Skip to contentSkip to conversation

5.17.2-rc1 review

PATCH In review Greg Kroah-Hartman 16 participants 1,147 messages Apr 5, 2022 – Apr 6, 2022 Page 8 of 9

Subsystems (11)
Stable branch VFS MIPS Crypto API +7 more subsystems
lore
Share & tools
lore
author maintainer
#953 lore raw

PATCH 949/1126 ALSA: hda/realtek: Add alc256-samsung-headphone fixup

In reply to Greg Kroah-Hartman

From: Matt Kramer <[email protected]>

[ Upstream commit ef248d9bd616b04df8be25539a4dc5db4b6c56f4 ]

This fixes the near-silence of the headphone jack on the ALC256-based Samsung Galaxy Book Flex Alpha (NP730QCJ). The magic verbs were found through trial and error, using known ALC298 hacks as inspiration. The fixup is auto-enabled only when the NP730QCJ is detected. It can be manually enabled using model=alc256-samsung-headphone.

---
 Documentation/sound/hd-audio/models.rst |  4 ++++
 sound/pci/hda/patch_realtek.c           | 11 +++++++++++
 2 files changed, 15 insertions(+)
Documentation/sound/hd-audio/models.rst +4 -0
index d25335993e55..9b52f50a6854 100644
@@ -261,6 +261,10 @@ alc-sense-combo
 huawei-mbx-stereo
     Enable initialization verbs for Huawei MBX stereo speakers;
     might be risky, try this at your own risk
+alc298-samsung-headphone
+    Samsung laptops with ALC298
+alc256-samsung-headphone
+    Samsung laptops with ALC256
 
 ALC66x/67x/892
 ==============
sound/pci/hda/patch_realtek.c +11 -0
index 6ff41d9b5bc9..16e90524a497 100644
@@ -6948,6 +6948,7 @@ enum {
 	ALC236_FIXUP_HP_MUTE_LED,
 	ALC236_FIXUP_HP_MUTE_LED_MICMUTE_VREF,
 	ALC298_FIXUP_SAMSUNG_HEADPHONE_VERY_QUIET,
+	ALC256_FIXUP_SAMSUNG_HEADPHONE_VERY_QUIET,
 	ALC295_FIXUP_ASUS_MIC_NO_PRESENCE,
 	ALC269VC_FIXUP_ACER_VCOPPERBOX_PINS,
 	ALC269VC_FIXUP_ACER_HEADSET_MIC,
@@ -8273,6 +8274,14 @@ static const struct hda_fixup alc269_fixups[] = {
 			{ }
 		},
 	},
+	[ALC256_FIXUP_SAMSUNG_HEADPHONE_VERY_QUIET] = {
+		.type = HDA_FIXUP_VERBS,
+		.v.verbs = (const struct hda_verb[]) {
+			{ 0x20, AC_VERB_SET_COEF_INDEX, 0x08},
+			{ 0x20, AC_VERB_SET_PROC_COEF, 0x2fcf},
+			{ }
+		},
+	},
 	[ALC295_FIXUP_ASUS_MIC_NO_PRESENCE] = {
 		.type = HDA_FIXUP_PINS,
 		.v.pins = (const struct hda_pintbl[]) {
@@ -9054,6 +9063,7 @@ static const struct snd_pci_quirk alc269_fixup_tbl[] = {
 	SND_PCI_QUIRK(0x144d, 0xc740, "Samsung Ativ book 8 (NP870Z5G)", ALC269_FIXUP_ATIV_BOOK_8),
 	SND_PCI_QUIRK(0x144d, 0xc812, "Samsung Notebook Pen S (NT950SBE-X58)", ALC298_FIXUP_SAMSUNG_HEADPHONE_VERY_QUIET),
 	SND_PCI_QUIRK(0x144d, 0xc830, "Samsung Galaxy Book Ion (NT950XCJ-X716A)", ALC298_FIXUP_SAMSUNG_HEADPHONE_VERY_QUIET),
+	SND_PCI_QUIRK(0x144d, 0xc832, "Samsung Galaxy Book Flex Alpha (NP730QCJ)", ALC256_FIXUP_SAMSUNG_HEADPHONE_VERY_QUIET),
 	SND_PCI_QUIRK(0x1458, 0xfa53, "Gigabyte BXBT-2807", ALC283_FIXUP_HEADSET_MIC),
 	SND_PCI_QUIRK(0x1462, 0xb120, "MSI Cubi MS-B120", ALC283_FIXUP_HEADSET_MIC),
 	SND_PCI_QUIRK(0x1462, 0xb171, "Cubi N 8GL (MS-B171)", ALC283_FIXUP_HEADSET_MIC),
@@ -9400,6 +9410,7 @@ static const struct hda_model_fixup alc269_fixup_models[] = {
 	{.id = ALC298_FIXUP_HUAWEI_MBX_STEREO, .name = "huawei-mbx-stereo"},
 	{.id = ALC256_FIXUP_MEDION_HEADSET_NO_PRESENCE, .name = "alc256-medion-headset"},
 	{.id = ALC298_FIXUP_SAMSUNG_HEADPHONE_VERY_QUIET, .name = "alc298-samsung-headphone"},
+	{.id = ALC256_FIXUP_SAMSUNG_HEADPHONE_VERY_QUIET, .name = "alc256-samsung-headphone"},
 	{.id = ALC255_FIXUP_XIAOMI_HEADSET_MIC, .name = "alc255-xiaomi-headset"},
 	{.id = ALC274_FIXUP_HP_MIC, .name = "alc274-hp-mic-detect"},
 	{.id = ALC245_FIXUP_HP_X360_AMP, .name = "alc245-hp-x360-amp"},
Signature · 1 line
2.34.1
author maintainer
#954 lore raw

PATCH 950/1126 KVM: SVM: Allow AVIC support on system w/ physical APIC ID > 255

In reply to Greg Kroah-Hartman

From: Suravee Suthikulpanit <[email protected]>

commit 4a204f7895878363ca8211f50ec610408c8c70aa upstream.

Expand KVM's mask for the AVIC host physical ID to the full 12 bits defined by the architecture. The number of bits consumed by hardware is model specific, e.g. early CPUs ignored bits 11:8, but there is no way for KVM to enumerate the "true" size. So, KVM must allow using all bits, else it risks rejecting completely legal x2APIC IDs on newer CPUs.

This means KVM relies on hardware to not assign x2APIC IDs that exceed the "true" width of the field, but presumably hardware is smart enough to tie the width to the max x2APIC ID. KVM also relies on hardware to support at least 8 bits, as the legacy xAPIC ID is writable by software. But, those assumptions are unavoidable due to the lack of any way to enumerate the "true" width.

Cc Maxim Levitsky <[email protected]>
Fixes 44a95dae1d22 ("KVM: x86: Detect and Initialize AVIC support")

Message-Id: <[email protected]>

---
 arch/x86/include/asm/svm.h |    2 +-
 arch/x86/kvm/svm/avic.c    |    7 +------
 arch/x86/kvm/svm/svm.h     |   11 +++++++++++
 3 files changed, 13 insertions(+), 7 deletions(-)
arch/x86/include/asm/svm.h +1 -1
@@ -226,7 +226,7 @@ struct __attribute__ ((__packed__)) vmcb
 #define AVIC_LOGICAL_ID_ENTRY_VALID_BIT			31
 #define AVIC_LOGICAL_ID_ENTRY_VALID_MASK		(1 << 31)
 
-#define AVIC_PHYSICAL_ID_ENTRY_HOST_PHYSICAL_ID_MASK	(0xFFULL)
+#define AVIC_PHYSICAL_ID_ENTRY_HOST_PHYSICAL_ID_MASK	GENMASK_ULL(11, 0)
 #define AVIC_PHYSICAL_ID_ENTRY_BACKING_PAGE_MASK	(0xFFFFFFFFFFULL << 12)
 #define AVIC_PHYSICAL_ID_ENTRY_IS_RUNNING_MASK		(1ULL << 62)
 #define AVIC_PHYSICAL_ID_ENTRY_VALID_MASK		(1ULL << 63)
arch/x86/kvm/svm/avic.c +1 -6
@@ -927,17 +927,12 @@ out:
 void avic_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 {
 	u64 entry;
-	/* ID = 0xff (broadcast), ID > 0xff (reserved) */
 	int h_physical_id = kvm_cpu_get_apicid(cpu);
 	struct vcpu_svm *svm = to_svm(vcpu);
 
 	lockdep_assert_preemption_disabled();
 
-	/*
-	 * Since the host physical APIC id is 8 bits,
-	 * we can support host APIC ID upto 255.
-	 */
-	if (WARN_ON(h_physical_id > AVIC_PHYSICAL_ID_ENTRY_HOST_PHYSICAL_ID_MASK))
+	if (WARN_ON(h_physical_id & ~AVIC_PHYSICAL_ID_ENTRY_HOST_PHYSICAL_ID_MASK))
 		return;
 
 	/*
arch/x86/kvm/svm/svm.h +11 -0
@@ -558,6 +558,17 @@ extern struct kvm_x86_nested_ops svm_nes
 
 /* avic.c */
 
+#define AVIC_LOGICAL_ID_ENTRY_GUEST_PHYSICAL_ID_MASK	(0xFF)
+#define AVIC_LOGICAL_ID_ENTRY_VALID_BIT			31
+#define AVIC_LOGICAL_ID_ENTRY_VALID_MASK		(1 << 31)
+
+#define AVIC_PHYSICAL_ID_ENTRY_HOST_PHYSICAL_ID_MASK	GENMASK_ULL(11, 0)
+#define AVIC_PHYSICAL_ID_ENTRY_BACKING_PAGE_MASK	(0xFFFFFFFFFFULL << 12)
+#define AVIC_PHYSICAL_ID_ENTRY_IS_RUNNING_MASK		(1ULL << 62)
+#define AVIC_PHYSICAL_ID_ENTRY_VALID_MASK		(1ULL << 63)
+
+#define VMCB_AVIC_APIC_BAR_MASK		0xFFFFFFFFFF000ULL
+
 int avic_ga_log_notifier(u32 ga_tag);
 void avic_vm_destroy(struct kvm *kvm);
 int avic_vm_init(struct kvm *kvm);
author maintainer
#955 lore raw

PATCH 951/1126 KVM: x86: Reinitialize context if host userspace toggles EFER.LME

In reply to Greg Kroah-Hartman

From: Paolo Bonzini <[email protected]>

commit d6174299365ddbbf491620c0b8c5ca1a6ef2eea5 upstream.

While the guest runs, EFER.LME cannot change unless CR0.PG is clear, and therefore EFER.NX is the only bit that can affect the MMU role. However, set_efer accepts a host-initiated change to EFER.LME even with CR0.PG=1. In that case, the MMU has to be reset.

Fixes 11988499e62b ("KVM: x86: Skip EFER vs. guest CPUID checks for host-initiated writes")
---
 arch/x86/kvm/mmu.h |    1 +
 arch/x86/kvm/x86.c |    3 +--
 2 files changed, 2 insertions(+), 2 deletions(-)
arch/x86/kvm/mmu.h +1 -0
@@ -48,6 +48,7 @@
 			       X86_CR4_SMEP | X86_CR4_SMAP | X86_CR4_PKE)
 
 #define KVM_MMU_CR0_ROLE_BITS (X86_CR0_PG | X86_CR0_WP)
+#define KVM_MMU_EFER_ROLE_BITS (EFER_LME | EFER_NX)
 
 static __always_inline u64 rsvd_bits(int s, int e)
 {
arch/x86/kvm/x86.c +1 -2
@@ -1656,8 +1656,7 @@ static int set_efer(struct kvm_vcpu *vcp
 		return r;
 	}
 
-	/* Update reserved bits */
-	if ((efer ^ old_efer) & EFER_NX)
+	if ((efer ^ old_efer) & KVM_MMU_EFER_ROLE_BITS)
 		kvm_mmu_reset_context(vcpu);
 
 	return 0;
author maintainer
#956 lore raw

PATCH 952/1126 KVM: x86/mmu: Use common TDP MMU zap helper for MMU notifier unmap hook

In reply to Greg Kroah-Hartman

From: Sean Christopherson <[email protected]>

commit 83b83a02073ec8d18c77a9bbe0881d710f7a9d32 upstream.

Use the common TDP MMU zap helper when handling an MMU notifier unmap event, the two flows are semantically identical. Consolidate the code in preparation for a future bug fix, as both kvm_tdp_mmu_unmap_gfn_range() and __kvm_tdp_mmu_zap_gfn_range() are guilty of not zapping SPTEs in invalid roots.

No functional change intended.

Message-Id: <[email protected]>

---
 arch/x86/kvm/mmu/tdp_mmu.c |    9 ++-------
 1 file changed, 2 insertions(+), 7 deletions(-)
arch/x86/kvm/mmu/tdp_mmu.c +2 -7
@@ -1032,13 +1032,8 @@ int kvm_tdp_mmu_map(struct kvm_vcpu *vcp
 bool kvm_tdp_mmu_unmap_gfn_range(struct kvm *kvm, struct kvm_gfn_range *range,
 				 bool flush)
 {
-	struct kvm_mmu_page *root;
-
-	for_each_tdp_mmu_root_yield_safe(kvm, root, range->slot->as_id, false)
-		flush = zap_gfn_range(kvm, root, range->start, range->end,
-				      range->may_block, flush, false);
-
-	return flush;
+	return __kvm_tdp_mmu_zap_gfn_range(kvm, range->slot->as_id, range->start,
+					   range->end, range->may_block, flush);
 }
 
 typedef bool (*tdp_handler_t)(struct kvm *kvm, struct tdp_iter *iter,
author maintainer
#957 lore raw

PATCH 953/1126 KVM: x86/mmu: Move "invalid" check out of kvm_tdp_mmu_get_root()

In reply to Greg Kroah-Hartman

From: Sean Christopherson <[email protected]>

commit 04dc4e6ce274fa729feda32aa957b27388a3870c upstream.

Move the check for an invalid root out of kvm_tdp_mmu_get_root() and into the one place it actually matters, tdp_mmu_next_root(), as the other user already has an implicit validity check. A future bug fix will need to get references to invalid roots to honor mmu_notifier requests; there's no point in forcing what will be a common path to open code getting a reference to a root.

No functional change intended.

Message-Id: <[email protected]>

---
 arch/x86/kvm/mmu/tdp_mmu.c |   12 ++++++++++--
 arch/x86/kvm/mmu/tdp_mmu.h |    3 ---
 2 files changed, 10 insertions(+), 5 deletions(-)
arch/x86/kvm/mmu/tdp_mmu.c +10 -2
@@ -121,9 +121,14 @@ static struct kvm_mmu_page *tdp_mmu_next
 		next_root = list_first_or_null_rcu(&kvm->arch.tdp_mmu_roots,
 						   typeof(*next_root), link);
 
-	while (next_root && !kvm_tdp_mmu_get_root(kvm, next_root))
+	while (next_root) {
+		if (!next_root->role.invalid &&
+		    kvm_tdp_mmu_get_root(kvm, next_root))
+			break;
+
 		next_root = list_next_or_null_rcu(&kvm->arch.tdp_mmu_roots,
 				&next_root->link, typeof(*next_root), link);
+	}
 
 	rcu_read_unlock();
 
@@ -200,7 +205,10 @@ hpa_t kvm_tdp_mmu_get_vcpu_root_hpa(stru
 
 	role = page_role_for_level(vcpu, vcpu->arch.mmu->shadow_root_level);
 
-	/* Check for an existing root before allocating a new one. */
+	/*
+	 * Check for an existing root before allocating a new one.  Note, the
+	 * role check prevents consuming an invalid root.
+	 */
 	for_each_tdp_mmu_root(kvm, root, kvm_mmu_role_as_id(role)) {
 		if (root->role.word == role.word &&
 		    kvm_tdp_mmu_get_root(kvm, root))
arch/x86/kvm/mmu/tdp_mmu.h +0 -3
@@ -10,9 +10,6 @@ hpa_t kvm_tdp_mmu_get_vcpu_root_hpa(stru
 __must_check static inline bool kvm_tdp_mmu_get_root(struct kvm *kvm,
 						     struct kvm_mmu_page *root)
 {
-	if (root->role.invalid)
-		return false;
-
 	return refcount_inc_not_zero(&root->tdp_mmu_root_count);
 }
 
author maintainer
#958 lore raw

PATCH 954/1126 KVM: x86/mmu: Zap _all_ roots when unmapping gfn range in TDP MMU

In reply to Greg Kroah-Hartman

From: Sean Christopherson <[email protected]>

commit d62007edf01f5c11f75d0f4b1e538fc52a5b1982 upstream.

Zap both valid and invalid roots when zapping/unmapping a gfn range, as KVM must ensure it holds no references to the freed page after returning from the unmap operation. Most notably, the TDP MMU doesn't zap invalid roots in mmu_notifier callbacks. This leads to use-after-free and other issues if the mmu_notifier runs to completion while an invalid root zapper yields as KVM fails to honor the requirement that there must be _no_ references to the page after the mmu_notifier returns.

The bug is most easily reproduced by hacking KVM to cause a collision between set_nx_huge_pages() and kvm_mmu_notifier_release(), but the bug exists between kvm_mmu_notifier_invalidate_range_start() and memslot updates as well. Invalidating a root ensures pages aren't accessible by the guest, and KVM won't read or write page data itself, but KVM will trigger e.g. kvm_set_pfn_dirty() when zapping SPTEs, and thus completing a zap of an invalid root _after_ the mmu_notifier returns is fatal.

  WARNING: CPU: 24 PID: 1496 at arch/x86/kvm/../../../virt/kvm/kvm_main.c:173 [kvm]
  RIP: 0010:kvm_is_zone_device_pfn+0x96/0xa0 [kvm]
  Call Trace:
   <TASK>
   kvm_set_pfn_dirty+0xa8/0xe0 [kvm]
   __handle_changed_spte+0x2ab/0x5e0 [kvm]
   __handle_changed_spte+0x2ab/0x5e0 [kvm]
   __handle_changed_spte+0x2ab/0x5e0 [kvm]
   zap_gfn_range+0x1f3/0x310 [kvm]
   kvm_tdp_mmu_zap_invalidated_roots+0x50/0x90 [kvm]
   kvm_mmu_zap_all_fast+0x177/0x1a0 [kvm]
   set_nx_huge_pages+0xb4/0x190 [kvm]
   param_attr_store+0x70/0x100
   module_attr_store+0x19/0x30
   kernfs_fop_write_iter+0x119/0x1b0
   new_sync_write+0x11c/0x1b0
   vfs_write+0x1cc/0x270
   ksys_write+0x5f/0xe0
   do_syscall_64+0x38/0xc0
   entry_SYSCALL_64_after_hwframe+0x44/0xae
   </TASK>
Fixes b7cccd397f31 ("KVM: x86/mmu: Fast invalidation for TDP MMU")
Cc Ben Gardon <[email protected]>

Message-Id: <[email protected]>

---
 arch/x86/kvm/mmu/tdp_mmu.c |   39 ++++++++++++++++++++++++---------------
 1 file changed, 24 insertions(+), 15 deletions(-)
arch/x86/kvm/mmu/tdp_mmu.c +24 -15
@@ -99,15 +99,18 @@ void kvm_tdp_mmu_put_root(struct kvm *kv
 }
 
 /*
- * Finds the next valid root after root (or the first valid root if root
- * is NULL), takes a reference on it, and returns that next root. If root
- * is not NULL, this thread should have already taken a reference on it, and
- * that reference will be dropped. If no valid root is found, this
- * function will return NULL.
+ * Returns the next root after @prev_root (or the first root if @prev_root is
+ * NULL).  A reference to the returned root is acquired, and the reference to
+ * @prev_root is released (the caller obviously must hold a reference to
+ * @prev_root if it's non-NULL).
+ *
+ * If @only_valid is true, invalid roots are skipped.
+ *
+ * Returns NULL if the end of tdp_mmu_roots was reached.
  */
 static struct kvm_mmu_page *tdp_mmu_next_root(struct kvm *kvm,
 					      struct kvm_mmu_page *prev_root,
-					      bool shared)
+					      bool shared, bool only_valid)
 {
 	struct kvm_mmu_page *next_root;
 
@@ -122,7 +125,7 @@ static struct kvm_mmu_page *tdp_mmu_next
 						   typeof(*next_root), link);
 
 	while (next_root) {
-		if (!next_root->role.invalid &&
+		if ((!only_valid || !next_root->role.invalid) &&
 		    kvm_tdp_mmu_get_root(kvm, next_root))
 			break;
 
@@ -148,13 +151,19 @@ static struct kvm_mmu_page *tdp_mmu_next
  * mode. In the unlikely event that this thread must free a root, the lock
  * will be temporarily dropped and reacquired in write mode.
  */
-#define for_each_tdp_mmu_root_yield_safe(_kvm, _root, _as_id, _shared)	\
-	for (_root = tdp_mmu_next_root(_kvm, NULL, _shared);		\
-	     _root;							\
-	     _root = tdp_mmu_next_root(_kvm, _root, _shared))		\
-		if (kvm_mmu_page_as_id(_root) != _as_id) {		\
+#define __for_each_tdp_mmu_root_yield_safe(_kvm, _root, _as_id, _shared, _only_valid)\
+	for (_root = tdp_mmu_next_root(_kvm, NULL, _shared, _only_valid);	\
+	     _root;								\
+	     _root = tdp_mmu_next_root(_kvm, _root, _shared, _only_valid))	\
+		if (kvm_mmu_page_as_id(_root) != _as_id) {			\
 		} else
 
+#define for_each_valid_tdp_mmu_root_yield_safe(_kvm, _root, _as_id, _shared)	\
+	__for_each_tdp_mmu_root_yield_safe(_kvm, _root, _as_id, _shared, true)
+
+#define for_each_tdp_mmu_root_yield_safe(_kvm, _root, _as_id, _shared)		\
+	__for_each_tdp_mmu_root_yield_safe(_kvm, _root, _as_id, _shared, false)
+
 #define for_each_tdp_mmu_root(_kvm, _root, _as_id)				\
 	list_for_each_entry_rcu(_root, &_kvm->arch.tdp_mmu_roots, link,		\
 				lockdep_is_held_type(&kvm->mmu_lock, 0) ||	\
@@ -1224,7 +1233,7 @@ bool kvm_tdp_mmu_wrprot_slot(struct kvm
 
 	lockdep_assert_held_read(&kvm->mmu_lock);
 
-	for_each_tdp_mmu_root_yield_safe(kvm, root, slot->as_id, true)
+	for_each_valid_tdp_mmu_root_yield_safe(kvm, root, slot->as_id, true)
 		spte_set |= wrprot_gfn_range(kvm, root, slot->base_gfn,
 			     slot->base_gfn + slot->npages, min_level);
 
@@ -1294,7 +1303,7 @@ bool kvm_tdp_mmu_clear_dirty_slot(struct
 
 	lockdep_assert_held_read(&kvm->mmu_lock);
 
-	for_each_tdp_mmu_root_yield_safe(kvm, root, slot->as_id, true)
+	for_each_valid_tdp_mmu_root_yield_safe(kvm, root, slot->as_id, true)
 		spte_set |= clear_dirty_gfn_range(kvm, root, slot->base_gfn,
 				slot->base_gfn + slot->npages);
 
@@ -1419,7 +1428,7 @@ void kvm_tdp_mmu_zap_collapsible_sptes(s
 
 	lockdep_assert_held_read(&kvm->mmu_lock);
 
-	for_each_tdp_mmu_root_yield_safe(kvm, root, slot->as_id, true)
+	for_each_valid_tdp_mmu_root_yield_safe(kvm, root, slot->as_id, true)
 		zap_collapsible_spte_range(kvm, root, slot);
 }
 
author maintainer
#959 lore raw

PATCH 955/1126 KVM: x86/mmu: Check for present SPTE when clearing dirty bit in TDP MMU

In reply to Greg Kroah-Hartman

From: Sean Christopherson <[email protected]>

commit 3354ef5a592d219364cf442c2f784ce7ad7629fd upstream.

Explicitly check for present SPTEs when clearing dirty bits in the TDP MMU. This isn't strictly required for correctness, as setting the dirty bit in a defunct SPTE will not change the SPTE from !PRESENT to PRESENT. However, the guarded MMU_WARN_ON() in spte_ad_need_write_protect() would complain if anyone actually turned on KVM's MMU debugging.

Fixes a6a0b05da9f3 ("kvm: x86/mmu: Support dirty logging for the TDP MMU")
Cc Ben Gardon <[email protected]>

Message-Id: <[email protected]>

---
 arch/x86/kvm/mmu/tdp_mmu.c |    3 +++
 1 file changed, 3 insertions(+)
arch/x86/kvm/mmu/tdp_mmu.c +3 -0
@@ -1261,6 +1261,9 @@ retry:
 		if (tdp_mmu_iter_cond_resched(kvm, &iter, false, true))
 			continue;
 
+		if (!is_shadow_present_pte(iter.old_spte))
+			continue;
+
 		if (spte_ad_need_write_protect(iter.old_spte)) {
 			if (is_writable_pte(iter.old_spte))
 				new_spte = iter.old_spte & ~PT_WRITABLE_MASK;
author maintainer
#960 lore raw

PATCH 956/1126 KVM: x86: hyper-v: Drop redundant ex parameter from kvm_hv_send_ipi()

In reply to Greg Kroah-Hartman

From: Vitaly Kuznetsov <[email protected]>

commit 50e523dd79f6a856d793ce5711719abe27cffbf2 upstream.

'struct kvm_hv_hcall' has all the required information already, there's no need to pass 'ex' additionally.

No functional change intended.

Message-Id: <[email protected]>

---
 arch/x86/kvm/hyperv.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
arch/x86/kvm/hyperv.c +4 -4
@@ -1875,7 +1875,7 @@ static void kvm_send_ipi_to_many(struct
 	}
 }
 
-static u64 kvm_hv_send_ipi(struct kvm_vcpu *vcpu, struct kvm_hv_hcall *hc, bool ex)
+static u64 kvm_hv_send_ipi(struct kvm_vcpu *vcpu, struct kvm_hv_hcall *hc)
 {
 	struct kvm *kvm = vcpu->kvm;
 	struct hv_send_ipi_ex send_ipi_ex;
@@ -1889,7 +1889,7 @@ static u64 kvm_hv_send_ipi(struct kvm_vc
 	u32 vector;
 	bool all_cpus;
 
-	if (!ex) {
+	if (hc->code == HVCALL_SEND_IPI) {
 		if (!hc->fast) {
 			if (unlikely(kvm_read_guest(kvm, hc->ingpa, &send_ipi,
 						    sizeof(send_ipi))))
@@ -2279,14 +2279,14 @@ int kvm_hv_hypercall(struct kvm_vcpu *vc
 			ret = HV_STATUS_INVALID_HYPERCALL_INPUT;
 			break;
 		}
-		ret = kvm_hv_send_ipi(vcpu, &hc, false);
+		ret = kvm_hv_send_ipi(vcpu, &hc);
 		break;
 	case HVCALL_SEND_IPI_EX:
 		if (unlikely(hc.fast || hc.rep)) {
 			ret = HV_STATUS_INVALID_HYPERCALL_INPUT;
 			break;
 		}
-		ret = kvm_hv_send_ipi(vcpu, &hc, true);
+		ret = kvm_hv_send_ipi(vcpu, &hc);
 		break;
 	case HVCALL_POST_DEBUG_DATA:
 	case HVCALL_RETRIEVE_DEBUG_DATA:
author maintainer
#961 lore raw

PATCH 957/1126 KVM: x86: hyper-v: Drop redundant ex parameter from kvm_hv_flush_tlb()

In reply to Greg Kroah-Hartman

From: Vitaly Kuznetsov <[email protected]>

commit 82c1ead0d678af31e5d883656c12096a0004178b upstream.

'struct kvm_hv_hcall' has all the required information already, there's no need to pass 'ex' additionally.

No functional change intended.

Message-Id: <[email protected]>

---
 arch/x86/kvm/hyperv.c |   23 ++++++-----------------
 1 file changed, 6 insertions(+), 17 deletions(-)
arch/x86/kvm/hyperv.c +6 -17
@@ -1750,7 +1750,7 @@ struct kvm_hv_hcall {
 	sse128_t xmm[HV_HYPERCALL_MAX_XMM_REGISTERS];
 };
 
-static u64 kvm_hv_flush_tlb(struct kvm_vcpu *vcpu, struct kvm_hv_hcall *hc, bool ex)
+static u64 kvm_hv_flush_tlb(struct kvm_vcpu *vcpu, struct kvm_hv_hcall *hc)
 {
 	int i;
 	gpa_t gpa;
@@ -1765,7 +1765,8 @@ static u64 kvm_hv_flush_tlb(struct kvm_v
 	int sparse_banks_len;
 	bool all_cpus;
 
-	if (!ex) {
+	if (hc->code == HVCALL_FLUSH_VIRTUAL_ADDRESS_LIST ||
+	    hc->code == HVCALL_FLUSH_VIRTUAL_ADDRESS_SPACE) {
 		if (hc->fast) {
 			flush.address_space = hc->ingpa;
 			flush.flags = hc->outgpa;
@@ -2247,32 +2248,20 @@ int kvm_hv_hypercall(struct kvm_vcpu *vc
 				kvm_hv_hypercall_complete_userspace;
 		return 0;
 	case HVCALL_FLUSH_VIRTUAL_ADDRESS_LIST:
-		if (unlikely(!hc.rep_cnt || hc.rep_idx)) {
-			ret = HV_STATUS_INVALID_HYPERCALL_INPUT;
-			break;
-		}
-		ret = kvm_hv_flush_tlb(vcpu, &hc, false);
-		break;
-	case HVCALL_FLUSH_VIRTUAL_ADDRESS_SPACE:
-		if (unlikely(hc.rep)) {
-			ret = HV_STATUS_INVALID_HYPERCALL_INPUT;
-			break;
-		}
-		ret = kvm_hv_flush_tlb(vcpu, &hc, false);
-		break;
 	case HVCALL_FLUSH_VIRTUAL_ADDRESS_LIST_EX:
 		if (unlikely(!hc.rep_cnt || hc.rep_idx)) {
 			ret = HV_STATUS_INVALID_HYPERCALL_INPUT;
 			break;
 		}
-		ret = kvm_hv_flush_tlb(vcpu, &hc, true);
+		ret = kvm_hv_flush_tlb(vcpu, &hc);
 		break;
+	case HVCALL_FLUSH_VIRTUAL_ADDRESS_SPACE:
 	case HVCALL_FLUSH_VIRTUAL_ADDRESS_SPACE_EX:
 		if (unlikely(hc.rep)) {
 			ret = HV_STATUS_INVALID_HYPERCALL_INPUT;
 			break;
 		}
-		ret = kvm_hv_flush_tlb(vcpu, &hc, true);
+		ret = kvm_hv_flush_tlb(vcpu, &hc);
 		break;
 	case HVCALL_SEND_IPI:
 		if (unlikely(hc.rep)) {
author maintainer
#962 lore raw

PATCH 958/1126 KVM: x86: hyper-v: Fix the maximum number of sparse banks for XMM fast TLB flush hypercalls

In reply to Greg Kroah-Hartman

From: Vitaly Kuznetsov <[email protected]>

commit 7321f47eada53a395fb3086d49297eebb19e8e58 upstream.

When TLB flush hypercalls (HVCALL_FLUSH_VIRTUAL_ADDRESS_{LIST,SPACE}_EX are issued in 'XMM fast' mode, the maximum number of allowed sparse_banks is not 'HV_HYPERCALL_MAX_XMM_REGISTERS - 1' (5) but twice as many (10) as each XMM register is 128 bit long and can hold two 64 bit long banks.

Cc [email protected] # 5.14.x
Fixes 5974565bc26d ("KVM: x86: kvm_hv_flush_tlb use inputs from XMM registers")

Message-Id: <[email protected]>

---
 arch/x86/kvm/hyperv.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
arch/x86/kvm/hyperv.c +2 -1
@@ -1820,7 +1820,8 @@ static u64 kvm_hv_flush_tlb(struct kvm_v
 
 		if (!all_cpus) {
 			if (hc->fast) {
-				if (sparse_banks_len > HV_HYPERCALL_MAX_XMM_REGISTERS - 1)
+				/* XMM0 is already consumed, each XMM holds two sparse banks. */
+				if (sparse_banks_len > 2 * (HV_HYPERCALL_MAX_XMM_REGISTERS - 1))
 					return HV_STATUS_INVALID_HYPERCALL_INPUT;
 				for (i = 0; i < sparse_banks_len; i += 2) {
 					sparse_banks[i] = sse128_lo(hc->xmm[i / 2 + 1]);
author maintainer
#963 lore raw

PATCH 959/1126 KVM: x86: hyper-v: HVCALL_SEND_IPI_EX is an XMM fast hypercall

In reply to Greg Kroah-Hartman

From: Vitaly Kuznetsov <[email protected]>

commit 47d3e5cdfe607ec6883eb0faa7acf05b8cb3f92a upstream.

It has been proven on practice that at least Windows Server 2019 tries using HVCALL_SEND_IPI_EX in 'XMM fast' mode when it has more than 64 vCPUs and it needs to send an IPI to a vCPU > 63. Similarly to other XMM Fast hypercalls (HVCALL_FLUSH_VIRTUAL_ADDRESS_{LIST,SPACE}{,_EX}), this information is missing in TLFS as of 6.0b. Currently, KVM returns an error (HV_STATUS_INVALID_HYPERCALL_INPUT) and Windows crashes.

Note, HVCALL_SEND_IPI is a 'standard' fast hypercall (not 'XMM fast') as all its parameters fit into RDX:R8 and this is handled by KVM correctly.

Cc [email protected] # 5.14.x: 3244867af8c0: KVM: x86: Ignore sparse banks size for an "all CPUs", non-sparse IPI req
Cc [email protected] # 5.14.x
Fixes d8f5537a8816 ("KVM: hyper-v: Advertise support for fast XMM hypercalls")

Message-Id: <[email protected]>

---
 arch/x86/kvm/hyperv.c |   52 ++++++++++++++++++++++++++++++++------------------
 1 file changed, 34 insertions(+), 18 deletions(-)
arch/x86/kvm/hyperv.c +34 -18
@@ -1890,6 +1890,7 @@ static u64 kvm_hv_send_ipi(struct kvm_vc
 	int sparse_banks_len;
 	u32 vector;
 	bool all_cpus;
+	int i;
 
 	if (hc->code == HVCALL_SEND_IPI) {
 		if (!hc->fast) {
@@ -1910,9 +1911,15 @@ static u64 kvm_hv_send_ipi(struct kvm_vc
 
 		trace_kvm_hv_send_ipi(vector, sparse_banks[0]);
 	} else {
-		if (unlikely(kvm_read_guest(kvm, hc->ingpa, &send_ipi_ex,
-					    sizeof(send_ipi_ex))))
-			return HV_STATUS_INVALID_HYPERCALL_INPUT;
+		if (!hc->fast) {
+			if (unlikely(kvm_read_guest(kvm, hc->ingpa, &send_ipi_ex,
+						    sizeof(send_ipi_ex))))
+				return HV_STATUS_INVALID_HYPERCALL_INPUT;
+		} else {
+			send_ipi_ex.vector = (u32)hc->ingpa;
+			send_ipi_ex.vp_set.format = hc->outgpa;
+			send_ipi_ex.vp_set.valid_bank_mask = sse128_lo(hc->xmm[0]);
+		}
 
 		trace_kvm_hv_send_ipi_ex(send_ipi_ex.vector,
 					 send_ipi_ex.vp_set.format,
@@ -1920,8 +1927,7 @@ static u64 kvm_hv_send_ipi(struct kvm_vc
 
 		vector = send_ipi_ex.vector;
 		valid_bank_mask = send_ipi_ex.vp_set.valid_bank_mask;
-		sparse_banks_len = bitmap_weight(&valid_bank_mask, 64) *
-			sizeof(sparse_banks[0]);
+		sparse_banks_len = bitmap_weight(&valid_bank_mask, 64);
 
 		all_cpus = send_ipi_ex.vp_set.format == HV_GENERIC_SET_ALL;
 
@@ -1931,12 +1937,27 @@ static u64 kvm_hv_send_ipi(struct kvm_vc
 		if (!sparse_banks_len)
 			goto ret_success;
 
-		if (kvm_read_guest(kvm,
-				   hc->ingpa + offsetof(struct hv_send_ipi_ex,
-							vp_set.bank_contents),
-				   sparse_banks,
-				   sparse_banks_len))
-			return HV_STATUS_INVALID_HYPERCALL_INPUT;
+		if (!hc->fast) {
+			if (kvm_read_guest(kvm,
+					   hc->ingpa + offsetof(struct hv_send_ipi_ex,
+								vp_set.bank_contents),
+					   sparse_banks,
+					   sparse_banks_len * sizeof(sparse_banks[0])))
+				return HV_STATUS_INVALID_HYPERCALL_INPUT;
+		} else {
+			/*
+			 * The lower half of XMM0 is already consumed, each XMM holds
+			 * two sparse banks.
+			 */
+			if (sparse_banks_len > (2 * HV_HYPERCALL_MAX_XMM_REGISTERS - 1))
+				return HV_STATUS_INVALID_HYPERCALL_INPUT;
+			for (i = 0; i < sparse_banks_len; i++) {
+				if (i % 2)
+					sparse_banks[i] = sse128_lo(hc->xmm[(i + 1) / 2]);
+				else
+					sparse_banks[i] = sse128_hi(hc->xmm[i / 2]);
+			}
+		}
 	}
 
 check_and_send_ipi:
@@ -2098,6 +2119,7 @@ static bool is_xmm_fast_hypercall(struct
 	case HVCALL_FLUSH_VIRTUAL_ADDRESS_SPACE:
 	case HVCALL_FLUSH_VIRTUAL_ADDRESS_LIST_EX:
 	case HVCALL_FLUSH_VIRTUAL_ADDRESS_SPACE_EX:
+	case HVCALL_SEND_IPI_EX:
 		return true;
 	}
 
@@ -2265,14 +2287,8 @@ int kvm_hv_hypercall(struct kvm_vcpu *vc
 		ret = kvm_hv_flush_tlb(vcpu, &hc);
 		break;
 	case HVCALL_SEND_IPI:
-		if (unlikely(hc.rep)) {
-			ret = HV_STATUS_INVALID_HYPERCALL_INPUT;
-			break;
-		}
-		ret = kvm_hv_send_ipi(vcpu, &hc);
-		break;
 	case HVCALL_SEND_IPI_EX:
-		if (unlikely(hc.fast || hc.rep)) {
+		if (unlikely(hc.rep)) {
 			ret = HV_STATUS_INVALID_HYPERCALL_INPUT;
 			break;
 		}
author maintainer
#964 lore raw

PATCH 960/1126 powerpc/kasan: Fix early region not updated correctly

In reply to Greg Kroah-Hartman

From: Chen Jingwen <[email protected]>

commit dd75080aa8409ce10d50fb58981c6b59bf8707d3 upstream.

The shadow's page table is not updated when PTE_RPN_SHIFT is 24 and PAGE_SHIFT is 12. It not only causes false positives but also false negative as shown the following text.

Fix it by bringing the logic of kasan_early_shadow_page_entry here.

1. False Positive:
==================================================================

BUG: KASAN: vmalloc-out-of-bounds in pcpu_alloc+0x508/0xa50
Write of size 16 at addr f57f3be0 by task swapper/0/1

CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.15.0-12267-gdebe436e77c7 #1
Call Trace:
[c80d1c20] [c07fe7b8] dump_stack_lvl+0x4c/0x6c (unreliable)
[c80d1c40] [c02ff668] print_address_description.constprop.0+0x88/0x300
[c80d1c70] [c02ff45c] kasan_report+0x1ec/0x200
[c80d1cb0] [c0300b20] kasan_check_range+0x160/0x2f0
[c80d1cc0] [c03018a4] memset+0x34/0x90
[c80d1ce0] [c0280108] pcpu_alloc+0x508/0xa50
[c80d1d40] [c02fd7bc] __kmem_cache_create+0xfc/0x570
[c80d1d70] [c0283d64] kmem_cache_create_usercopy+0x274/0x3e0
[c80d1db0] [c2036580] init_sd+0xc4/0x1d0
[c80d1de0] [c00044a0] do_one_initcall+0xc0/0x33c
[c80d1eb0] [c2001624] kernel_init_freeable+0x2c8/0x384
[c80d1ef0] [c0004b14] kernel_init+0x24/0x170
[c80d1f10] [c001b26c] ret_from_kernel_thread+0x5c/0x64

Memory state around the buggy address:
 f57f3a80: f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8
 f57f3b00: f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8
>f57f3b80: f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8
                                               ^
 f57f3c00: f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8
 f57f3c80: f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8
==================================================================

2. False Negative (with KASAN tests):
==================================================================
Before fix:
ok 45 - kmalloc_double_kzfree
# vmalloc_oob: EXPECTATION FAILED at lib/test_kasan.c:1039 KASAN failure expected in "((volatile char *)area)[3100]", but none occurred not ok 46 - vmalloc_oob
not ok 1 - kasan

==================================================================
After fix:
ok 1 - kasan

---
 arch/powerpc/mm/kasan/kasan_init_32.c |    3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
arch/powerpc/mm/kasan/kasan_init_32.c +1 -2
@@ -83,13 +83,12 @@ void __init
 kasan_update_early_region(unsigned long k_start, unsigned long k_end, pte_t pte)
 {
 	unsigned long k_cur;
-	phys_addr_t pa = __pa(kasan_early_shadow_page);
 
 	for (k_cur = k_start; k_cur != k_end; k_cur += PAGE_SIZE) {
 		pmd_t *pmd = pmd_off_k(k_cur);
 		pte_t *ptep = pte_offset_kernel(pmd, k_cur);
 
-		if ((pte_val(*ptep) & PTE_RPN_MASK) != pa)
+		if (pte_page(*ptep) != virt_to_page(lm_alias(kasan_early_shadow_page)))
 			continue;
 
 		__set_pte_at(&init_mm, k_cur, ptep, pte, 0);
author maintainer
#965 lore raw

PATCH 961/1126 powerpc/tm: Fix more userspace r13 corruption

In reply to Greg Kroah-Hartman

From: Nicholas Piggin <[email protected]>

commit 9d71165d3934e607070c4e48458c0cf161b1baea upstream.

Commit cf13435b730a ("powerpc/tm: Fix userspace r13 corruption") fixes a problem in treclaim where a SLB miss can occur on the thread_struct->ckpt_regs while SCRATCH0 is live with the saved user r13 value, clobbering it with the kernel r13 and ultimately resulting in kernel r13 being stored in ckpt_regs.

There is an equivalent problem in trechkpt where the user r13 value is loaded into r13 from chkpt_regs to be recheckpointed, but a SLB miss could occur on ckpt_regs accesses after that, which will result in r13 being clobbered with a kernel value and that will get recheckpointed and then restored to user registers.

The same memory page is accessed right before this critical window where a SLB miss could cause corruption, so hitting the bug requires the SLB entry be removed within a small window of instructions, which is possible if a SLB related MCE hits there. PAPR also permits the hypervisor to discard this SLB entry (because slb_shadow->persistent is only set to SLB_NUM_BOLTED) although it's not known whether any implementations would do this (KVM does not). So this is an extremely unlikely bug, only found by inspection.

Fix this by also storing user r13 in a temporary location on the kernel stack and don't change the r13 register from kernel r13 until the RI=0 critical section that does not fault.

The SCRATCH0 change is not strictly part of the fix, it's only used in the RI=0 section so it does not have the same problem as the previous SCRATCH0 bug.

Fixes 98ae22e15b43 ("powerpc: Add helper functions for transactional memory context switching")
---
 arch/powerpc/kernel/tm.S |   25 ++++++++++++++++---------
 1 file changed, 16 insertions(+), 9 deletions(-)
arch/powerpc/kernel/tm.S +16 -9
@@ -443,7 +443,8 @@ restore_gprs:
 
 	REST_GPR(0, r7)				/* GPR0 */
 	REST_GPRS(2, 4, r7)			/* GPR2-4 */
-	REST_GPRS(8, 31, r7)			/* GPR8-31 */
+	REST_GPRS(8, 12, r7)			/* GPR8-12 */
+	REST_GPRS(14, 31, r7)			/* GPR14-31 */
 
 	/* Load up PPR and DSCR here so we don't run with user values for long */
 	mtspr	SPRN_DSCR, r5
@@ -479,18 +480,24 @@ restore_gprs:
 	REST_GPR(6, r7)
 
 	/*
-	 * Store r1 and r5 on the stack so that we can access them after we
-	 * clear MSR RI.
+	 * Store user r1 and r5 and r13 on the stack (in the unused save
+	 * areas / compiler reserved areas), so that we can access them after
+	 * we clear MSR RI.
 	 */
 
 	REST_GPR(5, r7)
 	std	r5, -8(r1)
-	ld	r5, GPR1(r7)
+	ld	r5, GPR13(r7)
 	std	r5, -16(r1)
+	ld	r5, GPR1(r7)
+	std	r5, -24(r1)
 
 	REST_GPR(7, r7)
 
-	/* Clear MSR RI since we are about to use SCRATCH0. EE is already off */
+	/* Stash the stack pointer away for use after recheckpoint */
+	std	r1, PACAR1(r13)
+
+	/* Clear MSR RI since we are about to clobber r13. EE is already off */
 	li	r5, 0
 	mtmsrd	r5, 1
 
@@ -501,9 +508,9 @@ restore_gprs:
 	 * until we turn MSR RI back on.
 	 */
 
-	SET_SCRATCH0(r1)
 	ld	r5, -8(r1)
-	ld	r1, -16(r1)
+	ld	r13, -16(r1)
+	ld	r1, -24(r1)
 
 	/* Commit register state as checkpointed state: */
 	TRECHKPT
@@ -519,9 +526,9 @@ restore_gprs:
 	 */
 
 	GET_PACA(r13)
-	GET_SCRATCH0(r1)
+	ld	r1, PACAR1(r13)
 
-	/* R1 is restored, so we are recoverable again.  EE is still off */
+	/* R13, R1 is restored, so we are recoverable again.  EE is still off */
 	li	r4, MSR_RI
 	mtmsrd	r4, 1
 
author maintainer
#966 lore raw

PATCH 962/1126 powerpc/lib/sstep: Fix sthcx instruction

In reply to Greg Kroah-Hartman

From: Anders Roxell <[email protected]>

commit a633cb1edddaa643fadc70abc88f89a408fa834a upstream.

Looks like there been a copy paste mistake when added the instruction 'stbcx' twice and one was probably meant to be 'sthcx'. Changing to 'sthcx' from 'stbcx'.

Fixes 350779a29f11 ("powerpc: Handle most loads and stores in instruction emulation code")
Cc [email protected] # v4.14+
---
 arch/powerpc/lib/sstep.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/powerpc/lib/sstep.c +1 -1
@@ -3389,7 +3389,7 @@ int emulate_loadstore(struct pt_regs *re
 			__put_user_asmx(op->val, ea, err, "stbcx.", cr);
 			break;
 		case 2:
-			__put_user_asmx(op->val, ea, err, "stbcx.", cr);
+			__put_user_asmx(op->val, ea, err, "sthcx.", cr);
 			break;
 #endif
 		case 4:
author maintainer
#967 lore raw

PATCH 963/1126 powerpc/lib/sstep: Fix build errors with newer binutils

In reply to Greg Kroah-Hartman

From: Anders Roxell <[email protected]>

commit 8219d31effa7be5dbc7ff915d7970672e028c701 upstream.

Building tinyconfig with gcc (Debian 11.2.0-16) and assembler (Debian 2.37.90.20220207) the following build error shows up:

{standard input}: Assembler messages:
{standard input}:10576: Error: unrecognized opcode: stbcx.' {standard input}:10680: Error: unrecognized opcode: lharx' {standard input}:10694: Error: unrecognized opcode: `lbarx'

Rework to add assembler directives [1] around the instruction. The problem with this might be that we can trick a power6 into single-stepping through an stbcx. for instance, and it will execute that in kernel mode.

[1] https://sourceware.org/binutils/docs/as/PowerPC_002dPseudo.html#PowerPC_002dPseudo

Fixes 350779a29f11 ("powerpc: Handle most loads and stores in instruction emulation code")
Cc [email protected] # v4.14+
---
 arch/powerpc/lib/sstep.c |    6 ++++++
 1 file changed, 6 insertions(+)
arch/powerpc/lib/sstep.c +6 -0
@@ -1097,7 +1097,10 @@ NOKPROBE_SYMBOL(emulate_dcbz);
 
 #define __put_user_asmx(x, addr, err, op, cr)		\
 	__asm__ __volatile__(				\
+		".machine push\n"			\
+		".machine power8\n"			\
 		"1:	" op " %2,0,%3\n"		\
+		".machine pop\n"			\
 		"	mfcr	%1\n"			\
 		"2:\n"					\
 		".section .fixup,\"ax\"\n"		\
@@ -1110,7 +1113,10 @@ NOKPROBE_SYMBOL(emulate_dcbz);
 
 #define __get_user_asmx(x, addr, err, op)		\
 	__asm__ __volatile__(				\
+		".machine push\n"			\
+		".machine power8\n"			\
 		"1:	"op" %1,0,%2\n"			\
+		".machine pop\n"			\
 		"2:\n"					\
 		".section .fixup,\"ax\"\n"		\
 		"3:	li	%0,%3\n"		\
author maintainer
#968 lore raw

PATCH 964/1126 powerpc: Add set_memory_{p/np}() and remove set_memory_attr()

In reply to Greg Kroah-Hartman

From: Christophe Leroy <[email protected]>

commit f222ab83df92acf72691a2021e1f0d99880dcdf1 upstream.

set_memory_attr() was implemented by commit 4d1755b6a762 ("powerpc/mm:
implement set_memory_attr()") because the set_memory_xx() couldn't be used at that time to modify memory "on the fly" as explained it the commit.

