Skip to contentSkip to conversation

iio: adc: max9611: explicitly cast gain_selectors

PATCH NAK Sasha Levin 5 participants 213 messages Nov 8, 2019 – Nov 11, 2019 Page 2 of 2

Subsystems (33)
lore
Share & tools
lore
author maintainer
#185 lore raw

PATCH 184/205 ARM: dts: sunxi: Fix I2C bus warnings

In reply to Sasha Levin

From: Rob Herring <[email protected]>

[ Upstream commit 0729b4af5753b65aa031f58c435da53dbbf56d19 ]

dtc has new checks for I2C buses. Fix the warnings in unit-addresses.

arch/arm/boot/dts/sun8i-a23-gt90h-v4.dtb: Warning (i2c_bus_reg): /soc@1c00000/i2c@1c2ac00/touchscreen@0: I2C bus unit address format error, expected "40" arch/arm/boot/dts/sun8i-a23-inet86dz.dtb: Warning (i2c_bus_reg): /soc@1c00000/i2c@1c2ac00/touchscreen@0: I2C bus unit address format error, expected "40" arch/arm/boot/dts/sun8i-a23-polaroid-mid2407pxe03.dtb: Warning (i2c_bus_reg): /soc@1c00000/i2c@1c2ac00/touchscreen@0: I2C bus unit address format error, expected "40" arch/arm/boot/dts/sun8i-a23-polaroid-mid2809pxe04.dtb: Warning (i2c_bus_reg): /soc@1c00000/i2c@1c2ac00/touchscreen@0: I2C bus unit address format error, expected "40" arch/arm/boot/dts/sun8i-a33-ga10h-v1.1.dtb: Warning (i2c_bus_reg): /soc@1c00000/i2c@1c2ac00/touchscreen@0: I2C bus unit address format error, expected "40" arch/arm/boot/dts/sun8i-a33-inet-d978-rev2.dtb: Warning (i2c_bus_reg): /soc@1c00000/i2c@1c2ac00/touchscreen@0: missing or empty reg property arch/arm/boot/dts/sun8i-a33-ippo-q8h-v1.2.dtb: Warning (i2c_bus_reg): /soc@1c00000/i2c@1c2ac00/touchscreen@0: missing or empty reg property arch/arm/boot/dts/sun8i-a33-q8-tablet.dtb: Warning (i2c_bus_reg): /soc@1c00000/i2c@1c2ac00/touchscreen@0: missing or empty reg property arch/arm/boot/dts/sun5i-a13-utoo-p66.dtb: Warning (i2c_bus_reg): /soc@1c00000/i2c@1c2b000/touchscreen: I2C bus unit address format error, expected "40" arch/arm/boot/dts/sun5i-a13-difrnce-dit4350.dtb: Warning (i2c_bus_reg): /soc@1c00000/i2c@1c2b000/touchscreen: missing or empty reg property arch/arm/boot/dts/sun5i-a13-empire-electronix-m712.dtb: Warning (i2c_bus_reg): /soc@1c00000/i2c@1c2b000/touchscreen: missing or empty reg property arch/arm/boot/dts/sun5i-a13-inet-98v-rev2.dtb: Warning (i2c_bus_reg): /soc@1c00000/i2c@1c2b000/touchscreen: missing or empty reg property arch/arm/boot/dts/sun5i-a13-q8-tablet.dtb: Warning (i2c_bus_reg): /soc@1c00000/i2c@1c2b000/touchscreen: missing or empty reg property

---
 arch/arm/boot/dts/sun5i-reference-design-tablet.dtsi | 3 ++-
 arch/arm/boot/dts/sun8i-reference-design-tablet.dtsi | 3 ++-
 arch/arm/boot/dts/sun8i-v40-bananapi-m2-berry.dts    | 2 +-
 3 files changed, 5 insertions(+), 3 deletions(-)
arch/arm/boot/dts/sun5i-reference-design-tablet.dtsi +2 -1
index 8acbaab14fe51..d2a2eb8b3f262 100644
@@ -92,7 +92,8 @@
 	 */
 	clock-frequency = <400000>;
 
-	touchscreen: touchscreen {
+	touchscreen: touchscreen@40 {
+		reg = <0x40>;
 		interrupt-parent = <&pio>;
 		interrupts = <6 11 IRQ_TYPE_EDGE_FALLING>; /* EINT11 (PG11) */
 		pinctrl-names = "default";
arch/arm/boot/dts/sun8i-reference-design-tablet.dtsi +2 -1
index 880096c7e2523..5e8a95af89b8c 100644
@@ -69,7 +69,8 @@
 	 */
 	clock-frequency = <400000>;
 
-	touchscreen: touchscreen@0 {
+	touchscreen: touchscreen@40 {
+		reg = <0x40>;
 		interrupt-parent = <&pio>;
 		interrupts = <1 5 IRQ_TYPE_EDGE_FALLING>; /* PB5 */
 		pinctrl-names = "default";
arch/arm/boot/dts/sun8i-v40-bananapi-m2-berry.dts +1 -1
index 35859d8f3267f..bf97f6244c233 100644
@@ -95,7 +95,7 @@
 &i2c0 {
 	status = "okay";
 
-	axp22x: pmic@68 {
+	axp22x: pmic@34 {
 		compatible = "x-powers,axp221";
 		reg = <0x34>;
 		interrupt-parent = <&nmi_intc>;
Signature · 1 line
2.20.1
author maintainer
#186 lore raw

PATCH 185/205 pinctrl: at91: don't use the same irqchip with multiple gpiochips

In reply to Sasha Levin

From: Ludovic Desroches <[email protected]>

[ Upstream commit 0c3dfa176912b5f87732545598200fb55e9c1978 ]

Sharing the same irqchip with multiple gpiochips is not a good practice. For instance, when installing hooks, we change the state of the irqchip. The initial state of the irqchip for the second gpiochip to register is then disrupted.

---
 drivers/pinctrl/pinctrl-at91.c | 28 ++++++++++++++--------------
 1 file changed, 14 insertions(+), 14 deletions(-)
drivers/pinctrl/pinctrl-at91.c +14 -14
index 50f0ec42c6372..fad0e132ead84 100644
@@ -1574,16 +1574,6 @@ void at91_pinctrl_gpio_resume(void)
 #define gpio_irq_set_wake	NULL
 #endif /* CONFIG_PM */
 
-static struct irq_chip gpio_irqchip = {
-	.name		= "GPIO",
-	.irq_ack	= gpio_irq_ack,
-	.irq_disable	= gpio_irq_mask,
-	.irq_mask	= gpio_irq_mask,
-	.irq_unmask	= gpio_irq_unmask,
-	/* .irq_set_type is set dynamically */
-	.irq_set_wake	= gpio_irq_set_wake,
-};
-
 static void gpio_irq_handler(struct irq_desc *desc)
 {
 	struct irq_chip *chip = irq_desc_get_chip(desc);
@@ -1624,12 +1614,22 @@ static int at91_gpio_of_irq_setup(struct platform_device *pdev,
 	struct gpio_chip	*gpiochip_prev = NULL;
 	struct at91_gpio_chip   *prev = NULL;
 	struct irq_data		*d = irq_get_irq_data(at91_gpio->pioc_virq);
+	struct irq_chip		*gpio_irqchip;
 	int ret, i;
 
+	gpio_irqchip = devm_kzalloc(&pdev->dev, sizeof(*gpio_irqchip), GFP_KERNEL);
+	if (!gpio_irqchip)
+		return -ENOMEM;
+
 	at91_gpio->pioc_hwirq = irqd_to_hwirq(d);
 
-	/* Setup proper .irq_set_type function */
-	gpio_irqchip.irq_set_type = at91_gpio->ops->irq_type;
+	gpio_irqchip->name = "GPIO";
+	gpio_irqchip->irq_ack = gpio_irq_ack;
+	gpio_irqchip->irq_disable = gpio_irq_mask;
+	gpio_irqchip->irq_mask = gpio_irq_mask;
+	gpio_irqchip->irq_unmask = gpio_irq_unmask;
+	gpio_irqchip->irq_set_wake = gpio_irq_set_wake,
+	gpio_irqchip->irq_set_type = at91_gpio->ops->irq_type;
 
 	/* Disable irqs of this PIO controller */
 	writel_relaxed(~0, at91_gpio->regbase + PIO_IDR);
@@ -1640,7 +1640,7 @@ static int at91_gpio_of_irq_setup(struct platform_device *pdev,
 	 * interrupt.
 	 */
 	ret = gpiochip_irqchip_add(&at91_gpio->chip,
-				   &gpio_irqchip,
+				   gpio_irqchip,
 				   0,
 				   handle_edge_irq,
 				   IRQ_TYPE_NONE);
@@ -1658,7 +1658,7 @@ static int at91_gpio_of_irq_setup(struct platform_device *pdev,
 	if (!gpiochip_prev) {
 		/* Then register the chain on the parent IRQ */
 		gpiochip_set_chained_irqchip(&at91_gpio->chip,
-					     &gpio_irqchip,
+					     gpio_irqchip,
 					     at91_gpio->pioc_virq,
 					     gpio_irq_handler);
 		return 0;
Signature · 1 line
2.20.1
author maintainer
#187 lore raw

