Skip to contentSkip to conversation

5.5.9-stable review

PATCH In review Greg Kroah-Hartman 7 participants 206 messages Mar 10, 2020 – Mar 25, 2020 Page 2 of 2

Subsystems (4)
lore
Share & tools
lore
author maintainer
#164 lore raw

PATCH 161/189 RDMA/odp: Ensure the mm is still alive before creating an implicit child

In reply to Greg Kroah-Hartman

From: Jason Gunthorpe <[email protected]>

commit a4e63bce1414df7ab6eb82ca9feb8494ce13e554 upstream.

Registration of a mmu_notifier requires the caller to hold a mmget() on the mm as registration is not permitted to race with exit_mmap(). There is a BUG_ON inside the mmu_notifier to guard against this.

Normally creating a umem is done against current which implicitly holds the mmget(), however an implicit ODP child is created from a pagefault work queue and is not guaranteed to have a mmget().

Call mmget() around this registration and abort faulting if the MM has gone to exit_mmap().

Before the patch below the notifier was registered when the implicit ODP parent was created, so there was no chance to register a notifier outside of current.

Fixes c571feca2dc9 ("RDMA/odp: use mmu_notifier_get/put for 'struct ib_ucontext_per_mm'")
---
 drivers/infiniband/core/umem_odp.c |   24 +++++++++++++++++++-----
 1 file changed, 19 insertions(+), 5 deletions(-)
drivers/infiniband/core/umem_odp.c +19 -5
@@ -187,14 +187,28 @@ ib_umem_odp_alloc_child(struct ib_umem_o
 	odp_data->page_shift = PAGE_SHIFT;
 	odp_data->notifier.ops = ops;
 
+	/*
+	 * A mmget must be held when registering a notifier, the owming_mm only
+	 * has a mm_grab at this point.
+	 */
+	if (!mmget_not_zero(umem->owning_mm)) {
+		ret = -EFAULT;
+		goto out_free;
+	}
+
 	odp_data->tgid = get_pid(root->tgid);
 	ret = ib_init_umem_odp(odp_data, ops);
-	if (ret) {
-		put_pid(odp_data->tgid);
-		kfree(odp_data);
-		return ERR_PTR(ret);
-	}
+	if (ret)
+		goto out_tgid;
+	mmput(umem->owning_mm);
 	return odp_data;
+
+out_tgid:
+	put_pid(odp_data->tgid);
+	mmput(umem->owning_mm);
+out_free:
+	kfree(odp_data);
+	return ERR_PTR(ret);
 }
 EXPORT_SYMBOL(ib_umem_odp_alloc_child);
 
author maintainer
#165 lore raw

PATCH 162/189 RDMA/nldev: Fix crash when set a QP to a new counter but QPN is missing

In reply to Greg Kroah-Hartman

From: Mark Zhang <[email protected]>

commit 78f34a16c28654cb47791257006f90d0948f2f0c upstream.

This fixes the kernel crash when a RDMA_NLDEV_CMD_STAT_SET command is received, but the QP number parameter is not available.

iwpm_register_pid: Unable to send a nlmsg (client = 2) infiniband syz1: RDMA CMA: cma_listen_on_dev, error -98

  general protection fault, probably for non-canonical address 0xdffffc0000000000: 0000 [#1] PREEMPT SMP KASAN
  KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007]
  CPU: 0 PID: 9754 Comm: syz-executor069 Not tainted 5.6.0-rc2-syzkaller #0
  Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
  RIP: 0010:nla_get_u32 include/net/netlink.h:1474 [inline]
  RIP: 0010:nldev_stat_set_doit+0x63c/0xb70 drivers/infiniband/core/nldev.c:1760
  Code: fc 01 0f 84 58 03 00 00 e8 41 83 bf fb 4c 8b a3 58 fd ff ff 48 b8 00 00 00 00 00 fc ff df 49 8d 7c 24 04 48 89 fa 48 c1 ea 03 <0f> b6 14 02 48 89 f8 83 e0 07 83 c0 03 38 d0 7c 08 84 d2 0f 85 6d
  RSP: 0018:ffffc900068bf350 EFLAGS: 00010247
  RAX: dffffc0000000000 RBX: ffffc900068bf728 RCX: ffffffff85b60470
  RDX: 0000000000000000 RSI: ffffffff85b6047f RDI: 0000000000000004
  RBP: ffffc900068bf750 R08: ffff88808c3ee140 R09: ffff8880a25e6010
  R10: ffffed10144bcddc R11: ffff8880a25e6ee3 R12: 0000000000000000
  R13: ffff88809acb0000 R14: ffff888092a42c80 R15: 000000009ef2e29a
  FS:  0000000001ff0880(0000) GS:ffff8880ae800000(0000) knlGS:0000000000000000
  CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  CR2: 00007f4733e34000 CR3: 00000000a9b27000 CR4: 00000000001406f0
  DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
  DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
  Call Trace:
    rdma_nl_rcv_msg drivers/infiniband/core/netlink.c:195 [inline]
    rdma_nl_rcv_skb drivers/infiniband/core/netlink.c:239 [inline]
    rdma_nl_rcv+0x5d9/0x980 drivers/infiniband/core/netlink.c:259
    netlink_unicast_kernel net/netlink/af_netlink.c:1303 [inline]
    netlink_unicast+0x59e/0x7e0 net/netlink/af_netlink.c:1329
    netlink_sendmsg+0x91c/0xea0 net/netlink/af_netlink.c:1918
    sock_sendmsg_nosec net/socket.c:652 [inline]
    sock_sendmsg+0xd7/0x130 net/socket.c:672
    ____sys_sendmsg+0x753/0x880 net/socket.c:2343
    ___sys_sendmsg+0x100/0x170 net/socket.c:2397
    __sys_sendmsg+0x105/0x1d0 net/socket.c:2430
    __do_sys_sendmsg net/socket.c:2439 [inline]
    __se_sys_sendmsg net/socket.c:2437 [inline]
    __x64_sys_sendmsg+0x78/0xb0 net/socket.c:2437
    do_syscall_64+0xfa/0x790 arch/x86/entry/common.c:294
    entry_SYSCALL_64_after_hwframe+0x49/0xbe
  RIP: 0033:0x4403d9
  Code: 18 89 d0 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 0f 83 fb 13 fc ff c3 66 2e 0f 1f 84 00 00 00 00
  RSP: 002b:00007ffc0efbc5c8 EFLAGS: 00000246 ORIG_RAX: 000000000000002e
  RAX: ffffffffffffffda RBX: 00000000004002c8 RCX: 00000000004403d9
  RDX: 0000000000000000 RSI: 0000000020000240 RDI: 0000000000000004
  RBP: 00000000006ca018 R08: 0000000000000008 R09: 00000000004002c8
  R10: 000000000000004a R11: 0000000000000246 R12: 0000000000401c60
  R13: 0000000000401cf0 R14: 0000000000000000 R15: 0000000000000000
Fixes b389327df905 ("RDMA/nldev: Allow counter manual mode configration through RDMA netlink")
Reported-by [email protected]
---
 drivers/infiniband/core/nldev.c |    2 ++
 1 file changed, 2 insertions(+)
drivers/infiniband/core/nldev.c +2 -0
@@ -1756,6 +1756,8 @@ static int nldev_stat_set_doit(struct sk
 		if (ret)
 			goto err_msg;
 	} else {
+		if (!tb[RDMA_NLDEV_ATTR_RES_LQPN])
+			goto err_msg;
 		qpn = nla_get_u32(tb[RDMA_NLDEV_ATTR_RES_LQPN]);
 		if (tb[RDMA_NLDEV_ATTR_STAT_COUNTER_ID]) {
 			cntn = nla_get_u32(tb[RDMA_NLDEV_ATTR_STAT_COUNTER_ID]);
author maintainer
#166 lore raw

PATCH 163/189 RDMA/siw: Fix failure handling during device creation

In reply to Greg Kroah-Hartman

From: Bernard Metzler <[email protected]>

commit 12e5eef0f4d8087ea7b559f6630be08ffea2d851 upstream.

A failing call to ib_device_set_netdev() during device creation caused system crash due to xa_destroy of uninitialized xarray hit by device deallocation. Fixed by moving xarray initialization before potential device deallocation.

Fixes bdcf26bf9b3a ("rdma/siw: network and RDMA core interface")
Reported-by [email protected]
---
 drivers/infiniband/sw/siw/siw_main.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/infiniband/sw/siw/siw_main.c +3 -3
@@ -388,6 +388,9 @@ static struct siw_device *siw_device_cre
 		{ .max_segment_size = SZ_2G };
 	base_dev->num_comp_vectors = num_possible_cpus();
 
+	xa_init_flags(&sdev->qp_xa, XA_FLAGS_ALLOC1);
+	xa_init_flags(&sdev->mem_xa, XA_FLAGS_ALLOC1);
+
 	ib_set_device_ops(base_dev, &siw_device_ops);
 	rv = ib_device_set_netdev(base_dev, netdev, 1);
 	if (rv)
@@ -415,9 +418,6 @@ static struct siw_device *siw_device_cre
 	sdev->attrs.max_srq_wr = SIW_MAX_SRQ_WR;
 	sdev->attrs.max_srq_sge = SIW_MAX_SGE;
 
-	xa_init_flags(&sdev->qp_xa, XA_FLAGS_ALLOC1);
-	xa_init_flags(&sdev->mem_xa, XA_FLAGS_ALLOC1);
-
 	INIT_LIST_HEAD(&sdev->cep_list);
 	INIT_LIST_HEAD(&sdev->qp_list);
 
author maintainer
#167 lore raw

PATCH 164/189 RDMA/iwcm: Fix iwcm work deallocation

In reply to Greg Kroah-Hartman

From: Bernard Metzler <[email protected]>

commit 810dbc69087b08fd53e1cdd6c709f385bc2921ad upstream.

The dealloc_work_entries() function must update the work_free_list pointer while freeing its entries, since potentially called again on same list. A second iteration of the work list caused system crash. This happens, if work allocation fails during cma_iw_listen() and free_cm_id() tries to free the list again during cleanup.

---
 drivers/infiniband/core/iwcm.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/infiniband/core/iwcm.c +3 -1
@@ -159,8 +159,10 @@ static void dealloc_work_entries(struct
 {
 	struct list_head *e, *tmp;
 
-	list_for_each_safe(e, tmp, &cm_id_priv->work_free_list)
+	list_for_each_safe(e, tmp, &cm_id_priv->work_free_list) {
+		list_del(e);
 		kfree(list_entry(e, struct iwcm_work, free_list));
+	}
 }
 
 static int alloc_work_entries(struct iwcm_id_private *cm_id_priv, int count)
author maintainer
#168 lore raw

PATCH 165/189 RDMA/core: Fix protection fault in ib_mr_pool_destroy

In reply to Greg Kroah-Hartman

From: Maor Gottlieb <[email protected]>

commit e38b55ea0443da35a50a3eb2079ad3612cf763b9 upstream.