But set_memory_attr() uses set_pte_at() which leads to warnings when CONFIG_DEBUG_VM is selected, because set_pte_at() is unexpected for updating existing page table entries.

The check could be bypassed by using __set_pte_at() instead, as it was the case before commit c988cfd38e48 ("powerpc/32:
use set_memory_attr()") but since commit 9f7853d7609d ("powerpc/mm:
Fix set_memory_*() against concurrent accesses") it is now possible to use set_memory_xx() functions to update page table entries "on the fly" because the update is now atomic.

For DEBUG_PAGEALLOC we need to clear and set back _PAGE_PRESENT. Add set_memory_np() and set_memory_p() for that.

Replace all uses of set_memory_attr() by the relevant set_memory_xx() and remove set_memory_attr().

Fixes c988cfd38e48 ("powerpc/32: use set_memory_attr()")

Depends-on: 9f7853d7609d ("powerpc/mm: Fix set_memory_*() against concurrent accesses")

---
 arch/powerpc/include/asm/set_memory.h |   12 +++++++++-
 arch/powerpc/mm/pageattr.c            |   39 +++++-----------------------------
 arch/powerpc/mm/pgtable_32.c          |   24 +++++++++-----------
 3 files changed, 28 insertions(+), 47 deletions(-)
arch/powerpc/include/asm/set_memory.h +11 -1
@@ -6,6 +6,8 @@
 #define SET_MEMORY_RW	1
 #define SET_MEMORY_NX	2
 #define SET_MEMORY_X	3
+#define SET_MEMORY_NP	4	/* Set memory non present */
+#define SET_MEMORY_P	5	/* Set memory present */
 
 int change_memory_attr(unsigned long addr, int numpages, long action);
 
@@ -29,6 +31,14 @@ static inline int set_memory_x(unsigned
 	return change_memory_attr(addr, numpages, SET_MEMORY_X);
 }
 
-int set_memory_attr(unsigned long addr, int numpages, pgprot_t prot);
+static inline int set_memory_np(unsigned long addr, int numpages)
+{
+	return change_memory_attr(addr, numpages, SET_MEMORY_NP);
+}
+
+static inline int set_memory_p(unsigned long addr, int numpages)
+{
+	return change_memory_attr(addr, numpages, SET_MEMORY_P);
+}
 
 #endif
arch/powerpc/mm/pageattr.c +6 -33
@@ -48,6 +48,12 @@ static int change_page_attr(pte_t *ptep,
 	case SET_MEMORY_X:
 		pte = pte_mkexec(pte);
 		break;
+	case SET_MEMORY_NP:
+		pte_update(&init_mm, addr, ptep, _PAGE_PRESENT, 0, 0);
+		break;
+	case SET_MEMORY_P:
+		pte_update(&init_mm, addr, ptep, 0, _PAGE_PRESENT, 0);
+		break;
 	default:
 		WARN_ON_ONCE(1);
 		break;
@@ -96,36 +102,3 @@ int change_memory_attr(unsigned long add
 	return apply_to_existing_page_range(&init_mm, start, size,
 					    change_page_attr, (void *)action);
 }
-
-/*
- * Set the attributes of a page:
- *
- * This function is used by PPC32 at the end of init to set final kernel memory
- * protection. It includes changing the maping of the page it is executing from
- * and data pages it is using.
- */
-static int set_page_attr(pte_t *ptep, unsigned long addr, void *data)
-{
-	pgprot_t prot = __pgprot((unsigned long)data);
-
-	spin_lock(&init_mm.page_table_lock);
-
-	set_pte_at(&init_mm, addr, ptep, pte_modify(*ptep, prot));
-	flush_tlb_kernel_range(addr, addr + PAGE_SIZE);
-
-	spin_unlock(&init_mm.page_table_lock);
-
-	return 0;
-}
-
-int set_memory_attr(unsigned long addr, int numpages, pgprot_t prot)
-{
-	unsigned long start = ALIGN_DOWN(addr, PAGE_SIZE);
-	unsigned long sz = numpages * PAGE_SIZE;
-
-	if (numpages <= 0)
-		return 0;
-
-	return apply_to_existing_page_range(&init_mm, start, sz, set_page_attr,
-					    (void *)pgprot_val(prot));
-}
arch/powerpc/mm/pgtable_32.c +11 -13
@@ -135,10 +135,12 @@ void mark_initmem_nx(void)
 	unsigned long numpages = PFN_UP((unsigned long)_einittext) -
 				 PFN_DOWN((unsigned long)_sinittext);
 
-	if (v_block_mapped((unsigned long)_sinittext))
+	if (v_block_mapped((unsigned long)_sinittext)) {
 		mmu_mark_initmem_nx();
-	else
-		set_memory_attr((unsigned long)_sinittext, numpages, PAGE_KERNEL);
+	} else {
+		set_memory_nx((unsigned long)_sinittext, numpages);
+		set_memory_rw((unsigned long)_sinittext, numpages);
+	}
 }
 
 #ifdef CONFIG_STRICT_KERNEL_RWX
@@ -152,18 +154,14 @@ void mark_rodata_ro(void)
 		return;
 	}
 
-	numpages = PFN_UP((unsigned long)_etext) -
-		   PFN_DOWN((unsigned long)_stext);
-
-	set_memory_attr((unsigned long)_stext, numpages, PAGE_KERNEL_ROX);
 	/*
-	 * mark .rodata as read only. Use __init_begin rather than __end_rodata
-	 * to cover NOTES and EXCEPTION_TABLE.
+	 * mark .text and .rodata as read only. Use __init_begin rather than
+	 * __end_rodata to cover NOTES and EXCEPTION_TABLE.
 	 */
 	numpages = PFN_UP((unsigned long)__init_begin) -
-		   PFN_DOWN((unsigned long)__start_rodata);
+		   PFN_DOWN((unsigned long)_stext);
 
-	set_memory_attr((unsigned long)__start_rodata, numpages, PAGE_KERNEL_RO);
+	set_memory_ro((unsigned long)_stext, numpages);
 
 	// mark_initmem_nx() should have already run by now
 	ptdump_check_wx();
@@ -179,8 +177,8 @@ void __kernel_map_pages(struct page *pag
 		return;
 
 	if (enable)
-		set_memory_attr(addr, numpages, PAGE_KERNEL);
+		set_memory_p(addr, numpages);
 	else
-		set_memory_attr(addr, numpages, __pgprot(0));
+		set_memory_np(addr, numpages);
 }
 #endif /* CONFIG_DEBUG_PAGEALLOC */
author maintainer
#969 lore raw

PATCH 965/1126 powerpc: Fix build errors with newer binutils

In reply to Greg Kroah-Hartman

From: Anders Roxell <[email protected]>

commit 8667d0d64dd1f84fd41b5897fd87fa9113ae05e3 upstream.

Building tinyconfig with gcc (Debian 11.2.0-16) and assembler (Debian 2.37.90.20220207) the following build error shows up:

{standard input}: Assembler messages:
{standard input}:1190: Error: unrecognized opcode: stbcix' {standard input}:1433: Error: unrecognized opcode: lwzcix' {standard input}:1453: Error: unrecognized opcode: stbcix' {standard input}:1460: Error: unrecognized opcode: stwcix' {standard input}:1596: Error: unrecognized opcode: `stbcix'
...

Rework to add assembler directives [1] around the instruction. Going through them one by one shows that the changes should be safe. Like __get_user_atomic_128_aligned() is only called in p9_hmi_special_emu(), which according to the name is specific to power9. And __raw_rm_read*() are only called in things that are powernv or book3s_hv specific.

[1] https://sourceware.org/binutils/docs/as/PowerPC_002dPseudo.html#PowerPC_002dPseudo

---
 arch/powerpc/include/asm/io.h        |   40 ++++++++++++++++++++++++++++-------
 arch/powerpc/include/asm/uaccess.h   |    3 ++
 arch/powerpc/platforms/powernv/rng.c |    6 ++++-
 3 files changed, 40 insertions(+), 9 deletions(-)
arch/powerpc/include/asm/io.h +32 -8
@@ -359,25 +359,37 @@ static inline void __raw_writeq_be(unsig
  */
 static inline void __raw_rm_writeb(u8 val, volatile void __iomem *paddr)
 {
-	__asm__ __volatile__("stbcix %0,0,%1"
+	__asm__ __volatile__(".machine push;   \
+			      .machine power6; \
+			      stbcix %0,0,%1;  \
+			      .machine pop;"
 		: : "r" (val), "r" (paddr) : "memory");
 }
 
 static inline void __raw_rm_writew(u16 val, volatile void __iomem *paddr)
 {
-	__asm__ __volatile__("sthcix %0,0,%1"
+	__asm__ __volatile__(".machine push;   \
+			      .machine power6; \
+			      sthcix %0,0,%1;  \
+			      .machine pop;"
 		: : "r" (val), "r" (paddr) : "memory");
 }
 
 static inline void __raw_rm_writel(u32 val, volatile void __iomem *paddr)
 {
-	__asm__ __volatile__("stwcix %0,0,%1"
+	__asm__ __volatile__(".machine push;   \
+			      .machine power6; \
+			      stwcix %0,0,%1;  \
+			      .machine pop;"
 		: : "r" (val), "r" (paddr) : "memory");
 }
 
 static inline void __raw_rm_writeq(u64 val, volatile void __iomem *paddr)
 {
-	__asm__ __volatile__("stdcix %0,0,%1"
+	__asm__ __volatile__(".machine push;   \
+			      .machine power6; \
+			      stdcix %0,0,%1;  \
+			      .machine pop;"
 		: : "r" (val), "r" (paddr) : "memory");
 }
 
@@ -389,7 +401,10 @@ static inline void __raw_rm_writeq_be(u6
 static inline u8 __raw_rm_readb(volatile void __iomem *paddr)
 {
 	u8 ret;
-	__asm__ __volatile__("lbzcix %0,0, %1"
+	__asm__ __volatile__(".machine push;   \
+			      .machine power6; \
+			      lbzcix %0,0, %1; \
+			      .machine pop;"
 			     : "=r" (ret) : "r" (paddr) : "memory");
 	return ret;
 }
@@ -397,7 +412,10 @@ static inline u8 __raw_rm_readb(volatile
 static inline u16 __raw_rm_readw(volatile void __iomem *paddr)
 {
 	u16 ret;
-	__asm__ __volatile__("lhzcix %0,0, %1"
+	__asm__ __volatile__(".machine push;   \
+			      .machine power6; \
+			      lhzcix %0,0, %1; \
+			      .machine pop;"
 			     : "=r" (ret) : "r" (paddr) : "memory");
 	return ret;
 }
@@ -405,7 +423,10 @@ static inline u16 __raw_rm_readw(volatil
 static inline u32 __raw_rm_readl(volatile void __iomem *paddr)
 {
 	u32 ret;
-	__asm__ __volatile__("lwzcix %0,0, %1"
+	__asm__ __volatile__(".machine push;   \
+			      .machine power6; \
+			      lwzcix %0,0, %1; \
+			      .machine pop;"
 			     : "=r" (ret) : "r" (paddr) : "memory");
 	return ret;
 }
@@ -413,7 +434,10 @@ static inline u32 __raw_rm_readl(volatil
 static inline u64 __raw_rm_readq(volatile void __iomem *paddr)
 {
 	u64 ret;
-	__asm__ __volatile__("ldcix %0,0, %1"
+	__asm__ __volatile__(".machine push;   \
+			      .machine power6; \
+			      ldcix %0,0, %1;  \
+			      .machine pop;"
 			     : "=r" (ret) : "r" (paddr) : "memory");
 	return ret;
 }
arch/powerpc/include/asm/uaccess.h +3 -0
@@ -125,8 +125,11 @@ do {								\
  */
 #define __get_user_atomic_128_aligned(kaddr, uaddr, err)		\
 	__asm__ __volatile__(				\
+		".machine push\n"			\
+		".machine altivec\n"			\
 		"1:	lvx  0,0,%1	# get user\n"	\
 		" 	stvx 0,0,%2	# put kernel\n"	\
+		".machine pop\n"			\
 		"2:\n"					\
 		".section .fixup,\"ax\"\n"		\
 		"3:	li %0,%3\n"			\
arch/powerpc/platforms/powernv/rng.c +5 -1
@@ -43,7 +43,11 @@ static unsigned long rng_whiten(struct p
 	unsigned long parity;
 
 	/* Calculate the parity of the value */
-	asm ("popcntd %0,%1" : "=r" (parity) : "r" (val));
+	asm (".machine push;   \
+	      .machine power7; \
+	      popcntd %0,%1;   \
+	      .machine pop;"
+	     : "=r" (parity) : "r" (val));
 
 	/* xor our value with the previous mask */
 	val ^= rng->mask;
author maintainer
#970 lore raw

PATCH 966/1126 drm/dp: Fix off-by-one in register cache size

In reply to Greg Kroah-Hartman

From: Kees Cook <[email protected]>

commit d4da1f27396fb1dde079447a3612f4f512caed07 upstream.

The pcon_dsc_dpcd array holds 13 registers (0x92 through 0x9E). Fix the math to calculate the max size. Found from a -Warray-bounds build:

drivers/gpu/drm/drm_dp_helper.c: In function 'drm_dp_pcon_dsc_bpp_incr':
drivers/gpu/drm/drm_dp_helper.c:3130:28: error: array subscript 12 is outside array bounds of 'const u8[12]' {aka 'const unsigned char[12]'} [-Werror=array-bounds]

 3130 |         buf = pcon_dsc_dpcd[DP_PCON_DSC_BPP_INCR - DP_PCON_DSC_ENCODER];
      |               ~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
drivers/gpu/drm/drm_dp_helper.c:3126:39: note: while referencing 'pcon_dsc_dpcd'
 3126 | int drm_dp_pcon_dsc_bpp_incr(const u8 pcon_dsc_dpcd[DP_PCON_DSC_ENCODER_CAP_SIZE])
      |                              ~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
---
 include/drm/drm_dp_helper.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
include/drm/drm_dp_helper.h +1 -1
@@ -456,7 +456,7 @@ struct drm_panel;
 #define DP_FEC_CAPABILITY_1			0x091   /* 2.0 */
 
 /* DP-HDMI2.1 PCON DSC ENCODER SUPPORT */
-#define DP_PCON_DSC_ENCODER_CAP_SIZE        0xC	/* 0x9E - 0x92 */
+#define DP_PCON_DSC_ENCODER_CAP_SIZE        0xD	/* 0x92 through 0x9E */
 #define DP_PCON_DSC_ENCODER                 0x092
 # define DP_PCON_DSC_ENCODER_SUPPORTED      (1 << 0)
 # define DP_PCON_DSC_PPS_ENC_OVERRIDE       (1 << 1)
author maintainer
#971 lore raw

PATCH 967/1126 drm/i915: Treat SAGV block time 0 as SAGV disabled

In reply to Greg Kroah-Hartman

From: Ville Syrjälä <[email protected]>

commit 1937f3feb0e84089ae4065e09c871b8ab4676f01 upstream.

For modern platforms the spec explicitly states that a SAGV block time of zero means that SAGV is not supported. Let's extend that to all platforms. Supposedly there should be no systems where this isn't true, and it'll allow us to:
- use the same code regardless of older vs. newer platform
- wm latencies already treat 0 as disabled, so this fits well with other related code
- make it a bit more clear when SAGV is used vs. not
- avoid overflows from adding U32_MAX with a u16 wm0 latency value which could cause us to miscalculate the SAGV watermarks on tgl+

---
 drivers/gpu/drm/i915/intel_pm.c |   10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)
drivers/gpu/drm/i915/intel_pm.c +6 -4
@@ -3722,8 +3722,7 @@ skl_setup_sagv_block_time(struct drm_i91
 		MISSING_CASE(DISPLAY_VER(dev_priv));
 	}
 
-	/* Default to an unusable block time */
-	dev_priv->sagv_block_time_us = -1;
+	dev_priv->sagv_block_time_us = 0;
 }
 
 /*
@@ -5652,7 +5651,7 @@ static void skl_compute_plane_wm(const s
 	result->min_ddb_alloc = max(min_ddb_alloc, blocks) + 1;
 	result->enable = true;
 
-	if (DISPLAY_VER(dev_priv) < 12)
+	if (DISPLAY_VER(dev_priv) < 12 && dev_priv->sagv_block_time_us)
 		result->can_sagv = latency >= dev_priv->sagv_block_time_us;
 }
 
@@ -5683,7 +5682,10 @@ static void tgl_compute_sagv_wm(const st
 	struct drm_i915_private *dev_priv = to_i915(crtc_state->uapi.crtc->dev);
 	struct skl_wm_level *sagv_wm = &plane_wm->sagv.wm0;
 	struct skl_wm_level *levels = plane_wm->wm;
-	unsigned int latency = dev_priv->wm.skl_latency[0] + dev_priv->sagv_block_time_us;
+	unsigned int latency = 0;
+
+	if (dev_priv->sagv_block_time_us)
+		latency = dev_priv->sagv_block_time_us + dev_priv->wm.skl_latency[0];
 
 	skl_compute_plane_wm(crtc_state, 0, latency,
 			     wm_params, &levels[0],
author maintainer
#972 lore raw

PATCH 968/1126 drm/i915: Fix PSF GV point mask when SAGV is not possible

In reply to Greg Kroah-Hartman

From: Ville Syrjälä <[email protected]>

commit 3ef8b5e19ead5a79600ea55f9549658281415893 upstream.

Don't just mask off all the PSF GV points when SAGV gets disabled. This should in fact cause the Pcode to reject the request since at least one PSF point must remain enabled at all times.

Cc Stanislav Lisovskiy <[email protected]>
Fixes 192fbfb76744 ("drm/i915: Implement PSF GV point support")
(cherry picked from commit 0fed4ddd18f064d2359b430c6e83ee60dd1f49b1)
---
 drivers/gpu/drm/i915/display/intel_bw.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/gpu/drm/i915/display/intel_bw.c +2 -1
@@ -966,7 +966,8 @@ int intel_bw_atomic_check(struct intel_a
 	 * cause.
 	 */
 	if (!intel_can_enable_sagv(dev_priv, new_bw_state)) {
-		allowed_points = BIT(max_bw_point);
+		allowed_points &= ADLS_PSF_PT_MASK;
+		allowed_points |= BIT(max_bw_point);
 		drm_dbg_kms(&dev_priv->drm, "No SAGV, using single QGV point %d\n",
 			    max_bw_point);
 	}
author maintainer
#973 lore raw

PATCH 969/1126 drm/i915: Reject unsupported TMDS rates on ICL+

In reply to Greg Kroah-Hartman

From: Ville Syrjälä <[email protected]>

commit 9cddf03b2af07443bebdc73cba21acb360c079e8 upstream.

ICL+ PLLs can't genenerate certain frequencies. Running the PLL algorithms through for all frequencies 25-594MHz we see a gap just above 500 MHz. Specifically 500-522.8MHZ for TC PLLs, and 500-533.2 MHz for combo PHY PLLs. Reject those frequencies hdmi_port_clock_valid() so that we properly filter out unsupported modes and/or color depths for HDMI.

---
 drivers/gpu/drm/i915/display/intel_hdmi.c |    9 +++++++++
 1 file changed, 9 insertions(+)
drivers/gpu/drm/i915/display/intel_hdmi.c +9 -0
@@ -1836,6 +1836,7 @@ hdmi_port_clock_valid(struct intel_hdmi
 		      bool has_hdmi_sink)
 {
 	struct drm_i915_private *dev_priv = intel_hdmi_to_i915(hdmi);
+	enum phy phy = intel_port_to_phy(dev_priv, hdmi_to_dig_port(hdmi)->base.port);
 
 	if (clock < 25000)
 		return MODE_CLOCK_LOW;
@@ -1856,6 +1857,14 @@ hdmi_port_clock_valid(struct intel_hdmi
 	if (IS_CHERRYVIEW(dev_priv) && clock > 216000 && clock < 240000)
 		return MODE_CLOCK_RANGE;
 
+	/* ICL+ combo PHY PLL can't generate 500-533.2 MHz */
+	if (intel_phy_is_combo(dev_priv, phy) && clock > 500000 && clock < 533200)
+		return MODE_CLOCK_RANGE;
+
+	/* ICL+ TC PHY PLL can't generate 500-532.8 MHz */
+	if (intel_phy_is_tc(dev_priv, phy) && clock > 500000 && clock < 532800)
+		return MODE_CLOCK_RANGE;
+
 	/*
 	 * SNPS PHYs' MPLLB table-based programming can only handle a fixed
 	 * set of link rates.
author maintainer
#974 lore raw

PATCH 970/1126 scsi: qla2xxx: Refactor asynchronous command initialization

In reply to Greg Kroah-Hartman

From: Daniel Wagner <[email protected]>

commit d4523bd6fd5d3afa9f08a86038a8a92176089f5b upstream.

Move common open-coded asynchronous command initializing code such as setting up the timer and the done callback into one function. This is a preparation step and allows us later on to change the low level error flow handling at a central place.

---
 drivers/scsi/qla2xxx/qla_gbl.h    |    3 -
 drivers/scsi/qla2xxx/qla_gs.c     |   70 ++++++++++------------------------
 drivers/scsi/qla2xxx/qla_init.c   |   77 ++++++++++++--------------------------
 drivers/scsi/qla2xxx/qla_iocb.c   |   29 +++++++-------
 drivers/scsi/qla2xxx/qla_mbx.c    |   11 +----
 drivers/scsi/qla2xxx/qla_mid.c    |    5 --
 drivers/scsi/qla2xxx/qla_mr.c     |    7 +--
 drivers/scsi/qla2xxx/qla_target.c |    6 --
 8 files changed, 76 insertions(+), 132 deletions(-)
drivers/scsi/qla2xxx/qla_gbl.h +2 -1
@@ -316,7 +316,8 @@ extern int qla2x00_start_sp(srb_t *);
 extern int qla24xx_dif_start_scsi(srb_t *);
 extern int qla2x00_start_bidir(srb_t *, struct scsi_qla_host *, uint32_t);
 extern int qla2xxx_dif_start_scsi_mq(srb_t *);
-extern void qla2x00_init_timer(srb_t *sp, unsigned long tmo);
+extern void qla2x00_init_async_sp(srb_t *sp, unsigned long tmo,
+				  void (*done)(struct srb *, int));
 extern unsigned long qla2x00_get_async_timeout(struct scsi_qla_host *);
 
 extern void *qla2x00_alloc_iocbs(struct scsi_qla_host *, srb_t *);
drivers/scsi/qla2xxx/qla_gs.c +22 -48

202 lines not on this page · raw message

drivers/scsi/qla2xxx/qla_init.c +25 -52

183 lines not on this page · raw message

drivers/scsi/qla2xxx/qla_iocb.c +16 -13
@@ -2560,11 +2560,15 @@ qla24xx_tm_iocb(srb_t *sp, struct tsk_mg
 	}
 }
 
-void qla2x00_init_timer(srb_t *sp, unsigned long tmo)
+void
+qla2x00_init_async_sp(srb_t *sp, unsigned long tmo,
+		     void (*done)(struct srb *sp, int res))
 {
 	timer_setup(&sp->u.iocb_cmd.timer, qla2x00_sp_timeout, 0);
-	sp->u.iocb_cmd.timer.expires = jiffies + tmo * HZ;
+	sp->done = done;
 	sp->free = qla2x00_sp_free;
+	sp->u.iocb_cmd.timeout = qla2x00_async_iocb_timeout;
+	sp->u.iocb_cmd.timer.expires = jiffies + tmo * HZ;
 	if (IS_QLAFX00(sp->vha->hw) && sp->type == SRB_FXIOCB_DCMD)
 		init_completion(&sp->u.iocb_cmd.u.fxiocb.fxiocb_comp);
 	sp->start_timer = 1;
@@ -2672,11 +2676,11 @@ qla24xx_els_dcmd_iocb(scsi_qla_host_t *v
 	sp->type = SRB_ELS_DCMD;
 	sp->name = "ELS_DCMD";
 	sp->fcport = fcport;
-	elsio->timeout = qla2x00_els_dcmd_iocb_timeout;
-	qla2x00_init_timer(sp, ELS_DCMD_TIMEOUT);
-	init_completion(&sp->u.iocb_cmd.u.els_logo.comp);
-	sp->done = qla2x00_els_dcmd_sp_done;
+	qla2x00_init_async_sp(sp, ELS_DCMD_TIMEOUT,
+			      qla2x00_els_dcmd_sp_done);
 	sp->free = qla2x00_els_dcmd_sp_free;
+	sp->u.iocb_cmd.timeout = qla2x00_els_dcmd_iocb_timeout;
+	init_completion(&sp->u.iocb_cmd.u.els_logo.comp);
 
 	elsio->u.els_logo.els_logo_pyld = dma_alloc_coherent(&ha->pdev->dev,
 			    DMA_POOL_SIZE, &elsio->u.els_logo.els_logo_pyld_dma,
@@ -2993,17 +2997,16 @@ qla24xx_els_dcmd2_iocb(scsi_qla_host_t *
 	ql_dbg(ql_dbg_io, vha, 0x3073,
 	       "%s Enter: PLOGI portid=%06x\n", __func__, fcport->d_id.b24);
 
-	sp->type = SRB_ELS_DCMD;
-	sp->name = "ELS_DCMD";
-	sp->fcport = fcport;
-
-	elsio->timeout = qla2x00_els_dcmd2_iocb_timeout;
 	if (wait)
 		sp->flags = SRB_WAKEUP_ON_COMP;
 
-	qla2x00_init_timer(sp, ELS_DCMD_TIMEOUT + 2);
+	sp->type = SRB_ELS_DCMD;
+	sp->name = "ELS_DCMD";
+	sp->fcport = fcport;
+	qla2x00_init_async_sp(sp, ELS_DCMD_TIMEOUT + 2,
+			     qla2x00_els_dcmd2_sp_done);
+	sp->u.iocb_cmd.timeout = qla2x00_els_dcmd2_iocb_timeout;
 
-	sp->done = qla2x00_els_dcmd2_sp_done;
 	elsio->u.els_plogi.tx_size = elsio->u.els_plogi.rx_size = DMA_POOL_SIZE;
 
 	ptr = elsio->u.els_plogi.els_plogi_pyld =
drivers/scsi/qla2xxx/qla_mbx.c +4 -7
@@ -6483,19 +6483,16 @@ int qla24xx_send_mb_cmd(struct scsi_qla_
 	if (!sp)
 		goto done;
 
-	sp->type = SRB_MB_IOCB;
-	sp->name = mb_to_str(mcp->mb[0]);
-
 	c = &sp->u.iocb_cmd;
-	c->timeout = qla2x00_async_iocb_timeout;
 	init_completion(&c->u.mbx.comp);
 
-	qla2x00_init_timer(sp, qla2x00_get_async_timeout(vha) + 2);
+	sp->type = SRB_MB_IOCB;
+	sp->name = mb_to_str(mcp->mb[0]);
+	qla2x00_init_async_sp(sp, qla2x00_get_async_timeout(vha) + 2,
+			      qla2x00_async_mb_sp_done);
 
 	memcpy(sp->u.iocb_cmd.u.mbx.out_mb, mcp->mb, SIZEOF_IOCB_MB_REG);
 
-	sp->done = qla2x00_async_mb_sp_done;
-
 	rval = qla2x00_start_sp(sp);
 	if (rval != QLA_SUCCESS) {
 		ql_dbg(ql_dbg_mbx, vha, 0x1018,
drivers/scsi/qla2xxx/qla_mid.c +2 -3
@@ -972,9 +972,8 @@ int qla24xx_control_vp(scsi_qla_host_t *
 	sp->type = SRB_CTRL_VP;
 	sp->name = "ctrl_vp";
 	sp->comp = &comp;
-	sp->done = qla_ctrlvp_sp_done;
-	sp->u.iocb_cmd.timeout = qla2x00_async_iocb_timeout;
-	qla2x00_init_timer(sp, qla2x00_get_async_timeout(vha) + 2);
+	qla2x00_init_async_sp(sp, qla2x00_get_async_timeout(vha) + 2,
+			      qla_ctrlvp_sp_done);
 	sp->u.iocb_cmd.u.ctrlvp.cmd = cmd;
 	sp->u.iocb_cmd.u.ctrlvp.vp_index = vp_index;
 
drivers/scsi/qla2xxx/qla_mr.c +3 -4
@@ -1793,11 +1793,11 @@ qlafx00_fx_disc(scsi_qla_host_t *vha, fc
 
 	sp->type = SRB_FXIOCB_DCMD;
 	sp->name = "fxdisc";
+	qla2x00_init_async_sp(sp, FXDISC_TIMEOUT,
+			      qla2x00_fxdisc_sp_done);
+	sp->u.iocb_cmd.timeout = qla2x00_fxdisc_iocb_timeout;
 
 	fdisc = &sp->u.iocb_cmd;
-	fdisc->timeout = qla2x00_fxdisc_iocb_timeout;
-	qla2x00_init_timer(sp, FXDISC_TIMEOUT);
-
 	switch (fx_type) {
 	case FXDISC_GET_CONFIG_INFO:
 	fdisc->u.fxiocb.flags =
@@ -1898,7 +1898,6 @@ qlafx00_fx_disc(scsi_qla_host_t *vha, fc
 	}
 
 	fdisc->u.fxiocb.req_func_type = cpu_to_le16(fx_type);
-	sp->done = qla2x00_fxdisc_sp_done;
 
 	rval = qla2x00_start_sp(sp);
 	if (rval != QLA_SUCCESS)
drivers/scsi/qla2xxx/qla_target.c +2 -4
@@ -656,12 +656,10 @@ int qla24xx_async_notify_ack(scsi_qla_ho
 
 	sp->type = type;
 	sp->name = "nack";
-
-	sp->u.iocb_cmd.timeout = qla2x00_async_iocb_timeout;
-	qla2x00_init_timer(sp, qla2x00_get_async_timeout(vha)+2);
+	qla2x00_init_async_sp(sp, qla2x00_get_async_timeout(vha) + 2,
+			      qla2x00_async_nack_sp_done);
 
 	sp->u.iocb_cmd.u.nack.ntfy = ntfy;
-	sp->done = qla2x00_async_nack_sp_done;
 
 	ql_dbg(ql_dbg_disc, vha, 0x20f4,
 	    "Async-%s %8phC hndl %x %s\n",
author maintainer
#975 lore raw

PATCH 971/1126 scsi: qla2xxx: Implement ref count for SRB

In reply to Greg Kroah-Hartman

From: Saurav Kashyap <[email protected]>

commit 31e6cdbe0eae37badceb5e0d4f06cf051432fd77 upstream.

The timeout handler and the done function are racing. When qla2x00_async_iocb_timeout() starts to run it can be preempted by the normal response path (via the firmware?). qla24xx_async_gpsc_sp_done() releases the SRB unconditionally. When scheduling back to qla2x00_async_iocb_timeout() qla24xx_async_abort_cmd() will access an freed sp->qpair pointer:

qla2xxx [0000:83:00.0]-2871:0: Async-gpsc timeout - hdl=63d portid=234500 50:06:0e:80:08:77:b6:21. qla2xxx [0000:83:00.0]-2853:0: Async done-gpsc res 0, WWPN 50:06:0e:80:08:77:b6:21
qla2xxx [0000:83:00.0]-2854:0: Async-gpsc OUT WWPN 20:45:00:27:f8:75:33:00 speeds=2c00 speed=0400. qla2xxx [0000:83:00.0]-28d8:0: qla24xx_handle_gpsc_event 50:06:0e:80:08:77:b6:21 DS 7 LS 6 rc 0 login 1|1 rscn 1|0 lid 5

  BUG: unable to handle kernel NULL pointer dereference at 0000000000000004
  IP: qla24xx_async_abort_cmd+0x1b/0x1c0 [qla2xxx]

Obvious solution to this is to introduce a reference counter. One reference is taken for the normal code path (the 'good' case) and one for the timeout path. As we always race between the normal good case and the timeout/abort handler we need to serialize it. Also we cannot assume any order between the handlers. Since this is slow path we can use proper synchronization via locks.

When we are able to cancel a timer (del_timer returns 1) we know there can't be any error handling in progress because the timeout handler hasn't expired yet, thus we can safely decrement the refcounter by one.

If we are not able to cancel the timer, we know an abort handler is running. We have to make sure we call sp->done() in the abort handlers before calling kref_put().

---
 drivers/scsi/qla2xxx/qla_bsg.c    |    6 +-
 drivers/scsi/qla2xxx/qla_def.h    |    5 ++
 drivers/scsi/qla2xxx/qla_edif.c   |    3 -
 drivers/scsi/qla2xxx/qla_gbl.h    |    1 
 drivers/scsi/qla2xxx/qla_gs.c     |   85 +++++++++++++++++++++++++-------------
 drivers/scsi/qla2xxx/qla_init.c   |   70 +++++++++++++++++++++----------
 drivers/scsi/qla2xxx/qla_inline.h |    2 
 drivers/scsi/qla2xxx/qla_iocb.c   |   41 ++++++++++++++----
 drivers/scsi/qla2xxx/qla_mbx.c    |    4 +
 drivers/scsi/qla2xxx/qla_mid.c    |    4 +
 drivers/scsi/qla2xxx/qla_mr.c     |    4 +
 drivers/scsi/qla2xxx/qla_os.c     |   14 ++++--
 drivers/scsi/qla2xxx/qla_target.c |    4 -
 13 files changed, 173 insertions(+), 70 deletions(-)
drivers/scsi/qla2xxx/qla_bsg.c +4 -2
@@ -29,7 +29,8 @@ void qla2x00_bsg_job_done(srb_t *sp, int
 	    "%s: sp hdl %x, result=%x bsg ptr %p\n",
 	    __func__, sp->handle, res, bsg_job);
 
-	sp->free(sp);
+	/* ref: INIT */
+	kref_put(&sp->cmd_kref, qla2x00_sp_release);
 
 	bsg_reply->result = res;
 	bsg_job_done(bsg_job, bsg_reply->result,
@@ -3013,7 +3014,8 @@ qla24xx_bsg_timeout(struct bsg_job *bsg_
 
 done:
 	spin_unlock_irqrestore(&ha->hardware_lock, flags);
-	sp->free(sp);
+	/* ref: INIT */
+	kref_put(&sp->cmd_kref, qla2x00_sp_release);
 	return 0;
 }
 
drivers/scsi/qla2xxx/qla_def.h +5 -0
@@ -726,6 +726,11 @@ typedef struct srb {
 	 * code.
 	 */
 	void (*put_fn)(struct kref *kref);
+
+	/*
+	 * Report completion for asynchronous commands.
+	 */
+	void (*async_done)(struct srb *sp, int res);
 } srb_t;
 
 #define GET_CMD_SP(sp) (sp->u.scmd.cmd)
drivers/scsi/qla2xxx/qla_edif.c +2 -1
@@ -2146,7 +2146,8 @@ edif_doorbell_show(struct device *dev, s
 
 static void qla_noop_sp_done(srb_t *sp, int res)
 {
-	sp->free(sp);
+	/* ref: INIT */
+	kref_put(&sp->cmd_kref, qla2x00_sp_release);
 }
 
 /*
drivers/scsi/qla2xxx/qla_gbl.h +1 -0
@@ -333,6 +333,7 @@ extern int qla24xx_get_one_block_sg(uint
 extern int qla24xx_configure_prot_mode(srb_t *, uint16_t *);
 extern int qla24xx_issue_sa_replace_iocb(scsi_qla_host_t *vha,
 	struct qla_work_evt *e);
+void qla2x00_sp_release(struct kref *kref);
 
 /*
  * Global Function Prototypes in qla_mbx.c source file.
drivers/scsi/qla2xxx/qla_gs.c +57 -28

289 lines not on this page · raw message

drivers/scsi/qla2xxx/qla_init.c +49 -21

236 lines not on this page · raw message

drivers/scsi/qla2xxx/qla_inline.h +2 -0
@@ -184,6 +184,8 @@ static void qla2xxx_init_sp(srb_t *sp, s
 	sp->vha = vha;
 	sp->qpair = qpair;
 	sp->cmd_type = TYPE_SRB;
+	/* ref : INIT - normal flow */
+	kref_init(&sp->cmd_kref);
 	INIT_LIST_HEAD(&sp->elem);
 }
 
drivers/scsi/qla2xxx/qla_iocb.c +32 -9

102 lines not on this page · raw message

drivers/scsi/qla2xxx/qla_mbx.c +3 -1
@@ -6479,6 +6479,7 @@ int qla24xx_send_mb_cmd(struct scsi_qla_
 	if (!vha->hw->flags.fw_started)
 		goto done;
 
+	/* ref: INIT */
 	sp = qla2x00_get_sp(vha, NULL, GFP_KERNEL);
 	if (!sp)
 		goto done;
@@ -6524,7 +6525,8 @@ int qla24xx_send_mb_cmd(struct scsi_qla_
 	}
 
 done_free_sp:
-	sp->free(sp);
+	/* ref: INIT */
+	kref_put(&sp->cmd_kref, qla2x00_sp_release);
 done:
 	return rval;
 }
drivers/scsi/qla2xxx/qla_mid.c +3 -1
@@ -965,6 +965,7 @@ int qla24xx_control_vp(scsi_qla_host_t *
 	if (vp_index == 0 || vp_index >= ha->max_npiv_vports)
 		return QLA_PARAMETER_ERROR;
 
+	/* ref: INIT */
 	sp = qla2x00_get_sp(base_vha, NULL, GFP_KERNEL);
 	if (!sp)
 		return rval;
@@ -1007,6 +1008,7 @@ int qla24xx_control_vp(scsi_qla_host_t *
 		break;
 	}
 done:
-	sp->free(sp);
+	/* ref: INIT */
+	kref_put(&sp->cmd_kref, qla2x00_sp_release);
 	return rval;
 }
drivers/scsi/qla2xxx/qla_mr.c +3 -1
@@ -1787,6 +1787,7 @@ qlafx00_fx_disc(scsi_qla_host_t *vha, fc
 	struct register_host_info *preg_hsi;
 	struct new_utsname *p_sysid = NULL;
 
+	/* ref: INIT */
 	sp = qla2x00_get_sp(vha, fcport, GFP_KERNEL);
 	if (!sp)
 		goto done;
@@ -1973,7 +1974,8 @@ done_unmap_req:
 		dma_free_coherent(&ha->pdev->dev, fdisc->u.fxiocb.req_len,
 		    fdisc->u.fxiocb.req_addr, fdisc->u.fxiocb.req_dma_handle);
 done_free_sp:
-	sp->free(sp);
+	/* ref: INIT */
+	kref_put(&sp->cmd_kref, qla2x00_sp_release);
 done:
 	return rval;
 }
drivers/scsi/qla2xxx/qla_os.c +10 -4
@@ -728,7 +728,8 @@ void qla2x00_sp_compl(srb_t *sp, int res
 	struct scsi_cmnd *cmd = GET_CMD_SP(sp);
 	struct completion *comp = sp->comp;
 
-	sp->free(sp);
+	/* kref: INIT */
+	kref_put(&sp->cmd_kref, qla2x00_sp_release);
 	cmd->result = res;
 	CMD_SP(cmd) = NULL;
 	scsi_done(cmd);
@@ -819,7 +820,8 @@ void qla2xxx_qpair_sp_compl(srb_t *sp, i
 	struct scsi_cmnd *cmd = GET_CMD_SP(sp);
 	struct completion *comp = sp->comp;
 
-	sp->free(sp);
+	/* ref: INIT */
+	kref_put(&sp->cmd_kref, qla2x00_sp_release);
 	cmd->result = res;
 	CMD_SP(cmd) = NULL;
 	scsi_done(cmd);
@@ -919,6 +921,7 @@ qla2xxx_queuecommand(struct Scsi_Host *h
 		goto qc24_target_busy;
 
 	sp = scsi_cmd_priv(cmd);
+	/* ref: INIT */
 	qla2xxx_init_sp(sp, vha, vha->hw->base_qpair, fcport);
 
 	sp->u.scmd.cmd = cmd;
@@ -938,7 +941,8 @@ qla2xxx_queuecommand(struct Scsi_Host *h
 	return 0;
 
 qc24_host_busy_free_sp:
-	sp->free(sp);
+	/* ref: INIT */
+	kref_put(&sp->cmd_kref, qla2x00_sp_release);
 
 qc24_target_busy:
 	return SCSI_MLQUEUE_TARGET_BUSY;
@@ -1008,6 +1012,7 @@ qla2xxx_mqueuecommand(struct Scsi_Host *
 		goto qc24_target_busy;
 
 	sp = scsi_cmd_priv(cmd);
+	/* ref: INIT */
 	qla2xxx_init_sp(sp, vha, qpair, fcport);
 
 	sp->u.scmd.cmd = cmd;
@@ -1026,7 +1031,8 @@ qla2xxx_mqueuecommand(struct Scsi_Host *
 	return 0;
 
 qc24_host_busy_free_sp:
-	sp->free(sp);
+	/* ref: INIT */
+	kref_put(&sp->cmd_kref, qla2x00_sp_release);
 
 qc24_target_busy:
 	return SCSI_MLQUEUE_TARGET_BUSY;
drivers/scsi/qla2xxx/qla_target.c +2 -2
@@ -620,7 +620,7 @@ static void qla2x00_async_nack_sp_done(s
 	}
 	spin_unlock_irqrestore(&vha->hw->tgt.sess_lock, flags);
 
-	sp->free(sp);
+	kref_put(&sp->cmd_kref, qla2x00_sp_release);
 }
 
 int qla24xx_async_notify_ack(scsi_qla_host_t *vha, fc_port_t *fcport,
@@ -672,7 +672,7 @@ int qla24xx_async_notify_ack(scsi_qla_ho
 	return rval;
 
 done_free_sp:
-	sp->free(sp);
+	kref_put(&sp->cmd_kref, qla2x00_sp_release);
 done:
 	fcport->flags &= ~FCF_ASYNC_SENT;
 	return rval;
author maintainer
#976 lore raw

PATCH 972/1126 scsi: qla2xxx: Fix stuck session in gpdb

In reply to Greg Kroah-Hartman

From: Quinn Tran <[email protected]>

commit 725d3a0d31a51c0debf970011e05f585e805165b upstream.

Fix stuck sessions in get port database. When a thread is in the process of re-establishing a session, a flag is set to prevent multiple threads / triggers from doing the same task. This flag was left on, where any attempt to relogin was locked out. Clear this flag, if the attempt has failed.

---
 drivers/scsi/qla2xxx/qla_init.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/scsi/qla2xxx/qla_init.c +3 -3
@@ -1332,9 +1332,9 @@ int qla24xx_async_gpdb(struct scsi_qla_h
 	if (!vha->flags.online || (fcport->flags & FCF_ASYNC_SENT) ||
 	    fcport->loop_id == FC_NO_LOOP_ID) {
 		ql_log(ql_log_warn, vha, 0xffff,
-		    "%s: %8phC - not sending command.\n",
-		    __func__, fcport->port_name);
-		return rval;
+		    "%s: %8phC online %d flags %x - not sending command.\n",
+		    __func__, fcport->port_name, vha->flags.online, fcport->flags);
+		goto done;
 	}
 
 	sp = qla2x00_get_sp(vha, fcport, GFP_KERNEL);
author maintainer
#977 lore raw

PATCH 973/1126 scsi: qla2xxx: Fix warning message due to adisc being flushed

In reply to Greg Kroah-Hartman

From: Quinn Tran <[email protected]>

commit 64f24af75b79cba3b86b0760e27e0fa904db570f upstream.

Fix warning message due to adisc being flushed. Linux kernel triggered a warning message where a different error code type is not matching up with the expected type. Add additional translation of one error code type to another.

WARNING: CPU: 2 PID: 1131623 at drivers/scsi/qla2xxx/qla_init.c:498
qla2x00_async_adisc_sp_done+0x294/0x2b0 [qla2xxx]
CPU: 2 PID: 1131623 Comm: drmgr Not tainted 5.13.0-rc1-autotest #1

..
GPR28: c000000aaa9c8890 c0080000079ab678 c00000140a104800 c00000002bd19000 NIP [c00800000790857c] qla2x00_async_adisc_sp_done+0x294/0x2b0 [qla2xxx] LR [c008000007908578] qla2x00_async_adisc_sp_done+0x290/0x2b0 [qla2xxx]

Call Trace:
[c00000001cdc3620] [c008000007908578] qla2x00_async_adisc_sp_done+0x290/0x2b0 [qla2xxx] (unreliable)
[c00000001cdc3710] [c0080000078f3080] __qla2x00_abort_all_cmds+0x1b8/0x580 [qla2xxx]
[c00000001cdc3840] [c0080000078f589c] qla2x00_abort_all_cmds+0x34/0xd0 [qla2xxx]
[c00000001cdc3880] [c0080000079153d8] qla2x00_abort_isp_cleanup+0x3f0/0x570 [qla2xxx]
[c00000001cdc3920] [c0080000078fb7e8] qla2x00_remove_one+0x3d0/0x480 [qla2xxx]
[c00000001cdc39b0] [c00000000071c274] pci_device_remove+0x64/0x120
[c00000001cdc39f0] [c0000000007fb818] device_release_driver_internal+0x168/0x2a0
[c00000001cdc3a30] [c00000000070e304] pci_stop_bus_device+0xb4/0x100
[c00000001cdc3a70] [c00000000070e4f0] pci_stop_and_remove_bus_device+0x20/0x40
[c00000001cdc3aa0] [c000000000073940] pci_hp_remove_devices+0x90/0x130
[c00000001cdc3b30] [c0080000070704d0] disable_slot+0x38/0x90 [rpaphp] [
c00000001cdc3b60] [c00000000073eb4c] power_write_file+0xcc/0x180
[c00000001cdc3be0] [c0000000007354bc] pci_slot_attr_store+0x3c/0x60
[c00000001cdc3c00] [c00000000055f820] sysfs_kf_write+0x60/0x80 [c00000001cdc3c20]
[c00000000055df10] kernfs_fop_write_iter+0x1a0/0x290
[c00000001cdc3c70] [c000000000447c4c] new_sync_write+0x14c/0x1d0
[c00000001cdc3d10] [c00000000044b134] vfs_write+0x224/0x330
[c00000001cdc3d60] [c00000000044b3f4] ksys_write+0x74/0x130
[c00000001cdc3db0] [c00000000002df70] system_call_exception+0x150/0x2d0
[c00000001cdc3e10] [c00000000000d45c] system_call_common+0xec/0x278
---
 drivers/scsi/qla2xxx/qla_init.c |    6 ++++++
 1 file changed, 6 insertions(+)
drivers/scsi/qla2xxx/qla_init.c +6 -0
@@ -295,6 +295,8 @@ static void qla2x00_async_login_sp_done(
 		ea.iop[0] = lio->u.logio.iop[0];
 		ea.iop[1] = lio->u.logio.iop[1];
 		ea.sp = sp;
+		if (res)
+			ea.data[0] = MBS_COMMAND_ERROR;
 		qla24xx_handle_plogi_done_event(vha, &ea);
 	}
 
@@ -557,6 +559,8 @@ static void qla2x00_async_adisc_sp_done(
 	ea.iop[1] = lio->u.logio.iop[1];
 	ea.fcport = sp->fcport;
 	ea.sp = sp;
+	if (res)
+		ea.data[0] = MBS_COMMAND_ERROR;
 
 	qla24xx_handle_adisc_event(vha, &ea);
 	/* ref: INIT */
@@ -1237,6 +1241,8 @@ static void qla2x00_async_prli_sp_done(s
 		ea.sp = sp;
 		if (res == QLA_OS_TIMER_EXPIRED)
 			ea.data[0] = QLA_OS_TIMER_EXPIRED;
+		else if (res)
+			ea.data[0] = MBS_COMMAND_ERROR;
 
 		qla24xx_handle_prli_done_event(vha, &ea);
 	}
author maintainer
#978 lore raw

PATCH 974/1126 scsi: qla2xxx: Fix scheduling while atomic

In reply to Greg Kroah-Hartman

From: Quinn Tran <[email protected]>

commit afd438ff874ca40b74321b3fa19bd61adfd7ca0c upstream.

The driver makes a call into midlayer (fc_remote_port_delete) which can put the thread to sleep. The thread that originates the call is in interrupt context. The combination of the two trigger a crash. Schedule the call in non-interrupt context where it is more safe.

kernel: BUG: scheduling while atomic: swapper/7/0/0x00010000
kernel: Call Trace:
kernel:  <IRQ>
kernel:  dump_stack+0x66/0x81
kernel:  __schedule_bug.cold.90+0x5/0x1d
kernel:  __schedule+0x7af/0x960
kernel:  schedule+0x28/0x80
kernel:  schedule_timeout+0x26d/0x3b0
kernel:  wait_for_completion+0xb4/0x140
kernel:  ? wake_up_q+0x70/0x70
kernel:  __wait_rcu_gp+0x12c/0x160
kernel:  ? sdev_evt_alloc+0xc0/0x180 [scsi_mod]
kernel:  synchronize_sched+0x6c/0x80
kernel:  ? call_rcu_bh+0x20/0x20
kernel:  ? __bpf_trace_rcu_invoke_callback+0x10/0x10
kernel:  sdev_evt_alloc+0xfd/0x180 [scsi_mod]
kernel:  starget_for_each_device+0x85/0xb0 [scsi_mod]
kernel:  ? scsi_init_io+0x360/0x3d0 [scsi_mod]
kernel:  scsi_init_io+0x388/0x3d0 [scsi_mod]
kernel:  device_for_each_child+0x54/0x90
kernel:  fc_remote_port_delete+0x70/0xe0 [scsi_transport_fc]
kernel:  qla2x00_schedule_rport_del+0x62/0xf0 [qla2xxx]
kernel:  qla2x00_mark_device_lost+0x9c/0xd0 [qla2xxx]
kernel:  qla24xx_handle_plogi_done_event+0x55f/0x570 [qla2xxx]
kernel:  qla2x00_async_login_sp_done+0xd2/0x100 [qla2xxx]
kernel:  qla24xx_logio_entry+0x13a/0x3c0 [qla2xxx]
kernel:  qla24xx_process_response_queue+0x306/0x400 [qla2xxx]
kernel:  qla24xx_msix_rsp_q+0x3f/0xb0 [qla2xxx]
kernel:  __handle_irq_event_percpu+0x40/0x180
kernel:  handle_irq_event_percpu+0x30/0x80
kernel:  handle_irq_event+0x36/0x60
---
 drivers/scsi/qla2xxx/qla_init.c |    7 +------
 1 file changed, 1 insertion(+), 6 deletions(-)
drivers/scsi/qla2xxx/qla_init.c +1 -6
@@ -2231,12 +2231,7 @@ qla24xx_handle_plogi_done_event(struct s
 		ql_dbg(ql_dbg_disc, vha, 0x20eb, "%s %d %8phC cmd error %x\n",
 		    __func__, __LINE__, ea->fcport->port_name, ea->data[1]);
 
-		ea->fcport->flags &= ~FCF_ASYNC_SENT;
-		qla2x00_set_fcport_disc_state(ea->fcport, DSC_LOGIN_FAILED);
-		if (ea->data[1] & QLA_LOGIO_LOGIN_RETRIED)
-			set_bit(RELOGIN_NEEDED, &vha->dpc_flags);
-		else
-			qla2x00_mark_device_lost(vha, ea->fcport, 1);
+		qlt_schedule_sess_for_deletion(ea->fcport);
 		break;
 	case MBS_LOOP_ID_USED:
 		/* data[1] = IO PARAM 1 = nport ID  */
author maintainer
#979 lore raw

PATCH 975/1126 scsi: qla2xxx: Fix premature hw access after PCI error

In reply to Greg Kroah-Hartman

From: Quinn Tran <[email protected]>

commit e35920ab7874d5e2faeb4f958a74bfa793f1ce5a upstream.

