions to clear the pending PMI bit
+	 * in the paca.
+	 */
+	if (IS_ENABLED(CONFIG_PPC_IRQ_SOFT_MASK_DEBUG))
+		WARN_ON_ONCE(mfmsr() & MSR_EE);
+
+	get_paca()->irq_happened &= ~PACA_IRQ_PMI;
+}
+
+static inline bool pmi_irq_pending(void)
+{
+	/*
+	 * Invoked from PMU callback functions to check if there is a pending
+	 * PMI bit in the paca.
+	 */
+	if (get_paca()->irq_happened & PACA_IRQ_PMI)
+		return true;
+
+	return false;
+}
+
 #ifdef CONFIG_PPC_BOOK3S
 /*
  * To support disabling and enabling of irq with PMI, set of
@@ -408,6 +444,10 @@ static inline void do_hard_irq_enable(void)
 	BUILD_BUG();
 }
 
+static inline void clear_pmi_irq_pending(void) { }
+static inline void set_pmi_irq_pending(void) { }
+static inline bool pmi_irq_pending(void) { return false; }
+
 static inline void irq_soft_mask_regs_set_state(struct pt_regs *regs, unsigned long val)
 {
 }
arch/powerpc/perf/core-book3s.c +57 -1

98 lines not on this page · raw message

Signature · 1 line
2.34.1
author maintainer
#463 lore raw

PATCH 463/1039 powerpc/modules: Dont WARN on first module allocation attempt

In reply to Greg Kroah-Hartman

From: Christophe Leroy <[email protected]>

[ Upstream commit f1797e4de1146009c888bcf8b6bb6648d55394f1 ]

module_alloc() first tries to allocate module text within 24 bits direct jump from kernel text, and tries a wider allocation if first one fails.

When first allocation fails the following is observed in kernel logs:

vmap allocation for size 2400256 failed: use vmalloc=<size> to increase size
systemd-udevd: vmalloc error: size 2395133, vm_struct allocation failed, mode:0xcc0(GFP_KERNEL), nodemask=(null)

  CPU: 0 PID: 127 Comm: systemd-udevd Tainted: G        W         5.15.5-gentoo-PowerMacG4 #9
  Call Trace:
  [e2a53a50] [c0ba0048] dump_stack_lvl+0x80/0xb0 (unreliable)
  [e2a53a70] [c0540128] warn_alloc+0x11c/0x2b4
  [e2a53b50] [c0531be8] __vmalloc_node_range+0xd8/0x64c
  [e2a53c10] [c00338c0] module_alloc+0xa0/0xac
  [e2a53c40] [c027a368] load_module+0x2ae0/0x8148
  [e2a53e30] [c027fc78] sys_finit_module+0xfc/0x130
  [e2a53f30] [c0035098] ret_from_syscall+0x0/0x28
  ...

Add __GFP_NOWARN flag to first allocation so that no warning appears when it fails.

---
 arch/powerpc/kernel/module.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)
arch/powerpc/kernel/module.c +6 -5
index ed04a3ba66fe8..40a583e9d3c70 100644
@@ -90,16 +90,17 @@ int module_finalize(const Elf_Ehdr *hdr,
 }
 
 static __always_inline void *
-__module_alloc(unsigned long size, unsigned long start, unsigned long end)
+__module_alloc(unsigned long size, unsigned long start, unsigned long end, bool nowarn)
 {
 	pgprot_t prot = strict_module_rwx_enabled() ? PAGE_KERNEL : PAGE_KERNEL_EXEC;
+	gfp_t gfp = GFP_KERNEL | (nowarn ? __GFP_NOWARN : 0);
 
 	/*
 	 * Don't do huge page allocations for modules yet until more testing
 	 * is done. STRICT_MODULE_RWX may require extra work to support this
 	 * too.
 	 */
-	return __vmalloc_node_range(size, 1, start, end, GFP_KERNEL, prot,
+	return __vmalloc_node_range(size, 1, start, end, gfp, prot,
 				    VM_FLUSH_RESET_PERMS | VM_NO_HUGE_VMAP,
 				    NUMA_NO_NODE, __builtin_return_address(0));
 }
@@ -114,13 +115,13 @@ void *module_alloc(unsigned long size)
 
 	/* First try within 32M limit from _etext to avoid branch trampolines */
 	if (MODULES_VADDR < PAGE_OFFSET && MODULES_END > limit)
-		ptr = __module_alloc(size, limit, MODULES_END);
+		ptr = __module_alloc(size, limit, MODULES_END, true);
 
 	if (!ptr)
-		ptr = __module_alloc(size, MODULES_VADDR, MODULES_END);
+		ptr = __module_alloc(size, MODULES_VADDR, MODULES_END, false);
 
 	return ptr;
 #else
-	return __module_alloc(size, VMALLOC_START, VMALLOC_END);
+	return __module_alloc(size, VMALLOC_START, VMALLOC_END, false);
 #endif
 }
Signature · 1 line
2.34.1
author maintainer
#464 lore raw

PATCH 464/1039 powerpc/32s: Fix shift-out-of-bounds in KASAN init

In reply to Greg Kroah-Hartman

From: Christophe Leroy <[email protected]>

[ Upstream commit af11dee4361b3519981fa04d014873f9d9edd6ac ]

================================================================================

UBSAN: shift-out-of-bounds in arch/powerpc/mm/kasan/book3s_32.c:22:23
shift exponent -1 is negative
CPU: 0 PID: 0 Comm: swapper Not tainted 5.15.5-gentoo-PowerMacG4 #9
Call Trace:
[c214be60] [c0ba0048] dump_stack_lvl+0x80/0xb0 (unreliable)
[c214be80] [c0b99288] ubsan_epilogue+0x10/0x5c
[c214be90] [c0b98fe0] __ubsan_handle_shift_out_of_bounds+0x94/0x138
[c214bf00] [c1c0f010] kasan_init_region+0xd8/0x26c
[c214bf30] [c1c0ed84] kasan_init+0xc0/0x198
[c214bf70] [c1c08024] setup_arch+0x18/0x54c
[c214bfc0] [c1c037f0] start_kernel+0x90/0x33c

[c214bff0] [00003610] 0x3610
================================================================================

This happens when the directly mapped memory is a power of 2.

Fix it by checking the shift and set the result to 0 when shift is -1

---
 arch/powerpc/mm/kasan/book3s_32.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
arch/powerpc/mm/kasan/book3s_32.c +2 -1
index 202bd260a0095..35b287b0a8da4 100644
@@ -19,7 +19,8 @@ int __init kasan_init_region(void *start, size_t size)
 	block = memblock_alloc(k_size, k_size_base);
 
 	if (block && k_size_base >= SZ_128K && k_start == ALIGN(k_start, k_size_base)) {
-		int k_size_more = 1 << (ffs(k_size - k_size_base) - 1);
+		int shift = ffs(k_size - k_size_base);
+		int k_size_more = shift ? 1 << (shift - 1) : 0;
 
 		setbat(-1, k_start, __pa(block), k_size_base, PAGE_KERNEL);
 		if (k_size_more >= SZ_128K)
Signature · 1 line
2.34.1
author maintainer
#465 lore raw

PATCH 465/1039 clocksource: Avoid accidental unstable marking of clocksources

In reply to Greg Kroah-Hartman

From: Waiman Long <[email protected]>

[ Upstream commit c86ff8c55b8ae68837b2fa59dc0c203907e9a15f ]

Since commit db3a34e17433 ("clocksource: Retry clock read if long delays detected") and commit 2e27e793e280 ("clocksource: Reduce clocksource-skew threshold"), it is found that tsc clocksource fallback to hpet can sometimes happen on both Intel and AMD systems especially when they are running stressful benchmarking workloads. Of the 23 systems tested with a v5.14 kernel, 10 of them have switched to hpet clock source during the test run.

The result of falling back to hpet is a drastic reduction of performance when running benchmarks. For example, the fio performance tests can drop up to 70% whereas the iperf3 performance can drop up to 80%.

4 hpet fallbacks happened during bootup. They were:

  [    8.749399] clocksource: timekeeping watchdog on CPU13: hpet read-back delay of 263750ns, attempt 4, marking unstable
  [   12.044610] clocksource: timekeeping watchdog on CPU19: hpet read-back delay of 186166ns, attempt 4, marking unstable
  [   17.336941] clocksource: timekeeping watchdog on CPU28: hpet read-back delay of 182291ns, attempt 4, marking unstable
  [   17.518565] clocksource: timekeeping watchdog on CPU34: hpet read-back delay of 252196ns, attempt 4, marking unstable

Other fallbacks happen when the systems were running stressful benchmarks. For example:

  [ 2685.867873] clocksource: timekeeping watchdog on CPU117: hpet read-back delay of 57269ns, attempt 4, marking unstable
  [46215.471228] clocksource: timekeeping watchdog on CPU8: hpet read-back delay of 61460ns, attempt 4, marking unstable

Commit 2e27e793e280 ("clocksource: Reduce clocksource-skew threshold"), changed the skew margin from 100us to 50us. I think this is too small and can easily be exceeded when running some stressful workloads on a thermally stressed system. So it is switched back to 100us.

Even a maximum skew margin of 100us may be too small in for some systems when booting up especially if those systems are under thermal stress. To eliminate the case that the large skew is due to the system being too busy slowing down the reading of both the watchdog and the clocksource, an extra consecutive read of watchdog clock is being done to check this.

The consecutive watchdog read delay is compared against WATCHDOG_MAX_SKEW/2. If the delay exceeds the limit, we assume that the system is just too busy. A warning will be printed to the console and the clock skew check is skipped for this round.

Fixes db3a34e17433 ("clocksource: Retry clock read if long delays detected")
Fixes 2e27e793e280 ("clocksource: Reduce clocksource-skew threshold")
---
 kernel/time/clocksource.c | 50 ++++++++++++++++++++++++++++++++-------
 1 file changed, 41 insertions(+), 9 deletions(-)
kernel/time/clocksource.c +41 -9
index b8a14d2fb5ba6..bcad1a1e5dcf1 100644
@@ -107,7 +107,7 @@ static u64 suspend_start;
  * This delay could be due to SMIs, NMIs, or to VCPU preemptions.  Used as
  * a lower bound for cs->uncertainty_margin values when registering clocks.
  */
-#define WATCHDOG_MAX_SKEW (50 * NSEC_PER_USEC)
+#define WATCHDOG_MAX_SKEW (100 * NSEC_PER_USEC)
 
 #ifdef CONFIG_CLOCKSOURCE_WATCHDOG
 static void clocksource_watchdog_work(struct work_struct *work);
@@ -205,17 +205,24 @@ EXPORT_SYMBOL_GPL(max_cswd_read_retries);
 static int verify_n_cpus = 8;
 module_param(verify_n_cpus, int, 0644);
 
-static bool cs_watchdog_read(struct clocksource *cs, u64 *csnow, u64 *wdnow)
+enum wd_read_status {
+	WD_READ_SUCCESS,
+	WD_READ_UNSTABLE,
+	WD_READ_SKIP
+};
+
+static enum wd_read_status cs_watchdog_read(struct clocksource *cs, u64 *csnow, u64 *wdnow)
 {
 	unsigned int nretries;
-	u64 wd_end, wd_delta;
-	int64_t wd_delay;
+	u64 wd_end, wd_end2, wd_delta;
+	int64_t wd_delay, wd_seq_delay;
 
 	for (nretries = 0; nretries <= max_cswd_read_retries; nretries++) {
 		local_irq_disable();
 		*wdnow = watchdog->read(watchdog);
 		*csnow = cs->read(cs);
 		wd_end = watchdog->read(watchdog);
+		wd_end2 = watchdog->read(watchdog);
 		local_irq_enable();
 
 		wd_delta = clocksource_delta(wd_end, *wdnow, watchdog->mask);
@@ -226,13 +233,34 @@ static bool cs_watchdog_read(struct clocksource *cs, u64 *csnow, u64 *wdnow)
 				pr_warn("timekeeping watchdog on CPU%d: %s retried %d times before success\n",
 					smp_processor_id(), watchdog->name, nretries);
 			}
-			return true;
+			return WD_READ_SUCCESS;
 		}
+
+		/*
+		 * Now compute delay in consecutive watchdog read to see if
+		 * there is too much external interferences that cause
+		 * significant delay in reading both clocksource and watchdog.
+		 *
+		 * If consecutive WD read-back delay > WATCHDOG_MAX_SKEW/2,
+		 * report system busy, reinit the watchdog and skip the current
+		 * watchdog test.
+		 */
+		wd_delta = clocksource_delta(wd_end2, wd_end, watchdog->mask);
+		wd_seq_delay = clocksource_cyc2ns(wd_delta, watchdog->mult, watchdog->shift);
+		if (wd_seq_delay > WATCHDOG_MAX_SKEW/2)
+			goto skip_test;
 	}
 
 	pr_warn("timekeeping watchdog on CPU%d: %s read-back delay of %lldns, attempt %d, marking unstable\n",
 		smp_processor_id(), watchdog->name, wd_delay, nretries);
-	return false;
+	return WD_READ_UNSTABLE;
+
+skip_test:
+	pr_info("timekeeping watchdog on CPU%d: %s wd-wd read-back delay of %lldns\n",
+		smp_processor_id(), watchdog->name, wd_seq_delay);
+	pr_info("wd-%s-wd read-back delay of %lldns, clock-skew test skipped!\n",
+		cs->name, wd_delay);
+	return WD_READ_SKIP;
 }
 
 static u64 csnow_mid;
@@ -356,6 +384,7 @@ static void clocksource_watchdog(struct timer_list *unused)
 	int next_cpu, reset_pending;
 	int64_t wd_nsec, cs_nsec;
 	struct clocksource *cs;
+	enum wd_read_status read_ret;
 	u32 md;
 
 	spin_lock(&watchdog_lock);
@@ -373,9 +402,12 @@ static void clocksource_watchdog(struct timer_list *unused)
 			continue;
 		}
 
-		if (!cs_watchdog_read(cs, &csnow, &wdnow)) {
-			/* Clock readout unreliable, so give it up. */
-			__clocksource_unstable(cs);
+		read_ret = cs_watchdog_read(cs, &csnow, &wdnow);
+
+		if (read_ret != WD_READ_SUCCESS) {
+			if (read_ret == WD_READ_UNSTABLE)
+				/* Clock readout unreliable, so give it up. */
+				__clocksource_unstable(cs);
 			continue;
 		}
 
Signature · 1 line
2.34.1
author maintainer
#466 lore raw

PATCH 466/1039 ALSA: oss: fix compile error when OSS_DEBUG is enabled

In reply to Greg Kroah-Hartman

From: Bixuan Cui <[email protected]>

[ Upstream commit 8e7daf318d97f25e18b2fc7eb5909e34cd903575 ]

Fix compile error when OSS_DEBUG is enabled:

    sound/core/oss/pcm_oss.c: In function 'snd_pcm_oss_set_trigger':
    sound/core/oss/pcm_oss.c:2055:10: error: 'substream' undeclared (first
    use in this function); did you mean 'csubstream'?
      pcm_dbg(substream->pcm, "pcm_oss: trigger = 0x%x\n", trigger);
              ^
Fixes 61efcee8608c ("ALSA: oss: Use standard printk helpers")
---
 sound/core/oss/pcm_oss.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
sound/core/oss/pcm_oss.c +1 -1
index 20a0a4771b9a8..3ee9edf858156 100644
@@ -2065,7 +2065,7 @@ static int snd_pcm_oss_set_trigger(struct snd_pcm_oss_file *pcm_oss_file, int tr
 	int err, cmd;
 
 #ifdef OSS_DEBUG
-	pcm_dbg(substream->pcm, "pcm_oss: trigger = 0x%x\n", trigger);
+	pr_debug("pcm_oss: trigger = 0x%x\n", trigger);
 #endif
 	
 	psubstream = pcm_oss_file->streams[SNDRV_PCM_STREAM_PLAYBACK];
Signature · 1 line
2.34.1
author maintainer
#467 lore raw

PATCH 467/1039 ALSA: usb-audio: Drop superfluous 0 in Presonus Studio 1810cs ID

In reply to Greg Kroah-Hartman

From: Takashi Iwai <[email protected]>

[ Upstream commit 1e583aef12aa74afd37c1418255cc4b74e023236 ]

The vendor ID of Presonus Studio 1810c had a superfluous '0' in its USB ID. Drop it.

Fixes 8dc5efe3d17c ("ALSA: usb-audio: Add support for Presonus Studio 1810c")
---
 sound/usb/format.c       | 2 +-
 sound/usb/mixer_quirks.c | 2 +-
 sound/usb/quirks.c       | 2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)
sound/usb/format.c +1 -1
index f5e676a51b30d..405dc0bf6678c 100644
@@ -375,7 +375,7 @@ static int parse_uac2_sample_rate_range(struct snd_usb_audio *chip,
 		for (rate = min; rate <= max; rate += res) {
 
 			/* Filter out invalid rates on Presonus Studio 1810c */
-			if (chip->usb_id == USB_ID(0x0194f, 0x010c) &&
+			if (chip->usb_id == USB_ID(0x194f, 0x010c) &&
 			    !s1810c_valid_sample_rate(fp, rate))
 				goto skip_rate;
 
sound/usb/mixer_quirks.c +1 -1
index 823b6b8de942d..d48729e6a3b0a 100644
@@ -3254,7 +3254,7 @@ int snd_usb_mixer_apply_create_quirk(struct usb_mixer_interface *mixer)
 		err = snd_rme_controls_create(mixer);
 		break;
 
-	case USB_ID(0x0194f, 0x010c): /* Presonus Studio 1810c */
+	case USB_ID(0x194f, 0x010c): /* Presonus Studio 1810c */
 		err = snd_sc1810_init_mixer(mixer);
 		break;
 	case USB_ID(0x2a39, 0x3fb0): /* RME Babyface Pro FS */
sound/usb/quirks.c +1 -1
index 64e1c20311ed4..ab9f3da49941f 100644
@@ -1290,7 +1290,7 @@ int snd_usb_apply_interface_quirk(struct snd_usb_audio *chip,
 	if (chip->usb_id == USB_ID(0x0763, 0x2012))
 		return fasttrackpro_skip_setting_quirk(chip, iface, altno);
 	/* presonus studio 1810c: skip altsets incompatible with device_setup */
-	if (chip->usb_id == USB_ID(0x0194f, 0x010c))
+	if (chip->usb_id == USB_ID(0x194f, 0x010c))
 		return s1810c_skip_setting_quirk(chip, iface, altno);
 
 
Signature · 1 line
2.34.1
author maintainer
#468 lore raw

PATCH 468/1039 ASoC: amd: Fix dependency for SPI master

In reply to Greg Kroah-Hartman

From: Lucas Tanure <[email protected]>

[ Upstream commit 19a628d8f1a6c16263d8037a918427207c8a95a0 ]

Set SPI_MASTER as dependency as is using CS35L41 SPI driver

Fixes 96792fdd77cd1 ("ASoC: amd: enable vangogh platform machine driver build")
---
 sound/soc/amd/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
sound/soc/amd/Kconfig +1 -1
index 2c6af3f8f2961..ff535370e525a 100644
@@ -68,7 +68,7 @@ config SND_SOC_AMD_VANGOGH_MACH
 	tristate "AMD Vangogh support for NAU8821 CS35L41"
 	select SND_SOC_NAU8821
 	select SND_SOC_CS35L41_SPI
-	depends on SND_SOC_AMD_ACP5x && I2C
+	depends on SND_SOC_AMD_ACP5x && I2C && SPI_MASTER
 	help
 	  This option enables machine driver for Vangogh platform
 	  using NAU8821 and CS35L41 codecs.
Signature · 1 line
2.34.1
author maintainer
#469 lore raw

PATCH 469/1039 misc: at25: Make driver OF independent again

In reply to Greg Kroah-Hartman

From: Andy Shevchenko <[email protected]>

[ Upstream commit 5b557298d7d09cce04e0565a535fbca63661724a ]

The commit f60e7074902a ("misc: at25: Make use of device property API") made a good job by enabling the driver for non-OF platforms, but the recent commit 604288bc6196 ("nvmem: eeprom: at25: fix type compiler warnings") brought that back.

Restore greatness of the driver once again.

Fixes eab61fb1cc2e ("nvmem: eeprom: at25: fram discovery simplification")
Fixes fd307a4ad332 ("nvmem: prepare basics for FRAM support")
---
 drivers/misc/eeprom/at25.c | 13 ++++++-------
 1 file changed, 6 insertions(+), 7 deletions(-)
drivers/misc/eeprom/at25.c +6 -7
index b38978a3b3ffa..9193b812bc07e 100644
@@ -17,8 +17,6 @@
 #include <linux/spi/spi.h>
 #include <linux/spi/eeprom.h>
 #include <linux/property.h>
-#include <linux/of.h>
-#include <linux/of_device.h>
 #include <linux/math.h>
 
 /*
@@ -380,13 +378,14 @@ static int at25_probe(struct spi_device *spi)
 	int			sr;
 	u8 id[FM25_ID_LEN];
 	u8 sernum[FM25_SN_LEN];
+	bool is_fram;
 	int i;
-	const struct of_device_id *match;
-	bool is_fram = 0;
 
-	match = of_match_device(of_match_ptr(at25_of_match), &spi->dev);
-	if (match && !strcmp(match->compatible, "cypress,fm25"))
-		is_fram = 1;
+	err = device_property_match_string(&spi->dev, "compatible", "cypress,fm25");
+	if (err >= 0)
+		is_fram = true;
+	else
+		is_fram = false;
 
 	at25 = devm_kzalloc(&spi->dev, sizeof(struct at25_data), GFP_KERNEL);
 	if (!at25)
Signature · 1 line
2.34.1
author maintainer
#470 lore raw

PATCH 470/1039 char/mwave: Adjust io port register size

In reply to Greg Kroah-Hartman

From: Kees Cook <[email protected]>

[ Upstream commit f5912cc19acd7c24b2dbf65a6340bf194244f085 ]

Using MKWORD() on a byte-sized variable results in OOB read. Expand the size of the reserved area so both MKWORD and MKBYTE continue to work without overflow. Silences this warning on a -Warray-bounds build:

drivers/char/mwave/3780i.h:346:22: error: array subscript 'short unsigned int[0]' is partly outside array bounds of 'DSP_ISA_SLAVE_CONTROL[1]' [-Werror=array-bounds]
346 | #define MKWORD(var) (*((unsigned short *)(&var)))

      |                     ~^~~~~~~~~~~~~~~~~~~~~~~~~~~~

drivers/char/mwave/3780i.h:356:40: note: in definition of macro 'OutWordDsp'
356 | #define OutWordDsp(index,value) outw(value,usDspBaseIO+index)
| ^~~~~
drivers/char/mwave/3780i.c:373:41: note: in expansion of macro 'MKWORD'
373 | OutWordDsp(DSP_IsaSlaveControl, MKWORD(rSlaveControl));
| ^~~~~~
drivers/char/mwave/3780i.c:358:31: note: while referencing 'rSlaveControl'

  358 |         DSP_ISA_SLAVE_CONTROL rSlaveControl;
      |                               ^~~~~~~~~~~~~
---
 drivers/char/mwave/3780i.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/char/mwave/3780i.h +1 -1
index 9ccb6b270b071..95164246afd1a 100644
@@ -68,7 +68,7 @@ typedef struct {
 	unsigned char ClockControl:1;	/* RW: Clock control: 0=normal, 1=stop 3780i clocks */
 	unsigned char SoftReset:1;	/* RW: Soft reset 0=normal, 1=soft reset active */
 	unsigned char ConfigMode:1;	/* RW: Configuration mode, 0=normal, 1=config mode */
-	unsigned char Reserved:5;	/* 0: Reserved */
+	unsigned short Reserved:13;	/* 0: Reserved */
 } DSP_ISA_SLAVE_CONTROL;
 
 
Signature · 1 line
2.34.1
author maintainer
#471 lore raw

PATCH 471/1039 binder: fix handling of error during copy

In reply to Greg Kroah-Hartman

From: Todd Kjos <[email protected]>

[ Upstream commit fe6b1869243f23a485a106c214bcfdc7aa0ed593 ]

If a memory copy function fails to copy the whole buffer, a positive integar with the remaining bytes is returned. In binder_translate_fd_array() this can result in an fd being skipped due to the failed copy, but the loop continues processing fds since the early return condition expects a negative integer on error.

Fix by returning "ret > 0 ? -EINVAL : ret" to handle this case.