PATCH 186/205 ARM: dts: sun9i: Fix I2C bus warnings

In reply to Sasha Levin

From: Rob Herring <[email protected]>

[ Upstream commit 57a83c5222c1b5e7b3acc72c6e60fce00a38991a ]

dtc has new checks for I2C buses. The sun9i-a80 dts file has a node named 'i2c' which causes a false positive warning. As the node is a RSB bus, correct the node name to be 'rsb' to fix the warnings.

arch/arm/boot/dts/sun9i-a80-cubieboard4.dtb: Warning (i2c_bus_reg): /soc/i2c@8003400/codec@e89:reg: I2C address must be less than 10-bits, got "0xe89" arch/arm/boot/dts/sun9i-a80-cubieboard4.dtb: Warning (i2c_bus_reg): /soc/i2c@8003400/pmic@745:reg: I2C address must be less than 10-bits, got "0x745" arch/arm/boot/dts/sun9i-a80-optimus.dtb: Warning (i2c_bus_reg): /soc/i2c@8003400/codec@e89:reg: I2C address must be less than 10-bits, got "0xe89" arch/arm/boot/dts/sun9i-a80-optimus.dtb: Warning (i2c_bus_reg): /soc/i2c@8003400/pmic@745:reg: I2C address must be less than 10-bits, got "0x745"

---
 arch/arm/boot/dts/sun9i-a80.dtsi | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm/boot/dts/sun9i-a80.dtsi +1 -1
index 25591d6883ef2..d9532fb1ef650 100644
@@ -1196,7 +1196,7 @@
 			};
 		};
 
-		r_rsb: i2c@8003400 {
+		r_rsb: rsb@8003400 {
 			compatible = "allwinner,sun8i-a23-rsb";
 			reg = <0x08003400 0x400>;
 			interrupts = <GIC_SPI 39 IRQ_TYPE_LEVEL_HIGH>;
Signature · 1 line
2.20.1
author maintainer
#188 lore raw

PATCH 187/205 android: binder: no outgoing transaction when thread todo has transaction

In reply to Sasha Levin

From: Sherry Yang <[email protected]>

[ Upstream commit 44b73962cb25f1c8170ea695c4564b05a75e1fd4 ]

When a process dies, failed reply is sent to the sender of any transaction queued on a dead thread's todo list. The sender asserts that the received failed reply corresponds to the head of the transaction stack. This assert can fail if the dead thread is allowed to send outgoing transactions when there is already a transaction on its todo list, because this new transaction can end up on the transaction stack of the original sender. The following steps illustrate how this assertion can fail.

1. Thread1 sends txn19 to Thread2 (T1->transaction_stack=txn19, T2->todo+=txn19)
2. Without processing todo list, Thread2 sends txn20 to Thread1 (T1->todo+=txn20, T2->transaction_stack=txn20)
3. T1 processes txn20 on its todo list (T1->transaction_stack=txn20->txn19, T1->todo=<empty>)
4. T2 dies, T2->todo cleanup attempts to send failed reply for txn19, but T1->transaction_stack points to txn20 -- assertion failes

Step 2. is the incorrect behavior. When there is a transaction on a thread's todo list, this thread should not be able to send any outgoing synchronous transactions. Only the head of the todo list needs to be checked because only threads that are waiting for proc work can directly receive work from another thread, and no work is allowed to be queued on such a thread without waking up the thread. This patch also enforces that a thread is not waiting for proc work when a work is directly enqueued to its todo list.

---
 drivers/android/binder.c | 44 +++++++++++++++++++++++++++++-----------
 1 file changed, 32 insertions(+), 12 deletions(-)
drivers/android/binder.c +32 -12
index 6e04e7a707a12..cf4367135a00b 100644
@@ -822,6 +822,7 @@ static void
 binder_enqueue_deferred_thread_work_ilocked(struct binder_thread *thread,
 					    struct binder_work *work)
 {
+	WARN_ON(!list_empty(&thread->waiting_thread_node));
 	binder_enqueue_work_ilocked(work, &thread->todo);
 }
 
@@ -839,6 +840,7 @@ static void
 binder_enqueue_thread_work_ilocked(struct binder_thread *thread,
 				   struct binder_work *work)
 {
+	WARN_ON(!list_empty(&thread->waiting_thread_node));
 	binder_enqueue_work_ilocked(work, &thread->todo);
 	thread->process_todo = true;
 }
@@ -1270,19 +1272,12 @@ static int binder_inc_node_nilocked(struct binder_node *node, int strong,
 		} else
 			node->local_strong_refs++;
 		if (!node->has_strong_ref && target_list) {
+			struct binder_thread *thread = container_of(target_list,
+						    struct binder_thread, todo);
 			binder_dequeue_work_ilocked(&node->work);
-			/*
-			 * Note: this function is the only place where we queue
-			 * directly to a thread->todo without using the
-			 * corresponding binder_enqueue_thread_work() helper
-			 * functions; in this case it's ok to not set the
-			 * process_todo flag, since we know this node work will
-			 * always be followed by other work that starts queue
-			 * processing: in case of synchronous transactions, a
-			 * BR_REPLY or BR_ERROR; in case of oneway
-			 * transactions, a BR_TRANSACTION_COMPLETE.
-			 */
-			binder_enqueue_work_ilocked(&node->work, target_list);
+			BUG_ON(&thread->todo != target_list);
+			binder_enqueue_deferred_thread_work_ilocked(thread,
+								   &node->work);
 		}
 	} else {
 		if (!internal)
@@ -2733,6 +2728,7 @@ static void binder_transaction(struct binder_proc *proc,
 {
 	int ret;
 	struct binder_transaction *t;
+	struct binder_work *w;
 	struct binder_work *tcomplete;
 	binder_size_t *offp, *off_end, *off_start;
 	binder_size_t off_min;
@@ -2874,6 +2870,29 @@ static void binder_transaction(struct binder_proc *proc,
 			goto err_invalid_target_handle;
 		}
 		binder_inner_proc_lock(proc);
+
+		w = list_first_entry_or_null(&thread->todo,
+					     struct binder_work, entry);
+		if (!(tr->flags & TF_ONE_WAY) && w &&
+		    w->type == BINDER_WORK_TRANSACTION) {
+			/*
+			 * Do not allow new outgoing transaction from a
+			 * thread that has a transaction at the head of
+			 * its todo list. Only need to check the head
+			 * because binder_select_thread_ilocked picks a
+			 * thread from proc->waiting_threads to enqueue
+			 * the transaction, and nothing is queued to the
+			 * todo list while the thread is on waiting_threads.
+			 */
+			binder_user_error("%d:%d new transaction not allowed when there is a transaction on thread todo\n",
+					  proc->pid, thread->pid);
+			binder_inner_proc_unlock(proc);
+			return_error = BR_FAILED_REPLY;
+			return_error_param = -EPROTO;
+			return_error_line = __LINE__;
+			goto err_bad_todo_list;
+		}
+
 		if (!(tr->flags & TF_ONE_WAY) && thread->transaction_stack) {
 			struct binder_transaction *tmp;
 
@@ -3256,6 +3275,7 @@ err_alloc_tcomplete_failed:
 	kfree(t);
 	binder_stats_deleted(BINDER_STAT_TRANSACTION);
 err_alloc_t_failed:
+err_bad_todo_list:
 err_bad_call_stack:
 err_empty_call_stack:
 err_dead_binder:
Signature · 1 line
2.20.1
author maintainer
#189 lore raw

PATCH 188/205 cxgb4: Fix endianness issue in t4_fwcache()

In reply to Sasha Levin

From: Ganesh Goudar <[email protected]>

[ Upstream commit 0dc235afc59a226d951352b0adf4a89b532a9d13 ]

Do not put host-endian 0 or 1 into big endian feild.

---
 drivers/net/ethernet/chelsio/cxgb4/t4_hw.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/chelsio/cxgb4/t4_hw.c +1 -1
index 5fe5d16dee724..8350c0c9b89d1 100644
@@ -3889,7 +3889,7 @@ int t4_fwcache(struct adapter *adap, enum fw_params_param_dev_fwcache op)
 	c.param[0].mnem =
 		cpu_to_be32(FW_PARAMS_MNEM_V(FW_PARAMS_MNEM_DEV) |
 			    FW_PARAMS_PARAM_X_V(FW_PARAMS_PARAM_DEV_FWCACHE));
-	c.param[0].val = (__force __be32)op;
+	c.param[0].val = cpu_to_be32(op);
 
 	return t4_wr_mbox(adap, adap->mbox, &c, sizeof(c), NULL);
 }
Signature · 1 line
2.20.1
author maintainer
#190 lore raw

PATCH 189/205 arm64: fix for bad_mode() handler to always result in panic

In reply to Sasha Levin

From: Hari Vyas <[email protected]>

[ Upstream commit e4ba15debcfd27f60d43da940a58108783bff2a6 ]

The bad_mode() handler is called if we encounter an uunknown exception, with the expectation that the subsequent call to panic() will halt the system. Unfortunately, if the exception calling bad_mode() is taken from EL0, then the call to die() can end up killing the current user task and calling schedule() instead of falling through to panic().