After a recoverable PCI error has been detected and recovered, qla driver needs to check to see if the error condition still persist and/or wait for the OS to give the resume signal.

Sep  8 22:26:03 localhost kernel: WARNING: CPU: 9 PID: 124606 at qla_tmpl.c:440
qla27xx_fwdt_entry_t266+0x55/0x60 [qla2xxx]
Sep  8 22:26:03 localhost kernel: RIP: 0010:qla27xx_fwdt_entry_t266+0x55/0x60
[qla2xxx]
Sep  8 22:26:03 localhost kernel: Call Trace:
Sep  8 22:26:03 localhost kernel: ? qla27xx_walk_template+0xb1/0x1b0 [qla2xxx]
Sep  8 22:26:03 localhost kernel: ? qla27xx_execute_fwdt_template+0x12a/0x160
[qla2xxx]
Sep  8 22:26:03 localhost kernel: ? qla27xx_fwdump+0xa0/0x1c0 [qla2xxx]
Sep  8 22:26:03 localhost kernel: ? qla2xxx_pci_mmio_enabled+0xfb/0x120
[qla2xxx]
Sep  8 22:26:03 localhost kernel: ? report_mmio_enabled+0x44/0x80
Sep  8 22:26:03 localhost kernel: ? report_slot_reset+0x80/0x80
Sep  8 22:26:03 localhost kernel: ? pci_walk_bus+0x70/0x90
Sep  8 22:26:03 localhost kernel: ? aer_dev_correctable_show+0xc0/0xc0
Sep  8 22:26:03 localhost kernel: ? pcie_do_recovery+0x1bb/0x240
Sep  8 22:26:03 localhost kernel: ? aer_recover_work_func+0xaa/0xd0
Sep  8 22:26:03 localhost kernel: ? process_one_work+0x1a7/0x360
..
Sep  8 22:26:03 localhost kernel: qla2xxx [0000:42:00.2]-8041:22: detected PCI
disconnect.
Sep  8 22:26:03 localhost kernel: qla2xxx [0000:42:00.2]-107ff:22:
qla27xx_fwdt_entry_t262: dump ram MB failed. Area 5h start 198013h end 198013h
Sep  8 22:26:03 localhost kernel: qla2xxx [0000:42:00.2]-107ff:22: Unable to
capture FW dump
Sep  8 22:26:03 localhost kernel: qla2xxx [0000:42:00.2]-1015:22: cmd=0x0,
waited 5221 msecs
Sep  8 22:26:03 localhost kernel: qla2xxx [0000:42:00.2]-680d:22: mmio
enabled returning.
Sep  8 22:26:03 localhost kernel: qla2xxx [0000:42:00.2]-d04c:22: MBX

Command timeout for cmd 0, iocontrol=ffffffff jiffies=10140f2e5 mb[0-3]=[0xffff 0xffff 0xffff 0xffff]

---
 drivers/scsi/qla2xxx/qla_os.c   |   10 +++++++++-
 drivers/scsi/qla2xxx/qla_tmpl.c |    9 +++++++--
 2 files changed, 16 insertions(+), 3 deletions(-)
drivers/scsi/qla2xxx/qla_os.c +9 -1
@@ -7639,7 +7639,7 @@ qla2xxx_pci_error_detected(struct pci_de
 
 	switch (state) {
 	case pci_channel_io_normal:
-		ha->flags.eeh_busy = 0;
+		qla_pci_set_eeh_busy(vha);
 		if (ql2xmqsupport || ql2xnvmeenable) {
 			set_bit(QPAIR_ONLINE_CHECK_NEEDED, &vha->dpc_flags);
 			qla2xxx_wake_dpc(vha);
@@ -7680,9 +7680,16 @@ qla2xxx_pci_mmio_enabled(struct pci_dev
 	       "mmio enabled\n");
 
 	ha->pci_error_state = QLA_PCI_MMIO_ENABLED;
+
 	if (IS_QLA82XX(ha))
 		return PCI_ERS_RESULT_RECOVERED;
 
+	if (qla2x00_isp_reg_stat(ha)) {
+		ql_log(ql_log_info, base_vha, 0x803f,
+		    "During mmio enabled, PCI/Register disconnect still detected.\n");
+		goto out;
+	}
+
 	spin_lock_irqsave(&ha->hardware_lock, flags);
 	if (IS_QLA2100(ha) || IS_QLA2200(ha)){
 		stat = rd_reg_word(&reg->hccr);
@@ -7704,6 +7711,7 @@ qla2xxx_pci_mmio_enabled(struct pci_dev
 		    "RISC paused -- mmio_enabled, Dumping firmware.\n");
 		qla2xxx_dump_fw(base_vha);
 	}
+out:
 	/* set PCI_ERS_RESULT_NEED_RESET to trigger call to qla2xxx_pci_slot_reset */
 	ql_dbg(ql_dbg_aer, base_vha, 0x600d,
 	       "mmio enabled returning.\n");
drivers/scsi/qla2xxx/qla_tmpl.c +7 -2
@@ -435,8 +435,13 @@ qla27xx_fwdt_entry_t266(struct scsi_qla_
 {
 	ql_dbg(ql_dbg_misc, vha, 0xd20a,
 	    "%s: reset risc [%lx]\n", __func__, *len);
-	if (buf)
-		WARN_ON_ONCE(qla24xx_soft_reset(vha->hw) != QLA_SUCCESS);
+	if (buf) {
+		if (qla24xx_soft_reset(vha->hw) != QLA_SUCCESS) {
+			ql_dbg(ql_dbg_async, vha, 0x5001,
+			    "%s: unable to soft reset\n", __func__);
+			return INVALID_ENTRY;
+		}
+	}
 
 	return qla27xx_next_entry(ent);
 }
author maintainer
#980 lore raw

PATCH 976/1126 scsi: qla2xxx: Fix wrong FDMI data for 64G adapter

In reply to Greg Kroah-Hartman

From: Bikash Hazarika <[email protected]>

commit 1cfbbacbee2d6ea3816386a483e3c7a96e5bd657 upstream.

Corrected transmission speed mask values for FC.

Supported Speed: 16 32 20 Gb/s ===> Should be 64 instead of 20
Supported Speed: 16G 32G 48G ===> Should be 64G instead of 48G

---
 drivers/scsi/qla2xxx/qla_def.h |    6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
drivers/scsi/qla2xxx/qla_def.h +5 -1
@@ -2891,7 +2891,11 @@ struct ct_fdmi2_hba_attributes {
 #define FDMI_PORT_SPEED_8GB		0x10
 #define FDMI_PORT_SPEED_16GB		0x20
 #define FDMI_PORT_SPEED_32GB		0x40
-#define FDMI_PORT_SPEED_64GB		0x80
+#define FDMI_PORT_SPEED_20GB		0x80
+#define FDMI_PORT_SPEED_40GB		0x100
+#define FDMI_PORT_SPEED_128GB		0x200
+#define FDMI_PORT_SPEED_64GB		0x400
+#define FDMI_PORT_SPEED_256GB		0x800
 #define FDMI_PORT_SPEED_UNKNOWN		0x8000
 
 #define FC_CLASS_2	0x04
author maintainer
#981 lore raw

PATCH 977/1126 scsi: qla2xxx: Fix warning for missing error code

In reply to Greg Kroah-Hartman

From: Nilesh Javali <[email protected]>

commit 14cb838d245ae0d523b2f7804af5a02c22e79f5a upstream.

Fix smatch-reported warning message:

drivers/scsi/qla2xxx/qla_target.c:3324 qlt_xmit_response() warn: missing error code 'res'

---
 drivers/scsi/qla2xxx/qla_target.c |    1 +
 1 file changed, 1 insertion(+)
drivers/scsi/qla2xxx/qla_target.c +1 -0
@@ -3318,6 +3318,7 @@ int qlt_xmit_response(struct qla_tgt_cmd
 			"RESET-RSP online/active/old-count/new-count = %d/%d/%d/%d.\n",
 			vha->flags.online, qla2x00_reset_active(vha),
 			cmd->reset_count, qpair->chip_reset);
+		res = 0;
 		goto out_unmap_unlock;
 	}
 
author maintainer
#982 lore raw

PATCH 978/1126 scsi: qla2xxx: Fix device reconnect in loop topology

In reply to Greg Kroah-Hartman

From: Arun Easi <[email protected]>

commit 8ad4be3d15cf144b5834bdb00d5bbe4050938dc7 upstream.

A device logout in loop topology initiates a device connection teardown which loses the FW device handle. In loop topo, the device handle is not regrabbed leading to device login failures and eventually to loss of the device. Fix this by taking the main login path that does it.

---
 drivers/scsi/qla2xxx/qla_init.c |   15 +++++++++++++++
 drivers/scsi/qla2xxx/qla_os.c   |    5 +++++
 2 files changed, 20 insertions(+)
drivers/scsi/qla2xxx/qla_init.c +15 -0
@@ -974,6 +974,9 @@ static void qla24xx_handle_gnl_done_even
 				set_bit(RELOGIN_NEEDED, &vha->dpc_flags);
 			}
 			break;
+		case ISP_CFG_NL:
+			qla24xx_fcport_handle_login(vha, fcport);
+			break;
 		default:
 			break;
 		}
@@ -1563,6 +1566,11 @@ static void qla_chk_n2n_b4_login(struct
 	u8 login = 0;
 	int rc;
 
+	ql_dbg(ql_dbg_disc, vha, 0x307b,
+	    "%s %8phC DS %d LS %d lid %d retries=%d\n",
+	    __func__, fcport->port_name, fcport->disc_state,
+	    fcport->fw_login_state, fcport->loop_id, fcport->login_retry);
+
 	if (qla_tgt_mode_enabled(vha))
 		return;
 
@@ -5604,6 +5612,13 @@ qla2x00_configure_local_loop(scsi_qla_ho
 			memcpy(fcport->node_name, new_fcport->node_name,
 			    WWN_SIZE);
 			fcport->scan_state = QLA_FCPORT_FOUND;
+			if (fcport->login_retry == 0) {
+				fcport->login_retry = vha->hw->login_retry_count;
+				ql_dbg(ql_dbg_disc, vha, 0x2135,
+				    "Port login retry %8phN, lid 0x%04x retry cnt=%d.\n",
+				    fcport->port_name, fcport->loop_id,
+				    fcport->login_retry);
+			}
 			found++;
 			break;
 		}
drivers/scsi/qla2xxx/qla_os.c +5 -0
@@ -5533,6 +5533,11 @@ void qla2x00_relogin(struct scsi_qla_hos
 					ea.fcport = fcport;
 					qla24xx_handle_relogin_event(vha, &ea);
 				} else if (vha->hw->current_topology ==
+					 ISP_CFG_NL &&
+					IS_QLA2XXX_MIDTYPE(vha->hw)) {
+					(void)qla24xx_fcport_handle_login(vha,
+									fcport);
+				} else if (vha->hw->current_topology ==
 				    ISP_CFG_NL) {
 					fcport->login_retry--;
 					status =
author maintainer
#983 lore raw

PATCH 979/1126 scsi: qla2xxx: edif: Fix clang warning

In reply to Greg Kroah-Hartman

From: Quinn Tran <[email protected]>

commit 73825fd7a37c1a685e9e9e27c9dc91ef1f3e2971 upstream.

Silence compile warning due to unaligned memory access.

qla_edif.c:713:45: warning: taking address of packed member 'u' of class or
structure 'auth_complete_cmd' may result in an unaligned pointer value [-Waddress-of-packed-member]

    fcport = qla2x00_find_fcport_by_pid(vha, &appplogiok.u.d_id);
---
 drivers/scsi/qla2xxx/qla_edif.c |   22 +++++++++++++++++++---
 1 file changed, 19 insertions(+), 3 deletions(-)
drivers/scsi/qla2xxx/qla_edif.c +19 -3
@@ -668,6 +668,11 @@ qla_edif_app_authok(scsi_qla_host_t *vha
 	    bsg_job->request_payload.sg_cnt, &appplogiok,
 	    sizeof(struct auth_complete_cmd));
 
+	/* silent unaligned access warning */
+	portid.b.domain = appplogiok.u.d_id.b.domain;
+	portid.b.area   = appplogiok.u.d_id.b.area;
+	portid.b.al_pa  = appplogiok.u.d_id.b.al_pa;
+
 	switch (appplogiok.type) {
 	case PL_TYPE_WWPN:
 		fcport = qla2x00_find_fcport_by_wwpn(vha,
@@ -678,7 +683,7 @@ qla_edif_app_authok(scsi_qla_host_t *vha
 			    __func__, appplogiok.u.wwpn);
 		break;
 	case PL_TYPE_DID:
-		fcport = qla2x00_find_fcport_by_pid(vha, &appplogiok.u.d_id);
+		fcport = qla2x00_find_fcport_by_pid(vha, &portid);
 		if (!fcport)
 			ql_dbg(ql_dbg_edif, vha, 0x911d,
 			    "%s d_id lookup failed: %x\n", __func__,
@@ -777,6 +782,11 @@ qla_edif_app_authfail(scsi_qla_host_t *v
 	    bsg_job->request_payload.sg_cnt, &appplogifail,
 	    sizeof(struct auth_complete_cmd));
 
+	/* silent unaligned access warning */
+	portid.b.domain = appplogifail.u.d_id.b.domain;
+	portid.b.area   = appplogifail.u.d_id.b.area;
+	portid.b.al_pa  = appplogifail.u.d_id.b.al_pa;
+
 	/*
 	 * TODO: edif: app has failed this plogi. Inform driver to
 	 * take any action (if any).
@@ -788,7 +798,7 @@ qla_edif_app_authfail(scsi_qla_host_t *v
 		SET_DID_STATUS(bsg_reply->result, DID_OK);
 		break;
 	case PL_TYPE_DID:
-		fcport = qla2x00_find_fcport_by_pid(vha, &appplogifail.u.d_id);
+		fcport = qla2x00_find_fcport_by_pid(vha, &portid);
 		if (!fcport)
 			ql_dbg(ql_dbg_edif, vha, 0x911d,
 			    "%s d_id lookup failed: %x\n", __func__,
@@ -1253,6 +1263,7 @@ qla24xx_sadb_update(struct bsg_job *bsg_
 	int result = 0;
 	struct qla_sa_update_frame sa_frame;
 	struct srb_iocb *iocb_cmd;
+	port_id_t portid;
 
 	ql_dbg(ql_dbg_edif + ql_dbg_verbose, vha, 0x911d,
 	    "%s entered, vha: 0x%p\n", __func__, vha);
@@ -1276,7 +1287,12 @@ qla24xx_sadb_update(struct bsg_job *bsg_
 		goto done;
 	}
 
-	fcport = qla2x00_find_fcport_by_pid(vha, &sa_frame.port_id);
+	/* silent unaligned access warning */
+	portid.b.domain = sa_frame.port_id.b.domain;
+	portid.b.area   = sa_frame.port_id.b.area;
+	portid.b.al_pa  = sa_frame.port_id.b.al_pa;
+
+	fcport = qla2x00_find_fcport_by_pid(vha, &portid);
 	if (fcport) {
 		found = 1;
 		if (sa_frame.flags == QLA_SA_UPDATE_FLAGS_TX_KEY)
author maintainer
#984 lore raw

PATCH 980/1126 scsi: qla2xxx: Fix T10 PI tag escape and IP guard options for 28XX adapters

In reply to Greg Kroah-Hartman

From: Joe Carnuccio <[email protected]>

commit 4c103a802c69fca63976af6b372ccba39ed74370 upstream.

28XX adapters are capable of detecting both T10 PI tag escape values as well as IP guard. This was missed due to the adapter type missed in the corresponding macros. Fix this by adding support for 28xx in those macros.

---
 drivers/scsi/qla2xxx/qla_def.h |    6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/scsi/qla2xxx/qla_def.h +4 -2
@@ -4271,8 +4271,10 @@ struct qla_hw_data {
 #define QLA_ABTS_WAIT_ENABLED(_sp) \
 	(QLA_NVME_IOS(_sp) && QLA_ABTS_FW_ENABLED(_sp->fcport->vha->hw))
 
-#define IS_PI_UNINIT_CAPABLE(ha)	(IS_QLA83XX(ha) || IS_QLA27XX(ha))
-#define IS_PI_IPGUARD_CAPABLE(ha)	(IS_QLA83XX(ha) || IS_QLA27XX(ha))
+#define IS_PI_UNINIT_CAPABLE(ha)	(IS_QLA83XX(ha) || IS_QLA27XX(ha) || \
+					 IS_QLA28XX(ha))
+#define IS_PI_IPGUARD_CAPABLE(ha)	(IS_QLA83XX(ha) || IS_QLA27XX(ha) || \
+					 IS_QLA28XX(ha))
 #define IS_PI_DIFB_DIX0_CAPABLE(ha)	(0)
 #define IS_PI_SPLIT_DET_CAPABLE_HBA(ha)	(IS_QLA83XX(ha) || IS_QLA27XX(ha) || \
 					IS_QLA28XX(ha))
author maintainer
#985 lore raw

PATCH 981/1126 scsi: qla2xxx: Add devids and conditionals for 28xx

In reply to Greg Kroah-Hartman

From: Joe Carnuccio <[email protected]>

commit 0d6a536cb1fcabb6c3e9c94871c8d0b29bb5813b upstream.

This is an update to the original 28xx adapter enablement. Add a bunch of conditionals that are applicable for 28xx.

---
 drivers/scsi/qla2xxx/qla_attr.c   |    7 ++-----
 drivers/scsi/qla2xxx/qla_init.c   |    8 +++-----
 drivers/scsi/qla2xxx/qla_mbx.c    |   14 +++++++++++---
 drivers/scsi/qla2xxx/qla_os.c     |    3 +--
 drivers/scsi/qla2xxx/qla_sup.c    |    4 ++--
 drivers/scsi/qla2xxx/qla_target.c |    3 +--
 6 files changed, 20 insertions(+), 19 deletions(-)
drivers/scsi/qla2xxx/qla_attr.c +2 -5
@@ -555,7 +555,7 @@ qla2x00_sysfs_read_vpd(struct file *filp
 	if (!capable(CAP_SYS_ADMIN))
 		return -EINVAL;
 
-	if (IS_NOCACHE_VPD_TYPE(ha))
+	if (!IS_NOCACHE_VPD_TYPE(ha))
 		goto skip;
 
 	faddr = ha->flt_region_vpd << 2;
@@ -745,7 +745,7 @@ qla2x00_sysfs_write_reset(struct file *f
 		ql_log(ql_log_info, vha, 0x706f,
 		    "Issuing MPI reset.\n");
 
-		if (IS_QLA83XX(ha) || IS_QLA27XX(ha) || IS_QLA28XX(ha)) {
+		if (IS_QLA83XX(ha)) {
 			uint32_t idc_control;
 
 			qla83xx_idc_lock(vha, 0);
@@ -1056,9 +1056,6 @@ qla2x00_free_sysfs_attr(scsi_qla_host_t
 			continue;
 		if (iter->type == 3 && !(IS_CNA_CAPABLE(ha)))
 			continue;
-		if (iter->type == 0x27 &&
-		    (!IS_QLA27XX(ha) || !IS_QLA28XX(ha)))
-			continue;
 
 		sysfs_remove_bin_file(&host->shost_gendev.kobj,
 		    iter->attr);
drivers/scsi/qla2xxx/qla_init.c +3 -5
@@ -3492,7 +3492,7 @@ qla2x00_alloc_fw_dump(scsi_qla_host_t *v
 		mem_size = (ha->fw_memory_size - 0x11000 + 1) *
 		    sizeof(uint16_t);
 	} else if (IS_FWI2_CAPABLE(ha)) {
-		if (IS_QLA83XX(ha) || IS_QLA27XX(ha) || IS_QLA28XX(ha))
+		if (IS_QLA83XX(ha))
 			fixed_size = offsetof(struct qla83xx_fw_dump, ext_mem);
 		else if (IS_QLA81XX(ha))
 			fixed_size = offsetof(struct qla81xx_fw_dump, ext_mem);
@@ -3504,8 +3504,7 @@ qla2x00_alloc_fw_dump(scsi_qla_host_t *v
 		mem_size = (ha->fw_memory_size - 0x100000 + 1) *
 		    sizeof(uint32_t);
 		if (ha->mqenable) {
-			if (!IS_QLA83XX(ha) && !IS_QLA27XX(ha) &&
-			    !IS_QLA28XX(ha))
+			if (!IS_QLA83XX(ha))
 				mq_size = sizeof(struct qla2xxx_mq_chain);
 			/*
 			 * Allocate maximum buffer size for all queues - Q0.
@@ -4066,8 +4065,7 @@ enable_82xx_npiv:
 			    ha->fw_major_version, ha->fw_minor_version,
 			    ha->fw_subminor_version);
 
-			if (IS_QLA83XX(ha) || IS_QLA27XX(ha) ||
-			    IS_QLA28XX(ha)) {
+			if (IS_QLA83XX(ha)) {
 				ha->flags.fac_supported = 0;
 				rval = QLA_SUCCESS;
 			}
drivers/scsi/qla2xxx/qla_mbx.c +11 -3
@@ -9,6 +9,12 @@
 #include <linux/delay.h>
 #include <linux/gfp.h>
 
+#ifdef CONFIG_PPC
+#define IS_PPCARCH      true
+#else
+#define IS_PPCARCH      false
+#endif
+
 static struct mb_cmd_name {
 	uint16_t cmd;
 	const char *str;
@@ -728,6 +734,9 @@ again:
 				vha->min_supported_speed =
 				    nv->min_supported_speed;
 			}
+
+			if (IS_PPCARCH)
+				mcp->mb[11] |= BIT_4;
 		}
 
 		if (ha->flags.exlogins_enabled)
@@ -3029,8 +3038,7 @@ qla2x00_get_resource_cnts(scsi_qla_host_
 		ha->orig_fw_iocb_count = mcp->mb[10];
 		if (ha->flags.npiv_supported)
 			ha->max_npiv_vports = mcp->mb[11];
-		if (IS_QLA81XX(ha) || IS_QLA83XX(ha) || IS_QLA27XX(ha) ||
-		    IS_QLA28XX(ha))
+		if (IS_QLA81XX(ha) || IS_QLA83XX(ha))
 			ha->fw_max_fcf_count = mcp->mb[12];
 	}
 
@@ -5621,7 +5629,7 @@ qla2x00_get_data_rate(scsi_qla_host_t *v
 	mcp->out_mb = MBX_1|MBX_0;
 	mcp->in_mb = MBX_2|MBX_1|MBX_0;
 	if (IS_QLA83XX(ha) || IS_QLA27XX(ha) || IS_QLA28XX(ha))
-		mcp->in_mb |= MBX_3;
+		mcp->in_mb |= MBX_4|MBX_3;
 	mcp->tov = MBX_TOV_SECONDS;
 	mcp->flags = 0;
 	rval = qla2x00_mailbox_command(vha, mcp);
drivers/scsi/qla2xxx/qla_os.c +1 -2
@@ -3754,8 +3754,7 @@ qla2x00_unmap_iobases(struct qla_hw_data
 		if (ha->mqiobase)
 			iounmap(ha->mqiobase);
 
-		if ((IS_QLA83XX(ha) || IS_QLA27XX(ha) || IS_QLA28XX(ha)) &&
-		    ha->msixbase)
+		if (ha->msixbase)
 			iounmap(ha->msixbase);
 	}
 }
drivers/scsi/qla2xxx/qla_sup.c +2 -2
@@ -844,7 +844,7 @@ qla2xxx_get_flt_info(scsi_qla_host_t *vh
 				ha->flt_region_nvram = start;
 			break;
 		case FLT_REG_IMG_PRI_27XX:
-			if (IS_QLA27XX(ha) && !IS_QLA28XX(ha))
+			if (IS_QLA27XX(ha) || IS_QLA28XX(ha))
 				ha->flt_region_img_status_pri = start;
 			break;
 		case FLT_REG_IMG_SEC_27XX:
@@ -1356,7 +1356,7 @@ next:
 		    flash_data_addr(ha, faddr), le32_to_cpu(*dwptr));
 		if (ret) {
 			ql_dbg(ql_dbg_user, vha, 0x7006,
-			    "Failed slopw write %x (%x)\n", faddr, *dwptr);
+			    "Failed slow write %x (%x)\n", faddr, *dwptr);
 			break;
 		}
 	}
drivers/scsi/qla2xxx/qla_target.c +1 -2
@@ -7220,8 +7220,7 @@ qlt_probe_one_stage1(struct scsi_qla_hos
 	if (!QLA_TGT_MODE_ENABLED())
 		return;
 
-	if  ((ql2xenablemsix == 0) || IS_QLA83XX(ha) || IS_QLA27XX(ha) ||
-	    IS_QLA28XX(ha)) {
+	if  (ha->mqenable || IS_QLA83XX(ha) || IS_QLA27XX(ha) || IS_QLA28XX(ha)) {
 		ISP_ATIO_Q_IN(base_vha) = &ha->mqiobase->isp25mq.atio_q_in;
 		ISP_ATIO_Q_OUT(base_vha) = &ha->mqiobase->isp25mq.atio_q_out;
 	} else {
author maintainer
#986 lore raw

PATCH 982/1126 scsi: qla2xxx: Check for firmware dump already collected

In reply to Greg Kroah-Hartman

From: Joe Carnuccio <[email protected]>

commit cfbafad7c6032d449a5a07f2d273acd2437bbc6a upstream.

While allocating firmware dump, check if dump is already collected and do not re-allocate the buffer.

---
 drivers/scsi/qla2xxx/qla_init.c |    8 ++++++++
 1 file changed, 8 insertions(+)
drivers/scsi/qla2xxx/qla_init.c +8 -0
@@ -3482,6 +3482,14 @@ qla2x00_alloc_fw_dump(scsi_qla_host_t *v
 	struct rsp_que *rsp = ha->rsp_q_map[0];
 	struct qla2xxx_fw_dump *fw_dump;
 
+	if (ha->fw_dump) {
+		ql_dbg(ql_dbg_init, vha, 0x00bd,
+		    "Firmware dump already allocated.\n");
+		return;
+	}
+
+	ha->fw_dumped = 0;
+	ha->fw_dump_cap_flags = 0;
 	dump_size = fixed_size = mem_size = eft_size = fce_size = mq_size = 0;
 	req_q_size = rsp_q_size = 0;
 
author maintainer
#987 lore raw

PATCH 983/1126 scsi: qla2xxx: Suppress a kernel complaint in qla_create_qpair()

In reply to Greg Kroah-Hartman

From: Saurav Kashyap <[email protected]>

commit a60447e7d451df42c7bde43af53b34f10f34f469 upstream.

[   12.323788] BUG: using smp_processor_id() in preemptible [00000000] code: systemd-udevd/1020
[   12.332297] caller is qla2xxx_create_qpair+0x32a/0x5d0 [qla2xxx]
[   12.338417] CPU: 7 PID: 1020 Comm: systemd-udevd Tainted: G          I      --------- ---  5.14.0-29.el9.x86_64 #1
[   12.348827] Hardware name: Dell Inc. PowerEdge R610/0F0XJ6, BIOS 6.6.0 05/22/2018
[   12.356356] Call Trace:
[   12.358821]  dump_stack_lvl+0x34/0x44
[   12.362514]  check_preemption_disabled+0xd9/0xe0
[   12.367164]  qla2xxx_create_qpair+0x32a/0x5d0 [qla2xxx]
[   12.372481]  qla2x00_probe_one+0xa3a/0x1b80 [qla2xxx]
[   12.377617]  ? _raw_spin_lock_irqsave+0x19/0x40
[   12.384284]  local_pci_probe+0x42/0x80
[   12.390162]  ? pci_match_device+0xd7/0x110
[   12.396366]  pci_device_probe+0xfd/0x1b0
[   12.402372]  really_probe+0x1e7/0x3e0
[   12.408114]  __driver_probe_device+0xfe/0x180
[   12.414544]  driver_probe_device+0x1e/0x90
[   12.420685]  __driver_attach+0xc0/0x1c0
[   12.426536]  ? __device_attach_driver+0xe0/0xe0
[   12.433061]  ? __device_attach_driver+0xe0/0xe0
[   12.439538]  bus_for_each_dev+0x78/0xc0
[   12.445294]  bus_add_driver+0x12b/0x1e0
[   12.451021]  driver_register+0x8f/0xe0
[   12.456631]  ? 0xffffffffc07bc000
[   12.461773]  qla2x00_module_init+0x1be/0x229 [qla2xxx]
[   12.468776]  do_one_initcall+0x44/0x200
[   12.474401]  ? load_module+0xad3/0xba0
[   12.479908]  ? kmem_cache_alloc_trace+0x45/0x410
[   12.486268]  do_init_module+0x5c/0x280
[   12.491730]  __do_sys_init_module+0x12e/0x1b0
[   12.497785]  do_syscall_64+0x3b/0x90
[   12.503029]  entry_SYSCALL_64_after_hwframe+0x44/0xae
[   12.509764] RIP: 0033:0x7f554f73ab2e
---
 drivers/scsi/qla2xxx/qla_init.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/scsi/qla2xxx/qla_init.c +1 -1
@@ -9413,7 +9413,7 @@ struct qla_qpair *qla2xxx_create_qpair(s
 		qpair->rsp->req = qpair->req;
 		qpair->rsp->qpair = qpair;
 		/* init qpair to this cpu. Will adjust at run time. */
-		qla_cpu_update(qpair, smp_processor_id());
+		qla_cpu_update(qpair, raw_smp_processor_id());
 
 		if (IS_T10_PI_CAPABLE(ha) && ql2xenabledif) {
 			if (ha->fw_attributes & BIT_4)
author maintainer
#988 lore raw

PATCH 984/1126 scsi: qla2xxx: Fix disk failure to rediscover

In reply to Greg Kroah-Hartman

From: Quinn Tran <[email protected]>

commit 6a45c8e137d4e2c72eecf1ac7cf64f2fdfcead99 upstream.

User experienced some of the LUN failed to get rediscovered after long cable pull test. The issue is triggered by a race condition between driver setting session online state vs starting the LUN scan process at the same time. Current code set the online state after notifying the session is available. In this case, trigger to start the LUN scan process happened before driver could set the session in online state. LUN scan ends up with failure due to the session online check was failing.

Set the online state before reporting of the availability of the session.

---
 drivers/scsi/qla2xxx/qla_init.c |    5 +++--
 drivers/scsi/qla2xxx/qla_nvme.c |    5 +++++
 2 files changed, 8 insertions(+), 2 deletions(-)
drivers/scsi/qla2xxx/qla_init.c +3 -2
@@ -5758,6 +5758,8 @@ qla2x00_reg_remote_port(scsi_qla_host_t
 	if (atomic_read(&fcport->state) == FCS_ONLINE)
 		return;
 
+	qla2x00_set_fcport_state(fcport, FCS_ONLINE);
+
 	rport_ids.node_name = wwn_to_u64(fcport->node_name);
 	rport_ids.port_name = wwn_to_u64(fcport->port_name);
 	rport_ids.port_id = fcport->d_id.b.domain << 16 |
@@ -5858,6 +5860,7 @@ qla2x00_update_fcport(scsi_qla_host_t *v
 		qla2x00_reg_remote_port(vha, fcport);
 		break;
 	case MODE_TARGET:
+		qla2x00_set_fcport_state(fcport, FCS_ONLINE);
 		if (!vha->vha_tgt.qla_tgt->tgt_stop &&
 			!vha->vha_tgt.qla_tgt->tgt_stopped)
 			qlt_fc_port_added(vha, fcport);
@@ -5875,8 +5878,6 @@ qla2x00_update_fcport(scsi_qla_host_t *v
 	if (NVME_TARGET(vha->hw, fcport))
 		qla_nvme_register_remote(vha, fcport);
 
-	qla2x00_set_fcport_state(fcport, FCS_ONLINE);
-
 	if (IS_IIDMA_CAPABLE(vha->hw) && vha->hw->flags.gpsc_supported) {
 		if (fcport->id_changed) {
 			fcport->id_changed = 0;
drivers/scsi/qla2xxx/qla_nvme.c +5 -0
@@ -37,6 +37,11 @@ int qla_nvme_register_remote(struct scsi
 		(fcport->nvme_flag & NVME_FLAG_REGISTERED))
 		return 0;
 
+	if (atomic_read(&fcport->state) == FCS_ONLINE)
+		return 0;
+
+	qla2x00_set_fcport_state(fcport, FCS_ONLINE);
+
 	fcport->nvme_flag &= ~NVME_FLAG_RESETTING;
 
 	memset(&req, 0, sizeof(struct nvme_fc_port_info));
author maintainer
#989 lore raw

PATCH 985/1126 scsi: qla2xxx: Fix incorrect reporting of task management failure

In reply to Greg Kroah-Hartman

From: Quinn Tran <[email protected]>

commit 58ca5999e0367d131de82a75257fbfd5aed0195d upstream.

User experienced no task management error while target device is responding with error. The RSP_CODE field in the status IOCB is in little endian. Driver assumes it's big endian and it picked up erroneous data.

Convert the data back to big endian as is on the wire.

---
 drivers/scsi/qla2xxx/qla_isr.c |    1 +
 1 file changed, 1 insertion(+)
drivers/scsi/qla2xxx/qla_isr.c +1 -0
@@ -2498,6 +2498,7 @@ qla24xx_tm_iocb_entry(scsi_qla_host_t *v
 		iocb->u.tmf.data = QLA_FUNCTION_FAILED;
 	} else if ((le16_to_cpu(sts->scsi_status) &
 	    SS_RESPONSE_INFO_LEN_VALID)) {
+		host_to_fcp_swap(sts->data, sizeof(sts->data));
 		if (le32_to_cpu(sts->rsp_data_len) < 4) {
 			ql_log(ql_log_warn, fcport->vha, 0x503b,
 			    "Async-%s error - hdl=%x not enough response(%d).\n",
author maintainer
#990 lore raw

PATCH 986/1126 scsi: qla2xxx: Fix hang due to session stuck

In reply to Greg Kroah-Hartman

From: Quinn Tran <[email protected]>

commit c02aada06d19a215c8291bd968a99a270e96f734 upstream.

User experienced device lost. The log shows Get port data base command was queued up, failed, and requeued again. Every time it is requeued, it set the FCF_ASYNC_ACTIVE. This prevents any recovery code from occurring because driver thinks a recovery is in progress for this session. In essence, this session is hung. The reason it gets into this place is the session deletion got in front of this call due to link perturbation.

Break the requeue cycle and exit. The session deletion code will trigger a session relogin.

---
 drivers/scsi/qla2xxx/qla_def.h  |    4 ++++
 drivers/scsi/qla2xxx/qla_init.c |   19 +++++++++++++++++--
 2 files changed, 21 insertions(+), 2 deletions(-)
drivers/scsi/qla2xxx/qla_def.h +4 -0
@@ -5438,4 +5438,8 @@ struct ql_vnd_tgt_stats_resp {
 #include "qla_gbl.h"
 #include "qla_dbg.h"
 #include "qla_inline.h"
+
+#define IS_SESSION_DELETED(_fcport) (_fcport->disc_state == DSC_DELETE_PEND || \
+				      _fcport->disc_state == DSC_DELETED)
+
 #endif
drivers/scsi/qla2xxx/qla_init.c +17 -2
@@ -575,6 +575,14 @@ qla2x00_async_adisc(struct scsi_qla_host
 	struct srb_iocb *lio;
 	int rval = QLA_FUNCTION_FAILED;
 
+	if (IS_SESSION_DELETED(fcport)) {
+		ql_log(ql_log_warn, vha, 0xffff,
+		       "%s: %8phC is being delete - not sending command.\n",
+		       __func__, fcport->port_name);
+		fcport->flags &= ~FCF_ASYNC_ACTIVE;
+		return rval;
+	}
+
 	if (!vha->flags.online || (fcport->flags & FCF_ASYNC_SENT))
 		return rval;
 
@@ -1338,8 +1346,15 @@ int qla24xx_async_gpdb(struct scsi_qla_h
 	struct port_database_24xx *pd;
 	struct qla_hw_data *ha = vha->hw;
 
-	if (!vha->flags.online || (fcport->flags & FCF_ASYNC_SENT) ||
-	    fcport->loop_id == FC_NO_LOOP_ID) {
+	if (IS_SESSION_DELETED(fcport)) {
+		ql_log(ql_log_warn, vha, 0xffff,
+		       "%s: %8phC is being delete - not sending command.\n",
+		       __func__, fcport->port_name);
+		fcport->flags &= ~FCF_ASYNC_ACTIVE;
+		return rval;
+	}
+
+	if (!vha->flags.online || fcport->flags & FCF_ASYNC_SENT) {
 		ql_log(ql_log_warn, vha, 0xffff,
 		    "%s: %8phC online %d flags %x - not sending command.\n",
 		    __func__, fcport->port_name, vha->flags.online, fcport->flags);
author maintainer
#991 lore raw

PATCH 987/1126 scsi: qla2xxx: Fix laggy FC remote port session recovery

In reply to Greg Kroah-Hartman

From: Quinn Tran <[email protected]>

commit 713b415726f100f6644971e75ebfe1edbef1a390 upstream.

For session recovery, driver relies on the dpc thread to initiate certain operations. The dpc thread runs exclusively without the Mailbox interface being occupied. A recent code change for heartbeat check via mailbox cmd 0 is preventing the dpc thread from carrying out its operation. This patch allows the higher priority error recovery to run first before running the lower priority heartbeat check.

---
 drivers/scsi/qla2xxx/qla_def.h |    1 +
 drivers/scsi/qla2xxx/qla_os.c  |   20 +++++++++++++++++---
 2 files changed, 18 insertions(+), 3 deletions(-)
drivers/scsi/qla2xxx/qla_def.h +1 -0
@@ -4621,6 +4621,7 @@ struct qla_hw_data {
 	struct workqueue_struct *wq;
 	struct work_struct heartbeat_work;
 	struct qlfc_fw fw_buf;
+	unsigned long last_heartbeat_run_jiffies;
 
 	/* FCP_CMND priority support */
 	struct qla_fcp_prio_cfg *fcp_prio_cfg;
drivers/scsi/qla2xxx/qla_os.c +17 -3
@@ -7209,7 +7209,7 @@ skip:
 	return do_heartbeat;
 }
 
-static void qla_heart_beat(struct scsi_qla_host *vha)
+static void qla_heart_beat(struct scsi_qla_host *vha, u16 dpc_started)
 {
 	struct qla_hw_data *ha = vha->hw;
 
@@ -7219,8 +7219,19 @@ static void qla_heart_beat(struct scsi_q
 	if (vha->hw->flags.eeh_busy || qla2x00_chip_is_down(vha))
 		return;
 
-	if (qla_do_heartbeat(vha))
+	/*
+	 * dpc thread cannot run if heartbeat is running at the same time.
+	 * We also do not want to starve heartbeat task. Therefore, do
+	 * heartbeat task at least once every 5 seconds.
+	 */
+	if (dpc_started &&
+	    time_before(jiffies, ha->last_heartbeat_run_jiffies + 5 * HZ))
+		return;
+
+	if (qla_do_heartbeat(vha)) {
+		ha->last_heartbeat_run_jiffies = jiffies;
 		queue_work(ha->wq, &ha->heartbeat_work);
+	}
 }
 
 /**************************************************************************
@@ -7411,6 +7422,8 @@ qla2x00_timer(struct timer_list *t)
 		start_dpc++;
 	}
 
+	/* borrowing w to signify dpc will run */
+	w = 0;
 	/* Schedule the DPC routine if needed */
 	if ((test_bit(ISP_ABORT_NEEDED, &vha->dpc_flags) ||
 	    test_bit(LOOP_RESYNC_NEEDED, &vha->dpc_flags) ||
@@ -7443,9 +7456,10 @@ qla2x00_timer(struct timer_list *t)
 		    test_bit(RELOGIN_NEEDED, &vha->dpc_flags),
 		    test_bit(PROCESS_PUREX_IOCB, &vha->dpc_flags));
 		qla2xxx_wake_dpc(vha);
+		w = 1;
 	}
 
-	qla_heart_beat(vha);
+	qla_heart_beat(vha, w);
 
 	qla2x00_restart_timer(vha, WATCH_INTERVAL);
 }
author maintainer
#992 lore raw

PATCH 988/1126 scsi: qla2xxx: Fix missed DMA unmap for NVMe ls requests

In reply to Greg Kroah-Hartman

From: Arun Easi <[email protected]>

commit c85ab7d9e27a80e48d5b7d7fb2fe2b0fdb2de523 upstream.

At NVMe ELS request time, request structure is DMA mapped and never unmapped. Fix this by calling the unmap on ELS completion.

Fixes e84067d74301 ("scsi: qla2xxx: Add FC-NVMe F/W initialization and transport registration")
---
 drivers/scsi/qla2xxx/qla_nvme.c |   17 +++++++++++++++++
 1 file changed, 17 insertions(+)
drivers/scsi/qla2xxx/qla_nvme.c +17 -0
@@ -175,6 +175,18 @@ out:
 	qla2xxx_rel_qpair_sp(sp->qpair, sp);
 }
 
+static void qla_nvme_ls_unmap(struct srb *sp, struct nvmefc_ls_req *fd)
+{
+	if (sp->flags & SRB_DMA_VALID) {
+		struct srb_iocb *nvme = &sp->u.iocb_cmd;
+		struct qla_hw_data *ha = sp->fcport->vha->hw;
+
+		dma_unmap_single(&ha->pdev->dev, nvme->u.nvme.cmd_dma,
+				 fd->rqstlen, DMA_TO_DEVICE);
+		sp->flags &= ~SRB_DMA_VALID;
+	}
+}
+
 static void qla_nvme_release_ls_cmd_kref(struct kref *kref)
 {
 	struct srb *sp = container_of(kref, struct srb, cmd_kref);
@@ -191,6 +203,8 @@ static void qla_nvme_release_ls_cmd_kref
 	spin_unlock_irqrestore(&priv->cmd_lock, flags);
 
 	fd = priv->fd;
+
+	qla_nvme_ls_unmap(sp, fd);
 	fd->done(fd, priv->comp_status);
 out:
 	qla2x00_rel_sp(sp);
@@ -361,6 +375,8 @@ static int qla_nvme_ls_req(struct nvme_f
 	dma_sync_single_for_device(&ha->pdev->dev, nvme->u.nvme.cmd_dma,
 	    fd->rqstlen, DMA_TO_DEVICE);
 
+	sp->flags |= SRB_DMA_VALID;
+
 	rval = qla2x00_start_sp(sp);
 	if (rval != QLA_SUCCESS) {
 		ql_log(ql_log_warn, vha, 0x700e,
@@ -368,6 +384,7 @@ static int qla_nvme_ls_req(struct nvme_f
 		wake_up(&sp->nvme_ls_waitq);
 		sp->priv = NULL;
 		priv->sp = NULL;
+		qla_nvme_ls_unmap(sp, fd);
 		qla2x00_rel_sp(sp);
 		return rval;
 	}
author maintainer
#993 lore raw

PATCH 989/1126 scsi: qla2xxx: Fix crash during module load unload test

In reply to Greg Kroah-Hartman

From: Arun Easi <[email protected]>

commit 0972252450f90db56dd5415a20e2aec21a08d036 upstream.

During purex packet handling the driver was incorrectly freeing a pre-allocated structure. Fix this by skipping that entry.

System crashed with the following stack during a module unload test.

Call Trace:
	sbitmap_init_node+0x7f/0x1e0
	sbitmap_queue_init_node+0x24/0x150
	blk_mq_init_bitmaps+0x3d/0xa0
	blk_mq_init_tags+0x68/0x90
	blk_mq_alloc_map_and_rqs+0x44/0x120
	blk_mq_alloc_set_map_and_rqs+0x63/0x150
	blk_mq_alloc_tag_set+0x11b/0x230
	scsi_add_host_with_dma.cold+0x3f/0x245
	qla2x00_probe_one+0xd5a/0x1b80 [qla2xxx]

Call Trace with slub_debug and debug kernel:

	kasan_report_invalid_free+0x50/0x80
	__kasan_slab_free+0x137/0x150
	slab_free_freelist_hook+0xc6/0x190
	kfree+0xe8/0x2e0
	qla2x00_free_device+0x3bb/0x5d0 [qla2xxx]
	qla2x00_remove_one+0x668/0xcf0 [qla2xxx]
---
 drivers/scsi/qla2xxx/qla_os.c |    2 ++
 1 file changed, 2 insertions(+)
drivers/scsi/qla2xxx/qla_os.c +2 -0
@@ -3896,6 +3896,8 @@ qla24xx_free_purex_list(struct purex_lis
 	spin_lock_irqsave(&list->lock, flags);
 	list_for_each_entry_safe(item, next, &list->head, list) {
 		list_del(&item->list);
+		if (item == &item->vha->default_item)
+			continue;
 		kfree(item);
 	}
 	spin_unlock_irqrestore(&list->lock, flags);
author maintainer
#994 lore raw

PATCH 990/1126 scsi: qla2xxx: Fix N2N inconsistent PLOGI

In reply to Greg Kroah-Hartman

From: Quinn Tran <[email protected]>

commit c13ce47c64ea8f14e77eecb40d1e7c2ac667f898 upstream.

For N2N topology, ELS Passthrough is used to send PLOGI. On failure of ELS pass through PLOGI, driver flipped over to using LLIOCB PLOGI for N2N. This is not consistent. Delete the session to restart the connection where ELS pass through PLOGI would be used consistently.

---
 drivers/scsi/qla2xxx/qla_iocb.c |    8 +++-----
 1 file changed, 3 insertions(+), 5 deletions(-)
drivers/scsi/qla2xxx/qla_iocb.c +3 -5
@@ -2943,6 +2943,7 @@ static void qla2x00_els_dcmd2_sp_done(sr
 					set_bit(ISP_ABORT_NEEDED,
 					    &vha->dpc_flags);
 					qla2xxx_wake_dpc(vha);
+					break;
 				}
 				fallthrough;
 			default:
@@ -2952,9 +2953,7 @@ static void qla2x00_els_dcmd2_sp_done(sr
 				    fw_status[0], fw_status[1], fw_status[2]);
 
 				fcport->flags &= ~FCF_ASYNC_SENT;
-				qla2x00_set_fcport_disc_state(fcport,
-				    DSC_LOGIN_FAILED);
-				set_bit(RELOGIN_NEEDED, &vha->dpc_flags);
+				qlt_schedule_sess_for_deletion(fcport);
 				break;
 			}
 			break;
@@ -2966,8 +2965,7 @@ static void qla2x00_els_dcmd2_sp_done(sr
 			    fw_status[0], fw_status[1], fw_status[2]);
 
 			sp->fcport->flags &= ~FCF_ASYNC_SENT;
-			qla2x00_set_fcport_disc_state(fcport, DSC_LOGIN_FAILED);
-			set_bit(RELOGIN_NEEDED, &vha->dpc_flags);
+			qlt_schedule_sess_for_deletion(fcport);
 			break;
 		}
 
author maintainer
#995 lore raw

PATCH 991/1126 scsi: qla2xxx: Fix stuck session of PRLI reject

In reply to Greg Kroah-Hartman

From: Quinn Tran <[email protected]>

commit f3502e2e98a92981601edc3dadf4b0f43c79836b upstream.

Remove stale recovery code that prevents normal path recovery.

---
 drivers/scsi/qla2xxx/qla_init.c |    7 -------
 1 file changed, 7 deletions(-)
drivers/scsi/qla2xxx/qla_init.c +0 -7
@@ -2104,13 +2104,6 @@ qla24xx_handle_prli_done_event(struct sc
 		qla24xx_post_gpdb_work(vha, ea->fcport, 0);
 		break;
 	default:
-		if ((ea->iop[0] == LSC_SCODE_ELS_REJECT) &&
-		    (ea->iop[1] == 0x50000)) {   /* reson 5=busy expl:0x0 */
-			set_bit(RELOGIN_NEEDED, &vha->dpc_flags);
-			ea->fcport->fw_login_state = DSC_LS_PLOGI_COMP;
-			break;
-		}
-
 		sp = ea->sp;
 		ql_dbg(ql_dbg_disc, vha, 0x2118,
 		       "%s %d %8phC priority %s, fc4type %x prev try %s\n",
author maintainer
#996 lore raw

PATCH 992/1126 scsi: qla2xxx: Reduce false trigger to login

In reply to Greg Kroah-Hartman

From: Quinn Tran <[email protected]>

commit d2646eed7b19a206912f49101178cbbaa507256c upstream.