Fix NULL pointer dereference in the error flow of ib_create_qp_user when accessing to uninitialized list pointers - rdma_mrs and sig_mrs. The following crash from syzkaller revealed it.

kasan: GPF could be caused by NULL-ptr deref or user memory access

  general protection fault: 0000 [#1] SMP KASAN PTI
  CPU: 1 PID: 23167 Comm: syz-executor.1 Not tainted 5.5.0-rc5 #2
  Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
  rel-1.12.1-0-ga5cab58e9a3f-prebuilt.qemu.org 04/01/2014
  RIP: 0010:ib_mr_pool_destroy+0x81/0x1f0
  Code: 00 00 fc ff df 49 c1 ec 03 4d 01 fc e8 a8 ea 72 fe 41 80 3c 24 00
  0f 85 62 01 00 00 48 8b 13 48 89 d6 4c 8d 6a c8 48 c1 ee 03 <42> 80 3c
  3e 00 0f 85 34 01 00 00 48 8d 7a 08 4c 8b 02 48 89 fe 48
  RSP: 0018:ffffc9000951f8b0 EFLAGS: 00010046
  RAX: 0000000000040000 RBX: ffff88810f268038 RCX: ffffffff82c41628
  RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffffc9000951f850
  RBP: ffff88810f268020 R08: 0000000000000004 R09: fffff520012a3f0a
  R10: 0000000000000001 R11: fffff520012a3f0a R12: ffffed1021e4d007
  R13: ffffffffffffffc8 R14: 0000000000000246 R15: dffffc0000000000
  FS:  00007f54bc788700(0000) GS:ffff88811b100000(0000)
  knlGS:0000000000000000
  CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  CR2: 0000000000000000 CR3: 0000000116920002 CR4: 0000000000360ee0
  DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
  DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
  Call Trace:
   rdma_rw_cleanup_mrs+0x15/0x30
   ib_destroy_qp_user+0x674/0x7d0
   ib_create_qp_user+0xb01/0x11c0
   create_qp+0x1517/0x2130
   ib_uverbs_create_qp+0x13e/0x190
   ib_uverbs_write+0xaa5/0xdf0
   __vfs_write+0x7c/0x100
   vfs_write+0x168/0x4a0
   ksys_write+0xc8/0x200
   do_syscall_64+0x9c/0x390
   entry_SYSCALL_64_after_hwframe+0x44/0xa9
  RIP: 0033:0x465b49
  Code: f7 d8 64 89 02 b8 ff ff ff ff c3 66 0f 1f 44 00 00 48 89 f8 48 89
  f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01
  f0 ff ff 73 01 c3 48 c7 c1 bc ff ff ff f7 d8 64 89 01 48
  RSP: 002b:00007f54bc787c58 EFLAGS: 00000246 ORIG_RAX: 0000000000000001
  RAX: ffffffffffffffda RBX: 000000000073bf00 RCX: 0000000000465b49
  RDX: 0000000000000040 RSI: 0000000020000540 RDI: 0000000000000003
  RBP: 00007f54bc787c70 R08: 0000000000000000 R09: 0000000000000000
  R10: 0000000000000000 R11: 0000000000000246 R12: 00007f54bc7886bc
  R13: 00000000004ca2ec R14: 000000000070ded0 R15: 0000000000000005
---
 drivers/infiniband/core/core_priv.h  |   15 +++++++++++++++
 drivers/infiniband/core/uverbs_cmd.c |   10 ----------
 drivers/infiniband/core/verbs.c      |   10 ----------
 3 files changed, 15 insertions(+), 20 deletions(-)
drivers/infiniband/core/core_priv.h +15 -0
@@ -338,6 +338,21 @@ static inline struct ib_qp *_ib_create_q
 	qp->pd = pd;
 	qp->uobject = uobj;
 	qp->real_qp = qp;
+
+	qp->qp_type = attr->qp_type;
+	qp->qp_context = attr->qp_context;
+	qp->rwq_ind_tbl = attr->rwq_ind_tbl;
+	qp->send_cq = attr->send_cq;
+	qp->recv_cq = attr->recv_cq;
+	qp->srq = attr->srq;
+	qp->rwq_ind_tbl = attr->rwq_ind_tbl;
+	qp->event_handler = attr->event_handler;
+
+	atomic_set(&qp->usecnt, 0);
+	spin_lock_init(&qp->mr_lock);
+	INIT_LIST_HEAD(&qp->rdma_mrs);
+	INIT_LIST_HEAD(&qp->sig_mrs);
+
 	/*
 	 * We don't track XRC QPs for now, because they don't have PD
 	 * and more importantly they are created internaly by driver,
drivers/infiniband/core/uverbs_cmd.c +0 -10
@@ -1433,17 +1433,7 @@ static int create_qp(struct uverbs_attr_
 		if (ret)
 			goto err_cb;
 
-		qp->pd		  = pd;
-		qp->send_cq	  = attr.send_cq;
-		qp->recv_cq	  = attr.recv_cq;
-		qp->srq		  = attr.srq;
-		qp->rwq_ind_tbl	  = ind_tbl;
-		qp->event_handler = attr.event_handler;
-		qp->qp_context	  = attr.qp_context;
-		qp->qp_type	  = attr.qp_type;
-		atomic_set(&qp->usecnt, 0);
 		atomic_inc(&pd->usecnt);
-		qp->port = 0;
 		if (attr.send_cq)
 			atomic_inc(&attr.send_cq->usecnt);
 		if (attr.recv_cq)
drivers/infiniband/core/verbs.c +0 -10
@@ -1182,16 +1182,6 @@ struct ib_qp *ib_create_qp_user(struct i
 	if (ret)
 		goto err;
 
-	qp->qp_type    = qp_init_attr->qp_type;
-	qp->rwq_ind_tbl = qp_init_attr->rwq_ind_tbl;
-
-	atomic_set(&qp->usecnt, 0);
-	qp->mrs_used = 0;
-	spin_lock_init(&qp->mr_lock);
-	INIT_LIST_HEAD(&qp->rdma_mrs);
-	INIT_LIST_HEAD(&qp->sig_mrs);
-	qp->port = 0;
-
 	if (qp_init_attr->qp_type == IB_QPT_XRC_TGT) {
 		struct ib_qp *xrc_qp =
 			create_xrc_qp_user(qp, qp_init_attr, udata);
author maintainer
#169 lore raw

PATCH 166/189 regulator: stm32-vrefbuf: fix a possible overshoot when re-enabling

In reply to Greg Kroah-Hartman

From: Fabrice Gasnier <[email protected]>

commit 02fbabd5f4ed182d2c616e49309f5a3efd9ec671 upstream.

There maybe an overshoot, when disabling, then re-enabling vrefbuf too quickly. VREFBUF is used by ADC/DAC on some boards. When re-enabling too quickly, an overshoot on the reference voltage make the conversions inaccurate for a short period of time.
- Don't put the VREFBUF in HiZ when disabling, to force an active discharge.
- Enforce a 1ms OFF/ON delay

Fixes 0cdbf481e927 ("regulator: Add support for stm32-vrefbuf")

Message-Id: <[email protected]>

---
 drivers/regulator/stm32-vrefbuf.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/regulator/stm32-vrefbuf.c +2 -1
@@ -88,7 +88,7 @@ static int stm32_vrefbuf_disable(struct
 	}
 
 	val = readl_relaxed(priv->base + STM32_VREFBUF_CSR);
-	val = (val & ~STM32_ENVR) | STM32_HIZ;
+	val &= ~STM32_ENVR;
 	writel_relaxed(val, priv->base + STM32_VREFBUF_CSR);
 
 	pm_runtime_mark_last_busy(priv->dev);
@@ -175,6 +175,7 @@ static const struct regulator_desc stm32
 	.volt_table = stm32_vrefbuf_voltages,
 	.n_voltages = ARRAY_SIZE(stm32_vrefbuf_voltages),
 	.ops = &stm32_vrefbuf_volt_ops,
+	.off_on_delay = 1000,
 	.type = REGULATOR_VOLTAGE,
 	.owner = THIS_MODULE,
 };
author maintainer
#170 lore raw

PATCH 167/189 regulator: qcom_spmi: Fix docs for PM8004

In reply to Greg Kroah-Hartman

From: Petr Vorel <[email protected]>

commit a5b0cda136f4f420a8e24e50d19dfcef2f81df2e upstream.

Fixes 2e36e140b8b8 ("regulator: qcom_spmi: Add support for PM8004 regulators")
---
 Documentation/devicetree/bindings/regulator/qcom,spmi-regulator.txt |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
Documentation/devicetree/bindings/regulator/qcom,spmi-regulator.txt +1 -1
@@ -161,7 +161,7 @@ The regulator node houses sub-nodes for
 sub-node is identified using the node's name, with valid values listed for each
 of the PMICs below.
 
-pm8005:
+pm8004:
 	s2, s5
 
 pm8005:
author maintainer
#171 lore raw

PATCH 168/189 RMDA/cm: Fix missing ib_cm_destroy_id() in ib_cm_insert_listen()

In reply to Greg Kroah-Hartman

From: Jason Gunthorpe <[email protected]>

commit c14dfddbd869bf0c2bafb7ef260c41d9cebbcfec upstream.

The algorithm pre-allocates a cm_id since allocation cannot be done while holding the cm.lock spinlock, however it doesn't free it on one error path, leading to a memory leak.

Fixes 067b171b8679 ("IB/cm: Share listening CM IDs")
---
 drivers/infiniband/core/cm.c |    1 +
 1 file changed, 1 insertion(+)
drivers/infiniband/core/cm.c +1 -0
@@ -1202,6 +1202,7 @@ struct ib_cm_id *ib_cm_insert_listen(str
 			/* Sharing an ib_cm_id with different handlers is not
 			 * supported */
 			spin_unlock_irqrestore(&cm.lock, flags);
+			ib_destroy_cm_id(cm_id);
 			return ERR_PTR(-EINVAL);
 		}
 		refcount_inc(&cm_id_priv->refcount);
author maintainer
#172 lore raw

PATCH 169/189 IB/mlx5: Fix implicit ODP race

In reply to Greg Kroah-Hartman

From: Artemy Kovalyov <[email protected]>

commit de5ed007a03d71daaa505f5daa4d3666530c7090 upstream.

Following race may occur because of the call_srcu and the placement of the synchronize_srcu vs the xa_erase.

CPU0 CPU1

mlx5_ib_free_implicit_mr: destroy_unused_implicit_child_mr:
xa_erase(odp_mkeys)
synchronize_srcu()

				    xa_lock(implicit_children)
				    if (still in xarray)
				       atomic_inc()
				       call_srcu()
				    xa_unlock(implicit_children)

xa_erase(implicit_children):
xa_lock(implicit_children)
__xa_erase()
xa_unlock(implicit_children)

flush_workqueue()

				   [..]
				    free_implicit_child_mr_rcu:
				     (via call_srcu)
				      queue_work()

WARN_ON(atomic_read())

				   [..]
				    free_implicit_child_mr_work:
				     (via wq)
				      free_implicit_child_mr()

mlx5_mr_cache_invalidate()
mlx5_ib_update_xlt() <-- UMR QP fail atomic_dec()

The wait_event() solves the race because it blocks until free_implicit_child_mr_work() completes.