---
 drivers/android/binder.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/android/binder.c +2 -2
index c75fb600740cc..7d29d3d931a79 100644
@@ -2269,8 +2269,8 @@ static int binder_translate_fd_array(struct binder_fd_array_object *fda,
 		if (!ret)
 			ret = binder_translate_fd(fd, offset, t, thread,
 						  in_reply_to);
-		if (ret < 0)
-			return ret;
+		if (ret)
+			return ret > 0 ? -EINVAL : ret;
 	}
 	return 0;
 }
Signature · 1 line
2.34.1
author maintainer
#472 lore raw

PATCH 472/1039 binder: avoid potential data leakage when copying txn

In reply to Greg Kroah-Hartman

From: Todd Kjos <[email protected]>

[ Upstream commit 6d98eb95b450a75adb4516a1d33652dc78d2b20c ]

Transactions are copied from the sender to the target first and objects like BINDER_TYPE_PTR and BINDER_TYPE_FDA are then fixed up. This means there is a short period where the sender's version of these objects are visible to the target prior to the fixups.

Instead of copying all of the data first, copy data only after any needed fixups have been applied.

---
 drivers/android/binder.c | 94 ++++++++++++++++++++++++++++++----------
 1 file changed, 70 insertions(+), 24 deletions(-)
drivers/android/binder.c +70 -24
index 7d29d3d931a79..99ae919255f4d 100644
@@ -1608,15 +1608,21 @@ static void binder_cleanup_transaction(struct binder_transaction *t,
 /**
  * binder_get_object() - gets object and checks for valid metadata
  * @proc:	binder_proc owning the buffer
+ * @u:		sender's user pointer to base of buffer
  * @buffer:	binder_buffer that we're parsing.
  * @offset:	offset in the @buffer at which to validate an object.
  * @object:	struct binder_object to read into
  *
- * Return:	If there's a valid metadata object at @offset in @buffer, the
+ * Copy the binder object at the given offset into @object. If @u is
+ * provided then the copy is from the sender's buffer. If not, then
+ * it is copied from the target's @buffer.
+ *
+ * Return:	If there's a valid metadata object at @offset, the
  *		size of that object. Otherwise, it returns zero. The object
  *		is read into the struct binder_object pointed to by @object.
  */
 static size_t binder_get_object(struct binder_proc *proc,
+				const void __user *u,
 				struct binder_buffer *buffer,
 				unsigned long offset,
 				struct binder_object *object)
@@ -1626,10 +1632,16 @@ static size_t binder_get_object(struct binder_proc *proc,
 	size_t object_size = 0;
 
 	read_size = min_t(size_t, sizeof(*object), buffer->data_size - offset);
-	if (offset > buffer->data_size || read_size < sizeof(*hdr) ||
-	    binder_alloc_copy_from_buffer(&proc->alloc, object, buffer,
-					  offset, read_size))
+	if (offset > buffer->data_size || read_size < sizeof(*hdr))
 		return 0;
+	if (u) {
+		if (copy_from_user(object, u + offset, read_size))
+			return 0;
+	} else {
+		if (binder_alloc_copy_from_buffer(&proc->alloc, object, buffer,
+						  offset, read_size))
+			return 0;
+	}
 
 	/* Ok, now see if we read a complete object. */
 	hdr = &object->hdr;
@@ -1702,7 +1714,7 @@ static struct binder_buffer_object *binder_validate_ptr(
 					  b, buffer_offset,
 					  sizeof(object_offset)))
 		return NULL;
-	object_size = binder_get_object(proc, b, object_offset, object);
+	object_size = binder_get_object(proc, NULL, b, object_offset, object);
 	if (!object_size || object->hdr.type != BINDER_TYPE_PTR)
 		return NULL;
 	if (object_offsetp)
@@ -1767,7 +1779,8 @@ static bool binder_validate_fixup(struct binder_proc *proc,
 		unsigned long buffer_offset;
 		struct binder_object last_object;
 		struct binder_buffer_object *last_bbo;
-		size_t object_size = binder_get_object(proc, b, last_obj_offset,
+		size_t object_size = binder_get_object(proc, NULL, b,
+						       last_obj_offset,
 						       &last_object);
 		if (object_size != sizeof(*last_bbo))
 			return false;
@@ -1882,7 +1895,7 @@ static void binder_transaction_buffer_release(struct binder_proc *proc,
 		if (!binder_alloc_copy_from_buffer(&proc->alloc, &object_offset,
 						   buffer, buffer_offset,
 						   sizeof(object_offset)))
-			object_size = binder_get_object(proc, buffer,
+			object_size = binder_get_object(proc, NULL, buffer,
 							object_offset, &object);
 		if (object_size == 0) {
 			pr_err("transaction release %d bad object at offset %lld, size %zd\n",
@@ -2455,6 +2468,7 @@ static void binder_transaction(struct binder_proc *proc,
 	binder_size_t off_start_offset, off_end_offset;
 	binder_size_t off_min;
 	binder_size_t sg_buf_offset, sg_buf_end_offset;
+	binder_size_t user_offset = 0;
 	struct binder_proc *target_proc = NULL;
 	struct binder_thread *target_thread = NULL;
 	struct binder_node *target_node = NULL;
@@ -2469,6 +2483,8 @@ static void binder_transaction(struct binder_proc *proc,
 	int t_debug_id = atomic_inc_return(&binder_last_id);
 	char *secctx = NULL;
 	u32 secctx_sz = 0;
+	const void __user *user_buffer = (const void __user *)
+				(uintptr_t)tr->data.ptr.buffer;
 
 	e = binder_transaction_log_add(&binder_transaction_log);
 	e->debug_id = t_debug_id;
@@ -2780,19 +2796,6 @@ static void binder_transaction(struct binder_proc *proc,
 	t->buffer->clear_on_free = !!(t->flags & TF_CLEAR_BUF);
 	trace_binder_transaction_alloc_buf(t->buffer);
 
-	if (binder_alloc_copy_user_to_buffer(
-				&target_proc->alloc,
-				t->buffer, 0,
-				(const void __user *)
-					(uintptr_t)tr->data.ptr.buffer,
-				tr->data_size)) {
-		binder_user_error("%d:%d got transaction with invalid data ptr\n",
-				proc->pid, thread->pid);
-		return_error = BR_FAILED_REPLY;
-		return_error_param = -EFAULT;
-		return_error_line = __LINE__;
-		goto err_copy_data_failed;
-	}
 	if (binder_alloc_copy_user_to_buffer(
 				&target_proc->alloc,
 				t->buffer,
@@ -2837,6 +2840,7 @@ static void binder_transaction(struct binder_proc *proc,
 		size_t object_size;
 		struct binder_object object;
 		binder_size_t object_offset;
+		binder_size_t copy_size;
 
 		if (binder_alloc_copy_from_buffer(&target_proc->alloc,
 						  &object_offset,
@@ -2848,8 +2852,27 @@ static void binder_transaction(struct binder_proc *proc,
 			return_error_line = __LINE__;
 			goto err_bad_offset;
 		}
-		object_size = binder_get_object(target_proc, t->buffer,
-						object_offset, &object);
+
+		/*
+		 * Copy the source user buffer up to the next object
+		 * that will be processed.
+		 */
+		copy_size = object_offset - user_offset;
+		if (copy_size && (user_offset > object_offset ||
+				binder_alloc_copy_user_to_buffer(
+					&target_proc->alloc,
+					t->buffer, user_offset,
+					user_buffer + user_offset,
+					copy_size))) {
+			binder_user_error("%d:%d got transaction with invalid data ptr\n",
+					proc->pid, thread->pid);
+			return_error = BR_FAILED_REPLY;
+			return_error_param = -EFAULT;
+			return_error_line = __LINE__;
+			goto err_copy_data_failed;
+		}
+		object_size = binder_get_object(target_proc, user_buffer,
+				t->buffer, object_offset, &object);
 		if (object_size == 0 || object_offset < off_min) {
 			binder_user_error("%d:%d got transaction with invalid offset (%lld, min %lld max %lld) or object.\n",
 					  proc->pid, thread->pid,
@@ -2861,6 +2884,11 @@ static void binder_transaction(struct binder_proc *proc,
 			return_error_line = __LINE__;
 			goto err_bad_offset;
 		}
+		/*
+		 * Set offset to the next buffer fragment to be
+		 * copied
+		 */
+		user_offset = object_offset + object_size;
 
 		hdr = &object.hdr;
 		off_min = object_offset + object_size;
@@ -2956,9 +2984,14 @@ static void binder_transaction(struct binder_proc *proc,
 			}
 			ret = binder_translate_fd_array(fda, parent, t, thread,
 							in_reply_to);
-			if (ret < 0) {
+			if (!ret)
+				ret = binder_alloc_copy_to_buffer(&target_proc->alloc,
+								  t->buffer,
+								  object_offset,
+								  fda, sizeof(*fda));
+			if (ret) {
 				return_error = BR_FAILED_REPLY;
-				return_error_param = ret;
+				return_error_param = ret > 0 ? -EINVAL : ret;
 				return_error_line = __LINE__;
 				goto err_translate_failed;
 			}
@@ -3028,6 +3061,19 @@ static void binder_transaction(struct binder_proc *proc,
 			goto err_bad_object_type;
 		}
 	}
+	/* Done processing objects, copy the rest of the buffer */
+	if (binder_alloc_copy_user_to_buffer(
+				&target_proc->alloc,
+				t->buffer, user_offset,
+				user_buffer + user_offset,
+				tr->data_size - user_offset)) {
+		binder_user_error("%d:%d got transaction with invalid data ptr\n",
+				proc->pid, thread->pid);
+		return_error = BR_FAILED_REPLY;
+		return_error_param = -EFAULT;
+		return_error_line = __LINE__;
+		goto err_copy_data_failed;
+	}
 	if (t->buffer->oneway_spam_suspect)
 		tcomplete->type = BINDER_WORK_TRANSACTION_ONEWAY_SPAM_SUSPECT;
 	else
Signature · 1 line
2.34.1
author maintainer
#473 lore raw

PATCH 473/1039 openrisc: Add clone3 ABI wrapper

In reply to Greg Kroah-Hartman

From: Stafford Horne <[email protected]>

[ Upstream commit 433fe39f674d58bc7a3e8254a5d2ffc290b7e04e ]

Like fork and clone the clone3 syscall needs a wrapper to save callee saved registers, which is required by the OpenRISC ABI. This came up after auditing code following a discussion with Rob Landley and Arnd Bergmann [0].

Tested with the clone3 kselftests and there were no issues.

[0] https://lore.kernel.org/all/[email protected]/T/#m9c0cdb2703813b9df4da04cf6b30de1f1aa89944

Fixes 07e83dfbe16c ("openrisc: Enable the clone3 syscall")
Cc Rob Landley <[email protected]>
Cc Arnd Bergmann <[email protected]>
---
 arch/openrisc/include/asm/syscalls.h | 2 ++
 arch/openrisc/kernel/entry.S         | 5 +++++
 2 files changed, 7 insertions(+)
arch/openrisc/include/asm/syscalls.h +2 -0
index 3a7eeae6f56a8..aa1c7e98722e3 100644
@@ -22,9 +22,11 @@ asmlinkage long sys_or1k_atomic(unsigned long type, unsigned long *v1,
 
 asmlinkage long __sys_clone(unsigned long clone_flags, unsigned long newsp,
 			void __user *parent_tid, void __user *child_tid, int tls);
+asmlinkage long __sys_clone3(struct clone_args __user *uargs, size_t size);
 asmlinkage long __sys_fork(void);
 
 #define sys_clone __sys_clone
+#define sys_clone3 __sys_clone3
 #define sys_fork __sys_fork
 
 #endif /* __ASM_OPENRISC_SYSCALLS_H */
arch/openrisc/kernel/entry.S +5 -0
index 59c6d3aa7081e..dc5b45e9e72b5 100644
@@ -1170,6 +1170,11 @@ ENTRY(__sys_clone)
 	l.j	_fork_save_extra_regs_and_call
 	 l.nop
 
+ENTRY(__sys_clone3)
+	l.movhi	r29,hi(sys_clone3)
+	l.j	_fork_save_extra_regs_and_call
+	 l.ori	r29,r29,lo(sys_clone3)
+
 ENTRY(__sys_fork)
 	l.movhi	r29,hi(sys_fork)
 	l.ori	r29,r29,lo(sys_fork)
Signature · 1 line
2.34.1
author maintainer
#474 lore raw

PATCH 474/1039 uio: uio_dmem_genirq: Catch the Exception

In reply to Greg Kroah-Hartman

From: Jiasheng Jiang <[email protected]>

[ Upstream commit eec91694f927d1026974444eb6a3adccd4f1cbc2 ]

The return value of dma_set_coherent_mask() is not always 0. To catch the exception in case that dma is not support the mask.

Fixes 0a0c3b5a24bd ("Add new uio device for dynamic memory allocation")
---
 drivers/uio/uio_dmem_genirq.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
drivers/uio/uio_dmem_genirq.c +5 -1
index 6b5cfa5b06733..1106f33764047 100644
@@ -188,7 +188,11 @@ static int uio_dmem_genirq_probe(struct platform_device *pdev)
 		return -ENOMEM;
 	}
 
-	dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32));
+	ret = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32));
+	if (ret) {
+		dev_err(&pdev->dev, "DMA enable failed\n");
+		return ret;
+	}
 
 	priv->uioinfo = uioinfo;
 	spin_lock_init(&priv->lock);
Signature · 1 line
2.34.1
author maintainer
#475 lore raw

PATCH 475/1039 iommu: Extend mutex lock scope in iommu_probe_device()

In reply to Greg Kroah-Hartman

From: Lu Baolu <[email protected]>

[ Upstream commit 556f99ac886635e8da15528995f06d1d7028cfca ]

Extend the scope of holding group->mutex so that it can cover the default domain check/attachment and direct mappings of reserved regions.

Cc Ashish Mhetre <[email protected]>
Fixes 211ff31b3d33b ("iommu: Fix race condition during default domain allocation")
---
 drivers/iommu/iommu.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/iommu/iommu.c +2 -1
index dd7863e453a5f..8b86406b71627 100644
@@ -288,11 +288,11 @@ int iommu_probe_device(struct device *dev)
 	 */
 	mutex_lock(&group->mutex);
 	iommu_alloc_default_domain(group, dev);
-	mutex_unlock(&group->mutex);
 
 	if (group->default_domain) {
 		ret = __iommu_attach_device(group->default_domain, dev);
 		if (ret) {
+			mutex_unlock(&group->mutex);
 			iommu_group_put(group);
 			goto err_release;
 		}
@@ -300,6 +300,7 @@ int iommu_probe_device(struct device *dev)
 
 	iommu_create_device_direct_mappings(group, dev);
 
+	mutex_unlock(&group->mutex);
 	iommu_group_put(group);
 
 	if (ops->probe_finalize)
Signature · 1 line
2.34.1
author maintainer
#476 lore raw

PATCH 476/1039 iommu/io-pgtable-arm: Fix table descriptor paddr formatting

In reply to Greg Kroah-Hartman

From: Hector Martin <[email protected]>

[ Upstream commit 9abe2ac834851a7d0b0756e295cf7a292c45ca53 ]

Table descriptors were being installed without properly formatting the address using paddr_to_iopte, which does not match up with the iopte_deref in __arm_lpae_map. This is incorrect for the LPAE pte format, as it does not handle the high bits properly.

This was found on Apple T6000 DARTs, which require a new pte format (different shift); adding support for that to paddr_to_iopte/iopte_to_paddr caused it to break badly, as even <48-bit addresses would end up incorrect in that case.

Fixes 6c89928ff7a0 ("iommu/io-pgtable-arm: Support 52-bit physical address")
---
 drivers/iommu/io-pgtable-arm.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)
drivers/iommu/io-pgtable-arm.c +5 -4
index dd9e47189d0d9..94ff319ae8acc 100644
@@ -315,11 +315,12 @@ static int arm_lpae_init_pte(struct arm_lpae_io_pgtable *data,
 static arm_lpae_iopte arm_lpae_install_table(arm_lpae_iopte *table,
 					     arm_lpae_iopte *ptep,
 					     arm_lpae_iopte curr,
-					     struct io_pgtable_cfg *cfg)
+					     struct arm_lpae_io_pgtable *data)
 {
 	arm_lpae_iopte old, new;
+	struct io_pgtable_cfg *cfg = &data->iop.cfg;
 
-	new = __pa(table) | ARM_LPAE_PTE_TYPE_TABLE;
+	new = paddr_to_iopte(__pa(table), data) | ARM_LPAE_PTE_TYPE_TABLE;
 	if (cfg->quirks & IO_PGTABLE_QUIRK_ARM_NS)
 		new |= ARM_LPAE_PTE_NSTABLE;
 
@@ -380,7 +381,7 @@ static int __arm_lpae_map(struct arm_lpae_io_pgtable *data, unsigned long iova,
 		if (!cptep)
 			return -ENOMEM;
 
-		pte = arm_lpae_install_table(cptep, ptep, 0, cfg);
+		pte = arm_lpae_install_table(cptep, ptep, 0, data);
 		if (pte)
 			__arm_lpae_free_pages(cptep, tblsz, cfg);
 	} else if (!cfg->coherent_walk && !(pte & ARM_LPAE_PTE_SW_SYNC)) {
@@ -592,7 +593,7 @@ static size_t arm_lpae_split_blk_unmap(struct arm_lpae_io_pgtable *data,
 		__arm_lpae_init_pte(data, blk_paddr, pte, lvl, 1, &tablep[i]);
 	}
 
-	pte = arm_lpae_install_table(tablep, ptep, blk_pte, cfg);
+	pte = arm_lpae_install_table(tablep, ptep, blk_pte, data);
 	if (pte != blk_pte) {
 		__arm_lpae_free_pages(tablep, tablesz, cfg);
 		/*
Signature · 1 line
2.34.1
author maintainer
#477 lore raw

PATCH 477/1039 ASoC: SOF: Intel: fix build issue related to CODEC_PROBE_ENTRIES

In reply to Greg Kroah-Hartman

From: Kai Vehmanen <[email protected]>

[ Upstream commit 9a83dfcc5ae8230fbf12b63e281d5bb8450ec0e7 ]

Fix following error:
sound/soc/sof/intel/hda-codec.c:132:35: error: use of undeclared identifier 'CODEC_PROBE_RETRIES'

Found with config: i386-randconfig-r033-20211202
(https://download.01.org/0day-ci/archive/20211203/[email protected]/config)

---
 sound/soc/sof/intel/hda-codec.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
sound/soc/sof/intel/hda-codec.c +2 -1
index 13cd96e6724a4..2f3f4a733d9e6 100644
@@ -20,9 +20,10 @@
 #include "../../codecs/hdac_hda.h"
 #endif /* CONFIG_SND_SOC_SOF_HDA_AUDIO_CODEC */
 
+#define CODEC_PROBE_RETRIES	3
+
 #if IS_ENABLED(CONFIG_SND_SOC_SOF_HDA_AUDIO_CODEC)
 #define IDISP_VID_INTEL	0x80860000
-#define CODEC_PROBE_RETRIES 3
 
 /* load the legacy HDA codec driver */
 static int request_codec_module(struct hda_codec *codec)
Signature · 1 line
2.34.1
author maintainer
#478 lore raw

PATCH 478/1039 scsi: core: Fix scsi_device_max_queue_depth()

In reply to Greg Kroah-Hartman

From: Bart Van Assche <[email protected]>

[ Upstream commit 4bc3bffc1a885eb5cb259e4a25146a4c7b1034e3 ]

The comment above scsi_device_max_queue_depth() and also the description of commit ca4453213951 ("scsi: core: Make sure sdev->queue_depth is <= max(shost->can_queue, 1024)") contradict the implementation of the function scsi_device_max_queue_depth(). Additionally, the maximum queue depth of a SCSI LUN never exceeds host->can_queue. Fix scsi_device_max_queue_depth() by changing max_t() into min_t().

Fixes ca4453213951 ("scsi: core: Make sure sdev->queue_depth is <= max(shost->can_queue, 1024)")
Cc Hannes Reinecke <[email protected]>
Cc Sumanesh Samanta <[email protected]>
---
 drivers/scsi/scsi.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/scsi/scsi.c +2 -2
index f6af1562cba49..10e5bffc34aaf 100644
@@ -201,11 +201,11 @@ void scsi_finish_command(struct scsi_cmnd *cmd)
 
 
 /*
- * 1024 is big enough for saturating the fast scsi LUN now
+ * 1024 is big enough for saturating fast SCSI LUNs.
  */
 int scsi_device_max_queue_depth(struct scsi_device *sdev)
 {
-	return max_t(int, sdev->host->can_queue, 1024);
+	return min_t(int, sdev->host->can_queue, 1024);
 }
 
 /**
Signature · 1 line
2.34.1
author maintainer
#479 lore raw

PATCH 479/1039 scsi: ufs: Fix race conditions related to driver data

In reply to Greg Kroah-Hartman

From: Bart Van Assche <[email protected]>

[ Upstream commit 21ad0e49085deb22c094f91f9da57319a97188e4 ]

The driver data pointer must be set before any callbacks are registered that use that pointer. Hence move the initialization of that pointer from after the ufshcd_init() call to inside ufshcd_init().

---
 drivers/scsi/ufs/tc-dwc-g210-pci.c | 1 -
 drivers/scsi/ufs/ufshcd-pci.c      | 2 --
 drivers/scsi/ufs/ufshcd-pltfrm.c   | 2 --
 drivers/scsi/ufs/ufshcd.c          | 7 +++++++
 4 files changed, 7 insertions(+), 5 deletions(-)
drivers/scsi/ufs/tc-dwc-g210-pci.c +0 -1
index 679289e1a78e6..7b08e2e07cc5f 100644
@@ -110,7 +110,6 @@ tc_dwc_g210_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 		return err;
 	}
 
-	pci_set_drvdata(pdev, hba);
 	pm_runtime_put_noidle(&pdev->dev);
 	pm_runtime_allow(&pdev->dev);
 
drivers/scsi/ufs/ufshcd-pci.c +0 -2
index f725248ba57f4..f76692053ca17 100644
@@ -538,8 +538,6 @@ ufshcd_pci_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 		return err;
 	}
 
-	pci_set_drvdata(pdev, hba);
-
 	hba->vops = (struct ufs_hba_variant_ops *)id->driver_data;
 
 	err = ufshcd_init(hba, mmio_base, pdev->irq);
drivers/scsi/ufs/ufshcd-pltfrm.c +0 -2
index eaeae83b999fd..8b16bbbcb806c 100644
@@ -361,8 +361,6 @@ int ufshcd_pltfrm_init(struct platform_device *pdev,
 		goto dealloc_host;
 	}
 
-	platform_set_drvdata(pdev, hba);
-
 	pm_runtime_set_active(&pdev->dev);
 	pm_runtime_enable(&pdev->dev);
 
drivers/scsi/ufs/ufshcd.c +7 -0
index 13c09dbd99b92..4394081806978 100644
@@ -9486,6 +9486,13 @@ int ufshcd_init(struct ufs_hba *hba, void __iomem *mmio_base, unsigned int irq)
 	struct device *dev = hba->dev;
 	char eh_wq_name[sizeof("ufs_eh_wq_00")];
 
+	/*
+	 * dev_set_drvdata() must be called before any callbacks are registered
+	 * that use dev_get_drvdata() (frequency scaling, clock scaling, hwmon,
+	 * sysfs).
+	 */
+	dev_set_drvdata(dev, hba);
+
 	if (!mmio_base) {
 		dev_err(hba->dev,
 		"Invalid memory reference for mmio_base is NULL\n");
Signature · 1 line
2.34.1
author maintainer
#480 lore raw

PATCH 480/1039 RDMA/qedr: Fix reporting max_{send/recv}_wr attrs

In reply to Greg Kroah-Hartman

From: Kamal Heib <[email protected]>

[ Upstream commit b1a4da64bfc189510e08df1ccb1c589e667dc7a3 ]

Fix the wrongly reported max_send_wr and max_recv_wr attributes for user QP by making sure to save their valuse on QP creation, so when query QP is called the attributes will be reported correctly.

---
 drivers/infiniband/hw/qedr/verbs.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/infiniband/hw/qedr/verbs.c +2 -0
index 9100009f0a23d..a53476653b0d9 100644
@@ -1931,6 +1931,7 @@ static int qedr_create_user_qp(struct qedr_dev *dev,
 	/* db offset was calculated in copy_qp_uresp, now set in the user q */
 	if (qedr_qp_has_sq(qp)) {
 		qp->usq.db_addr = ctx->dpi_addr + uresp.sq_db_offset;
+		qp->sq.max_wr = attrs->cap.max_send_wr;
 		rc = qedr_db_recovery_add(dev, qp->usq.db_addr,
 					  &qp->usq.db_rec_data->db_data,
 					  DB_REC_WIDTH_32B,
@@ -1941,6 +1942,7 @@ static int qedr_create_user_qp(struct qedr_dev *dev,
 
 	if (qedr_qp_has_rq(qp)) {
 		qp->urq.db_addr = ctx->dpi_addr + uresp.rq_db_offset;
+		qp->rq.max_wr = attrs->cap.max_recv_wr;
 		rc = qedr_db_recovery_add(dev, qp->urq.db_addr,
 					  &qp->urq.db_rec_data->db_data,
 					  DB_REC_WIDTH_32B,
Signature · 1 line
2.34.1
author maintainer
#481 lore raw

PATCH 481/1039 PCI/MSI: Fix pci_irq_vector()/pci_irq_get_affinity()

In reply to Greg Kroah-Hartman

From: Thomas Gleixner <[email protected]>

[ Upstream commit 29bbc35e29d9b6347780dcacde2deb4b39344167 ]

pci_irq_vector() and pci_irq_get_affinity() use the list position to find the MSI-X descriptor at a given index. That's correct for the normal case where the entry number is the same as the list position.