While a session is in the middle of a relogin, a late RSCN can be delivered from switch. RSCN trigger fabric scan where the scan logic can trigger another session login while a login is in progress. Reduce the extra trigger to prevent multiple logins to the same session.

---
 drivers/scsi/qla2xxx/qla_init.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/scsi/qla2xxx/qla_init.c +3 -2
@@ -1644,7 +1644,8 @@ int qla24xx_fcport_handle_login(struct s
 	    fcport->login_gen, fcport->loop_id, fcport->scan_state,
 	    fcport->fc4_type);
 
-	if (fcport->scan_state != QLA_FCPORT_FOUND)
+	if (fcport->scan_state != QLA_FCPORT_FOUND ||
+	    fcport->disc_state == DSC_DELETE_PEND)
 		return 0;
 
 	if ((fcport->loop_id != FC_NO_LOOP_ID) &&
@@ -1665,7 +1666,7 @@ int qla24xx_fcport_handle_login(struct s
 	if (vha->host->active_mode == MODE_TARGET && !N2N_TOPO(vha->hw))
 		return 0;
 
-	if (fcport->flags & FCF_ASYNC_SENT) {
+	if (fcport->flags & (FCF_ASYNC_SENT | FCF_ASYNC_ACTIVE)) {
 		set_bit(RELOGIN_NEEDED, &vha->dpc_flags);
 		return 0;
 	}
author maintainer
#997 lore raw

PATCH 993/1126 scsi: qla2xxx: Use correct feature type field during RFF_ID processing

In reply to Greg Kroah-Hartman

From: Manish Rangankar <[email protected]>

commit a7e05f7a1bcbe4ee055479242de46c5c16ab03b1 upstream.

During SNS Register FC-4 Features (RFF_ID) the initiator driver was sending incorrect type field for NVMe supported device. Use correct feature type field.

Fixes e374f9f59281 ("scsi: qla2xxx: Migrate switch registration commands away from mailbox interface")
---
 drivers/scsi/qla2xxx/qla_gs.c |    5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
drivers/scsi/qla2xxx/qla_gs.c +2 -3
@@ -676,8 +676,7 @@ qla2x00_rff_id(scsi_qla_host_t *vha, u8
 		return (QLA_SUCCESS);
 	}
 
-	return qla_async_rffid(vha, &vha->d_id, qlt_rff_id(vha),
-	    FC4_TYPE_FCP_SCSI);
+	return qla_async_rffid(vha, &vha->d_id, qlt_rff_id(vha), type);
 }
 
 static int qla_async_rffid(scsi_qla_host_t *vha, port_id_t *d_id,
@@ -729,7 +728,7 @@ static int qla_async_rffid(scsi_qla_host
 	/* Prepare CT arguments -- port_id, FC-4 feature, FC-4 type */
 	ct_req->req.rff_id.port_id = port_id_to_be_id(*d_id);
 	ct_req->req.rff_id.fc4_feature = fc4feature;
-	ct_req->req.rff_id.fc4_type = fc4type;		/* SCSI - FCP */
+	ct_req->req.rff_id.fc4_type = fc4type;		/* SCSI-FCP or FC-NVMe */
 
 	sp->u.iocb_cmd.u.ctarg.req_size = RFF_ID_REQ_SIZE;
 	sp->u.iocb_cmd.u.ctarg.rsp_size = RFF_ID_RSP_SIZE;
author maintainer
#998 lore raw

PATCH 994/1126 platform: chrome: Split trace include file

In reply to Greg Kroah-Hartman

From: Gwendal Grignou <[email protected]>

commit eabd9a3807e17e211690e6c40f1405b427b64c48 upstream.

cros_ec_trace.h defined 5 tracing events, 2 for cros_ec_proto and 3 for cros_ec_sensorhub_ring.
These 2 files are in different kernel modules, the traces are defined twice in the kernel which leads to problem enabling only some traces.

Move sensorhub traces from cros_ec_trace.h to cros_ec_sensorhub_trace.h and enable them only in cros_ec_sensorhub kernel module.

Check we can now enable any single traces: without this patch, we can only enable all sensorhub traces or none.

Fixes d453ceb6549a ("platform/chrome: sensorhub: Add trace events for sample")
---
 drivers/platform/chrome/Makefile                  |    3 
 drivers/platform/chrome/cros_ec_sensorhub_ring.c  |    3 
 drivers/platform/chrome/cros_ec_sensorhub_trace.h |  123 ++++++++++++++++++++++
 drivers/platform/chrome/cros_ec_trace.h           |   95 ----------------
 4 files changed, 127 insertions(+), 97 deletions(-)
 create mode 100644 drivers/platform/chrome/cros_ec_sensorhub_trace.h
drivers/platform/chrome/Makefile +2 -1
@@ -2,6 +2,7 @@
 
 # tell define_trace.h where to find the cros ec trace header
 CFLAGS_cros_ec_trace.o:=		-I$(src)
+CFLAGS_cros_ec_sensorhub_ring.o:=	-I$(src)
 
 obj-$(CONFIG_CHROMEOS_LAPTOP)		+= chromeos_laptop.o
 obj-$(CONFIG_CHROMEOS_PSTORE)		+= chromeos_pstore.o
@@ -20,7 +21,7 @@ obj-$(CONFIG_CROS_EC_CHARDEV)		+= cros_e
 obj-$(CONFIG_CROS_EC_LIGHTBAR)		+= cros_ec_lightbar.o
 obj-$(CONFIG_CROS_EC_VBC)		+= cros_ec_vbc.o
 obj-$(CONFIG_CROS_EC_DEBUGFS)		+= cros_ec_debugfs.o
-cros-ec-sensorhub-objs			:= cros_ec_sensorhub.o cros_ec_sensorhub_ring.o cros_ec_trace.o
+cros-ec-sensorhub-objs			:= cros_ec_sensorhub.o cros_ec_sensorhub_ring.o
 obj-$(CONFIG_CROS_EC_SENSORHUB)		+= cros-ec-sensorhub.o
 obj-$(CONFIG_CROS_EC_SYSFS)		+= cros_ec_sysfs.o
 obj-$(CONFIG_CROS_USBPD_LOGGER)		+= cros_usbpd_logger.o
drivers/platform/chrome/cros_ec_sensorhub_ring.c +2 -1
@@ -17,7 +17,8 @@
 #include <linux/sort.h>
 #include <linux/slab.h>
 
-#include "cros_ec_trace.h"
+#define CREATE_TRACE_POINTS
+#include "cros_ec_sensorhub_trace.h"
 
 /* Precision of fixed point for the m values from the filter */
 #define M_PRECISION BIT(23)
drivers/platform/chrome/cros_ec_sensorhub_trace.h new file +123 -0

123 lines not on this page · raw message

drivers/platform/chrome/cros_ec_trace.h +0 -95

107 lines not on this page · raw message

author maintainer
#999 lore raw

PATCH 995/1126 MIPS: crypto: Fix CRC32 code

In reply to Greg Kroah-Hartman

From: Paul Cercueil <[email protected]>

commit 41022eff9c2d21e658c7a6fcd31005bf514d28b7 upstream.

Commit 67512a8cf5a7 ("MIPS: Avoid macro redefinitions") changed how the MIPS register macros were defined, in order to allow the code to compile under LLVM/Clang.

The MIPS CRC32 code however wasn't updated accordingly, causing a build bug when using a MIPS32r6 toolchain without CRC support.

Update the CRC32 code to use the macros correctly, to fix the build failures.

Fixes 67512a8cf5a7 ("MIPS: Avoid macro redefinitions")
---
 arch/mips/crypto/crc32-mips.c |   46 +++++++++++++++++++++++-------------------
 1 file changed, 26 insertions(+), 20 deletions(-)
arch/mips/crypto/crc32-mips.c +26 -20
@@ -28,7 +28,7 @@ enum crc_type {
 };
 
 #ifndef TOOLCHAIN_SUPPORTS_CRC
-#define _ASM_MACRO_CRC32(OP, SZ, TYPE)					  \
+#define _ASM_SET_CRC(OP, SZ, TYPE)					  \
 _ASM_MACRO_3R(OP, rt, rs, rt2,						  \
 	".ifnc	\\rt, \\rt2\n\t"					  \
 	".error	\"invalid operands \\\"" #OP " \\rt,\\rs,\\rt2\\\"\"\n\t" \
@@ -37,30 +37,36 @@ _ASM_MACRO_3R(OP, rt, rs, rt2,						  \
 			  ((SZ) <<  6) | ((TYPE) << 8))			  \
 	_ASM_INSN32_IF_MM(0x00000030 | (__rs << 16) | (__rt << 21) |	  \
 			  ((SZ) << 14) | ((TYPE) << 3)))
-_ASM_MACRO_CRC32(crc32b,  0, 0);
-_ASM_MACRO_CRC32(crc32h,  1, 0);
-_ASM_MACRO_CRC32(crc32w,  2, 0);
-_ASM_MACRO_CRC32(crc32d,  3, 0);
-_ASM_MACRO_CRC32(crc32cb, 0, 1);
-_ASM_MACRO_CRC32(crc32ch, 1, 1);
-_ASM_MACRO_CRC32(crc32cw, 2, 1);
-_ASM_MACRO_CRC32(crc32cd, 3, 1);
-#define _ASM_SET_CRC ""
+#define _ASM_UNSET_CRC(op, SZ, TYPE) ".purgem " #op "\n\t"
 #else /* !TOOLCHAIN_SUPPORTS_CRC */
-#define _ASM_SET_CRC ".set\tcrc\n\t"
+#define _ASM_SET_CRC(op, SZ, TYPE) ".set\tcrc\n\t"
+#define _ASM_UNSET_CRC(op, SZ, TYPE)
 #endif
 
-#define _CRC32(crc, value, size, type)		\
-do {						\
-	__asm__ __volatile__(			\
-		".set	push\n\t"		\
-		_ASM_SET_CRC			\
-		#type #size "	%0, %1, %0\n\t"	\
-		".set	pop"			\
-		: "+r" (crc)			\
-		: "r" (value));			\
+#define __CRC32(crc, value, op, SZ, TYPE)		\
+do {							\
+	__asm__ __volatile__(				\
+		".set	push\n\t"			\
+		_ASM_SET_CRC(op, SZ, TYPE)		\
+		#op "	%0, %1, %0\n\t"			\
+		_ASM_UNSET_CRC(op, SZ, TYPE)		\
+		".set	pop"				\
+		: "+r" (crc)				\
+		: "r" (value));				\
 } while (0)
 
+#define _CRC32_crc32b(crc, value)	__CRC32(crc, value, crc32b, 0, 0)
+#define _CRC32_crc32h(crc, value)	__CRC32(crc, value, crc32h, 1, 0)
+#define _CRC32_crc32w(crc, value)	__CRC32(crc, value, crc32w, 2, 0)
+#define _CRC32_crc32d(crc, value)	__CRC32(crc, value, crc32d, 3, 0)
+#define _CRC32_crc32cb(crc, value)	__CRC32(crc, value, crc32cb, 0, 1)
+#define _CRC32_crc32ch(crc, value)	__CRC32(crc, value, crc32ch, 1, 1)
+#define _CRC32_crc32cw(crc, value)	__CRC32(crc, value, crc32cw, 2, 1)
+#define _CRC32_crc32cd(crc, value)	__CRC32(crc, value, crc32cd, 3, 1)
+
+#define _CRC32(crc, value, size, op) \
+	_CRC32_##op##size(crc, value)
+
 #define CRC32(crc, value, size) \
 	_CRC32(crc, value, size, crc32)
 
author maintainer
#1000 lore raw

PATCH 996/1126 KVM: x86: Check lapic_in_kernel() before attempting to set a SynIC irq

In reply to Greg Kroah-Hartman

From: Vitaly Kuznetsov <[email protected]>

commit 7ec37d1cbe17d8189d9562178d8b29167fe1c31a upstream.

When KVM_CAP_HYPERV_SYNIC{,2} is activated, KVM already checks for irqchip_in_kernel() so normally SynIC irqs should never be set. It is, however, possible for a misbehaving VMM to write to SYNIC/STIMER MSRs causing erroneous behavior.

The immediate issue being fixed is that kvm_irq_delivery_to_apic() (kvm_irq_delivery_to_apic_fast()) crashes when called with 'irq.shorthand = APIC_DEST_SELF' and 'src == NULL'.

Message-Id: <[email protected]>

---
 arch/x86/kvm/hyperv.c |    3 +++
 1 file changed, 3 insertions(+)
arch/x86/kvm/hyperv.c +3 -0
@@ -446,6 +446,9 @@ static int synic_set_irq(struct kvm_vcpu
 	struct kvm_lapic_irq irq;
 	int ret, vector;
 
+	if (KVM_BUG_ON(!lapic_in_kernel(vcpu), vcpu->kvm))
+		return -EINVAL;
+
 	if (sint >= ARRAY_SIZE(synic->sint))
 		return -EINVAL;
 
author maintainer
#1001 lore raw

PATCH 997/1126 KVM: x86: Avoid theoretical NULL pointer dereference in kvm_irq_delivery_to_apic_fast()

In reply to Greg Kroah-Hartman

From: Vitaly Kuznetsov <[email protected]>

commit 00b5f37189d24ac3ed46cb7f11742094778c46ce upstream.

When kvm_irq_delivery_to_apic_fast() is called with APIC_DEST_SELF shorthand, 'src' must not be NULL. Crash the VM with KVM_BUG_ON() instead of crashing the host.

Message-Id: <[email protected]>

---
 arch/x86/kvm/lapic.c |    4 ++++
 1 file changed, 4 insertions(+)
arch/x86/kvm/lapic.c +4 -0
@@ -992,6 +992,10 @@ bool kvm_irq_delivery_to_apic_fast(struc
 	*r = -1;
 
 	if (irq->shorthand == APIC_DEST_SELF) {
+		if (KVM_BUG_ON(!src, kvm)) {
+			*r = 0;
+			return true;
+		}
 		*r = kvm_apic_set_irq(src->vcpu, irq, dest_map);
 		return true;
 	}
author maintainer
#1002 lore raw

PATCH 998/1126 KVM: x86: Forbid VMM to set SYNIC/STIMER MSRs when SynIC wasnt activated

In reply to Greg Kroah-Hartman

From: Vitaly Kuznetsov <[email protected]>

commit b1e34d325397a33d97d845e312d7cf2a8b646b44 upstream.

Setting non-zero values to SYNIC/STIMER MSRs activates certain features, this should not happen when KVM_CAP_HYPERV_SYNIC{,2} was not activated.

Note, it would've been better to forbid writing anything to SYNIC/STIMER MSRs, including zeroes, however, at least QEMU tries clearing HV_X64_MSR_STIMER0_CONFIG without SynIC. HV_X64_MSR_EOM MSR is somewhat 'special' as writing zero there triggers an action, this also should not happen when SynIC wasn't activated.

Message-Id: <[email protected]>

---
 arch/x86/kvm/hyperv.c |    9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)
arch/x86/kvm/hyperv.c +6 -3
@@ -236,7 +236,7 @@ static int synic_set_msr(struct kvm_vcpu
 	struct kvm_vcpu *vcpu = hv_synic_to_vcpu(synic);
 	int ret;
 
-	if (!synic->active && !host)
+	if (!synic->active && (!host || data))
 		return 1;
 
 	trace_kvm_hv_synic_set_msr(vcpu->vcpu_id, msr, data, host);
@@ -282,6 +282,9 @@ static int synic_set_msr(struct kvm_vcpu
 	case HV_X64_MSR_EOM: {
 		int i;
 
+		if (!synic->active)
+			break;
+
 		for (i = 0; i < ARRAY_SIZE(synic->sint); i++)
 			kvm_hv_notify_acked_sint(vcpu, i);
 		break;
@@ -661,7 +664,7 @@ static int stimer_set_config(struct kvm_
 	struct kvm_vcpu_hv *hv_vcpu = to_hv_vcpu(vcpu);
 	struct kvm_vcpu_hv_synic *synic = to_hv_synic(vcpu);
 
-	if (!synic->active && !host)
+	if (!synic->active && (!host || config))
 		return 1;
 
 	if (unlikely(!host && hv_vcpu->enforce_cpuid && new_config.direct_mode &&
@@ -690,7 +693,7 @@ static int stimer_set_count(struct kvm_v
 	struct kvm_vcpu *vcpu = hv_stimer_to_vcpu(stimer);
 	struct kvm_vcpu_hv_synic *synic = to_hv_synic(vcpu);
 
-	if (!synic->active && !host)
+	if (!synic->active && (!host || count))
 		return 1;
 
 	trace_kvm_hv_stimer_set_count(hv_stimer_to_vcpu(stimer)->vcpu_id,
author maintainer
#1003 lore raw

PATCH 999/1126 KVM: x86/mmu: do compare-and-exchange of gPTE via the user address

In reply to Greg Kroah-Hartman

From: Paolo Bonzini <[email protected]>

commit 2a8859f373b0a86f0ece8ec8312607eacf12485d upstream.

FNAME(cmpxchg_gpte) is an inefficient mess. It is at least decent if it can go through get_user_pages_fast(), but if it cannot then it tries to use memremap(); that is not just terribly slow, it is also wrong because it assumes that the VM_PFNMAP VMA is contiguous.

The right way to do it would be to do the same thing as hva_to_pfn_remapped() does since commit add6a0cd1c5b ("KVM: MMU: try to fix up page faults before giving up", 2016-07-05), using follow_pte() and fixup_user_fault() to determine the correct address to use for memremap(). To do this, one could for example extract hva_to_pfn() for use outside virt/kvm/kvm_main.c. But really there is no reason to do that either, because there is already a perfectly valid address to do the cmpxchg() on, only it is a userspace address. That means doing user_access_begin()/user_access_end() and writing the code in assembly to handle exceptions correctly. Worse, the guest PTE can be 8-byte even on i686 so there is the extra complication of using cmpxchg8b to account for. But at least it is an efficient mess.

(Thanks to Linus for suggesting improvement on the inline assembly).

Reported-by [email protected]
Fixes bd53cb35a3e9 ("X86/KVM: Handle PFNs outside of kernel reach when touching GPTEs")
---
 arch/x86/kvm/mmu/paging_tmpl.h |   74 ++++++++++++++++++-----------------------
 1 file changed, 34 insertions(+), 40 deletions(-)
arch/x86/kvm/mmu/paging_tmpl.h +34 -40
@@ -34,9 +34,8 @@
 	#define PT_HAVE_ACCESSED_DIRTY(mmu) true
 	#ifdef CONFIG_X86_64
 	#define PT_MAX_FULL_LEVELS PT64_ROOT_MAX_LEVEL
-	#define CMPXCHG cmpxchg
+	#define CMPXCHG "cmpxchgq"
 	#else
-	#define CMPXCHG cmpxchg64
 	#define PT_MAX_FULL_LEVELS 2
 	#endif
 #elif PTTYPE == 32
@@ -52,7 +51,7 @@
 	#define PT_GUEST_DIRTY_SHIFT PT_DIRTY_SHIFT
 	#define PT_GUEST_ACCESSED_SHIFT PT_ACCESSED_SHIFT
 	#define PT_HAVE_ACCESSED_DIRTY(mmu) true
-	#define CMPXCHG cmpxchg
+	#define CMPXCHG "cmpxchgl"
 #elif PTTYPE == PTTYPE_EPT
 	#define pt_element_t u64
 	#define guest_walker guest_walkerEPT
@@ -65,7 +64,9 @@
 	#define PT_GUEST_DIRTY_SHIFT 9
 	#define PT_GUEST_ACCESSED_SHIFT 8
 	#define PT_HAVE_ACCESSED_DIRTY(mmu) ((mmu)->ept_ad)
-	#define CMPXCHG cmpxchg64
+	#ifdef CONFIG_X86_64
+	#define CMPXCHG "cmpxchgq"
+	#endif
 	#define PT_MAX_FULL_LEVELS PT64_ROOT_MAX_LEVEL
 #else
 	#error Invalid PTTYPE value
@@ -147,43 +148,36 @@ static int FNAME(cmpxchg_gpte)(struct kv
 			       pt_element_t __user *ptep_user, unsigned index,
 			       pt_element_t orig_pte, pt_element_t new_pte)
 {
-	int npages;
-	pt_element_t ret;
-	pt_element_t *table;
-	struct page *page;
-
-	npages = get_user_pages_fast((unsigned long)ptep_user, 1, FOLL_WRITE, &page);
-	if (likely(npages == 1)) {
-		table = kmap_atomic(page);
-		ret = CMPXCHG(&table[index], orig_pte, new_pte);
-		kunmap_atomic(table);
-
-		kvm_release_page_dirty(page);
-	} else {
-		struct vm_area_struct *vma;
-		unsigned long vaddr = (unsigned long)ptep_user & PAGE_MASK;
-		unsigned long pfn;
-		unsigned long paddr;
-
-		mmap_read_lock(current->mm);
-		vma = find_vma_intersection(current->mm, vaddr, vaddr + PAGE_SIZE);
-		if (!vma || !(vma->vm_flags & VM_PFNMAP)) {
-			mmap_read_unlock(current->mm);
-			return -EFAULT;
-		}
-		pfn = ((vaddr - vma->vm_start) >> PAGE_SHIFT) + vma->vm_pgoff;
-		paddr = pfn << PAGE_SHIFT;
-		table = memremap(paddr, PAGE_SIZE, MEMREMAP_WB);
-		if (!table) {
-			mmap_read_unlock(current->mm);
-			return -EFAULT;
-		}
-		ret = CMPXCHG(&table[index], orig_pte, new_pte);
-		memunmap(table);
-		mmap_read_unlock(current->mm);
-	}
+	signed char r;
+
+	if (!user_access_begin(ptep_user, sizeof(pt_element_t)))
+		return -EFAULT;
+
+#ifdef CMPXCHG
+	asm volatile("1:" LOCK_PREFIX CMPXCHG " %[new], %[ptr]\n"
+		     "setnz %b[r]\n"
+		     "2:"
+		     _ASM_EXTABLE_TYPE_REG(1b, 2b, EX_TYPE_EFAULT_REG, %k[r])
+		     : [ptr] "+m" (*ptep_user),
+		       [old] "+a" (orig_pte),
+		       [r] "=q" (r)
+		     : [new] "r" (new_pte)
+		     : "memory");
+#else
+	asm volatile("1:" LOCK_PREFIX "cmpxchg8b %[ptr]\n"
+		     "setnz %b[r]\n"
+		     "2:"
+		     _ASM_EXTABLE_TYPE_REG(1b, 2b, EX_TYPE_EFAULT_REG, %k[r])
+		     : [ptr] "+m" (*ptep_user),
+		       [old] "+A" (orig_pte),
+		       [r] "=q" (r)
+		     : [new_lo] "b" ((u32)new_pte),
+		       [new_hi] "c" ((u32)(new_pte >> 32))
+		     : "memory");
+#endif
 
-	return (ret != orig_pte);
+	user_access_end();
+	return r;
 }
 
 static bool FNAME(prefetch_invalid_gpte)(struct kvm_vcpu *vcpu,
author maintainer
#1004 lore raw

PATCH 1000/1126 KVM: Prevent module exit until all VMs are freed

In reply to Greg Kroah-Hartman

From: David Matlack <[email protected]>

commit 5f6de5cbebee925a612856fce6f9182bb3eee0db upstream.

Tie the lifetime the KVM module to the lifetime of each VM via kvm.users_count. This way anything that grabs a reference to the VM via kvm_get_kvm() cannot accidentally outlive the KVM module.

Prior to this commit, the lifetime of the KVM module was tied to the lifetime of /dev/kvm file descriptors, VM file descriptors, and vCPU file descriptors by their respective file_operations "owner" field. This approach is insufficient because references grabbed via kvm_get_kvm() do not prevent closing any of the aforementioned file descriptors.

This fixes a long standing theoretical bug in KVM that at least affects async page faults. kvm_setup_async_pf() grabs a reference via kvm_get_kvm(), and drops it in an asynchronous work callback. Nothing prevents the VM file descriptor from being closed and the KVM module from being unloaded before this callback runs.

Fixes af585b921e5d ("KVM: Halt vcpu if page it tries to access is swapped out")
Fixes 3d3aab1b973b ("KVM: set owner of cpu and vm file operations")
[ Based on a patch from Ben implemented for Google's kernel. ]

Message-Id: <[email protected]>

---
 virt/kvm/kvm_main.c |   13 +++++++++++++
 1 file changed, 13 insertions(+)
virt/kvm/kvm_main.c +13 -0
@@ -117,6 +117,8 @@ EXPORT_SYMBOL_GPL(kvm_debugfs_dir);
 
 static const struct file_operations stat_fops_per_vm;
 
+static struct file_operations kvm_chardev_ops;
+
 static long kvm_vcpu_ioctl(struct file *file, unsigned int ioctl,
 			   unsigned long arg);
 #ifdef CONFIG_KVM_COMPAT
@@ -1137,6 +1139,16 @@ static struct kvm *kvm_create_vm(unsigne
 	preempt_notifier_inc();
 	kvm_init_pm_notifier(kvm);
 
+	/*
+	 * When the fd passed to this ioctl() is opened it pins the module,
+	 * but try_module_get() also prevents getting a reference if the module
+	 * is in MODULE_STATE_GOING (e.g. if someone ran "rmmod --wait").
+	 */
+	if (!try_module_get(kvm_chardev_ops.owner)) {
+		r = -ENODEV;
+		goto out_err;
+	}
+
 	return kvm;
 
 out_err:
@@ -1226,6 +1238,7 @@ static void kvm_destroy_vm(struct kvm *k
 	preempt_notifier_dec();
 	hardware_disable_all();
 	mmdrop(mm);
+	module_put(kvm_chardev_ops.owner);
 }
 
 void kvm_get_kvm(struct kvm *kvm)
author maintainer
#1005 lore raw

PATCH 1001/1126 KVM: x86: fix sending PV IPI

In reply to Greg Kroah-Hartman

From: Li RongQing <[email protected]>

commit c15e0ae42c8e5a61e9aca8aac920517cf7b3e94e upstream.

If apic_id is less than min, and (max - apic_id) is greater than KVM_IPI_CLUSTER_SIZE, then the third check condition is satisfied but the new apic_id does not fit the bitmask. In this case __send_ipi_mask should send the IPI.

This is mostly theoretical, but it can happen if the apic_ids on three iterations of the loop are for example 1, KVM_IPI_CLUSTER_SIZE, 0.

Fixes aaffcfd1e82 ("KVM: X86: Implement PV IPIs in linux guest")

Message-Id: <[email protected]>

---
 arch/x86/kernel/kvm.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/kernel/kvm.c +1 -1
@@ -517,7 +517,7 @@ static void __send_ipi_mask(const struct
 		} else if (apic_id < min && max - apic_id < KVM_IPI_CLUSTER_SIZE) {
 			ipi_bitmap <<= min - apic_id;
 			min = apic_id;
-		} else if (apic_id < min + KVM_IPI_CLUSTER_SIZE) {
+		} else if (apic_id > min && apic_id < min + KVM_IPI_CLUSTER_SIZE) {
 			max = apic_id < max ? max : apic_id;
 		} else {
 			ret = kvm_hypercall4(KVM_HC_SEND_IPI, (unsigned long)ipi_bitmap,
author maintainer
#1006 lore raw

PATCH 1002/1126 KVM: SVM: fix panic on out-of-bounds guest IRQ

In reply to Greg Kroah-Hartman

From: Yi Wang <[email protected]>

commit a80ced6ea514000d34bf1239d47553de0d1ee89e upstream.

As guest_irq is coming from KVM_IRQFD API call, it may trigger crash in svm_update_pi_irte() due to out-of-bounds:

crash> bt
PID: 22218 TASK: ffff951a6ad74980 CPU: 73 COMMAND: "vcpu8"
#0 [ffffb1ba6707fa40] machine_kexec at ffffffff8565b397 #1 [ffffb1ba6707fa90] __crash_kexec at ffffffff85788a6d #2 [ffffb1ba6707fb58] crash_kexec at ffffffff8578995d #3 [ffffb1ba6707fb70] oops_end at ffffffff85623c0d #4 [ffffb1ba6707fb90] no_context at ffffffff856692c9 #5 [ffffb1ba6707fbf8] exc_page_fault at ffffffff85f95b51 #6 [ffffb1ba6707fc50] asm_exc_page_fault at ffffffff86000ace

    [exception RIP: svm_update_pi_irte+227]
    RIP: ffffffffc0761b53  RSP: ffffb1ba6707fd08  RFLAGS: 00010086
    RAX: ffffb1ba6707fd78  RBX: ffffb1ba66d91000  RCX: 0000000000000001
    RDX: 00003c803f63f1c0  RSI: 000000000000019a  RDI: ffffb1ba66db2ab8
    RBP: 000000000000019a   R8: 0000000000000040   R9: ffff94ca41b82200
    R10: ffffffffffffffcf  R11: 0000000000000001  R12: 0000000000000001
    R13: 0000000000000001  R14: ffffffffffffffcf  R15: 000000000000005f
    ORIG_RAX: ffffffffffffffff  CS: 0010  SS: 0018
 #7 [ffffb1ba6707fdb8] kvm_irq_routing_update at ffffffffc09f19a1 [kvm]
 #8 [ffffb1ba6707fde0] kvm_set_irq_routing at ffffffffc09f2133 [kvm]
 #9 [ffffb1ba6707fe18] kvm_vm_ioctl at ffffffffc09ef544 [kvm]
    RIP: 00007f143c36488b  RSP: 00007f143a4e04b8  RFLAGS: 00000246
    RAX: ffffffffffffffda  RBX: 00007f05780041d0  RCX: 00007f143c36488b
    RDX: 00007f05780041d0  RSI: 000000004008ae6a  RDI: 0000000000000020
    RBP: 00000000000004e8   R8: 0000000000000008   R9: 00007f05780041e0
    R10: 00007f0578004560  R11: 0000000000000246  R12: 00000000000004e0
    R13: 000000000000001a  R14: 00007f1424001c60  R15: 00007f0578003bc0
    ORIG_RAX: 0000000000000010  CS: 0033  SS: 002b

Vmx have been fix this in commit 3a8b0677fc61 (KVM: VMX: Do not BUG() on out-of-bounds guest IRQ), so we can just copy source from that to fix this.

Message-Id: <[email protected]>

---
 arch/x86/kvm/svm/avic.c |   10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)
arch/x86/kvm/svm/avic.c +8 -2
@@ -783,7 +783,7 @@ int svm_update_pi_irte(struct kvm *kvm,
 {
 	struct kvm_kernel_irq_routing_entry *e;
 	struct kvm_irq_routing_table *irq_rt;
-	int idx, ret = -EINVAL;
+	int idx, ret = 0;
 
 	if (!kvm_arch_has_assigned_device(kvm) ||
 	    !irq_remapping_cap(IRQ_POSTING_CAP))
@@ -794,7 +794,13 @@ int svm_update_pi_irte(struct kvm *kvm,
 
 	idx = srcu_read_lock(&kvm->irq_srcu);
 	irq_rt = srcu_dereference(kvm->irq_routing, &kvm->irq_srcu);
-	WARN_ON(guest_irq >= irq_rt->nr_rt_entries);
+
+	if (guest_irq >= irq_rt->nr_rt_entries ||
+		hlist_empty(&irq_rt->map[guest_irq])) {
+		pr_warn_once("no route for guest_irq %u/%u (broken user space?)\n",
+			     guest_irq, irq_rt->nr_rt_entries);
+		goto out;
+	}
 
 	hlist_for_each_entry(e, &irq_rt->map[guest_irq], link) {
 		struct vcpu_data vcpu_info;
author maintainer
#1007 lore raw

PATCH 1003/1126 KVM: avoid double put_page with gfn-to-pfn cache

In reply to Greg Kroah-Hartman

From: David Woodhouse <[email protected]>

commit 79593c086eb95eb2886f36ee6f78a1d6845e1bdf upstream.

If the cache's user host virtual address becomes invalid, there is still a path from kvm_gfn_to_pfn_cache_refresh() where __release_gpc() could release the pfn but the gpc->pfn field has not been overwritten with an error value. If this happens, kvm_gfn_to_pfn_cache_unmap will call put_page again on the same page.

Fixes 982ed0de4753 ("KVM: Reinstate gfn_to_pfn_cache with invalidation support")
---
 virt/kvm/pfncache.c |    1 +
 1 file changed, 1 insertion(+)
virt/kvm/pfncache.c +1 -0
@@ -191,6 +191,7 @@ int kvm_gfn_to_pfn_cache_refresh(struct
 		gpc->uhva = gfn_to_hva_memslot(gpc->memslot, gfn);
 
 		if (kvm_is_error_hva(gpc->uhva)) {
+			gpc->pfn = KVM_PFN_ERR_FAULT;
 			ret = -EFAULT;
 			goto out;
 		}
author maintainer
#1008 lore raw

PATCH 1004/1126 ubifs: rename_whiteout: Fix double free for whiteout_ui->data

In reply to Greg Kroah-Hartman

From: Zhihao Cheng <[email protected]>

commit 40a8f0d5e7b3999f096570edab71c345da812e3e upstream.

'whiteout_ui->data' will be freed twice if space budget fail for rename whiteout operation as following process:

rename_whiteout
dev = kmalloc
whiteout_ui->data = dev
kfree(whiteout_ui->data) // Free first time
iput(whiteout)

    ubifs_free_inode
      kfree(ui->data)	    // Double free!

KASAN reports:
==================================================================

BUG: KASAN: double-free or invalid-free in ubifs_free_inode+0x4f/0x70
Call Trace:
  kfree+0x117/0x490
  ubifs_free_inode+0x4f/0x70 [ubifs]
  i_callback+0x30/0x60
  rcu_do_batch+0x366/0xac0
  __do_softirq+0x133/0x57f

Allocated by task 1506:
  kmem_cache_alloc_trace+0x3c2/0x7a0
  do_rename+0x9b7/0x1150 [ubifs]
  ubifs_rename+0x106/0x1f0 [ubifs]
  do_syscall_64+0x35/0x80

Freed by task 1506:
  kfree+0x117/0x490
  do_rename.cold+0x53/0x8a [ubifs]
  ubifs_rename+0x106/0x1f0 [ubifs]
  do_syscall_64+0x35/0x80

The buggy address belongs to the object at ffff88810238bed8 which

belongs to the cache kmalloc-8 of size 8
==================================================================

Let ubifs_free_inode() free 'whiteout_ui->data'. BTW, delete unused assignment 'whiteout_ui->data_len = 0', process 'ubifs_evict_inode() -> ubifs_jnl_delete_inode() -> ubifs_jnl_write_inode()' doesn't need it (because 'inc_nlink(whiteout)' won't be excuted by 'goto out_release',
and the nlink of whiteout inode is 0).

Fixes 9e0a1fff8db56ea ("ubifs: Implement RENAME_WHITEOUT")
---
 fs/ubifs/dir.c |    2 --
 1 file changed, 2 deletions(-)
fs/ubifs/dir.c +0 -2
@@ -1425,8 +1425,6 @@ static int do_rename(struct inode *old_d
 
 		err = ubifs_budget_space(c, &wht_req);
 		if (err) {
-			kfree(whiteout_ui->data);
-			whiteout_ui->data_len = 0;
 			iput(whiteout);
 			goto out_release;
 		}
author maintainer
#1009 lore raw

PATCH 1005/1126 ubifs: Fix deadlock in concurrent rename whiteout and inode writeback

In reply to Greg Kroah-Hartman

From: Zhihao Cheng <[email protected]>

commit afd427048047e8efdedab30e8888044e2be5aa9c upstream.

Following hung tasks:

[   77.028764] task:kworker/u8:4    state:D stack:    0 pid:  132
[   77.028820] Call Trace:
[   77.029027]  schedule+0x8c/0x1b0
[   77.029067]  mutex_lock+0x50/0x60
[   77.029074]  ubifs_write_inode+0x68/0x1f0 [ubifs]
[   77.029117]  __writeback_single_inode+0x43c/0x570
[   77.029128]  writeback_sb_inodes+0x259/0x740
[   77.029148]  wb_writeback+0x107/0x4d0
[   77.029163]  wb_workfn+0x162/0x7b0
[   92.390442] task:aa              state:D stack:    0 pid: 1506
[   92.390448] Call Trace:
[   92.390458]  schedule+0x8c/0x1b0
[   92.390461]  wb_wait_for_completion+0x82/0xd0
[   92.390469]  __writeback_inodes_sb_nr+0xb2/0x110
[   92.390472]  writeback_inodes_sb_nr+0x14/0x20
[   92.390476]  ubifs_budget_space+0x705/0xdd0 [ubifs]
[   92.390503]  do_rename.cold+0x7f/0x187 [ubifs]
[   92.390549]  ubifs_rename+0x8b/0x180 [ubifs]
[   92.390571]  vfs_rename+0xdb2/0x1170
[   92.390580]  do_renameat2+0x554/0x770

, are caused by concurrent rename whiteout and inode writeback processes:
rename_whiteout(Thread 1) wb_workfn(Thread2)
ubifs_rename
do_rename
lock_4_inodes (Hold ui_mutex)
ubifs_budget_space
make_free_space
shrink_liability
__writeback_inodes_sb_nr
bdi_split_work_to_wbs (Queue new wb work)
wb_do_writeback(wb work)
__writeback_single_inode
ubifs_write_inode
LOCK(ui_mutex)
↑
wb_wait_for_completion (Wait wb work) <-- deadlock!

Reproducer (Detail program in [Link]):
1. SYS_renameat2("/mp/dir/file", "/mp/dir/whiteout", RENAME_WHITEOUT)
2. Consume out of space before kernel(mdelay) doing budget for whiteout

Fix it by doing whiteout space budget before locking ubifs inodes. BTW, it also fixes wrong goto tag 'out_release' in whiteout budget error handling path(It should at least recover dir i_size and unlock 4 ubifs inodes).

---
 fs/ubifs/dir.c |   25 +++++++++++++++----------
 1 file changed, 15 insertions(+), 10 deletions(-)
fs/ubifs/dir.c +15 -10
@@ -1324,6 +1324,7 @@ static int do_rename(struct inode *old_d
 
 	if (flags & RENAME_WHITEOUT) {
 		union ubifs_dev_desc *dev = NULL;
+		struct ubifs_budget_req wht_req;
 
 		dev = kmalloc(sizeof(union ubifs_dev_desc), GFP_NOFS);
 		if (!dev) {
@@ -1345,6 +1346,20 @@ static int do_rename(struct inode *old_d
 		whiteout_ui->data = dev;
 		whiteout_ui->data_len = ubifs_encode_dev(dev, MKDEV(0, 0));
 		ubifs_assert(c, !whiteout_ui->dirty);
+
+		memset(&wht_req, 0, sizeof(struct ubifs_budget_req));
+		wht_req.dirtied_ino = 1;
+		wht_req.dirtied_ino_d = ALIGN(whiteout_ui->data_len, 8);
+		/*
+		 * To avoid deadlock between space budget (holds ui_mutex and
+		 * waits wb work) and writeback work(waits ui_mutex), do space
+		 * budget before ubifs inodes locked.
+		 */
+		err = ubifs_budget_space(c, &wht_req);
+		if (err) {
+			iput(whiteout);
+			goto out_release;
+		}
 	}
 
 	lock_4_inodes(old_dir, new_dir, new_inode, whiteout);
@@ -1419,16 +1434,6 @@ static int do_rename(struct inode *old_d
 	}
 
 	if (whiteout) {
-		struct ubifs_budget_req wht_req = { .dirtied_ino = 1,
-				.dirtied_ino_d = \
-				ALIGN(ubifs_inode(whiteout)->data_len, 8) };
-
-		err = ubifs_budget_space(c, &wht_req);
-		if (err) {
-			iput(whiteout);
-			goto out_release;
-		}
-
 		inc_nlink(whiteout);
 		mark_inode_dirty(whiteout);
 
author maintainer
#1010 lore raw

PATCH 1006/1126 ubifs: Add missing iput if do_tmpfile() failed in rename whiteout

In reply to Greg Kroah-Hartman

From: Zhihao Cheng <[email protected]>

commit 716b4573026bcbfa7b58ed19fe15554bac66b082 upstream.

whiteout inode should be put when do_tmpfile() failed if inode has been initialized. Otherwise we will get following warning during umount:
UBIFS error (ubi0:0 pid 1494): ubifs_assert_failed [ubifs]: UBIFS assert failed: c->bi.dd_growth == 0, in fs/ubifs/super.c:1930
VFS: Busy inodes after unmount of ubifs. Self-destruct in 5 seconds.

Fixes 9e0a1fff8db56ea ("ubifs: Implement RENAME_WHITEOUT")
---
 fs/ubifs/dir.c |    2 ++
 1 file changed, 2 insertions(+)
fs/ubifs/dir.c +2 -0
@@ -432,6 +432,8 @@ out_inode:
 	make_bad_inode(inode);
 	if (!instantiated)
 		iput(inode);
+	else if (whiteout)
+		iput(*whiteout);
 out_budg:
 	ubifs_release_budget(c, &req);
 	if (!instantiated)
author maintainer
#1011 lore raw

PATCH 1007/1126 ubifs: Rename whiteout atomically

In reply to Greg Kroah-Hartman

From: Zhihao Cheng <[email protected]>

commit 278d9a243635f26c05ad95dcf9c5a593b9e04dc6 upstream.

Currently, rename whiteout has 3 steps:
1. create tmpfile(which associates old dentry to tmpfile inode) for whiteout, and store tmpfile to disk
2. link whiteout, associate whiteout inode to old dentry agagin and store old dentry, old inode, new dentry on disk
3. writeback dirty whiteout inode to disk

Suddenly power-cut or error occurring(eg. ENOSPC returned by budget, memory allocation failure) during above steps may cause kinds of problems:
Problem 1: ENOSPC returned by whiteout space budget (before step 2),
old dentry will disappear after rename syscall, whiteout file cannot be found either.

ls dir // we get file, whiteout
rename(dir/file, dir/whiteout, REANME_WHITEOUT) ENOSPC = ubifs_budget_space(&wht_req) // return
ls dir // empty (no file, no whiteout)
Problem 2: Power-cut happens before step 3, whiteout inode with 'nlink=1'
is not stored on disk, whiteout dentry(old dentry) is written on disk, whiteout file is lost on next mount (We get "dead directory entry" after executing 'ls -l' on whiteout file).

Now, we use following 3 steps to finish rename whiteout:
1. create an in-mem inode with 'nlink = 1' as whiteout
2. ubifs_jnl_rename (Write on disk to finish associating old dentry to whiteout inode, associating new dentry with old inode)
3. iput(whiteout)

Rely writing in-mem inode on disk by ubifs_jnl_rename() to finish rename whiteout, which avoids middle disk state caused by suddenly power-cut and error occurring.

Fixes 9e0a1fff8db56ea ("ubifs: Implement RENAME_WHITEOUT")
---
 fs/ubifs/dir.c     |  144 +++++++++++++++++++++++++++++++++--------------------
 fs/ubifs/journal.c |   52 ++++++++++++++++---
 2 files changed, 136 insertions(+), 60 deletions(-)
fs/ubifs/dir.c +92 -52
@@ -349,8 +349,56 @@ out_budg:
 	return err;
 }
 
-static int do_tmpfile(struct inode *dir, struct dentry *dentry,
-		      umode_t mode, struct inode **whiteout)
+static struct inode *create_whiteout(struct inode *dir, struct dentry *dentry)
+{
+	int err;
+	umode_t mode = S_IFCHR | WHITEOUT_MODE;
+	struct inode *inode;
+	struct ubifs_info *c = dir->i_sb->s_fs_info;
+	struct fscrypt_name nm;
+
+	/*
+	 * Create an inode('nlink = 1') for whiteout without updating journal,
+	 * let ubifs_jnl_rename() store it on flash to complete rename whiteout
+	 * atomically.
+	 */
+
+	dbg_gen("dent '%pd', mode %#hx in dir ino %lu",
+		dentry, mode, dir->i_ino);
+
+	err = fscrypt_setup_filename(dir, &dentry->d_name, 0, &nm);
+	if (err)
+		return ERR_PTR(err);
+
+	inode = ubifs_new_inode(c, dir, mode);
+	if (IS_ERR(inode)) {
+		err = PTR_ERR(inode);
+		goto out_free;
+	}
+
+	init_special_inode(inode, inode->i_mode, WHITEOUT_DEV);
+	ubifs_assert(c, inode->i_op == &ubifs_file_inode_operations);
+
+	err = ubifs_init_security(dir, inode, &dentry->d_name);
+	if (err)
+		goto out_inode;
+
+	/* The dir size is updated by do_rename. */
+	insert_inode_hash(inode);
+
+	return inode;
+
+out_inode:
+	make_bad_inode(inode);
+	iput(inode);
+out_free:
+	fscrypt_free_filename(&nm);
+	ubifs_err(c, "cannot create whiteout file, error %d", err);
+	return ERR_PTR(err);
+}
+
+static int ubifs_tmpfile(struct user_namespace *mnt_userns, struct inode *dir,
+			 struct dentry *dentry, umode_t mode)
 {
 	struct inode *inode;
 	struct ubifs_info *c = dir->i_sb->s_fs_info;
@@ -392,25 +440,13 @@ static int do_tmpfile(struct inode *dir,
 	}
 	ui = ubifs_inode(inode);
 
-	if (whiteout) {
-		init_special_inode(inode, inode->i_mode, WHITEOUT_DEV);
-		ubifs_assert(c, inode->i_op == &ubifs_file_inode_operations);
-	}
-
 	err = ubifs_init_security(dir, inode, &dentry->d_name);
 	if (err)
 		goto out_inode;
 
 	mutex_lock(&ui->ui_mutex);
 	insert_inode_hash(inode);
-
-	if (whiteout) {
-		mark_inode_dirty(inode);
-		drop_nlink(inode);
-		*whiteout = inode;
-	} else {
-		d_tmpfile(dentry, inode);
-	}
+	d_tmpfile(dentry, inode);
 	ubifs_assert(c, ui->dirty);
 
 	instantiated = 1;
@@ -432,8 +468,6 @@ out_inode:
 	make_bad_inode(inode);
 	if (!instantiated)
 		iput(inode);
-	else if (whiteout)
-		iput(*whiteout);
 out_budg:
 	ubifs_release_budget(c, &req);
 	if (!instantiated)
@@ -443,12 +477,6 @@ out_budg:
 	return err;
 }
 
-static int ubifs_tmpfile(struct user_namespace *mnt_userns, struct inode *dir,
-			 struct dentry *dentry, umode_t mode)
-{
-	return do_tmpfile(dir, dentry, mode, NULL);
-}
-
 /**
  * vfs_dent_type - get VFS directory entry type.
  * @type: UBIFS directory entry type
@@ -1266,17 +1294,19 @@ static int do_rename(struct inode *old_d
 					.dirtied_ino = 3 };
 	struct ubifs_budget_req ino_req = { .dirtied_ino = 1,
 			.dirtied_ino_d = ALIGN(old_inode_ui->data_len, 8) };
+	struct ubifs_budget_req wht_req;
 	struct timespec64 time;
 	unsigned int saved_nlink;
 	struct fscrypt_name old_nm, new_nm;
 
 	/*
-	 * Budget request settings: deletion direntry, new direntry, removing
-	 * the old inode, and changing old and new parent directory inodes.
+	 * Budget request settings:
+	 *   req: deletion direntry, new direntry, removing the old inode,
+	 *   and changing old and new parent directory inodes.
+	 *
+	 *   wht_req: new whiteout inode for RENAME_WHITEOUT.
 	 *
-	 * However, this operation also marks the target inode as dirty and
-	 * does not write it, so we allocate budget for the target inode
-	 * separately.
+	 *   ino_req: marks the target inode as dirty and does not write it.
 	 */
 
 	dbg_gen("dent '%pd' ino %lu in dir ino %lu to dent '%pd' in dir ino %lu flags 0x%x",
@@ -1326,7 +1356,6 @@ static int do_rename(struct inode *old_d
 
 	if (flags & RENAME_WHITEOUT) {
 		union ubifs_dev_desc *dev = NULL;
-		struct ubifs_budget_req wht_req;
 
 		dev = kmalloc(sizeof(union ubifs_dev_desc), GFP_NOFS);
 		if (!dev) {
@@ -1334,24 +1363,26 @@ static int do_rename(struct inode *old_d
 			goto out_release;
 		}
 
-		err = do_tmpfile(old_dir, old_dentry, S_IFCHR | WHITEOUT_MODE, &whiteout);
-		if (err) {
+		/*
+		 * The whiteout inode without dentry is pinned in memory,
+		 * umount won't happen during rename process because we
+		 * got parent dentry.
+		 */
+		whiteout = create_whiteout(old_dir, old_dentry);
+		if (IS_ERR(whiteout)) {
+			err = PTR_ERR(whiteout);
 			kfree(dev);
 			goto out_release;
 		}
 
-		spin_lock(&whiteout->i_lock);
-		whiteout->i_state |= I_LINKABLE;
-		spin_unlock(&whiteout->i_lock);
-
 		whiteout_ui = ubifs_inode(whiteout);
 		whiteout_ui->data = dev;
 		whiteout_ui->data_len = ubifs_encode_dev(dev, MKDEV(0, 0));
 		ubifs_assert(c, !whiteout_ui->dirty);
 
 		memset(&wht_req, 0, sizeof(struct ubifs_budget_req));
-		wht_req.dirtied_ino = 1;
-		wht_req.dirtied_ino_d = ALIGN(whiteout_ui->data_len, 8);
+		wht_req.new_ino = 1;
+		wht_req.new_ino_d = ALIGN(whiteout_ui->data_len, 8);
 		/*
 		 * To avoid deadlock between space budget (holds ui_mutex and
 		 * waits wb work) and writeback work(waits ui_mutex), do space
@@ -1359,6 +1390,11 @@ static int do_rename(struct inode *old_d
 		 */
 		err = ubifs_budget_space(c, &wht_req);
 		if (err) {
+			/*
+			 * Whiteout inode can not be written on flash by
+			 * ubifs_jnl_write_inode(), because it's neither
+			 * dirty nor zero-nlink.
+			 */
 			iput(whiteout);
 			goto out_release;
 		}
@@ -1433,17 +1469,11 @@ static int do_rename(struct inode *old_d
 		sync = IS_DIRSYNC(old_dir) || IS_DIRSYNC(new_dir);
 		if (unlink && IS_SYNC(new_inode))
 			sync = 1;
-	}
-
-	if (whiteout) {
-		inc_nlink(whiteout);
-		mark_inode_dirty(whiteout);
-
-		spin_lock(&whiteout->i_lock);
-		whiteout->i_state &= ~I_LINKABLE;
-		spin_unlock(&whiteout->i_lock);
-
-		iput(whiteout);
+		/*
+		 * S_SYNC flag of whiteout inherits from the old_dir, and we
+		 * have already checked the old dir inode. So there is no need
+		 * to check whiteout.
+		 */
 	}
 
 	err = ubifs_jnl_rename(c, old_dir, old_inode, &old_nm, new_dir,
@@ -1454,6 +1484,11 @@ static int do_rename(struct inode *old_d
 	unlock_4_inodes(old_dir, new_dir, new_inode, whiteout);
 	ubifs_release_budget(c, &req);
 
+	if (whiteout) {
+		ubifs_release_budget(c, &wht_req);
+		iput(whiteout);
+	}
+
 	mutex_lock(&old_inode_ui->ui_mutex);
 	release = old_inode_ui->dirty;
 	mark_inode_dirty_sync(old_inode);
@@ -1462,11 +1497,16 @@ static int do_rename(struct inode *old_d
 	if (release)
 		ubifs_release_budget(c, &ino_req);
 	if (IS_SYNC(old_inode))
-		err = old_inode->i_sb->s_op->write_inode(old_inode, NULL);
+		/*
+		 * Rename finished here. Although old inode cannot be updated
+		 * on flash, old ctime is not a big problem, don't return err
+		 * code to userspace.
+		 */
+		old_inode->i_sb->s_op->write_inode(old_inode, NULL);
 
 	fscrypt_free_filename(&old_nm);
 	fscrypt_free_filename(&new_nm);
-	return err;
+	return 0;
 
 out_cancel:
 	if (unlink) {
@@ -1487,11 +1527,11 @@ out_cancel:
 				inc_nlink(old_dir);
 		}
 	}
+	unlock_4_inodes(old_dir, new_dir, new_inode, whiteout);
 	if (whiteout) {
-		drop_nlink(whiteout);
+		ubifs_release_budget(c, &wht_req);
 		iput(whiteout);
 	}
-	unlock_4_inodes(old_dir, new_dir, new_inode, whiteout);
 out_release:
 	ubifs_release_budget(c, &ino_req);
 	ubifs_release_budget(c, &req);
fs/ubifs/journal.c +44 -8

108 lines not on this page · raw message

author maintainer
#1012 lore raw

PATCH 1008/1126 ubifs: Fix ui->dirty race between do_tmpfile() and writeback work

In reply to Greg Kroah-Hartman

From: Zhihao Cheng <[email protected]>

commit 60eb3b9c9f11206996f57cb89521824304b305ad upstream.