---
 drivers/infiniband/hw/mlx5/mlx5_ib.h |    1 +
 drivers/infiniband/hw/mlx5/odp.c     |   17 +++++++----------
 2 files changed, 8 insertions(+), 10 deletions(-)
drivers/infiniband/hw/mlx5/mlx5_ib.h +1 -0
@@ -629,6 +629,7 @@ struct mlx5_ib_mr {
 
 	/* For ODP and implicit */
 	atomic_t		num_deferred_work;
+	wait_queue_head_t       q_deferred_work;
 	struct xarray		implicit_children;
 	union {
 		struct rcu_head rcu;
drivers/infiniband/hw/mlx5/odp.c +7 -10
@@ -197,7 +197,8 @@ static void free_implicit_child_mr(struc
 	mr->parent = NULL;
 	mlx5_mr_cache_free(mr->dev, mr);
 	ib_umem_odp_release(odp);
-	atomic_dec(&imr->num_deferred_work);
+	if (atomic_dec_and_test(&imr->num_deferred_work))
+		wake_up(&imr->q_deferred_work);
 }
 
 static void free_implicit_child_mr_work(struct work_struct *work)
@@ -516,6 +517,7 @@ struct mlx5_ib_mr *mlx5_ib_alloc_implici
 	imr->umem = &umem_odp->umem;
 	imr->is_odp_implicit = true;
 	atomic_set(&imr->num_deferred_work, 0);
+	init_waitqueue_head(&imr->q_deferred_work);
 	xa_init(&imr->implicit_children);
 
 	err = mlx5_ib_update_xlt(imr, 0,
@@ -573,10 +575,7 @@ void mlx5_ib_free_implicit_mr(struct mlx
 	 * under xa_lock while the child is in the xarray. Thus at this point
 	 * it is only decreasing, and all work holding it is now on the wq.
 	 */
-	if (atomic_read(&imr->num_deferred_work)) {
-		flush_workqueue(system_unbound_wq);
-		WARN_ON(atomic_read(&imr->num_deferred_work));
-	}
+	wait_event(imr->q_deferred_work, !atomic_read(&imr->num_deferred_work));
 
 	/*
 	 * Fence the imr before we destroy the children. This allows us to
@@ -607,10 +606,7 @@ void mlx5_ib_fence_odp_mr(struct mlx5_ib
 	/* Wait for all running page-fault handlers to finish. */
 	synchronize_srcu(&mr->dev->odp_srcu);
 
-	if (atomic_read(&mr->num_deferred_work)) {
-		flush_workqueue(system_unbound_wq);
-		WARN_ON(atomic_read(&mr->num_deferred_work));
-	}
+	wait_event(mr->q_deferred_work, !atomic_read(&mr->num_deferred_work));
 
 	dma_fence_odp_mr(mr);
 }
@@ -1682,7 +1678,8 @@ static void destroy_prefetch_work(struct
 	u32 i;
 
 	for (i = 0; i < work->num_sge; ++i)
-		atomic_dec(&work->frags[i].mr->num_deferred_work);
+		if (atomic_dec_and_test(&work->frags[i].mr->num_deferred_work))
+			wake_up(&work->frags[i].mr->q_deferred_work);
 	kvfree(work);
 }
 
author maintainer
#173 lore raw

PATCH 170/189 IB/hfi1, qib: Ensure RCU is locked when accessing list

In reply to Greg Kroah-Hartman

From: Dennis Dalessandro <[email protected]>

commit 817a68a6584aa08e323c64283fec5ded7be84759 upstream.

The packet handling function, specifically the iteration of the qp list for mad packet processing misses locking RCU before running through the list. Not only is this incorrect, but the list_for_each_entry_rcu() call can not be called with a conditional check for lock dependency. Remedy this by invoking the rcu lock and unlock around the critical section.

This brings MAD packet processing in line with what is done for non-MAD packets.

---
 drivers/infiniband/hw/hfi1/verbs.c    |    4 +++-
 drivers/infiniband/hw/qib/qib_verbs.c |    2 ++
 2 files changed, 5 insertions(+), 1 deletion(-)
drivers/infiniband/hw/hfi1/verbs.c +3 -1
@@ -515,10 +515,11 @@ static inline void hfi1_handle_packet(st
 				       opa_get_lid(packet->dlid, 9B));
 		if (!mcast)
 			goto drop;
+		rcu_read_lock();
 		list_for_each_entry_rcu(p, &mcast->qp_list, list) {
 			packet->qp = p->qp;
 			if (hfi1_do_pkey_check(packet))
-				goto drop;
+				goto unlock_drop;
 			spin_lock_irqsave(&packet->qp->r_lock, flags);
 			packet_handler = qp_ok(packet);
 			if (likely(packet_handler))
@@ -527,6 +528,7 @@ static inline void hfi1_handle_packet(st
 				ibp->rvp.n_pkt_drops++;
 			spin_unlock_irqrestore(&packet->qp->r_lock, flags);
 		}
+		rcu_read_unlock();
 		/*
 		 * Notify rvt_multicast_detach() if it is waiting for us
 		 * to finish.
drivers/infiniband/hw/qib/qib_verbs.c +2 -0
@@ -329,8 +329,10 @@ void qib_ib_rcv(struct qib_ctxtdata *rcd
 		if (mcast == NULL)
 			goto drop;
 		this_cpu_inc(ibp->pmastats->n_multicast_rcv);
+		rcu_read_lock();
 		list_for_each_entry_rcu(p, &mcast->qp_list, list)
 			qib_qp_rcv(rcd, hdr, 1, data, tlen, p->qp);
+		rcu_read_unlock();
 		/*
 		 * Notify rvt_multicast_detach() if it is waiting for us
 		 * to finish.
author maintainer
#174 lore raw

PATCH 171/189 ARM: imx: build v7_cpu_resume() unconditionally

In reply to Greg Kroah-Hartman

From: Ahmad Fatoum <[email protected]>

commit 512a928affd51c2dc631401e56ad5ee5d5dd68b6 upstream.

This function is not only needed by the platform suspend code, but is also reused as the CPU resume function when the ARM cores can be powered down completely in deep idle, which is the case on i.MX6SX and i.MX6UL(L).

Providing the static inline stub whenever CONFIG_SUSPEND is disabled means that those platforms will hang on resume from cpuidle if suspend is disabled.

So there are two problems:

- The static inline stub masks the linker error
- The function is not available where needed

Fix both by just building the function unconditionally, when CONFIG_SOC_IMX6 is enabled. The actual code is three instructions long, so it's arguably ok to just leave it in for all i.MX6 kernel configurations.

Fixes 05136f0897b5 ("ARM: imx: support arm power off in cpuidle for i.mx6sx")
---
 arch/arm/mach-imx/Makefile       |    2 ++
 arch/arm/mach-imx/common.h       |    4 ++--
 arch/arm/mach-imx/resume-imx6.S  |   24 ++++++++++++++++++++++++
 arch/arm/mach-imx/suspend-imx6.S |   14 --------------
 4 files changed, 28 insertions(+), 16 deletions(-)
arch/arm/mach-imx/Makefile +2 -0
@@ -91,6 +91,8 @@ AFLAGS_suspend-imx6.o :=-Wa,-march=armv7
 obj-$(CONFIG_SOC_IMX6) += suspend-imx6.o
 obj-$(CONFIG_SOC_IMX53) += suspend-imx53.o
 endif
+AFLAGS_resume-imx6.o :=-Wa,-march=armv7-a
+obj-$(CONFIG_SOC_IMX6) += resume-imx6.o
 obj-$(CONFIG_SOC_IMX6) += pm-imx6.o
 
 obj-$(CONFIG_SOC_IMX1) += mach-imx1.o
arch/arm/mach-imx/common.h +2 -2
@@ -109,17 +109,17 @@ void imx_cpu_die(unsigned int cpu);
 int imx_cpu_kill(unsigned int cpu);
 
 #ifdef CONFIG_SUSPEND
-void v7_cpu_resume(void);
 void imx53_suspend(void __iomem *ocram_vbase);
 extern const u32 imx53_suspend_sz;
 void imx6_suspend(void __iomem *ocram_vbase);
 #else
-static inline void v7_cpu_resume(void) {}
 static inline void imx53_suspend(void __iomem *ocram_vbase) {}
 static const u32 imx53_suspend_sz;
 static inline void imx6_suspend(void __iomem *ocram_vbase) {}
 #endif
 
+void v7_cpu_resume(void);
+
 void imx6_pm_ccm_init(const char *ccm_compat);
 void imx6q_pm_init(void);
 void imx6dl_pm_init(void);
arch/arm/mach-imx/resume-imx6.S new file +24 -0
@@ -0,0 +1,24 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * Copyright 2014 Freescale Semiconductor, Inc.
+ */
+
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+#include <asm/asm-offsets.h>
+#include <asm/hardware/cache-l2x0.h>
+#include "hardware.h"
+
+/*
+ * The following code must assume it is running from physical address
+ * where absolute virtual addresses to the data section have to be
+ * turned into relative ones.
+ */
+
+ENTRY(v7_cpu_resume)
+	bl	v7_invalidate_l1
+#ifdef CONFIG_CACHE_L2X0
+	bl	l2c310_early_resume
+#endif
+	b	cpu_resume
+ENDPROC(v7_cpu_resume)
arch/arm/mach-imx/suspend-imx6.S +0 -14
@@ -327,17 +327,3 @@ resume:
 
 	ret	lr
 ENDPROC(imx6_suspend)
-
-/*
- * The following code must assume it is running from physical address
- * where absolute virtual addresses to the data section have to be
- * turned into relative ones.
- */
-
-ENTRY(v7_cpu_resume)
-	bl	v7_invalidate_l1
-#ifdef CONFIG_CACHE_L2X0
-	bl	l2c310_early_resume
-#endif
-	b	cpu_resume
-ENDPROC(v7_cpu_resume)
author maintainer
#175 lore raw

PATCH 172/189 ARM: dts: imx7d: fix opp-supported-hw

In reply to Greg Kroah-Hartman

From: Peng Fan <[email protected]>

commit 54d6477dca3b65b7b77a903fe60a9447bc836e7f upstream.

Per i.MX7D Document Number: IMX7DCEC Rev. 6, 03/2019, there are only consumer/industrial parts, and 1.2GHz is only support in consumer parts.

So exclude automotive from 792/996MHz/1.2GHz and exclude industrial from 1.2GHz.