Remove the die() call altogether, since we really want to bring down the machine in this "impossible" case.

---
 arch/arm64/kernel/traps.c | 1 -
 1 file changed, 1 deletion(-)
arch/arm64/kernel/traps.c +0 -1
index a0099be4311ae..c8dc3a3640e7e 100644
@@ -611,7 +611,6 @@ asmlinkage void bad_mode(struct pt_regs *regs, int reason, unsigned int esr)
 		handler[reason], smp_processor_id(), esr,
 		esr_get_class_string(esr));
 
-	die("Oops - bad mode", regs, 0);
 	local_daif_mask();
 	panic("bad mode");
 }
Signature · 1 line
2.20.1
author maintainer
#191 lore raw

PATCH 190/205 block, bfq: inject other-queue I/O into seeky idle queues on NCQ flash

In reply to Sasha Levin

From: Paolo Valente <[email protected]>

[ Upstream commit d0edc2473be9d70f999282e1ca7863ad6ae704dc ]

The Achilles' heel of BFQ is its failing to reach a high throughput with sync random I/O on flash storage with internal queueing, in case the processes doing I/O have differentiated weights.

The cause of this failure is as follows. If at least two processes do sync I/O, and have a different weight from each other, then BFQ plugs I/O dispatching every time one of these processes, while it is being served, remains temporarily without pending I/O requests. This plugging is necessary to guarantee that every process enjoys a bandwidth proportional to its weight; but it empties the internal queue(s) of the drive. And this kills throughput with random I/O. So, if some processes have differentiated weights and do both sync and random I/O, the end result is a throughput collapse.

This commit tries to counter this problem by injecting the service of other processes, in a controlled way, while the process in service happens to have no I/O. This injection is performed only if the medium is non rotational and performs internal queueing, and the process in service does random I/O (service injection might be beneficial for sequential I/O too, we'll work on that).

As an example of the benefits of this commit, on a PLEXTOR PX-256M5S SSD, and with five processes having differentiated weights and doing sync random 4KB I/O, this commit makes the throughput with bfq grow by 400%, from 25 to 100MB/s. This higher throughput is 10MB/s lower than that reached with none. As some less random I/O is added to the mix, the throughput becomes equal to or higher than that with none.

This commit is a very first attempt to recover throughput without losing control, and certainly has many limitations. One is, e.g., that the processes whose service is injected are not chosen so as to distribute the extra bandwidth they receive in accordance to their weights. Thus there might be loss of weighted fairness in some cases. Anyway, this loss concerns extra service, which would not have been received at all without this commit. Other limitations and issues will probably show up with usage.

---
 block/bfq-iosched.c | 68 +++++++++++++++++++++++++++++++++++++++++----
 block/bfq-iosched.h | 26 +++++++++++++++++
 2 files changed, 88 insertions(+), 6 deletions(-)
block/bfq-iosched.c +62 -6
index d8d2ac294b0c0..35ddaa820737c 100644
@@ -3195,6 +3195,13 @@ static unsigned long bfq_bfqq_softrt_next_start(struct bfq_data *bfqd,
 		    jiffies + nsecs_to_jiffies(bfqq->bfqd->bfq_slice_idle) + 4);
 }
 
+static bool bfq_bfqq_injectable(struct bfq_queue *bfqq)
+{
+	return BFQQ_SEEKY(bfqq) && bfqq->wr_coeff == 1 &&
+		blk_queue_nonrot(bfqq->bfqd->queue) &&
+		bfqq->bfqd->hw_tag;
+}
+
 /**
  * bfq_bfqq_expire - expire a queue.
  * @bfqd: device owning the queue.
@@ -3304,6 +3311,8 @@ void bfq_bfqq_expire(struct bfq_data *bfqd,
 	if (ref == 1) /* bfqq is gone, no more actions on it */
 		return;
 
+	bfqq->injected_service = 0;
+
 	/* mark bfqq as waiting a request only if a bic still points to it */
 	if (!bfq_bfqq_busy(bfqq) &&
 	    reason != BFQQE_BUDGET_TIMEOUT &&
@@ -3642,6 +3651,30 @@ static bool bfq_bfqq_must_idle(struct bfq_queue *bfqq)
 	return RB_EMPTY_ROOT(&bfqq->sort_list) && bfq_better_to_idle(bfqq);
 }
 
+static struct bfq_queue *bfq_choose_bfqq_for_injection(struct bfq_data *bfqd)
+{
+	struct bfq_queue *bfqq;
+
+	/*
+	 * A linear search; but, with a high probability, very few
+	 * steps are needed to find a candidate queue, i.e., a queue
+	 * with enough budget left for its next request. In fact:
+	 * - BFQ dynamically updates the budget of every queue so as
+	 *   to accommodate the expected backlog of the queue;
+	 * - if a queue gets all its requests dispatched as injected
+	 *   service, then the queue is removed from the active list
+	 *   (and re-added only if it gets new requests, but with
+	 *   enough budget for its new backlog).
+	 */
+	list_for_each_entry(bfqq, &bfqd->active_list, bfqq_list)
+		if (!RB_EMPTY_ROOT(&bfqq->sort_list) &&
+		    bfq_serv_to_charge(bfqq->next_rq, bfqq) <=
+		    bfq_bfqq_budget_left(bfqq))
+			return bfqq;
+
+	return NULL;
+}
+
 /*
  * Select a queue for service.  If we have a current queue in service,
  * check whether to continue servicing it, or retrieve and set a new one.
@@ -3723,10 +3756,19 @@ check_queue:
 	 * No requests pending. However, if the in-service queue is idling
 	 * for a new request, or has requests waiting for a completion and
 	 * may idle after their completion, then keep it anyway.
+	 *
+	 * Yet, to boost throughput, inject service from other queues if
+	 * possible.
 	 */
 	if (bfq_bfqq_wait_request(bfqq) ||
 	    (bfqq->dispatched != 0 && bfq_better_to_idle(bfqq))) {
-		bfqq = NULL;
+		if (bfq_bfqq_injectable(bfqq) &&
+		    bfqq->injected_service * bfqq->inject_coeff <
+		    bfqq->entity.service * 10)
+			bfqq = bfq_choose_bfqq_for_injection(bfqd);
+		else
+			bfqq = NULL;
+
 		goto keep_queue;
 	}
 
@@ -3816,6 +3858,14 @@ static struct request *bfq_dispatch_rq_from_bfqq(struct bfq_data *bfqd,
 
 	bfq_dispatch_remove(bfqd->queue, rq);
 
+	if (bfqq != bfqd->in_service_queue) {
+		if (likely(bfqd->in_service_queue))
+			bfqd->in_service_queue->injected_service +=
+				bfq_serv_to_charge(rq, bfqq);
+
+		goto return_rq;
+	}
+
 	/*
 	 * If weight raising has to terminate for bfqq, then next
 	 * function causes an immediate update of bfqq's weight,
@@ -3834,13 +3884,12 @@ static struct request *bfq_dispatch_rq_from_bfqq(struct bfq_data *bfqd,
 	 * belongs to CLASS_IDLE and other queues are waiting for
 	 * service.
 	 */
-	if (bfqd->busy_queues > 1 && bfq_class_idle(bfqq))
-		goto expire;
-
-	return rq;
+	if (!(bfqd->busy_queues > 1 && bfq_class_idle(bfqq)))
+		goto return_rq;
 
-expire:
 	bfq_bfqq_expire(bfqd, bfqq, false, BFQQE_BUDGET_EXHAUSTED);
+
+return_rq:
 	return rq;
 }
 