'ui->dirty' is not protected by 'ui_mutex' in function do_tmpfile() which may race with ubifs_write_inode[wb_workfn] to access/update 'ui->dirty', finally dirty space is released twice.

open(O_TMPFILE) wb_workfn
do_tmpfile
ubifs_budget_space(ino_req = { .dirtied_ino = 1}) d_tmpfile // mark inode(tmpfile) dirty ubifs_jnl_update // without holding tmpfile's ui_mutex

    mark_inode_clean(ui)
      if (ui->dirty)
        ubifs_release_dirty_inode_budget(ui)  // release first time
                                   ubifs_write_inode
				     mutex_lock(&ui->ui_mutex)
                                     ubifs_release_dirty_inode_budget(ui)
				     // release second time
				     mutex_unlock(&ui->ui_mutex)
      ui->dirty = 0

Run generic/476 can reproduce following message easily (See reproducer in [Link]):

UBIFS error (ubi0:0 pid 2578): ubifs_assert_failed [ubifs]: UBIFS assert
failed: c->bi.dd_growth >= 0, in fs/ubifs/budget.c:554
UBIFS warning (ubi0:0 pid 2578): ubifs_ro_mode [ubifs]: switched to read-only mode, error -22

  Workqueue: writeback wb_workfn (flush-ubifs_0_0)
  Call Trace:
    ubifs_ro_mode+0x54/0x60 [ubifs]
    ubifs_assert_failed+0x4b/0x80 [ubifs]
    ubifs_release_budget+0x468/0x5a0 [ubifs]
    ubifs_release_dirty_inode_budget+0x53/0x80 [ubifs]
    ubifs_write_inode+0x121/0x1f0 [ubifs]
    ...
    wb_workfn+0x283/0x7b0

Fix it by holding tmpfile ubifs inode lock during ubifs_jnl_update(). Similar problem exists in whiteout renaming, but previous fix("ubifs:
Rename whiteout atomically") has solved the problem.

---
 fs/ubifs/dir.c |   60 ++++++++++++++++++++++++++++-----------------------------
 1 file changed, 30 insertions(+), 30 deletions(-)
fs/ubifs/dir.c +30 -30
@@ -397,6 +397,32 @@ out_free:
 	return ERR_PTR(err);
 }
 
+/**
+ * lock_2_inodes - a wrapper for locking two UBIFS inodes.
+ * @inode1: first inode
+ * @inode2: second inode
+ *
+ * We do not implement any tricks to guarantee strict lock ordering, because
+ * VFS has already done it for us on the @i_mutex. So this is just a simple
+ * wrapper function.
+ */
+static void lock_2_inodes(struct inode *inode1, struct inode *inode2)
+{
+	mutex_lock_nested(&ubifs_inode(inode1)->ui_mutex, WB_MUTEX_1);
+	mutex_lock_nested(&ubifs_inode(inode2)->ui_mutex, WB_MUTEX_2);
+}
+
+/**
+ * unlock_2_inodes - a wrapper for unlocking two UBIFS inodes.
+ * @inode1: first inode
+ * @inode2: second inode
+ */
+static void unlock_2_inodes(struct inode *inode1, struct inode *inode2)
+{
+	mutex_unlock(&ubifs_inode(inode2)->ui_mutex);
+	mutex_unlock(&ubifs_inode(inode1)->ui_mutex);
+}
+
 static int ubifs_tmpfile(struct user_namespace *mnt_userns, struct inode *dir,
 			 struct dentry *dentry, umode_t mode)
 {
@@ -404,7 +430,7 @@ static int ubifs_tmpfile(struct user_nam
 	struct ubifs_info *c = dir->i_sb->s_fs_info;
 	struct ubifs_budget_req req = { .new_ino = 1, .new_dent = 1};
 	struct ubifs_budget_req ino_req = { .dirtied_ino = 1 };
-	struct ubifs_inode *ui, *dir_ui = ubifs_inode(dir);
+	struct ubifs_inode *ui;
 	int err, instantiated = 0;
 	struct fscrypt_name nm;
 
@@ -452,18 +478,18 @@ static int ubifs_tmpfile(struct user_nam
 	instantiated = 1;
 	mutex_unlock(&ui->ui_mutex);
 
-	mutex_lock(&dir_ui->ui_mutex);
+	lock_2_inodes(dir, inode);
 	err = ubifs_jnl_update(c, dir, &nm, inode, 1, 0);
 	if (err)
 		goto out_cancel;
-	mutex_unlock(&dir_ui->ui_mutex);
+	unlock_2_inodes(dir, inode);
 
 	ubifs_release_budget(c, &req);
 
 	return 0;
 
 out_cancel:
-	mutex_unlock(&dir_ui->ui_mutex);
+	unlock_2_inodes(dir, inode);
 out_inode:
 	make_bad_inode(inode);
 	if (!instantiated)
@@ -690,32 +716,6 @@ static int ubifs_dir_release(struct inod
 	return 0;
 }
 
-/**
- * lock_2_inodes - a wrapper for locking two UBIFS inodes.
- * @inode1: first inode
- * @inode2: second inode
- *
- * We do not implement any tricks to guarantee strict lock ordering, because
- * VFS has already done it for us on the @i_mutex. So this is just a simple
- * wrapper function.
- */
-static void lock_2_inodes(struct inode *inode1, struct inode *inode2)
-{
-	mutex_lock_nested(&ubifs_inode(inode1)->ui_mutex, WB_MUTEX_1);
-	mutex_lock_nested(&ubifs_inode(inode2)->ui_mutex, WB_MUTEX_2);
-}
-
-/**
- * unlock_2_inodes - a wrapper for unlocking two UBIFS inodes.
- * @inode1: first inode
- * @inode2: second inode
- */
-static void unlock_2_inodes(struct inode *inode1, struct inode *inode2)
-{
-	mutex_unlock(&ubifs_inode(inode2)->ui_mutex);
-	mutex_unlock(&ubifs_inode(inode1)->ui_mutex);
-}
-
 static int ubifs_link(struct dentry *old_dentry, struct inode *dir,
 		      struct dentry *dentry)
 {
author maintainer
#1013 lore raw

PATCH 1009/1126 ubifs: Rectify space amount budget for mkdir/tmpfile operations

In reply to Greg Kroah-Hartman

From: Zhihao Cheng <[email protected]>

commit a6dab6607d4681d227905d5198710b575dbdb519 upstream.

UBIFS should make sure the flash has enough space to store dirty (Data that is newer than disk) data (in memory), space budget is exactly designed to do that. If space budget calculates less data than we need, 'make_reservation()' will do more work(return -ENOSPC if no free space lelf, sometimes we can see "cannot reserve xxx bytes in jhead xxx, error -28" in ubifs error messages) with ubifs inodes locked, which may effect other syscalls.

A simple way to decide how much space do we need when make a budget:
See how much space is needed by 'make_reservation()' in ubifs_jnl_xxx() function according to corresponding operation.

It's better to report ENOSPC in ubifs_budget_space(), as early as we can.

Fixes 474b93704f32163 ("ubifs: Implement O_TMPFILE")
Fixes 1e51764a3c2ac05 ("UBIFS: add new flash file system")
---
 fs/ubifs/dir.c |   12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)
fs/ubifs/dir.c +8 -4
@@ -428,15 +428,18 @@ static int ubifs_tmpfile(struct user_nam
 {
 	struct inode *inode;
 	struct ubifs_info *c = dir->i_sb->s_fs_info;
-	struct ubifs_budget_req req = { .new_ino = 1, .new_dent = 1};
+	struct ubifs_budget_req req = { .new_ino = 1, .new_dent = 1,
+					.dirtied_ino = 1};
 	struct ubifs_budget_req ino_req = { .dirtied_ino = 1 };
 	struct ubifs_inode *ui;
 	int err, instantiated = 0;
 	struct fscrypt_name nm;
 
 	/*
-	 * Budget request settings: new dirty inode, new direntry,
-	 * budget for dirtied inode will be released via writeback.
+	 * Budget request settings: new inode, new direntry, changing the
+	 * parent directory inode.
+	 * Allocate budget separately for new dirtied inode, the budget will
+	 * be released via writeback.
 	 */
 
 	dbg_gen("dent '%pd', mode %#hx in dir ino %lu",
@@ -979,7 +982,8 @@ static int ubifs_mkdir(struct user_names
 	struct ubifs_inode *dir_ui = ubifs_inode(dir);
 	struct ubifs_info *c = dir->i_sb->s_fs_info;
 	int err, sz_change;
-	struct ubifs_budget_req req = { .new_ino = 1, .new_dent = 1 };
+	struct ubifs_budget_req req = { .new_ino = 1, .new_dent = 1,
+					.dirtied_ino = 1};
 	struct fscrypt_name nm;
 
 	/*
author maintainer
#1014 lore raw

PATCH 1010/1126 ubifs: setflags: Make dirtied_ino_d 8 bytes aligned

In reply to Greg Kroah-Hartman

From: Zhihao Cheng <[email protected]>

commit 1b83ec057db16b4d0697dc21ef7a9743b6041f72 upstream.

Make 'ui->data_len' aligned with 8 bytes before it is assigned to dirtied_ino_d. Since 8871d84c8f8b0c6b("ubifs: convert to fileattr") applied, 'setflags()' only affects regular files and directories, only xattr inode, symlink inode and special inode(pipe/char_dev/block_dev) have none- zero 'ui->data_len' field, so assertion '!(req->dirtied_ino_d & 7)' cannot fail in ubifs_budget_space(). To avoid assertion fails in future evolution(eg. setflags can operate special inodes), it's better to make dirtied_ino_d 8 bytes aligned, after all aligned size is still zero for regular files.

Fixes 1e51764a3c2ac05a ("UBIFS: add new flash file system")
---
 fs/ubifs/ioctl.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/ubifs/ioctl.c +1 -1
@@ -108,7 +108,7 @@ static int setflags(struct inode *inode,
 	struct ubifs_inode *ui = ubifs_inode(inode);
 	struct ubifs_info *c = inode->i_sb->s_fs_info;
 	struct ubifs_budget_req req = { .dirtied_ino = 1,
-					.dirtied_ino_d = ui->data_len };
+			.dirtied_ino_d = ALIGN(ui->data_len, 8) };
 
 	err = ubifs_budget_space(c, &req);
 	if (err)
author maintainer
#1015 lore raw

PATCH 1011/1126 ubifs: Fix read out-of-bounds in ubifs_wbuf_write_nolock()

In reply to Greg Kroah-Hartman

From: Zhihao Cheng <[email protected]>

commit 4f2262a334641e05f645364d5ade1f565c85f20b upstream.

Function ubifs_wbuf_write_nolock() may access buf out of bounds in following process:

ubifs_wbuf_write_nolock():
aligned_len = ALIGN(len, 8); // Assume len = 4089, aligned_len = 4096
if (aligned_len <= wbuf->avail) ... // Not satisfy

  if (wbuf->used) {
    ubifs_leb_write()  // Fill some data in avail wbuf
    len -= wbuf->avail;   // len is still not 8-bytes aligned
    aligned_len -= wbuf->avail;
  }
  n = aligned_len >> c->max_write_shift;
  if (n) {
    n <<= c->max_write_shift;
    err = ubifs_leb_write(c, wbuf->lnum, buf + written,
                          wbuf->offs, n);
    // n > len, read out of bounds less than 8(n-len) bytes

}

, which can be catched by KASAN:
=========================================================

  BUG: KASAN: slab-out-of-bounds in ecc_sw_hamming_calculate+0x1dc/0x7d0
  Read of size 4 at addr ffff888105594ff8 by task kworker/u8:4/128
  Workqueue: writeback wb_workfn (flush-ubifs_0_0)
  Call Trace:
    kasan_report.cold+0x81/0x165
    nand_write_page_swecc+0xa9/0x160
    ubifs_leb_write+0xf2/0x1b0 [ubifs]
    ubifs_wbuf_write_nolock+0x421/0x12c0 [ubifs]
    write_head+0xdc/0x1c0 [ubifs]
    ubifs_jnl_write_inode+0x627/0x960 [ubifs]
    wb_workfn+0x8af/0xb80

Function ubifs_wbuf_write_nolock() accepts that parameter 'len' is not 8 bytes aligned, the 'len' represents the true length of buf (which is allocated in 'ubifs_jnl_xxx', eg. ubifs_jnl_write_inode), so ubifs_wbuf_write_nolock() must handle the length read from 'buf' carefully to write leb safely.

Fetch a reproducer in [Link].

---
 fs/ubifs/io.c |   34 ++++++++++++++++++++++++++++++----
 1 file changed, 30 insertions(+), 4 deletions(-)
fs/ubifs/io.c +30 -4
@@ -854,16 +854,42 @@ int ubifs_wbuf_write_nolock(struct ubifs
 	 */
 	n = aligned_len >> c->max_write_shift;
 	if (n) {
-		n <<= c->max_write_shift;
+		int m = n - 1;
+
 		dbg_io("write %d bytes to LEB %d:%d", n, wbuf->lnum,
 		       wbuf->offs);
-		err = ubifs_leb_write(c, wbuf->lnum, buf + written,
-				      wbuf->offs, n);
+
+		if (m) {
+			/* '(n-1)<<c->max_write_shift < len' is always true. */
+			m <<= c->max_write_shift;
+			err = ubifs_leb_write(c, wbuf->lnum, buf + written,
+					      wbuf->offs, m);
+			if (err)
+				goto out;
+			wbuf->offs += m;
+			aligned_len -= m;
+			len -= m;
+			written += m;
+		}
+
+		/*
+		 * The non-written len of buf may be less than 'n' because
+		 * parameter 'len' is not 8 bytes aligned, so here we read
+		 * min(len, n) bytes from buf.
+		 */
+		n = 1 << c->max_write_shift;
+		memcpy(wbuf->buf, buf + written, min(len, n));
+		if (n > len) {
+			ubifs_assert(c, n - len < 8);
+			ubifs_pad(c, wbuf->buf + len, n - len);
+		}
+
+		err = ubifs_leb_write(c, wbuf->lnum, wbuf->buf, wbuf->offs, n);
 		if (err)
 			goto out;
 		wbuf->offs += n;
 		aligned_len -= n;
-		len -= n;
+		len -= min(len, n);
 		written += n;
 	}
 
author maintainer
#1016 lore raw

PATCH 1012/1126 ubifs: Fix to add refcount once page is set private

In reply to Greg Kroah-Hartman

From: Zhihao Cheng <[email protected]>

commit 3b67db8a6ca83e6ff90b756d3da0c966f61cd37b upstream.

MM defined the rule [1] very clearly that once page was set with PG_private flag, we should increment the refcount in that page, also main flows like pageout(), migrate_page() will assume there is one additional page reference count if page_has_private() returns true. Otherwise, we may get a BUG in page migration:

page:0000000080d05b9d refcount:-1 mapcount:0 mapping:000000005f4d82a8 index:0xe2 pfn:0x14c12
aops:ubifs_file_address_operations [ubifs] ino:8f1 dentry name:"f30e"
flags: 0x1fffff80002405(locked|uptodate|owner_priv_1|private|node=0| zone=1|lastcpupid=0x1fffff)

  page dumped because: VM_BUG_ON_PAGE(page_count(page) != 0)
  ------------[ cut here ]------------
  kernel BUG at include/linux/page_ref.h:184!
  invalid opcode: 0000 [#1] SMP
  CPU: 3 PID: 38 Comm: kcompactd0 Not tainted 5.15.0-rc5
  RIP: 0010:migrate_page_move_mapping+0xac3/0xe70
  Call Trace:
    ubifs_migrate_page+0x22/0xc0 [ubifs]
    move_to_new_page+0xb4/0x600
    migrate_pages+0x1523/0x1cc0
    compact_zone+0x8c5/0x14b0
    kcompactd+0x2bc/0x560
    kthread+0x18c/0x1e0
    ret_from_fork+0x1f/0x30

Before the time, we should make clean a concept, what does refcount means in page gotten from grab_cache_page_write_begin(). There are 2 situations:
Situation 1: refcount is 3, page is created by __page_cache_alloc.
TYPE_A - the write process is using this page
TYPE_B - page is assigned to one certain mapping by calling

	   __add_to_page_cache_locked()

TYPE_C - page is added into pagevec list corresponding current cpu by
calling lru_cache_add()
Situation 2: refcount is 2, page is gotten from the mapping's tree
TYPE_B - page has been assigned to one certain mapping TYPE_A - the write process is using this page (by calling
page_cache_get_speculative())
Filesystem releases one refcount by calling put_page() in xxx_write_end(), the released refcount corresponds to TYPE_A (write task is using it). If there are any processes using a page, page migration process will skip the page by judging whether expected_page_refs() equals to page refcount.

The BUG is caused by following process:
PA(cpu 0) kcompactd(cpu 1)
compact_zone
ubifs_write_begin
page_a = grab_cache_page_write_begin
add_to_page_cache_lru
lru_cache_add
pagevec_add // put page into cpu 0's pagevec
(refcnf = 3, for page creation process)
ubifs_write_end
SetPagePrivate(page_a) // doesn't increase page count ! unlock_page(page_a)
put_page(page_a) // refcnt = 2

				[...]

    PB(cpu 0)

filemap_read
filemap_get_pages
add_to_page_cache_lru
lru_cache_add
__pagevec_lru_add // traverse all pages in cpu 0's pagevec
__pagevec_lru_add_fn
SetPageLRU(page_a)
isolate_migratepages
isolate_migratepages_block
get_page_unless_zero(page_a) // refcnt = 3
list_add(page_a, from_list)
migrate_pages(from_list)
__unmap_and_move
move_to_new_page
ubifs_migrate_page(page_a)
migrate_page_move_mapping
expected_page_refs get 3
(migration[1] + mapping[1] + private[1])
release_pages
put_page_testzero(page_a) // refcnt = 3
page_ref_freeze // refcnt = 0
page_ref_dec_and_test(0 - 1 = -1)
page_ref_unfreeze
VM_BUG_ON_PAGE(-1 != 0, page)

UBIFS doesn't increase the page refcount after setting private flag, which leads to page migration task believes the page is not used by any other processes, so the page is migrated. This causes concurrent accessing on page refcount between put_page() called by other process(eg. read process calls lru_cache_add) and page_ref_unfreeze() called by migration task.

Actually zhangjun has tried to fix this problem [2] by recalculating page refcnt in ubifs_migrate_page(). It's better to follow MM rules [1], because just like Kirill suggested in [2], we need to check all users of page_has_private() helper. Like f2fs does in [3], fix it by adding/deleting refcount when setting/clearing private for a page. BTW, according to [4], we set 'page->private' as 1 because ubifs just simply SetPagePrivate(). And, [5] provided a common helper to set/clear page private, ubifs can use this helper following the example of iomap, afs, btrfs, etc.

Jump [6] to find a reproducer.

[1] https://lore.kernel.org/lkml/[email protected]
[2] https://www.spinics.net/lists/linux-mtd/msg04018.html
[3] http://lkml.iu.edu/hypermail/linux/kernel/1903.0/03313.html
[4] https://lore.kernel.org/linux-f2fs-devel/[email protected]
[5] https://lore.kernel.org/all/[email protected]
[6] https://bugzilla.kernel.org/show_bug.cgi?id=214961

Fixes 1e51764a3c2ac0 ("UBIFS: add new flash file system")
---
 fs/ubifs/file.c |   14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)
fs/ubifs/file.c +7 -7
@@ -570,7 +570,7 @@ static int ubifs_write_end(struct file *
 	}
 
 	if (!PagePrivate(page)) {
-		SetPagePrivate(page);
+		attach_page_private(page, (void *)1);
 		atomic_long_inc(&c->dirty_pg_cnt);
 		__set_page_dirty_nobuffers(page);
 	}
@@ -947,7 +947,7 @@ static int do_writepage(struct page *pag
 		release_existing_page_budget(c);
 
 	atomic_long_dec(&c->dirty_pg_cnt);
-	ClearPagePrivate(page);
+	detach_page_private(page);
 	ClearPageChecked(page);
 
 	kunmap(page);
@@ -1304,7 +1304,7 @@ static void ubifs_invalidatepage(struct
 		release_existing_page_budget(c);
 
 	atomic_long_dec(&c->dirty_pg_cnt);
-	ClearPagePrivate(page);
+	detach_page_private(page);
 	ClearPageChecked(page);
 }
 
@@ -1471,8 +1471,8 @@ static int ubifs_migrate_page(struct add
 		return rc;
 
 	if (PagePrivate(page)) {
-		ClearPagePrivate(page);
-		SetPagePrivate(newpage);
+		detach_page_private(page);
+		attach_page_private(newpage, (void *)1);
 	}
 
 	if (mode != MIGRATE_SYNC_NO_COPY)
@@ -1496,7 +1496,7 @@ static int ubifs_releasepage(struct page
 		return 0;
 	ubifs_assert(c, PagePrivate(page));
 	ubifs_assert(c, 0);
-	ClearPagePrivate(page);
+	detach_page_private(page);
 	ClearPageChecked(page);
 	return 1;
 }
@@ -1567,7 +1567,7 @@ static vm_fault_t ubifs_vm_page_mkwrite(
 	else {
 		if (!PageChecked(page))
 			ubifs_convert_page_budget(c);
-		SetPagePrivate(page);
+		attach_page_private(page, (void *)1);
 		atomic_long_inc(&c->dirty_pg_cnt);
 		__set_page_dirty_nobuffers(page);
 	}
author maintainer
#1017 lore raw

PATCH 1013/1126 ubifs: rename_whiteout: correct old_dir size computing

In reply to Greg Kroah-Hartman

From: Baokun Li <[email protected]>

commit 705757274599e2e064dd3054aabc74e8af31a095 upstream.

When renaming the whiteout file, the old whiteout file is not deleted. Therefore, we add the old dentry size to the old dir like XFS. Otherwise, an error may be reported due to fscki->calc_sz != fscki->size in check_indes.

Fixes 9e0a1fff8db56ea ("ubifs: Implement RENAME_WHITEOUT")
---
 fs/ubifs/dir.c |    3 +++
 1 file changed, 3 insertions(+)
fs/ubifs/dir.c +3 -0
@@ -1402,6 +1402,9 @@ static int do_rename(struct inode *old_d
 			iput(whiteout);
 			goto out_release;
 		}
+
+		/* Add the old_dentry size to the old_dir size. */
+		old_sz -= CALC_DENT_SIZE(fname_len(&old_nm));
 	}
 
 	lock_4_inodes(old_dir, new_dir, new_inode, whiteout);
author maintainer
#1018 lore raw

PATCH 1014/1126 nvme: allow duplicate NSIDs for private namespaces

In reply to Greg Kroah-Hartman

From: Sungup Moon <[email protected]>

commit 5974ea7ce0f9a5987fc8cf5e08ad6e3e70bb542e upstream.

A NVMe subsystem with multiple controller can have private namespaces that use the same NSID under some conditions:

"If Namespace Management, ANA Reporting, or NVM Sets are supported, the
NSIDs shall be unique within the NVM subsystem. If the Namespace Management, ANA Reporting, and NVM Sets are not supported, then NSIDs:
a) for shared namespace shall be unique; and
b) for private namespace are not required to be unique."

Reference: Section 6.1.6 NSID and Namespace Usage; NVM Express 1.4c spec.

Make sure this specific setup is supported in Linux.

Fixes 9ad1927a3bc2 ("nvme: always search for namespace head")

[hch: refactored and fixed the controller vs subsystem based naming
conflict]

---
 drivers/nvme/host/core.c      |   15 ++++++++++-----
 drivers/nvme/host/multipath.c |    7 ++++---
 drivers/nvme/host/nvme.h      |   19 +++++++++++++++++++
 include/linux/nvme.h          |    1 +
 4 files changed, 34 insertions(+), 8 deletions(-)
drivers/nvme/host/core.c +10 -5
@@ -3574,15 +3574,20 @@ static const struct attribute_group *nvm
 	NULL,
 };
 
-static struct nvme_ns_head *nvme_find_ns_head(struct nvme_subsystem *subsys,
+static struct nvme_ns_head *nvme_find_ns_head(struct nvme_ctrl *ctrl,
 		unsigned nsid)
 {
 	struct nvme_ns_head *h;
 
-	lockdep_assert_held(&subsys->lock);
+	lockdep_assert_held(&ctrl->subsys->lock);
 
-	list_for_each_entry(h, &subsys->nsheads, entry) {
-		if (h->ns_id != nsid)
+	list_for_each_entry(h, &ctrl->subsys->nsheads, entry) {
+		/*
+		 * Private namespaces can share NSIDs under some conditions.
+		 * In that case we can't use the same ns_head for namespaces
+		 * with the same NSID.
+		 */
+		if (h->ns_id != nsid || !nvme_is_unique_nsid(ctrl, h))
 			continue;
 		if (!list_empty(&h->list) && nvme_tryget_ns_head(h))
 			return h;
@@ -3750,7 +3755,7 @@ static int nvme_init_ns_head(struct nvme
 	int ret = 0;
 
 	mutex_lock(&ctrl->subsys->lock);
-	head = nvme_find_ns_head(ctrl->subsys, nsid);
+	head = nvme_find_ns_head(ctrl, nsid);
 	if (!head) {
 		head = nvme_alloc_ns_head(ctrl, nsid, ids);
 		if (IS_ERR(head)) {
drivers/nvme/host/multipath.c +4 -3
@@ -504,10 +504,11 @@ int nvme_mpath_alloc_disk(struct nvme_ct
 
 	/*
 	 * Add a multipath node if the subsystems supports multiple controllers.
-	 * We also do this for private namespaces as the namespace sharing data could
-	 * change after a rescan.
+	 * We also do this for private namespaces as the namespace sharing flag
+	 * could change after a rescan.
 	 */
-	if (!(ctrl->subsys->cmic & NVME_CTRL_CMIC_MULTI_CTRL) || !multipath)
+	if (!(ctrl->subsys->cmic & NVME_CTRL_CMIC_MULTI_CTRL) ||
+	    !nvme_is_unique_nsid(ctrl, head) || !multipath)
 		return 0;
 
 	head->disk = blk_alloc_disk(ctrl->numa_node);
drivers/nvme/host/nvme.h +19 -0
@@ -716,6 +716,25 @@ static inline bool nvme_check_ready(stru
 		return queue_live;
 	return __nvme_check_ready(ctrl, rq, queue_live);
 }
+
+/*
+ * NSID shall be unique for all shared namespaces, or if at least one of the
+ * following conditions is met:
+ *   1. Namespace Management is supported by the controller
+ *   2. ANA is supported by the controller
+ *   3. NVM Set are supported by the controller
+ *
+ * In other case, private namespace are not required to report a unique NSID.
+ */
+static inline bool nvme_is_unique_nsid(struct nvme_ctrl *ctrl,
+		struct nvme_ns_head *head)
+{
+	return head->shared ||
+		(ctrl->oacs & NVME_CTRL_OACS_NS_MNGT_SUPP) ||
+		(ctrl->subsys->cmic & NVME_CTRL_CMIC_ANA) ||
+		(ctrl->ctratt & NVME_CTRL_CTRATT_NVM_SETS);
+}
+
 int nvme_submit_sync_cmd(struct request_queue *q, struct nvme_command *cmd,
 		void *buf, unsigned bufflen);
 int __nvme_submit_sync_cmd(struct request_queue *q, struct nvme_command *cmd,
include/linux/nvme.h +1 -0
@@ -337,6 +337,7 @@ enum {
 	NVME_CTRL_ONCS_TIMESTAMP		= 1 << 6,
 	NVME_CTRL_VWC_PRESENT			= 1 << 0,
 	NVME_CTRL_OACS_SEC_SUPP                 = 1 << 0,
+	NVME_CTRL_OACS_NS_MNGT_SUPP		= 1 << 3,
 	NVME_CTRL_OACS_DIRECTIVES		= 1 << 5,
 	NVME_CTRL_OACS_DBBUF_SUPP		= 1 << 8,
 	NVME_CTRL_LPA_CMD_EFFECTS_LOG		= 1 << 1,
author maintainer
#1019 lore raw

PATCH 1015/1126 nvme: fix the read-only state for zoned namespaces with unsupposed features

In reply to Greg Kroah-Hartman

From: Pankaj Raghav <[email protected]>

commit 726be2c72efc0a64c206e854b8996ad3ab9c7507 upstream.

commit 2f4c9ba23b88 ("nvme: export zoned namespaces without Zone Append support read-only") marks zoned namespaces without append support read-only. It does iso by setting NVME_NS_FORCE_RO in ns->flags in nvme_update_zone_info and checking for that flag later in nvme_update_disk_info to mark the disk as read-only.

But commit 73d90386b559 ("nvme: cleanup zone information initialization") rearranged nvme_update_disk_info to be called before nvme_update_zone_info and thus not marking the disk as read-only. The call order cannot be just reverted because nvme_update_zone_info sets certain queue parameters such as zone_write_granularity that depend on the prior call to nvme_update_disk_info.

Remove the call to set_disk_ro in nvme_update_disk_info. and call set_disk_ro after nvme_update_zone_info and nvme_update_disk_info to set the permission for ZNS drives correctly. The same applies to the multipath disk path.

Fixes 73d90386b559 ("nvme: cleanup zone information initialization")
---
 drivers/nvme/host/core.c |    8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)
drivers/nvme/host/core.c +5 -3
@@ -1857,9 +1857,6 @@ static void nvme_update_disk_info(struct
 	nvme_config_discard(disk, ns);
 	blk_queue_max_write_zeroes_sectors(disk->queue,
 					   ns->ctrl->max_zeroes_sectors);
-
-	set_disk_ro(disk, (id->nsattr & NVME_NS_ATTR_RO) ||
-		test_bit(NVME_NS_FORCE_RO, &ns->flags));
 }
 
 static inline bool nvme_first_scan(struct gendisk *disk)
@@ -1918,6 +1915,8 @@ static int nvme_update_ns_info(struct nv
 			goto out_unfreeze;
 	}
 
+	set_disk_ro(ns->disk, (id->nsattr & NVME_NS_ATTR_RO) ||
+		test_bit(NVME_NS_FORCE_RO, &ns->flags));
 	set_bit(NVME_NS_READY, &ns->flags);
 	blk_mq_unfreeze_queue(ns->disk->queue);
 
@@ -1930,6 +1929,9 @@ static int nvme_update_ns_info(struct nv
 	if (nvme_ns_head_multipath(ns->head)) {
 		blk_mq_freeze_queue(ns->head->disk->queue);
 		nvme_update_disk_info(ns->head->disk, ns, id);
+		set_disk_ro(ns->head->disk,
+			    (id->nsattr & NVME_NS_ATTR_RO) ||
+				    test_bit(NVME_NS_FORCE_RO, &ns->flags));
 		nvme_mpath_revalidate_paths(ns);
 		blk_stack_limits(&ns->head->disk->queue->limits,
 				 &ns->queue->limits, 0);
author maintainer
#1020 lore raw

PATCH 1016/1126 wireguard: queueing: use CFI-safe ptr_ring cleanup function

In reply to Greg Kroah-Hartman

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

commit ec59f128a9bd4255798abb1e06ac3b442f46ef68 upstream.

We make too nuanced use of ptr_ring to entirely move to the skb_array wrappers, but we at least should avoid the naughty function pointer cast when cleaning up skbs. Otherwise RAP/CFI will honk at us. This patch uses the __skb_array_destroy_skb wrapper for the cleanup, rather than directly providing kfree_skb, which is what other drivers in the same situation do too.

Fixes 886fcee939ad ("wireguard: receive: use ring buffer for incoming handshakes")
---
 drivers/net/wireguard/queueing.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/wireguard/queueing.c +2 -1
@@ -4,6 +4,7 @@
  */
 
 #include "queueing.h"
+#include <linux/skb_array.h>
 
 struct multicore_worker __percpu *
 wg_packet_percpu_multicore_worker_alloc(work_func_t function, void *ptr)
@@ -42,7 +43,7 @@ void wg_packet_queue_free(struct crypt_q
 {
 	free_percpu(queue->worker);
 	WARN_ON(!purge && !__ptr_ring_empty(&queue->ring));
-	ptr_ring_cleanup(&queue->ring, purge ? (void(*)(void*))kfree_skb : NULL);
+	ptr_ring_cleanup(&queue->ring, purge ? __skb_array_destroy_skb : NULL);
 }
 
 #define NEXT(skb) ((skb)->prev)
author maintainer
#1021 lore raw

PATCH 1017/1126 wireguard: socket: free skb in send6 when ipv6 is disabled

In reply to Greg Kroah-Hartman

From: Wang Hai <[email protected]>

commit bbbf962d9460194993ee1943a793a0a0af4a7fbf upstream.

I got a memory leak report:

unreferenced object 0xffff8881191fc040 (size 232):
comm "kworker/u17:0", pid 23193, jiffies 4295238848 (age 3464.870s) hex dump (first 32 bytes):

    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
  backtrace:
    [<ffffffff814c3ef4>] slab_post_alloc_hook+0x84/0x3b0
    [<ffffffff814c8977>] kmem_cache_alloc_node+0x167/0x340
    [<ffffffff832974fb>] __alloc_skb+0x1db/0x200
    [<ffffffff82612b5d>] wg_socket_send_buffer_to_peer+0x3d/0xc0
    [<ffffffff8260e94a>] wg_packet_send_handshake_initiation+0xfa/0x110
    [<ffffffff8260ec81>] wg_packet_handshake_send_worker+0x21/0x30
    [<ffffffff8119c558>] process_one_work+0x2e8/0x770
    [<ffffffff8119ca2a>] worker_thread+0x4a/0x4b0
    [<ffffffff811a88e0>] kthread+0x120/0x160
    [<ffffffff8100242f>] ret_from_fork+0x1f/0x30

In function wg_socket_send_buffer_as_reply_to_skb() or wg_socket_send_ buffer_to_peer(), the semantics of send6() is required to free skb. But when CONFIG_IPV6 is disable, kfree_skb() is missing. This patch adds it to fix this bug.

Fixes e7096c131e51 ("net: WireGuard secure network tunnel")
---
 drivers/net/wireguard/socket.c |    1 +
 1 file changed, 1 insertion(+)
drivers/net/wireguard/socket.c +1 -0
@@ -160,6 +160,7 @@ out:
 	rcu_read_unlock_bh();
 	return ret;
 #else
+	kfree_skb(skb);
 	return -EAFNOSUPPORT;
 #endif
 }
author maintainer
#1022 lore raw

PATCH 1018/1126 wireguard: socket: ignore v6 endpoints when ipv6 is disabled

In reply to Greg Kroah-Hartman

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

commit 77fc73ac89be96ec8f39e8efa53885caa7cb3645 upstream.

The previous commit fixed a memory leak on the send path in the event that IPv6 is disabled at compile time, but how did a packet even arrive there to begin with? It turns out we have previously allowed IPv6 endpoints even when IPv6 support is disabled at compile time. This is awkward and inconsistent. Instead, let's just ignore all things IPv6, the same way we do other malformed endpoints, in the case where IPv6 is disabled.

Fixes e7096c131e51 ("net: WireGuard secure network tunnel")
---
 drivers/net/wireguard/socket.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/wireguard/socket.c +2 -2
@@ -242,7 +242,7 @@ int wg_socket_endpoint_from_skb(struct e
 		endpoint->addr4.sin_addr.s_addr = ip_hdr(skb)->saddr;
 		endpoint->src4.s_addr = ip_hdr(skb)->daddr;
 		endpoint->src_if4 = skb->skb_iif;
-	} else if (skb->protocol == htons(ETH_P_IPV6)) {
+	} else if (IS_ENABLED(CONFIG_IPV6) && skb->protocol == htons(ETH_P_IPV6)) {
 		endpoint->addr6.sin6_family = AF_INET6;
 		endpoint->addr6.sin6_port = udp_hdr(skb)->source;
 		endpoint->addr6.sin6_addr = ipv6_hdr(skb)->saddr;
@@ -285,7 +285,7 @@ void wg_socket_set_peer_endpoint(struct
 		peer->endpoint.addr4 = endpoint->addr4;
 		peer->endpoint.src4 = endpoint->src4;
 		peer->endpoint.src_if4 = endpoint->src_if4;
-	} else if (endpoint->addr.sa_family == AF_INET6) {
+	} else if (IS_ENABLED(CONFIG_IPV6) && endpoint->addr.sa_family == AF_INET6) {
 		peer->endpoint.addr6 = endpoint->addr6;
 		peer->endpoint.src6 = endpoint->src6;
 	} else {
author maintainer
#1023 lore raw

PATCH 1019/1126 XArray: Fix xas_create_range() when multi-order entry present

In reply to Greg Kroah-Hartman

From: Matthew Wilcox (Oracle) <[email protected]>

commit 3e3c658055c002900982513e289398a1aad4a488 upstream.

If there is already an entry present that is of order >= XA_CHUNK_SHIFT when we call xas_create_range(), xas_create_range() will misinterpret that entry as a node and dereference xa_node->parent, generally leading to a crash that looks something like this:

general protection fault, probably for non-canonical address 0xdffffc0000000001:
0000 [#1] PREEMPT SMP KASAN
KASAN: null-ptr-deref in range [0x0000000000000008-0x000000000000000f]
CPU: 0 PID: 32 Comm: khugepaged Not tainted 5.17.0-rc8-syzkaller-00003-g56e337f2cf13 #0
RIP: 0010:xa_parent_locked include/linux/xarray.h:1207 [inline]
RIP: 0010:xas_create_range+0x2d9/0x6e0 lib/xarray.c:725

It's deterministically reproducable once you know what the problem is, but producing it in a live kernel requires khugepaged to hit a race. While the problem has been present since xas_create_range() was introduced, I'm not aware of a way to hit it before the page cache was converted to use multi-index entries.

Fixes 6b24ca4a1a8d ("mm: Use multi-index entries in the page cache")
Reported-by [email protected]
---
 lib/test_xarray.c |   22 ++++++++++++++++++++++
 lib/xarray.c      |    2 ++
 2 files changed, 24 insertions(+)
lib/test_xarray.c +22 -0
@@ -1463,6 +1463,25 @@ unlock:
 	XA_BUG_ON(xa, !xa_empty(xa));
 }
 
+static noinline void check_create_range_5(struct xarray *xa,
+		unsigned long index, unsigned int order)
+{
+	XA_STATE_ORDER(xas, xa, index, order);
+	unsigned int i;
+
+	xa_store_order(xa, index, order, xa_mk_index(index), GFP_KERNEL);
+
+	for (i = 0; i < order + 10; i++) {
+		do {
+			xas_lock(&xas);
+			xas_create_range(&xas);
+			xas_unlock(&xas);
+		} while (xas_nomem(&xas, GFP_KERNEL));
+	}
+
+	xa_destroy(xa);
+}
+
 static noinline void check_create_range(struct xarray *xa)
 {
 	unsigned int order;
@@ -1490,6 +1509,9 @@ static noinline void check_create_range(
 		check_create_range_4(xa, (3U << order) + 1, order);
 		check_create_range_4(xa, (3U << order) - 1, order);
 		check_create_range_4(xa, (1U << 24) + 1, order);
+
+		check_create_range_5(xa, 0, order);
+		check_create_range_5(xa, (1U << order), order);
 	}
 
 	check_create_range_3();
lib/xarray.c +2 -0
@@ -722,6 +722,8 @@ void xas_create_range(struct xa_state *x
 
 		for (;;) {
 			struct xa_node *node = xas->xa_node;
+			if (node->shift >= shift)
+				break;
 			xas->xa_node = xa_parent_locked(xas->xa, node);
 			xas->xa_offset = node->offset - 1;
 			if (node->offset != 0)
author maintainer
#1024 lore raw

PATCH 1020/1126 can: mcba_usb: mcba_usb_start_xmit(): fix double dev_kfree_skb in error path

In reply to Greg Kroah-Hartman

From: Hangyu Hua <[email protected]>

commit 04c9b00ba83594a29813d6b1fb8fdc93a3915174 upstream.

There is no need to call dev_kfree_skb() when usb_submit_urb() fails because can_put_echo_skb() deletes original skb and can_free_echo_skb() deletes the cloned skb.

Fixes 51f3baad7de9 ("can: mcba_usb: Add support for Microchip CAN BUS Analyzer")
---
 drivers/net/can/usb/mcba_usb.c |    1 -
 1 file changed, 1 deletion(-)
drivers/net/can/usb/mcba_usb.c +0 -1
@@ -364,7 +364,6 @@ static netdev_tx_t mcba_usb_start_xmit(s
 xmit_failed:
 	can_free_echo_skb(priv->netdev, ctx->ndx, NULL);
 	mcba_usb_free_ctx(ctx);
-	dev_kfree_skb(skb);
 	stats->tx_dropped++;
 
 	return NETDEV_TX_OK;
author maintainer
#1025 lore raw

PATCH 1021/1126 can: mcba_usb: properly check endpoint type

In reply to Greg Kroah-Hartman

From: Pavel Skripkin <[email protected]>

commit 136bed0bfd3bc9c95c88aafff2d22ecb3a919f23 upstream.

Syzbot reported warning in usb_submit_urb() which is caused by wrong endpoint type. We should check that in endpoint is actually present to prevent this warning.

Found pipes are now saved to struct mcba_priv and code uses them directly instead of making pipes in place.

Fail log:

| usb 5-1: BOGUS urb xfer, pipe 3 != type 1
| WARNING: CPU: 1 PID: 49 at drivers/usb/core/urb.c:502 usb_submit_urb+0xed2/0x18a0 drivers/usb/core/urb.c:502
| Modules linked in:
| CPU: 1 PID: 49 Comm: kworker/1:2 Not tainted 5.17.0-rc6-syzkaller-00184-g38f80f42147f #0
| Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.14.0-2 04/01/2014
| Workqueue: usb_hub_wq hub_event
| RIP: 0010:usb_submit_urb+0xed2/0x18a0 drivers/usb/core/urb.c:502
| ...
| Call Trace:
|  <TASK>
|  mcba_usb_start drivers/net/can/usb/mcba_usb.c:662 [inline]
|  mcba_usb_probe+0x8a3/0xc50 drivers/net/can/usb/mcba_usb.c:858
|  usb_probe_interface+0x315/0x7f0 drivers/usb/core/driver.c:396
|  call_driver_probe drivers/base/dd.c:517 [inline]
Fixes 51f3baad7de9 ("can: mcba_usb: Add support for Microchip CAN BUS Analyzer")
Reported-and-tested-by [email protected]
---
 drivers/net/can/usb/mcba_usb.c |   26 ++++++++++++++++----------
 1 file changed, 16 insertions(+), 10 deletions(-)
drivers/net/can/usb/mcba_usb.c +16 -10
@@ -33,10 +33,6 @@
 #define MCBA_USB_RX_BUFF_SIZE 64
 #define MCBA_USB_TX_BUFF_SIZE (sizeof(struct mcba_usb_msg))
 
-/* MCBA endpoint numbers */
-#define MCBA_USB_EP_IN 1
-#define MCBA_USB_EP_OUT 1
-
 /* Microchip command id */
 #define MBCA_CMD_RECEIVE_MESSAGE 0xE3
 #define MBCA_CMD_I_AM_ALIVE_FROM_CAN 0xF5
@@ -83,6 +79,8 @@ struct mcba_priv {
 	atomic_t free_ctx_cnt;
 	void *rxbuf[MCBA_MAX_RX_URBS];
 	dma_addr_t rxbuf_dma[MCBA_MAX_RX_URBS];
+	int rx_pipe;
+	int tx_pipe;
 };
 
 /* CAN frame */
@@ -268,10 +266,8 @@ static netdev_tx_t mcba_usb_xmit(struct
 
 	memcpy(buf, usb_msg, MCBA_USB_TX_BUFF_SIZE);
 
-	usb_fill_bulk_urb(urb, priv->udev,
-			  usb_sndbulkpipe(priv->udev, MCBA_USB_EP_OUT), buf,
-			  MCBA_USB_TX_BUFF_SIZE, mcba_usb_write_bulk_callback,
-			  ctx);
+	usb_fill_bulk_urb(urb, priv->udev, priv->tx_pipe, buf, MCBA_USB_TX_BUFF_SIZE,
+			  mcba_usb_write_bulk_callback, ctx);
 
 	urb->transfer_flags |= URB_NO_TRANSFER_DMA_MAP;
 	usb_anchor_urb(urb, &priv->tx_submitted);
@@ -607,7 +603,7 @@ static void mcba_usb_read_bulk_callback(
 resubmit_urb:
 
 	usb_fill_bulk_urb(urb, priv->udev,
-			  usb_rcvbulkpipe(priv->udev, MCBA_USB_EP_OUT),
+			  priv->rx_pipe,
 			  urb->transfer_buffer, MCBA_USB_RX_BUFF_SIZE,
 			  mcba_usb_read_bulk_callback, priv);
 
@@ -652,7 +648,7 @@ static int mcba_usb_start(struct mcba_pr
 		urb->transfer_dma = buf_dma;
 
 		usb_fill_bulk_urb(urb, priv->udev,
-				  usb_rcvbulkpipe(priv->udev, MCBA_USB_EP_IN),
+				  priv->rx_pipe,
 				  buf, MCBA_USB_RX_BUFF_SIZE,
 				  mcba_usb_read_bulk_callback, priv);
 		urb->transfer_flags |= URB_NO_TRANSFER_DMA_MAP;
@@ -806,6 +802,13 @@ static int mcba_usb_probe(struct usb_int
 	struct mcba_priv *priv;
 	int err;
 	struct usb_device *usbdev = interface_to_usbdev(intf);
+	struct usb_endpoint_descriptor *in, *out;
+
+	err = usb_find_common_endpoints(intf->cur_altsetting, &in, &out, NULL, NULL);
+	if (err) {
+		dev_err(&intf->dev, "Can't find endpoints\n");
+		return err;
+	}
 
 	netdev = alloc_candev(sizeof(struct mcba_priv), MCBA_MAX_TX_URBS);
 	if (!netdev) {
@@ -851,6 +854,9 @@ static int mcba_usb_probe(struct usb_int
 		goto cleanup_free_candev;
 	}
 
+	priv->rx_pipe = usb_rcvbulkpipe(priv->udev, in->bEndpointAddress);
+	priv->tx_pipe = usb_sndbulkpipe(priv->udev, out->bEndpointAddress);
+
 	devm_can_led_init(netdev);
 
 	/* Start USB dev only if we have successfully registered CAN device */
author maintainer
#1026 lore raw

PATCH 1022/1126 can: mcp251xfd: mcp251xfd_register_get_dev_id(): fix return of error value

In reply to Greg Kroah-Hartman

From: Tom Rix <[email protected]>

commit fa7b514d2b2894e052b8e94c7a29feb98e90093f upstream.