Fixes d7bfba7296ca ("ARM: dts: imx7d: Update cpufreq OPP table")
Cc Leonard Crestez <[email protected]>
---
 arch/arm/boot/dts/imx7d.dtsi |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
arch/arm/boot/dts/imx7d.dtsi +3 -3
@@ -44,7 +44,7 @@
 			opp-hz = /bits/ 64 <792000000>;
 			opp-microvolt = <1000000>;
 			clock-latency-ns = <150000>;
-			opp-supported-hw = <0xd>, <0xf>;
+			opp-supported-hw = <0xd>, <0x7>;
 			opp-suspend;
 		};
 
@@ -52,7 +52,7 @@
 			opp-hz = /bits/ 64 <996000000>;
 			opp-microvolt = <1100000>;
 			clock-latency-ns = <150000>;
-			opp-supported-hw = <0xc>, <0xf>;
+			opp-supported-hw = <0xc>, <0x7>;
 			opp-suspend;
 		};
 
@@ -60,7 +60,7 @@
 			opp-hz = /bits/ 64 <1200000000>;
 			opp-microvolt = <1225000>;
 			clock-latency-ns = <150000>;
-			opp-supported-hw = <0x8>, <0xf>;
+			opp-supported-hw = <0x8>, <0x3>;
 			opp-suspend;
 		};
 	};
author maintainer
#176 lore raw

PATCH 173/189 ARM: dts: am437x-idk-evm: Fix incorrect OPP node names

In reply to Greg Kroah-Hartman

From: Suman Anna <[email protected]>

commit 31623468be0bf57617b8057dcd335693935a9491 upstream.

The commit 337c6c9a69af ("ARM: dts: am437x-idk-evm: Disable OPP50 for MPU") adjusts couple of OPP nodes defined in the common am4372.dtsi file, but used outdated node names. This results in these getting treated as new OPP nodes with missing properties.

Fix this properly by using the correct node names as updated in commit b9cb2ba71848 ("ARM: dts: Use - instead of @ for DT OPP entries for TI SoCs").

Fixes 337c6c9a69af ("ARM: dts: am437x-idk-evm: Disable OPP50 for MPU")
---
 arch/arm/boot/dts/am437x-idk-evm.dts |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/arm/boot/dts/am437x-idk-evm.dts +2 -2
@@ -526,11 +526,11 @@
 	 * Supply voltage supervisor on board will not allow opp50 so
 	 * disable it and set opp100 as suspend OPP.
 	 */
-	opp50@300000000 {
+	opp50-300000000 {
 		status = "disabled";
 	};
 
-	opp100@600000000 {
+	opp100-600000000 {
 		opp-suspend;
 	};
 };
author maintainer
#177 lore raw

PATCH 174/189 ARM: dts: dra7xx-clocks: Fixup IPU1 mux clock parent source

In reply to Greg Kroah-Hartman

From: Suman Anna <[email protected]>

commit 78722d37b2b4cf9178295e2aa5510880e6135fd7 upstream.

The IPU1 functional clock is the output of a mux clock (represented by ipu1_gfclk_mux previously) and the clock source for this has been updated to be sourced from dpll_core_h22x2_ck in commit 39879c7d963e ("ARM: dts: dra7xx-clocks: Source IPU1 functional clock from CORE DPLL"). ipu1_gfclk_mux is an obsolete clock now with the clkctrl conversion, and this clock source parenting is lost during the new clkctrl layout conversion.

Remove this stale clock and fix up the clock source for this mux clock using the latest equivalent clkctrl clock. This restores the previous logic and ensures that the IPU1 continues to run at the same frequency of IPU2 and independent of the ABE DPLL.

Fixes b5f8ffbb6fad ("ARM: dts: dra7: convert to use new clkctrl layout")
---
 arch/arm/boot/dts/dra7xx-clocks.dtsi |   12 ++----------
 1 file changed, 2 insertions(+), 10 deletions(-)
arch/arm/boot/dts/dra7xx-clocks.dtsi +2 -10
@@ -796,16 +796,6 @@
 		clock-div = <1>;
 	};
 
-	ipu1_gfclk_mux: ipu1_gfclk_mux@520 {
-		#clock-cells = <0>;
-		compatible = "ti,mux-clock";
-		clocks = <&dpll_abe_m2x2_ck>, <&dpll_core_h22x2_ck>;
-		ti,bit-shift = <24>;
-		reg = <0x0520>;
-		assigned-clocks = <&ipu1_gfclk_mux>;
-		assigned-clock-parents = <&dpll_core_h22x2_ck>;
-	};
-
 	dummy_ck: dummy_ck {
 		#clock-cells = <0>;
 		compatible = "fixed-clock";
@@ -1564,6 +1554,8 @@
 			compatible = "ti,clkctrl";
 			reg = <0x20 0x4>;
 			#clock-cells = <2>;
+			assigned-clocks = <&ipu1_clkctrl DRA7_IPU1_MMU_IPU1_CLKCTRL 24>;
+			assigned-clock-parents = <&dpll_core_h22x2_ck>;
 		};
 
 		ipu_clkctrl: ipu-clkctrl@50 {
author maintainer
#178 lore raw

PATCH 175/189 ARM: dts: imx6dl-colibri-eval-v3: fix sram compatible properties

In reply to Greg Kroah-Hartman

From: Johan Hovold <[email protected]>

commit bcbf53a0dab50980867476994f6079c1ec5bb3a3 upstream.

The sram-node compatible properties have mistakingly combined the model-specific string with the generic "mtd-ram" string.

Note that neither "cy7c1019dv33-10zsxi, mtd-ram" or "cy7c1019dv33-10zsxi" are used by any in-kernel driver and they are not present in any binding.

The physmap driver will however bind to platform devices that specify "mtd-ram".

Fixes fc48e76489fd ("ARM: dts: imx6: Add support for Toradex Colibri iMX6 module")
Cc Sanchayan Maity <[email protected]>
Cc Marcel Ziswiler <[email protected]>
Cc Shawn Guo <[email protected]>
---
 arch/arm/boot/dts/imx6dl-colibri-eval-v3.dts |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/arm/boot/dts/imx6dl-colibri-eval-v3.dts +2 -2
@@ -275,7 +275,7 @@
 
 	/* SRAM on Colibri nEXT_CS0 */
 	sram@0,0 {
-		compatible = "cypress,cy7c1019dv33-10zsxi, mtd-ram";
+		compatible = "cypress,cy7c1019dv33-10zsxi", "mtd-ram";
 		reg = <0 0 0x00010000>;
 		#address-cells = <1>;
 		#size-cells = <1>;
@@ -286,7 +286,7 @@
 
 	/* SRAM on Colibri nEXT_CS1 */
 	sram@1,0 {
-		compatible = "cypress,cy7c1019dv33-10zsxi, mtd-ram";
+		compatible = "cypress,cy7c1019dv33-10zsxi", "mtd-ram";
 		reg = <1 0 0x00010000>;
 		#address-cells = <1>;
 		#size-cells = <1>;
#179 lore raw

Re: [PATCH 5.5 175/189] ARM: dts: imx6dl-colibri-eval-v3: fix sram compatible properties

Quoted 14 lines · Greg Kroah-Hartman

From: Johan Hovold <[email protected]>

commit bcbf53a0dab50980867476994f6079c1ec5bb3a3 upstream.

The sram-node compatible properties have mistakingly combined the model-specific string with the generic "mtd-ram" string.

Note that neither "cy7c1019dv33-10zsxi, mtd-ram" or "cy7c1019dv33-10zsxi" are used by any in-kernel driver and they are not present in any binding.

The physmap driver will however bind to platform devices that specify "mtd-ram".

Quoted 8 lines
Fixes fc48e76489fd ("ARM: dts: imx6: Add support for Toradex Colibri iMX6 module")
Cc Sanchayan Maity <[email protected]>
Cc Marcel Ziswiler <[email protected]>
Cc Shawn Guo <[email protected]>

This was never meant to go into stable so I didn't add a stable CC-tag.

It causes a driver to bind to the corresponding platform devices, which have so far been unbound and may therefore have unwanted side-effects.

I don't think it's stable material either way.

Quoted 18 lines
---
 arch/arm/boot/dts/imx6dl-colibri-eval-v3.dts |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/arm/boot/dts/imx6dl-colibri-eval-v3.dts +1 -1
@@ -275,7 +275,7 @@
 
 	/* SRAM on Colibri nEXT_CS0 */
 	sram@0,0 {
-		compatible = "cypress,cy7c1019dv33-10zsxi, mtd-ram";
+		compatible = "cypress,cy7c1019dv33-10zsxi", "mtd-ram";
 		reg = <0 0 0x00010000>;
 		#address-cells = <1>;
 		#size-cells = <1>;

arch/arm/boot/dts/imx6dl-colibri-eval-v3.dts line 292

Quoted 8 lines
arch/arm/boot/dts/imx6dl-colibri-eval-v3.dts +1 -1
@@ -286,7 +286,7 @@
 
 	/* SRAM on Colibri nEXT_CS1 */
 	sram@1,0 {
-		compatible = "cypress,cy7c1019dv33-10zsxi, mtd-ram";
+		compatible = "cypress,cy7c1019dv33-10zsxi", "mtd-ram";
 		reg = <1 0 0x00010000>;
 		#address-cells = <1>;
 		#size-cells = <1>;

Johan

author maintainer
#180 lore raw

Re: [PATCH 5.5 175/189] ARM: dts: imx6dl-colibri-eval-v3: fix sram compatible properties

Quoted 24 lines · Johan Hovold
Quoted 14 lines · Greg Kroah-Hartman

From: Johan Hovold <[email protected]>

commit bcbf53a0dab50980867476994f6079c1ec5bb3a3 upstream.

The sram-node compatible properties have mistakingly combined the model-specific string with the generic "mtd-ram" string.

Note that neither "cy7c1019dv33-10zsxi, mtd-ram" or "cy7c1019dv33-10zsxi" are used by any in-kernel driver and they are not present in any binding.

The physmap driver will however bind to platform devices that specify "mtd-ram".

Quoted 8 lines
Fixes fc48e76489fd ("ARM: dts: imx6: Add support for Toradex Colibri iMX6 module")
Cc Sanchayan Maity <[email protected]>
Cc Marcel Ziswiler <[email protected]>
Cc Shawn Guo <[email protected]>
Quoted 6 lines

This was never meant to go into stable so I didn't add a stable CC-tag.

It causes a driver to bind to the corresponding platform devices, which have so far been unbound and may therefore have unwanted side-effects.

I don't think it's stable material either way.

Thanks, now dropped from all kernel trees.

greg k-h

author maintainer
#181 lore raw

PATCH 176/189 ARM: dts: imx7-colibri: Fix frequency for sd/mmc

In reply to Greg Kroah-Hartman

From: Oleksandr Suvorov <[email protected]>

commit 2773fe1d31c42ffae2a9cb9a6055d99dd86e2fee upstream.