But it's wrong for cases where MSI-X was allocated with an entries array describing sparse entry numbers into the hardware message descriptor table. That's inconsistent at best.

Make it always check the entry number because that's what the zero base index really means. This change won't break existing users which use a sparse entries array for allocation because these users retrieve the Linux interrupt number from the entries array after allocation and none of them uses pci_irq_vector() or pci_irq_get_affinity().

---
 drivers/pci/msi.c | 26 ++++++++++++++++++--------
 1 file changed, 18 insertions(+), 8 deletions(-)
drivers/pci/msi.c +18 -8
index d84cf30bb2790..8465221be6d21 100644
@@ -1194,19 +1194,24 @@ EXPORT_SYMBOL(pci_free_irq_vectors);
 
 /**
  * pci_irq_vector - return Linux IRQ number of a device vector
- * @dev: PCI device to operate on
- * @nr: device-relative interrupt vector index (0-based).
+ * @dev:	PCI device to operate on
+ * @nr:		Interrupt vector index (0-based)
+ *
+ * @nr has the following meanings depending on the interrupt mode:
+ *   MSI-X:	The index in the MSI-X vector table
+ *   MSI:	The index of the enabled MSI vectors
+ *   INTx:	Must be 0
+ *
+ * Return: The Linux interrupt number or -EINVAl if @nr is out of range.
  */
 int pci_irq_vector(struct pci_dev *dev, unsigned int nr)
 {
 	if (dev->msix_enabled) {
 		struct msi_desc *entry;
-		int i = 0;
 
 		for_each_pci_msi_entry(entry, dev) {
-			if (i == nr)
+			if (entry->msi_attrib.entry_nr == nr)
 				return entry->irq;
-			i++;
 		}
 		WARN_ON_ONCE(1);
 		return -EINVAL;
@@ -1230,17 +1235,22 @@ EXPORT_SYMBOL(pci_irq_vector);
  * pci_irq_get_affinity - return the affinity of a particular MSI vector
  * @dev:	PCI device to operate on
  * @nr:		device-relative interrupt vector index (0-based).
+ *
+ * @nr has the following meanings depending on the interrupt mode:
+ *   MSI-X:	The index in the MSI-X vector table
+ *   MSI:	The index of the enabled MSI vectors
+ *   INTx:	Must be 0
+ *
+ * Return: A cpumask pointer or NULL if @nr is out of range
  */
 const struct cpumask *pci_irq_get_affinity(struct pci_dev *dev, int nr)
 {
 	if (dev->msix_enabled) {
 		struct msi_desc *entry;
-		int i = 0;
 
 		for_each_pci_msi_entry(entry, dev) {
-			if (i == nr)
+			if (entry->msi_attrib.entry_nr == nr)
 				return &entry->affinity->mask;
-			i++;
 		}
 		WARN_ON_ONCE(1);
 		return NULL;
Signature · 1 line
2.34.1
author maintainer
#482 lore raw

PATCH 482/1039 powerpc/powermac: Add additional missing lockdep_register_key()

In reply to Greg Kroah-Hartman

From: Christophe Leroy <[email protected]>

[ Upstream commit b149d5d45ac9171ed699a256f026c8ebef901112 ]

Commit df1f679d19ed ("powerpc/powermac: Add missing lockdep_register_key()") fixed a problem that was causing a WARNING.

There are two other places in the same file with the same problem originating from commit 9e607f72748d ("i2c_powermac: shut up lockdep warning").

Add missing lockdep_register_key()

Fixes 9e607f72748d ("i2c_powermac: shut up lockdep warning")

Depends-on: df1f679d19ed ("powerpc/powermac: Add missing lockdep_register_key()")

---
 arch/powerpc/platforms/powermac/low_i2c.c | 2 ++
 1 file changed, 2 insertions(+)
arch/powerpc/platforms/powermac/low_i2c.c +2 -0
index f77a59b5c2e1a..09bfe4b8f25aa 100644
@@ -810,6 +810,7 @@ static void __init pmu_i2c_probe(void)
 		bus->hostdata = bus + 1;
 		bus->xfer = pmu_i2c_xfer;
 		mutex_init(&bus->mutex);
+		lockdep_register_key(&bus->lock_key);
 		lockdep_set_class(&bus->mutex, &bus->lock_key);
 		bus->flags = pmac_i2c_multibus;
 		list_add(&bus->link, &pmac_i2c_busses);
@@ -933,6 +934,7 @@ static void __init smu_i2c_probe(void)
 		bus->hostdata = bus + 1;
 		bus->xfer = smu_i2c_xfer;
 		mutex_init(&bus->mutex);
+		lockdep_register_key(&bus->lock_key);
 		lockdep_set_class(&bus->mutex, &bus->lock_key);
 		bus->flags = 0;
 		list_add(&bus->link, &pmac_i2c_busses);
Signature · 1 line
2.34.1
author maintainer
#483 lore raw

PATCH 483/1039 iommu/arm-smmu-qcom: Fix TTBR0 read

In reply to Greg Kroah-Hartman

From: Rob Clark <[email protected]>

[ Upstream commit c31112fbd4077a51a14ff338038c82e9571dc821 ]

It is a 64b register, lets not lose the upper bits.

Fixes ab5df7b953d8 ("iommu/arm-smmu-qcom: Add an adreno-smmu-priv callback to get pagefault info")
---
 drivers/iommu/arm/arm-smmu/arm-smmu-qcom.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/iommu/arm/arm-smmu/arm-smmu-qcom.c +1 -1
index ca736b065dd0b..40c91dd368a4d 100644
@@ -51,7 +51,7 @@ static void qcom_adreno_smmu_get_fault_info(const void *cookie,
 	info->fsynr1 = arm_smmu_cb_read(smmu, cfg->cbndx, ARM_SMMU_CB_FSYNR1);
 	info->far = arm_smmu_cb_readq(smmu, cfg->cbndx, ARM_SMMU_CB_FAR);
 	info->cbfrsynra = arm_smmu_gr1_read(smmu, ARM_SMMU_GR1_CBFRSYNRA(cfg->cbndx));
-	info->ttbr0 = arm_smmu_cb_read(smmu, cfg->cbndx, ARM_SMMU_CB_TTBR0);
+	info->ttbr0 = arm_smmu_cb_readq(smmu, cfg->cbndx, ARM_SMMU_CB_TTBR0);
 	info->contextidr = arm_smmu_cb_read(smmu, cfg->cbndx, ARM_SMMU_CB_CONTEXTIDR);
 }
 
Signature · 1 line
2.34.1
author maintainer
#484 lore raw

PATCH 484/1039 RDMA/core: Let ib_find_gid() continue search even after empty entry

In reply to Greg Kroah-Hartman

From: Avihai Horon <[email protected]>

[ Upstream commit 483d805191a23191f8294bbf9b4e94836f5d92e4 ]

Currently, ib_find_gid() will stop searching after encountering the first empty GID table entry. This behavior is wrong since neither IB nor RoCE spec enforce tightly packed GID tables.

For example, when a valid GID entry exists at index N, and if a GID entry is empty at index N-1, ib_find_gid() will fail to find the valid entry.

Fix it by making ib_find_gid() continue searching even after encountering missing entries.

---
 drivers/infiniband/core/device.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/infiniband/core/device.c +2 -1
index 22a4adda7981d..a311df07b1bdb 100644
@@ -2461,7 +2461,8 @@ int ib_find_gid(struct ib_device *device, union ib_gid *gid,
 		     ++i) {
 			ret = rdma_query_gid(device, port, i, &tmp_gid);
 			if (ret)
-				return ret;
+				continue;
+
 			if (!memcmp(&tmp_gid, gid, sizeof *gid)) {
 				*port_num = port;
 				if (index)
Signature · 1 line
2.34.1
author maintainer
#485 lore raw

PATCH 485/1039 RDMA/cma: Let cma_resolve_ib_dev() continue search even after empty entry

In reply to Greg Kroah-Hartman

From: Avihai Horon <[email protected]>

[ Upstream commit 20679094a0161c94faf77e373fa3f7428a8e14bd ]

Currently, when cma_resolve_ib_dev() searches for a matching GID it will stop searching after encountering the first empty GID table entry. This behavior is wrong since neither IB nor RoCE spec enforce tightly packed GID tables.

For example, when the matching valid GID entry exists at index N, and if a GID entry is empty at index N-1, cma_resolve_ib_dev() will fail to find the matching valid entry.

Fix it by making cma_resolve_ib_dev() continue searching even after encountering missing entries.

---
 drivers/infiniband/core/cma.c | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)
drivers/infiniband/core/cma.c +9 -3
index 835ac54d4a24c..a3834ef691910 100644
@@ -766,6 +766,7 @@ static int cma_resolve_ib_dev(struct rdma_id_private *id_priv)
 	unsigned int p;
 	u16 pkey, index;
 	enum ib_port_state port_state;
+	int ret;
 	int i;
 
 	cma_dev = NULL;
@@ -784,9 +785,14 @@ static int cma_resolve_ib_dev(struct rdma_id_private *id_priv)
 
 			if (ib_get_cached_port_state(cur_dev->device, p, &port_state))
 				continue;
-			for (i = 0; !rdma_query_gid(cur_dev->device,
-						    p, i, &gid);
-			     i++) {
+
+			for (i = 0; i < cur_dev->device->port_data[p].immutable.gid_tbl_len;
+			     ++i) {
+				ret = rdma_query_gid(cur_dev->device, p, i,
+						     &gid);
+				if (ret)
+					continue;
+
 				if (!memcmp(&gid, dgid, sizeof(gid))) {
 					cma_dev = cur_dev;
 					sgid = gid;
Signature · 1 line
2.34.1
author maintainer
#486 lore raw

PATCH 486/1039 ASoC: rt5663: Handle device_property_read_u32_array error codes

In reply to Greg Kroah-Hartman

From: Jiasheng Jiang <[email protected]>

[ Upstream commit 2167c0b205960607fb136b4bb3c556a62be1569a ]

The return value of device_property_read_u32_array() is not always 0. To catch the exception in case that devm_kzalloc failed and the rt5663->imp_table was NULL, which caused the failure of device_property_read_u32_array.

Fixes 450f0f6a8fb4 ("ASoC: rt5663: Add the manual offset field to compensate the DC offset")
---
 sound/soc/codecs/rt5663.c | 12 +++++++++---
 1 file changed, 9 insertions(+), 3 deletions(-)
sound/soc/codecs/rt5663.c +9 -3
index 0389b2bb360e2..2138f62e6af5d 100644
@@ -3461,6 +3461,7 @@ static void rt5663_calibrate(struct rt5663_priv *rt5663)
 static int rt5663_parse_dp(struct rt5663_priv *rt5663, struct device *dev)
 {
 	int table_size;
+	int ret;
 
 	device_property_read_u32(dev, "realtek,dc_offset_l_manual",
 		&rt5663->pdata.dc_offset_l_manual);
@@ -3477,9 +3478,11 @@ static int rt5663_parse_dp(struct rt5663_priv *rt5663, struct device *dev)
 		table_size = sizeof(struct impedance_mapping_table) *
 			rt5663->pdata.impedance_sensing_num;
 		rt5663->imp_table = devm_kzalloc(dev, table_size, GFP_KERNEL);
-		device_property_read_u32_array(dev,
+		ret = device_property_read_u32_array(dev,
 			"realtek,impedance_sensing_table",
 			(u32 *)rt5663->imp_table, table_size);
+		if (ret)
+			return ret;
 	}
 
 	return 0;
@@ -3504,8 +3507,11 @@ static int rt5663_i2c_probe(struct i2c_client *i2c,
 
 	if (pdata)
 		rt5663->pdata = *pdata;
-	else
-		rt5663_parse_dp(rt5663, &i2c->dev);
+	else {
+		ret = rt5663_parse_dp(rt5663, &i2c->dev);
+		if (ret)
+			return ret;
+	}
 
 	for (i = 0; i < ARRAY_SIZE(rt5663->supplies); i++)
 		rt5663->supplies[i].supply = rt5663_supply_names[i];
Signature · 1 line
2.34.1
author maintainer
#487 lore raw

PATCH 487/1039 of: unittest: fix warning on PowerPC frame size warning

In reply to Greg Kroah-Hartman

From: Jim Quinlan <[email protected]>

[ Upstream commit a8d61a9112ad0c9216ab45d050991e07bc4f3408 ]

The struct device variable "dev_bogus" was triggering this warning on a PowerPC build:

drivers/of/unittest.c: In function 'of_unittest_dma_ranges_one.constprop':
[...] >> The frame size of 1424 bytes is larger than 1024 bytes
[-Wframe-larger-than=]

This variable is now dynamically allocated.

---
 drivers/of/unittest.c | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)
drivers/of/unittest.c +12 -4
index 481ba8682ebf4..02c5cd06ad19c 100644
@@ -911,11 +911,18 @@ static void __init of_unittest_dma_ranges_one(const char *path,
 	if (!rc) {
 		phys_addr_t	paddr;
 		dma_addr_t	dma_addr;
-		struct device	dev_bogus;
+		struct device	*dev_bogus;
 
-		dev_bogus.dma_range_map = map;
-		paddr = dma_to_phys(&dev_bogus, expect_dma_addr);
-		dma_addr = phys_to_dma(&dev_bogus, expect_paddr);
+		dev_bogus = kzalloc(sizeof(struct device), GFP_KERNEL);
+		if (!dev_bogus) {
+			unittest(0, "kzalloc() failed\n");
+			kfree(map);
+			return;
+		}
+
+		dev_bogus->dma_range_map = map;
+		paddr = dma_to_phys(dev_bogus, expect_dma_addr);
+		dma_addr = phys_to_dma(dev_bogus, expect_paddr);
 
 		unittest(paddr == expect_paddr,
 			 "of_dma_get_range: wrong phys addr %pap (expecting %llx) on node %pOF\n",
@@ -925,6 +932,7 @@ static void __init of_unittest_dma_ranges_one(const char *path,
 			 &dma_addr, expect_dma_addr, np);
 
 		kfree(map);
+		kfree(dev_bogus);
 	}
 	of_node_put(np);
 #endif
Signature · 1 line
2.34.1
author maintainer
#488 lore raw

PATCH 488/1039 of: unittest: 64 bit dma address test requires arch support

In reply to Greg Kroah-Hartman

From: Frank Rowand <[email protected]>

[ Upstream commit 9fd4cf5d3571b27d746b8ead494a3f051485b679 ]

If an architecture does not support 64 bit dma addresses then testing for an expected dma address >= 0x100000000 will fail.

Fixes e0d072782c73 ("dma-mapping: introduce DMA range map, supplanting dma_pfn_offset")
---
 drivers/of/unittest.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/of/unittest.c +3 -2
index 02c5cd06ad19c..35af4fedc15de 100644
@@ -942,8 +942,9 @@ static void __init of_unittest_parse_dma_ranges(void)
 {
 	of_unittest_dma_ranges_one("/testcase-data/address-tests/device@70000000",
 		0x0, 0x20000000);
-	of_unittest_dma_ranges_one("/testcase-data/address-tests/bus@80000000/device@1000",
-		0x100000000, 0x20000000);
+	if (IS_ENABLED(CONFIG_ARCH_DMA_ADDR_T_64BIT))
+		of_unittest_dma_ranges_one("/testcase-data/address-tests/bus@80000000/device@1000",
+			0x100000000, 0x20000000);
 	of_unittest_dma_ranges_one("/testcase-data/address-tests/pci@90000000",
 		0x80000000, 0x20000000);
 }
Signature · 1 line
2.34.1
author maintainer
#489 lore raw

PATCH 489/1039 clk: stm32: Fix ltdcs clock turn off by clk_disable_unused() after system enter shell

In reply to Greg Kroah-Hartman

From: Dillon Min <[email protected]>

[ Upstream commit 6fc058a72f3b7b07fc4de6d66ad1f68951b00f6e ]

stm32's clk driver register two ltdc gate clk to clk core by clk_hw_register_gate() and clk_hw_register_composite()

first: 'stm32f429_gates[]', clk name is 'ltdc', which no user to use.
second: 'stm32f429_aux_clk[]', clk name is 'lcd-tft', used by ltdc driver

both of them point to the same offset of stm32's RCC register. after kernel enter console, clk core turn off ltdc's clk as 'stm32f429_gates[]' is no one to use. but, actually 'stm32f429_aux_clk[]' is in use.

stm32f469/746/769 have the same issue, fix it.

---
 drivers/clk/clk-stm32f4.c | 4 ----
 1 file changed, 4 deletions(-)
drivers/clk/clk-stm32f4.c +0 -4
index af46176ad0539..473dfe632cc57 100644
@@ -129,7 +129,6 @@ static const struct stm32f4_gate_data stm32f429_gates[] __initconst = {
 	{ STM32F4_RCC_APB2ENR, 20,	"spi5",		"apb2_div" },
 	{ STM32F4_RCC_APB2ENR, 21,	"spi6",		"apb2_div" },
 	{ STM32F4_RCC_APB2ENR, 22,	"sai1",		"apb2_div" },
-	{ STM32F4_RCC_APB2ENR, 26,	"ltdc",		"apb2_div" },
 };
 
 static const struct stm32f4_gate_data stm32f469_gates[] __initconst = {
@@ -211,7 +210,6 @@ static const struct stm32f4_gate_data stm32f469_gates[] __initconst = {
 	{ STM32F4_RCC_APB2ENR, 20,	"spi5",		"apb2_div" },
 	{ STM32F4_RCC_APB2ENR, 21,	"spi6",		"apb2_div" },
 	{ STM32F4_RCC_APB2ENR, 22,	"sai1",		"apb2_div" },
-	{ STM32F4_RCC_APB2ENR, 26,	"ltdc",		"apb2_div" },
 };
 
 static const struct stm32f4_gate_data stm32f746_gates[] __initconst = {
@@ -286,7 +284,6 @@ static const struct stm32f4_gate_data stm32f746_gates[] __initconst = {
 	{ STM32F4_RCC_APB2ENR, 21,	"spi6",		"apb2_div" },
 	{ STM32F4_RCC_APB2ENR, 22,	"sai1",		"apb2_div" },
 	{ STM32F4_RCC_APB2ENR, 23,	"sai2",		"apb2_div" },
-	{ STM32F4_RCC_APB2ENR, 26,	"ltdc",		"apb2_div" },
 };
 
 static const struct stm32f4_gate_data stm32f769_gates[] __initconst = {
@@ -364,7 +361,6 @@ static const struct stm32f4_gate_data stm32f769_gates[] __initconst = {
 	{ STM32F4_RCC_APB2ENR, 21,	"spi6",		"apb2_div" },
 	{ STM32F4_RCC_APB2ENR, 22,	"sai1",		"apb2_div" },
 	{ STM32F4_RCC_APB2ENR, 23,	"sai2",		"apb2_div" },
-	{ STM32F4_RCC_APB2ENR, 26,	"ltdc",		"apb2_div" },
 	{ STM32F4_RCC_APB2ENR, 30,	"mdio",		"apb2_div" },
 };
 
Signature · 1 line
2.34.1
author maintainer
#490 lore raw

PATCH 490/1039 mips: add SYS_HAS_CPU_MIPS64_R5 config for MIPS Release 5 support

In reply to Greg Kroah-Hartman

From: Lukas Bulwahn <[email protected]>

[ Upstream commit fd4eb90b164442cb1e9909f7845e12a0835ac699 ]

Commit ab7c01fdc3cf ("mips: Add MIPS Release 5 support") adds the two configs CPU_MIPS32_R5 and CPU_MIPS64_R5, which depend on the corresponding SYS_HAS_CPU_MIPS32_R5 and SYS_HAS_CPU_MIPS64_R5, respectively.

The config SYS_HAS_CPU_MIPS32_R5 was already introduced with commit c5b367835cfc ("MIPS: Add support for XPA."); the config SYS_HAS_CPU_MIPS64_R5, however, was never introduced.

Hence, ./scripts/checkkconfigsymbols.py warns:

SYS_HAS_CPU_MIPS64_R5
Referencing files: arch/mips/Kconfig, arch/mips/include/asm/cpu-type.h

Add the definition for config SYS_HAS_CPU_MIPS64_R5 under the assumption that SYS_HAS_CPU_MIPS64_R5 follows the same pattern as the existing SYS_HAS_CPU_MIPS32_R5 and SYS_HAS_CPU_MIPS64_R6.

Fixes ab7c01fdc3cf ("mips: Add MIPS Release 5 support")
---
 arch/mips/Kconfig | 4 ++++
 1 file changed, 4 insertions(+)
arch/mips/Kconfig +4 -0
index 0215dc1529e9a..91ce1c57af005 100644
@@ -1907,6 +1907,10 @@ config SYS_HAS_CPU_MIPS64_R1
 config SYS_HAS_CPU_MIPS64_R2
 	bool
 
+config SYS_HAS_CPU_MIPS64_R5
+	bool
+	select ARCH_HAS_SYNC_DMA_FOR_CPU if DMA_NONCOHERENT
+
 config SYS_HAS_CPU_MIPS64_R6
 	bool
 	select ARCH_HAS_SYNC_DMA_FOR_CPU if DMA_NONCOHERENT
Signature · 1 line
2.34.1
author maintainer
#491 lore raw

PATCH 491/1039 mips: fix Kconfig reference to PHYS_ADDR_T_64BIT

In reply to Greg Kroah-Hartman

From: Lukas Bulwahn <[email protected]>

[ Upstream commit a670c82d9ca4f1e7385d9d6f26ff41a50fbdd944 ]

Commit d4a451d5fc84 ("arch: remove the ARCH_PHYS_ADDR_T_64BIT config symbol") removes config ARCH_PHYS_ADDR_T_64BIT with all instances of that config refactored appropriately. Since then, it is recommended to use the config PHYS_ADDR_T_64BIT instead.

Commit 171543e75272 ("MIPS: Disallow CPU_SUPPORTS_HUGEPAGES for XPA,EVA") introduces the expression "!(32BIT && (ARCH_PHYS_ADDR_T_64BIT || EVA))" for config CPU_SUPPORTS_HUGEPAGES, which unintentionally refers to the non-existing symbol ARCH_PHYS_ADDR_T_64BIT instead of the intended PHYS_ADDR_T_64BIT.

Fix this Kconfig reference to the intended PHYS_ADDR_T_64BIT.

This issue was identified with the script ./scripts/checkkconfigsymbols.py. I then reported it on the mailing list and Paul confirmed the mistake in the linked email thread.

Fixes 171543e75272 ("MIPS: Disallow CPU_SUPPORTS_HUGEPAGES for XPA,EVA")
---
 arch/mips/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/mips/Kconfig +1 -1
index 91ce1c57af005..c5826236d913a 100644
@@ -2069,7 +2069,7 @@ config CPU_SUPPORTS_ADDRWINCFG
 	bool
 config CPU_SUPPORTS_HUGEPAGES
 	bool
-	depends on !(32BIT && (ARCH_PHYS_ADDR_T_64BIT || EVA))
+	depends on !(32BIT && (PHYS_ADDR_T_64BIT || EVA))
 config MIPS_PGD_C0_CONTEXT
 	bool
 	depends on 64BIT
Signature · 1 line
2.34.1
author maintainer
#492 lore raw

PATCH 492/1039 dmaengine: pxa/mmp: stop referencing config->slave_id

In reply to Greg Kroah-Hartman

From: Arnd Bergmann <[email protected]>

[ Upstream commit 134c37fa250a87a7e77c80a7c59ae16c462e46e0 ]

The last driver referencing the slave_id on Marvell PXA and MMP platforms was the SPI driver, but this stopped doing so a long time ago, so the TODO from the earlier patch can no be removed.