Clang static analysis reports this issue:

| mcp251xfd-core.c:1813:7: warning: The left operand
|   of '&' is a garbage value
|   FIELD_GET(MCP251XFD_REG_DEVID_ID_MASK, dev_id),
|   ^                                      ~~~~~~

dev_id is set in a successful call to mcp251xfd_register_get_dev_id(). Though the status of calls made by mcp251xfd_register_get_dev_id() are checked and handled, their status' are not returned. So return err.

Fixes 55e5b97f003e ("can: mcp25xxfd: add driver for Microchip MCP25xxFD SPI CAN")
---
 drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c +1 -1
@@ -1657,7 +1657,7 @@ mcp251xfd_register_get_dev_id(const stru
  out_kfree_buf_rx:
 	kfree(buf_rx);
 
-	return 0;
+	return err;
 }
 
 #define MCP251XFD_QUIRK_ACTIVE(quirk) \
author maintainer
#1027 lore raw

PATCH 1023/1126 XArray: Include bitmap.h from xarray.h

In reply to Greg Kroah-Hartman

From: Matthew Wilcox (Oracle) <[email protected]>

commit 22f56b8e890d4e2835951b437bb6eeebfd1cb18b upstream.

xas_find_chunk() calls find_next_bit(), which is defined in find.h, included from bitmap.h. Inside the kernel, this isn't a problem because bitmap.h is included from cpumask.h which is dragged in (eventually) by gfp.h. When building the test-suite, that doesn't happen, so we need to include bitmap.h explicitly.

Fixes 4ade0818cf04 ("tools: sync tools/bitmap with mother linux")
---
 include/linux/xarray.h | 1 +
 1 file changed, 1 insertion(+)
include/linux/xarray.h +1 -0
index d6d5da6ed735..66e28bc1a023 100644
@@ -9,6 +9,7 @@
  * See Documentation/core-api/xarray.rst for how to use the XArray.
  */
 
+#include <linux/bitmap.h>
 #include <linux/bug.h>
 #include <linux/compiler.h>
 #include <linux/gfp.h>
Signature · 1 line
2.35.1
author maintainer
#1028 lore raw

PATCH 1024/1126 XArray: Update the LRU list in xas_split()

In reply to Greg Kroah-Hartman

From: Matthew Wilcox (Oracle) <[email protected]>

commit 3ed4bb77156da0bc732847c8c9df92454c1fbeea upstream.

When splitting a value entry, we may need to add the new nodes to the LRU list and remove the parent node from the LRU list. The WARN_ON checks in shadow_lru_isolate() catch this oversight. This bug was latent until we stopped splitting folios in shrink_page_list() with commit 820c4e2e6f51 ("mm/vmscan: Free non-shmem folios without splitting them"). That allows the creation of large shadow entries, and subsequently when trying to page in a small page, we will split the large shadow entry in __filemap_add_folio().

Fixes 8fc75643c5e1 ("XArray: add xas_split")
---
 lib/xarray.c |    2 ++
 1 file changed, 2 insertions(+)
lib/xarray.c +2 -0
@@ -1081,6 +1081,7 @@ void xas_split(struct xa_state *xas, voi
 					xa_mk_node(child));
 			if (xa_is_value(curr))
 				values--;
+			xas_update(xas, child);
 		} else {
 			unsigned int canon = offset - xas->xa_sibs;
 
@@ -1095,6 +1096,7 @@ void xas_split(struct xa_state *xas, voi
 	} while (offset-- > xas->xa_offset);
 
 	node->nr_values += values;
+	xas_update(xas, node);
 }
 EXPORT_SYMBOL_GPL(xas_split);
 #endif
author maintainer
#1029 lore raw

PATCH 1025/1126 modpost: restore the warning message for missing symbol versions

In reply to Greg Kroah-Hartman

From: Masahiro Yamada <[email protected]>

commit bf5c0c2231bcab677e5cdfb7f73e6c79f6d8c2d4 upstream.

This log message was accidentally chopped off.

I was wondering why this happened, but checking the ML log, Mark precisely followed my suggestion [1].

I just used "..." because I was too lazy to type the sentence fully. Sorry for the confusion.

[1]: https://lore.kernel.org/all/CAK7LNAR6bXXk9-ZzZYpTqzFqdYbQsZHmiWspu27rtsFxvfRuVA@mail.gmail.com/

Fixes 4a6795933a89 ("kbuild: modpost: Explicitly warn about unprototyped symbols")
---
 scripts/mod/modpost.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
scripts/mod/modpost.c +1 -1
@@ -669,7 +669,7 @@ static void handle_modversion(const stru
 	unsigned int crc;
 
 	if (sym->st_shndx == SHN_UNDEF) {
-		warn("EXPORT symbol \"%s\" [%s%s] version ...\n"
+		warn("EXPORT symbol \"%s\" [%s%s] version generation failed, symbol will not be versioned.\n"
 		     "Is \"%s\" prototyped in <asm/asm-prototypes.h>?\n",
 		     symname, mod->name, mod->is_vmlinux ? "" : ".ko",
 		     symname);
author maintainer
#1030 lore raw

PATCH 1026/1126 rtc: gamecube: Fix refcount leak in gamecube_rtc_read_offset_from_sram

In reply to Greg Kroah-Hartman

From: Miaoqian Lin <[email protected]>

commit 4b2dc39ca024990abe36ad5d145c4fe0c06afd34 upstream.

The of_find_compatible_node() function returns a node pointer with refcount incremented, We should use of_node_put() on it when done Add the missing of_node_put() to release the refcount.

Fixes 86559400b3ef ("rtc: gamecube: Add a RTC driver for the GameCube, Wii and Wii U")
---
 drivers/rtc/rtc-gamecube.c |    1 +
 1 file changed, 1 insertion(+)
drivers/rtc/rtc-gamecube.c +1 -0
@@ -235,6 +235,7 @@ static int gamecube_rtc_read_offset_from
 	}
 
 	ret = of_address_to_resource(np, 0, &res);
+	of_node_put(np);
 	if (ret) {
 		pr_err("no io memory range found\n");
 		return -1;
author maintainer
#1031 lore raw

PATCH 1027/1126 rtc: check if __rtc_read_time was successful

In reply to Greg Kroah-Hartman

From: Tom Rix <[email protected]>

commit 915593a7a663b2ad08b895a5f3ba8b19d89d4ebf upstream.

Clang static analysis reports this issue interface.c:810:8: warning: Passed-by-value struct
argument contains uninitialized data

  now = rtc_tm_to_ktime(tm);
      ^~~~~~~~~~~~~~~~~~~

tm is set by a successful call to __rtc_read_time() but its return status is not checked. Check if it was successful before setting the enabled flag. Move the decl of err to function scope.

Fixes 2b2f5ff00f63 ("rtc: interface: ignore expired timers when enqueuing new timers")
---
 drivers/rtc/interface.c |    7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)
drivers/rtc/interface.c +5 -2
@@ -804,9 +804,13 @@ static int rtc_timer_enqueue(struct rtc_
 	struct timerqueue_node *next = timerqueue_getnext(&rtc->timerqueue);
 	struct rtc_time tm;
 	ktime_t now;
+	int err;
+
+	err = __rtc_read_time(rtc, &tm);
+	if (err)
+		return err;
 
 	timer->enabled = 1;
-	__rtc_read_time(rtc, &tm);
 	now = rtc_tm_to_ktime(tm);
 
 	/* Skip over expired timers */
@@ -820,7 +824,6 @@ static int rtc_timer_enqueue(struct rtc_
 	trace_rtc_timer_enqueue(timer);
 	if (!next || ktime_before(timer->node.expires, next->expires)) {
 		struct rtc_wkalrm alarm;
-		int err;
 
 		alarm.time = rtc_ktime_to_tm(timer->node.expires);
 		alarm.enabled = 1;
author maintainer
#1032 lore raw

PATCH 1028/1126 loop: fix ioctl calls using compat_loop_info

In reply to Greg Kroah-Hartman

From: Carlos Llamas <[email protected]>

commit f941c51eeac7ebe0f8ec30943bf78e7f60aad039 upstream.

Support for cryptoloop was deleted in commit 47e9624616c8 ("block:
remove support for cryptoloop and the xor transfer"), making the usage of loop_info->lo_encrypt_type obsolete. However, this member was also removed from the compat_loop_info definition and this breaks userspace ioctl calls for 32-bit binaries and CONFIG_COMPAT=y.

This patch restores the compat_loop_info->lo_encrypt_type member and marks it obsolete as well as in the uapi header definitions.

Fixes 47e9624616c8 ("block: remove support for cryptoloop and the xor transfer")
---
 drivers/block/loop.c      |    1 +
 include/uapi/linux/loop.h |    4 ++--
 2 files changed, 3 insertions(+), 2 deletions(-)
drivers/block/loop.c +1 -0
@@ -1592,6 +1592,7 @@ struct compat_loop_info {
 	compat_ulong_t	lo_inode;       /* ioctl r/o */
 	compat_dev_t	lo_rdevice;     /* ioctl r/o */
 	compat_int_t	lo_offset;
+	compat_int_t	lo_encrypt_type;        /* obsolete, ignored */
 	compat_int_t	lo_encrypt_key_size;    /* ioctl w/o */
 	compat_int_t	lo_flags;       /* ioctl r/o */
 	char		lo_name[LO_NAME_SIZE];
include/uapi/linux/loop.h +2 -2
@@ -45,7 +45,7 @@ struct loop_info {
 	unsigned long	   lo_inode; 		/* ioctl r/o */
 	__kernel_old_dev_t lo_rdevice; 		/* ioctl r/o */
 	int		   lo_offset;
-	int		   lo_encrypt_type;
+	int		   lo_encrypt_type;		/* obsolete, ignored */
 	int		   lo_encrypt_key_size; 	/* ioctl w/o */
 	int		   lo_flags;
 	char		   lo_name[LO_NAME_SIZE];
@@ -61,7 +61,7 @@ struct loop_info64 {
 	__u64		   lo_offset;
 	__u64		   lo_sizelimit;/* bytes, 0 == max available */
 	__u32		   lo_number;			/* ioctl r/o */
-	__u32		   lo_encrypt_type;
+	__u32		   lo_encrypt_type;		/* obsolete, ignored */
 	__u32		   lo_encrypt_key_size;		/* ioctl w/o */
 	__u32		   lo_flags;
 	__u8		   lo_file_name[LO_NAME_SIZE];
author maintainer
#1033 lore raw

PATCH 1029/1126 gfs2: gfs2_setattr_size error path fix

In reply to Greg Kroah-Hartman

From: Andreas Gruenbacher <[email protected]>

commit 7336905a89f19173bf9301cd50a24421162f417c upstream.

When gfs2_setattr_size() fails, it calls gfs2_rs_delete(ip, NULL) to get rid of any reservations the inode may have. Instead, it should pass in the inode's write count as the second parameter to allow gfs2_rs_delete() to figure out if the inode has any writers left.

In a next step, there are two instances of gfs2_rs_delete(ip, NULL) left where we know that there can be no other users of the inode. Replace those with gfs2_rs_deltree(&ip->i_res) to avoid the unnecessary write count check.

With that, gfs2_rs_delete() is only called with the inode's actual write count, so get rid of the second parameter.

Fixes a097dc7e24cb ("GFS2: Make rgrp reservations part of the gfs2_inode structure")
---
 fs/gfs2/bmap.c  |    2 +-
 fs/gfs2/file.c  |    2 +-
 fs/gfs2/inode.c |    2 +-
 fs/gfs2/rgrp.c  |    7 ++++---
 fs/gfs2/rgrp.h  |    2 +-
 fs/gfs2/super.c |    2 +-
 6 files changed, 9 insertions(+), 8 deletions(-)
fs/gfs2/bmap.c +1 -1
@@ -2146,7 +2146,7 @@ int gfs2_setattr_size(struct inode *inod
 
 	ret = do_shrink(inode, newsize);
 out:
-	gfs2_rs_delete(ip, NULL);
+	gfs2_rs_delete(ip);
 	gfs2_qa_put(ip);
 	return ret;
 }
fs/gfs2/file.c +1 -1
@@ -706,7 +706,7 @@ static int gfs2_release(struct inode *in
 
 	if (file->f_mode & FMODE_WRITE) {
 		if (gfs2_rs_active(&ip->i_res))
-			gfs2_rs_delete(ip, &inode->i_writecount);
+			gfs2_rs_delete(ip);
 		gfs2_qa_put(ip);
 	}
 	return 0;
fs/gfs2/inode.c +1 -1
@@ -793,7 +793,7 @@ fail_free_inode:
 		if (free_vfs_inode) /* else evict will do the put for us */
 			gfs2_glock_put(ip->i_gl);
 	}
-	gfs2_rs_delete(ip, NULL);
+	gfs2_rs_deltree(&ip->i_res);
 	gfs2_qa_put(ip);
 fail_free_acls:
 	posix_acl_release(default_acl);
fs/gfs2/rgrp.c +4 -3
@@ -680,13 +680,14 @@ void gfs2_rs_deltree(struct gfs2_blkrese
 /**
  * gfs2_rs_delete - delete a multi-block reservation
  * @ip: The inode for this reservation
- * @wcount: The inode's write count, or NULL
  *
  */
-void gfs2_rs_delete(struct gfs2_inode *ip, atomic_t *wcount)
+void gfs2_rs_delete(struct gfs2_inode *ip)
 {
+	struct inode *inode = &ip->i_inode;
+
 	down_write(&ip->i_rw_mutex);
-	if ((wcount == NULL) || (atomic_read(wcount) <= 1))
+	if (atomic_read(&inode->i_writecount) <= 1)
 		gfs2_rs_deltree(&ip->i_res);
 	up_write(&ip->i_rw_mutex);
 }
fs/gfs2/rgrp.h +1 -1
@@ -45,7 +45,7 @@ extern int gfs2_alloc_blocks(struct gfs2
 			     bool dinode, u64 *generation);
 
 extern void gfs2_rs_deltree(struct gfs2_blkreserv *rs);
-extern void gfs2_rs_delete(struct gfs2_inode *ip, atomic_t *wcount);
+extern void gfs2_rs_delete(struct gfs2_inode *ip);
 extern void __gfs2_free_blocks(struct gfs2_inode *ip, struct gfs2_rgrpd *rgd,
 			       u64 bstart, u32 blen, int meta);
 extern void gfs2_free_meta(struct gfs2_inode *ip, struct gfs2_rgrpd *rgd,
fs/gfs2/super.c +1 -1
@@ -1396,7 +1396,7 @@ out:
 	truncate_inode_pages_final(&inode->i_data);
 	if (ip->i_qadata)
 		gfs2_assert_warn(sdp, ip->i_qadata->qa_ref == 0);
-	gfs2_rs_delete(ip, NULL);
+	gfs2_rs_deltree(&ip->i_res);
 	gfs2_ordered_del_inode(ip);
 	clear_inode(inode);
 	gfs2_dir_hash_inval(ip);
author maintainer
#1034 lore raw

PATCH 1030/1126 gfs2: Fix gfs2_file_buffered_write endless loop workaround

In reply to Greg Kroah-Hartman

From: Andreas Gruenbacher <[email protected]>

commit 46f3e0421ccb5474b5c006b0089b9dfd42534bb6 upstream.

Since commit 554c577cee95b, gfs2_file_buffered_write() can accidentally return a truncated iov_iter, which might confuse callers. Fix that.

Fixes 554c577cee95b ("gfs2: Prevent endless loops in gfs2_file_buffered_write")
---
 fs/gfs2/file.c |    1 +
 1 file changed, 1 insertion(+)
fs/gfs2/file.c +1 -0
@@ -1083,6 +1083,7 @@ out_uninit:
 	gfs2_holder_uninit(gh);
 	if (statfs_gh)
 		kfree(statfs_gh);
+	from->count = orig_count - read;
 	return read ? read : ret;
 }
 
author maintainer
#1035 lore raw

PATCH 1031/1126 gfs2: Make sure FITRIM minlen is rounded up to fs block size

In reply to Greg Kroah-Hartman

From: Andrew Price <[email protected]>

commit 27ca8273fda398638ca994a207323a85b6d81190 upstream.

Per fstrim(8) we must round up the minlen argument to the fs block size. The current calculation doesn't take into account devices that have a discard granularity and requested minlen less than 1 fs block, so the value can get shifted away to zero in the translation to fs blocks.

The zero minlen passed to gfs2_rgrp_send_discards() then allows sb_issue_discard() to be called with nr_sects == 0 which returns -EINVAL and results in gfs2_rgrp_send_discards() returning -EIO.

Make sure minlen is never < 1 fs block by taking the max of the requested minlen and the fs block size before comparing to the device's discard granularity and shifting to fs blocks.

Fixes 076f0faa764ab ("GFS2: Fix FITRIM argument handling")
---
 fs/gfs2/rgrp.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
fs/gfs2/rgrp.c +2 -1
@@ -1416,7 +1416,8 @@ int gfs2_fitrim(struct file *filp, void
 
 	start = r.start >> bs_shift;
 	end = start + (r.len >> bs_shift);
-	minlen = max_t(u64, r.minlen,
+	minlen = max_t(u64, r.minlen, sdp->sd_sb.sb_bsize);
+	minlen = max_t(u64, minlen,
 		       q->limits.discard_granularity) >> bs_shift;
 
 	if (end <= start || minlen > sdp->sd_max_rg_data)
author maintainer
#1036 lore raw

PATCH 1032/1126 net: hns3: fix the concurrency between functions reading debugfs

In reply to Greg Kroah-Hartman

From: Yufeng Mo <[email protected]>

commit 9c9a04212fa380d2e7d1412bb281309955c0a781 upstream.

Currently, the debugfs mechanism is that all functions share a global variable to save the pointer for obtaining data. When different functions concurrently access the same file node, repeated release exceptions occur. Therefore, the granularity of the pointer for storing the obtained data is adjusted to be private for each function.

Fixes 5e69ea7ee2a6 ("net: hns3: refactor the debugfs process")
---
 drivers/net/ethernet/hisilicon/hns3/hnae3.h        |    1 +
 drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c |   15 +++++++++++----
 drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.h |    1 -
 3 files changed, 12 insertions(+), 5 deletions(-)
drivers/net/ethernet/hisilicon/hns3/hnae3.h +1 -0
@@ -844,6 +844,7 @@ struct hnae3_handle {
 	struct dentry *hnae3_dbgfs;
 	/* protects concurrent contention between debugfs commands */
 	struct mutex dbgfs_lock;
+	char **dbgfs_buf;
 
 	/* Network interface message level enabled bits */
 	u32 msg_enable;
drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.c +11 -4
@@ -1227,7 +1227,7 @@ static ssize_t hns3_dbg_read(struct file
 		return ret;
 
 	mutex_lock(&handle->dbgfs_lock);
-	save_buf = &hns3_dbg_cmd[index].buf;
+	save_buf = &handle->dbgfs_buf[index];
 
 	if (!test_bit(HNS3_NIC_STATE_INITED, &priv->state) ||
 	    test_bit(HNS3_NIC_STATE_RESETTING, &priv->state)) {
@@ -1332,6 +1332,13 @@ int hns3_dbg_init(struct hnae3_handle *h
 	int ret;
 	u32 i;
 
+	handle->dbgfs_buf = devm_kcalloc(&handle->pdev->dev,
+					 ARRAY_SIZE(hns3_dbg_cmd),
+					 sizeof(*handle->dbgfs_buf),
+					 GFP_KERNEL);
+	if (!handle->dbgfs_buf)
+		return -ENOMEM;
+
 	hns3_dbg_dentry[HNS3_DBG_DENTRY_COMMON].dentry =
 				debugfs_create_dir(name, hns3_dbgfs_root);
 	handle->hnae3_dbgfs = hns3_dbg_dentry[HNS3_DBG_DENTRY_COMMON].dentry;
@@ -1380,9 +1387,9 @@ void hns3_dbg_uninit(struct hnae3_handle
 	u32 i;
 
 	for (i = 0; i < ARRAY_SIZE(hns3_dbg_cmd); i++)
-		if (hns3_dbg_cmd[i].buf) {
-			kvfree(hns3_dbg_cmd[i].buf);
-			hns3_dbg_cmd[i].buf = NULL;
+		if (handle->dbgfs_buf[i]) {
+			kvfree(handle->dbgfs_buf[i]);
+			handle->dbgfs_buf[i] = NULL;
 		}
 
 	mutex_destroy(&handle->dbgfs_lock);
drivers/net/ethernet/hisilicon/hns3/hns3_debugfs.h +0 -1
@@ -49,7 +49,6 @@ struct hns3_dbg_cmd_info {
 	enum hnae3_dbg_cmd cmd;
 	enum hns3_dbg_dentry_type dentry;
 	u32 buf_len;
-	char *buf;
 	int (*init)(struct hnae3_handle *handle, unsigned int cmd);
 };
 
author maintainer
#1037 lore raw

PATCH 1033/1126 net: hns3: fix software vlan talbe of vlan 0 inconsistent with hardware

In reply to Greg Kroah-Hartman

From: Guangbin Huang <[email protected]>

commit 7ed258f12ec5ce855f15cdfb5710361dc82fe899 upstream.

When user delete vlan 0, as driver will not delete vlan 0 for hardware in function hclge_set_vlan_filter_hw(), so vlan 0 in software vlan talbe should not be deleted.

Fixes fe4144d47eef ("net: hns3: sync VLAN filter entries when kill VLAN ID failed")
---
 drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_main.c +3 -3
@@ -10314,11 +10314,11 @@ int hclge_set_vlan_filter(struct hnae3_h
 	}
 
 	if (!ret) {
-		if (is_kill)
-			hclge_rm_vport_vlan_table(vport, vlan_id, false);
-		else
+		if (!is_kill)
 			hclge_add_vport_vlan_table(vport, vlan_id,
 						   writen_to_tbl);
+		else if (is_kill && vlan_id != 0)
+			hclge_rm_vport_vlan_table(vport, vlan_id, false);
 	} else if (is_kill) {
 		/* when remove hw vlan filter failed, record the vlan id,
 		 * and try to remove it from hw later, to be consistence
author maintainer
#1038 lore raw

PATCH 1034/1126 rxrpc: fix some null-ptr-deref bugs in server_key.c

In reply to Greg Kroah-Hartman

From: Xiaolong Huang <[email protected]>

commit ff8376ade4f668130385839cef586a0990f8ef87 upstream.

Some function calls are not implemented in rxrpc_no_security, there are preparse_server_key, free_preparse_server_key and destroy_server_key. When rxrpc security type is rxrpc_no_security, user can easily trigger a null-ptr-deref bug via ioctl. So judgment should be added to prevent it

The crash log:

user@syzkaller:~$ ./rxrpc_preparse_s
[   37.956878][T15626] BUG: kernel NULL pointer dereference, address: 0000000000000000
[   37.957645][T15626] #PF: supervisor instruction fetch in kernel mode
[   37.958229][T15626] #PF: error_code(0x0010) - not-present page
[   37.958762][T15626] PGD 4aadf067 P4D 4aadf067 PUD 4aade067 PMD 0
[   37.959321][T15626] Oops: 0010 [#1] PREEMPT SMP
[   37.959739][T15626] CPU: 0 PID: 15626 Comm: rxrpc_preparse_ Not tainted 5.17.0-01442-gb47d5a4f6b8d #43
[   37.960588][T15626] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.13.0-1ubuntu1 04/01/2014
[   37.961474][T15626] RIP: 0010:0x0
[   37.961787][T15626] Code: Unable to access opcode bytes at RIP 0xffffffffffffffd6.
[   37.962480][T15626] RSP: 0018:ffffc9000d9abdc0 EFLAGS: 00010286
[   37.963018][T15626] RAX: ffffffff84335200 RBX: ffff888012a1ce80 RCX: 0000000000000000
[   37.963727][T15626] RDX: 0000000000000000 RSI: ffffffff84a736dc RDI: ffffc9000d9abe48
[   37.964425][T15626] RBP: ffffc9000d9abe48 R08: 0000000000000000 R09: 0000000000000002
[   37.965118][T15626] R10: 000000000000000a R11: f000000000000000 R12: ffff888013145680
[   37.965836][T15626] R13: 0000000000000000 R14: ffffffffffffffec R15: ffff8880432aba80
[   37.966441][T15626] FS:  00007f2177907700(0000) GS:ffff88803ec00000(0000) knlGS:0000000000000000
[   37.966979][T15626] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   37.967384][T15626] CR2: ffffffffffffffd6 CR3: 000000004aaf1000 CR4: 00000000000006f0
[   37.967864][T15626] Call Trace:
[   37.968062][T15626]  <TASK>
[   37.968240][T15626]  rxrpc_preparse_s+0x59/0x90
[   37.968541][T15626]  key_create_or_update+0x174/0x510
[   37.968863][T15626]  __x64_sys_add_key+0x139/0x1d0
[   37.969165][T15626]  do_syscall_64+0x35/0xb0
[   37.969451][T15626]  entry_SYSCALL_64_after_hwframe+0x44/0xae
[   37.969824][T15626] RIP: 0033:0x43a1f9
---
 net/rxrpc/server_key.c |    7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)
net/rxrpc/server_key.c +5 -2
@@ -84,6 +84,9 @@ static int rxrpc_preparse_s(struct key_p
 
 	prep->payload.data[1] = (struct rxrpc_security *)sec;
 
+	if (!sec->preparse_server_key)
+		return -EINVAL;
+
 	return sec->preparse_server_key(prep);
 }
 
@@ -91,7 +94,7 @@ static void rxrpc_free_preparse_s(struct
 {
 	const struct rxrpc_security *sec = prep->payload.data[1];
 
-	if (sec)
+	if (sec && sec->free_preparse_server_key)
 		sec->free_preparse_server_key(prep);
 }
 
@@ -99,7 +102,7 @@ static void rxrpc_destroy_s(struct key *
 {
 	const struct rxrpc_security *sec = key->payload.data[1];
 
-	if (sec)
+	if (sec && sec->destroy_server_key)
 		sec->destroy_server_key(key);
 }
 
author maintainer
#1039 lore raw

PATCH 1035/1126 rxrpc: Fix call timer start racing with call destruction

In reply to Greg Kroah-Hartman

From: David Howells <[email protected]>

commit 4a7f62f91933c8ae5308f9127fd8ea48188b6bc3 upstream.

The rxrpc_call struct has a timer used to handle various timed events relating to a call. This timer can get started from the packet input routines that are run in softirq mode with just the RCU read lock held. Unfortunately, because only the RCU read lock is held - and neither ref or other lock is taken - the call can start getting destroyed at the same time a packet comes in addressed to that call. This causes the timer - which was already stopped - to get restarted. Later, the timer dispatch code may then oops if the timer got deallocated first.

Fix this by trying to take a ref on the rxrpc_call struct and, if successful, passing that ref along to the timer. If the timer was already running, the ref is discarded.

The timer completion routine can then pass the ref along to the call's work item when it queues it. If the timer or work item where already queued/running, the extra ref is discarded.

---
 include/trace/events/rxrpc.h |    8 +++++++-
 net/rxrpc/ar-internal.h      |   15 +++++++--------
 net/rxrpc/call_event.c       |    2 +-
 net/rxrpc/call_object.c      |   40 +++++++++++++++++++++++++++++++++++-----
 4 files changed, 50 insertions(+), 15 deletions(-)
include/trace/events/rxrpc.h +7 -1
@@ -83,12 +83,15 @@ enum rxrpc_call_trace {
 	rxrpc_call_error,
 	rxrpc_call_got,
 	rxrpc_call_got_kernel,
+	rxrpc_call_got_timer,
 	rxrpc_call_got_userid,
 	rxrpc_call_new_client,
 	rxrpc_call_new_service,
 	rxrpc_call_put,
 	rxrpc_call_put_kernel,
 	rxrpc_call_put_noqueue,
+	rxrpc_call_put_notimer,
+	rxrpc_call_put_timer,
 	rxrpc_call_put_userid,
 	rxrpc_call_queued,
 	rxrpc_call_queued_ref,
@@ -278,12 +281,15 @@ enum rxrpc_tx_point {
 	EM(rxrpc_call_error,			"*E*") \
 	EM(rxrpc_call_got,			"GOT") \
 	EM(rxrpc_call_got_kernel,		"Gke") \
+	EM(rxrpc_call_got_timer,		"GTM") \
 	EM(rxrpc_call_got_userid,		"Gus") \
 	EM(rxrpc_call_new_client,		"NWc") \
 	EM(rxrpc_call_new_service,		"NWs") \
 	EM(rxrpc_call_put,			"PUT") \
 	EM(rxrpc_call_put_kernel,		"Pke") \
-	EM(rxrpc_call_put_noqueue,		"PNQ") \
+	EM(rxrpc_call_put_noqueue,		"PnQ") \
+	EM(rxrpc_call_put_notimer,		"PnT") \
+	EM(rxrpc_call_put_timer,		"PTM") \
 	EM(rxrpc_call_put_userid,		"Pus") \
 	EM(rxrpc_call_queued,			"QUE") \
 	EM(rxrpc_call_queued_ref,		"QUR") \
net/rxrpc/ar-internal.h +7 -8
@@ -777,14 +777,12 @@ void rxrpc_propose_ACK(struct rxrpc_call
 		       enum rxrpc_propose_ack_trace);
 void rxrpc_process_call(struct work_struct *);
 
-static inline void rxrpc_reduce_call_timer(struct rxrpc_call *call,
-					   unsigned long expire_at,
-					   unsigned long now,
-					   enum rxrpc_timer_trace why)
-{
-	trace_rxrpc_timer(call, why, now);
-	timer_reduce(&call->timer, expire_at);
-}
+void rxrpc_reduce_call_timer(struct rxrpc_call *call,
+			     unsigned long expire_at,
+			     unsigned long now,
+			     enum rxrpc_timer_trace why);
+
+void rxrpc_delete_call_timer(struct rxrpc_call *call);
 
 /*
  * call_object.c
@@ -808,6 +806,7 @@ void rxrpc_release_calls_on_socket(struc
 bool __rxrpc_queue_call(struct rxrpc_call *);
 bool rxrpc_queue_call(struct rxrpc_call *);
 void rxrpc_see_call(struct rxrpc_call *);
+bool rxrpc_try_get_call(struct rxrpc_call *call, enum rxrpc_call_trace op);
 void rxrpc_get_call(struct rxrpc_call *, enum rxrpc_call_trace);
 void rxrpc_put_call(struct rxrpc_call *, enum rxrpc_call_trace);
 void rxrpc_cleanup_call(struct rxrpc_call *);
net/rxrpc/call_event.c +1 -1
@@ -310,7 +310,7 @@ recheck_state:
 	}
 
 	if (call->state == RXRPC_CALL_COMPLETE) {
-		del_timer_sync(&call->timer);
+		rxrpc_delete_call_timer(call);
 		goto out_put;
 	}
 
net/rxrpc/call_object.c +35 -5
@@ -53,10 +53,30 @@ static void rxrpc_call_timer_expired(str
 
 	if (call->state < RXRPC_CALL_COMPLETE) {
 		trace_rxrpc_timer(call, rxrpc_timer_expired, jiffies);
-		rxrpc_queue_call(call);
+		__rxrpc_queue_call(call);
+	} else {
+		rxrpc_put_call(call, rxrpc_call_put);
+	}
+}
+
+void rxrpc_reduce_call_timer(struct rxrpc_call *call,
+			     unsigned long expire_at,
+			     unsigned long now,
+			     enum rxrpc_timer_trace why)
+{
+	if (rxrpc_try_get_call(call, rxrpc_call_got_timer)) {
+		trace_rxrpc_timer(call, why, now);
+		if (timer_reduce(&call->timer, expire_at))
+			rxrpc_put_call(call, rxrpc_call_put_notimer);
 	}
 }
 
+void rxrpc_delete_call_timer(struct rxrpc_call *call)
+{
+	if (del_timer_sync(&call->timer))
+		rxrpc_put_call(call, rxrpc_call_put_timer);
+}
+
 static struct lock_class_key rxrpc_call_user_mutex_lock_class_key;
 
 /*
@@ -463,6 +483,17 @@ void rxrpc_see_call(struct rxrpc_call *c
 	}
 }
 
+bool rxrpc_try_get_call(struct rxrpc_call *call, enum rxrpc_call_trace op)
+{
+	const void *here = __builtin_return_address(0);
+	int n = atomic_fetch_add_unless(&call->usage, 1, 0);
+
+	if (n == 0)
+		return false;
+	trace_rxrpc_call(call->debug_id, op, n, here, NULL);
+	return true;
+}
+
 /*
  * Note the addition of a ref on a call.
  */
@@ -510,8 +541,7 @@ void rxrpc_release_call(struct rxrpc_soc
 	spin_unlock_bh(&call->lock);
 
 	rxrpc_put_call_slot(call);
-
-	del_timer_sync(&call->timer);
+	rxrpc_delete_call_timer(call);
 
 	/* Make sure we don't get any more notifications */
 	write_lock_bh(&rx->recvmsg_lock);
@@ -618,6 +648,8 @@ static void rxrpc_destroy_call(struct wo
 	struct rxrpc_call *call = container_of(work, struct rxrpc_call, processor);
 	struct rxrpc_net *rxnet = call->rxnet;
 
+	rxrpc_delete_call_timer(call);
+
 	rxrpc_put_connection(call->conn);
 	rxrpc_put_peer(call->peer);
 	kfree(call->rxtx_buffer);
@@ -652,8 +684,6 @@ void rxrpc_cleanup_call(struct rxrpc_cal
 
 	memset(&call->sock_node, 0xcd, sizeof(call->sock_node));
 
-	del_timer_sync(&call->timer);
-
 	ASSERTCMP(call->state, ==, RXRPC_CALL_COMPLETE);
 	ASSERT(test_bit(RXRPC_CALL_RELEASED, &call->flags));
 
author maintainer
#1040 lore raw

PATCH 1036/1126 mailbox: imx: fix wakeup failure from freeze mode

In reply to Greg Kroah-Hartman

From: Robin Gong <[email protected]>

commit 892cb524ae8a27bf5e42f711318371acd9a9f74a upstream.

Since IRQF_NO_SUSPEND used for imx mailbox driver, that means this irq can't be used for wakeup source so that can't wakeup from freeze mode. Add pm_system_wakeup() to wakeup from freeze mode.

Fixes b7b2796b9b31e("mailbox: imx: ONLY IPC MU needs IRQF_NO_SUSPEND flag")
---
 drivers/mailbox/imx-mailbox.c |    9 +++++++++
 1 file changed, 9 insertions(+)
drivers/mailbox/imx-mailbox.c +9 -0
@@ -14,6 +14,7 @@
 #include <linux/module.h>
 #include <linux/of_device.h>
 #include <linux/pm_runtime.h>
+#include <linux/suspend.h>
 #include <linux/slab.h>
 
 #define IMX_MU_CHANS		16
@@ -76,6 +77,7 @@ struct imx_mu_priv {
 	const struct imx_mu_dcfg	*dcfg;
 	struct clk		*clk;
 	int			irq;
+	bool			suspend;
 
 	u32 xcr[4];
 
@@ -334,6 +336,9 @@ static irqreturn_t imx_mu_isr(int irq, v
 		return IRQ_NONE;
 	}
 
+	if (priv->suspend)
+		pm_system_wakeup();
+
 	return IRQ_HANDLED;
 }
 
@@ -702,6 +707,8 @@ static int __maybe_unused imx_mu_suspend
 			priv->xcr[i] = imx_mu_read(priv, priv->dcfg->xCR[i]);
 	}
 
+	priv->suspend = true;
+
 	return 0;
 }
 
@@ -723,6 +730,8 @@ static int __maybe_unused imx_mu_resume_
 			imx_mu_write(priv, priv->xcr[i], priv->dcfg->xCR[i]);
 	}
 
+	priv->suspend = false;
+
 	return 0;
 }
 
author maintainer
#1041 lore raw

PATCH 1037/1126 crypto: x86/poly1305 - Fixup SLS

In reply to Greg Kroah-Hartman

From: Peter Zijlstra <[email protected]>

commit 7ed7aa4de9421229be6d331ed52d5cd09c99f409 upstream.

Due to being a perl generated asm file, it got missed by the mass convertion script.

arch/x86/crypto/poly1305-x86_64-cryptogams.o: warning: objtool: poly1305_init_x86_64()+0x3a: missing int3 after ret arch/x86/crypto/poly1305-x86_64-cryptogams.o: warning: objtool: poly1305_blocks_x86_64()+0xf2: missing int3 after ret arch/x86/crypto/poly1305-x86_64-cryptogams.o: warning: objtool: poly1305_emit_x86_64()+0x37: missing int3 after ret arch/x86/crypto/poly1305-x86_64-cryptogams.o: warning: objtool: __poly1305_block()+0x6d: missing int3 after ret arch/x86/crypto/poly1305-x86_64-cryptogams.o: warning: objtool: __poly1305_init_avx()+0x1e8: missing int3 after ret arch/x86/crypto/poly1305-x86_64-cryptogams.o: warning: objtool: poly1305_blocks_avx()+0x18a: missing int3 after ret arch/x86/crypto/poly1305-x86_64-cryptogams.o: warning: objtool: poly1305_blocks_avx()+0xaf8: missing int3 after ret arch/x86/crypto/poly1305-x86_64-cryptogams.o: warning: objtool: poly1305_emit_avx()+0x99: missing int3 after ret arch/x86/crypto/poly1305-x86_64-cryptogams.o: warning: objtool: poly1305_blocks_avx2()+0x18a: missing int3 after ret arch/x86/crypto/poly1305-x86_64-cryptogams.o: warning: objtool: poly1305_blocks_avx2()+0x776: missing int3 after ret arch/x86/crypto/poly1305-x86_64-cryptogams.o: warning: objtool: poly1305_blocks_avx512()+0x18a: missing int3 after ret arch/x86/crypto/poly1305-x86_64-cryptogams.o: warning: objtool: poly1305_blocks_avx512()+0x796: missing int3 after ret arch/x86/crypto/poly1305-x86_64-cryptogams.o: warning: objtool: poly1305_blocks_avx512()+0x10bd: missing int3 after ret

Fixes f94909ceb1ed ("x86: Prepare asm files for straight-line-speculation")
---
 arch/x86/crypto/poly1305-x86_64-cryptogams.pl |   38 +++++++++++++-------------
 1 file changed, 19 insertions(+), 19 deletions(-)
arch/x86/crypto/poly1305-x86_64-cryptogams.pl +19 -19
@@ -297,7 +297,7 @@ ___
 $code.=<<___;
 	mov	\$1,%eax
 .Lno_key:
-	ret
+	RET
 ___
 &end_function("poly1305_init_x86_64");
 
@@ -373,7 +373,7 @@ $code.=<<___;
 .cfi_adjust_cfa_offset	-48
 .Lno_data:
 .Lblocks_epilogue:
-	ret
+	RET
 .cfi_endproc
 ___
 &end_function("poly1305_blocks_x86_64");
@@ -399,7 +399,7 @@ $code.=<<___;
 	mov	%rax,0($mac)	# write result
 	mov	%rcx,8($mac)
 
-	ret
+	RET
 ___
 &end_function("poly1305_emit_x86_64");
 if ($avx) {
@@ -429,7 +429,7 @@ ___
 	&poly1305_iteration();
 $code.=<<___;
 	pop $ctx
-	ret
+	RET
 .size	__poly1305_block,.-__poly1305_block
 
 .type	__poly1305_init_avx,\@abi-omnipotent
@@ -594,7 +594,7 @@ __poly1305_init_avx:
 
 	lea	-48-64($ctx),$ctx	# size [de-]optimization
 	pop %rbp
-	ret
+	RET
 .size	__poly1305_init_avx,.-__poly1305_init_avx
 ___
 
@@ -747,7 +747,7 @@ $code.=<<___;
 .cfi_restore	%rbp
 .Lno_data_avx:
 .Lblocks_avx_epilogue:
-	ret
+	RET
 .cfi_endproc
 
 .align	32
@@ -1452,7 +1452,7 @@ $code.=<<___	if (!$win64);
 ___
 $code.=<<___;
 	vzeroupper
-	ret
+	RET
 .cfi_endproc
 ___
 &end_function("poly1305_blocks_avx");
@@ -1508,7 +1508,7 @@ $code.=<<___;
 	mov	%rax,0($mac)	# write result
 	mov	%rcx,8($mac)
 
-	ret
+	RET
 ___
 &end_function("poly1305_emit_avx");
 
@@ -1675,7 +1675,7 @@ $code.=<<___;
 .cfi_restore 	%rbp
 .Lno_data_avx2$suffix:
 .Lblocks_avx2_epilogue$suffix:
-	ret
+	RET
 .cfi_endproc
 
 .align	32
@@ -2201,7 +2201,7 @@ $code.=<<___	if (!$win64);
 ___
 $code.=<<___;
 	vzeroupper
-	ret
+	RET
 .cfi_endproc
 ___
 if($avx > 2 && $avx512) {
@@ -2792,7 +2792,7 @@ $code.=<<___	if (!$win64);
 .cfi_def_cfa_register	%rsp
 ___
 $code.=<<___;
-	ret
+	RET
 .cfi_endproc
 ___
 
@@ -2893,7 +2893,7 @@ $code.=<<___	if ($flavour =~ /elf32/);
 ___
 $code.=<<___;
 	mov	\$1,%eax
-	ret
+	RET
 .size	poly1305_init_base2_44,.-poly1305_init_base2_44
 ___
 {
@@ -3010,7 +3010,7 @@ poly1305_blocks_vpmadd52:
 	jnz		.Lblocks_vpmadd52_4x
 
 .Lno_data_vpmadd52:
-	ret
+	RET
 .size	poly1305_blocks_vpmadd52,.-poly1305_blocks_vpmadd52
 ___
 }
@@ -3451,7 +3451,7 @@ poly1305_blocks_vpmadd52_4x:
 	vzeroall
 
 .Lno_data_vpmadd52_4x:
-	ret
+	RET
 .size	poly1305_blocks_vpmadd52_4x,.-poly1305_blocks_vpmadd52_4x
 ___
 }
@@ -3824,7 +3824,7 @@ $code.=<<___;
 	vzeroall
 
 .Lno_data_vpmadd52_8x:
-	ret
+	RET
 .size	poly1305_blocks_vpmadd52_8x,.-poly1305_blocks_vpmadd52_8x
 ___
 }
@@ -3861,7 +3861,7 @@ poly1305_emit_base2_44:
 	mov	%rax,0($mac)	# write result
 	mov	%rcx,8($mac)
 
-	ret
+	RET
 .size	poly1305_emit_base2_44,.-poly1305_emit_base2_44
 ___
 }	}	}
@@ -3916,7 +3916,7 @@ xor128_encrypt_n_pad:
 
 .Ldone_enc:
 	mov	$otp,%rax
-	ret
+	RET
 .size	xor128_encrypt_n_pad,.-xor128_encrypt_n_pad
 
 .globl	xor128_decrypt_n_pad
@@ -3967,7 +3967,7 @@ xor128_decrypt_n_pad:
 
 .Ldone_dec:
 	mov	$otp,%rax
-	ret
+	RET
 .size	xor128_decrypt_n_pad,.-xor128_decrypt_n_pad
 ___
 }
@@ -4109,7 +4109,7 @@ avx_handler:
 	pop	%rbx
 	pop	%rdi
 	pop	%rsi
-	ret
+	RET
 .size	avx_handler,.-avx_handler
 
 .section	.pdata
author maintainer
#1042 lore raw

PATCH 1038/1126 crypto: arm/aes-neonbs-cbc - Select generic cbc and aes

In reply to Greg Kroah-Hartman

From: Herbert Xu <[email protected]>

commit c8bd296cca3434b13b28b074eaeb78a23284de77 upstream.

The algorithm __cbc-aes-neonbs requires a fallback so we need to select the config options for them or otherwise it will fail to register on boot-up.

Fixes 00b99ad2bac2 ("crypto: arm/aes-neonbs - Use generic cbc...")
---
 arch/arm/crypto/Kconfig |    2 ++
 1 file changed, 2 insertions(+)
arch/arm/crypto/Kconfig +2 -0
@@ -102,6 +102,8 @@ config CRYPTO_AES_ARM_BS
 	depends on KERNEL_MODE_NEON
 	select CRYPTO_SKCIPHER
 	select CRYPTO_LIB_AES
+	select CRYPTO_AES
+	select CRYPTO_CBC
 	select CRYPTO_SIMD
 	help
 	  Use a faster and more secure NEON based implementation of AES in CBC,
author maintainer
#1043 lore raw

PATCH 1039/1126 watch_queue: Free the page array when watch_queue is dismantled

In reply to Greg Kroah-Hartman

From: Eric Dumazet <[email protected]>

commit b490207017ba237d97b735b2aa66dc241ccd18f5 upstream.

Commit 7ea1a0124b6d ("watch_queue: Free the alloc bitmap when the watch_queue is torn down") took care of the bitmap, but not the page array.

  BUG: memory leak
  unreferenced object 0xffff88810d9bc140 (size 32):
  comm "syz-executor335", pid 3603, jiffies 4294946994 (age 12.840s)
  hex dump (first 32 bytes):
    40 a7 40 04 00 ea ff ff 00 00 00 00 00 00 00 00  @.@.............
    00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
  backtrace:
     kmalloc_array include/linux/slab.h:621 [inline]
     kcalloc include/linux/slab.h:652 [inline]
     watch_queue_set_size+0x12f/0x2e0 kernel/watch_queue.c:251
     pipe_ioctl+0x82/0x140 fs/pipe.c:632
     vfs_ioctl fs/ioctl.c:51 [inline]
     __do_sys_ioctl fs/ioctl.c:874 [inline]
     __se_sys_ioctl fs/ioctl.c:860 [inline]
     __x64_sys_ioctl+0xfc/0x140 fs/ioctl.c:860
     do_syscall_x64 arch/x86/entry/common.c:50 [inline]
Reported-by [email protected]
Fixes c73be61cede5 ("pipe: Add general notification queue support")
Cc Jann Horn <[email protected]>
---
 kernel/watch_queue.c |    1 +
 1 file changed, 1 insertion(+)
kernel/watch_queue.c +1 -0
@@ -370,6 +370,7 @@ static void __put_watch_queue(struct kre
 
 	for (i = 0; i < wqueue->nr_pages; i++)
 		__free_page(wqueue->notes[i]);
+	kfree(wqueue->notes);
 	bitmap_free(wqueue->notes_bitmap);
 
 	wfilter = rcu_access_pointer(wqueue->filter);
author maintainer
#1044 lore raw

PATCH 1040/1126 pinctrl: pinconf-generic: Print arguments for bias-pull-*

In reply to Greg Kroah-Hartman

From: Chen-Yu Tsai <[email protected]>

commit 188e5834b930acd03ad3cf7c5e7aa24db9665a29 upstream.

The bias-pull-* properties, or PIN_CONFIG_BIAS_PULL_* pin config parameters, accept optional arguments in ohms denoting the strength of the pin bias.

Print these values out in debugfs as well.

---
 drivers/pinctrl/pinconf-generic.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/pinctrl/pinconf-generic.c +3 -3
@@ -30,10 +30,10 @@ static const struct pin_config_item conf
 	PCONFDUMP(PIN_CONFIG_BIAS_BUS_HOLD, "input bias bus hold", NULL, false),
 	PCONFDUMP(PIN_CONFIG_BIAS_DISABLE, "input bias disabled", NULL, false),
 	PCONFDUMP(PIN_CONFIG_BIAS_HIGH_IMPEDANCE, "input bias high impedance", NULL, false),
-	PCONFDUMP(PIN_CONFIG_BIAS_PULL_DOWN, "input bias pull down", NULL, false),
+	PCONFDUMP(PIN_CONFIG_BIAS_PULL_DOWN, "input bias pull down", "ohms", true),
 	PCONFDUMP(PIN_CONFIG_BIAS_PULL_PIN_DEFAULT,
-				"input bias pull to pin specific state", NULL, false),
-	PCONFDUMP(PIN_CONFIG_BIAS_PULL_UP, "input bias pull up", NULL, false),
+				"input bias pull to pin specific state", "ohms", true),
+	PCONFDUMP(PIN_CONFIG_BIAS_PULL_UP, "input bias pull up", "ohms", true),
 	PCONFDUMP(PIN_CONFIG_DRIVE_OPEN_DRAIN, "output drive open drain", NULL, false),
 	PCONFDUMP(PIN_CONFIG_DRIVE_OPEN_SOURCE, "output drive open source", NULL, false),
 	PCONFDUMP(PIN_CONFIG_DRIVE_PUSH_PULL, "output drive push pull", NULL, false),
author maintainer
#1045 lore raw

PATCH 1041/1126 watchdog: rti-wdt: Add missing pm_runtime_disable() in probe function

In reply to Greg Kroah-Hartman

From: Miaoqian Lin <[email protected]>

commit d055ef3a2c6919cff504ae3b710c96318d545fd2 upstream.

If the probe fails, we should use pm_runtime_disable() to balance pm_runtime_enable().

---
 drivers/watchdog/rti_wdt.c |    1 +
 1 file changed, 1 insertion(+)
drivers/watchdog/rti_wdt.c +1 -0
@@ -228,6 +228,7 @@ static int rti_wdt_probe(struct platform
 	ret = pm_runtime_get_sync(dev);
 	if (ret) {
 		pm_runtime_put_noidle(dev);
+		pm_runtime_disable(&pdev->dev);
 		return dev_err_probe(dev, ret, "runtime pm failed\n");
 	}
 
author maintainer
#1046 lore raw

PATCH 1042/1126 net: sparx5: uses, depends on BRIDGE or !BRIDGE

In reply to Greg Kroah-Hartman

From: Randy Dunlap <[email protected]>

commit f9512d654f62604664251dedd437a22fe484974a upstream.