SD/MMC on Colibri iMX7S/D modules successfully support 200Mhz frequency in HS200 mode.

Removing the unnecessary max-frequency limit significantly increases the performance:

== before fix ====

root@colibri-imx7-emmc:~# hdparm -t /dev/mmcblk0
/dev/mmcblk0:
 Timing buffered disk reads: 252 MB in  3.02 seconds =  83.54 MB/sec
==================

=== after fix ====

root@colibri-imx7-emmc:~# hdparm -t /dev/mmcblk0
/dev/mmcblk0:
 Timing buffered disk reads: 408 MB in  3.00 seconds = 135.94 MB/sec
==================
Fixes f928a4a377e4 ("ARM: dts: imx7: add Toradex Colibri iMX7D 1GB (eMMC) support")
---
 arch/arm/boot/dts/imx7-colibri.dtsi |    1 -
 1 file changed, 1 deletion(-)
arch/arm/boot/dts/imx7-colibri.dtsi +0 -1
@@ -336,7 +336,6 @@
 	assigned-clock-rates = <400000000>;
 	bus-width = <8>;
 	fsl,tuning-step = <2>;
-	max-frequency = <100000000>;
 	vmmc-supply = <&reg_module_3v3>;
 	vqmmc-supply = <&reg_DCDC3>;
 	non-removable;
author maintainer
#182 lore raw

PATCH 177/189 hwmon: (adt7462) Fix an error return in ADT7462_REG_VOLT()

In reply to Greg Kroah-Hartman

From: Dan Carpenter <[email protected]>

commit 44f2f882909fedfc3a56e4b90026910456019743 upstream.

This is only called from adt7462_update_device(). The caller expects it to return zero on error. I fixed a similar issue earlier in commit a4bf06d58f21 ("hwmon: (adt7462) ADT7462_REG_VOLT_MAX() should return 0") but I missed this one.

---
 drivers/hwmon/adt7462.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/hwmon/adt7462.c +1 -1
@@ -413,7 +413,7 @@ static int ADT7462_REG_VOLT(struct adt74
 			return 0x95;
 		break;
 	}
-	return -ENODEV;
+	return 0;
 }
 
 /* Provide labels for sysfs */
author maintainer
#183 lore raw

PATCH 178/189 dma-buf: free dmabuf->name in dma_buf_release()

In reply to Greg Kroah-Hartman

From: Cong Wang <[email protected]>

commit d1f37226431f5d9657aa144a40f2383adbcf27e1 upstream.

dma-buf name can be set via DMA_BUF_SET_NAME ioctl, but once set it never gets freed.

Free it in dma_buf_release().

Fixes bb2bb9030425 ("dma-buf: add DMA_BUF_SET_NAME ioctls")
Reported-by [email protected]
Cc Greg Hackmann <[email protected]>
Cc Chenbo Feng <[email protected]>
Cc Sumit Semwal <[email protected]>
---
 drivers/dma-buf/dma-buf.c |    1 +
 1 file changed, 1 insertion(+)
drivers/dma-buf/dma-buf.c +1 -0
@@ -108,6 +108,7 @@ static int dma_buf_release(struct inode
 		dma_resv_fini(dmabuf->resv);
 
 	module_put(dmabuf->owner);
+	kfree(dmabuf->name);
 	kfree(dmabuf);
 	return 0;
 }
author maintainer
#184 lore raw

PATCH 179/189 dmaengine: coh901318: Fix a double lock bug in dma_tc_handle()

In reply to Greg Kroah-Hartman

From: Dan Carpenter <[email protected]>

commit 36d5d22090d13fd3a7a8c9663a711cbe6970aac8 upstream.

The caller is already holding the lock so this will deadlock.

Fixes 0b58828c923e ("DMAENGINE: COH 901 318 remove irq counting")
---
 drivers/dma/coh901318.c |    4 ----
 1 file changed, 4 deletions(-)
drivers/dma/coh901318.c +0 -4
@@ -1947,8 +1947,6 @@ static void dma_tc_handle(struct coh9013
 		return;
 	}
 
-	spin_lock(&cohc->lock);
-
 	/*
 	 * When we reach this point, at least one queue item
 	 * should have been moved over from cohc->queue to
@@ -1969,8 +1967,6 @@ static void dma_tc_handle(struct coh9013
 	if (coh901318_queue_start(cohc) == NULL)
 		cohc->busy = 0;
 
-	spin_unlock(&cohc->lock);
-
 	/*
 	 * This tasklet will remove items from cohc->active
 	 * and thus terminates them.
author maintainer
#185 lore raw

PATCH 180/189 sched/fair: Fix statistics for find_idlest_group()

In reply to Greg Kroah-Hartman

From: Vincent Guittot <[email protected]>

commit 289de35984815576793f579ec27248609e75976e upstream.

sgs->group_weight is not set while gathering statistics in update_sg_wakeup_stats(). This means that a group can be classified as fully busy with 0 running tasks if utilization is high enough.

This path is mainly used for fork and exec.

---
 kernel/sched/fair.c |    2 ++
 1 file changed, 2 insertions(+)
kernel/sched/fair.c +2 -0
@@ -8315,6 +8315,8 @@ static inline void update_sg_wakeup_stat
 
 	sgs->group_capacity = group->sgc->capacity;
 
+	sgs->group_weight = group->group_weight;
+
 	sgs->group_type = group_classify(sd->imbalance_pct, group, sgs);
 
 	/*
author maintainer
#186 lore raw

PATCH 181/189 arm64: dts: meson: fix gxm-khadas-vim2 wifi

In reply to Greg Kroah-Hartman

From: Christian Hewitt <[email protected]>

commit 146033562e7e5d1c9aae9653986806664995f1d5 upstream.

before

[6.418252] brcmfmac: F1 signature read @0x18000000=0x17224356
[6.435663] brcmfmac: brcmf_fw_alloc_request: using brcm/brcmfmac4356-sdio for chip BCM4356/2
[6.551259] brcmfmac: brcmf_sdiod_ramrw: membytes transfer failed
[6.551275] brcmfmac: brcmf_sdio_verifymemory: error -84 on reading 2048 membytes at 0x00184000
[6.551352] brcmfmac: brcmf_sdio_download_firmware: dongle image file download failed

after

[6.657165] brcmfmac: F1 signature read @0x18000000=0x17224356
[6.660807] brcmfmac: brcmf_fw_alloc_request: using brcm/brcmfmac4356-sdio for chip BCM4356/2
[6.918643] brcmfmac: brcmf_fw_alloc_request: using brcm/brcmfmac4356-sdio for chip BCM4356/2
[6.918734] brcmfmac: brcmf_c_process_clm_blob: no clm_blob available (err=-2), device may have limited channels available
[6.922724] brcmfmac: brcmf_c_preinit_dcmds: Firmware: BCM4356/2 wl0: Jun 16 2015 14:25:06 version 7.35.184.r1 (TOB) (r559293) FWID 01-b22ae69c
Fixes adc52bf7ef16 ("arm64: dts: meson: fix mmc v2 chips max frequencies")
---
 arch/arm64/boot/dts/amlogic/meson-gxm-khadas-vim2.dts |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm64/boot/dts/amlogic/meson-gxm-khadas-vim2.dts +1 -1
@@ -327,7 +327,7 @@
 	#size-cells = <0>;
 
 	bus-width = <4>;
-	max-frequency = <50000000>;
+	max-frequency = <60000000>;
 
 	non-removable;
 	disable-wp;
author maintainer
#187 lore raw

PATCH 182/189 bus: ti-sysc: Fix 1-wire reset quirk

In reply to Greg Kroah-Hartman

From: Tony Lindgren <[email protected]>

commit aec551c7a00fb7eae049c0c4cc3208ca53e26355 upstream.

Because of the i2c quirk we have the reset quirks named in a confusing way. Let's fix the 1-wire quirk accordinlyg. Then let's switch to using better naming later on.

Fixes 4e23be473e30 ("bus: ti-sysc: Add support for module specific reset quirks")
---
 drivers/bus/ti-sysc.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/bus/ti-sysc.c +2 -2
@@ -1406,7 +1406,7 @@ static void sysc_init_revision_quirks(st
 }
 
 /* 1-wire needs module's internal clocks enabled for reset */
-static void sysc_clk_enable_quirk_hdq1w(struct sysc *ddata)
+static void sysc_pre_reset_quirk_hdq1w(struct sysc *ddata)
 {
 	int offset = 0x0c;	/* HDQ_CTRL_STATUS */
 	u16 val;
@@ -1494,7 +1494,7 @@ static void sysc_init_module_quirks(stru
 		return;
 
 	if (ddata->cfg.quirks & SYSC_MODULE_QUIRK_HDQ1W) {
-		ddata->clk_enable_quirk = sysc_clk_enable_quirk_hdq1w;
+		ddata->clk_disable_quirk = sysc_pre_reset_quirk_hdq1w;
 
 		return;
 	}
author maintainer
#188 lore raw

PATCH 183/189 dt-bindings: arm: fsl: fix APF6Dev compatible

In reply to Greg Kroah-Hartman

From: Sébastien Szymanski <[email protected]>

commit ab4562f4dd92c455f6b313717af5e7d72a55d7b4 upstream.

APF6 Dev compatible is armadeus,imx6dl-apf6dev and not armadeus,imx6dl-apf6dldev.

Fixes 3d735471d066 ("dt-bindings: arm: Document Armadeus SoM and Dev boards devicetree binding")
---
 Documentation/devicetree/bindings/arm/fsl.yaml |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
Documentation/devicetree/bindings/arm/fsl.yaml +1 -1
@@ -139,7 +139,7 @@ properties:
         items:
           - enum:
               - armadeus,imx6dl-apf6      # APF6 (Solo) SoM
-              - armadeus,imx6dl-apf6dldev # APF6 (Solo) SoM on APF6Dev board
+              - armadeus,imx6dl-apf6dev   # APF6 (Solo) SoM on APF6Dev board
               - eckelmann,imx6dl-ci4x10
               - emtrion,emcon-mx6         # emCON-MX6S or emCON-MX6DL SoM
               - emtrion,emcon-mx6-avari   # emCON-MX6S or emCON-MX6DL SoM on Avari Base
author maintainer
#189 lore raw

PATCH 184/189 EDAC/synopsys: Do not print an error with back-to-back snprintf() calls

In reply to Greg Kroah-Hartman

From: Sherry Sun <[email protected]>

commit dfc6014e3b60713f375d0601d7549eed224c4615 upstream.

handle_error() currently calls snprintf() a couple of times in succession to output the message for a CE/UE, therefore overwriting each part of the message which was formatted with the previous snprintf() call. As a result, only the part of the message from the last snprintf() call will be printed.

The simplest and most effective way to fix this problem is to combine the whole string into one which to supply to a single snprintf() call.