Fixes b729bf34535e ("spi/pxa2xx: Don't use slave_id of dma_slave_config")
Fixes 13b3006b8ebd ("dma: mmp_pdma: add filter function")
---
 drivers/dma/mmp_pdma.c | 6 ------
 drivers/dma/pxa_dma.c  | 7 -------
 2 files changed, 13 deletions(-)
drivers/dma/mmp_pdma.c +0 -6
index a23563cd118b7..5a53d7fcef018 100644
@@ -727,12 +727,6 @@ static int mmp_pdma_config_write(struct dma_chan *dchan,
 
 	chan->dir = direction;
 	chan->dev_addr = addr;
-	/* FIXME: drivers should be ported over to use the filter
-	 * function. Once that's done, the following two lines can
-	 * be removed.
-	 */
-	if (cfg->slave_id)
-		chan->drcmr = cfg->slave_id;
 
 	return 0;
 }
drivers/dma/pxa_dma.c +0 -7
index 52d04641e3611..6078cc81892e4 100644
@@ -909,13 +909,6 @@ static void pxad_get_config(struct pxad_chan *chan,
 		*dcmd |= PXA_DCMD_BURST16;
 	else if (maxburst == 32)
 		*dcmd |= PXA_DCMD_BURST32;
-
-	/* FIXME: drivers should be ported over to use the filter
-	 * function. Once that's done, the following two lines can
-	 * be removed.
-	 */
-	if (chan->cfg.slave_id)
-		chan->drcmr = chan->cfg.slave_id;
 }
 
 static struct dma_async_tx_descriptor *
Signature · 1 line
2.34.1
author maintainer
#493 lore raw

PATCH 493/1039 iommu/amd: Restore GA log/tail pointer on host resume

In reply to Greg Kroah-Hartman

From: Maxim Levitsky <[email protected]>

[ Upstream commit a8d4a37d1bb93608501d0d0545f902061152669a ]

This will give IOMMU GA log a chance to work after resume from s3/s4.

Fixes 8bda0cfbdc1a6 ("iommu/amd: Detect and initialize guest vAPIC log")
---
 drivers/iommu/amd/init.c | 31 +++++++++++++++----------------
 1 file changed, 15 insertions(+), 16 deletions(-)
drivers/iommu/amd/init.c +15 -16
index 1eacd43cb4368..8dae85fcfc2eb 100644
@@ -806,16 +806,27 @@ static int iommu_ga_log_enable(struct amd_iommu *iommu)
 {
 #ifdef CONFIG_IRQ_REMAP
 	u32 status, i;
+	u64 entry;
 
 	if (!iommu->ga_log)
 		return -EINVAL;
 
-	status = readl(iommu->mmio_base + MMIO_STATUS_OFFSET);
-
 	/* Check if already running */
-	if (status & (MMIO_STATUS_GALOG_RUN_MASK))
+	status = readl(iommu->mmio_base + MMIO_STATUS_OFFSET);
+	if (WARN_ON(status & (MMIO_STATUS_GALOG_RUN_MASK)))
 		return 0;
 
+	entry = iommu_virt_to_phys(iommu->ga_log) | GA_LOG_SIZE_512;
+	memcpy_toio(iommu->mmio_base + MMIO_GA_LOG_BASE_OFFSET,
+		    &entry, sizeof(entry));
+	entry = (iommu_virt_to_phys(iommu->ga_log_tail) &
+		 (BIT_ULL(52)-1)) & ~7ULL;
+	memcpy_toio(iommu->mmio_base + MMIO_GA_LOG_TAIL_OFFSET,
+		    &entry, sizeof(entry));
+	writel(0x00, iommu->mmio_base + MMIO_GA_HEAD_OFFSET);
+	writel(0x00, iommu->mmio_base + MMIO_GA_TAIL_OFFSET);
+
+
 	iommu_feature_enable(iommu, CONTROL_GAINT_EN);
 	iommu_feature_enable(iommu, CONTROL_GALOG_EN);
 
@@ -825,7 +836,7 @@ static int iommu_ga_log_enable(struct amd_iommu *iommu)
 			break;
 	}
 
-	if (i >= LOOP_TIMEOUT)
+	if (WARN_ON(i >= LOOP_TIMEOUT))
 		return -EINVAL;
 #endif /* CONFIG_IRQ_REMAP */
 	return 0;
@@ -834,8 +845,6 @@ static int iommu_ga_log_enable(struct amd_iommu *iommu)
 static int iommu_init_ga_log(struct amd_iommu *iommu)
 {
 #ifdef CONFIG_IRQ_REMAP
-	u64 entry;
-
 	if (!AMD_IOMMU_GUEST_IR_VAPIC(amd_iommu_guest_ir))
 		return 0;
 
@@ -849,16 +858,6 @@ static int iommu_init_ga_log(struct amd_iommu *iommu)
 	if (!iommu->ga_log_tail)
 		goto err_out;
 
-	entry = iommu_virt_to_phys(iommu->ga_log) | GA_LOG_SIZE_512;
-	memcpy_toio(iommu->mmio_base + MMIO_GA_LOG_BASE_OFFSET,
-		    &entry, sizeof(entry));
-	entry = (iommu_virt_to_phys(iommu->ga_log_tail) &
-		 (BIT_ULL(52)-1)) & ~7ULL;
-	memcpy_toio(iommu->mmio_base + MMIO_GA_LOG_TAIL_OFFSET,
-		    &entry, sizeof(entry));
-	writel(0x00, iommu->mmio_base + MMIO_GA_HEAD_OFFSET);
-	writel(0x00, iommu->mmio_base + MMIO_GA_TAIL_OFFSET);
-
 	return 0;
 err_out:
 	free_ga_log(iommu);
Signature · 1 line
2.34.1
author maintainer
#494 lore raw

PATCH 494/1039 iommu/amd: X2apic mode: re-enable after resume

In reply to Greg Kroah-Hartman

From: Maxim Levitsky <[email protected]>

[ Upstream commit 01b297a48a26bcb96769505ac948db4603b72bd1 ]

Otherwise it is guaranteed to not work after the resume...

Fixes 66929812955bb ("iommu/amd: Add support for X2APIC IOMMU interrupts")
---
 drivers/iommu/amd/init.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/iommu/amd/init.c +4 -1
index 8dae85fcfc2eb..b905604f434e1 100644
@@ -2172,7 +2172,6 @@ static int iommu_setup_intcapxt(struct amd_iommu *iommu)
 		return ret;
 	}
 
-	iommu_feature_enable(iommu, CONTROL_INTCAPXT_EN);
 	return 0;
 }
 
@@ -2195,6 +2194,10 @@ static int iommu_init_irq(struct amd_iommu *iommu)
 
 	iommu->int_enabled = true;
 enable_faults:
+
+	if (amd_iommu_xt_mode == IRQ_REMAP_X2APIC_MODE)
+		iommu_feature_enable(iommu, CONTROL_INTCAPXT_EN);
+
 	iommu_feature_enable(iommu, CONTROL_EVT_INT_EN);
 
 	if (iommu->ppr_log != NULL)
Signature · 1 line
2.34.1
author maintainer
#495 lore raw

PATCH 495/1039 iommu/amd: X2apic mode: setup the INTX registers on mask/unmask

In reply to Greg Kroah-Hartman

From: Maxim Levitsky <[email protected]>

[ Upstream commit 4691f79d62a637958f7b5f55c232a65399500b7a ]

This is more logically correct and will also allow us to to use mask/unmask logic to restore INTX setttings after the resume from s3/s4.

Fixes 66929812955bb ("iommu/amd: Add support for X2APIC IOMMU interrupts")
---
 drivers/iommu/amd/init.c | 65 ++++++++++++++++++++--------------------
 1 file changed, 33 insertions(+), 32 deletions(-)
drivers/iommu/amd/init.c +33 -32
index b905604f434e1..9e895bb8086a6 100644
@@ -2015,48 +2015,18 @@ union intcapxt {
 	};
 } __attribute__ ((packed));
 
-/*
- * There isn't really any need to mask/unmask at the irqchip level because
- * the 64-bit INTCAPXT registers can be updated atomically without tearing
- * when the affinity is being updated.
- */
-static void intcapxt_unmask_irq(struct irq_data *data)
-{
-}
-
-static void intcapxt_mask_irq(struct irq_data *data)
-{
-}
 
 static struct irq_chip intcapxt_controller;
 
 static int intcapxt_irqdomain_activate(struct irq_domain *domain,
 				       struct irq_data *irqd, bool reserve)
 {
-	struct amd_iommu *iommu = irqd->chip_data;
-	struct irq_cfg *cfg = irqd_cfg(irqd);
-	union intcapxt xt;
-
-	xt.capxt = 0ULL;
-	xt.dest_mode_logical = apic->dest_mode_logical;
-	xt.vector = cfg->vector;
-	xt.destid_0_23 = cfg->dest_apicid & GENMASK(23, 0);
-	xt.destid_24_31 = cfg->dest_apicid >> 24;
-
-	/**
-	 * Current IOMMU implemtation uses the same IRQ for all
-	 * 3 IOMMU interrupts.
-	 */
-	writeq(xt.capxt, iommu->mmio_base + MMIO_INTCAPXT_EVT_OFFSET);
-	writeq(xt.capxt, iommu->mmio_base + MMIO_INTCAPXT_PPR_OFFSET);
-	writeq(xt.capxt, iommu->mmio_base + MMIO_INTCAPXT_GALOG_OFFSET);
 	return 0;
 }
 
 static void intcapxt_irqdomain_deactivate(struct irq_domain *domain,
 					  struct irq_data *irqd)
 {
-	intcapxt_mask_irq(irqd);
 }
 
 
@@ -2090,6 +2060,38 @@ static void intcapxt_irqdomain_free(struct irq_domain *domain, unsigned int virq
 	irq_domain_free_irqs_top(domain, virq, nr_irqs);
 }
 
+
+static void intcapxt_unmask_irq(struct irq_data *irqd)
+{
+	struct amd_iommu *iommu = irqd->chip_data;
+	struct irq_cfg *cfg = irqd_cfg(irqd);
+	union intcapxt xt;
+
+	xt.capxt = 0ULL;
+	xt.dest_mode_logical = apic->dest_mode_logical;
+	xt.vector = cfg->vector;
+	xt.destid_0_23 = cfg->dest_apicid & GENMASK(23, 0);
+	xt.destid_24_31 = cfg->dest_apicid >> 24;
+
+	/**
+	 * Current IOMMU implementation uses the same IRQ for all
+	 * 3 IOMMU interrupts.
+	 */
+	writeq(xt.capxt, iommu->mmio_base + MMIO_INTCAPXT_EVT_OFFSET);
+	writeq(xt.capxt, iommu->mmio_base + MMIO_INTCAPXT_PPR_OFFSET);
+	writeq(xt.capxt, iommu->mmio_base + MMIO_INTCAPXT_GALOG_OFFSET);
+}
+
+static void intcapxt_mask_irq(struct irq_data *irqd)
+{
+	struct amd_iommu *iommu = irqd->chip_data;
+
+	writeq(0, iommu->mmio_base + MMIO_INTCAPXT_EVT_OFFSET);
+	writeq(0, iommu->mmio_base + MMIO_INTCAPXT_PPR_OFFSET);
+	writeq(0, iommu->mmio_base + MMIO_INTCAPXT_GALOG_OFFSET);
+}
+
+
 static int intcapxt_set_affinity(struct irq_data *irqd,
 				 const struct cpumask *mask, bool force)
 {
@@ -2099,8 +2101,7 @@ static int intcapxt_set_affinity(struct irq_data *irqd,
 	ret = parent->chip->irq_set_affinity(parent, mask, force);
 	if (ret < 0 || ret == IRQ_SET_MASK_OK_DONE)
 		return ret;
-
-	return intcapxt_irqdomain_activate(irqd->domain, irqd, false);
+	return 0;
 }
 
 static struct irq_chip intcapxt_controller = {
Signature · 1 line
2.34.1
author maintainer
#496 lore raw

PATCH 496/1039 iommu/amd: X2apic mode: mask/unmask interrupts on suspend/resume

In reply to Greg Kroah-Hartman

From: Maxim Levitsky <[email protected]>

[ Upstream commit 1980105e3cfc2215c75b4f6b172661d675c467d1 ]

Use IRQCHIP_MASK_ON_SUSPEND to make the core irq code to mask the iommu interrupt on suspend and unmask it on the resume.

Since now the unmask function updates the INTX settings, that will restore them on resume from s3/s4.

Since IRQCHIP_MASK_ON_SUSPEND is only effective for interrupts which are not wakeup sources, remove IRQCHIP_SKIP_SET_WAKE flag and instead implement a dummy .irq_set_wake which doesn't allow the interrupt to become a wakeup source.

Fixes 66929812955bb ("iommu/amd: Add support for X2APIC IOMMU interrupts")
---
 drivers/iommu/amd/init.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
drivers/iommu/amd/init.c +7 -1
index 9e895bb8086a6..b94822fc2c9f7 100644
@@ -2104,6 +2104,11 @@ static int intcapxt_set_affinity(struct irq_data *irqd,
 	return 0;
 }
 
+static int intcapxt_set_wake(struct irq_data *irqd, unsigned int on)
+{
+	return on ? -EOPNOTSUPP : 0;
+}
+
 static struct irq_chip intcapxt_controller = {
 	.name			= "IOMMU-MSI",
 	.irq_unmask		= intcapxt_unmask_irq,
@@ -2111,7 +2116,8 @@ static struct irq_chip intcapxt_controller = {
 	.irq_ack		= irq_chip_ack_parent,
 	.irq_retrigger		= irq_chip_retrigger_hierarchy,
 	.irq_set_affinity       = intcapxt_set_affinity,
-	.flags			= IRQCHIP_SKIP_SET_WAKE,
+	.irq_set_wake		= intcapxt_set_wake,
+	.flags			= IRQCHIP_MASK_ON_SUSPEND,
 };
 
 static const struct irq_domain_ops intcapxt_domain_ops = {
Signature · 1 line
2.34.1
author maintainer
#497 lore raw

PATCH 497/1039 iommu/amd: Remove useless irq affinity notifier

In reply to Greg Kroah-Hartman

From: Maxim Levitsky <[email protected]>

[ Upstream commit 575f5cfb13c84f324f9898383fa4a5694e53c9ef ]

iommu->intcapxt_notify field is no longer used
after a switch to a separate domain was done

Fixes d1adcfbb520c ("iommu/amd: Fix IOMMU interrupt generation in X2APIC mode")
---
 drivers/iommu/amd/amd_iommu_types.h | 2 --
 1 file changed, 2 deletions(-)
drivers/iommu/amd/amd_iommu_types.h +0 -2
index 867535eb0ce97..ffc89c4fb1205 100644
@@ -645,8 +645,6 @@ struct amd_iommu {
 	/* DebugFS Info */
 	struct dentry *debugfs;
 #endif
-	/* IRQ notifier for IntCapXT interrupt */
-	struct irq_affinity_notify intcapxt_notify;
 };
 
 static inline struct amd_iommu *dev_to_amd_iommu(struct device *dev)
Signature · 1 line
2.34.1
author maintainer
#498 lore raw

PATCH 498/1039 ASoC: Intel: catpt: Test dmaengine_submit() result before moving on

In reply to Greg Kroah-Hartman

From: Cezary Rojewski <[email protected]>

[ Upstream commit 2a9a72e290d4a4741e673f86b9fba9bfb319786d ]

After calling dmaengine_submit(), the submitted transfer descriptor belongs to the DMA engine. Pointer to that descriptor may no longer be valid after the call and should be tested before awaiting transfer completion.

---
 sound/soc/intel/catpt/dsp.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)
sound/soc/intel/catpt/dsp.c +12 -2
index 9c5fd18f2600f..346bec0003066 100644
@@ -65,6 +65,7 @@ static int catpt_dma_memcpy(struct catpt_dev *cdev, struct dma_chan *chan,
 {
 	struct dma_async_tx_descriptor *desc;
 	enum dma_status status;
+	int ret;
 
 	desc = dmaengine_prep_dma_memcpy(chan, dst_addr, src_addr, size,
 					 DMA_CTRL_ACK);
@@ -77,13 +78,22 @@ static int catpt_dma_memcpy(struct catpt_dev *cdev, struct dma_chan *chan,
 	catpt_updatel_shim(cdev, HMDC,
 			   CATPT_HMDC_HDDA(CATPT_DMA_DEVID, chan->chan_id),
 			   CATPT_HMDC_HDDA(CATPT_DMA_DEVID, chan->chan_id));
-	dmaengine_submit(desc);
+
+	ret = dma_submit_error(dmaengine_submit(desc));
+	if (ret) {
+		dev_err(cdev->dev, "submit tx failed: %d\n", ret);
+		goto clear_hdda;
+	}
+
 	status = dma_wait_for_async_tx(desc);
+	ret = (status == DMA_COMPLETE) ? 0 : -EPROTO;
+
+clear_hdda:
 	/* regardless of status, disable access to HOST memory in demand mode */
 	catpt_updatel_shim(cdev, HMDC,
 			   CATPT_HMDC_HDDA(CATPT_DMA_DEVID, chan->chan_id), 0);
 
-	return (status == DMA_COMPLETE) ? 0 : -EPROTO;
+	return ret;
 }
 
 int catpt_dma_memcpy_todsp(struct catpt_dev *cdev, struct dma_chan *chan,
Signature · 1 line
2.34.1
author maintainer
#499 lore raw

PATCH 499/1039 iommu/iova: Fix race between FQ timeout and teardown

In reply to Greg Kroah-Hartman

From: Xiongfeng Wang <[email protected]>

[ Upstream commit d7061627d701c90e1cac1e1e60c45292f64f3470 ]

It turns out to be possible for hotplugging out a device to reach the stage of tearing down the device's group and default domain before the domain's flush queue has drained naturally. At this point, it is then possible for the timeout to expire just before the del_timer() call in free_iova_flush_queue(), such that we then proceed to free the FQ resources while fq_flush_timeout() is still accessing them on another CPU. Crashes due to this have been observed in the wild while removing NVMe devices.

Close the race window by using del_timer_sync() to safely wait for any active timeout handler to finish before we start to free things. We already avoid any locking in free_iova_flush_queue() since the FQ is supposed to be inactive anyway, so the potential deadlock scenario does not apply.

---
 drivers/iommu/iova.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/iommu/iova.c +1 -2
index 9e8bc802ac053..920fcc27c9a1e 100644
@@ -83,8 +83,7 @@ static void free_iova_flush_queue(struct iova_domain *iovad)
 	if (!has_iova_flush_queue(iovad))
 		return;
 
-	if (timer_pending(&iovad->fq_timer))
-		del_timer(&iovad->fq_timer);
+	del_timer_sync(&iovad->fq_timer);
 
 	fq_destroy_all_entries(iovad);
 
Signature · 1 line
2.34.1
author maintainer
#500 lore raw

PATCH 500/1039 ASoC: mediatek: mt8195: correct default value

In reply to Greg Kroah-Hartman

From: Trevor Wu <[email protected]>

[ Upstream commit 30e693ee82d20361f2caacca3b68c79e1a7cb16c ]

mt8195_cg_patch is used to reset the default value of audio cg, so the register value could be consistent with CCF reference count. Nevertheless, AUDIO_TOP_CON1[1:0] is used to control an internal mux, and it's expected to keep the default value 0.

This patch corrects the default value in case an unexpected behavior happens in the future.

Fixes 6746cc8582599 ("ASoC: mediatek: mt8195: add platform driver")
---
 sound/soc/mediatek/mt8195/mt8195-afe-pcm.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
sound/soc/mediatek/mt8195/mt8195-afe-pcm.c +1 -1
index 2bb05a828e8d2..15b4cae2524c1 100644
@@ -3028,7 +3028,7 @@ static const struct reg_sequence mt8195_afe_reg_defaults[] = {
 
 static const struct reg_sequence mt8195_cg_patch[] = {
 	{ AUDIO_TOP_CON0, 0xfffffffb },
-	{ AUDIO_TOP_CON1, 0xfffffffa },
+	{ AUDIO_TOP_CON1, 0xfffffff8 },
 };
 
 static int mt8195_afe_init_registers(struct mtk_base_afe *afe)
Signature · 1 line
2.34.1
author maintainer
#501 lore raw

PATCH 501/1039 counter: 104-quad-8: Fix persistent enabled events bug

In reply to Greg Kroah-Hartman

From: William Breathitt Gray <[email protected]>

[ Upstream commit c95cc0d95702523f8f361b802c9b7d4eeae07f5d ]

A bug exists if the user executes a COUNTER_ADD_WATCH_IOCTL ioctl call, and then executes a COUNTER_DISABLE_EVENTS_IOCTL ioctl call. Disabling the events should disable the 104-QUAD-8 interrupts, but because of this bug the interrupts are not disabling.

The reason this bug is occurring is because quad8_events_configure() is called when COUNTER_DISABLE_EVENTS_IOCTL is handled, but the next_irq_trigger[] array has not been cleared before it is checked in the loop.

This patch fixes the bug by removing the next_irq_trigger array and instead utilizing a different algorithm of walking the events_list list for the current requested events. When a COUNTER_DISABLE_EVENTS_IOCTL is handled, events_list will be empty and thus all device channels end up with interrupts disabled.

---
 drivers/counter/104-quad-8.c | 82 +++++++++++++++++-------------------
 1 file changed, 39 insertions(+), 43 deletions(-)
drivers/counter/104-quad-8.c +39 -43
index 1cbd60aaed697..a97027db0446d 100644
@@ -14,6 +14,7 @@
 #include <linux/interrupt.h>
 #include <linux/isa.h>
 #include <linux/kernel.h>
+#include <linux/list.h>
 #include <linux/module.h>
 #include <linux/moduleparam.h>
 #include <linux/types.h>
@@ -44,7 +45,6 @@ MODULE_PARM_DESC(irq, "ACCES 104-QUAD-8 interrupt line numbers");
  * @ab_enable:		array of A and B inputs enable configurations
  * @preset_enable:	array of set_to_preset_on_index attribute configurations
  * @irq_trigger:	array of current IRQ trigger function configurations
- * @next_irq_trigger:	array of next IRQ trigger function configurations
  * @synchronous_mode:	array of index function synchronous mode configurations
  * @index_polarity:	array of index function polarity configurations
  * @cable_fault_enable:	differential encoder cable status enable configurations
@@ -61,7 +61,6 @@ struct quad8 {
 	unsigned int ab_enable[QUAD8_NUM_COUNTERS];
 	unsigned int preset_enable[QUAD8_NUM_COUNTERS];
 	unsigned int irq_trigger[QUAD8_NUM_COUNTERS];
-	unsigned int next_irq_trigger[QUAD8_NUM_COUNTERS];
 	unsigned int synchronous_mode[QUAD8_NUM_COUNTERS];
 	unsigned int index_polarity[QUAD8_NUM_COUNTERS];
 	unsigned int cable_fault_enable;
@@ -390,7 +389,6 @@ static int quad8_action_read(struct counter_device *counter,
 }
 
 enum {
-	QUAD8_EVENT_NONE = -1,
 	QUAD8_EVENT_CARRY = 0,
 	QUAD8_EVENT_COMPARE = 1,
 	QUAD8_EVENT_CARRY_BORROW = 2,
@@ -402,34 +400,49 @@ static int quad8_events_configure(struct counter_device *counter)
 	struct quad8 *const priv = counter->priv;
 	unsigned long irq_enabled = 0;
 	unsigned long irqflags;
-	size_t channel;
+	struct counter_event_node *event_node;
+	unsigned int next_irq_trigger;
 	unsigned long ior_cfg;
 	unsigned long base_offset;
 
 	spin_lock_irqsave(&priv->lock, irqflags);
 
-	/* Enable interrupts for the requested channels, disable for the rest */
-	for (channel = 0; channel < QUAD8_NUM_COUNTERS; channel++) {
-		if (priv->next_irq_trigger[channel] == QUAD8_EVENT_NONE)
-			continue;
+	list_for_each_entry(event_node, &counter->events_list, l) {
+		switch (event_node->event) {
+		case COUNTER_EVENT_OVERFLOW:
+			next_irq_trigger = QUAD8_EVENT_CARRY;
+			break;
+		case COUNTER_EVENT_THRESHOLD:
+			next_irq_trigger = QUAD8_EVENT_COMPARE;
+			break;
+		case COUNTER_EVENT_OVERFLOW_UNDERFLOW:
+			next_irq_trigger = QUAD8_EVENT_CARRY_BORROW;
+			break;
+		case COUNTER_EVENT_INDEX:
+			next_irq_trigger = QUAD8_EVENT_INDEX;
+			break;
+		default:
+			/* should never reach this path */
+			spin_unlock_irqrestore(&priv->lock, irqflags);
+			return -EINVAL;
+		}
 
-		if (priv->irq_trigger[channel] != priv->next_irq_trigger[channel]) {
-			/* Save new IRQ function configuration */
-			priv->irq_trigger[channel] = priv->next_irq_trigger[channel];
+		/* Skip configuration if it is the same as previously set */
+		if (priv->irq_trigger[event_node->channel] == next_irq_trigger)
+			continue;
 
-			/* Load configuration to I/O Control Register */
-			ior_cfg = priv->ab_enable[channel] |
-				  priv->preset_enable[channel] << 1 |
-				  priv->irq_trigger[channel] << 3;
-			base_offset = priv->base + 2 * channel + 1;
-			outb(QUAD8_CTR_IOR | ior_cfg, base_offset);
-		}
+		/* Save new IRQ function configuration */
+		priv->irq_trigger[event_node->channel] = next_irq_trigger;
 