@@ -4246,6 +4295,13 @@ static void bfq_init_bfqq(struct bfq_data *bfqd, struct bfq_queue *bfqq,
 			bfq_mark_bfqq_has_short_ttime(bfqq);
 		bfq_mark_bfqq_sync(bfqq);
 		bfq_mark_bfqq_just_created(bfqq);
+		/*
+		 * Aggressively inject a lot of service: up to 90%.
+		 * This coefficient remains constant during bfqq life,
+		 * but this behavior might be changed, after enough
+		 * testing and tuning.
+		 */
+		bfqq->inject_coeff = 1;
 	} else
 		bfq_clear_bfqq_sync(bfqq);
 
block/bfq-iosched.h +26 -0
index d5e9e60cb1a5f..a41e9884f2dd2 100644
@@ -351,6 +351,32 @@ struct bfq_queue {
 	unsigned long split_time; /* time of last split */
 
 	unsigned long first_IO_time; /* time of first I/O for this queue */
+
+	/* max service rate measured so far */
+	u32 max_service_rate;
+	/*
+	 * Ratio between the service received by bfqq while it is in
+	 * service, and the cumulative service (of requests of other
+	 * queues) that may be injected while bfqq is empty but still
+	 * in service. To increase precision, the coefficient is
+	 * measured in tenths of unit. Here are some example of (1)
+	 * ratios, (2) resulting percentages of service injected
+	 * w.r.t. to the total service dispatched while bfqq is in
+	 * service, and (3) corresponding values of the coefficient:
+	 * 1 (50%) -> 10
+	 * 2 (33%) -> 20
+	 * 10 (9%) -> 100
+	 * 9.9 (9%) -> 99
+	 * 1.5 (40%) -> 15
+	 * 0.5 (66%) -> 5
+	 * 0.1 (90%) -> 1
+	 *
+	 * So, if the coefficient is lower than 10, then
+	 * injected service is more than bfqq service.
+	 */
+	unsigned int inject_coeff;
+	/* amount of service injected in current service slot */
+	unsigned int injected_service;
 };
 
 /**
Signature · 1 line
2.20.1
author maintainer
#192 lore raw

PATCH 191/205 blok, bfq: do not plug I/O if all queues are weight-raised

In reply to Sasha Levin

From: Paolo Valente <[email protected]>

[ Upstream commit c8765de0adfcaaf4ffb2d951e07444f00ffa9453 ]

To reduce latency for interactive and soft real-time applications, bfq privileges the bfq_queues containing the I/O of these applications. These privileged queues, referred-to as weight-raised queues, get a much higher share of the device throughput w.r.t. non-privileged queues. To preserve this higher share, the I/O of any non-weight-raised queue must be plugged whenever a sync weight-raised queue, while being served, remains temporarily empty. To attain this goal, bfq simply plugs any I/O (from any queue), if a sync weight-raised queue remains empty while in service.

Unfortunately, this plugging typically lowers throughput with random I/O, on devices with internal queueing (because it reduces the filling level of the internal queues of the device).

This commit addresses this issue by restricting the cases where plugging is performed: if a sync weight-raised queue remains empty while in service, then I/O plugging is performed only if some of the active bfq_queues are *not* weight-raised (which is actually the only circumstance where plugging is needed to preserve the higher share of the throughput of weight-raised queues). This restriction proved able to boost throughput in really many use cases needing only maximum throughput.

---
 block/bfq-iosched.c | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)
block/bfq-iosched.c +8 -2
index 35ddaa820737c..66b1ebc21ce4f 100644
@@ -3593,7 +3593,12 @@ static bool bfq_better_to_idle(struct bfq_queue *bfqq)
 	 * whether bfqq is being weight-raised, because
 	 * bfq_symmetric_scenario() does not take into account also
 	 * weight-raised queues (see comments on
-	 * bfq_weights_tree_add()).
+	 * bfq_weights_tree_add()). In particular, if bfqq is being
+	 * weight-raised, it is important to idle only if there are
+	 * other, non-weight-raised queues that may steal throughput
+	 * to bfqq. Actually, we should be even more precise, and
+	 * differentiate between interactive weight raising and
+	 * soft real-time weight raising.
 	 *
 	 * As a side note, it is worth considering that the above
 	 * device-idling countermeasures may however fail in the
@@ -3605,7 +3610,8 @@ static bool bfq_better_to_idle(struct bfq_queue *bfqq)
 	 * to let requests be served in the desired order until all
 	 * the requests already queued in the device have been served.
 	 */
-	asymmetric_scenario = bfqq->wr_coeff > 1 ||
+	asymmetric_scenario = (bfqq->wr_coeff > 1 &&
+			       bfqd->wr_busy_queues < bfqd->busy_queues) ||
 		!bfq_symmetric_scenario(bfqd);
 
 	/*
Signature · 1 line
2.20.1
author maintainer
#193 lore raw

PATCH 192/205 arm64: dts: meson: Fix erroneous SPI bus warnings

In reply to Sasha Levin

From: Rob Herring <[email protected]>

[ Upstream commit 68ecb5c1920c5b98b1e717fd2349fba2ee5d4031 ]

dtc has new checks for SPI buses. The meson dts files have a node named spi' which causes false positive warnings. As the node is a pinctrl child node, change the node name to be 'spi-pins' to fix the warnings.

arch/arm64/boot/dts/amlogic/meson-gxbb-nanopi-k2.dtb: Warning (spi_bus_bridge): /soc/periphs@c8834000/pinctrl@4b0/spi: incorrect #address-cells for SPI bus

---
 arch/arm64/boot/dts/amlogic/meson-gxbb.dtsi | 2 +-
 arch/arm64/boot/dts/amlogic/meson-gxl.dtsi  | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
arch/arm64/boot/dts/amlogic/meson-gxbb.dtsi +1 -1
index 98cbba6809caa..1ade7e486828c 100644
@@ -390,7 +390,7 @@
 			};
 		};
 
-		spi_pins: spi {
+		spi_pins: spi-pins {
 			mux {
 				groups = "spi_miso",
 					"spi_mosi",
arch/arm64/boot/dts/amlogic/meson-gxl.dtsi +1 -1
index c87a80e9bcc6a..8f0bb3c44bd6d 100644
@@ -337,7 +337,7 @@
 			};
 		};
 
-		spi_pins: spi {
+		spi_pins: spi-pins {
 			mux {
 				groups = "spi_miso",
 					"spi_mosi",
Signature · 7 lines
2.20.1


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

PATCH 193/205 power: supply: ab8500_fg: silence uninitialized variable warnings

In reply to Sasha Levin

From: Dan Carpenter <[email protected]>

[ Upstream commit 54baff8d4e5dce2cef61953b1dc22079cda1ddb1 ]

If kstrtoul() fails then we print "charge_full" when it's uninitialized. The debug printk doesn't add anything so I deleted it and cleaned these two functions up a bit.

---
 drivers/power/supply/ab8500_fg.c | 31 ++++++++++++-------------------
 1 file changed, 12 insertions(+), 19 deletions(-)
drivers/power/supply/ab8500_fg.c +12 -19
index 02356f9b5f22a..8bb89c697c1eb 100644
@@ -2433,17 +2433,14 @@ static ssize_t charge_full_store(struct ab8500_fg *di, const char *buf,
 				 size_t count)
 {
 	unsigned long charge_full;
-	ssize_t ret;
+	int ret;
 
 	ret = kstrtoul(buf, 10, &charge_full);
+	if (ret)
+		return ret;
 
-	dev_dbg(di->dev, "Ret %zd charge_full %lu", ret, charge_full);
-
-	if (!ret) {
-		di->bat_cap.max_mah = (int) charge_full;
-		ret = count;
-	}
-	return ret;
+	di->bat_cap.max_mah = (int) charge_full;
+	return count;
 }
 
 static ssize_t charge_now_show(struct ab8500_fg *di, char *buf)
@@ -2455,20 +2452,16 @@ static ssize_t charge_now_store(struct ab8500_fg *di, const char *buf,
 				 size_t count)
 {
 	unsigned long charge_now;
-	ssize_t ret;
+	int ret;
 
 	ret = kstrtoul(buf, 10, &charge_now);
+	if (ret)
+		return ret;
 
-	dev_dbg(di->dev, "Ret %zd charge_now %lu was %d",
-		ret, charge_now, di->bat_cap.prev_mah);
-
-	if (!ret) {
-		di->bat_cap.user_mah = (int) charge_now;
-		di->flags.user_cap = true;
-		ret = count;
-		queue_delayed_work(di->fg_wq, &di->fg_periodic_work, 0);
-	}
-	return ret;
+	di->bat_cap.user_mah = (int) charge_now;
+	di->flags.user_cap = true;
+	queue_delayed_work(di->fg_wq, &di->fg_periodic_work, 0);
+	return count;
 }
 
 static struct ab8500_fg_sysfs_entry charge_full_attr =
Signature · 1 line
2.20.1
author maintainer
#195 lore raw

PATCH 194/205 power: reset: at91-poweroff: do not procede if at91_shdwc is allocated

In reply to Sasha Levin

From: Claudiu Beznea <[email protected]>

[ Upstream commit 9f1e44774be578fb92776add95f1fcaf8284d692 ]

There should be only one instance of struct shdwc in the system. This is referenced through at91_shdwc. Return in probe if at91_shdwc is already allocated.