[ bp: Massage. ]

Fixes b500b4a029d57 ("EDAC, synopsys: Add ECC support for ZynqMP DDR controller")
Cc Manish Narani <[email protected]>
---
 drivers/edac/synopsys_edac.c |   22 +++++++---------------
 1 file changed, 7 insertions(+), 15 deletions(-)
drivers/edac/synopsys_edac.c +7 -15
@@ -479,20 +479,14 @@ static void handle_error(struct mem_ctl_
 		pinf = &p->ceinfo;
 		if (!priv->p_data->quirks) {
 			snprintf(priv->message, SYNPS_EDAC_MSG_SIZE,
-				 "DDR ECC error type:%s Row %d Bank %d Col %d ",
-				  "CE", pinf->row, pinf->bank, pinf->col);
-			snprintf(priv->message, SYNPS_EDAC_MSG_SIZE,
-				 "Bit Position: %d Data: 0x%08x\n",
+				 "DDR ECC error type:%s Row %d Bank %d Col %d Bit Position: %d Data: 0x%08x",
+				 "CE", pinf->row, pinf->bank, pinf->col,
 				 pinf->bitpos, pinf->data);
 		} else {
 			snprintf(priv->message, SYNPS_EDAC_MSG_SIZE,
-				 "DDR ECC error type:%s Row %d Bank %d Col %d ",
-				  "CE", pinf->row, pinf->bank, pinf->col);
-			snprintf(priv->message, SYNPS_EDAC_MSG_SIZE,
-				 "BankGroup Number %d Block Number %d ",
-				 pinf->bankgrpnr, pinf->blknr);
-			snprintf(priv->message, SYNPS_EDAC_MSG_SIZE,
-				 "Bit Position: %d Data: 0x%08x\n",
+				 "DDR ECC error type:%s Row %d Bank %d Col %d BankGroup Number %d Block Number %d Bit Position: %d Data: 0x%08x",
+				 "CE", pinf->row, pinf->bank, pinf->col,
+				 pinf->bankgrpnr, pinf->blknr,
 				 pinf->bitpos, pinf->data);
 		}
 
@@ -509,10 +503,8 @@ static void handle_error(struct mem_ctl_
 				"UE", pinf->row, pinf->bank, pinf->col);
 		} else {
 			snprintf(priv->message, SYNPS_EDAC_MSG_SIZE,
-				 "DDR ECC error type :%s Row %d Bank %d Col %d ",
-				 "UE", pinf->row, pinf->bank, pinf->col);
-			snprintf(priv->message, SYNPS_EDAC_MSG_SIZE,
-				 "BankGroup Number %d Block Number %d",
+				 "DDR ECC error type :%s Row %d Bank %d Col %d BankGroup Number %d Block Number %d",
+				 "UE", pinf->row, pinf->bank, pinf->col,
 				 pinf->bankgrpnr, pinf->blknr);
 		}
 
author maintainer
#190 lore raw

PATCH 185/189 powerpc: fix hardware PMU exception bug on PowerVM compatibility mode systems

In reply to Greg Kroah-Hartman

From: Desnes A. Nunes do Rosario <[email protected]>

commit fc37a1632d40c80c067eb1bc235139f5867a2667 upstream.

PowerVM systems running compatibility mode on a few Power8 revisions are still vulnerable to the hardware defect that loses PMU exceptions arriving prior to a context switch.

The software fix for this issue is enabled through the CPU_FTR_PMAO_BUG cpu_feature bit, nevertheless this bit also needs to be set for PowerVM compatibility mode systems.

---
 arch/powerpc/kernel/cputable.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
arch/powerpc/kernel/cputable.c +3 -1
@@ -2193,11 +2193,13 @@ static struct cpu_spec * __init setup_cp
 		 * oprofile_cpu_type already has a value, then we are
 		 * possibly overriding a real PVR with a logical one,
 		 * and, in that case, keep the current value for
-		 * oprofile_cpu_type.
+		 * oprofile_cpu_type. Futhermore, let's ensure that the
+		 * fix for the PMAO bug is enabled on compatibility mode.
 		 */
 		if (old.oprofile_cpu_type != NULL) {
 			t->oprofile_cpu_type = old.oprofile_cpu_type;
 			t->oprofile_type = old.oprofile_type;
+			t->cpu_features |= old.cpu_features & CPU_FTR_PMAO_BUG;
 		}
 	}
 
author maintainer
#191 lore raw

PATCH 186/189 efi/x86: Align GUIDs to their size in the mixed mode runtime wrapper

In reply to Greg Kroah-Hartman

From: Ard Biesheuvel <[email protected]>

commit 63056e8b5ebf41d52170e9f5ba1fc83d1855278c upstream.

Hans reports that his mixed mode systems running v5.6-rc1 kernels hit the WARN_ON() in virt_to_phys_or_null_size(), caused by the fact that efi_guid_t objects on the vmap'ed stack happen to be misaligned with respect to their sizes. As a quick (i.e., backportable) fix, copy GUID pointer arguments to the local stack into a buffer that is naturally aligned to its size, so that it is guaranteed to cover only one physical page.

Note that on x86, we cannot rely on the stack pointer being aligned the way the compiler expects, so we need to allocate an 8-byte aligned buffer of sufficient size, and copy the GUID into that buffer at an offset that is aligned to 16 bytes.

Fixes f6697df36bdf0bf7 ("x86/efi: Prevent mixed mode boot corruption with CONFIG_VMAP_STACK=y")
Cc Ingo Molnar <[email protected]>
Cc Thomas Gleixner <[email protected]>
---
 arch/x86/platform/efi/efi_64.c |   25 +++++++++++++++++++++----
 1 file changed, 21 insertions(+), 4 deletions(-)
arch/x86/platform/efi/efi_64.c +21 -4
@@ -791,6 +791,8 @@ static efi_status_t
 efi_thunk_get_variable(efi_char16_t *name, efi_guid_t *vendor,
 		       u32 *attr, unsigned long *data_size, void *data)
 {
+	u8 buf[24] __aligned(8);
+	efi_guid_t *vnd = PTR_ALIGN((efi_guid_t *)buf, sizeof(*vnd));
 	efi_status_t status;
 	u32 phys_name, phys_vendor, phys_attr;
 	u32 phys_data_size, phys_data;
@@ -798,8 +800,10 @@ efi_thunk_get_variable(efi_char16_t *nam
 
 	spin_lock_irqsave(&efi_runtime_lock, flags);
 
+	*vnd = *vendor;
+
 	phys_data_size = virt_to_phys_or_null(data_size);
-	phys_vendor = virt_to_phys_or_null(vendor);
+	phys_vendor = virt_to_phys_or_null(vnd);
 	phys_name = virt_to_phys_or_null_size(name, efi_name_size(name));
 	phys_attr = virt_to_phys_or_null(attr);
 	phys_data = virt_to_phys_or_null_size(data, *data_size);
@@ -816,14 +820,18 @@ static efi_status_t
 efi_thunk_set_variable(efi_char16_t *name, efi_guid_t *vendor,
 		       u32 attr, unsigned long data_size, void *data)
 {
+	u8 buf[24] __aligned(8);
+	efi_guid_t *vnd = PTR_ALIGN((efi_guid_t *)buf, sizeof(*vnd));
 	u32 phys_name, phys_vendor, phys_data;
 	efi_status_t status;
 	unsigned long flags;
 
 	spin_lock_irqsave(&efi_runtime_lock, flags);
 
+	*vnd = *vendor;
+
 	phys_name = virt_to_phys_or_null_size(name, efi_name_size(name));
-	phys_vendor = virt_to_phys_or_null(vendor);
+	phys_vendor = virt_to_phys_or_null(vnd);
 	phys_data = virt_to_phys_or_null_size(data, data_size);
 
 	/* If data_size is > sizeof(u32) we've got problems */
@@ -840,6 +848,8 @@ efi_thunk_set_variable_nonblocking(efi_c
 				   u32 attr, unsigned long data_size,
 				   void *data)
 {
+	u8 buf[24] __aligned(8);
+	efi_guid_t *vnd = PTR_ALIGN((efi_guid_t *)buf, sizeof(*vnd));
 	u32 phys_name, phys_vendor, phys_data;
 	efi_status_t status;
 	unsigned long flags;
@@ -847,8 +857,10 @@ efi_thunk_set_variable_nonblocking(efi_c
 	if (!spin_trylock_irqsave(&efi_runtime_lock, flags))
 		return EFI_NOT_READY;
 
+	*vnd = *vendor;
+
 	phys_name = virt_to_phys_or_null_size(name, efi_name_size(name));
-	phys_vendor = virt_to_phys_or_null(vendor);
+	phys_vendor = virt_to_phys_or_null(vnd);
 	phys_data = virt_to_phys_or_null_size(data, data_size);
 
 	/* If data_size is > sizeof(u32) we've got problems */
@@ -865,14 +877,18 @@ efi_thunk_get_next_variable(unsigned lon
 			    efi_char16_t *name,
 			    efi_guid_t *vendor)
 {
+	u8 buf[24] __aligned(8);
+	efi_guid_t *vnd = PTR_ALIGN((efi_guid_t *)buf, sizeof(*vnd));
 	efi_status_t status;
 	u32 phys_name_size, phys_name, phys_vendor;
 	unsigned long flags;
 
 	spin_lock_irqsave(&efi_runtime_lock, flags);
 
+	*vnd = *vendor;
+
 	phys_name_size = virt_to_phys_or_null(name_size);
-	phys_vendor = virt_to_phys_or_null(vendor);
+	phys_vendor = virt_to_phys_or_null(vnd);
 	phys_name = virt_to_phys_or_null_size(name, *name_size);
 
 	status = efi_thunk(get_next_variable, phys_name_size,
@@ -880,6 +896,7 @@ efi_thunk_get_next_variable(unsigned lon
 
 	spin_unlock_irqrestore(&efi_runtime_lock, flags);
 
+	*vendor = *vnd;
 	return status;
 }
 
author maintainer
#192 lore raw

PATCH 187/189 efi/x86: Handle by-ref arguments covering multiple pages in mixed mode

In reply to Greg Kroah-Hartman

From: Ard Biesheuvel <[email protected]>

commit 8319e9d5ad98ffccd19f35664382c73cea216193 upstream.

The mixed mode runtime wrappers are fragile when it comes to how the memory referred to by its pointer arguments are laid out in memory, due to the fact that it translates these addresses to physical addresses that the runtime services can dereference when running in 1:1 mode. Since vmalloc'ed pages (including the vmap'ed stack) are not contiguous in the physical address space, this scheme only works if the referenced memory objects do not cross page boundaries.

Currently, the mixed mode runtime service wrappers require that all by-ref arguments that live in the vmalloc space have a size that is a power of 2, and are aligned to that same value. While this is a sensible way to construct an object that is guaranteed not to cross a page boundary, it is overly strict when it comes to checking whether a given object violates this requirement, as we can simply take the physical address of the first and the last byte, and verify that they point into the same physical page.