-		/* Reset next IRQ trigger function configuration */
-		priv->next_irq_trigger[channel] = QUAD8_EVENT_NONE;
+		/* Load configuration to I/O Control Register */
+		ior_cfg = priv->ab_enable[event_node->channel] |
+			  priv->preset_enable[event_node->channel] << 1 |
+			  priv->irq_trigger[event_node->channel] << 3;
+		base_offset = priv->base + 2 * event_node->channel + 1;
+		outb(QUAD8_CTR_IOR | ior_cfg, base_offset);
 
 		/* Enable IRQ line */
-		irq_enabled |= BIT(channel);
+		irq_enabled |= BIT(event_node->channel);
 	}
 
 	outb(irq_enabled, priv->base + QUAD8_REG_INDEX_INTERRUPT);
@@ -442,35 +455,20 @@ static int quad8_events_configure(struct counter_device *counter)
 static int quad8_watch_validate(struct counter_device *counter,
 				const struct counter_watch *watch)
 {
-	struct quad8 *const priv = counter->priv;
+	struct counter_event_node *event_node;
 
 	if (watch->channel > QUAD8_NUM_COUNTERS - 1)
 		return -EINVAL;
 
 	switch (watch->event) {
 	case COUNTER_EVENT_OVERFLOW:
-		if (priv->next_irq_trigger[watch->channel] == QUAD8_EVENT_NONE)
-			priv->next_irq_trigger[watch->channel] = QUAD8_EVENT_CARRY;
-		else if (priv->next_irq_trigger[watch->channel] != QUAD8_EVENT_CARRY)
-			return -EINVAL;
-		return 0;
 	case COUNTER_EVENT_THRESHOLD:
-		if (priv->next_irq_trigger[watch->channel] == QUAD8_EVENT_NONE)
-			priv->next_irq_trigger[watch->channel] = QUAD8_EVENT_COMPARE;
-		else if (priv->next_irq_trigger[watch->channel] != QUAD8_EVENT_COMPARE)
-			return -EINVAL;
-		return 0;
 	case COUNTER_EVENT_OVERFLOW_UNDERFLOW:
-		if (priv->next_irq_trigger[watch->channel] == QUAD8_EVENT_NONE)
-			priv->next_irq_trigger[watch->channel] = QUAD8_EVENT_CARRY_BORROW;
-		else if (priv->next_irq_trigger[watch->channel] != QUAD8_EVENT_CARRY_BORROW)
-			return -EINVAL;
-		return 0;
 	case COUNTER_EVENT_INDEX:
-		if (priv->next_irq_trigger[watch->channel] == QUAD8_EVENT_NONE)
-			priv->next_irq_trigger[watch->channel] = QUAD8_EVENT_INDEX;
-		else if (priv->next_irq_trigger[watch->channel] != QUAD8_EVENT_INDEX)
-			return -EINVAL;
+		list_for_each_entry(event_node, &counter->next_events_list, l)
+			if (watch->channel == event_node->channel &&
+				watch->event != event_node->event)
+				return -EINVAL;
 		return 0;
 	default:
 		return -EINVAL;
@@ -1183,8 +1181,6 @@ static int quad8_probe(struct device *dev, unsigned int id)
 		outb(QUAD8_CTR_IOR, base_offset + 1);
 		/* Disable index function; negative index polarity */
 		outb(QUAD8_CTR_IDR, base_offset + 1);
-		/* Initialize next IRQ trigger function configuration */
-		priv->next_irq_trigger[i] = QUAD8_EVENT_NONE;
 	}
 	/* Disable Differential Encoder Cable Status for all channels */
 	outb(0xFF, base[id] + QUAD8_DIFF_ENCODER_CABLE_STATUS);
Signature · 1 line
2.34.1
author maintainer
#502 lore raw

PATCH 502/1039 of: fdt: Aggregate the processing of "linux,usable-memory-range"

In reply to Greg Kroah-Hartman

From: Zhen Lei <[email protected]>

[ Upstream commit 8347b41748c3019157312fbe7f8a6792ae396eb7 ]

Currently, we parse the "linux,usable-memory-range" property in early_init_dt_scan_chosen(), to obtain the specified memory range of the crash kernel. We then reserve the required memory after early_init_dt_scan_memory() has identified all available physical memory. Because the two pieces of code are separated far, the readability and maintainability are reduced. So bring them together.

(change the prototype of early_init_dt_check_for_usable_mem_range(), in order to use it outside)

To: [email protected]
To: [email protected]

---
 drivers/of/fdt.c | 19 +++++++++++++------
 1 file changed, 13 insertions(+), 6 deletions(-)
drivers/of/fdt.c +13 -6
index bdca35284cebd..5a238a933eb29 100644
@@ -965,18 +965,22 @@ static void __init early_init_dt_check_for_elfcorehdr(unsigned long node)
 		 elfcorehdr_addr, elfcorehdr_size);
 }
 
-static phys_addr_t cap_mem_addr;
-static phys_addr_t cap_mem_size;
+static unsigned long chosen_node_offset = -FDT_ERR_NOTFOUND;
 
 /**
  * early_init_dt_check_for_usable_mem_range - Decode usable memory range
  * location from flat tree
- * @node: reference to node containing usable memory range location ('chosen')
  */
-static void __init early_init_dt_check_for_usable_mem_range(unsigned long node)
+static void __init early_init_dt_check_for_usable_mem_range(void)
 {
 	const __be32 *prop;
 	int len;
+	phys_addr_t cap_mem_addr;
+	phys_addr_t cap_mem_size;
+	unsigned long node = chosen_node_offset;
+
+	if ((long)node < 0)
+		return;
 
 	pr_debug("Looking for usable-memory-range property... ");
 
@@ -989,6 +993,8 @@ static void __init early_init_dt_check_for_usable_mem_range(unsigned long node)
 
 	pr_debug("cap_mem_start=%pa cap_mem_size=%pa\n", &cap_mem_addr,
 		 &cap_mem_size);
+
+	memblock_cap_memory_range(cap_mem_addr, cap_mem_size);
 }
 
 #ifdef CONFIG_SERIAL_EARLYCON
@@ -1137,9 +1143,10 @@ int __init early_init_dt_scan_chosen(unsigned long node, const char *uname,
 	    (strcmp(uname, "chosen") != 0 && strcmp(uname, "chosen@0") != 0))
 		return 0;
 
+	chosen_node_offset = node;
+
 	early_init_dt_check_for_initrd(node);
 	early_init_dt_check_for_elfcorehdr(node);
-	early_init_dt_check_for_usable_mem_range(node);
 
 	/* Retrieve command line */
 	p = of_get_flat_dt_prop(node, "bootargs", &l);
@@ -1275,7 +1282,7 @@ void __init early_init_dt_scan_nodes(void)
 	of_scan_flat_dt(early_init_dt_scan_memory, NULL);
 
 	/* Handle linux,usable-memory-range property */
-	memblock_cap_memory_range(cap_mem_addr, cap_mem_size);
+	early_init_dt_check_for_usable_mem_range();
 }
 
 bool __init early_init_dt_scan(void *params)
Signature · 9 lines
2.34.1




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

PATCH 503/1039 efi: apply memblock cap after memblock_add()

In reply to Greg Kroah-Hartman

From: Pingfan Liu <[email protected]>

[ Upstream commit b398123bff3bcbc1facb0f29bf6e7b9f1bc55931 ]

On arm64, during kdump kernel saves vmcore, it runs into the following bug:
...

[   15.148919] usercopy: Kernel memory exposure attempt detected from SLUB object 'kmem_cache_node' (offset 0, size 4096)!
[   15.159707] ------------[ cut here ]------------
[   15.164311] kernel BUG at mm/usercopy.c:99!
[   15.168482] Internal error: Oops - BUG: 0 [#1] SMP
[   15.173261] Modules linked in: xfs libcrc32c crct10dif_ce ghash_ce sha2_ce sha256_arm64 sha1_ce sbsa_gwdt ast i2c_algo_bit drm_vram_helper drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops cec drm_ttm_helper ttm drm nvme nvme_core xgene_hwmon i2c_designware_platform i2c_designware_core dm_mirror dm_region_hash dm_log dm_mod overlay squashfs zstd_decompress loop
[   15.206186] CPU: 0 PID: 542 Comm: cp Not tainted 5.16.0-rc4 #1
[   15.212006] Hardware name: GIGABYTE R272-P30-JG/MP32-AR0-JG, BIOS F12 (SCP: 1.5.20210426) 05/13/2021
[   15.221125] pstate: 60400009 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[   15.228073] pc : usercopy_abort+0x9c/0xa0
[   15.232074] lr : usercopy_abort+0x9c/0xa0
[   15.236070] sp : ffff8000121abba0
[   15.239371] x29: ffff8000121abbb0 x28: 0000000000003000 x27: 0000000000000000
[   15.246494] x26: 0000000080000400 x25: 0000ffff885c7000 x24: 0000000000000000
[   15.253617] x23: 000007ff80400000 x22: ffff07ff80401000 x21: 0000000000000001
[   15.260739] x20: 0000000000001000 x19: ffff07ff80400000 x18: ffffffffffffffff
[   15.267861] x17: 656a626f2042554c x16: 53206d6f72662064 x15: 6574636574656420
[   15.274983] x14: 74706d6574746120 x13: 2129363930342065 x12: 7a6973202c302074
[   15.282105] x11: ffffc8b041d1b148 x10: 00000000ffff8000 x9 : ffffc8b04012812c
[   15.289228] x8 : 00000000ffff7fff x7 : ffffc8b041d1b148 x6 : 0000000000000000
[   15.296349] x5 : 0000000000000000 x4 : 0000000000007fff x3 : 0000000000000000
[   15.303471] x2 : 0000000000000000 x1 : ffff07ff8c064800 x0 : 000000000000006b
[   15.310593] Call trace:
[   15.313027]  usercopy_abort+0x9c/0xa0
[   15.316677]  __check_heap_object+0xd4/0xf0
[   15.320762]  __check_object_size.part.0+0x160/0x1e0
[   15.325628]  __check_object_size+0x2c/0x40
[   15.329711]  copy_oldmem_page+0x7c/0x140
[   15.333623]  read_from_oldmem.part.0+0xfc/0x1c0
[   15.338142]  __read_vmcore.constprop.0+0x23c/0x350
[   15.342920]  read_vmcore+0x28/0x34
[   15.346309]  proc_reg_read+0xb4/0xf0
[   15.349871]  vfs_read+0xb8/0x1f0
[   15.353088]  ksys_read+0x74/0x100
[   15.356390]  __arm64_sys_read+0x28/0x34

...

This bug introduced by commit b261dba2fdb2 ("arm64: kdump: Remove custom linux,usable-memory-range handling"), which moves memblock_cap_memory_range() to fdt, but it breaches the rules that memblock_cap_memory_range() should come after memblock_add() etc as said in commit e888fa7bb882 ("memblock: Check memory add/cap ordering").

As a consequence, the virtual address set up by copy_oldmem_page() does not bail out from the test of virt_addr_valid() in check_heap_object(), and finally hits the BUG_ON().

Since memblock allocator has no idea about when the memblock is fully populated, while efi_init() is aware, so tackling this issue by calling the interface early_init_dt_check_for_usable_mem_range() exposed by of/fdt.

Fixes b261dba2fdb2 ("arm64: kdump: Remove custom linux,usable-memory-range handling")
Cc Rob Herring <[email protected]>
Cc Zhen Lei <[email protected]>
Cc Catalin Marinas <[email protected]>
Cc Will Deacon <[email protected]>
Cc Andrew Morton <[email protected]>
Cc Mike Rapoport <[email protected]>
Cc Geert Uytterhoeven <[email protected]>
Cc Frank Rowand <[email protected]>
Cc Ard Biesheuvel <[email protected]>
Cc Nick Terrell <[email protected]>

To: [email protected]
To: [email protected]

---
 drivers/firmware/efi/efi-init.c | 5 +++++
 drivers/of/fdt.c                | 2 +-
 include/linux/of_fdt.h          | 2 ++
 3 files changed, 8 insertions(+), 1 deletion(-)
drivers/firmware/efi/efi-init.c +5 -0
index b19ce1a83f91a..b2c829e95bd14 100644
@@ -235,6 +235,11 @@ void __init efi_init(void)
 	}
 
 	reserve_regions();
+	/*
+	 * For memblock manipulation, the cap should come after the memblock_add().
+	 * And now, memblock is fully populated, it is time to do capping.
+	 */
+	early_init_dt_check_for_usable_mem_range();
 	efi_esrt_init();
 	efi_mokvar_table_init();
 
drivers/of/fdt.c +1 -1
index 5a238a933eb29..65af475dfa950 100644
@@ -971,7 +971,7 @@ static unsigned long chosen_node_offset = -FDT_ERR_NOTFOUND;
  * early_init_dt_check_for_usable_mem_range - Decode usable memory range
  * location from flat tree
  */
-static void __init early_init_dt_check_for_usable_mem_range(void)
+void __init early_init_dt_check_for_usable_mem_range(void)
 {
 	const __be32 *prop;
 	int len;
include/linux/of_fdt.h +2 -0
index cf48983d3c867..ad09beb6d13c4 100644
@@ -62,6 +62,7 @@ extern int early_init_dt_scan_chosen(unsigned long node, const char *uname,
 				     int depth, void *data);
 extern int early_init_dt_scan_memory(unsigned long node, const char *uname,
 				     int depth, void *data);
+extern void early_init_dt_check_for_usable_mem_range(void);
 extern int early_init_dt_scan_chosen_stdout(void);
 extern void early_init_fdt_scan_reserved_mem(void);
 extern void early_init_fdt_reserve_self(void);
@@ -86,6 +87,7 @@ extern void unflatten_and_copy_device_tree(void);
 extern void early_init_devtree(void *);
 extern void early_get_first_memblock_info(void *, phys_addr_t *);
 #else /* CONFIG_OF_EARLY_FLATTREE */
+static inline void early_init_dt_check_for_usable_mem_range(void) {}
 static inline int early_init_dt_scan_chosen_stdout(void) { return -ENODEV; }
 static inline void early_init_fdt_scan_reserved_mem(void) {}
 static inline void early_init_fdt_reserve_self(void) {}
Signature · 9 lines
2.34.1




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

PATCH 504/1039 scsi: block: pm: Always set request queue runtime active in blk_post_runtime_resume()

In reply to Greg Kroah-Hartman

From: Alan Stern <[email protected]>

[ Upstream commit 6e1fcab00a23f7fe9f4fe9704905a790efa1eeab ]

John Garry reported a deadlock that occurs when trying to access a runtime-suspended SATA device. For obscure reasons, the rescan procedure causes the link to be hard-reset, which disconnects the device.

The rescan tries to carry out a runtime resume when accessing the device. scsi_rescan_device() holds the SCSI device lock and won't release it until it can put commands onto the device's block queue. This can't happen until the queue is successfully runtime-resumed or the device is unregistered. But the runtime resume fails because the device is disconnected, and __scsi_remove_device() can't do the unregistration because it can't get the device lock.

The best way to resolve this deadlock appears to be to allow the block queue to start running again even after an unsuccessful runtime resume. The idea is that the driver or the SCSI error handler will need to be able to use the queue to resolve the runtime resume failure.

This patch removes the err argument to blk_post_runtime_resume() and makes the routine act as though the resume was successful always. This fixes the deadlock.

---
 block/blk-pm.c         | 22 +++++++---------------
 drivers/scsi/scsi_pm.c |  2 +-
 include/linux/blk-pm.h |  2 +-
 3 files changed, 9 insertions(+), 17 deletions(-)
block/blk-pm.c +7 -15
index 17bd020268d42..2dad62cc15727 100644
@@ -163,27 +163,19 @@ EXPORT_SYMBOL(blk_pre_runtime_resume);
 /**
  * blk_post_runtime_resume - Post runtime resume processing
  * @q: the queue of the device
- * @err: return value of the device's runtime_resume function
  *
  * Description:
- *    Update the queue's runtime status according to the return value of the
- *    device's runtime_resume function. If the resume was successful, call
- *    blk_set_runtime_active() to do the real work of restarting the queue.
+ *    For historical reasons, this routine merely calls blk_set_runtime_active()
+ *    to do the real work of restarting the queue.  It does this regardless of
+ *    whether the device's runtime-resume succeeded; even if it failed the
+ *    driver or error handler will need to communicate with the device.
  *
  *    This function should be called near the end of the device's
  *    runtime_resume callback.
  */
-void blk_post_runtime_resume(struct request_queue *q, int err)
+void blk_post_runtime_resume(struct request_queue *q)
 {
-	if (!q->dev)
-		return;
-	if (!err) {
-		blk_set_runtime_active(q);
-	} else {
-		spin_lock_irq(&q->queue_lock);
-		q->rpm_status = RPM_SUSPENDED;
-		spin_unlock_irq(&q->queue_lock);
-	}
+	blk_set_runtime_active(q);
 }
 EXPORT_SYMBOL(blk_post_runtime_resume);
 
@@ -201,7 +193,7 @@ EXPORT_SYMBOL(blk_post_runtime_resume);
  * runtime PM status and re-enable peeking requests from the queue. It
  * should be called before first request is added to the queue.
  *
- * This function is also called by blk_post_runtime_resume() for successful
+ * This function is also called by blk_post_runtime_resume() for
  * runtime resumes.  It does everything necessary to restart the queue.
  */
 void blk_set_runtime_active(struct request_queue *q)
drivers/scsi/scsi_pm.c +1 -1
index b5a858c29488a..f06ca9d2a597d 100644
@@ -181,7 +181,7 @@ static int sdev_runtime_resume(struct device *dev)
 	blk_pre_runtime_resume(sdev->request_queue);
 	if (pm && pm->runtime_resume)
 		err = pm->runtime_resume(dev);
-	blk_post_runtime_resume(sdev->request_queue, err);
+	blk_post_runtime_resume(sdev->request_queue);
 
 	return err;
 }
include/linux/blk-pm.h +1 -1
index b80c65aba2493..2580e05a8ab67 100644
@@ -14,7 +14,7 @@ extern void blk_pm_runtime_init(struct request_queue *q, struct device *dev);
 extern int blk_pre_runtime_suspend(struct request_queue *q);
 extern void blk_post_runtime_suspend(struct request_queue *q, int err);
 extern void blk_pre_runtime_resume(struct request_queue *q);
-extern void blk_post_runtime_resume(struct request_queue *q, int err);
+extern void blk_post_runtime_resume(struct request_queue *q);
 extern void blk_set_runtime_active(struct request_queue *q);
 #else
 static inline void blk_pm_runtime_init(struct request_queue *q,
Signature · 1 line
2.34.1
author maintainer
#505 lore raw

PATCH 505/1039 phy: uniphier-usb3ss: fix unintended writing zeros to PHY register

In reply to Greg Kroah-Hartman

From: Ryuta NAKANISHI <[email protected]>

[ Upstream commit 898c7a9ec81620125f2463714a0f4dea18ad6e54 ]

Similar to commit 4a90bbb478db ("phy: uniphier-pcie: Fix updating phy parameters"), in function uniphier_u3ssphy_set_param(), unintentionally write zeros to other fields when writing PHY registers.

Fixes 5ab43d0f8697 ("phy: socionext: add USB3 PHY driver for UniPhier SoC")
---
 drivers/phy/socionext/phy-uniphier-usb3ss.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)
drivers/phy/socionext/phy-uniphier-usb3ss.c +6 -4
index 6700645bcbe6b..3b5ffc16a6947 100644
@@ -22,11 +22,13 @@
 #include <linux/reset.h>
 
 #define SSPHY_TESTI		0x0
-#define SSPHY_TESTO		0x4
 #define TESTI_DAT_MASK		GENMASK(13, 6)
 #define TESTI_ADR_MASK		GENMASK(5, 1)
 #define TESTI_WR_EN		BIT(0)
 
+#define SSPHY_TESTO		0x4
+#define TESTO_DAT_MASK		GENMASK(7, 0)
+
 #define PHY_F(regno, msb, lsb) { (regno), (msb), (lsb) }
 
 #define CDR_CPD_TRIM	PHY_F(7, 3, 0)	/* RxPLL charge pump current */
@@ -84,12 +86,12 @@ static void uniphier_u3ssphy_set_param(struct uniphier_u3ssphy_priv *priv,
 	val  = FIELD_PREP(TESTI_DAT_MASK, 1);
 	val |= FIELD_PREP(TESTI_ADR_MASK, p->field.reg_no);
 	uniphier_u3ssphy_testio_write(priv, val);
-	val = readl(priv->base + SSPHY_TESTO);
+	val = readl(priv->base + SSPHY_TESTO) & TESTO_DAT_MASK;
 
 	/* update value */
-	val &= ~FIELD_PREP(TESTI_DAT_MASK, field_mask);
+	val &= ~field_mask;
 	data = field_mask & (p->value << p->field.lsb);
-	val  = FIELD_PREP(TESTI_DAT_MASK, data);
+	val  = FIELD_PREP(TESTI_DAT_MASK, data | val);
 	val |= FIELD_PREP(TESTI_ADR_MASK, p->field.reg_no);
 	uniphier_u3ssphy_testio_write(priv, val);
 	uniphier_u3ssphy_testio_write(priv, val | TESTI_WR_EN);
Signature · 1 line
2.34.1
author maintainer
#506 lore raw

PATCH 506/1039 ASoC: mediatek: Check for error clk pointer

In reply to Greg Kroah-Hartman

From: Jiasheng Jiang <[email protected]>

[ Upstream commit 9de2b9286a6dd16966959b3cb34fc2ddfd39213e ]

Yes, you are right and now the return code depending on the init_clks().

Fixes 6078c651947a ("soc: mediatek: Refine scpsys to support multiple platform")
---
 drivers/soc/mediatek/mtk-scpsys.c | 15 +++++++++++----
 1 file changed, 11 insertions(+), 4 deletions(-)
drivers/soc/mediatek/mtk-scpsys.c +11 -4
index ca75b14931ec9..670cc82d17dc2 100644
@@ -411,12 +411,17 @@ out:
 	return ret;
 }
 
-static void init_clks(struct platform_device *pdev, struct clk **clk)
+static int init_clks(struct platform_device *pdev, struct clk **clk)
 {
 	int i;
 
-	for (i = CLK_NONE + 1; i < CLK_MAX; i++)
+	for (i = CLK_NONE + 1; i < CLK_MAX; i++) {
 		clk[i] = devm_clk_get(&pdev->dev, clk_names[i]);
+		if (IS_ERR(clk[i]))
+			return PTR_ERR(clk[i]);
+	}
+
+	return 0;
 }
 
 static struct scp *init_scp(struct platform_device *pdev,
@@ -426,7 +431,7 @@ static struct scp *init_scp(struct platform_device *pdev,
 {
 	struct genpd_onecell_data *pd_data;
 	struct resource *res;
-	int i, j;
+	int i, j, ret;
 	struct scp *scp;
 	struct clk *clk[CLK_MAX];
 
@@ -481,7 +486,9 @@ static struct scp *init_scp(struct platform_device *pdev,
 
 	pd_data->num_domains = num;
 
-	init_clks(pdev, clk);
+	ret = init_clks(pdev, clk);
+	if (ret)
+		return ERR_PTR(ret);
 
 	for (i = 0; i < num; i++) {
 		struct scp_domain *scpd = &scp->domains[i];
Signature · 1 line
2.34.1
author maintainer
#507 lore raw

PATCH 507/1039 powerpc/64s: Mask NIP before checking against SRR0

In reply to Greg Kroah-Hartman

From: Michael Ellerman <[email protected]>

[ Upstream commit 314f6c23dd8d417281eb9e8a516dd98036f2e7b3 ]

When CONFIG_PPC_RFI_SRR_DEBUG=y we check that NIP and SRR0 match when returning from interrupts. This can trigger falsely if NIP has either of its two low bits set via sigreturn or ptrace, while SRR0 has its low two bits masked in hardware.

As a quick fix make sure to mask the low bits before doing the check.