---
 drivers/power/reset/at91-sama5d2_shdwc.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/power/reset/at91-sama5d2_shdwc.c +3 -0
index 0206cce328b3d..d9493e893d64e 100644
@@ -246,6 +246,9 @@ static int __init at91_shdwc_probe(struct platform_device *pdev)
 	if (!pdev->dev.of_node)
 		return -ENODEV;
 
+	if (at91_shdwc)
+		return -EBUSY;
+
 	at91_shdwc = devm_kzalloc(&pdev->dev, sizeof(*at91_shdwc), GFP_KERNEL);
 	if (!at91_shdwc)
 		return -ENOMEM;
Signature · 1 line
2.20.1
author maintainer
#196 lore raw

PATCH 195/205 power: supply: max8998-charger: Fix platform data retrieval

In reply to Sasha Levin

From: Tomasz Figa <[email protected]>

[ Upstream commit cb90a2c6f77fe9b43d1e3f759bb2f13fe7fa1811 ]

Since the max8998 MFD driver supports instantiation by DT, platform data retrieval is handled in MFD probe and cell drivers should get use the pdata field of max8998_dev struct to obtain them.

Fixes ee999fb3f17f ("mfd: max8998: Add support for Device Tree")
---
 drivers/power/supply/max8998_charger.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/power/supply/max8998_charger.c +1 -1
index cad7d1a8feec7..aa65e6c36c55e 100644
@@ -86,7 +86,7 @@ static const struct power_supply_desc max8998_battery_desc = {
 static int max8998_battery_probe(struct platform_device *pdev)
 {
 	struct max8998_dev *iodev = dev_get_drvdata(pdev->dev.parent);
-	struct max8998_platform_data *pdata = dev_get_platdata(iodev->dev);
+	struct max8998_platform_data *pdata = iodev->pdata;
 	struct power_supply_config psy_cfg = {};
 	struct max8998_battery_data *max8998;
 	struct i2c_client *i2c;
Signature · 1 line
2.20.1
author maintainer
#197 lore raw

PATCH 196/205 component: fix loop condition to call unbind() if bind() fails

In reply to Sasha Levin

From: Banajit Goswami <[email protected]>

[ Upstream commit bdae566d5d9733b6e32b378668b84eadf28a94d4 ]

During component_bind_all(), if bind() fails for any particular component associated with a master, unbind() should be called for all previous components in that master's match array, whose bind() might have completed successfully. As per the current logic, if bind() fails for the component at position 'n' in the master's match array, it would start calling unbind() from component in 'n'th position itself and work backwards, and will always skip calling unbind() for component in 0th position in the master's match array.
Fix this by updating the loop condition, and the logic to refer to the components in master's match array, so that unbind() is called for all components starting from 'n-1'st position in the array, until (and including) component in 0th position.

---
 drivers/base/component.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/base/component.c +3 -3
index 8946dfee4768e..e8d676fad0c95 100644
@@ -536,9 +536,9 @@ int component_bind_all(struct device *master_dev, void *data)
 		}
 
 	if (ret != 0) {
-		for (; i--; )
-			if (!master->match->compare[i].duplicate) {
-				c = master->match->compare[i].component;
+		for (; i > 0; i--)
+			if (!master->match->compare[i - 1].duplicate) {
+				c = master->match->compare[i - 1].component;
 				component_unbind(c, master, data);
 			}
 	}
Signature · 1 line
2.20.1
author maintainer
#198 lore raw

PATCH 197/205 kernfs: Fix range checks in kernfs_get_target_path

In reply to Sasha Levin

From: Bernd Edlinger <[email protected]>

[ Upstream commit a75e78f21f9ad4b810868c89dbbabcc3931591ca ]

The terminating NUL byte is only there because the buffer is allocated with kzalloc(PAGE_SIZE, GFP_KERNEL), but since the range-check is off-by-one, and PAGE_SIZE==PATH_MAX, the returned string may not be zero-terminated if it is exactly PATH_MAX characters long. Furthermore also the initial loop may theoretically exceed PATH_MAX and cause a fault.

---
 fs/kernfs/symlink.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
fs/kernfs/symlink.c +4 -1
index 305b220af45d3..162f43b80c84c 100644
@@ -72,6 +72,9 @@ static int kernfs_get_target_path(struct kernfs_node *parent,
 		if (base == kn)
 			break;
 
+		if ((s - path) + 3 >= PATH_MAX)
+			return -ENAMETOOLONG;
+
 		strcpy(s, "../");
 		s += 3;
 		base = base->parent;
@@ -88,7 +91,7 @@ static int kernfs_get_target_path(struct kernfs_node *parent,
 	if (len < 2)
 		return -EINVAL;
 	len--;
-	if ((s - path) + len > PATH_MAX)
+	if ((s - path) + len >= PATH_MAX)
 		return -ENAMETOOLONG;
 
 	/* reverse fillup of target string from target to base */
Signature · 1 line
2.20.1
author maintainer
#199 lore raw

PATCH 198/205 ip_gre: fix parsing gre header in ipgre_err

In reply to Sasha Levin

From: Haishuang Yan <[email protected]>

[ Upstream commit b0350d51f001e6edc13ee4f253b98b50b05dd401 ]

gre_parse_header stops parsing when csum_err is encountered, which means tpi->key is undefined and ip_tunnel_lookup will return NULL improperly.

This patch introduce a NULL pointer as csum_err parameter. Even when csum_err is encountered, it won't return error and continue parsing gre header as expected.

Fixes 9f57c67c379d ("gre: Remove support for sharing GRE protocol hook.")
---
 net/ipv4/gre_demux.c | 7 ++++---
 net/ipv4/ip_gre.c    | 9 +++------
 2 files changed, 7 insertions(+), 9 deletions(-)
net/ipv4/gre_demux.c +4 -3
index f21ea6125fc2d..511b32ea25331 100644
@@ -87,13 +87,14 @@ int gre_parse_header(struct sk_buff *skb, struct tnl_ptk_info *tpi,
 
 	options = (__be32 *)(greh + 1);
 	if (greh->flags & GRE_CSUM) {
-		if (skb_checksum_simple_validate(skb)) {
+		if (!skb_checksum_simple_validate(skb)) {
+			skb_checksum_try_convert(skb, IPPROTO_GRE, 0,
+						 null_compute_pseudo);
+		} else if (csum_err) {
 			*csum_err = true;
 			return -EINVAL;
 		}
 
-		skb_checksum_try_convert(skb, IPPROTO_GRE, 0,
-					 null_compute_pseudo);
 		options++;
 	}
 
net/ipv4/ip_gre.c +3 -6
index 0b87558f265eb..9d5db4e40b2a9 100644
@@ -232,13 +232,10 @@ static void gre_err(struct sk_buff *skb, u32 info)
 	const int type = icmp_hdr(skb)->type;
 	const int code = icmp_hdr(skb)->code;
 	struct tnl_ptk_info tpi;
-	bool csum_err = false;
 
-	if (gre_parse_header(skb, &tpi, &csum_err, htons(ETH_P_IP),
-			     iph->ihl * 4) < 0) {
-		if (!csum_err)		/* ignore csum errors. */
-			return;
-	}
+	if (gre_parse_header(skb, &tpi, NULL, htons(ETH_P_IP),
+			     iph->ihl * 4) < 0)
+		return;
 
 	if (type == ICMP_DEST_UNREACH && code == ICMP_FRAG_NEEDED) {
 		ipv4_update_pmtu(skb, dev_net(skb->dev), info,
Signature · 1 line
2.20.1
author maintainer
#200 lore raw

PATCH 199/205 scsi: ufshcd: Fix NULL pointer dereference for in ufshcd_init

In reply to Sasha Levin

From: Vivek Gautam <[email protected]>

[ Upstream commit eebcc19646489b68399ce7b35d9c38eb9f4ec40f ]

Error paths in ufshcd_init() ufshcd_hba_exit() killed clk_scaling workqueue when the workqueue is actually created quite late in ufshcd_init(). So, we end up getting NULL pointer dereference in such error paths. Fix this by moving clk_scaling initialization and kill codes to two separate methods, and call them at required places.