When this check fails, we emit a WARN(), but then simply proceed with the call, which could cause data corruption if the next physical page belongs to a mapping that is entirely unrelated.

Given that with vmap'ed stacks, this condition is much more likely to trigger, let's relax the condition a bit, but fail the runtime service call if it does trigger.

Fixes f6697df36bdf0bf7 ("x86/efi: Prevent mixed mode boot corruption with CONFIG_VMAP_STACK=y")
Cc Ingo Molnar <[email protected]>
Cc Thomas Gleixner <[email protected]>
---
 arch/x86/platform/efi/efi_64.c |   45 +++++++++++++++++++++++------------------
 1 file changed, 26 insertions(+), 19 deletions(-)
arch/x86/platform/efi/efi_64.c +26 -19
@@ -316,7 +316,7 @@ void efi_sync_low_kernel_mappings(void)
 static inline phys_addr_t
 virt_to_phys_or_null_size(void *va, unsigned long size)
 {
-	bool bad_size;
+	phys_addr_t pa;
 
 	if (!va)
 		return 0;
@@ -324,16 +324,13 @@ virt_to_phys_or_null_size(void *va, unsi
 	if (virt_addr_valid(va))
 		return virt_to_phys(va);
 
-	/*
-	 * A fully aligned variable on the stack is guaranteed not to
-	 * cross a page bounary. Try to catch strings on the stack by
-	 * checking that 'size' is a power of two.
-	 */
-	bad_size = size > PAGE_SIZE || !is_power_of_2(size);
+	pa = slow_virt_to_phys(va);
 
-	WARN_ON(!IS_ALIGNED((unsigned long)va, size) || bad_size);
+	/* check if the object crosses a page boundary */
+	if (WARN_ON((pa ^ (pa + size - 1)) & PAGE_MASK))
+		return 0;
 
-	return slow_virt_to_phys(va);
+	return pa;
 }
 
 #define virt_to_phys_or_null(addr)				\
@@ -808,8 +805,11 @@ efi_thunk_get_variable(efi_char16_t *nam
 	phys_attr = virt_to_phys_or_null(attr);
 	phys_data = virt_to_phys_or_null_size(data, *data_size);
 
-	status = efi_thunk(get_variable, phys_name, phys_vendor,
-			   phys_attr, phys_data_size, phys_data);
+	if (!phys_name || (data && !phys_data))
+		status = EFI_INVALID_PARAMETER;
+	else
+		status = efi_thunk(get_variable, phys_name, phys_vendor,
+				   phys_attr, phys_data_size, phys_data);
 
 	spin_unlock_irqrestore(&efi_runtime_lock, flags);
 
@@ -834,9 +834,11 @@ efi_thunk_set_variable(efi_char16_t *nam
 	phys_vendor = virt_to_phys_or_null(vnd);
 	phys_data = virt_to_phys_or_null_size(data, data_size);
 
-	/* If data_size is > sizeof(u32) we've got problems */
-	status = efi_thunk(set_variable, phys_name, phys_vendor,
-			   attr, data_size, phys_data);
+	if (!phys_name || !phys_data)
+		status = EFI_INVALID_PARAMETER;
+	else
+		status = efi_thunk(set_variable, phys_name, phys_vendor,
+				   attr, data_size, phys_data);
 
 	spin_unlock_irqrestore(&efi_runtime_lock, flags);
 
@@ -863,9 +865,11 @@ efi_thunk_set_variable_nonblocking(efi_c
 	phys_vendor = virt_to_phys_or_null(vnd);
 	phys_data = virt_to_phys_or_null_size(data, data_size);
 
-	/* If data_size is > sizeof(u32) we've got problems */
-	status = efi_thunk(set_variable, phys_name, phys_vendor,
-			   attr, data_size, phys_data);
+	if (!phys_name || !phys_data)
+		status = EFI_INVALID_PARAMETER;
+	else
+		status = efi_thunk(set_variable, phys_name, phys_vendor,
+				   attr, data_size, phys_data);
 
 	spin_unlock_irqrestore(&efi_runtime_lock, flags);
 
@@ -891,8 +895,11 @@ efi_thunk_get_next_variable(unsigned lon
 	phys_vendor = virt_to_phys_or_null(vnd);
 	phys_name = virt_to_phys_or_null_size(name, *name_size);
 
-	status = efi_thunk(get_next_variable, phys_name_size,
-			   phys_name, phys_vendor);
+	if (!phys_name)
+		status = EFI_INVALID_PARAMETER;
+	else
+		status = efi_thunk(get_next_variable, phys_name_size,
+				   phys_name, phys_vendor);
 
 	spin_unlock_irqrestore(&efi_runtime_lock, flags);
 
author maintainer
#193 lore raw

PATCH 188/189 efi: READ_ONCE rng seed size before munmap

In reply to Greg Kroah-Hartman

From: Jason A. Donenfeld <[email protected]>

commit be36f9e7517e17810ec369626a128d7948942259 upstream.

This function is consistent with using size instead of seed->size (except for one place that this patch fixes), but it reads seed->size without using READ_ONCE, which means the compiler might still do something unwanted. So, this commit simply adds the READ_ONCE wrapper.

---
 drivers/firmware/efi/efi.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/firmware/efi/efi.c +2 -2
@@ -552,7 +552,7 @@ int __init efi_config_parse_tables(void
 
 		seed = early_memremap(efi.rng_seed, sizeof(*seed));
 		if (seed != NULL) {
-			size = seed->size;
+			size = READ_ONCE(seed->size);
 			early_memunmap(seed, sizeof(*seed));
 		} else {
 			pr_err("Could not map UEFI random seed!\n");
@@ -562,7 +562,7 @@ int __init efi_config_parse_tables(void
 					      sizeof(*seed) + size);
 			if (seed != NULL) {
 				pr_notice("seeding entropy pool\n");
-				add_bootloader_randomness(seed->bits, seed->size);
+				add_bootloader_randomness(seed->bits, size);
 				early_memunmap(seed, sizeof(*seed) + size);
 			} else {
 				pr_err("Could not map UEFI random seed!\n");
author maintainer
#194 lore raw

PATCH 189/189 net: stmmac: fix notifier registration

In reply to Greg Kroah-Hartman

From: Aaro Koskinen <[email protected]>

commit 474a31e13a4e9749fb3ee55794d69d0f17ee0998 upstream.

We cannot register the same netdev notifier multiple times when probing stmmac devices. Register the notifier only once in module init, and also make debugfs creation/deletion safe against simultaneous notifier call.

Fixes 481a7d154cbb ("stmmac: debugfs entry name is not be changed when udev rename device name.")
---
 drivers/net/ethernet/stmicro/stmmac/stmmac_main.c |   13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)
drivers/net/ethernet/stmicro/stmmac/stmmac_main.c +8 -5
@@ -4289,6 +4289,8 @@ static void stmmac_init_fs(struct net_de
 {
 	struct stmmac_priv *priv = netdev_priv(dev);
 
+	rtnl_lock();
+
 	/* Create per netdev entries */
 	priv->dbgfs_dir = debugfs_create_dir(dev->name, stmmac_fs_dir);
 
@@ -4300,14 +4302,13 @@ static void stmmac_init_fs(struct net_de
 	debugfs_create_file("dma_cap", 0444, priv->dbgfs_dir, dev,
 			    &stmmac_dma_cap_fops);
 
-	register_netdevice_notifier(&stmmac_notifier);
+	rtnl_unlock();
 }
 
 static void stmmac_exit_fs(struct net_device *dev)
 {
 	struct stmmac_priv *priv = netdev_priv(dev);
 
-	unregister_netdevice_notifier(&stmmac_notifier);
 	debugfs_remove_recursive(priv->dbgfs_dir);
 }
 #endif /* CONFIG_DEBUG_FS */
@@ -4825,14 +4826,14 @@ int stmmac_dvr_remove(struct device *dev
 
 	netdev_info(priv->dev, "%s: removing driver", __func__);
 
-#ifdef CONFIG_DEBUG_FS
-	stmmac_exit_fs(ndev);
-#endif
 	stmmac_stop_all_dma(priv);
 
 	stmmac_mac_set(priv, priv->ioaddr, false);
 	netif_carrier_off(ndev);
 	unregister_netdev(ndev);
+#ifdef CONFIG_DEBUG_FS
+	stmmac_exit_fs(ndev);
+#endif
 	phylink_destroy(priv->phylink);
 	if (priv->plat->stmmac_rst)
 		reset_control_assert(priv->plat->stmmac_rst);
@@ -5052,6 +5053,7 @@ static int __init stmmac_init(void)
 	/* Create debugfs main directory if it doesn't exist yet */
 	if (!stmmac_fs_dir)
 		stmmac_fs_dir = debugfs_create_dir(STMMAC_RESOURCE_NAME, NULL);
+	register_netdevice_notifier(&stmmac_notifier);
 #endif
 
 	return 0;
@@ -5060,6 +5062,7 @@ static int __init stmmac_init(void)
 static void __exit stmmac_exit(void)
 {
 #ifdef CONFIG_DEBUG_FS
+	unregister_netdevice_notifier(&stmmac_notifier);
 	debugfs_remove_recursive(stmmac_fs_dir);
 #endif
 }
#195 lore raw
In reply to Greg Kroah-Hartman
Quoted 8 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.5.9 release. There are 189 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Thu, 12 Mar 2020 12:34:10 +0000. Anything received after that time might be too late.

block/bfq-wf2q.c: In function 'bfq_forget_entity':
./include/linux/kernel.h:987:51: error: 'struct bfq_group' has no member named 'entity'
987 | BUILD_BUG_ON_MSG(!__same_type(*(ptr), ((type *)0)->member) && \

Guess this is the same problem that is also seen with v5.4.25-rc1.

Guenter

author maintainer
#196 lore raw
Quoted 10 lines · Guenter Roeck
Quoted 8 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.5.9 release. There are 189 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Thu, 12 Mar 2020 12:34:10 +0000. Anything received after that time might be too late.

Quoted 5 lines

block/bfq-wf2q.c: In function 'bfq_forget_entity':
./include/linux/kernel.h:987:51: error: 'struct bfq_group' has no member named 'entity'
987 | BUILD_BUG_ON_MSG(!__same_type(*(ptr), ((type *)0)->member) && \

Guess this is the same problem that is also seen with v5.4.25-rc1.

Thanks for the report, I should have that now fixed up for 5.5.y and 5.4.y.

greg k-h

maintainer
#197 lore raw
In reply to Greg Kroah-Hartman
Quoted 9 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.5.9 release. There are 189 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Thu, 12 Mar 2020 12:34:10 +0000. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

Quoted 8 lines
	https://www.kernel.org/pub/linux/kernel/v5.x/stable-review/patch-5.5.9-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-5.5.y
and the diffstat can be found below.

thanks,

greg k-h

All tests are passing for Tegra ...