Fix build errors when BRIDGE=m and SPARX5_SWITCH=y:

riscv64-linux-ld: drivers/net/ethernet/microchip/sparx5/sparx5_switchdev.o: in function `.L305':
sparx5_switchdev.c:(.text+0xdb0): undefined reference to `br_vlan_enabled'
riscv64-linux-ld: drivers/net/ethernet/microchip/sparx5/sparx5_switchdev.o: in function `.L283':
sparx5_switchdev.c:(.text+0xee0): undefined reference to `br_vlan_enabled'

Fixes 3cfa11bac9bb ("net: sparx5: add the basic sparx5 driver")
Cc Horatiu Vultur <[email protected]>
Cc Lars Povlsen <[email protected]>
Cc Steen Hegelund <[email protected]>
Cc Paolo Abeni <[email protected]>
---
 drivers/net/ethernet/microchip/sparx5/Kconfig |    1 +
 1 file changed, 1 insertion(+)
drivers/net/ethernet/microchip/sparx5/Kconfig +1 -0
@@ -5,6 +5,7 @@ config SPARX5_SWITCH
 	depends on OF
 	depends on ARCH_SPARX5 || COMPILE_TEST
 	depends on PTP_1588_CLOCK_OPTIONAL
+	depends on BRIDGE || BRIDGE=n
 	select PHYLINK
 	select PHY_SPARX5_SERDES
 	select RESET_CONTROLLER
author maintainer
#1047 lore raw

PATCH 1043/1126 pinctrl: nuvoton: npcm7xx: Rename DS() macro to DSTR()

In reply to Greg Kroah-Hartman

From: Jonathan Neuschäfer <[email protected]>

commit 603501c16431c56f74eaef1ee1390f60a30c2187 upstream.

The name "DS" is defined in arch/x86/um/shared/sysdep/ptrace_64.h, which results in a compiler warning when build-testing on ARCH=um. Rename this driver's "DS" macro to DSTR so avoid this collision.

---
 drivers/pinctrl/nuvoton/pinctrl-npcm7xx.c |  156 +++++++++++++++---------------
 1 file changed, 78 insertions(+), 78 deletions(-)
drivers/pinctrl/nuvoton/pinctrl-npcm7xx.c +78 -78
@@ -904,7 +904,7 @@ static struct npcm7xx_func npcm7xx_funcs
 #define DRIVE_STRENGTH_HI_SHIFT		12
 #define DRIVE_STRENGTH_MASK		0x0000FF00
 
-#define DS(lo, hi)	(((lo) << DRIVE_STRENGTH_LO_SHIFT) | \
+#define DSTR(lo, hi)	(((lo) << DRIVE_STRENGTH_LO_SHIFT) | \
 			 ((hi) << DRIVE_STRENGTH_HI_SHIFT))
 #define DSLO(x)		(((x) >> DRIVE_STRENGTH_LO_SHIFT) & 0xF)
 #define DSHI(x)		(((x) >> DRIVE_STRENGTH_HI_SHIFT) & 0xF)
@@ -924,31 +924,31 @@ struct npcm7xx_pincfg {
 static const struct npcm7xx_pincfg pincfg[] = {
 	/*		PIN	  FUNCTION 1		   FUNCTION 2		  FUNCTION 3	    FLAGS */
 	NPCM7XX_PINCFG(0,	 iox1, MFSEL1, 30,	  none, NONE, 0,	none, NONE, 0,	     0),
-	NPCM7XX_PINCFG(1,	 iox1, MFSEL1, 30,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12)),
-	NPCM7XX_PINCFG(2,	 iox1, MFSEL1, 30,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12)),
+	NPCM7XX_PINCFG(1,	 iox1, MFSEL1, 30,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12)),
+	NPCM7XX_PINCFG(2,	 iox1, MFSEL1, 30,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12)),
 	NPCM7XX_PINCFG(3,	 iox1, MFSEL1, 30,	  none, NONE, 0,	none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(4,	 iox2, MFSEL3, 14,	 smb1d, I2CSEGSEL, 7,	none, NONE, 0,	     SLEW),
 	NPCM7XX_PINCFG(5,	 iox2, MFSEL3, 14,	 smb1d, I2CSEGSEL, 7,	none, NONE, 0,	     SLEW),
 	NPCM7XX_PINCFG(6,	 iox2, MFSEL3, 14,	 smb2d, I2CSEGSEL, 10,  none, NONE, 0,       SLEW),
 	NPCM7XX_PINCFG(7,	 iox2, MFSEL3, 14,	 smb2d, I2CSEGSEL, 10,  none, NONE, 0,       SLEW),
-	NPCM7XX_PINCFG(8,      lkgpo1, FLOCKR1, 4,        none, NONE, 0,	none, NONE, 0,	     DS(8, 12)),
-	NPCM7XX_PINCFG(9,      lkgpo2, FLOCKR1, 8,        none, NONE, 0,	none, NONE, 0,	     DS(8, 12)),
-	NPCM7XX_PINCFG(10,	 ioxh, MFSEL3, 18,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12)),
-	NPCM7XX_PINCFG(11,	 ioxh, MFSEL3, 18,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12)),
+	NPCM7XX_PINCFG(8,      lkgpo1, FLOCKR1, 4,        none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12)),
+	NPCM7XX_PINCFG(9,      lkgpo2, FLOCKR1, 8,        none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12)),
+	NPCM7XX_PINCFG(10,	 ioxh, MFSEL3, 18,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12)),
+	NPCM7XX_PINCFG(11,	 ioxh, MFSEL3, 18,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12)),
 	NPCM7XX_PINCFG(12,	 gspi, MFSEL1, 24,	 smb5b, I2CSEGSEL, 19,  none, NONE, 0,	     SLEW),
 	NPCM7XX_PINCFG(13,	 gspi, MFSEL1, 24,	 smb5b, I2CSEGSEL, 19,  none, NONE, 0,	     SLEW),
 	NPCM7XX_PINCFG(14,	 gspi, MFSEL1, 24,	 smb5c, I2CSEGSEL, 20,	none, NONE, 0,	     SLEW),
 	NPCM7XX_PINCFG(15,	 gspi, MFSEL1, 24,	 smb5c, I2CSEGSEL, 20,	none, NONE, 0,	     SLEW),
-	NPCM7XX_PINCFG(16,     lkgpo0, FLOCKR1, 0,        none, NONE, 0,	none, NONE, 0,	     DS(8, 12)),
-	NPCM7XX_PINCFG(17,      pspi2, MFSEL3, 13,     smb4den, I2CSEGSEL, 23,  none, NONE, 0,       DS(8, 12)),
-	NPCM7XX_PINCFG(18,      pspi2, MFSEL3, 13,	 smb4b, I2CSEGSEL, 14,  none, NONE, 0,	     DS(8, 12)),
-	NPCM7XX_PINCFG(19,      pspi2, MFSEL3, 13,	 smb4b, I2CSEGSEL, 14,  none, NONE, 0,	     DS(8, 12)),
+	NPCM7XX_PINCFG(16,     lkgpo0, FLOCKR1, 0,        none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12)),
+	NPCM7XX_PINCFG(17,      pspi2, MFSEL3, 13,     smb4den, I2CSEGSEL, 23,  none, NONE, 0,       DSTR(8, 12)),
+	NPCM7XX_PINCFG(18,      pspi2, MFSEL3, 13,	 smb4b, I2CSEGSEL, 14,  none, NONE, 0,	     DSTR(8, 12)),
+	NPCM7XX_PINCFG(19,      pspi2, MFSEL3, 13,	 smb4b, I2CSEGSEL, 14,  none, NONE, 0,	     DSTR(8, 12)),
 	NPCM7XX_PINCFG(20,	smb4c, I2CSEGSEL, 15,    smb15, MFSEL3, 8,      none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(21,	smb4c, I2CSEGSEL, 15,    smb15, MFSEL3, 8,      none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(22,      smb4d, I2CSEGSEL, 16,	 smb14, MFSEL3, 7,      none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(23,      smb4d, I2CSEGSEL, 16,	 smb14, MFSEL3, 7,      none, NONE, 0,	     0),
-	NPCM7XX_PINCFG(24,	 ioxh, MFSEL3, 18,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12)),
-	NPCM7XX_PINCFG(25,	 ioxh, MFSEL3, 18,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12)),
+	NPCM7XX_PINCFG(24,	 ioxh, MFSEL3, 18,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12)),
+	NPCM7XX_PINCFG(25,	 ioxh, MFSEL3, 18,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12)),
 	NPCM7XX_PINCFG(26,	 smb5, MFSEL1, 2,	  none, NONE, 0,	none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(27,	 smb5, MFSEL1, 2,	  none, NONE, 0,	none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(28,	 smb4, MFSEL1, 1,	  none, NONE, 0,	none, NONE, 0,	     0),
@@ -964,12 +964,12 @@ static const struct npcm7xx_pincfg pincf
 	NPCM7XX_PINCFG(39,	smb3b, I2CSEGSEL, 11,	  none, NONE, 0,	none, NONE, 0,	     SLEW),
 	NPCM7XX_PINCFG(40,	smb3b, I2CSEGSEL, 11,	  none, NONE, 0,	none, NONE, 0,	     SLEW),
 	NPCM7XX_PINCFG(41,  bmcuart0a, MFSEL1, 9,         none, NONE, 0,	none, NONE, 0,	     0),
-	NPCM7XX_PINCFG(42,  bmcuart0a, MFSEL1, 9,         none, NONE, 0,	none, NONE, 0,	     DS(2, 4) | GPO),
+	NPCM7XX_PINCFG(42,  bmcuart0a, MFSEL1, 9,         none, NONE, 0,	none, NONE, 0,	     DSTR(2, 4) | GPO),
 	NPCM7XX_PINCFG(43,      uart1, MFSEL1, 10,	 jtag2, MFSEL4, 0,  bmcuart1, MFSEL3, 24,    0),
 	NPCM7XX_PINCFG(44,      uart1, MFSEL1, 10,	 jtag2, MFSEL4, 0,  bmcuart1, MFSEL3, 24,    0),
 	NPCM7XX_PINCFG(45,      uart1, MFSEL1, 10,	 jtag2, MFSEL4, 0,	none, NONE, 0,	     0),
-	NPCM7XX_PINCFG(46,      uart1, MFSEL1, 10,	 jtag2, MFSEL4, 0,	none, NONE, 0,	     DS(2, 8)),
-	NPCM7XX_PINCFG(47,      uart1, MFSEL1, 10,	 jtag2, MFSEL4, 0,	none, NONE, 0,	     DS(2, 8)),
+	NPCM7XX_PINCFG(46,      uart1, MFSEL1, 10,	 jtag2, MFSEL4, 0,	none, NONE, 0,	     DSTR(2, 8)),
+	NPCM7XX_PINCFG(47,      uart1, MFSEL1, 10,	 jtag2, MFSEL4, 0,	none, NONE, 0,	     DSTR(2, 8)),
 	NPCM7XX_PINCFG(48,	uart2, MFSEL1, 11,   bmcuart0b, MFSEL4, 1,      none, NONE, 0,	     GPO),
 	NPCM7XX_PINCFG(49,	uart2, MFSEL1, 11,   bmcuart0b, MFSEL4, 1,      none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(50,	uart2, MFSEL1, 11,	  none, NONE, 0,        none, NONE, 0,	     0),
@@ -979,8 +979,8 @@ static const struct npcm7xx_pincfg pincf
 	NPCM7XX_PINCFG(54,	uart2, MFSEL1, 11,	  none, NONE, 0,        none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(55,	uart2, MFSEL1, 11,	  none, NONE, 0,        none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(56,	r1err, MFSEL1, 12,	  none, NONE, 0,	none, NONE, 0,	     0),
-	NPCM7XX_PINCFG(57,       r1md, MFSEL1, 13,        none, NONE, 0,        none, NONE, 0,       DS(2, 4)),
-	NPCM7XX_PINCFG(58,       r1md, MFSEL1, 13,        none, NONE, 0,	none, NONE, 0,	     DS(2, 4)),
+	NPCM7XX_PINCFG(57,       r1md, MFSEL1, 13,        none, NONE, 0,        none, NONE, 0,       DSTR(2, 4)),
+	NPCM7XX_PINCFG(58,       r1md, MFSEL1, 13,        none, NONE, 0,	none, NONE, 0,	     DSTR(2, 4)),
 	NPCM7XX_PINCFG(59,	smb3d, I2CSEGSEL, 13,	  none, NONE, 0,	none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(60,	smb3d, I2CSEGSEL, 13,	  none, NONE, 0,	none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(61,      uart1, MFSEL1, 10,	  none, NONE, 0,	none, NONE, 0,     GPO),
@@ -1003,19 +1003,19 @@ static const struct npcm7xx_pincfg pincf
 	NPCM7XX_PINCFG(77,    fanin13, MFSEL2, 13,        none, NONE, 0,	none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(78,    fanin14, MFSEL2, 14,        none, NONE, 0,	none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(79,    fanin15, MFSEL2, 15,        none, NONE, 0,	none, NONE, 0,	     0),
-	NPCM7XX_PINCFG(80,	 pwm0, MFSEL2, 16,        none, NONE, 0,	none, NONE, 0,	     DS(4, 8)),
-	NPCM7XX_PINCFG(81,	 pwm1, MFSEL2, 17,        none, NONE, 0,	none, NONE, 0,	     DS(4, 8)),
-	NPCM7XX_PINCFG(82,	 pwm2, MFSEL2, 18,        none, NONE, 0,	none, NONE, 0,	     DS(4, 8)),
-	NPCM7XX_PINCFG(83,	 pwm3, MFSEL2, 19,        none, NONE, 0,	none, NONE, 0,	     DS(4, 8)),
-	NPCM7XX_PINCFG(84,         r2, MFSEL1, 14,        none, NONE, 0,        none, NONE, 0,	     DS(8, 12) | SLEW),
-	NPCM7XX_PINCFG(85,         r2, MFSEL1, 14,        none, NONE, 0,        none, NONE, 0,	     DS(8, 12) | SLEW),
-	NPCM7XX_PINCFG(86,         r2, MFSEL1, 14,        none, NONE, 0,        none, NONE, 0,	     DS(8, 12) | SLEW),
+	NPCM7XX_PINCFG(80,	 pwm0, MFSEL2, 16,        none, NONE, 0,	none, NONE, 0,	     DSTR(4, 8)),
+	NPCM7XX_PINCFG(81,	 pwm1, MFSEL2, 17,        none, NONE, 0,	none, NONE, 0,	     DSTR(4, 8)),
+	NPCM7XX_PINCFG(82,	 pwm2, MFSEL2, 18,        none, NONE, 0,	none, NONE, 0,	     DSTR(4, 8)),
+	NPCM7XX_PINCFG(83,	 pwm3, MFSEL2, 19,        none, NONE, 0,	none, NONE, 0,	     DSTR(4, 8)),
+	NPCM7XX_PINCFG(84,         r2, MFSEL1, 14,        none, NONE, 0,        none, NONE, 0,	     DSTR(8, 12) | SLEW),
+	NPCM7XX_PINCFG(85,         r2, MFSEL1, 14,        none, NONE, 0,        none, NONE, 0,	     DSTR(8, 12) | SLEW),
+	NPCM7XX_PINCFG(86,         r2, MFSEL1, 14,        none, NONE, 0,        none, NONE, 0,	     DSTR(8, 12) | SLEW),
 	NPCM7XX_PINCFG(87,         r2, MFSEL1, 14,        none, NONE, 0,        none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(88,         r2, MFSEL1, 14,        none, NONE, 0,        none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(89,         r2, MFSEL1, 14,        none, NONE, 0,        none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(90,      r2err, MFSEL1, 15,        none, NONE, 0,        none, NONE, 0,       0),
-	NPCM7XX_PINCFG(91,       r2md, MFSEL1, 16,	  none, NONE, 0,        none, NONE, 0,	     DS(2, 4)),
-	NPCM7XX_PINCFG(92,       r2md, MFSEL1, 16,	  none, NONE, 0,        none, NONE, 0,	     DS(2, 4)),
+	NPCM7XX_PINCFG(91,       r2md, MFSEL1, 16,	  none, NONE, 0,        none, NONE, 0,	     DSTR(2, 4)),
+	NPCM7XX_PINCFG(92,       r2md, MFSEL1, 16,	  none, NONE, 0,        none, NONE, 0,	     DSTR(2, 4)),
 	NPCM7XX_PINCFG(93,    ga20kbc, MFSEL1, 17,	 smb5d, I2CSEGSEL, 21,  none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(94,    ga20kbc, MFSEL1, 17,	 smb5d, I2CSEGSEL, 21,  none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(95,	  lpc, NONE, 0,		  espi, MFSEL4, 8,      gpio, MFSEL1, 26,    0),
@@ -1061,34 +1061,34 @@ static const struct npcm7xx_pincfg pincf
 	NPCM7XX_PINCFG(133,	smb10, MFSEL4, 13,	  none, NONE, 0,	none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(134,	smb11, MFSEL4, 14,	  none, NONE, 0,	none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(135,	smb11, MFSEL4, 14,	  none, NONE, 0,	none, NONE, 0,	     0),
-	NPCM7XX_PINCFG(136,	  sd1, MFSEL3, 12,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW),
-	NPCM7XX_PINCFG(137,	  sd1, MFSEL3, 12,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW),
-	NPCM7XX_PINCFG(138,	  sd1, MFSEL3, 12,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW),
-	NPCM7XX_PINCFG(139,	  sd1, MFSEL3, 12,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW),
-	NPCM7XX_PINCFG(140,	  sd1, MFSEL3, 12,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW),
+	NPCM7XX_PINCFG(136,	  sd1, MFSEL3, 12,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW),
+	NPCM7XX_PINCFG(137,	  sd1, MFSEL3, 12,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW),
+	NPCM7XX_PINCFG(138,	  sd1, MFSEL3, 12,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW),
+	NPCM7XX_PINCFG(139,	  sd1, MFSEL3, 12,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW),
+	NPCM7XX_PINCFG(140,	  sd1, MFSEL3, 12,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW),
 	NPCM7XX_PINCFG(141,	  sd1, MFSEL3, 12,	  none, NONE, 0,	none, NONE, 0,	     0),
-	NPCM7XX_PINCFG(142,	  sd1, MFSEL3, 12,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW),
+	NPCM7XX_PINCFG(142,	  sd1, MFSEL3, 12,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW),
 	NPCM7XX_PINCFG(143,       sd1, MFSEL3, 12,      sd1pwr, MFSEL4, 5,      none, NONE, 0,       0),
-	NPCM7XX_PINCFG(144,	 pwm4, MFSEL2, 20,	  none, NONE, 0,	none, NONE, 0,	     DS(4, 8)),
-	NPCM7XX_PINCFG(145,	 pwm5, MFSEL2, 21,	  none, NONE, 0,	none, NONE, 0,	     DS(4, 8)),
-	NPCM7XX_PINCFG(146,	 pwm6, MFSEL2, 22,	  none, NONE, 0,	none, NONE, 0,	     DS(4, 8)),
-	NPCM7XX_PINCFG(147,	 pwm7, MFSEL2, 23,	  none, NONE, 0,	none, NONE, 0,	     DS(4, 8)),
-	NPCM7XX_PINCFG(148,	 mmc8, MFSEL3, 11,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW),
-	NPCM7XX_PINCFG(149,	 mmc8, MFSEL3, 11,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW),
-	NPCM7XX_PINCFG(150,	 mmc8, MFSEL3, 11,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW),
-	NPCM7XX_PINCFG(151,	 mmc8, MFSEL3, 11,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW),
-	NPCM7XX_PINCFG(152,	  mmc, MFSEL3, 10,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW),
+	NPCM7XX_PINCFG(144,	 pwm4, MFSEL2, 20,	  none, NONE, 0,	none, NONE, 0,	     DSTR(4, 8)),
+	NPCM7XX_PINCFG(145,	 pwm5, MFSEL2, 21,	  none, NONE, 0,	none, NONE, 0,	     DSTR(4, 8)),
+	NPCM7XX_PINCFG(146,	 pwm6, MFSEL2, 22,	  none, NONE, 0,	none, NONE, 0,	     DSTR(4, 8)),
+	NPCM7XX_PINCFG(147,	 pwm7, MFSEL2, 23,	  none, NONE, 0,	none, NONE, 0,	     DSTR(4, 8)),
+	NPCM7XX_PINCFG(148,	 mmc8, MFSEL3, 11,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW),
+	NPCM7XX_PINCFG(149,	 mmc8, MFSEL3, 11,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW),
+	NPCM7XX_PINCFG(150,	 mmc8, MFSEL3, 11,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW),
+	NPCM7XX_PINCFG(151,	 mmc8, MFSEL3, 11,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW),
+	NPCM7XX_PINCFG(152,	  mmc, MFSEL3, 10,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW),
 	NPCM7XX_PINCFG(153,     mmcwp, FLOCKR1, 24,       none, NONE, 0,	none, NONE, 0,	     0),  /* Z1/A1 */
-	NPCM7XX_PINCFG(154,	  mmc, MFSEL3, 10,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW),
+	NPCM7XX_PINCFG(154,	  mmc, MFSEL3, 10,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW),
 	NPCM7XX_PINCFG(155,     mmccd, MFSEL3, 25,      mmcrst, MFSEL4, 6,      none, NONE, 0,       0),  /* Z1/A1 */
-	NPCM7XX_PINCFG(156,	  mmc, MFSEL3, 10,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW),
-	NPCM7XX_PINCFG(157,	  mmc, MFSEL3, 10,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW),
-	NPCM7XX_PINCFG(158,	  mmc, MFSEL3, 10,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW),
-	NPCM7XX_PINCFG(159,	  mmc, MFSEL3, 10,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW),
-
-	NPCM7XX_PINCFG(160,    clkout, MFSEL1, 21,        none, NONE, 0,        none, NONE, 0,	     DS(8, 12) | SLEW),
-	NPCM7XX_PINCFG(161,	  lpc, NONE, 0,		  espi, MFSEL4, 8,      gpio, MFSEL1, 26,    DS(8, 12)),
-	NPCM7XX_PINCFG(162,    serirq, NONE, 0,           gpio, MFSEL1, 31,	none, NONE, 0,	     DS(8, 12)),
+	NPCM7XX_PINCFG(156,	  mmc, MFSEL3, 10,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW),
+	NPCM7XX_PINCFG(157,	  mmc, MFSEL3, 10,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW),
+	NPCM7XX_PINCFG(158,	  mmc, MFSEL3, 10,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW),
+	NPCM7XX_PINCFG(159,	  mmc, MFSEL3, 10,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW),
+
+	NPCM7XX_PINCFG(160,    clkout, MFSEL1, 21,        none, NONE, 0,        none, NONE, 0,	     DSTR(8, 12) | SLEW),
+	NPCM7XX_PINCFG(161,	  lpc, NONE, 0,		  espi, MFSEL4, 8,      gpio, MFSEL1, 26,    DSTR(8, 12)),
+	NPCM7XX_PINCFG(162,    serirq, NONE, 0,           gpio, MFSEL1, 31,	none, NONE, 0,	     DSTR(8, 12)),
 	NPCM7XX_PINCFG(163,	  lpc, NONE, 0,		  espi, MFSEL4, 8,      gpio, MFSEL1, 26,    0),
 	NPCM7XX_PINCFG(164,	  lpc, NONE, 0,		  espi, MFSEL4, 8,      gpio, MFSEL1, 26,    SLEWLPC),
 	NPCM7XX_PINCFG(165,	  lpc, NONE, 0,		  espi, MFSEL4, 8,      gpio, MFSEL1, 26,    SLEWLPC),
@@ -1101,25 +1101,25 @@ static const struct npcm7xx_pincfg pincf
 	NPCM7XX_PINCFG(172,	 smb6, MFSEL3, 1,	  none, NONE, 0,	none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(173,	 smb7, MFSEL3, 2,	  none, NONE, 0,	none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(174,	 smb7, MFSEL3, 2,	  none, NONE, 0,	none, NONE, 0,	     0),
-	NPCM7XX_PINCFG(175,	pspi1, MFSEL3, 4,       faninx, MFSEL3, 3,      none, NONE, 0,	     DS(8, 12)),
-	NPCM7XX_PINCFG(176,     pspi1, MFSEL3, 4,       faninx, MFSEL3, 3,      none, NONE, 0,	     DS(8, 12)),
-	NPCM7XX_PINCFG(177,     pspi1, MFSEL3, 4,       faninx, MFSEL3, 3,      none, NONE, 0,	     DS(8, 12)),
-	NPCM7XX_PINCFG(178,	   r1, MFSEL3, 9,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW),
-	NPCM7XX_PINCFG(179,	   r1, MFSEL3, 9,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW),
-	NPCM7XX_PINCFG(180,	   r1, MFSEL3, 9,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW),
+	NPCM7XX_PINCFG(175,	pspi1, MFSEL3, 4,       faninx, MFSEL3, 3,      none, NONE, 0,	     DSTR(8, 12)),
+	NPCM7XX_PINCFG(176,     pspi1, MFSEL3, 4,       faninx, MFSEL3, 3,      none, NONE, 0,	     DSTR(8, 12)),
+	NPCM7XX_PINCFG(177,     pspi1, MFSEL3, 4,       faninx, MFSEL3, 3,      none, NONE, 0,	     DSTR(8, 12)),
+	NPCM7XX_PINCFG(178,	   r1, MFSEL3, 9,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW),
+	NPCM7XX_PINCFG(179,	   r1, MFSEL3, 9,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW),
+	NPCM7XX_PINCFG(180,	   r1, MFSEL3, 9,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW),
 	NPCM7XX_PINCFG(181,	   r1, MFSEL3, 9,	  none, NONE, 0,	none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(182,	   r1, MFSEL3, 9,	  none, NONE, 0,	none, NONE, 0,	     0),
-	NPCM7XX_PINCFG(183,     spi3, MFSEL4, 16,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW),
-	NPCM7XX_PINCFG(184,     spi3, MFSEL4, 16,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW | GPO),
-	NPCM7XX_PINCFG(185,     spi3, MFSEL4, 16,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW | GPO),
-	NPCM7XX_PINCFG(186,     spi3, MFSEL4, 16,	  none, NONE, 0,	none, NONE, 0,	     DS(8, 12)),
-	NPCM7XX_PINCFG(187,   spi3cs1, MFSEL4, 17,        none, NONE, 0,	none, NONE, 0,	     DS(8, 12)),
-	NPCM7XX_PINCFG(188,  spi3quad, MFSEL4, 20,     spi3cs2, MFSEL4, 18,     none, NONE, 0,    DS(8, 12) | SLEW),
-	NPCM7XX_PINCFG(189,  spi3quad, MFSEL4, 20,     spi3cs3, MFSEL4, 19,     none, NONE, 0,    DS(8, 12) | SLEW),
-	NPCM7XX_PINCFG(190,      gpio, FLOCKR1, 20,   nprd_smi, NONE, 0,	none, NONE, 0,	     DS(2, 4)),
-	NPCM7XX_PINCFG(191,	 none, NONE, 0,		  none, NONE, 0,	none, NONE, 0,	     DS(8, 12)),  /* XX */
+	NPCM7XX_PINCFG(183,     spi3, MFSEL4, 16,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW),
+	NPCM7XX_PINCFG(184,     spi3, MFSEL4, 16,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW | GPO),
+	NPCM7XX_PINCFG(185,     spi3, MFSEL4, 16,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW | GPO),
+	NPCM7XX_PINCFG(186,     spi3, MFSEL4, 16,	  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12)),
+	NPCM7XX_PINCFG(187,   spi3cs1, MFSEL4, 17,        none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12)),
+	NPCM7XX_PINCFG(188,  spi3quad, MFSEL4, 20,     spi3cs2, MFSEL4, 18,     none, NONE, 0,    DSTR(8, 12) | SLEW),
+	NPCM7XX_PINCFG(189,  spi3quad, MFSEL4, 20,     spi3cs3, MFSEL4, 19,     none, NONE, 0,    DSTR(8, 12) | SLEW),
+	NPCM7XX_PINCFG(190,      gpio, FLOCKR1, 20,   nprd_smi, NONE, 0,	none, NONE, 0,	     DSTR(2, 4)),
+	NPCM7XX_PINCFG(191,	 none, NONE, 0,		  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12)),  /* XX */
 
-	NPCM7XX_PINCFG(192,	 none, NONE, 0,		  none, NONE, 0,	none, NONE, 0,	     DS(8, 12)),  /* XX */
+	NPCM7XX_PINCFG(192,	 none, NONE, 0,		  none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12)),  /* XX */
 	NPCM7XX_PINCFG(193,	   r1, MFSEL3, 9,	  none, NONE, 0,	none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(194,	smb0b, I2CSEGSEL, 0,	  none, NONE, 0,	none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(195,	smb0b, I2CSEGSEL, 0,	  none, NONE, 0,	none, NONE, 0,	     0),
@@ -1130,11 +1130,11 @@ static const struct npcm7xx_pincfg pincf
 	NPCM7XX_PINCFG(200,        r2, MFSEL1, 14,        none, NONE, 0,        none, NONE, 0,       0),
 	NPCM7XX_PINCFG(201,	   r1, MFSEL3, 9,	  none, NONE, 0,	none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(202,	smb0c, I2CSEGSEL, 1,	  none, NONE, 0,	none, NONE, 0,	     0),
-	NPCM7XX_PINCFG(203,    faninx, MFSEL3, 3,         none, NONE, 0,	none, NONE, 0,	     DS(8, 12)),
+	NPCM7XX_PINCFG(203,    faninx, MFSEL3, 3,         none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12)),
 	NPCM7XX_PINCFG(204,	  ddc, NONE, 0,           gpio, MFSEL3, 22,	none, NONE, 0,	     SLEW),
 	NPCM7XX_PINCFG(205,	  ddc, NONE, 0,           gpio, MFSEL3, 22,	none, NONE, 0,	     SLEW),
-	NPCM7XX_PINCFG(206,	  ddc, NONE, 0,           gpio, MFSEL3, 22,	none, NONE, 0,	     DS(4, 8)),
-	NPCM7XX_PINCFG(207,	  ddc, NONE, 0,           gpio, MFSEL3, 22,	none, NONE, 0,	     DS(4, 8)),
+	NPCM7XX_PINCFG(206,	  ddc, NONE, 0,           gpio, MFSEL3, 22,	none, NONE, 0,	     DSTR(4, 8)),
+	NPCM7XX_PINCFG(207,	  ddc, NONE, 0,           gpio, MFSEL3, 22,	none, NONE, 0,	     DSTR(4, 8)),
 	NPCM7XX_PINCFG(208,       rg2, MFSEL4, 24,         ddr, MFSEL3, 26,     none, NONE, 0,       0),
 	NPCM7XX_PINCFG(209,       rg2, MFSEL4, 24,         ddr, MFSEL3, 26,     none, NONE, 0,       0),
 	NPCM7XX_PINCFG(210,       rg2, MFSEL4, 24,         ddr, MFSEL3, 26,     none, NONE, 0,       0),
@@ -1146,20 +1146,20 @@ static const struct npcm7xx_pincfg pincf
 	NPCM7XX_PINCFG(216,   rg2mdio, MFSEL4, 23,         ddr, MFSEL3, 26,     none, NONE, 0,       0),
 	NPCM7XX_PINCFG(217,   rg2mdio, MFSEL4, 23,         ddr, MFSEL3, 26,     none, NONE, 0,       0),
 	NPCM7XX_PINCFG(218,     wdog1, MFSEL3, 19,        none, NONE, 0,	none, NONE, 0,	     0),
-	NPCM7XX_PINCFG(219,     wdog2, MFSEL3, 20,        none, NONE, 0,	none, NONE, 0,	     DS(4, 8)),
+	NPCM7XX_PINCFG(219,     wdog2, MFSEL3, 20,        none, NONE, 0,	none, NONE, 0,	     DSTR(4, 8)),
 	NPCM7XX_PINCFG(220,	smb12, MFSEL3, 5,	  none, NONE, 0,	none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(221,	smb12, MFSEL3, 5,	  none, NONE, 0,	none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(222,     smb13, MFSEL3, 6,         none, NONE, 0,	none, NONE, 0,	     0),
 	NPCM7XX_PINCFG(223,     smb13, MFSEL3, 6,         none, NONE, 0,	none, NONE, 0,	     0),
 
 	NPCM7XX_PINCFG(224,	 spix, MFSEL4, 27,        none, NONE, 0,	none, NONE, 0,	     SLEW),
-	NPCM7XX_PINCFG(225,	 spix, MFSEL4, 27,        none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW | GPO),
-	NPCM7XX_PINCFG(226,	 spix, MFSEL4, 27,        none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW | GPO),
-	NPCM7XX_PINCFG(227,	 spix, MFSEL4, 27,        none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW),
-	NPCM7XX_PINCFG(228,   spixcs1, MFSEL4, 28,        none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW),
-	NPCM7XX_PINCFG(229,	 spix, MFSEL4, 27,        none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW),
-	NPCM7XX_PINCFG(230,	 spix, MFSEL4, 27,        none, NONE, 0,	none, NONE, 0,	     DS(8, 12) | SLEW),
-	NPCM7XX_PINCFG(231,    clkreq, MFSEL4, 9,         none, NONE, 0,        none, NONE, 0,	     DS(8, 12)),
+	NPCM7XX_PINCFG(225,	 spix, MFSEL4, 27,        none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW | GPO),
+	NPCM7XX_PINCFG(226,	 spix, MFSEL4, 27,        none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW | GPO),
+	NPCM7XX_PINCFG(227,	 spix, MFSEL4, 27,        none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW),
+	NPCM7XX_PINCFG(228,   spixcs1, MFSEL4, 28,        none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW),
+	NPCM7XX_PINCFG(229,	 spix, MFSEL4, 27,        none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW),
+	NPCM7XX_PINCFG(230,	 spix, MFSEL4, 27,        none, NONE, 0,	none, NONE, 0,	     DSTR(8, 12) | SLEW),
+	NPCM7XX_PINCFG(231,    clkreq, MFSEL4, 9,         none, NONE, 0,        none, NONE, 0,	     DSTR(8, 12)),
 	NPCM7XX_PINCFG(253,	 none, NONE, 0,		  none, NONE, 0,	none, NONE, 0,	     GPI), /* SDHC1 power */
 	NPCM7XX_PINCFG(254,	 none, NONE, 0,		  none, NONE, 0,	none, NONE, 0,	     GPI), /* SDHC2 power */
 	NPCM7XX_PINCFG(255,	 none, NONE, 0,		  none, NONE, 0,	none, NONE, 0,	     GPI), /* DACOSEL */
author maintainer
#1048 lore raw

PATCH 1044/1126 pinctrl: nuvoton: npcm7xx: Use %zu printk format for ARRAY_SIZE()

In reply to Greg Kroah-Hartman

From: Jonathan Neuschäfer <[email protected]>

commit 9d0f18bca3b557ae5d2128661ac06d33b3f45c0a upstream.

When compile-testing on 64-bit architectures, GCC complains about the mismatch of types between the %d format specifier and value returned by ARRAY_LENGTH(). Use %zu, which is correct everywhere.

---
 drivers/pinctrl/nuvoton/pinctrl-npcm7xx.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/pinctrl/nuvoton/pinctrl-npcm7xx.c +1 -1
@@ -1560,7 +1560,7 @@ static int npcm7xx_get_groups_count(stru
 {
 	struct npcm7xx_pinctrl *npcm = pinctrl_dev_get_drvdata(pctldev);
 
-	dev_dbg(npcm->dev, "group size: %d\n", ARRAY_SIZE(npcm7xx_groups));
+	dev_dbg(npcm->dev, "group size: %zu\n", ARRAY_SIZE(npcm7xx_groups));
 	return ARRAY_SIZE(npcm7xx_groups);
 }
 
author maintainer
#1049 lore raw

PATCH 1045/1126 ASoC: rockchip: i2s_tdm: Fixup config for SND_SOC_DAIFMT_DSP_A/B

In reply to Greg Kroah-Hartman

From: Meng Tang <[email protected]>

commit 2f45a4e2897793cc6ae25f5fe78b485ce7fd01d0 upstream.

SND_SOC_DAIFMT_DSP_A: PCM delay 1 bit mode, L data MSB after FRM LRC
SND_SOC_DAIFMT_DSP_B: PCM no delay mode, L data MSB during FRM LRC

Fixes 081068fd64140 (ASoC: rockchip: add support for i2s-tdm controller)
---
 sound/soc/rockchip/rockchip_i2s_tdm.c |   10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)
sound/soc/rockchip/rockchip_i2s_tdm.c +5 -5
@@ -469,14 +469,14 @@ static int rockchip_i2s_tdm_set_fmt(stru
 		txcr_val = I2S_TXCR_IBM_NORMAL;
 		rxcr_val = I2S_RXCR_IBM_NORMAL;
 		break;
-	case SND_SOC_DAIFMT_DSP_A: /* PCM no delay mode */
-		txcr_val = I2S_TXCR_TFS_PCM;
-		rxcr_val = I2S_RXCR_TFS_PCM;
-		break;
-	case SND_SOC_DAIFMT_DSP_B: /* PCM delay 1 mode */
+	case SND_SOC_DAIFMT_DSP_A: /* PCM delay 1 mode */
 		txcr_val = I2S_TXCR_TFS_PCM | I2S_TXCR_PBM_MODE(1);
 		rxcr_val = I2S_RXCR_TFS_PCM | I2S_RXCR_PBM_MODE(1);
 		break;
+	case SND_SOC_DAIFMT_DSP_B: /* PCM no delay mode */
+		txcr_val = I2S_TXCR_TFS_PCM;
+		rxcr_val = I2S_RXCR_TFS_PCM;
+		break;
 	default:
 		ret = -EINVAL;
 		goto err_pm_put;
author maintainer
#1050 lore raw

PATCH 1046/1126 ASoC: mediatek: mt6358: add missing EXPORT_SYMBOLs

In reply to Greg Kroah-Hartman

From: Jiaxin Yu <[email protected]>

commit a7663c89f4193dbf717572e46e5a3251940dbdc8 upstream.

Fixes the following build errors when mt6358 is configured as module:

Quoted 4 lines
Quoted 4 lines

ERROR: modpost: "mt6358_set_mtkaif_protocol" [sound/soc/mediatek/mt8186/mt8186-mt6366-rt1019-rt5682s.ko] undefined!
ERROR: modpost: "mt6358_set_mtkaif_protocol" [sound/soc/mediatek/mt8186/mt8186-mt6366-da7219-max98357.ko] undefined!

---
 sound/soc/codecs/mt6358.c |    4 ++++
 1 file changed, 4 insertions(+)
sound/soc/codecs/mt6358.c +4 -0
@@ -107,6 +107,7 @@ int mt6358_set_mtkaif_protocol(struct sn
 	priv->mtkaif_protocol = mtkaif_protocol;
 	return 0;
 }
+EXPORT_SYMBOL_GPL(mt6358_set_mtkaif_protocol);
 
 static void playback_gpio_set(struct mt6358_priv *priv)
 {
@@ -273,6 +274,7 @@ int mt6358_mtkaif_calibration_enable(str
 			   1 << RG_AUD_PAD_TOP_DAT_MISO_LOOPBACK_SFT);
 	return 0;
 }
+EXPORT_SYMBOL_GPL(mt6358_mtkaif_calibration_enable);
 
 int mt6358_mtkaif_calibration_disable(struct snd_soc_component *cmpnt)
 {
@@ -296,6 +298,7 @@ int mt6358_mtkaif_calibration_disable(st
 	capture_gpio_reset(priv);
 	return 0;
 }
+EXPORT_SYMBOL_GPL(mt6358_mtkaif_calibration_disable);
 
 int mt6358_set_mtkaif_calibration_phase(struct snd_soc_component *cmpnt,
 					int phase_1, int phase_2)
@@ -310,6 +313,7 @@ int mt6358_set_mtkaif_calibration_phase(
 			   phase_2 << RG_AUD_PAD_TOP_PHASE_MODE2_SFT);
 	return 0;
 }
+EXPORT_SYMBOL_GPL(mt6358_set_mtkaif_calibration_phase);
 
 /* dl pga gain */
 enum {
author maintainer
#1051 lore raw

PATCH 1047/1126 ubi: Fix race condition between ctrl_cdev_ioctl and ubi_cdev_ioctl

In reply to Greg Kroah-Hartman

From: Baokun Li <[email protected]>

commit 3cbf0e392f173ba0ce425968c8374a6aa3e90f2e upstream.

Hulk Robot reported a KASAN report about use-after-free:
==================================================================

 BUG: KASAN: use-after-free in __list_del_entry_valid+0x13d/0x160
 Read of size 8 at addr ffff888035e37d98 by task ubiattach/1385
 [...]
 Call Trace:
  klist_dec_and_del+0xa7/0x4a0
  klist_put+0xc7/0x1a0
  device_del+0x4d4/0xed0
  cdev_device_del+0x1a/0x80
  ubi_attach_mtd_dev+0x2951/0x34b0 [ubi]
  ctrl_cdev_ioctl+0x286/0x2f0 [ubi]

 Allocated by task 1414:
  device_add+0x60a/0x18b0
  cdev_device_add+0x103/0x170
  ubi_create_volume+0x1118/0x1a10 [ubi]
  ubi_cdev_ioctl+0xb7f/0x1ba0 [ubi]

 Freed by task 1385:
  cdev_device_del+0x1a/0x80
  ubi_remove_volume+0x438/0x6c0 [ubi]
  ubi_cdev_ioctl+0xbf4/0x1ba0 [ubi]
 [...]
 ==================================================================

The lock held by ctrl_cdev_ioctl is ubi_devices_mutex, but the lock held by ubi_cdev_ioctl is ubi->device_mutex. Therefore, the two locks can be concurrent.

ctrl_cdev_ioctl contains two operations: ubi_attach and ubi_detach. ubi_detach is bug-free because it uses reference counting to prevent concurrency. However, uif_init and uif_close in ubi_attach may race with ubi_cdev_ioctl.

uif_init will race with ubi_cdev_ioctl as in the following stack.
cpu1 cpu2 cpu3
_______________________|________________________|______________________ ctrl_cdev_ioctl
ubi_attach_mtd_dev
uif_init
ubi_cdev_ioctl
ubi_create_volume
cdev_device_add
ubi_add_volume
// sysfs exist
kill_volumes

                                                    ubi_cdev_ioctl
                                                     ubi_remove_volume
                                                      cdev_device_del
                                                       // first free
    ubi_free_volume
     cdev_del
     // double free

cdev_device_del

And uif_close will race with ubi_cdev_ioctl as in the following stack.
cpu1 cpu2 cpu3
_______________________|________________________|______________________ ctrl_cdev_ioctl
ubi_attach_mtd_dev
uif_init
ubi_cdev_ioctl
ubi_create_volume
cdev_device_add
ubi_debugfs_init_dev
//error goto out_uif;
uif_close
kill_volumes

                                                    ubi_cdev_ioctl
                                                     ubi_remove_volume
                                                      cdev_device_del
                                                       // first free
    ubi_free_volume
    // double free

The cause of this problem is that commit 714fb87e8bc0 make device "available" before it becomes accessible via sysfs. Therefore, we roll back the modification. We will fix the race condition between ubi device creation and udev by removing ubi_get_device in vol_attribute_show and dev_attribute_show.This avoids accessing uninitialized ubi_devices[ubi_num].

ubi_get_device is used to prevent devices from being deleted during sysfs execution. However, now kernfs ensures that devices will not be deleted before all reference counting are released. The key process is shown in the following stack.

device_del
device_remove_attrs
device_remove_groups
sysfs_remove_groups
sysfs_remove_group
remove_files
kernfs_remove_by_name
kernfs_remove_by_name_ns
__kernfs_remove
kernfs_drain

Fixes 714fb87e8bc0 ("ubi: Fix race condition between ubi device creation and udev")
---
 drivers/mtd/ubi/build.c |    9 +--------
 drivers/mtd/ubi/vmt.c   |    8 +-------
 2 files changed, 2 insertions(+), 15 deletions(-)
drivers/mtd/ubi/build.c +1 -8
@@ -351,9 +351,6 @@ static ssize_t dev_attribute_show(struct
 	 * we still can use 'ubi->ubi_num'.
 	 */
 	ubi = container_of(dev, struct ubi_device, dev);
-	ubi = ubi_get_device(ubi->ubi_num);
-	if (!ubi)
-		return -ENODEV;
 
 	if (attr == &dev_eraseblock_size)
 		ret = sprintf(buf, "%d\n", ubi->leb_size);
@@ -382,7 +379,6 @@ static ssize_t dev_attribute_show(struct
 	else
 		ret = -EINVAL;
 
-	ubi_put_device(ubi);
 	return ret;
 }
 
@@ -979,9 +975,6 @@ int ubi_attach_mtd_dev(struct mtd_info *
 			goto out_detach;
 	}
 
-	/* Make device "available" before it becomes accessible via sysfs */
-	ubi_devices[ubi_num] = ubi;
-
 	err = uif_init(ubi);
 	if (err)
 		goto out_detach;
@@ -1026,6 +1019,7 @@ int ubi_attach_mtd_dev(struct mtd_info *
 	wake_up_process(ubi->bgt_thread);
 	spin_unlock(&ubi->wl_lock);
 
+	ubi_devices[ubi_num] = ubi;
 	ubi_notify_all(ubi, UBI_VOLUME_ADDED, NULL);
 	return ubi_num;
 
@@ -1034,7 +1028,6 @@ out_debugfs:
 out_uif:
 	uif_close(ubi);
 out_detach:
-	ubi_devices[ubi_num] = NULL;
 	ubi_wl_close(ubi);
 	ubi_free_all_volumes(ubi);
 	vfree(ubi->vtbl);
drivers/mtd/ubi/vmt.c +1 -7
@@ -56,16 +56,11 @@ static ssize_t vol_attribute_show(struct
 {
 	int ret;
 	struct ubi_volume *vol = container_of(dev, struct ubi_volume, dev);
-	struct ubi_device *ubi;
-
-	ubi = ubi_get_device(vol->ubi->ubi_num);
-	if (!ubi)
-		return -ENODEV;
+	struct ubi_device *ubi = vol->ubi;
 
 	spin_lock(&ubi->volumes_lock);
 	if (!ubi->volumes[vol->vol_id]) {
 		spin_unlock(&ubi->volumes_lock);
-		ubi_put_device(ubi);
 		return -ENODEV;
 	}
 	/* Take a reference to prevent volume removal */
@@ -103,7 +98,6 @@ static ssize_t vol_attribute_show(struct
 	vol->ref_count -= 1;
 	ubi_assert(vol->ref_count >= 0);
 	spin_unlock(&ubi->volumes_lock);
-	ubi_put_device(ubi);
 	return ret;
 }
 
author maintainer
#1052 lore raw

PATCH 1048/1126 ARM: iop32x: offset IRQ numbers by 1

In reply to Greg Kroah-Hartman

From: Arnd Bergmann <[email protected]>

commit 9d67412f24cc3a2c05f35f7c856addb07a2960ce upstream.

iop32x is one of the last platforms to use IRQ 0, and this has apparently stopped working in a 2014 cleanup without anyone noticing. This interrupt is used for the DMA engine, so most likely this has not actually worked in the past 7 years, but it's also not essential for using this board.

I'm splitting out this change from my GENERIC_IRQ_MULTI_HANDLER conversion so it can be backported if anyone cares.

Fixes a71b092a9c68 ("ARM: Convert handle_IRQ to use __handle_domain_irq")
[ardb: take +1 offset into account in mask/unmask and init as well]
---
 arch/arm/mach-iop32x/include/mach/entry-macro.S |    2 
 arch/arm/mach-iop32x/include/mach/irqs.h        |    2 
 arch/arm/mach-iop32x/irq.c                      |    6 +-
 arch/arm/mach-iop32x/irqs.h                     |   60 ++++++++++++------------
 4 files changed, 37 insertions(+), 33 deletions(-)
arch/arm/mach-iop32x/include/mach/entry-macro.S +1 -1
@@ -20,7 +20,7 @@
 	mrc     p6, 0, \irqstat, c8, c0, 0	@ Read IINTSRC
 	cmp     \irqstat, #0
 	clzne   \irqnr, \irqstat
-	rsbne   \irqnr, \irqnr, #31
+	rsbne   \irqnr, \irqnr, #32
 	.endm
 
 	.macro arch_ret_to_user, tmp1, tmp2
arch/arm/mach-iop32x/include/mach/irqs.h +1 -1
@@ -9,6 +9,6 @@
 #ifndef __IRQS_H
 #define __IRQS_H
 
-#define NR_IRQS			32
+#define NR_IRQS			33
 
 #endif
arch/arm/mach-iop32x/irq.c +3 -3
@@ -32,14 +32,14 @@ static void intstr_write(u32 val)
 static void
 iop32x_irq_mask(struct irq_data *d)
 {
-	iop32x_mask &= ~(1 << d->irq);
+	iop32x_mask &= ~(1 << (d->irq - 1));
 	intctl_write(iop32x_mask);
 }
 
 static void
 iop32x_irq_unmask(struct irq_data *d)
 {
-	iop32x_mask |= 1 << d->irq;
+	iop32x_mask |= 1 << (d->irq - 1);
 	intctl_write(iop32x_mask);
 }
 
@@ -65,7 +65,7 @@ void __init iop32x_init_irq(void)
 	    machine_is_em7210())
 		*IOP3XX_PCIIRSR = 0x0f;
 
-	for (i = 0; i < NR_IRQS; i++) {
+	for (i = 1; i < NR_IRQS; i++) {
 		irq_set_chip_and_handler(i, &ext_chip, handle_level_irq);
 		irq_clear_status_flags(i, IRQ_NOREQUEST | IRQ_NOPROBE);
 	}
arch/arm/mach-iop32x/irqs.h +32 -28

68 lines not on this page · raw message

author maintainer
#1053 lore raw

PATCH 1049/1126 block: Fix the maximum minor value is blk_alloc_ext_minor()

In reply to Greg Kroah-Hartman

From: Christophe JAILLET <[email protected]>

commit d1868328dec5ae2cf210111025fcbc71f78dd5ca upstream.

ida_alloc_range(..., min, max, ...) returns values from min to max, inclusive.