Fixes 401f1e4490ee ("scsi: ufs: don't suspend clock scaling during clock

gating")

Cc Bjorn Andersson <[email protected]>
Cc Subhash Jadavani <[email protected]>
Cc Matthias Kaehlcke <[email protected]>
Cc Evan Green <[email protected]>
Cc Martin K. Petersen <[email protected]>
---
 drivers/scsi/ufs/ufshcd.c | 53 +++++++++++++++++++++++++--------------
 1 file changed, 34 insertions(+), 19 deletions(-)
drivers/scsi/ufs/ufshcd.c +34 -19
index 4aaba3e030554..8bce755e0f5bc 100644
@@ -1772,6 +1772,34 @@ out:
 	return count;
 }
 
+static void ufshcd_init_clk_scaling(struct ufs_hba *hba)
+{
+	char wq_name[sizeof("ufs_clkscaling_00")];
+
+	if (!ufshcd_is_clkscaling_supported(hba))
+		return;
+
+	INIT_WORK(&hba->clk_scaling.suspend_work,
+		  ufshcd_clk_scaling_suspend_work);
+	INIT_WORK(&hba->clk_scaling.resume_work,
+		  ufshcd_clk_scaling_resume_work);
+
+	snprintf(wq_name, sizeof(wq_name), "ufs_clkscaling_%d",
+		 hba->host->host_no);
+	hba->clk_scaling.workq = create_singlethread_workqueue(wq_name);
+
+	ufshcd_clkscaling_init_sysfs(hba);
+}
+
+static void ufshcd_exit_clk_scaling(struct ufs_hba *hba)
+{
+	if (!ufshcd_is_clkscaling_supported(hba))
+		return;
+
+	destroy_workqueue(hba->clk_scaling.workq);
+	ufshcd_devfreq_remove(hba);
+}
+
 static void ufshcd_init_clk_gating(struct ufs_hba *hba)
 {
 	char wq_name[sizeof("ufs_clk_gating_00")];
@@ -6676,6 +6704,7 @@ out:
 	 */
 	if (ret && !ufshcd_eh_in_progress(hba) && !hba->pm_op_in_progress) {
 		pm_runtime_put_sync(hba->dev);
+		ufshcd_exit_clk_scaling(hba);
 		ufshcd_hba_exit(hba);
 	}
 
@@ -7223,12 +7252,9 @@ static void ufshcd_hba_exit(struct ufs_hba *hba)
 		ufshcd_variant_hba_exit(hba);
 		ufshcd_setup_vreg(hba, false);
 		ufshcd_suspend_clkscaling(hba);
-		if (ufshcd_is_clkscaling_supported(hba)) {
+		if (ufshcd_is_clkscaling_supported(hba))
 			if (hba->devfreq)
 				ufshcd_suspend_clkscaling(hba);
-			destroy_workqueue(hba->clk_scaling.workq);
-			ufshcd_devfreq_remove(hba);
-		}
 		ufshcd_setup_clocks(hba, false);
 		ufshcd_setup_hba_vreg(hba, false);
 		hba->is_powered = false;
@@ -7908,6 +7934,7 @@ void ufshcd_remove(struct ufs_hba *hba)
 	ufshcd_disable_intr(hba, hba->intr_mask);
 	ufshcd_hba_stop(hba, true);
 
+	ufshcd_exit_clk_scaling(hba);
 	ufshcd_exit_clk_gating(hba);
 	if (ufshcd_is_clkscaling_supported(hba))
 		device_remove_file(hba->dev, &hba->clk_scaling.enable_attr);
@@ -8079,6 +8106,8 @@ int ufshcd_init(struct ufs_hba *hba, void __iomem *mmio_base, unsigned int irq)
 
 	ufshcd_init_clk_gating(hba);
 
+	ufshcd_init_clk_scaling(hba);
+
 	/*
 	 * In order to avoid any spurious interrupt immediately after
 	 * registering UFS controller interrupt handler, clear any pending UFS
@@ -8117,21 +8146,6 @@ int ufshcd_init(struct ufs_hba *hba, void __iomem *mmio_base, unsigned int irq)
 		goto out_remove_scsi_host;
 	}
 
-	if (ufshcd_is_clkscaling_supported(hba)) {
-		char wq_name[sizeof("ufs_clkscaling_00")];
-
-		INIT_WORK(&hba->clk_scaling.suspend_work,
-			  ufshcd_clk_scaling_suspend_work);
-		INIT_WORK(&hba->clk_scaling.resume_work,
-			  ufshcd_clk_scaling_resume_work);
-
-		snprintf(wq_name, sizeof(wq_name), "ufs_clkscaling_%d",
-			 host->host_no);
-		hba->clk_scaling.workq = create_singlethread_workqueue(wq_name);
-
-		ufshcd_clkscaling_init_sysfs(hba);
-	}
-
 	/*
 	 * Set the default power management level for runtime and system PM.
 	 * Default power saving mode is to keep UFS link in Hibern8 state
@@ -8169,6 +8183,7 @@ int ufshcd_init(struct ufs_hba *hba, void __iomem *mmio_base, unsigned int irq)
 out_remove_scsi_host:
 	scsi_remove_host(hba->host);
 exit_gating:
+	ufshcd_exit_clk_scaling(hba);
 	ufshcd_exit_clk_gating(hba);
 out_disable:
 	hba->is_irq_enabled = false;
Signature · 1 line
2.20.1
author maintainer
#201 lore raw

PATCH 200/205 ARM: dts: rockchip: Fix erroneous SPI bus dtc warnings on rk3036

In reply to Sasha Levin

From: Rob Herring <[email protected]>

[ Upstream commit 131c3eb428ccd5f0c784b9edb4f72ec296a045d2 ]

dtc has new checks for SPI buses. The rk3036 dts file has a node named spi' which causes false positive warnings. As the node is a pinctrl child node, change the node name to be 'spi-pins' to fix the warnings.

arch/arm/boot/dts/rk3036-evb.dtb: Warning (spi_bus_bridge): /pinctrl/spi: incorrect #address-cells for SPI bus arch/arm/boot/dts/rk3036-kylin.dtb: Warning (spi_bus_bridge): /pinctrl/spi: incorrect #address-cells for SPI bus arch/arm/boot/dts/rk3036-evb.dtb: Warning (spi_bus_bridge): /pinctrl/spi: incorrect #size-cells for SPI bus arch/arm/boot/dts/rk3036-kylin.dtb: Warning (spi_bus_bridge): /pinctrl/spi: incorrect #size-cells for SPI bus

---
 arch/arm/boot/dts/rk3036.dtsi | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm/boot/dts/rk3036.dtsi +1 -1
index 67f57200d9a06..d560fc4051c5f 100644
@@ -733,7 +733,7 @@
 			/* no rts / cts for uart2 */
 		};
 
-		spi {
+		spi-pins {
 			spi_txd:spi-txd {
 				rockchip,pins = <1 29 RK_FUNC_3 &pcfg_pull_default>;
 			};
Signature · 1 line
2.20.1
author maintainer
#202 lore raw

PATCH 201/205 arm64: dts: rockchip: Fix I2C bus unit-address error on rk3399-puma-haikou

In reply to Sasha Levin

From: Rob Herring <[email protected]>

[ Upstream commit 501500e65fa96f899230d66153fefd780f08dd34 ]

dtc has new checks for I2C buses. Fix the warnings in unit-addresses.

arch/arm64/boot/dts/rockchip/rk3399-puma-haikou.dtb: Warning (i2c_bus_reg): /i2c@ff3d0000/codec@0a: I2C bus unit address format error, expected "a"

---
 arch/arm64/boot/dts/rockchip/rk3399-puma-haikou.dts | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm64/boot/dts/rockchip/rk3399-puma-haikou.dts +1 -1
index 8ce4a79d9360f..1e6a71066c163 100644
@@ -131,7 +131,7 @@
 	status = "okay";
 	clock-frequency = <400000>;
 
-	sgtl5000: codec@0a {
+	sgtl5000: codec@a {
 		compatible = "fsl,sgtl5000";
 		reg = <0x0a>;
 		clocks = <&sgtl5000_clk>;
Signature · 1 line
2.20.1
author maintainer
#203 lore raw

PATCH 202/205 ACPI / LPSS: Exclude I2C busses shared with PUNIT from pmc_atom_d3_mask

In reply to Sasha Levin

From: Hans de Goede <[email protected]>

[ Upstream commit 86b62e5cd8965d3056f9e9ccdec51631c37add81 ]

lpss_iosf_enter_d3_state() checks if all hw-blocks using the DMA controllers are in d3 before powering down the DMA controllers.

But on devices, where the I2C bus connected to the PMIC is shared by the PUNIT, the controller for that bus will never reach d3 since it has an effectively empty _PS3 method. Instead it appears to automatically power-down during S0i3 and we never see it as being in d3.

This causes the DMA controllers to never be powered-down on these devices, causing them to never reach S0i3. This commit uses the ACPI _SEM method to detect if an I2C bus is shared with the PUNIT and if it is, it removes it from the mask of devices which lpss_iosf_enter_d3_state() checks for.

This fixes these devices never reaching any S0ix states.