Fixes 59dc5bfca0cb ("powerpc/64s: avoid reloading (H)SRR registers if they are still valid")
---
 arch/powerpc/kernel/interrupt_64.S | 2 ++
 1 file changed, 2 insertions(+)
arch/powerpc/kernel/interrupt_64.S +2 -0
index ec950b08a8dcc..894588b2381e5 100644
@@ -30,6 +30,7 @@ COMPAT_SYS_CALL_TABLE:
 	.ifc \srr,srr
 	mfspr	r11,SPRN_SRR0
 	ld	r12,_NIP(r1)
+	clrrdi  r12,r12,2
 100:	tdne	r11,r12
 	EMIT_BUG_ENTRY 100b,__FILE__,__LINE__,(BUGFLAG_WARNING | BUGFLAG_ONCE)
 	mfspr	r11,SPRN_SRR1
@@ -39,6 +40,7 @@ COMPAT_SYS_CALL_TABLE:
 	.else
 	mfspr	r11,SPRN_HSRR0
 	ld	r12,_NIP(r1)
+	clrrdi  r12,r12,2
 100:	tdne	r11,r12
 	EMIT_BUG_ENTRY 100b,__FILE__,__LINE__,(BUGFLAG_WARNING | BUGFLAG_ONCE)
 	mfspr	r11,SPRN_HSRR1
Signature · 1 line
2.34.1
author maintainer
#508 lore raw

PATCH 508/1039 powerpc/64s: Use EMIT_WARN_ENTRY for SRR debug warnings

In reply to Greg Kroah-Hartman

From: Michael Ellerman <[email protected]>

[ Upstream commit fd1eaaaaa6864b5fb8f99880fcefb49760b8fe4e ]

When CONFIG_PPC_RFI_SRR_DEBUG=y we check the SRR values before returning from interrupts. This is done in asm using EMIT_BUG_ENTRY, and passing BUGFLAG_WARNING.

However that fails to create an exception table entry for the warning, and so do_program_check() fails the exception table search and proceeds to call _exception(), resulting in an oops like:

  Oops: Exception in kernel mode, sig: 5 [#1]
  LE PAGE_SIZE=64K MMU=Radix SMP NR_CPUS=2048 NUMA pSeries
  Modules linked in:
  CPU: 2 PID: 1204 Comm: sigreturn_unali Tainted: P                  5.16.0-rc2-00194-g91ca3d4f77c5 #12
  NIP:  c00000000000c5b0 LR: 0000000000000000 CTR: 0000000000000000
  ...
  NIP [c00000000000c5b0] system_call_common+0x150/0x268
  LR [0000000000000000] 0x0
  Call Trace:
  [c00000000db73e10] [c00000000000c558] system_call_common+0xf8/0x268 (unreliable)
  ...
  Instruction dump:
  7cc803a6 888d0931 2c240000 4082001c 38800000 988d0931 e8810170 e8a10178
  7c9a03a6 7cbb03a6 7d7a02a6 e9810170 <7f0b6088> 7d7b02a6 e9810178 7f0b6088

We should instead use EMIT_WARN_ENTRY, which creates an exception table entry for the warning, allowing the warning to be correctly recognised, and the code to resume after printing the warning.

Note however that because this warning is buried deep in the interrupt return path, we are not able to recover from it (due to MSR_RI being clear), so we still end up in die() with an unrecoverable exception.

Fixes 59dc5bfca0cb ("powerpc/64s: avoid reloading (H)SRR registers if they are still valid")
---
 arch/powerpc/kernel/interrupt_64.S | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
arch/powerpc/kernel/interrupt_64.S +4 -4
index 894588b2381e5..4b1ff94e67eb4 100644
@@ -32,21 +32,21 @@ COMPAT_SYS_CALL_TABLE:
 	ld	r12,_NIP(r1)
 	clrrdi  r12,r12,2
 100:	tdne	r11,r12
-	EMIT_BUG_ENTRY 100b,__FILE__,__LINE__,(BUGFLAG_WARNING | BUGFLAG_ONCE)
+	EMIT_WARN_ENTRY 100b,__FILE__,__LINE__,(BUGFLAG_WARNING | BUGFLAG_ONCE)
 	mfspr	r11,SPRN_SRR1
 	ld	r12,_MSR(r1)
 100:	tdne	r11,r12
-	EMIT_BUG_ENTRY 100b,__FILE__,__LINE__,(BUGFLAG_WARNING | BUGFLAG_ONCE)
+	EMIT_WARN_ENTRY 100b,__FILE__,__LINE__,(BUGFLAG_WARNING | BUGFLAG_ONCE)
 	.else
 	mfspr	r11,SPRN_HSRR0
 	ld	r12,_NIP(r1)
 	clrrdi  r12,r12,2
 100:	tdne	r11,r12
-	EMIT_BUG_ENTRY 100b,__FILE__,__LINE__,(BUGFLAG_WARNING | BUGFLAG_ONCE)
+	EMIT_WARN_ENTRY 100b,__FILE__,__LINE__,(BUGFLAG_WARNING | BUGFLAG_ONCE)
 	mfspr	r11,SPRN_HSRR1
 	ld	r12,_MSR(r1)
 100:	tdne	r11,r12
-	EMIT_BUG_ENTRY 100b,__FILE__,__LINE__,(BUGFLAG_WARNING | BUGFLAG_ONCE)
+	EMIT_WARN_ENTRY 100b,__FILE__,__LINE__,(BUGFLAG_WARNING | BUGFLAG_ONCE)
 	.endif
 #endif
 .endm
Signature · 1 line
2.34.1
author maintainer
#509 lore raw

PATCH 509/1039 phy: cadence: Sierra: Fix to get correct parent for mux clocks

In reply to Greg Kroah-Hartman

From: Swapnil Jakhade <[email protected]>

[ Upstream commit da08aab940092a050a4fb2857ed9479d2b0e03c4 ]

Fix get_parent() callback to return the correct index of the parent for PLL_CMNLC1 clock. Add a separate table of register values corresponding to the parent index for PLL_CMNLC1. Update set_parent() callback accordingly.

Fixes 28081b72859f ("phy: cadence: Sierra: Model PLL_CMNLC and PLL_CMNLC1 as clocks (mux clocks)")
---
 drivers/phy/cadence/phy-cadence-sierra.c | 31 ++++++++++++++++++++----
 1 file changed, 26 insertions(+), 5 deletions(-)
drivers/phy/cadence/phy-cadence-sierra.c +26 -5
index e93818e3991fd..3e2d096d54fd7 100644
@@ -215,7 +215,10 @@ static const int pll_mux_parent_index[][SIERRA_NUM_CMN_PLLC_PARENTS] = {
 	[CMN_PLLLC1] = { PLL1_REFCLK, PLL0_REFCLK },
 };
 
-static u32 cdns_sierra_pll_mux_table[] = { 0, 1 };
+static u32 cdns_sierra_pll_mux_table[][SIERRA_NUM_CMN_PLLC_PARENTS] = {
+	[CMN_PLLLC] = { 0, 1 },
+	[CMN_PLLLC1] = { 1, 0 },
+};
 
 struct cdns_sierra_inst {
 	struct phy *phy;
@@ -436,11 +439,25 @@ static const struct phy_ops ops = {
 static u8 cdns_sierra_pll_mux_get_parent(struct clk_hw *hw)
 {
 	struct cdns_sierra_pll_mux *mux = to_cdns_sierra_pll_mux(hw);
+	struct regmap_field *plllc1en_field = mux->plllc1en_field;
+	struct regmap_field *termen_field = mux->termen_field;
 	struct regmap_field *field = mux->pfdclk_sel_preg;
 	unsigned int val;
+	int index;
 
 	regmap_field_read(field, &val);
-	return clk_mux_val_to_index(hw, cdns_sierra_pll_mux_table, 0, val);
+
+	if (strstr(clk_hw_get_name(hw), clk_names[CDNS_SIERRA_PLL_CMNLC1])) {
+		index = clk_mux_val_to_index(hw, cdns_sierra_pll_mux_table[CMN_PLLLC1], 0, val);
+		if (index == 1) {
+			regmap_field_write(plllc1en_field, 1);
+			regmap_field_write(termen_field, 1);
+		}
+	} else {
+		index = clk_mux_val_to_index(hw, cdns_sierra_pll_mux_table[CMN_PLLLC], 0, val);
+	}
+
+	return index;
 }
 
 static int cdns_sierra_pll_mux_set_parent(struct clk_hw *hw, u8 index)
@@ -458,7 +475,11 @@ static int cdns_sierra_pll_mux_set_parent(struct clk_hw *hw, u8 index)
 		ret |= regmap_field_write(termen_field, 1);
 	}
 
-	val = cdns_sierra_pll_mux_table[index];
+	if (strstr(clk_hw_get_name(hw), clk_names[CDNS_SIERRA_PLL_CMNLC1]))
+		val = cdns_sierra_pll_mux_table[CMN_PLLLC1][index];
+	else
+		val = cdns_sierra_pll_mux_table[CMN_PLLLC][index];
+
 	ret |= regmap_field_write(field, val);
 
 	return ret;
@@ -496,8 +517,8 @@ static int cdns_sierra_pll_mux_register(struct cdns_sierra_phy *sp,
 	for (i = 0; i < num_parents; i++) {
 		clk = sp->input_clks[pll_mux_parent_index[clk_index][i]];
 		if (IS_ERR_OR_NULL(clk)) {
-			dev_err(dev, "No parent clock for derived_refclk\n");
-			return PTR_ERR(clk);
+			dev_err(dev, "No parent clock for PLL mux clocks\n");
+			return IS_ERR(clk) ? PTR_ERR(clk) : -ENOENT;
 		}
 		parent_names[i] = __clk_get_name(clk);
 	}
Signature · 1 line
2.34.1
author maintainer
#510 lore raw

PATCH 510/1039 iio: chemical: sunrise_co2: set val parameter only on success

In reply to Greg Kroah-Hartman

From: Tom Rix <[email protected]>

[ Upstream commit 38ac2f038666521f94d4fa37b5a9441cef832ccf ]

Clang static analysis reports this representative warning

sunrise_co2.c:410:9: warning: Assigned value is garbage or undefined

  *val = value;
       ^ ~~~~~

The ealier call to sunrise_read_word can fail without setting value. So defer setting val until we know the read was successful.

Fixes c397894e24f1 ("iio: chemical: Add Senseair Sunrise 006-0-007 driver")
---
 drivers/iio/chemical/sunrise_co2.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/iio/chemical/sunrise_co2.c +2 -2
index 233bd0f379c93..8440dc0c77cfe 100644
@@ -407,24 +407,24 @@ static int sunrise_read_raw(struct iio_dev *iio_dev,
 			mutex_lock(&sunrise->lock);
 			ret = sunrise_read_word(sunrise, SUNRISE_CO2_FILTERED_COMP_REG,
 						&value);
-			*val = value;
 			mutex_unlock(&sunrise->lock);
 
 			if (ret)
 				return ret;
 
+			*val = value;
 			return IIO_VAL_INT;
 
 		case IIO_TEMP:
 			mutex_lock(&sunrise->lock);
 			ret = sunrise_read_word(sunrise, SUNRISE_CHIP_TEMPERATURE_REG,
 						&value);
-			*val = value;
 			mutex_unlock(&sunrise->lock);
 
 			if (ret)
 				return ret;
 
+			*val = value;
 			return IIO_VAL_INT;
 
 		default:
Signature · 1 line
2.34.1
author maintainer
#511 lore raw

PATCH 511/1039 ASoC: samsung: idma: Check of ioremap return value

In reply to Greg Kroah-Hartman

From: Jiasheng Jiang <[email protected]>

[ Upstream commit 3ecb46755eb85456b459a1a9f952c52986bce8ec ]

Because of the potential failure of the ioremap(), the buf->area could be NULL.
Therefore, we need to check it and return -ENOMEM in order to transfer the error.

---
 sound/soc/samsung/idma.c | 2 ++
 1 file changed, 2 insertions(+)
sound/soc/samsung/idma.c +2 -0
index 66bcc2f97544b..c3f1b054e2389 100644
@@ -360,6 +360,8 @@ static int preallocate_idma_buffer(struct snd_pcm *pcm, int stream)
 	buf->addr = idma.lp_tx_addr;
 	buf->bytes = idma_hardware.buffer_bytes_max;
 	buf->area = (unsigned char * __force)ioremap(buf->addr, buf->bytes);
+	if (!buf->area)
+		return -ENOMEM;
 
 	return 0;
 }
Signature · 1 line
2.34.1
author maintainer
#512 lore raw

PATCH 512/1039 misc: lattice-ecp3-config: Fix task hung when firmware load failed

In reply to Greg Kroah-Hartman

From: Wei Yongjun <[email protected]>

[ Upstream commit fcee5ce50bdb21116711e38635e3865594af907e ]

When firmware load failed, kernel report task hung as follows:

INFO: task xrun:5191 blocked for more than 147 seconds.
      Tainted: G        W         5.16.0-rc5-next-20211220+ #11
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
task:xrun            state:D stack:    0 pid: 5191 ppid:   270 flags:0x00000004
Call Trace:
 __schedule+0xc12/0x4b50 kernel/sched/core.c:4986
 schedule+0xd7/0x260 kernel/sched/core.c:6369 (discriminator 1)
 schedule_timeout+0x7aa/0xa80 kernel/time/timer.c:1857
 wait_for_completion+0x181/0x290 kernel/sched/completion.c:85
 lattice_ecp3_remove+0x32/0x40 drivers/misc/lattice-ecp3-config.c:221
 spi_remove+0x72/0xb0 drivers/spi/spi.c:409

lattice_ecp3_remove() wait for signals from firmware loading, but when load failed, firmware_load() does not send this signal. This cause device remove hung. Fix it by sending signal even if load failed.

Fixes 781551df57c7 ("misc: Add Lattice ECP3 FPGA configuration via SPI")
---
 drivers/misc/lattice-ecp3-config.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)
drivers/misc/lattice-ecp3-config.c +6 -6
index 0f54730c7ed56..98828030b5a4d 100644
@@ -76,12 +76,12 @@ static void firmware_load(const struct firmware *fw, void *context)
 
 	if (fw == NULL) {
 		dev_err(&spi->dev, "Cannot load firmware, aborting\n");
-		return;
+		goto out;
 	}
 
 	if (fw->size == 0) {
 		dev_err(&spi->dev, "Error: Firmware size is 0!\n");
-		return;
+		goto out;
 	}
 
 	/* Fill dummy data (24 stuffing bits for commands) */
@@ -103,7 +103,7 @@ static void firmware_load(const struct firmware *fw, void *context)
 		dev_err(&spi->dev,
 			"Error: No supported FPGA detected (JEDEC_ID=%08x)!\n",
 			jedec_id);
-		return;
+		goto out;
 	}
 
 	dev_info(&spi->dev, "FPGA %s detected\n", ecp3_dev[i].name);
@@ -116,7 +116,7 @@ static void firmware_load(const struct firmware *fw, void *context)
 	buffer = kzalloc(fw->size + 8, GFP_KERNEL);
 	if (!buffer) {
 		dev_err(&spi->dev, "Error: Can't allocate memory!\n");
-		return;
+		goto out;
 	}
 
 	/*
@@ -155,7 +155,7 @@ static void firmware_load(const struct firmware *fw, void *context)
 			"Error: Timeout waiting for FPGA to clear (status=%08x)!\n",
 			status);
 		kfree(buffer);
-		return;
+		goto out;
 	}
 
 	dev_info(&spi->dev, "Configuring the FPGA...\n");
@@ -181,7 +181,7 @@ static void firmware_load(const struct firmware *fw, void *context)
 	release_firmware(fw);
 
 	kfree(buffer);
-
+out:
 	complete(&data->fw_loaded);
 }
 
Signature · 1 line
2.34.1
author maintainer
#513 lore raw

PATCH 513/1039 ASoC: mediatek: mt8195: correct pcmif BE dai control flow

In reply to Greg Kroah-Hartman

From: Trevor Wu <[email protected]>

[ Upstream commit 2355028c0c54c03afb66c589347f1dc9f6fe2e38 ]

Originally, the conditions for preventing reentry are not correct. dai->component->active is not the state specifically for pcmif dai, so it is not a correct condition to indicate the status of pcmif dai. On the other hand, snd_soc_dai_stream_actvie() in prepare ops for both playback and capture possibly return true at the first entry when these two streams are opened at the same time.

In the patch, I refer to the implementation in mt8192-dai-pcm.c. Clock and enabling bit for PCMIF are managed by DAPM, and the condition for prepare ops is replaced by the status of dai widget.

Fixes 1f95c019115c ("ASoC: mediatek: mt8195: support pcm in platform driver")
---
 sound/soc/mediatek/mt8195/mt8195-dai-pcm.c | 73 +++++++---------------
 sound/soc/mediatek/mt8195/mt8195-reg.h     |  1 +
 2 files changed, 22 insertions(+), 52 deletions(-)
sound/soc/mediatek/mt8195/mt8195-dai-pcm.c +21 -52
index 5d10d2c4c991c..151914c873acd 100644
@@ -80,8 +80,15 @@ static const struct snd_soc_dapm_widget mtk_dai_pcm_widgets[] = {
 			   mtk_dai_pcm_o001_mix,
 			   ARRAY_SIZE(mtk_dai_pcm_o001_mix)),
 
+	SND_SOC_DAPM_SUPPLY("PCM_EN", PCM_INTF_CON1,
+			    PCM_INTF_CON1_PCM_EN_SHIFT, 0, NULL, 0),
+
 	SND_SOC_DAPM_INPUT("PCM1_INPUT"),
 	SND_SOC_DAPM_OUTPUT("PCM1_OUTPUT"),
+
+	SND_SOC_DAPM_CLOCK_SUPPLY("aud_asrc11"),
+	SND_SOC_DAPM_CLOCK_SUPPLY("aud_asrc12"),
+	SND_SOC_DAPM_CLOCK_SUPPLY("aud_pcmif"),
 };
 
 static const struct snd_soc_dapm_route mtk_dai_pcm_routes[] = {
@@ -97,22 +104,18 @@ static const struct snd_soc_dapm_route mtk_dai_pcm_routes[] = {
 	{"PCM1 Playback", NULL, "O000"},
 	{"PCM1 Playback", NULL, "O001"},
 
+	{"PCM1 Playback", NULL, "PCM_EN"},
+	{"PCM1 Playback", NULL, "aud_asrc12"},
+	{"PCM1 Playback", NULL, "aud_pcmif"},
+
+	{"PCM1 Capture", NULL, "PCM_EN"},
+	{"PCM1 Capture", NULL, "aud_asrc11"},
+	{"PCM1 Capture", NULL, "aud_pcmif"},
+
 	{"PCM1_OUTPUT", NULL, "PCM1 Playback"},
 	{"PCM1 Capture", NULL, "PCM1_INPUT"},
 };
 
-static void mtk_dai_pcm_enable(struct mtk_base_afe *afe)
-{
-	regmap_update_bits(afe->regmap, PCM_INTF_CON1,
-			   PCM_INTF_CON1_PCM_EN, PCM_INTF_CON1_PCM_EN);
-}
-
-static void mtk_dai_pcm_disable(struct mtk_base_afe *afe)
-{
-	regmap_update_bits(afe->regmap, PCM_INTF_CON1,
-			   PCM_INTF_CON1_PCM_EN, 0x0);
-}
-
 static int mtk_dai_pcm_configure(struct snd_pcm_substream *substream,
 				 struct snd_soc_dai *dai)
 {
@@ -207,54 +210,22 @@ static int mtk_dai_pcm_configure(struct snd_pcm_substream *substream,
 }
 
 /* dai ops */
-static int mtk_dai_pcm_startup(struct snd_pcm_substream *substream,
-			       struct snd_soc_dai *dai)
-{
-	struct mtk_base_afe *afe = snd_soc_dai_get_drvdata(dai);
-	struct mt8195_afe_private *afe_priv = afe->platform_priv;
-
-	if (dai->component->active)
-		return 0;
-
-	mt8195_afe_enable_clk(afe, afe_priv->clk[MT8195_CLK_AUD_ASRC11]);
-	mt8195_afe_enable_clk(afe, afe_priv->clk[MT8195_CLK_AUD_ASRC12]);
-	mt8195_afe_enable_clk(afe, afe_priv->clk[MT8195_CLK_AUD_PCMIF]);
-
-	return 0;
-}
-
-static void mtk_dai_pcm_shutdown(struct snd_pcm_substream *substream,
-				 struct snd_soc_dai *dai)
-{
-	struct mtk_base_afe *afe = snd_soc_dai_get_drvdata(dai);
-	struct mt8195_afe_private *afe_priv = afe->platform_priv;
-
-	if (dai->component->active)
-		return;
-
-	mtk_dai_pcm_disable(afe);
-
-	mt8195_afe_disable_clk(afe, afe_priv->clk[MT8195_CLK_AUD_PCMIF]);
-	mt8195_afe_disable_clk(afe, afe_priv->clk[MT8195_CLK_AUD_ASRC12]);
-	mt8195_afe_disable_clk(afe, afe_priv->clk[MT8195_CLK_AUD_ASRC11]);
-}
-
 static int mtk_dai_pcm_prepare(struct snd_pcm_substream *substream,
 			       struct snd_soc_dai *dai)
 {
-	struct mtk_base_afe *afe = snd_soc_dai_get_drvdata(dai);
-	int ret = 0;
+	int ret;
 
-	if (snd_soc_dai_stream_active(dai, SNDRV_PCM_STREAM_PLAYBACK) &&
-	    snd_soc_dai_stream_active(dai, SNDRV_PCM_STREAM_CAPTURE))
+	dev_dbg(dai->dev, "%s(), id %d, stream %d, widget active p %d, c %d\n",
+		__func__, dai->id, substream->stream,
+		dai->playback_widget->active, dai->capture_widget->active);
+
+	if (dai->playback_widget->active || dai->capture_widget->active)
 		return 0;
 
 	ret = mtk_dai_pcm_configure(substream, dai);
 	if (ret)
 		return ret;
 
-	mtk_dai_pcm_enable(afe);
-
 	return 0;
 }
 
@@ -316,8 +287,6 @@ static int mtk_dai_pcm_set_fmt(struct snd_soc_dai *dai, unsigned int fmt)
 }
 
 static const struct snd_soc_dai_ops mtk_dai_pcm_ops = {
-	.startup	= mtk_dai_pcm_startup,
-	.shutdown	= mtk_dai_pcm_shutdown,
 	.prepare	= mtk_dai_pcm_prepare,
 	.set_fmt	= mtk_dai_pcm_set_fmt,
 };
sound/soc/mediatek/mt8195/mt8195-reg.h +1 -0
index d06f9cf85a4ec..d3871353db415 100644
@@ -2550,6 +2550,7 @@
 #define PCM_INTF_CON1_PCM_FMT(x)       (((x) & 0x3) << 1)
 #define PCM_INTF_CON1_PCM_FMT_MASK     (0x3 << 1)
 #define PCM_INTF_CON1_PCM_EN           BIT(0)
+#define PCM_INTF_CON1_PCM_EN_SHIFT     0
 
 /* PCM_INTF_CON2 */
 #define PCM_INTF_CON2_CLK_DOMAIN_SEL(x)   (((x) & 0x3) << 23)
Signature · 1 line
2.34.1
author maintainer
#514 lore raw

PATCH 514/1039 arm64: tegra: Remove non existent Tegra194 reset

In reply to Greg Kroah-Hartman

From: Sameer Pujar <[email protected]>

[ Upstream commit 146b3a77af8091cabbd1decc51d67799e69682d2 ]

Tegra194 does not really have "hda2codec_2x" related reset. Hence drop this entry to reflect actual HW.

Fixes 4878cc0c9fab ("arm64: tegra: Add HDA controller on Tegra194")
---
 arch/arm64/boot/dts/nvidia/tegra194.dtsi | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
arch/arm64/boot/dts/nvidia/tegra194.dtsi +2 -3
index 851e049b3519c..dcc0e55d6bdbb 100644
@@ -976,9 +976,8 @@
 				 <&bpmp TEGRA194_CLK_HDA2CODEC_2X>;
 			clock-names = "hda", "hda2hdmi", "hda2codec_2x";
 			resets = <&bpmp TEGRA194_RESET_HDA>,
-				 <&bpmp TEGRA194_RESET_HDA2HDMICODEC>,
-				 <&bpmp TEGRA194_RESET_HDA2CODEC_2X>;
-			reset-names = "hda", "hda2hdmi", "hda2codec_2x";
+				 <&bpmp TEGRA194_RESET_HDA2HDMICODEC>;
+			reset-names = "hda", "hda2hdmi";
 			power-domains = <&bpmp TEGRA194_POWER_DOMAIN_DISP>;
 			interconnects = <&mc TEGRA194_MEMORY_CLIENT_HDAR &emc>,
 					<&mc TEGRA194_MEMORY_CLIENT_HDAW &emc>;
Signature · 1 line
2.34.1
author maintainer
#515 lore raw

PATCH 515/1039 mips: lantiq: add support for clk_set_parent()

In reply to Greg Kroah-Hartman

From: Randy Dunlap <[email protected]>

[ Upstream commit 76f66dfd60dc5d2f9dec22d99091fea1035c5d03 ]

Provide a simple implementation of clk_set_parent() in the lantiq subarch so that callers of it will build without errors.

Fixes these build errors:

ERROR: modpost: "clk_set_parent" [sound/soc/jz4740/snd-soc-jz4740-i2s.ko] undefined!
ERROR: modpost: "clk_set_parent" [sound/soc/atmel/snd-soc-atmel-i2s.ko] undefined!