Test results for stable-v5.5:

    13 builds:	13 pass, 0 fail
    22 boots:	22 pass, 0 fail
    40 tests:	40 pass, 0 fail

Linux version: 5.5.9-rc1-g11e07aec0780
Boards tested: tegra124-jetson-tk1, tegra186-p2771-0000,

                tegra194-p2972-0000, tegra20-ventana,
                tegra210-p2371-2180, tegra210-p3450-0000,
                tegra30-cardhu-a04

Cheers
Jon

Signature · 1 line
nvpublic
author maintainer
#198 lore raw
Quoted 27 lines · Jon Hunter
Quoted 9 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.5.9 release. There are 189 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Thu, 12 Mar 2020 12:34:10 +0000. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

Quoted 8 lines
	https://www.kernel.org/pub/linux/kernel/v5.x/stable-review/patch-5.5.9-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-5.5.y
and the diffstat can be found below.

thanks,

greg k-h

All tests are passing for Tegra ...

Test results for stable-v5.5:

    13 builds:	13 pass, 0 fail
    22 boots:	22 pass, 0 fail
    40 tests:	40 pass, 0 fail
Quoted 5 lines

Linux version: 5.5.9-rc1-g11e07aec0780
Boards tested: tegra124-jetson-tk1, tegra186-p2771-0000,

                tegra194-p2972-0000, tegra20-ventana,
                tegra210-p2371-2180, tegra210-p3450-0000,
                tegra30-cardhu-a04

Thanks for testing all of these and letting me know.

greg k-h

#199 lore raw
In reply to Greg Kroah-Hartman
Quoted 9 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.5.9 release. There are 189 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Thu, 12 Mar 2020 12:34:10 +0000. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

Quoted 9 lines
	https://www.kernel.org/pub/linux/kernel/v5.x/stable-review/patch-5.5.9-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-5.5.y
and the diffstat can be found below.

thanks,

greg k-h

Compiled and booted on my test system. No dmesg regressions.

thanks,
-- Shuah

author maintainer
#200 lore raw
Quoted 15 lines · shuah
Quoted 14 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.5.9 release. There are 189 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Thu, 12 Mar 2020 12:34:10 +0000. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

	https://www.kernel.org/pub/linux/kernel/v5.x/stable-review/patch-5.5.9-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-5.5.y
and the diffstat can be found below.

Quoted 6 lines
Quoted 4 lines

thanks,

greg k-h

Compiled and booted on my test system. No dmesg regressions.

Thanks for testing all of these and letting me know.

greg k-h

#201 lore raw
In reply to Greg Kroah-Hartman
Quoted 8 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.5.9 release. There are 189 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Thu, 12 Mar 2020 12:34:10 +0000. Anything received after that time might be too late.

For v5.5.8-190-g11e07aec0780:

Build results:

	total: 157 pass: 152 fail: 5
Failed builds:
	csky:defconfig
	csky:allnoconfig
	csky:tinyconfig
	m68k:defconfig
	m68k:sun3_defconfig
Qemu test results:
	total: 423 pass: 418 fail: 5
Failed tests:
	arm:sx1:sx1_defconfig:initrd
	arm:sx1:sx1_defconfig:sd:rootfs
	arm:sx1:sx1_defconfig:flash32,26,3:rootfs
	q800:m68040:mac_defconfig:initrd
	q800:m68040:mac_defconfig:rootfs

csky:

kernel/fork.c:2588:2: error: #error clone3 requires copy_thread_tls support in arch
2588 | #error clone3 requires copy_thread_tls support in arch

m68k, arm:

block/bfq-wf2q.c: In function 'bfq_get_entity':
include/linux/kernel.h:987:51: error: 'struct bfq_group' has no member named 'entity'

Guenter

author maintainer
#202 lore raw
Quoted 30 lines · Guenter Roeck
Quoted 8 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.5.9 release. There are 189 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Thu, 12 Mar 2020 12:34:10 +0000. Anything received after that time might be too late.

For v5.5.8-190-g11e07aec0780:

Build results:

	total: 157 pass: 152 fail: 5
Failed builds:
	csky:defconfig
	csky:allnoconfig
	csky:tinyconfig
	m68k:defconfig
	m68k:sun3_defconfig
Qemu test results:
	total: 423 pass: 418 fail: 5
Failed tests:
	arm:sx1:sx1_defconfig:initrd
	arm:sx1:sx1_defconfig:sd:rootfs
	arm:sx1:sx1_defconfig:flash32,26,3:rootfs
	q800:m68040:mac_defconfig:initrd
	q800:m68040:mac_defconfig:rootfs
Quoted 4 lines

csky:

kernel/fork.c:2588:2: error: #error clone3 requires copy_thread_tls support in arch
2588 | #error clone3 requires copy_thread_tls support in arch

That's odd, let me dig...

author maintainer
#203 lore raw
In reply to Guenter Roeck
Quoted 30 lines · Guenter Roeck
Quoted 8 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.5.9 release. There are 189 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Thu, 12 Mar 2020 12:34:10 +0000. Anything received after that time might be too late.

For v5.5.8-190-g11e07aec0780:

Build results:

	total: 157 pass: 152 fail: 5
Failed builds:
	csky:defconfig
	csky:allnoconfig
	csky:tinyconfig
	m68k:defconfig
	m68k:sun3_defconfig
Qemu test results:
	total: 423 pass: 418 fail: 5
Failed tests:
	arm:sx1:sx1_defconfig:initrd
	arm:sx1:sx1_defconfig:sd:rootfs
	arm:sx1:sx1_defconfig:flash32,26,3:rootfs
	q800:m68040:mac_defconfig:initrd
	q800:m68040:mac_defconfig:rootfs
Quoted 4 lines

csky:

kernel/fork.c:2588:2: error: #error clone3 requires copy_thread_tls support in arch
2588 | #error clone3 requires copy_thread_tls support in arch

Ok, found this one. Sasha was wiring up csky syscalls and missed this one...

thanks,

greg k-h

#204 lore raw
In reply to Greg Kroah-Hartman
Quoted 10 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.5.9 release. There are 189 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Thu, 12 Mar 2020 12:34:10 +0000. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

Quoted 8 lines
        https://www.kernel.org/pub/linux/kernel/v5.x/stable-review/patch-5.5.9-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-5.5.y
and the diffstat can be found below.

thanks,

greg k-h

Results from Linaro’s test farm.
No regressions on arm64, arm, x86_64, and i386.

Summary
------------------------------------------------------------------------

kernel: 5.5.9-rc1
git repo: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git
git branch: linux-5.5.y
git commit: 11e07aec07807683209513e2ad4a41bd7ee8a250
git describe: v5.5.8-190-g11e07aec0780
Test details: https://qa-reports.linaro.org/lkft/linux-stable-rc-5.5-oe/build/v5.5.8-190-g11e07aec0780

No regressions (compared to build v5.5.8)

No fixes (compared to build v5.5.8)

Ran 27520 total tests in the following environments and test suites.

Environments
--------------
- dragonboard-410c
- hi6220-hikey
- i386
- juno-r2
- nxp-ls2088
- qemu_arm
- qemu_arm64
- qemu_i386
- qemu_x86_64
- x15
- x86

Test Suites
-----------
* build
* install-android-platform-tools-r2600
* install-android-platform-tools-r2800
* kselftest
* libgpiod
* libhugetlbfs
* linux-log-parser
* ltp-cap_bounds-tests
* ltp-commands-tests
* ltp-containers-tests
* ltp-cpuhotplug-tests
* ltp-crypto-tests
* ltp-fcntl-locktests-tests
* ltp-filecaps-tests
* ltp-fs-tests
* ltp-fs_bind-tests
* ltp-fs_perms_simple-tests
* ltp-fsx-tests
* ltp-hugetlb-tests
* ltp-math-tests
* ltp-mm-tests
* ltp-nptl-tests
* ltp-pty-tests
* ltp-securebits-tests
* perf
* ltp-dio-tests
* ltp-io-tests
* ltp-sched-tests
* ltp-syscalls-tests
* v4l2-compliance
* kvm-unit-tests
* ltp-cve-tests
* ltp-ipc-tests
* network-basic-tests
* spectre-meltdown-checker-test
* ltp-cap_bounds-64k-page_size-tests
* ltp-cap_bounds-kasan-tests
* ltp-commands-64k-page_size-tests
* ltp-commands-kasan-tests
* ltp-containers-64k-page_size-tests
* ltp-containers-kasan-tests
* ltp-cpuhotplug-64k-page_size-tests
* ltp-cpuhotplug-kasan-tests
* ltp-crypto-64k-page_size-tests
* ltp-crypto-kasan-tests
* ltp-cve-kasan-tests
* ltp-dio-64k-page_size-tests
* ltp-dio-kasan-tests
* ltp-fcntl-locktests-64k-page_size-tests
* ltp-fcntl-locktests-kasan-tests
* ltp-filecaps-64k-page_size-tests
* ltp-filecaps-kasan-tests
* ltp-fs-64k-page_size-tests
* ltp-fs-kasan-tests
* ltp-fs_bind-64k-page_size-tests
* ltp-fs_bind-kasan-tests
* ltp-fs_perms_simple-64k-page_size-tests
* ltp-fs_perms_simple-kasan-tests
* ltp-fsx-64k-page_size-tests
* ltp-fsx-kasan-tests
* ltp-hugetlb-64k-page_size-tests
* ltp-hugetlb-kasan-tests
* ltp-io-64k-page_size-tests
* ltp-io-kasan-tests
* ltp-ipc-64k-page_size-tests
* ltp-ipc-kasan-tests
* ltp-math-64k-page_size-tests
* ltp-math-kasan-tests
* ltp-mm-64k-page_size-tests
* ltp-mm-kasan-tests
* ltp-nptl-64k-page_size-tests
* ltp-nptl-kasan-tests
* ltp-pty-64k-page_size-tests
* ltp-pty-kasan-tests
* ltp-sched-64k-page_size-tests
* ltp-sched-kasan-tests
* ltp-securebits-64k-page_size-tests
* ltp-securebits-kasan-tests
* ltp-syscalls-64k-page_size-tests
* ltp-syscalls-compat-tests
* ltp-syscalls-kasan-tests
* ltp-open-posix-tests
* kselftest-vsyscall-mode-native
* kselftest-vsyscall-mode-none

Signature · 2 lines
Linaro LKFT
https://lkft.linaro.org
author maintainer
#205 lore raw
Quoted 17 lines · Naresh Kamboju
Quoted 15 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 5.5.9 release. There are 189 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Thu, 12 Mar 2020 12:34:10 +0000. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

        https://www.kernel.org/pub/linux/kernel/v5.x/stable-review/patch-5.5.9-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-5.5.y
and the diffstat can be found below.

Quoted 6 lines
Quoted 3 lines

thanks,

greg k-h

Results from Linaro’s test farm.
No regressions on arm64, arm, x86_64, and i386.

Thanks for testing all of these and letting me know.

greg k-h