---
 drivers/acpi/acpi_lpss.c | 22 ++++++++++++++++++++--
 1 file changed, 20 insertions(+), 2 deletions(-)
drivers/acpi/acpi_lpss.c +20 -2
index c651e206d7960..7eda27d43b482 100644
@@ -99,6 +99,9 @@ struct lpss_private_data {
 	u32 prv_reg_ctx[LPSS_PRV_REG_COUNT];
 };
 
+/* Devices which need to be in D3 before lpss_iosf_enter_d3_state() proceeds */
+static u32 pmc_atom_d3_mask = 0xfe000ffe;
+
 /* LPSS run time quirks */
 static unsigned int lpss_quirks;
 
@@ -175,6 +178,21 @@ static void byt_pwm_setup(struct lpss_private_data *pdata)
 
 static void byt_i2c_setup(struct lpss_private_data *pdata)
 {
+	const char *uid_str = acpi_device_uid(pdata->adev);
+	acpi_handle handle = pdata->adev->handle;
+	unsigned long long shared_host = 0;
+	acpi_status status;
+	long uid = 0;
+
+	/* Expected to always be true, but better safe then sorry */
+	if (uid_str)
+		uid = simple_strtol(uid_str, NULL, 10);
+
+	/* Detect I2C bus shared with PUNIT and ignore its d3 status */
+	status = acpi_evaluate_integer(handle, "_SEM", NULL, &shared_host);
+	if (ACPI_SUCCESS(status) && shared_host && uid)
+		pmc_atom_d3_mask &= ~(BIT_LPSS2_F1_I2C1 << (uid - 1));
+
 	lpss_deassert_reset(pdata);
 
 	if (readl(pdata->mmio_base + pdata->dev_desc->prv_offset))
@@ -894,7 +912,7 @@ static void lpss_iosf_enter_d3_state(void)
 	 * Here we read the values related to LPSS power island, i.e. LPSS
 	 * devices, excluding both LPSS DMA controllers, along with SCC domain.
 	 */
-	u32 func_dis, d3_sts_0, pmc_status, pmc_mask = 0xfe000ffe;
+	u32 func_dis, d3_sts_0, pmc_status;
 	int ret;
 
 	ret = pmc_atom_read(PMC_FUNC_DIS, &func_dis);
@@ -912,7 +930,7 @@ static void lpss_iosf_enter_d3_state(void)
 	 * Shutdown both LPSS DMA controllers if and only if all other devices
 	 * are already in D3hot.
 	 */
-	pmc_status = (~(d3_sts_0 | func_dis)) & pmc_mask;
+	pmc_status = (~(d3_sts_0 | func_dis)) & pmc_atom_d3_mask;
 	if (pmc_status)
 		goto exit;
 
Signature · 1 line
2.20.1
author maintainer
#204 lore raw

PATCH 203/205 netfilter: nf_tables: avoid BUG_ON usage

In reply to Sasha Levin

From: Florian Westphal <[email protected]>

[ Upstream commit fa5950e498e7face21a1761f327e6c1152f778c3 ]

None of these spots really needs to crash the kernel.
In one two cases we can jsut report error to userspace, in the other cases we can just use WARN_ON (and leak memory instead).

---
 net/netfilter/nf_tables_api.c | 9 ++++++---
 net/netfilter/nft_cmp.c       | 6 ++++--
 net/netfilter/nft_reject.c    | 6 ++++--
 3 files changed, 14 insertions(+), 7 deletions(-)
net/netfilter/nf_tables_api.c +6 -3
index 24fddf0322790..289d079008ee8 100644
@@ -1031,7 +1031,8 @@ static int nf_tables_deltable(struct net *net, struct sock *nlsk,
 
 static void nf_tables_table_destroy(struct nft_ctx *ctx)
 {
-	BUG_ON(ctx->table->use > 0);
+	if (WARN_ON(ctx->table->use > 0))
+		return;
 
 	rhltable_destroy(&ctx->table->chains_ht);
 	kfree(ctx->table->name);
@@ -1446,7 +1447,8 @@ static void nf_tables_chain_destroy(struct nft_ctx *ctx)
 {
 	struct nft_chain *chain = ctx->chain;
 
-	BUG_ON(chain->use > 0);
+	if (WARN_ON(chain->use > 0))
+		return;
 
 	/* no concurrent access possible anymore */
 	nf_tables_chain_free_chain_rules(chain);
@@ -7253,7 +7255,8 @@ int __nft_release_basechain(struct nft_ctx *ctx)
 {
 	struct nft_rule *rule, *nr;
 
-	BUG_ON(!nft_is_base_chain(ctx->chain));
+	if (WARN_ON(!nft_is_base_chain(ctx->chain)))
+		return 0;
 
 	nf_tables_unregister_hook(ctx->net, ctx->chain->table, ctx->chain);
 	list_for_each_entry_safe(rule, nr, &ctx->chain->rules, list) {
net/netfilter/nft_cmp.c +4 -2
index fa90a8402845d..79d48c1d06f4d 100644
@@ -79,7 +79,8 @@ static int nft_cmp_init(const struct nft_ctx *ctx, const struct nft_expr *expr,
 
 	err = nft_data_init(NULL, &priv->data, sizeof(priv->data), &desc,
 			    tb[NFTA_CMP_DATA]);
-	BUG_ON(err < 0);
+	if (err < 0)
+		return err;
 
 	priv->sreg = nft_parse_register(tb[NFTA_CMP_SREG]);
 	err = nft_validate_register_load(priv->sreg, desc.len);
@@ -129,7 +130,8 @@ static int nft_cmp_fast_init(const struct nft_ctx *ctx,
 
 	err = nft_data_init(NULL, &data, sizeof(data), &desc,
 			    tb[NFTA_CMP_DATA]);
-	BUG_ON(err < 0);
+	if (err < 0)
+		return err;
 
 	priv->sreg = nft_parse_register(tb[NFTA_CMP_SREG]);
 	err = nft_validate_register_load(priv->sreg, desc.len);
net/netfilter/nft_reject.c +4 -2
index 29f5bd2377b0d..b48e58cceeb72 100644
@@ -94,7 +94,8 @@ static u8 icmp_code_v4[NFT_REJECT_ICMPX_MAX + 1] = {
 
 int nft_reject_icmp_code(u8 code)
 {
-	BUG_ON(code > NFT_REJECT_ICMPX_MAX);
+	if (WARN_ON_ONCE(code > NFT_REJECT_ICMPX_MAX))
+		return ICMP_NET_UNREACH;
 
 	return icmp_code_v4[code];
 }
@@ -111,7 +112,8 @@ static u8 icmp_code_v6[NFT_REJECT_ICMPX_MAX + 1] = {
 
 int nft_reject_icmpv6_code(u8 code)
 {
-	BUG_ON(code > NFT_REJECT_ICMPX_MAX);
+	if (WARN_ON_ONCE(code > NFT_REJECT_ICMPX_MAX))
+		return ICMPV6_NOROUTE;
 
 	return icmp_code_v6[code];
 }
Signature · 1 line
2.20.1
author maintainer
#205 lore raw

PATCH 204/205 s390/qeth: limit csum offload erratum to L3 devices

In reply to Sasha Levin

From: Julian Wiedmann <[email protected]>

[ Upstream commit f231dc9dbd789b0f98a15941e3cebedb4ad72ad5 ]

Combined L3+L4 csum offload is only required for some L3 HW. So for L2 devices, don't offload the IP header csum calculation.

Reference-ID: JUP 394553

---
 drivers/s390/net/qeth_core.h    | 5 -----
 drivers/s390/net/qeth_l3_main.c | 5 +++++
 2 files changed, 5 insertions(+), 5 deletions(-)
drivers/s390/net/qeth_core.h +0 -5
index b2657582cfcfd..41a2f901ccee5 100644
@@ -902,11 +902,6 @@ static inline void qeth_tx_csum(struct sk_buff *skb, u8 *flags, int ipv)
 	if ((ipv == 4 && ip_hdr(skb)->protocol == IPPROTO_UDP) ||
 	    (ipv == 6 && ipv6_hdr(skb)->nexthdr == IPPROTO_UDP))
 		*flags |= QETH_HDR_EXT_UDP;
-	if (ipv == 4) {
-		/* some HW requires combined L3+L4 csum offload: */
-		*flags |= QETH_HDR_EXT_CSUM_HDR_REQ;
-		ip_hdr(skb)->check = 0;
-	}
 }
 
 static inline void qeth_put_buffer_pool_entry(struct qeth_card *card,
drivers/s390/net/qeth_l3_main.c +5 -0
index 9c5e801b3f6cb..c60660cb5a031 100644
@@ -2054,6 +2054,11 @@ static void qeth_l3_fill_header(struct qeth_card *card, struct qeth_hdr *hdr,
 
 	if (!skb_is_gso(skb) && skb->ip_summed == CHECKSUM_PARTIAL) {
 		qeth_tx_csum(skb, &hdr->hdr.l3.ext_flags, ipv);
+		/* some HW requires combined L3+L4 csum offload: */
+		if (ipv == 4) {
+			hdr->hdr.l3.ext_flags |= QETH_HDR_EXT_CSUM_HDR_REQ;
+			ip_hdr(skb)->check = 0;
+		}
 		if (card->options.performance_stats)
 			card->perf_stats.tx_csum++;
 	}
Signature · 1 line
2.20.1
#206 lore raw

Re: [PATCH AUTOSEL 4.19 204/205] s390/qeth: limit csum offload erratum to L3 devices

Quoted 7 lines · Sasha Levin

From: Julian Wiedmann <[email protected]>

[ Upstream commit f231dc9dbd789b0f98a15941e3cebedb4ad72ad5 ]

Combined L3+L4 csum offload is only required for some L3 HW. So for L2 devices, don't offload the IP header csum calculation.