Fixes 171bb2f19ed6 ("MIPS: Lantiq: Add initial support for Lantiq SoCs")

[email protected] --cc="John Crispin <[email protected]>" --cc="Jonathan Cameron <[email protected]>" --cc="Russell King <[email protected]>" --cc="Andy Shevchenko <[email protected]>" [email protected] --to="Thomas Bogendoerfer <[email protected]>"

---
 arch/mips/lantiq/clk.c | 6 ++++++
 1 file changed, 6 insertions(+)
arch/mips/lantiq/clk.c +6 -0
index 4916cccf378fd..7a623684d9b5e 100644
@@ -164,6 +164,12 @@ struct clk *clk_get_parent(struct clk *clk)
 }
 EXPORT_SYMBOL(clk_get_parent);
 
+int clk_set_parent(struct clk *clk, struct clk *parent)
+{
+	return 0;
+}
+EXPORT_SYMBOL(clk_set_parent);
+
 static inline u32 get_counter_resolution(void)
 {
 	u32 res;
Signature · 1 line
2.34.1
author maintainer
#516 lore raw

PATCH 516/1039 mips: bcm63xx: add support for clk_set_parent()

In reply to Greg Kroah-Hartman

From: Randy Dunlap <[email protected]>

[ Upstream commit 6f03055d508ff4feb8db02ba3df9303a1db8d381 ]

The MIPS BMC63XX subarch does not provide/support clk_set_parent(). This causes build errors in a few drivers, so add a simple implementation of that function so that callers of it will build without errors.

Fixes these build errors:

ERROR: modpost: "clk_set_parent" [sound/soc/jz4740/snd-soc-jz4740-i2s.ko] undefined!
ERROR: modpost: "clk_set_parent" [sound/soc/atmel/snd-soc-atmel-i2s.ko] undefined!

Fixes e7300d04bd08 ("MIPS: BCM63xx: Add support for the Broadcom BCM63xx family of SOCs." )
---
 arch/mips/bcm63xx/clk.c | 6 ++++++
 1 file changed, 6 insertions(+)
arch/mips/bcm63xx/clk.c +6 -0
index 1c91064cb448b..6e6756e8fa0a9 100644
@@ -387,6 +387,12 @@ struct clk *clk_get_parent(struct clk *clk)
 }
 EXPORT_SYMBOL(clk_get_parent);
 
+int clk_set_parent(struct clk *clk, struct clk *parent)
+{
+	return 0;
+}
+EXPORT_SYMBOL(clk_set_parent);
+
 unsigned long clk_get_rate(struct clk *clk)
 {
 	if (!clk)
Signature · 1 line
2.34.1
author maintainer
#517 lore raw

PATCH 517/1039 powerpc/xive: Add missing null check after calling kmalloc

In reply to Greg Kroah-Hartman

From: Ammar Faizi <[email protected]>

[ Upstream commit 18dbfcdedc802f9500b2c29794f22a31d27639c0 ]

Commit 930914b7d528fc ("powerpc/xive: Add a debugfs file to dump internal XIVE state") forgot to add a null check.

Add it.

Fixes 930914b7d528fc6b0249bffc00564100bcf6ef75 ("powerpc/xive: Add a debugfs file to dump internal XIVE state")
---
 arch/powerpc/sysdev/xive/spapr.c | 3 +++
 1 file changed, 3 insertions(+)
arch/powerpc/sysdev/xive/spapr.c +3 -0
index f143b6f111ac0..1179632560b8d 100644
@@ -653,6 +653,9 @@ static int xive_spapr_debug_show(struct seq_file *m, void *private)
 	struct xive_irq_bitmap *xibm;
 	char *buf = kmalloc(PAGE_SIZE, GFP_KERNEL);
 
+	if (!buf)
+		return -ENOMEM;
+
 	list_for_each_entry(xibm, &xive_irq_bitmaps, list) {
 		memset(buf, 0, PAGE_SIZE);
 		bitmap_print_to_pagebuf(true, buf, xibm->bitmap, xibm->count);
Signature · 1 line
2.34.1
author maintainer
#518 lore raw

PATCH 518/1039 ASoC: fsl_mqs: fix MODULE_ALIAS

In reply to Greg Kroah-Hartman

From: Alyssa Ross <[email protected]>

[ Upstream commit 9f3d45318dd9e739ed62e4218839a7a824d3cced ]

modprobe can't handle spaces in aliases.

Fixes 9e28f6532c61 ("ASoC: fsl_mqs: Add MQS component driver")
---
 sound/soc/fsl/fsl_mqs.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
sound/soc/fsl/fsl_mqs.c +1 -1
index 27b4536dce443..ceaecbe3a25e4 100644
@@ -337,4 +337,4 @@ module_platform_driver(fsl_mqs_driver);
 MODULE_AUTHOR("Shengjiu Wang <[email protected]>");
 MODULE_DESCRIPTION("MQS codec driver");
 MODULE_LICENSE("GPL v2");
-MODULE_ALIAS("platform: fsl-mqs");
+MODULE_ALIAS("platform:fsl-mqs");
Signature · 1 line
2.34.1
author maintainer
#519 lore raw

PATCH 519/1039 ALSA: hda/cs8409: Increase delay during jack detection

In reply to Greg Kroah-Hartman

From: Christian A. Ehrhardt <[email protected]>

[ Upstream commit 8cd07657177006b67cc1610e4466cc75ad781c05 ]

Commit c8b4f0865e82 reduced delays related to cs42l42 jack detection. However, the change was too aggressive. As a result internal speakers on DELL Inspirion 3501 are not detected.

Increase the delay in cs42l42_run_jack_detect() a bit.

Fixes c8b4f0865e82 ("ALSA: hda/cs8409: Remove unnecessary delays")
---
 sound/pci/hda/patch_cs8409.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
sound/pci/hda/patch_cs8409.c +2 -2
index 039b9f2f8e947..bf5d7f0c6ba55 100644
@@ -628,8 +628,8 @@ static void cs42l42_run_jack_detect(struct sub_codec *cs42l42)
 	cs8409_i2c_write(cs42l42, 0x1b74, 0x07);
 	cs8409_i2c_write(cs42l42, 0x131b, 0xFD);
 	cs8409_i2c_write(cs42l42, 0x1120, 0x80);
-	/* Wait ~100us*/
-	usleep_range(100, 200);
+	/* Wait ~20ms*/
+	usleep_range(20000, 25000);
 	cs8409_i2c_write(cs42l42, 0x111f, 0x77);
 	cs8409_i2c_write(cs42l42, 0x1120, 0xc0);
 }
Signature · 1 line
2.34.1
author maintainer
#520 lore raw

PATCH 520/1039 ALSA: hda/cs8409: Fix Jack detection after resume

In reply to Greg Kroah-Hartman

From: Christian A. Ehrhardt <[email protected]>

[ Upstream commit 57f234248ff925d88caedf4019ec84e6ecb83909 ]

The suspend code unconditionally sets ->hp_jack_in and ->mic_jack_in to zero but without reporting this status change to the HDA core. To compensate for this, always assume a status change on the first unsol event after boot or resume.

Fixes 424e531b47f8 ("ALSA: hda/cs8409: Ensure Type Detection is only run on startup when necessary")
---
 sound/pci/hda/patch_cs8409-tables.c | 3 +++
 sound/pci/hda/patch_cs8409.c        | 5 ++++-
 sound/pci/hda/patch_cs8409.h        | 1 +
 3 files changed, 8 insertions(+), 1 deletion(-)
sound/pci/hda/patch_cs8409-tables.c +3 -0
index 0fb0a428428b4..df0b4522babf7 100644
@@ -252,6 +252,7 @@ struct sub_codec cs8409_cs42l42_codec = {
 	.init_seq_num = ARRAY_SIZE(cs42l42_init_reg_seq),
 	.hp_jack_in = 0,
 	.mic_jack_in = 0,
+	.force_status_change = 1,
 	.paged = 1,
 	.suspended = 1,
 	.no_type_dect = 0,
@@ -443,6 +444,7 @@ struct sub_codec dolphin_cs42l42_0 = {
 	.init_seq_num = ARRAY_SIZE(dolphin_c0_init_reg_seq),
 	.hp_jack_in = 0,
 	.mic_jack_in = 0,
+	.force_status_change = 1,
 	.paged = 1,
 	.suspended = 1,
 	.no_type_dect = 0,
@@ -456,6 +458,7 @@ struct sub_codec dolphin_cs42l42_1 = {
 	.init_seq_num = ARRAY_SIZE(dolphin_c1_init_reg_seq),
 	.hp_jack_in = 0,
 	.mic_jack_in = 0,
+	.force_status_change = 1,
 	.paged = 1,
 	.suspended = 1,
 	.no_type_dect = 1,
sound/pci/hda/patch_cs8409.c +4 -1
index bf5d7f0c6ba55..aff2b5abb81ea 100644
@@ -636,7 +636,9 @@ static void cs42l42_run_jack_detect(struct sub_codec *cs42l42)
 
 static int cs42l42_handle_tip_sense(struct sub_codec *cs42l42, unsigned int reg_ts_status)
 {
-	int status_changed = 0;
+	int status_changed = cs42l42->force_status_change;
+
+	cs42l42->force_status_change = 0;
 
 	/* TIP_SENSE INSERT/REMOVE */
 	switch (reg_ts_status) {
@@ -791,6 +793,7 @@ static void cs42l42_suspend(struct sub_codec *cs42l42)
 	cs42l42->last_page = 0;
 	cs42l42->hp_jack_in = 0;
 	cs42l42->mic_jack_in = 0;
+	cs42l42->force_status_change = 1;
 
 	/* Put CS42L42 into Reset */
 	gpio_data = snd_hda_codec_read(codec, CS8409_PIN_AFG, 0, AC_VERB_GET_GPIO_DATA, 0);
sound/pci/hda/patch_cs8409.h +1 -0
index ade2b838590cf..d0b725c7285b6 100644
@@ -305,6 +305,7 @@ struct sub_codec {
 
 	unsigned int hp_jack_in:1;
 	unsigned int mic_jack_in:1;
+	unsigned int force_status_change:1;
 	unsigned int suspended:1;
 	unsigned int paged:1;
 	unsigned int last_page;
Signature · 1 line
2.34.1
author maintainer
#521 lore raw

PATCH 521/1039 cxl/core: Remove cxld_const_init in cxl_decoder_alloc()

In reply to Greg Kroah-Hartman

From: Nathan Chancellor <[email protected]>

[ Upstream commit be185c2988b48db65348d94168c793bdbc8d23c3 ]

Commit 48667f676189 ("cxl/core: Split decoder setup into alloc + add") aimed to fix a large stack frame warning but from v5 to v6, it introduced a new instance of the warning due to allocating cxld_const_init on the stack, which was done due to the use of const on the nr_target member of the cxl_decoder struct. With ARCH=arm allmodconfig minus CONFIG_KASAN:

GCC 11.2.0:

drivers/cxl/core/bus.c: In function ‘cxl_decoder_alloc’:
drivers/cxl/core/bus.c:523:1: error: the frame size of 1032 bytes is larger than 1024 bytes [-Werror=frame-larger-than=]
523 | }

      | ^

cc1: all warnings being treated as errors

Clang 12.0.1:

drivers/cxl/core/bus.c:486:21: error: stack frame size of 1056 bytes in function 'cxl_decoder_alloc' [-Werror,-Wframe-larger-than=] struct cxl_decoder *cxl_decoder_alloc(struct cxl_port *port, int nr_targets)

                    ^

1 error generated.

Revert that part of the change, which makes the stack frame of cxl_decoder_alloc() much more reasonable.

---
 drivers/cxl/core/bus.c | 6 ++----
 drivers/cxl/cxl.h      | 2 +-
 2 files changed, 3 insertions(+), 5 deletions(-)
drivers/cxl/core/bus.c +2 -4
index ebd061d039508..46ce58376580b 100644
@@ -485,9 +485,7 @@ out_unlock:
 
 struct cxl_decoder *cxl_decoder_alloc(struct cxl_port *port, int nr_targets)
 {
-	struct cxl_decoder *cxld, cxld_const_init = {
-		.nr_targets = nr_targets,
-	};
+	struct cxl_decoder *cxld;
 	struct device *dev;
 	int rc = 0;
 
@@ -497,13 +495,13 @@ struct cxl_decoder *cxl_decoder_alloc(struct cxl_port *port, int nr_targets)
 	cxld = kzalloc(struct_size(cxld, target, nr_targets), GFP_KERNEL);
 	if (!cxld)
 		return ERR_PTR(-ENOMEM);
-	memcpy(cxld, &cxld_const_init, sizeof(cxld_const_init));
 
 	rc = ida_alloc(&port->decoder_ida, GFP_KERNEL);
 	if (rc < 0)
 		goto err;
 
 	cxld->id = rc;
+	cxld->nr_targets = nr_targets;
 	dev = &cxld->dev;
 	device_initialize(dev);
 	device_set_pm_not_required(dev);
drivers/cxl/cxl.h +1 -1
index 3af704e9b448e..7c2b51746e318 100644
@@ -191,7 +191,7 @@ struct cxl_decoder {
 	int interleave_granularity;
 	enum cxl_decoder_type target_type;
 	unsigned long flags;
-	const int nr_targets;
+	int nr_targets;
 	struct cxl_dport *target[];
 };
 
Signature · 1 line
2.34.1
author maintainer
#522 lore raw

PATCH 522/1039 MIPS: fix local_{add,sub}_return on MIPS64

In reply to Greg Kroah-Hartman

From: Huang Pei <[email protected]>

[ Upstream commit 277c8cb3e8ac199f075bf9576ad286687ed17173 ]

Use "daddu/dsubu" for long int on MIPS64 instead of "addu/subu"

Fixes 7232311ef14c ("local_t: mips extension")
---
 arch/mips/include/asm/local.h | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)
arch/mips/include/asm/local.h +5 -4
index ecda7295ddcd1..3fa6340903882 100644
@@ -5,6 +5,7 @@
 #include <linux/percpu.h>
 #include <linux/bitops.h>
 #include <linux/atomic.h>
+#include <asm/asm.h>
 #include <asm/cmpxchg.h>
 #include <asm/compiler.h>
 #include <asm/war.h>
@@ -39,7 +40,7 @@ static __inline__ long local_add_return(long i, local_t * l)
 		"	.set	arch=r4000				\n"
 			__SYNC(full, loongson3_war) "			\n"
 		"1:"	__LL	"%1, %2		# local_add_return	\n"
-		"	addu	%0, %1, %3				\n"
+			__stringify(LONG_ADDU)	"	%0, %1, %3	\n"
 			__SC	"%0, %2					\n"
 		"	beqzl	%0, 1b					\n"
 		"	addu	%0, %1, %3				\n"
@@ -55,7 +56,7 @@ static __inline__ long local_add_return(long i, local_t * l)
 		"	.set	"MIPS_ISA_ARCH_LEVEL"			\n"
 			__SYNC(full, loongson3_war) "			\n"
 		"1:"	__LL	"%1, %2		# local_add_return	\n"
-		"	addu	%0, %1, %3				\n"
+			__stringify(LONG_ADDU)	"	%0, %1, %3	\n"
 			__SC	"%0, %2					\n"
 		"	beqz	%0, 1b					\n"
 		"	addu	%0, %1, %3				\n"
@@ -88,7 +89,7 @@ static __inline__ long local_sub_return(long i, local_t * l)
 		"	.set	arch=r4000				\n"
 			__SYNC(full, loongson3_war) "			\n"
 		"1:"	__LL	"%1, %2		# local_sub_return	\n"
-		"	subu	%0, %1, %3				\n"
+			__stringify(LONG_SUBU)	"	%0, %1, %3	\n"
 			__SC	"%0, %2					\n"
 		"	beqzl	%0, 1b					\n"
 		"	subu	%0, %1, %3				\n"
@@ -104,7 +105,7 @@ static __inline__ long local_sub_return(long i, local_t * l)
 		"	.set	"MIPS_ISA_ARCH_LEVEL"			\n"
 			__SYNC(full, loongson3_war) "			\n"
 		"1:"	__LL	"%1, %2		# local_sub_return	\n"
-		"	subu	%0, %1, %3				\n"
+			__stringify(LONG_SUBU)	"	%0, %1, %3	\n"
 			__SC	"%0, %2					\n"
 		"	beqz	%0, 1b					\n"
 		"	subu	%0, %1, %3				\n"
Signature · 1 line
2.34.1
author maintainer
#523 lore raw

PATCH 523/1039 RDMA/cxgb4: Set queue pair state when being queried

In reply to Greg Kroah-Hartman

From: Kamal Heib <[email protected]>

[ Upstream commit e375b9c92985e409c4bb95dd43d34915ea7f5e28 ]

The API for ib_query_qp requires the driver to set cur_qp_state on return, add the missing set.

---
 drivers/infiniband/hw/cxgb4/qp.c | 1 +
 1 file changed, 1 insertion(+)
drivers/infiniband/hw/cxgb4/qp.c +1 -0
index d20b4ef2c853d..ffbd9a89981e7 100644
@@ -2460,6 +2460,7 @@ int c4iw_ib_query_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 	memset(attr, 0, sizeof(*attr));
 	memset(init_attr, 0, sizeof(*init_attr));
 	attr->qp_state = to_ib_qp_state(qhp->attr.state);
+	attr->cur_qp_state = to_ib_qp_state(qhp->attr.state);
 	init_attr->cap.max_send_wr = qhp->attr.sq_num_entries;
 	init_attr->cap.max_recv_wr = qhp->attr.rq_num_entries;
 	init_attr->cap.max_send_sge = qhp->attr.sq_max_sges;
Signature · 1 line
2.34.1
author maintainer
#524 lore raw

PATCH 524/1039 clk: qcom: gcc-sc7280: Mark gcc_cfg_noc_lpass_clk always enabled

In reply to Greg Kroah-Hartman

From: Taniya Das <[email protected]>

[ Upstream commit 9c337073d9d81a145434b22f42dc3128ecd17730 ]

The gcc cfg noc lpass clock is required to be always enabled for the LPASS core and audio drivers to be functional.

Fixes a3cc092196ef ("clk: qcom: Add Global Clock controller (GCC) driver for SC7280")
---
 drivers/clk/qcom/gcc-sc7280.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/clk/qcom/gcc-sc7280.c +1 -1
index 8fb6bd69f240e..423627d49719c 100644
@@ -2917,7 +2917,7 @@ static struct clk_branch gcc_cfg_noc_lpass_clk = {
 		.enable_mask = BIT(0),
 		.hw.init = &(struct clk_init_data){
 			.name = "gcc_cfg_noc_lpass_clk",
-			.ops = &clk_branch2_ops,
+			.ops = &clk_branch2_aon_ops,
 		},
 	},
 };
Signature · 1 line
2.34.1
author maintainer
#525 lore raw

PATCH 525/1039 ASoC: imx-card: Need special setting for ak4497 on i.MX8MQ

In reply to Greg Kroah-Hartman

From: Shengjiu Wang <[email protected]>

[ Upstream commit 3349b3d0c63b8b6fcca58156d72407f0b2e101ac ]

The SAI on i.MX8MQ don't support one2one ratio for mclk:bclk, so the mclk frequency exceeds the supported range of codec for the case that sample rate is larger than 705kHZ and format is S32_LE. Update the supported width for such case.

Fixes aa736700f42f ("ASoC: imx-card: Add imx-card machine driver")
---
 sound/soc/fsl/imx-card.c | 17 ++++++++++++++++-
 1 file changed, 16 insertions(+), 1 deletion(-)
sound/soc/fsl/imx-card.c +16 -1
index 6f06afd23b16a..96ac1e465fade 100644
@@ -553,8 +553,23 @@ static int imx_card_parse_of(struct imx_card_data *data)
 			link_data->cpu_sysclk_id = FSL_SAI_CLK_MAST1;
 
 			/* sai may support mclk/bclk = 1 */
-			if (of_find_property(np, "fsl,mclk-equal-bclk", NULL))
+			if (of_find_property(np, "fsl,mclk-equal-bclk", NULL)) {
 				link_data->one2one_ratio = true;
+			} else {
+				int i;
+
+				/*
+				 * i.MX8MQ don't support one2one ratio, then
+				 * with ak4497 only 16bit case is supported.
+				 */
+				for (i = 0; i < ARRAY_SIZE(ak4497_fs_mul); i++) {
+					if (ak4497_fs_mul[i].rmin == 705600 &&
+					    ak4497_fs_mul[i].rmax == 768000) {
+						ak4497_fs_mul[i].wmin = 32;
+						ak4497_fs_mul[i].wmax = 32;
+					}
+				}
+			}
 		}
 
 		link->cpus->of_node = args.np;
Signature · 1 line
2.34.1
author maintainer
#526 lore raw

PATCH 526/1039 ASoC: imx-card: Fix mclk calculation issue for akcodec

In reply to Greg Kroah-Hartman

From: Shengjiu Wang <[email protected]>

[ Upstream commit f331ae5fa59fbfb748317b290648fc3f1a50d932 ]

Transfer the refined slots and slot_width to akcodec_get_mclk_rate() for mclk calculation, otherwise the mclk frequency does not match with the slots and slot_width for S16_LE format, because the default slot_width is 32.

Fixes aa736700f42f ("ASoC: imx-card: Add imx-card machine driver")
---
 sound/soc/fsl/imx-card.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
sound/soc/fsl/imx-card.c +4 -3
index 96ac1e465fade..84a8dcdee9316 100644
@@ -247,13 +247,14 @@ static bool codec_is_akcodec(unsigned int type)
 }
 
 static unsigned long akcodec_get_mclk_rate(struct snd_pcm_substream *substream,
-					   struct snd_pcm_hw_params *params)
+					   struct snd_pcm_hw_params *params,
+					   int slots, int slot_width)
 {
 	struct snd_soc_pcm_runtime *rtd = substream->private_data;
 	struct imx_card_data *data = snd_soc_card_get_drvdata(rtd->card);
 	const struct imx_card_plat_data *plat_data = data->plat_data;
 	struct dai_link_data *link_data = &data->link_data[rtd->num];
-	unsigned int width = link_data->slots * link_data->slot_width;
+	unsigned int width = slots * slot_width;
 	unsigned int rate = params_rate(params);
 	int i;
 
@@ -349,7 +350,7 @@ static int imx_aif_hw_params(struct snd_pcm_substream *substream,
 
 	/* Set MCLK freq */
 	if (codec_is_akcodec(plat_data->type))
-		mclk_freq = akcodec_get_mclk_rate(substream, params);
+		mclk_freq = akcodec_get_mclk_rate(substream, params, slots, slot_width);
 	else
 		mclk_freq = params_rate(params) * slots * slot_width;
 	/* Use the maximum freq from DSD512 (512*44100 = 22579200) */
Signature · 1 line
2.34.1
author maintainer
#527 lore raw

PATCH 527/1039 ASoC: imx-card: improve the sound quality for low rate

In reply to Greg Kroah-Hartman

From: Shengjiu Wang <[email protected]>

[ Upstream commit 3969341813eb56d2dfc39bb64229359a6ae3c195 ]

According to RM, on auto mode:
For codec AK4458 and AK4497, the lowest ratio of MLCK/FS is 256 if sample rate is 8kHz-48kHz,
For codec AK5558, the lowest ratio of MLCK/FS is 512 if sample rate is 8kHz-48kHz.

With these setting the sound quality for 8kHz-48kHz can be improved.