NACK, this has no relevance for stable.

Quoted 42 lines

Reference-ID: JUP 394553

---
 drivers/s390/net/qeth_core.h    | 5 -----
 drivers/s390/net/qeth_l3_main.c | 5 +++++
 2 files changed, 5 insertions(+), 5 deletions(-)
drivers/s390/net/qeth_core.h +0 -5
index b2657582cfcfd..41a2f901ccee5 100644
@@ -902,11 +902,6 @@ static inline void qeth_tx_csum(struct sk_buff *skb, u8 *flags, int ipv)
 	if ((ipv == 4 && ip_hdr(skb)->protocol == IPPROTO_UDP) ||
 	    (ipv == 6 && ipv6_hdr(skb)->nexthdr == IPPROTO_UDP))
 		*flags |= QETH_HDR_EXT_UDP;
-	if (ipv == 4) {
-		/* some HW requires combined L3+L4 csum offload: */
-		*flags |= QETH_HDR_EXT_CSUM_HDR_REQ;
-		ip_hdr(skb)->check = 0;
-	}
 }
 
 static inline void qeth_put_buffer_pool_entry(struct qeth_card *card,
drivers/s390/net/qeth_l3_main.c +5 -0
index 9c5e801b3f6cb..c60660cb5a031 100644
@@ -2054,6 +2054,11 @@ static void qeth_l3_fill_header(struct qeth_card *card, struct qeth_hdr *hdr,
 
 	if (!skb_is_gso(skb) && skb->ip_summed == CHECKSUM_PARTIAL) {
 		qeth_tx_csum(skb, &hdr->hdr.l3.ext_flags, ipv);
+		/* some HW requires combined L3+L4 csum offload: */
+		if (ipv == 4) {
+			hdr->hdr.l3.ext_flags |= QETH_HDR_EXT_CSUM_HDR_REQ;
+			ip_hdr(skb)->check = 0;
+		}
 		if (card->options.performance_stats)
 			card->perf_stats.tx_csum++;
 	}
author maintainer
#207 lore raw

Re: [PATCH AUTOSEL 4.19 204/205] s390/qeth: limit csum offload erratum to L3 devices

Quoted 10 lines · Julian Wiedmann
Quoted 7 lines · Sasha Levin

From: Julian Wiedmann <[email protected]>

[ Upstream commit f231dc9dbd789b0f98a15941e3cebedb4ad72ad5 ]

Combined L3+L4 csum offload is only required for some L3 HW. So for L2 devices, don't offload the IP header csum calculation.

NACK, this has no relevance for stable.

Sure, I'll drop it.

Do you have an idea why the centos and ubuntu folks might have backported this commit into their kernels?

Signature · 2 lines
Thanks,
Sasha
#208 lore raw

Re: [PATCH AUTOSEL 4.19 204/205] s390/qeth: limit csum offload erratum to L3 devices

Quoted 10 lines · Sasha Levin
Quoted 9 lines · Julian Wiedmann
Quoted 7 lines · Sasha Levin

From: Julian Wiedmann <[email protected]>

[ Upstream commit f231dc9dbd789b0f98a15941e3cebedb4ad72ad5 ]

Combined L3+L4 csum offload is only required for some L3 HW. So for L2 devices, don't offload the IP header csum calculation.

Quoted 7 lines
Quoted 1 line

NACK, this has no relevance for stable.

Sure, I'll drop it.

Do you have an idea why the centos and ubuntu folks might have backported this commit into their kernels?

No clue, I trust they have their own reasons.

maintainer
#209 lore raw

Re: [PATCH AUTOSEL 4.19 204/205] s390/qeth: limit csum offload erratum to L3 devices

Quoted 13 lines · Julian Wiedmann
Quoted 12 lines · Sasha Levin
Quoted 10 lines · Julian Wiedmann
Quoted 7 lines · Sasha Levin

From: Julian Wiedmann <[email protected]>

[ Upstream commit f231dc9dbd789b0f98a15941e3cebedb4ad72ad5 ]

Combined L3+L4 csum offload is only required for some L3 HW. So for L2 devices, don't offload the IP header csum calculation.

NACK, this has no relevance for stable.

Quoted 8 lines
Quoted 5 lines

Sure, I'll drop it.

Do you have an idea why the centos and ubuntu folks might have backported this commit into their kernels?

No clue, I trust they have their own reasons.

I cant see centos backporting anything unless they were asked to do so. And this really looks like a "bugfix" to me, why isn't this relevant for any older kernel versions?

thanks,

greg k-h

maintainer
#210 lore raw

Re: [PATCH AUTOSEL 4.19 204/205] s390/qeth: limit csum offload erratum to L3 devices

From: Greg KH <[email protected]>
Date: Fri, 8 Nov 2019 13:34:16 +0100

Quoted 20 lines · Julian Wiedmann
Quoted 19 lines · Sasha Levin
Quoted 17 lines · Julian Wiedmann
Quoted 10 lines · Sasha Levin
Quoted 7 lines

From: Julian Wiedmann <[email protected]>

[ Upstream commit f231dc9dbd789b0f98a15941e3cebedb4ad72ad5 ]

Combined L3+L4 csum offload is only required for some L3 HW. So for L2 devices, don't offload the IP header csum calculation.

NACK, this has no relevance for stable.

Sure, I'll drop it.

Do you have an idea why the centos and ubuntu folks might have backported this commit into their kernels?

Quoted 6 lines
Quoted 2 lines

No clue, I trust they have their own reasons.

I cant see centos backporting anything unless they were asked to do so. And this really looks like a "bugfix" to me, why isn't this relevant for any older kernel versions?

Yeah seriously, this looks entirely legit.

#211 lore raw

Re: [PATCH AUTOSEL 4.19 204/205] s390/qeth: limit csum offload erratum to L3 devices

Quoted 23 lines · David Miller

From: Greg KH <[email protected]>
Date: Fri, 8 Nov 2019 13:34:16 +0100

Quoted 20 lines · Julian Wiedmann
Quoted 19 lines · Sasha Levin
Quoted 17 lines · Julian Wiedmann
Quoted 10 lines · Sasha Levin
Quoted 7 lines

From: Julian Wiedmann <[email protected]>

[ Upstream commit f231dc9dbd789b0f98a15941e3cebedb4ad72ad5 ]

Combined L3+L4 csum offload is only required for some L3 HW. So for L2 devices, don't offload the IP header csum calculation.

NACK, this has no relevance for stable.

Sure, I'll drop it.

Do you have an idea why the centos and ubuntu folks might have backported this commit into their kernels?

Quoted 9 lines
Quoted 6 lines
Quoted 2 lines

No clue, I trust they have their own reasons.

I cant see centos backporting anything unless they were asked to do so. And this really looks like a "bugfix" to me, why isn't this relevant for any older kernel versions?

Yeah seriously, this looks entirely legit.

I can assure you this doesn't fix any actual bug, and there's zero user impact from _not_ having this patch.

author maintainer
#212 lore raw

PATCH 205/205 ath9k: Fix a locking bug in ath9k_add_interface()

In reply to Sasha Levin

From: Dan Carpenter <[email protected]>

[ Upstream commit 461cf036057477805a8a391e5fd0f5264a5e56a8 ]

We tried to revert commit d9c52fd17cb4 ("ath9k: fix tx99 with monitor mode interface") but accidentally missed part of the locking change.

The lock has to be held earlier so that we're holding it when we do "sc->tx99_vif = vif;" and also there in the current code there is a stray unlock before we have taken the lock.

Fixes 6df0580be8bc ("ath9k: add back support for using active monitor interfaces for tx99")
---
 drivers/net/wireless/ath/ath9k/main.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/net/wireless/ath/ath9k/main.c +1 -2
index c85f613e8ceb5..74f98bbaea889 100644
@@ -1251,6 +1251,7 @@ static int ath9k_add_interface(struct ieee80211_hw *hw,
 	struct ath_vif *avp = (void *)vif->drv_priv;
 	struct ath_node *an = &avp->mcast_node;
 
+	mutex_lock(&sc->mutex);
 	if (IS_ENABLED(CONFIG_ATH9K_TX99)) {
 		if (sc->cur_chan->nvifs >= 1) {
 			mutex_unlock(&sc->mutex);
@@ -1259,8 +1260,6 @@ static int ath9k_add_interface(struct ieee80211_hw *hw,
 		sc->tx99_vif = vif;
 	}
 
-	mutex_lock(&sc->mutex);
-
 	ath_dbg(common, CONFIG, "Attach a VIF of type: %d\n", vif->type);
 	sc->cur_chan->nvifs++;
 
Signature · 1 line
2.20.1