Fixes aa736700f42f ("ASoC: imx-card: Add imx-card machine driver")
---
 sound/soc/fsl/imx-card.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
sound/soc/fsl/imx-card.c +4 -4
index 84a8dcdee9316..c7503a5f7cfbc 100644
@@ -120,7 +120,7 @@ struct imx_card_data {
 
 static struct imx_akcodec_fs_mul ak4458_fs_mul[] = {
 	/* Normal, < 32kHz */
-	{ .rmin = 8000,   .rmax = 24000,  .wmin = 1024, .wmax = 1024, },
+	{ .rmin = 8000,   .rmax = 24000,  .wmin = 256,  .wmax = 1024, },
 	/* Normal, 32kHz */
 	{ .rmin = 32000,  .rmax = 32000,  .wmin = 256,  .wmax = 1024, },
 	/* Normal */
@@ -151,8 +151,8 @@ static struct imx_akcodec_fs_mul ak4497_fs_mul[] = {
 	 * Table 7      - mapping multiplier and speed mode
 	 * Tables 8 & 9 - mapping speed mode and LRCK fs
 	 */
-	{ .rmin = 8000,   .rmax = 32000,  .wmin = 1024, .wmax = 1024, }, /* Normal, <= 32kHz */
-	{ .rmin = 44100,  .rmax = 48000,  .wmin = 512,  .wmax = 512, }, /* Normal */
+	{ .rmin = 8000,   .rmax = 32000,  .wmin = 256,  .wmax = 1024, }, /* Normal, <= 32kHz */
+	{ .rmin = 44100,  .rmax = 48000,  .wmin = 256,  .wmax = 512, }, /* Normal */
 	{ .rmin = 88200,  .rmax = 96000,  .wmin = 256,  .wmax = 256, }, /* Double */
 	{ .rmin = 176400, .rmax = 192000, .wmin = 128,  .wmax = 128, }, /* Quad */
 	{ .rmin = 352800, .rmax = 384000, .wmin = 128,  .wmax = 128, }, /* Oct */
@@ -164,7 +164,7 @@ static struct imx_akcodec_fs_mul ak4497_fs_mul[] = {
  * (Table 4 from datasheet)
  */
 static struct imx_akcodec_fs_mul ak5558_fs_mul[] = {
-	{ .rmin = 8000,   .rmax = 32000,  .wmin = 1024, .wmax = 1024, },
+	{ .rmin = 8000,   .rmax = 32000,  .wmin = 512,  .wmax = 1024, },
 	{ .rmin = 44100,  .rmax = 48000,  .wmin = 512,  .wmax = 512, },
 	{ .rmin = 88200,  .rmax = 96000,  .wmin = 256,  .wmax = 256, },
 	{ .rmin = 176400, .rmax = 192000, .wmin = 128,  .wmax = 128, },
Signature · 1 line
2.34.1
author maintainer
#528 lore raw

PATCH 528/1039 ASoC: fsl_asrc: refine the check of available clock divider

In reply to Greg Kroah-Hartman

From: Shengjiu Wang <[email protected]>

[ Upstream commit 320386343451ab6a3577e0ee200dac56a6182944 ]

According to RM, the clock divider range is from 1 to 8, clock prescaling ratio may be any power of 2 from 1 to 128. So the supported divider is not all the value between 1 and 1024, just limited value in that range.

Create table for the supported divder and add function to check the clock divider is available by comparing with the table.

Fixes d0250cf4f2ab ("ASoC: fsl_asrc: Add an option to select internal ratio mode")
---
 sound/soc/fsl/fsl_asrc.c | 69 +++++++++++++++++++++++++++++++++-------
 1 file changed, 58 insertions(+), 11 deletions(-)
sound/soc/fsl/fsl_asrc.c +58 -11
index 24b41881a68f8..d7d1536a4f377 100644
@@ -19,6 +19,7 @@
 #include "fsl_asrc.h"
 
 #define IDEAL_RATIO_DECIMAL_DEPTH 26
+#define DIVIDER_NUM  64
 
 #define pair_err(fmt, ...) \
 	dev_err(&asrc->pdev->dev, "Pair %c: " fmt, 'A' + index, ##__VA_ARGS__)
@@ -101,6 +102,55 @@ static unsigned char clk_map_imx8qxp[2][ASRC_CLK_MAP_LEN] = {
 	},
 };
 
+/*
+ * According to RM, the divider range is 1 ~ 8,
+ * prescaler is power of 2 from 1 ~ 128.
+ */
+static int asrc_clk_divider[DIVIDER_NUM] = {
+	1,  2,  4,  8,  16,  32,  64,  128,  /* divider = 1 */
+	2,  4,  8, 16,  32,  64, 128,  256,  /* divider = 2 */
+	3,  6, 12, 24,  48,  96, 192,  384,  /* divider = 3 */
+	4,  8, 16, 32,  64, 128, 256,  512,  /* divider = 4 */
+	5, 10, 20, 40,  80, 160, 320,  640,  /* divider = 5 */
+	6, 12, 24, 48,  96, 192, 384,  768,  /* divider = 6 */
+	7, 14, 28, 56, 112, 224, 448,  896,  /* divider = 7 */
+	8, 16, 32, 64, 128, 256, 512, 1024,  /* divider = 8 */
+};
+
+/*
+ * Check if the divider is available for internal ratio mode
+ */
+static bool fsl_asrc_divider_avail(int clk_rate, int rate, int *div)
+{
+	u32 rem, i;
+	u64 n;
+
+	if (div)
+		*div = 0;
+
+	if (clk_rate == 0 || rate == 0)
+		return false;
+
+	n = clk_rate;
+	rem = do_div(n, rate);
+
+	if (div)
+		*div = n;
+
+	if (rem != 0)
+		return false;
+
+	for (i = 0; i < DIVIDER_NUM; i++) {
+		if (n == asrc_clk_divider[i])
+			break;
+	}
+
+	if (i == DIVIDER_NUM)
+		return false;
+
+	return true;
+}
+
 /**
  * fsl_asrc_sel_proc - Select the pre-processing and post-processing options
  * @inrate: input sample rate
@@ -330,12 +380,12 @@ static int fsl_asrc_config_pair(struct fsl_asrc_pair *pair, bool use_ideal_rate)
 	enum asrc_word_width input_word_width;
 	enum asrc_word_width output_word_width;
 	u32 inrate, outrate, indiv, outdiv;
-	u32 clk_index[2], div[2], rem[2];
+	u32 clk_index[2], div[2];
 	u64 clk_rate;
 	int in, out, channels;
 	int pre_proc, post_proc;
 	struct clk *clk;
-	bool ideal;
+	bool ideal, div_avail;
 
 	if (!config) {
 		pair_err("invalid pair config\n");
@@ -415,8 +465,7 @@ static int fsl_asrc_config_pair(struct fsl_asrc_pair *pair, bool use_ideal_rate)
 	clk = asrc_priv->asrck_clk[clk_index[ideal ? OUT : IN]];
 
 	clk_rate = clk_get_rate(clk);
-	rem[IN] = do_div(clk_rate, inrate);
-	div[IN] = (u32)clk_rate;
+	div_avail = fsl_asrc_divider_avail(clk_rate, inrate, &div[IN]);
 
 	/*
 	 * The divider range is [1, 1024], defined by the hardware. For non-
@@ -425,7 +474,7 @@ static int fsl_asrc_config_pair(struct fsl_asrc_pair *pair, bool use_ideal_rate)
 	 * only result in different converting speeds. So remainder does not
 	 * matter, as long as we keep the divider within its valid range.
 	 */
-	if (div[IN] == 0 || (!ideal && (div[IN] > 1024 || rem[IN] != 0))) {
+	if (div[IN] == 0 || (!ideal && !div_avail)) {
 		pair_err("failed to support input sample rate %dHz by asrck_%x\n",
 				inrate, clk_index[ideal ? OUT : IN]);
 		return -EINVAL;
@@ -436,13 +485,12 @@ static int fsl_asrc_config_pair(struct fsl_asrc_pair *pair, bool use_ideal_rate)
 	clk = asrc_priv->asrck_clk[clk_index[OUT]];
 	clk_rate = clk_get_rate(clk);
 	if (ideal && use_ideal_rate)
-		rem[OUT] = do_div(clk_rate, IDEAL_RATIO_RATE);
+		div_avail = fsl_asrc_divider_avail(clk_rate, IDEAL_RATIO_RATE, &div[OUT]);
 	else
-		rem[OUT] = do_div(clk_rate, outrate);
-	div[OUT] = clk_rate;
+		div_avail = fsl_asrc_divider_avail(clk_rate, outrate, &div[OUT]);
 
 	/* Output divider has the same limitation as the input one */
-	if (div[OUT] == 0 || (!ideal && (div[OUT] > 1024 || rem[OUT] != 0))) {
+	if (div[OUT] == 0 || (!ideal && !div_avail)) {
 		pair_err("failed to support output sample rate %dHz by asrck_%x\n",
 				outrate, clk_index[OUT]);
 		return -EINVAL;
@@ -621,8 +669,7 @@ static void fsl_asrc_select_clk(struct fsl_asrc_priv *asrc_priv,
 			clk_index = asrc_priv->clk_map[j][i];
 			clk_rate = clk_get_rate(asrc_priv->asrck_clk[clk_index]);
 			/* Only match a perfect clock source with no remainder */
-			if (clk_rate != 0 && (clk_rate / rate[j]) <= 1024 &&
-			    (clk_rate % rate[j]) == 0)
+			if (fsl_asrc_divider_avail(clk_rate, rate[j], NULL))
 				break;
 		}
 
Signature · 1 line
2.34.1
author maintainer
#529 lore raw

PATCH 529/1039 clk: bm1880: remove kfrees on static allocations

In reply to Greg Kroah-Hartman

From: Conor Dooley <[email protected]>

[ Upstream commit c861c1be3897845313a0df47804b1db37c7052e1 ]

bm1880_clk_unregister_pll & bm1880_clk_unregister_div both try to free statically allocated variables, so remove those kfrees.

For example, if we take L703 kfree(div_hw):
- div_hw is a bm1880_div_hw_clock pointer
- in bm1880_clk_register_plls this is pointed to an element of arg1:
struct bm1880_div_hw_clock *clks
- in the probe, where bm1880_clk_register_plls is called arg1 is bm1880_div_clks, defined on L371:
static struct bm1880_div_hw_clock bm1880_div_clks[]

Fixes 1ab4601da55b ("clk: Add common clock driver for BM1880 SoC")
---
 drivers/clk/clk-bm1880.c | 20 ++------------------
 1 file changed, 2 insertions(+), 18 deletions(-)
drivers/clk/clk-bm1880.c +2 -18
index e6d6599d310a1..fad78a22218e8 100644
@@ -522,14 +522,6 @@ static struct clk_hw *bm1880_clk_register_pll(struct bm1880_pll_hw_clock *pll_cl
 	return hw;
 }
 
-static void bm1880_clk_unregister_pll(struct clk_hw *hw)
-{
-	struct bm1880_pll_hw_clock *pll_hw = to_bm1880_pll_clk(hw);
-
-	clk_hw_unregister(hw);
-	kfree(pll_hw);
-}
-
 static int bm1880_clk_register_plls(struct bm1880_pll_hw_clock *clks,
 				    int num_clks,
 				    struct bm1880_clock_data *data)
@@ -555,7 +547,7 @@ static int bm1880_clk_register_plls(struct bm1880_pll_hw_clock *clks,
 
 err_clk:
 	while (i--)
-		bm1880_clk_unregister_pll(data->hw_data.hws[clks[i].pll.id]);
+		clk_hw_unregister(data->hw_data.hws[clks[i].pll.id]);
 
 	return PTR_ERR(hw);
 }
@@ -695,14 +687,6 @@ static struct clk_hw *bm1880_clk_register_div(struct bm1880_div_hw_clock *div_cl
 	return hw;
 }
 
-static void bm1880_clk_unregister_div(struct clk_hw *hw)
-{
-	struct bm1880_div_hw_clock *div_hw = to_bm1880_div_clk(hw);
-
-	clk_hw_unregister(hw);
-	kfree(div_hw);
-}
-
 static int bm1880_clk_register_divs(struct bm1880_div_hw_clock *clks,
 				    int num_clks,
 				    struct bm1880_clock_data *data)
@@ -729,7 +713,7 @@ static int bm1880_clk_register_divs(struct bm1880_div_hw_clock *clks,
 
 err_clk:
 	while (i--)
-		bm1880_clk_unregister_div(data->hw_data.hws[clks[i].div.id]);
+		clk_hw_unregister(data->hw_data.hws[clks[i].div.id]);
 
 	return PTR_ERR(hw);
 }
Signature · 1 line
2.34.1
author maintainer
#530 lore raw

PATCH 530/1039 of: base: Fix phandle argument length mismatch error message

In reply to Greg Kroah-Hartman

From: Baruch Siach <[email protected]>

[ Upstream commit 94a4950a4acff39b5847cc1fee4f65e160813493 ]

The cell_count field of of_phandle_iterator is the number of cells we expect in the phandle arguments list when cells_name is missing. The error message should show the number of cells we actually see.

---
 drivers/of/base.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/of/base.c +2 -2
index 61de453b885cb..81b956ab2348e 100644
@@ -1349,9 +1349,9 @@ int of_phandle_iterator_next(struct of_phandle_iterator *it)
 		 * property data length
 		 */
 		if (it->cur + count > it->list_end) {
-			pr_err("%pOF: %s = %d found %d\n",
+			pr_err("%pOF: %s = %d found %td\n",
 			       it->parent, it->cells_name,
-			       count, it->cell_count);
+			       count, it->list_end - it->cur);
 			goto err;
 		}
 	}
Signature · 1 line
2.34.1
author maintainer
#531 lore raw

PATCH 531/1039 of/fdt: Dont worry about non-memory region overlap for no-map

In reply to Greg Kroah-Hartman

From: Stephen Boyd <[email protected]>

[ Upstream commit da17d6905d29ddcdc04b2fdc37ed8cf1e8437cc8 ]

In commit 8a5a75e5e9e5 ("of/fdt: Make sure no-map does not remove already reserved regions") we returned -EBUSY when trying to mark regions as no-map when they intersect with reserved memory. The goal was to find bad no-map reserved memory DT nodes that would unmap the kernel text/data sections.

The problem is the reserved memory check will still trigger if the DT has a /memreserve/ that completely subsumes the no-map memory carveouts in the reserved memory node _and_ that region is also not part of the memory reg property. For example in sc7180.dtsi we have the following reserved-memory and memory node:

      memory@80000000 {
          /* We expect the bootloader to fill in the size */
          reg = <0 0x80000000 0 0>;
      };

      smem_mem: memory@80900000 {
              reg = <0x0 0x80900000 0x0 0x200000>;
              no-map;
      };

and the memreserve filled in by the bootloader is

      /memreserve/ 0x80800000 0x400000;

while the /memory node is transformed into

      memory@80000000 {
          /* The bootloader fills in the size, and adds another region */
          reg = <0 0x80000000 0 0x00800000>,
                <0 0x80c00000 0 0x7f200000>;
      };

The smem region is doubly reserved via /memreserve/ and by not being part of the /memory reg property. This leads to the following warning printed at boot.

OF: fdt: Reserved memory: failed to reserve memory for node 'memory@80900000': base 0x0000000080900000, size 2 MiB

Otherwise nothing really goes wrong because the smem region is not going to be mapped by the kernel's direct linear mapping given that it isn't part of the memory node. Therefore, let's only consider this to be a problem if we're trying to mark a region as no-map and it is actually memory that we're intending to keep out of the kernel's direct mapping but it's already been reserved.

Cc Douglas Anderson <[email protected]>
Cc Nicolas Boichat <[email protected]>
Cc Quentin Perret <[email protected]>
Cc Jan Kiszka <[email protected]>
Fixes 8a5a75e5e9e5 ("of/fdt: Make sure no-map does not remove already reserved regions")
---
 drivers/of/fdt.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/of/fdt.c +4 -2
index 65af475dfa950..7e868e5995b7e 100644
@@ -482,9 +482,11 @@ static int __init early_init_dt_reserve_memory_arch(phys_addr_t base,
 	if (nomap) {
 		/*
 		 * If the memory is already reserved (by another region), we
-		 * should not allow it to be marked nomap.
+		 * should not allow it to be marked nomap, but don't worry
+		 * if the region isn't memory as it won't be mapped.
 		 */
-		if (memblock_is_region_reserved(base, size))
+		if (memblock_overlaps_region(&memblock.memory, base, size) &&
+		    memblock_is_region_reserved(base, size))
 			return -EBUSY;
 
 		return memblock_mark_nomap(base, size);
Signature · 1 line
2.34.1
author maintainer
#532 lore raw

PATCH 532/1039 MIPS: compressed: Fix build with ZSTD compression

In reply to Greg Kroah-Hartman

From: Paul Cercueil <[email protected]>

[ Upstream commit c5c7440fe7f74645940d5c9e2c49cd7efb706a4f ]

Fix the following build issues:

mips64el-linux-ld: arch/mips/boot/compressed/decompress.o: in function `FSE_buildDTable_internal':
decompress.c:(.text.FSE_buildDTable_internal+0x2cc): undefined reference to `__clzdi2'
mips64el-linux-ld: arch/mips/boot/compressed/decompress.o: in function `BIT_initDStream':
decompress.c:(.text.BIT_initDStream+0x7c): undefined reference to `__clzdi2'
mips64el-linux-ld: decompress.c:(.text.BIT_initDStream+0x158): undefined reference to `__clzdi2'
mips64el-linux-ld: arch/mips/boot/compressed/decompress.o: in function `ZSTD_buildFSETable_body_default.constprop.0':
decompress.c:(.text.ZSTD_buildFSETable_body_default.constprop.0+0x2a8): undefined reference to `__clzdi2'
mips64el-linux-ld: arch/mips/boot/compressed/decompress.o: in function `FSE_readNCount_body_default':
decompress.c:(.text.FSE_readNCount_body_default+0x130): undefined reference to `__ctzdi2'
mips64el-linux-ld: decompress.c:(.text.FSE_readNCount_body_default+0x1a4): undefined reference to `__ctzdi2'
mips64el-linux-ld: decompress.c:(.text.FSE_readNCount_body_default+0x2e4): undefined reference to `__clzdi2'
mips64el-linux-ld: arch/mips/boot/compressed/decompress.o: in function `HUF_readStats_body_default':
decompress.c:(.text.HUF_readStats_body_default+0x184): undefined reference to `__clzdi2'
mips64el-linux-ld: decompress.c:(.text.HUF_readStats_body_default+0x1b4): undefined reference to `__clzdi2'
mips64el-linux-ld: arch/mips/boot/compressed/decompress.o: in function `ZSTD_DCtx_getParameter':
decompress.c:(.text.ZSTD_DCtx_getParameter+0x60): undefined reference to `__clzdi2'

Fixes a510b616131f ("MIPS: Add support for ZSTD-compressed kernels")
---
 arch/mips/boot/compressed/Makefile  | 2 +-
 arch/mips/boot/compressed/clz_ctz.c | 2 ++
 2 files changed, 3 insertions(+), 1 deletion(-)
 create mode 100644 arch/mips/boot/compressed/clz_ctz.c
arch/mips/boot/compressed/Makefile +1 -1
index f27cf31b41401..38e233f7fd7a4 100644
@@ -52,7 +52,7 @@ endif
 
 vmlinuzobjs-$(CONFIG_KERNEL_XZ) += $(obj)/ashldi3.o
 
-vmlinuzobjs-$(CONFIG_KERNEL_ZSTD) += $(obj)/bswapdi.o $(obj)/ashldi3.o
+vmlinuzobjs-$(CONFIG_KERNEL_ZSTD) += $(obj)/bswapdi.o $(obj)/ashldi3.o $(obj)/clz_ctz.o
 
 targets := $(notdir $(vmlinuzobjs-y))
 
arch/mips/boot/compressed/clz_ctz.c new file +2 -0
new file mode 100644
index 0000000000000..b4a1b6eb2f8ad
@@ -0,0 +1,2 @@
+// SPDX-License-Identifier: GPL-2.0-only
+#include "../../../../lib/clz_ctz.c"
Signature · 1 line
2.34.1
author maintainer
#533 lore raw

PATCH 533/1039 mailbox: fix gce_num of mt8192 driver data

In reply to Greg Kroah-Hartman

From: jason-jh.lin <[email protected]>

[ Upstream commit 35ca43710f792ce183312fdc7e4b2bb0b721a173 ]

Because mt8192 only have 1 gce, the gce_num should be 1.

Fixes 85dfdbfc13ea ("mailbox: cmdq: add multi-gce clocks support for mt8195")
---
 drivers/mailbox/mtk-cmdq-mailbox.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/mailbox/mtk-cmdq-mailbox.c +1 -1
index a8845b162dbfa..9aae13e9e050e 100644
@@ -658,7 +658,7 @@ static const struct gce_plat gce_plat_v5 = {
 	.thread_nr = 24,
 	.shift = 3,
 	.control_by_sw = true,
-	.gce_num = 2
+	.gce_num = 1
 };
 
 static const struct gce_plat gce_plat_v6 = {
Signature · 1 line
2.34.1
author maintainer
#534 lore raw

PATCH 534/1039 mailbox: imx: Fix an IS_ERR() vs NULL bug

In reply to Greg Kroah-Hartman

From: Dan Carpenter <[email protected]>

[ Upstream commit 05d06f37196b2e3abeff2b98b785c8803865e646 ]

The devm_kzalloc() function does not return error pointers, it returns NULL on failure.

Fixes 97961f78e8bc ("mailbox: imx: support i.MX8ULP S4 MU")
---
 drivers/mailbox/imx-mailbox.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/mailbox/imx-mailbox.c +2 -2
index ffe36a6bef9e0..544de2db64531 100644
@@ -563,8 +563,8 @@ static int imx_mu_probe(struct platform_device *pdev)
 		size = sizeof(struct imx_sc_rpc_msg_max);
 
 	priv->msg = devm_kzalloc(dev, size, GFP_KERNEL);
-	if (IS_ERR(priv->msg))
-		return PTR_ERR(priv->msg);
+	if (!priv->msg)
+		return -ENOMEM;
 
 	priv->clk = devm_clk_get(dev, NULL);
 	if (IS_ERR(priv->clk)) {
Signature · 1 line
2.34.1
author maintainer
#535 lore raw

PATCH 535/1039 mailbox: pcc: Avoid using the uninitialized variable dev

In reply to Greg Kroah-Hartman

From: Sudeep Holla <[email protected]>

[ Upstream commit 960c4056aadcf61983f8eaac159927a052f8cf01 ]

Smatch static checker warns:

  |  drivers/mailbox/pcc.c:292 pcc_mbox_request_channel()
  |  error: uninitialized symbol 'dev'.

Fix the same by using pr_err instead of dev_err as the variable 'dev' is uninitialized at that stage.

Fixes ce028702ddbc ("mailbox: pcc: Move bulk of PCCT parsing into pcc_mbox_probe")
Cc Jassi Brar <[email protected]>
---
 drivers/mailbox/pcc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/mailbox/pcc.c +1 -1
index 887a3704c12ec..e0a1ab3861f0d 100644
@@ -289,7 +289,7 @@ pcc_mbox_request_channel(struct mbox_client *cl, int subspace_id)
 	pchan = chan_info + subspace_id;
 	chan = pchan->chan.mchan;
 	if (IS_ERR(chan) || chan->cl) {
-		dev_err(dev, "Channel not found for idx: %d\n", subspace_id);
+		pr_err("Channel not found for idx: %d\n", subspace_id);
 		return ERR_PTR(-EBUSY);
 	}
 	dev = chan->mbox->dev;
Signature · 1 line
2.34.1
author maintainer
#536 lore raw

PATCH 536/1039 mailbox: pcc: Handle all PCC subtypes correctly in pcc_mbox_irq

In reply to Greg Kroah-Hartman

From: Sudeep Holla <[email protected]>

[ Upstream commit 7215a7857e796c655ae1184b313556102fa8bc40 ]

Commit c45ded7e1135 ("mailbox: pcc: Add support for PCCT extended PCC subspaces(type 3/4)") enabled the type3/4 of PCCT, but the change in pcc_mbox_irq breaks the other PCC subtypes.

The kernel reports a warning on an Ampere eMag server

-->8

 CPU: 0 PID: 0 Comm: swapper/0 Not tainted 5.16.0-rc4 #127
 Hardware name: MiTAC RAPTOR EV-883832-X3-0001/RAPTOR, BIOS 0.14 02/22/2019
 Call trace:
  dump_backtrace+0x0/0x200
  show_stack+0x20/0x30
  dump_stack_lvl+0x68/0x84
  dump_stack+0x18/0x34
  __report_bad_irq+0x54/0x17c
  note_interrupt+0x330/0x428
  handle_irq_event_percpu+0x90/0x98
  handle_irq_event+0x4c/0x148
  handle_fasteoi_irq+0xc4/0x188
  generic_handle_domain_irq+0x44/0x68
  gic_handle_irq+0x84/0x2ec
  call_on_irq_stack+0x28/0x34
  do_interrupt_handler+0x88/0x90
  el1_interrupt+0x48/0xb0
  el1h_64_irq_handler+0x18/0x28
  el1h_64_irq+0x7c/0x80
Fixes c45ded7e1135 ("mailbox: pcc: Add support for PCCT extended PCC subspaces(type 3/4)")
---
 drivers/mailbox/pcc.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)
drivers/mailbox/pcc.c +5 -3
index e0a1ab3861f0d..ed18936b8ce68 100644
@@ -241,9 +241,11 @@ static irqreturn_t pcc_mbox_irq(int irq, void *p)
 	if (ret)
 		return IRQ_NONE;
 
-	val &= pchan->cmd_complete.status_mask;
-	if (!val)
-		return IRQ_NONE;
+	if (val) { /* Ensure GAS exists and value is non-zero */
+		val &= pchan->cmd_complete.status_mask;
+		if (!val)
+			return IRQ_NONE;
+	}
 
 	ret = pcc_chan_reg_read(&pchan->error, &val);
 	if (ret)
Signature · 1 line
2.34.1