So, NR_EXT_DEVT is a valid idx returned by blk_alloc_ext_minor().

This is an issue because in device_add_disk(), this value is used in:
ddev->devt = MKDEV(disk->major, disk->first_minor);
and NR_EXT_DEVT is '(1 << MINORBITS)'.

So, should 'disk->first_minor' be NR_EXT_DEVT, it would overflow.

---
 block/genhd.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
block/genhd.c +1 -1
@@ -330,7 +330,7 @@ int blk_alloc_ext_minor(void)
 {
 	int idx;
 
-	idx = ida_alloc_range(&ext_devt_ida, 0, NR_EXT_DEVT, GFP_KERNEL);
+	idx = ida_alloc_range(&ext_devt_ida, 0, NR_EXT_DEVT - 1, GFP_KERNEL);
 	if (idx == -ENOSPC)
 		return -EBUSY;
 	return idx;
author maintainer
#1054 lore raw

PATCH 1050/1126 block: restore the old set_task_ioprio() behaviour wrt PF_EXITING

In reply to Greg Kroah-Hartman

From: Jiri Slaby <[email protected]>

commit 15583a563cd5a7358e975599b7de7caacd9e9ce9 upstream.

PF_EXITING tasks were silently ignored before the below commits.
Continue doing so. Otherwise python-psutil tests fail:
ERROR: psutil.tests.test_process.TestProcess.test_zombie_process
----------------------------------------------------------------------
Traceback (most recent call last):
File "/home/abuild/rpmbuild/BUILD/psutil-5.9.0/build/lib.linux-x86_64-3.9/psutil/_pslinux.py", line 1661, in wrapper
return fun(self, *args, **kwargs)
File "/home/abuild/rpmbuild/BUILD/psutil-5.9.0/build/lib.linux-x86_64-3.9/psutil/_pslinux.py", line 2133, in ionice_set
return cext.proc_ioprio_set(self.pid, ioclass, value)
ProcessLookupError: [Errno 3] No such process

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "/home/abuild/rpmbuild/BUILD/psutil-5.9.0/psutil/tests/test_process.py", line 1313, in test_zombie_process
succeed_or_zombie_p_exc(fun)
File "/home/abuild/rpmbuild/BUILD/psutil-5.9.0/psutil/tests/test_process.py", line 1288, in succeed_or_zombie_p_exc
return fun()
File "/home/abuild/rpmbuild/BUILD/psutil-5.9.0/build/lib.linux-x86_64-3.9/psutil/__init__.py", line 792, in ionice
return self._proc.ionice_set(ioclass, value)
File "/home/abuild/rpmbuild/BUILD/psutil-5.9.0/build/lib.linux-x86_64-3.9/psutil/_pslinux.py", line 1665, in wrapper
raise NoSuchProcess(self.pid, self._name)
psutil.NoSuchProcess: process no longer exists (pid=2057)

Cc Christoph Hellwig <[email protected]>
Cc Jan Kara <[email protected]>
Cc Jens Axboe <[email protected]>
Fixes 5fc11eebb4 (block: open code create_task_io_context in set_task_ioprio)
Fixes a957b61254 (block: fix error in handling dead task for ioprio setting)
---
 block/blk-ioc.c |    3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)
block/blk-ioc.c +1 -2
@@ -280,7 +280,6 @@ int set_task_ioprio(struct task_struct *
 
 		task_lock(task);
 		if (task->flags & PF_EXITING) {
-			err = -ESRCH;
 			kmem_cache_free(iocontext_cachep, ioc);
 			goto out;
 		}
@@ -292,7 +291,7 @@ int set_task_ioprio(struct task_struct *
 	task->io_context->ioprio = ioprio;
 out:
 	task_unlock(task);
-	return err;
+	return 0;
 }
 EXPORT_SYMBOL_GPL(set_task_ioprio);
 
author maintainer
#1055 lore raw

PATCH 1051/1126 Revert "virtio-pci: harden INTX interrupts"

In reply to Greg Kroah-Hartman

From: Jason Wang <[email protected]>

commit 7b79edfb862d6b1ecc66479419ae67a7db2d02e3 upstream.

This reverts commit 080cd7c3ac8701081d143a15ba17dd9475313188. Since the MSI-X interrupts hardening will be reverted in the next patch. We will rework the interrupt hardening in the future.

---
 drivers/virtio/virtio_pci_common.c |   23 ++---------------------
 drivers/virtio/virtio_pci_common.h |    1 -
 2 files changed, 2 insertions(+), 22 deletions(-)
drivers/virtio/virtio_pci_common.c +2 -21
@@ -30,16 +30,8 @@ void vp_disable_cbs(struct virtio_device
 	struct virtio_pci_device *vp_dev = to_vp_device(vdev);
 	int i;
 
-	if (vp_dev->intx_enabled) {
-		/*
-		 * The below synchronize() guarantees that any
-		 * interrupt for this line arriving after
-		 * synchronize_irq() has completed is guaranteed to see
-		 * intx_soft_enabled == false.
-		 */
-		WRITE_ONCE(vp_dev->intx_soft_enabled, false);
+	if (vp_dev->intx_enabled)
 		synchronize_irq(vp_dev->pci_dev->irq);
-	}
 
 	for (i = 0; i < vp_dev->msix_vectors; ++i)
 		disable_irq(pci_irq_vector(vp_dev->pci_dev, i));
@@ -51,16 +43,8 @@ void vp_enable_cbs(struct virtio_device
 	struct virtio_pci_device *vp_dev = to_vp_device(vdev);
 	int i;
 
-	if (vp_dev->intx_enabled) {
-		disable_irq(vp_dev->pci_dev->irq);
-		/*
-		 * The above disable_irq() provides TSO ordering and
-		 * as such promotes the below store to store-release.
-		 */
-		WRITE_ONCE(vp_dev->intx_soft_enabled, true);
-		enable_irq(vp_dev->pci_dev->irq);
+	if (vp_dev->intx_enabled)
 		return;
-	}
 
 	for (i = 0; i < vp_dev->msix_vectors; ++i)
 		enable_irq(pci_irq_vector(vp_dev->pci_dev, i));
@@ -113,9 +97,6 @@ static irqreturn_t vp_interrupt(int irq,
 	struct virtio_pci_device *vp_dev = opaque;
 	u8 isr;
 
-	if (!READ_ONCE(vp_dev->intx_soft_enabled))
-		return IRQ_NONE;
-
 	/* reading the ISR has the effect of also clearing it so it's very
 	 * important to save off the value. */
 	isr = ioread8(vp_dev->isr);
drivers/virtio/virtio_pci_common.h +0 -1
@@ -63,7 +63,6 @@ struct virtio_pci_device {
 	/* MSI-X support */
 	int msix_enabled;
 	int intx_enabled;
-	bool intx_soft_enabled;
 	cpumask_var_t *msix_affinity_masks;
 	/* Name strings for interrupts. This size should be enough,
 	 * and I'm too lazy to allocate each name separately. */
author maintainer
#1056 lore raw

PATCH 1052/1126 Revert "virtio_pci: harden MSI-X interrupts"

In reply to Greg Kroah-Hartman

From: Jason Wang <[email protected]>

commit eb4cecb453a19b34d5454b49532e09e9cb0c1529 upstream.

This reverts commit 9e35276a5344f74d4a3600fc4100b3dd251d5c56. Issue were reported for the drivers that are using affinity managed IRQ where manually toggling IRQ status is not expected. And we forget to enable the interrupts in the restore path as well.

In the future, we will rework on the interrupt hardening.

---
 drivers/virtio/virtio_pci_common.c |   27 ++++++---------------------
 drivers/virtio/virtio_pci_common.h |    6 ++----
 drivers/virtio/virtio_pci_legacy.c |    5 ++---
 drivers/virtio/virtio_pci_modern.c |    6 ++----
 4 files changed, 12 insertions(+), 32 deletions(-)
drivers/virtio/virtio_pci_common.c +6 -21
@@ -24,8 +24,8 @@ MODULE_PARM_DESC(force_legacy,
 		 "Force legacy mode for transitional virtio 1 devices");
 #endif
 
-/* disable irq handlers */
-void vp_disable_cbs(struct virtio_device *vdev)
+/* wait for pending irq handlers */
+void vp_synchronize_vectors(struct virtio_device *vdev)
 {
 	struct virtio_pci_device *vp_dev = to_vp_device(vdev);
 	int i;
@@ -34,20 +34,7 @@ void vp_disable_cbs(struct virtio_device
 		synchronize_irq(vp_dev->pci_dev->irq);
 
 	for (i = 0; i < vp_dev->msix_vectors; ++i)
-		disable_irq(pci_irq_vector(vp_dev->pci_dev, i));
-}
-
-/* enable irq handlers */
-void vp_enable_cbs(struct virtio_device *vdev)
-{
-	struct virtio_pci_device *vp_dev = to_vp_device(vdev);
-	int i;
-
-	if (vp_dev->intx_enabled)
-		return;
-
-	for (i = 0; i < vp_dev->msix_vectors; ++i)
-		enable_irq(pci_irq_vector(vp_dev->pci_dev, i));
+		synchronize_irq(pci_irq_vector(vp_dev->pci_dev, i));
 }
 
 /* the notify function used when creating a virt queue */
@@ -154,8 +141,7 @@ static int vp_request_msix_vectors(struc
 	snprintf(vp_dev->msix_names[v], sizeof *vp_dev->msix_names,
 		 "%s-config", name);
 	err = request_irq(pci_irq_vector(vp_dev->pci_dev, v),
-			  vp_config_changed, IRQF_NO_AUTOEN,
-			  vp_dev->msix_names[v],
+			  vp_config_changed, 0, vp_dev->msix_names[v],
 			  vp_dev);
 	if (err)
 		goto error;
@@ -174,8 +160,7 @@ static int vp_request_msix_vectors(struc
 		snprintf(vp_dev->msix_names[v], sizeof *vp_dev->msix_names,
 			 "%s-virtqueues", name);
 		err = request_irq(pci_irq_vector(vp_dev->pci_dev, v),
-				  vp_vring_interrupt, IRQF_NO_AUTOEN,
-				  vp_dev->msix_names[v],
+				  vp_vring_interrupt, 0, vp_dev->msix_names[v],
 				  vp_dev);
 		if (err)
 			goto error;
@@ -352,7 +337,7 @@ static int vp_find_vqs_msix(struct virti
 			 "%s-%s",
 			 dev_name(&vp_dev->vdev.dev), names[i]);
 		err = request_irq(pci_irq_vector(vp_dev->pci_dev, msix_vec),
-				  vring_interrupt, IRQF_NO_AUTOEN,
+				  vring_interrupt, 0,
 				  vp_dev->msix_names[msix_vec],
 				  vqs[i]);
 		if (err)
drivers/virtio/virtio_pci_common.h +2 -4
@@ -101,10 +101,8 @@ static struct virtio_pci_device *to_vp_d
 	return container_of(vdev, struct virtio_pci_device, vdev);
 }
 
-/* disable irq handlers */
-void vp_disable_cbs(struct virtio_device *vdev);
-/* enable irq handlers */
-void vp_enable_cbs(struct virtio_device *vdev);
+/* wait for pending irq handlers */
+void vp_synchronize_vectors(struct virtio_device *vdev);
 /* the notify function used when creating a virt queue */
 bool vp_notify(struct virtqueue *vq);
 /* the config->del_vqs() implementation */
drivers/virtio/virtio_pci_legacy.c +2 -3
@@ -98,8 +98,8 @@ static void vp_reset(struct virtio_devic
 	/* Flush out the status write, and flush in device writes,
 	 * including MSi-X interrupts, if any. */
 	vp_legacy_get_status(&vp_dev->ldev);
-	/* Disable VQ/configuration callbacks. */
-	vp_disable_cbs(vdev);
+	/* Flush pending VQ/configuration callbacks. */
+	vp_synchronize_vectors(vdev);
 }
 
 static u16 vp_config_vector(struct virtio_pci_device *vp_dev, u16 vector)
@@ -185,7 +185,6 @@ static void del_vq(struct virtio_pci_vq_
 }
 
 static const struct virtio_config_ops virtio_pci_config_ops = {
-	.enable_cbs	= vp_enable_cbs,
 	.get		= vp_get,
 	.set		= vp_set,
 	.get_status	= vp_get_status,
drivers/virtio/virtio_pci_modern.c +2 -4
@@ -172,8 +172,8 @@ static void vp_reset(struct virtio_devic
 	 */
 	while (vp_modern_get_status(mdev))
 		msleep(1);
-	/* Disable VQ/configuration callbacks. */
-	vp_disable_cbs(vdev);
+	/* Flush pending VQ/configuration callbacks. */
+	vp_synchronize_vectors(vdev);
 }
 
 static u16 vp_config_vector(struct virtio_pci_device *vp_dev, u16 vector)
@@ -380,7 +380,6 @@ static bool vp_get_shm_region(struct vir
 }
 
 static const struct virtio_config_ops virtio_pci_config_nodev_ops = {
-	.enable_cbs	= vp_enable_cbs,
 	.get		= NULL,
 	.set		= NULL,
 	.generation	= vp_generation,
@@ -398,7 +397,6 @@ static const struct virtio_config_ops vi
 };
 
 static const struct virtio_config_ops virtio_pci_config_ops = {
-	.enable_cbs	= vp_enable_cbs,
 	.get		= vp_get,
 	.set		= vp_set,
 	.generation	= vp_generation,
author maintainer
#1057 lore raw

PATCH 1053/1126 virtio: use virtio_device_ready() in virtio_device_restore()

In reply to Greg Kroah-Hartman

From: Stefano Garzarella <[email protected]>

commit 8d65bc9a5be3f23c5e2ab36b6b8ef40095165b18 upstream.

After waking up a suspended VM, the kernel prints the following trace for virtio drivers which do not directly call virtio_device_ready() in the .restore:

    PM: suspend exit
    irq 22: nobody cared (try booting with the "irqpoll" option)
    Call Trace:
     <IRQ>
     dump_stack_lvl+0x38/0x49
     dump_stack+0x10/0x12
     __report_bad_irq+0x3a/0xaf
     note_interrupt.cold+0xb/0x60
     handle_irq_event+0x71/0x80
     handle_fasteoi_irq+0x95/0x1e0
     __common_interrupt+0x6b/0x110
     common_interrupt+0x63/0xe0
     asm_common_interrupt+0x1e/0x40
     ? __do_softirq+0x75/0x2f3
     irq_exit_rcu+0x93/0xe0
     sysvec_apic_timer_interrupt+0xac/0xd0
     </IRQ>
     <TASK>
     asm_sysvec_apic_timer_interrupt+0x12/0x20
     arch_cpu_idle+0x12/0x20
     default_idle_call+0x39/0xf0
     do_idle+0x1b5/0x210
     cpu_startup_entry+0x20/0x30
     start_secondary+0xf3/0x100
     secondary_startup_64_no_verify+0xc3/0xcb
     </TASK>
    handlers:
    [<000000008f9bac49>] vp_interrupt
    [<000000008f9bac49>] vp_interrupt
    Disabling IRQ #22

This happens because we don't invoke .enable_cbs callback in virtio_device_restore(). That callback is used by some transports (e.g. virtio-pci) to enable interrupts.

Let's fix it, by calling virtio_device_ready() as we do in virtio_dev_probe(). This function calls .enable_cts callback and sets DRIVER_OK status bit.

This fix also avoids setting DRIVER_OK twice for those drivers that call virtio_device_ready() in the .restore.

Fixes d50497eb4e55 ("virtio_config: introduce a new .enable_cbs method")
---
 drivers/virtio/virtio.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/virtio/virtio.c +3 -2
@@ -526,8 +526,9 @@ int virtio_device_restore(struct virtio_
 			goto err;
 	}
 
-	/* Finally, tell the device we're all set */
-	virtio_add_status(dev, VIRTIO_CONFIG_S_DRIVER_OK);
+	/* If restore didn't do it, mark device DRIVER_OK ourselves. */
+	if (!(dev->config->get_status(dev) & VIRTIO_CONFIG_S_DRIVER_OK))
+		virtio_device_ready(dev);
 
 	virtio_config_enable(dev);
 
author maintainer
#1058 lore raw

PATCH 1054/1126 io_uring: remove poll entry from list when canceling all

In reply to Greg Kroah-Hartman

From: Jens Axboe <[email protected]>

commit 61bc84c4008812d784c398cfb54118c1ba396dfc upstream.

When the ring is exiting, as part of the shutdown, poll requests are removed. But io_poll_remove_all() does not remove entries when finding them, and since completions are done out-of-band, we can find and remove the same entry multiple times.

We do guard the poll execution by poll ownership, but that does not exclude us from reissuing a new one once the previous removal ownership goes away.

This can race with poll execution as well, where we then end up seeing req->apoll be NULL because a previous task_work requeue finished the request.

Remove the poll entry when we find it and get ownership of it. This prevents multiple invocations from finding it.

Fixes aa43477b0402 ("io_uring: poll rework")
---
 fs/io_uring.c |    1 +
 1 file changed, 1 insertion(+)
fs/io_uring.c +1 -0
@@ -5870,6 +5870,7 @@ static __cold bool io_poll_remove_all(st
 		list = &ctx->cancel_hash[i];
 		hlist_for_each_entry_safe(req, tmp, list, hash_node) {
 			if (io_match_task_safe(req, tsk, cancel_all)) {
+				hlist_del_init(&req->hash_node);
 				io_poll_cancel_req(req);
 				found = true;
 			}
author maintainer
#1059 lore raw

PATCH 1055/1126 io_uring: bump poll refs to full 31-bits

In reply to Greg Kroah-Hartman

From: Jens Axboe <[email protected]>

commit e2c0cb7c0cc72939b61a7efee376206725796625 upstream.

The previous commit:

1bc84c40088 ("io_uring: remove poll entry from list when canceling all")

removed a potential overflow condition for the poll references. They are currently limited to 20-bits, even if we have 31-bits available. The upper bit is used to mark for cancelation.

Bump the poll ref space to 31-bits, making that kind of situation much harder to trigger in general. We'll separately add overflow checking and handling.

Fixes aa43477b0402 ("io_uring: poll rework")
---
 fs/io_uring.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/io_uring.c +1 -1
@@ -5414,7 +5414,7 @@ struct io_poll_table {
 };
 
 #define IO_POLL_CANCEL_FLAG	BIT(31)
-#define IO_POLL_REF_MASK	((1u << 20)-1)
+#define IO_POLL_REF_MASK	GENMASK(30, 0)
 
 /*
  * If refs part of ->poll_refs (see IO_POLL_REF_MASK) is 0, it's free. We can
author maintainer
#1060 lore raw

PATCH 1056/1126 io_uring: fix memory leak of uid in files registration

In reply to Greg Kroah-Hartman

From: Pavel Begunkov <[email protected]>

commit c86d18f4aa93e0e66cda0e55827cd03eea6bc5f8 upstream.

When there are no files for __io_sqe_files_scm() to process in the range, it'll free everything and return. However, it forgets to put uid.

---
 fs/io_uring.c |    1 +
 1 file changed, 1 insertion(+)
fs/io_uring.c +1 -0
@@ -8241,6 +8241,7 @@ static int __io_sqe_files_scm(struct io_
 			fput(fpl->fp[i]);
 	} else {
 		kfree_skb(skb);
+		free_uid(fpl->user);
 		kfree(fpl);
 	}
 
author maintainer
#1061 lore raw

PATCH 1057/1126 riscv module: remove (NOLOAD)

In reply to Greg Kroah-Hartman

From: Fangrui Song <[email protected]>

commit 60210a3d86dc57ce4a76a366e7841dda746a33f7 upstream.

On ELF, (NOLOAD) sets the section type to SHT_NOBITS[1]. It is conceptually inappropriate for .plt, .got, and .got.plt sections which are always SHT_PROGBITS.

In GNU ld, if PLT entries are needed, .plt will be SHT_PROGBITS anyway and (NOLOAD) will be essentially ignored. In ld.lld, since
https://reviews.llvm.org/D118840 ("[ELF] Support (TYPE=<value>) to customize the output section type"), ld.lld will report a section type mismatch error (later changed to a warning). Just remove (NOLOAD) to fix the warning.

[1] https://lld.llvm.org/ELF/linker_script.html As of today, "The section should be marked as not loadable" on
https://sourceware.org/binutils/docs/ld/Output-Section-Type.html is outdated for ELF.

Fixes ab1ef68e5401 ("RISC-V: Add sections of PLT and GOT for kernel module")
---
 arch/riscv/include/asm/module.lds.h |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
arch/riscv/include/asm/module.lds.h +3 -3
@@ -2,8 +2,8 @@
 /* Copyright (C) 2017 Andes Technology Corporation */
 #ifdef CONFIG_MODULE_SECTIONS
 SECTIONS {
-	.plt (NOLOAD) : { BYTE(0) }
-	.got (NOLOAD) : { BYTE(0) }
-	.got.plt (NOLOAD) : { BYTE(0) }
+	.plt : { BYTE(0) }
+	.got : { BYTE(0) }
+	.got.plt : { BYTE(0) }
 }
 #endif
author maintainer
#1062 lore raw

PATCH 1058/1126 ACPI: CPPC: Avoid out of bounds access when parsing _CPC data

In reply to Greg Kroah-Hartman

From: Rafael J. Wysocki <[email protected]>

commit 40d8abf364bcab23bc715a9221a3c8623956257b upstream.

If the NumEntries field in the _CPC return package is less than 2, do not attempt to access the "Revision" element of that package, because it may not be present then.

---
 drivers/acpi/cppc_acpi.c |    5 +++++
 1 file changed, 5 insertions(+)
drivers/acpi/cppc_acpi.c +5 -0
@@ -676,6 +676,11 @@ int acpi_cppc_processor_probe(struct acp
 	cpc_obj = &out_obj->package.elements[0];
 	if (cpc_obj->type == ACPI_TYPE_INTEGER)	{
 		num_ent = cpc_obj->integer.value;
+		if (num_ent <= 1) {
+			pr_debug("Unexpected _CPC NumEntries value (%d) for CPU:%d\n",
+				 num_ent, pr->id);
+			goto out_free;
+		}
 	} else {
 		pr_debug("Unexpected entry type(%d) for NumEntries\n",
 				cpc_obj->type);
author maintainer
#1063 lore raw

PATCH 1059/1126 vhost: handle error while adding split ranges to iotlb

In reply to Greg Kroah-Hartman

From: Anirudh Rayabharam <[email protected]>

commit 03a91c9af2c42ae14afafb829a4b7e6589ab5892 upstream.

vhost_iotlb_add_range_ctx() handles the range [0, ULONG_MAX] by splitting it into two ranges and adding them separately. The return value of adding the first range to the iotlb is currently ignored. Check the return value and bail out in case of an error.

---
 drivers/vhost/iotlb.c |    6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
drivers/vhost/iotlb.c +5 -1
@@ -62,8 +62,12 @@ int vhost_iotlb_add_range_ctx(struct vho
 	 */
 	if (start == 0 && last == ULONG_MAX) {
 		u64 mid = last / 2;
+		int err = vhost_iotlb_add_range_ctx(iotlb, start, mid, addr,
+				perm, opaque);
+
+		if (err)
+			return err;
 
-		vhost_iotlb_add_range_ctx(iotlb, start, mid, addr, perm, opaque);
 		addr += mid + 1;
 		start = mid + 1;
 	}
author maintainer
#1064 lore raw

PATCH 1060/1126 spi: Fix Tegra QSPI example

In reply to Greg Kroah-Hartman

From: Jon Hunter <[email protected]>

commit 320689a1b543ca1396b3ed43bb18045e4a7ffd79 upstream.

When running dt_binding_check on the nvidia,tegra210-quad.yaml binding document the following error is reported ...

nvidia,tegra210-quad.example.dt.yaml:0:0: /example-0/spi@70410000/flash@0:
failed to match any schema with compatible: ['spi-nor']

Update the example in the binding document to fix the above error.

Fixes 9684752e5fe3 ("dt-bindings: spi: Add Tegra Quad SPI device tree binding")
---
 Documentation/devicetree/bindings/spi/nvidia,tegra210-quad.yaml |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
Documentation/devicetree/bindings/spi/nvidia,tegra210-quad.yaml +1 -1
@@ -106,7 +106,7 @@ examples:
             dma-names = "rx", "tx";
 
             flash@0 {
-                    compatible = "spi-nor";
+                    compatible = "jedec,spi-nor";
                     reg = <0>;
                     spi-max-frequency = <104000000>;
                     spi-tx-bus-width = <2>;
author maintainer
#1065 lore raw

PATCH 1061/1126 platform/chrome: cros_ec_typec: Check for EC device

In reply to Greg Kroah-Hartman

From: Prashant Malani <[email protected]>

commit ffebd90532728086007038986900426544e3df4e upstream.

The Type C ACPI device on older Chromebooks is not generated correctly (since their EC firmware doesn't support the new commands required). In such cases, the crafted ACPI device doesn't have an EC parent, and it is therefore not useful (it shouldn't be generated in the first place since the EC firmware doesn't support any of the Type C commands).

To handle devices which use these older firmware revisions, check for the parent EC device handle, and fail the probe if it's not found.

---
 drivers/platform/chrome/cros_ec_typec.c |    6 ++++++
 1 file changed, 6 insertions(+)
drivers/platform/chrome/cros_ec_typec.c +6 -0
@@ -1075,7 +1075,13 @@ static int cros_typec_probe(struct platf
 		return -ENOMEM;
 
 	typec->dev = dev;
+
 	typec->ec = dev_get_drvdata(pdev->dev.parent);
+	if (!typec->ec) {
+		dev_err(dev, "couldn't find parent EC device\n");
+		return -ENODEV;
+	}
+
 	platform_set_drvdata(pdev, typec);
 
 	ret = cros_typec_get_cmd_version(typec);
author maintainer
#1066 lore raw

PATCH 1062/1126 platform/x86: asus-wmi: Fix regression when probing for fan curve control

In reply to Greg Kroah-Hartman

From: Hans de Goede <[email protected]>

commit d717e4509af0380a94dbc28b61839df39f17e1eb upstream.

The fan curve control patches introduced a regression for at least the TUF FX506 and possibly other TUF series laptops that do not have support for fan curve control.

As part of the probing process, asus_wmi_evaluate_method_buf is called to get the factory default fan curve . The WMI management function returns 0 on certain laptops to indicate lack of fan curve control instead of ASUS_WMI_UNSUPPORTED_METHOD. This 0 is transformed to -ENODATA which results in failure when probing.

Fixes 0f0ac158d28f ("platform/x86: asus-wmi: Add support for custom fan curves")
Reported-and-tested-by Abhijeet V <[email protected]>
---
 drivers/platform/x86/asus-wmi.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/platform/x86/asus-wmi.c +1 -1
@@ -2059,7 +2059,7 @@ static int fan_boost_mode_check_present(
 	err = asus_wmi_get_devstate(asus, ASUS_WMI_DEVID_FAN_BOOST_MODE,
 				    &result);
 	if (err) {
-		if (err == -ENODEV)
+		if (err == -ENODEV || err == -ENODATA)
 			return 0;
 		else
 			return err;
author maintainer
#1067 lore raw

PATCH 1063/1126 can: isotp: restore accidentally removed MSG_PEEK feature

In reply to Greg Kroah-Hartman

From: Oliver Hartkopp <[email protected]>

[ Upstream commit e382fea8ae54f5bb62869c6b69b33993d43adeca ]

In commit 42bf50a1795a ("can: isotp: support MSG_TRUNC flag when reading from socket") a new check for recvmsg flags has been introduced that only checked for the flags that are handled in isotp_recvmsg() itself.

This accidentally removed the MSG_PEEK feature flag which is processed later in the call chain in __skb_try_recv_from_queue().

Add MSG_PEEK to the set of valid flags to restore the feature.

---
 net/can/isotp.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/can/isotp.c +1 -1
index ad61342d2e16..a95d171b3a64 100644
@@ -1009,7 +1009,7 @@ static int isotp_recvmsg(struct socket *sock, struct msghdr *msg, size_t size,
 	int noblock = flags & MSG_DONTWAIT;
 	int ret = 0;
 
-	if (flags & ~(MSG_DONTWAIT | MSG_TRUNC))
+	if (flags & ~(MSG_DONTWAIT | MSG_TRUNC | MSG_PEEK))
 		return -EINVAL;
 
 	if (!so->bound)
Signature · 1 line
2.34.1
author maintainer
#1068 lore raw

PATCH 1064/1126 proc: bootconfig: Add null pointer check

In reply to Greg Kroah-Hartman

From: Lv Ruyi <[email protected]>

commit bed5b60bf67ccd8957b8c0558fead30c4a3f5d3f upstream.

kzalloc is a memory allocation function which can return NULL when some internal memory errors happen. It is safer to add null pointer check.

Fixes c1a3c36017d4 ("proc: bootconfig: Add /proc/bootconfig to show boot config list")
---
 fs/proc/bootconfig.c |    2 ++
 1 file changed, 2 insertions(+)
fs/proc/bootconfig.c +2 -0
@@ -32,6 +32,8 @@ static int __init copy_xbc_key_value_lis
 	int ret = 0;
 
 	key = kzalloc(XBC_KEYLEN_MAX, GFP_KERNEL);
+	if (!key)
+		return -ENOMEM;
 
 	xbc_for_each_key_value(leaf, val) {
 		ret = xbc_node_compose_key(leaf, key, XBC_KEYLEN_MAX);
author maintainer
#1069 lore raw

PATCH 1065/1126 x86/fpu/xstate: Fix the ARCH_REQ_XCOMP_PERM implementation

In reply to Greg Kroah-Hartman

From: Yang Zhong <[email protected]>

commit 063452fd94d153d4eb38ad58f210f3d37a09cca4 upstream.

ARCH_REQ_XCOMP_PERM is supposed to add the requested feature to the permission bitmap of thread_group_leader()->fpu. But the code overwrites the bitmap with the requested feature bit only rather than adding it.

Fix the code to add the requested feature bit to the master bitmask.

Fixes db8268df0983 ("x86/arch_prctl: Add controls for dynamic XSTATE components")
Cc Paolo Bonzini <[email protected]>
---
 arch/x86/kernel/fpu/xstate.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/kernel/fpu/xstate.c +1 -1
@@ -1639,7 +1639,7 @@ static int __xstate_request_perm(u64 per
 
 	perm = guest ? &fpu->guest_perm : &fpu->perm;
 	/* Pairs with the READ_ONCE() in xstate_get_group_perm() */
-	WRITE_ONCE(perm->__state_perm, requested);
+	WRITE_ONCE(perm->__state_perm, mask);
 	/* Protected by sighand lock */
 	perm->__state_size = ksize;
 	perm->__user_state_size = usize;
author maintainer
#1070 lore raw

PATCH 1066/1126 x86/sev: Unroll string mmio with CC_ATTR_GUEST_UNROLL_STRING_IO

In reply to Greg Kroah-Hartman

From: Joerg Roedel <[email protected]>

commit 4009a4ac82dd95b8cd2b62bd30019476983f0aff upstream.

The io-specific memcpy/memset functions use string mmio accesses to do their work. Under SEV, the hypervisor can't emulate these instructions because they read/write directly from/to encrypted memory.

KVM will inject a page fault exception into the guest when it is asked to emulate string mmio instructions for an SEV guest:

  BUG: unable to handle page fault for address: ffffc90000065068
  #PF: supervisor read access in kernel mode
  #PF: error_code(0x0000) - not-present page
  PGD 8000100000067 P4D 8000100000067 PUD 80001000fb067 PMD 80001000fc067 PTE 80000000fed40173
  Oops: 0000 [#1] PREEMPT SMP NOPTI
  CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.17.0-rc7 #3

As string mmio for an SEV guest can not be supported by the hypervisor, unroll the instructions for CC_ATTR_GUEST_UNROLL_STRING_IO enabled kernels.

This issue appears when kernels are launched in recent libvirt-managed SEV virtual machines, because virt-install started to add a tpm-crb device to the guest by default and proactively because, raisins:

https://github.com/virt-manager/virt-manager/commit/eb58c09f488b0633ed1eea012cd311e48864401e

and as that commit says, the default adding of a TPM can be disabled with "virt-install ... --tpm none".

The kernel driver for tpm-crb uses memcpy_to/from_io() functions to access MMIO memory, resulting in a page-fault injected by KVM and crashing the kernel at boot.

[ bp: Massage and extend commit message. ]

Fixes d8aa7eea78a1 ('x86/mm: Add Secure Encrypted Virtualization (SEV) support')
---
 arch/x86/lib/iomem.c |   65 ++++++++++++++++++++++++++++++++++++++++++++-------
 1 file changed, 57 insertions(+), 8 deletions(-)
arch/x86/lib/iomem.c +57 -8
@@ -22,7 +22,7 @@ static __always_inline void rep_movs(voi
 		     : "memory");
 }
 
-void memcpy_fromio(void *to, const volatile void __iomem *from, size_t n)
+static void string_memcpy_fromio(void *to, const volatile void __iomem *from, size_t n)
 {
 	if (unlikely(!n))
 		return;
@@ -38,9 +38,8 @@ void memcpy_fromio(void *to, const volat
 	}
 	rep_movs(to, (const void *)from, n);
 }
-EXPORT_SYMBOL(memcpy_fromio);
 
-void memcpy_toio(volatile void __iomem *to, const void *from, size_t n)
+static void string_memcpy_toio(volatile void __iomem *to, const void *from, size_t n)
 {
 	if (unlikely(!n))
 		return;
@@ -56,14 +55,64 @@ void memcpy_toio(volatile void __iomem *
 	}
 	rep_movs((void *)to, (const void *) from, n);
 }
+
+static void unrolled_memcpy_fromio(void *to, const volatile void __iomem *from, size_t n)
+{
+	const volatile char __iomem *in = from;
+	char *out = to;
+	int i;
+
+	for (i = 0; i < n; ++i)
+		out[i] = readb(&in[i]);
+}
+
+static void unrolled_memcpy_toio(volatile void __iomem *to, const void *from, size_t n)
+{
+	volatile char __iomem *out = to;
+	const char *in = from;
+	int i;
+
+	for (i = 0; i < n; ++i)
+		writeb(in[i], &out[i]);
+}
+
+static void unrolled_memset_io(volatile void __iomem *a, int b, size_t c)
+{
+	volatile char __iomem *mem = a;
+	int i;
+
+	for (i = 0; i < c; ++i)
+		writeb(b, &mem[i]);
+}
+
+void memcpy_fromio(void *to, const volatile void __iomem *from, size_t n)
+{
+	if (cc_platform_has(CC_ATTR_GUEST_UNROLL_STRING_IO))
+		unrolled_memcpy_fromio(to, from, n);
+	else
+		string_memcpy_fromio(to, from, n);
+}
+EXPORT_SYMBOL(memcpy_fromio);
+
+void memcpy_toio(volatile void __iomem *to, const void *from, size_t n)
+{
+	if (cc_platform_has(CC_ATTR_GUEST_UNROLL_STRING_IO))
+		unrolled_memcpy_toio(to, from, n);
+	else
+		string_memcpy_toio(to, from, n);
+}
 EXPORT_SYMBOL(memcpy_toio);
 
 void memset_io(volatile void __iomem *a, int b, size_t c)
 {
-	/*
-	 * TODO: memset can mangle the IO patterns quite a bit.
-	 * perhaps it would be better to use a dumb one:
-	 */
-	memset((void *)a, b, c);
+	if (cc_platform_has(CC_ATTR_GUEST_UNROLL_STRING_IO)) {
+		unrolled_memset_io(a, b, c);
+	} else {
+		/*
+		 * TODO: memset can mangle the IO patterns quite a bit.
+		 * perhaps it would be better to use a dumb one:
+		 */
+		memset((void *)a, b, c);
+	}
 }
 EXPORT_SYMBOL(memset_io);
author maintainer
#1071 lore raw

PATCH 1067/1126 drm/connector: Fix typo in documentation

In reply to Greg Kroah-Hartman

From: Maxime Ripard <[email protected]>

commit dca384a3bf5af1c781cfa6aec63904bdb5018c36 upstream.

Commit 4adc33f36d80 ("drm/edid: Split deep color modes between RGB and YUV444") introduced two new variables in struct drm_display_info and their documentation, but the documentation part had a typo resulting in a doc build warning.

---
 include/drm/drm_connector.h |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
include/drm/drm_connector.h +2 -2
@@ -592,13 +592,13 @@ struct drm_display_info {
 	bool rgb_quant_range_selectable;
 
 	/**
-	 * @edid_hdmi_dc_rgb444_modes: Mask of supported hdmi deep color modes
+	 * @edid_hdmi_rgb444_dc_modes: Mask of supported hdmi deep color modes
 	 * in RGB 4:4:4. Even more stuff redundant with @bus_formats.
 	 */
 	u8 edid_hdmi_rgb444_dc_modes;
 
 	/**
-	 * @edid_hdmi_dc_ycbcr444_modes: Mask of supported hdmi deep color
+	 * @edid_hdmi_ycbcr444_dc_modes: Mask of supported hdmi deep color
 	 * modes in YCbCr 4:4:4. Even more stuff redundant with @bus_formats.
 	 */
 	u8 edid_hdmi_ycbcr444_dc_modes;
author maintainer
#1072 lore raw

PATCH 1068/1126 scsi: qla2xxx: Add qla2x00_async_done() for async routines

In reply to Greg Kroah-Hartman

From: Saurav Kashyap <[email protected]>

commit 49b729f58e7a98a006a8a0c1dcca8a1a4f58d2a8 upstream.

This done routine will delete the timer and check for its return value and decrease the reference count accordingly. This prevents boot hangs reported after commit 31e6cdbe0eae ("scsi: qla2xxx: Implement ref count for SRB") was merged.

---
 drivers/scsi/qla2xxx/qla_iocb.c |   17 ++++++++++++++++-
 1 file changed, 16 insertions(+), 1 deletion(-)
drivers/scsi/qla2xxx/qla_iocb.c +16 -1
@@ -2560,6 +2560,20 @@ qla24xx_tm_iocb(srb_t *sp, struct tsk_mg
 	}
 }
 
+static void
+qla2x00_async_done(struct srb *sp, int res)
+{
+	if (del_timer(&sp->u.iocb_cmd.timer)) {
+		/*
+		 * Successfully cancelled the timeout handler
+		 * ref: TMR
+		 */
+		if (kref_put(&sp->cmd_kref, qla2x00_sp_release))
+			return;
+	}
+	sp->async_done(sp, res);
+}
+
 void
 qla2x00_sp_release(struct kref *kref)
 {
@@ -2573,7 +2587,8 @@ qla2x00_init_async_sp(srb_t *sp, unsigne
 		     void (*done)(struct srb *sp, int res))
 {
 	timer_setup(&sp->u.iocb_cmd.timer, qla2x00_sp_timeout, 0);
-	sp->done = done;
+	sp->done = qla2x00_async_done;
+	sp->async_done = done;
 	sp->free = qla2x00_sp_free;
 	sp->u.iocb_cmd.timeout = qla2x00_async_iocb_timeout;
 	sp->u.iocb_cmd.timer.expires = jiffies + tmo * HZ;
author maintainer
#1073 lore raw

PATCH 1069/1126 staging: mt7621-dts: fix pinctrl-0 items to be size-1 items on ethernet

In reply to Greg Kroah-Hartman

From: Arınç ÜNAL <[email protected]>

commit 25e4f5220efead592c83200241e098e757d37e1f upstream.

Fix pinctrl-0 items under the ethernet node to be size-1 items. Current notation would be used on specifications with non-zero cells.

Fixes 0a93c0d75809 ("staging: mt7621-dts: fix pinctrl properties for ethernet")
---
 drivers/staging/mt7621-dts/mt7621.dtsi |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/staging/mt7621-dts/mt7621.dtsi +1 -1
@@ -326,7 +326,7 @@
 		mediatek,ethsys = <&sysc>;
 
 		pinctrl-names = "default";
-		pinctrl-0 = <&rgmii1_pins &rgmii2_pins &mdio_pins>;
+		pinctrl-0 = <&mdio_pins>, <&rgmii1_pins>, <&rgmii2_pins>;
 
 		gmac0: mac@0 {
 			compatible = "mediatek,eth-mac";
author maintainer
#1074 lore raw

PATCH 1070/1126 docs: fix make htmldocs warning in SCTP.rst

In reply to Greg Kroah-Hartman

From: Wan Jiabing <[email protected]>

commit 70868c6b8fd80db585da57a264c50a69af8fd3c3 upstream.

Fix following 'make htmldocs' warnings:
./Documentation/security/SCTP.rst:123: WARNING: Title underline too short. security_sctp_assoc_established()
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
./Documentation/security/SCTP.rst:123: WARNING: Title underline too short. security_sctp_assoc_established()
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
./Documentation/security/SCTP.rst:273: WARNING: Title underline too short. security_sctp_assoc_established()
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
./Documentation/security/SCTP.rst:273: WARNING: Title underline too short. security_sctp_assoc_established()
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Fixes 5e50f5d4ff31 ("security: add sctp_assoc_established hook")
---
 Documentation/security/SCTP.rst |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
Documentation/security/SCTP.rst +2 -2
@@ -120,7 +120,7 @@ calls **sctp_peeloff**\(3).
 
 
 security_sctp_assoc_established()
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 Called when a COOKIE ACK is received, and the peer secid will be
 saved into ``@asoc->peer_secid`` for client::
 
@@ -270,7 +270,7 @@ sockets sid and peer sid to that contain
 
 
 security_sctp_assoc_established()
-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 Called when a COOKIE ACK is received where it sets the connection's peer sid
 to that in ``@skb``::
 
author maintainer
#1075 lore raw

PATCH 1071/1126 arm64: mm: Drop const from conditional arm64_dma_phys_limit definition

In reply to Greg Kroah-Hartman

From: Will Deacon <[email protected]>

commit 770093459b9b333380aa71f2c31c60b14895c1df upstream.

Commit 031495635b46 ("arm64: Do not defer reserve_crashkernel() for platforms with no DMA memory zones") introduced different definitions for 'arm64_dma_phys_limit' depending on CONFIG_ZONE_DMA{,32} based on a late suggestion from Pasha. Sadly, this results in a build error when passing W=1:

| arch/arm64/mm/init.c:90:19: error: conflicting type qualifiers for 'arm64_dma_phys_limit'

Drop the 'const' for now and use '__ro_after_init' consistently.

---
 arch/arm64/mm/init.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm64/mm/init.c +1 -1
@@ -87,7 +87,7 @@ EXPORT_SYMBOL(memstart_addr);
 #if IS_ENABLED(CONFIG_ZONE_DMA) || IS_ENABLED(CONFIG_ZONE_DMA32)
 phys_addr_t __ro_after_init arm64_dma_phys_limit;
 #else
-const phys_addr_t arm64_dma_phys_limit = PHYS_MASK + 1;
+phys_addr_t __ro_after_init arm64_dma_phys_limit = PHYS_MASK + 1;
 #endif
 
 #ifdef CONFIG_KEXEC_CORE
author maintainer
#1076 lore raw

PATCH 1072/1126 ASoC: soc-compress: Change the check for codec_dai

In reply to Greg Kroah-Hartman

From: Jiasheng Jiang <[email protected]>

commit ccb4214f7f2a8b75acf493f31128e464ee1a3536 upstream.

It should be better to reverse the check on codec_dai and returned early in order to be easier to understand.

---
 sound/soc/soc-compress.c |   19 +++++++++++--------
 1 file changed, 11 insertions(+), 8 deletions(-)
sound/soc/soc-compress.c +11 -8
@@ -567,16 +567,19 @@ int snd_soc_new_compress(struct snd_soc_
 		return -EINVAL;
 	}
 
-	/* check client and interface hw capabilities */
-	if (codec_dai) {
-		if (snd_soc_dai_stream_valid(codec_dai, SNDRV_PCM_STREAM_PLAYBACK) &&
-		    snd_soc_dai_stream_valid(cpu_dai,   SNDRV_PCM_STREAM_PLAYBACK))
-			playback = 1;
-		if (snd_soc_dai_stream_valid(codec_dai, SNDRV_PCM_STREAM_CAPTURE) &&
-		    snd_soc_dai_stream_valid(cpu_dai,   SNDRV_PCM_STREAM_CAPTURE))
-			capture = 1;
+	if (!codec_dai) {
+		dev_err(rtd->card->dev, "Missing codec\n");
+		return -EINVAL;
 	}
 
+	/* check client and interface hw capabilities */
+	if (snd_soc_dai_stream_valid(codec_dai, SNDRV_PCM_STREAM_PLAYBACK) &&
+	    snd_soc_dai_stream_valid(cpu_dai,   SNDRV_PCM_STREAM_PLAYBACK))
+		playback = 1;
+	if (snd_soc_dai_stream_valid(codec_dai, SNDRV_PCM_STREAM_CAPTURE) &&
+	    snd_soc_dai_stream_valid(cpu_dai,   SNDRV_PCM_STREAM_CAPTURE))
+		capture = 1;
+
 	/*
 	 * Compress devices are unidirectional so only one of the directions
 	 * should be set, check for that (xor)
author maintainer
#1077 lore raw

PATCH 1073/1126 KVM: x86: SVM: fix avic spec based definitions again

In reply to Greg Kroah-Hartman

From: Maxim Levitsky <[email protected]>

commit 0dacc3df898e219fa774f39e5e10d686364e0a27 upstream.

Due to wrong rebase, commit
4a204f7895878 ("KVM: SVM: Allow AVIC support on system w/ physical APIC ID > 255")

moved avic spec #defines back to avic.c.

Move them back, and while at it extend AVIC_DOORBELL_PHYSICAL_ID_MASK to 12 bits as well (it will be used in nested avic)

Message-Id: <[email protected]>

---
 arch/x86/include/asm/svm.h |    8 +++++---
 arch/x86/kvm/svm/svm.h     |   11 -----------
 2 files changed, 5 insertions(+), 14 deletions(-)
arch/x86/include/asm/svm.h +5 -3
@@ -222,7 +222,7 @@ struct __attribute__ ((__packed__)) vmcb
 
 
 /* AVIC */
-#define AVIC_LOGICAL_ID_ENTRY_GUEST_PHYSICAL_ID_MASK	(0xFF)
+#define AVIC_LOGICAL_ID_ENTRY_GUEST_PHYSICAL_ID_MASK	(0xFFULL)
 #define AVIC_LOGICAL_ID_ENTRY_VALID_BIT			31
 #define AVIC_LOGICAL_ID_ENTRY_VALID_MASK		(1 << 31)
 
@@ -230,9 +230,11 @@ struct __attribute__ ((__packed__)) vmcb
 #define AVIC_PHYSICAL_ID_ENTRY_BACKING_PAGE_MASK	(0xFFFFFFFFFFULL << 12)
 #define AVIC_PHYSICAL_ID_ENTRY_IS_RUNNING_MASK		(1ULL << 62)
 #define AVIC_PHYSICAL_ID_ENTRY_VALID_MASK		(1ULL << 63)
-#define AVIC_PHYSICAL_ID_TABLE_SIZE_MASK		(0xFF)
+#define AVIC_PHYSICAL_ID_TABLE_SIZE_MASK		(0xFFULL)
 
-#define AVIC_DOORBELL_PHYSICAL_ID_MASK			(0xFF)
+#define AVIC_DOORBELL_PHYSICAL_ID_MASK			GENMASK_ULL(11, 0)
+
+#define VMCB_AVIC_APIC_BAR_MASK				0xFFFFFFFFFF000ULL
 
 #define AVIC_UNACCEL_ACCESS_WRITE_MASK		1
 #define AVIC_UNACCEL_ACCESS_OFFSET_MASK		0xFF0
arch/x86/kvm/svm/svm.h +0 -11
@@ -558,17 +558,6 @@ extern struct kvm_x86_nested_ops svm_nes
 
 /* avic.c */
 
-#define AVIC_LOGICAL_ID_ENTRY_GUEST_PHYSICAL_ID_MASK	(0xFF)
-#define AVIC_LOGICAL_ID_ENTRY_VALID_BIT			31
-#define AVIC_LOGICAL_ID_ENTRY_VALID_MASK		(1 << 31)
-
-#define AVIC_PHYSICAL_ID_ENTRY_HOST_PHYSICAL_ID_MASK	GENMASK_ULL(11, 0)
-#define AVIC_PHYSICAL_ID_ENTRY_BACKING_PAGE_MASK	(0xFFFFFFFFFFULL << 12)
-#define AVIC_PHYSICAL_ID_ENTRY_IS_RUNNING_MASK		(1ULL << 62)
-#define AVIC_PHYSICAL_ID_ENTRY_VALID_MASK		(1ULL << 63)
-
-#define VMCB_AVIC_APIC_BAR_MASK		0xFFFFFFFFFF000ULL
-
 int avic_ga_log_notifier(u32 ga_tag);
 void avic_vm_destroy(struct kvm *kvm);
 int avic_vm_init(struct kvm *kvm);
author maintainer
#1078 lore raw

PATCH 1074/1126 ax25: fix UAF bug in ax25_send_control()

In reply to Greg Kroah-Hartman

From: Duoming Zhou <[email protected]>

commit 5352a761308397a0e6250fdc629bb3f615b94747 upstream.

There are UAF bugs in ax25_send_control(), when we call ax25_release() to deallocate ax25_dev. The possible race condition is shown below:

      (Thread 1)              |     (Thread 2)
ax25_dev_device_up() //(1)    |
                              | ax25_kill_by_device()
ax25_bind()          //(2)    |
ax25_connect()                | ...
 ax25->state = AX25_STATE_1   |
 ...                          | ax25_dev_device_down() //(3)
      (Thread 3)
ax25_release()                |
 ax25_dev_put()  //(4) FREE   |
 case AX25_STATE_1:           |
  ax25_send_control()         |
   alloc_skb()       //USE    |

The refcount of ax25_dev increases in position (1) and (2), and decreases in position (3) and (4). The ax25_dev will be freed before dereference sites in ax25_send_control().

The following is part of the report:

[  102.297448] BUG: KASAN: use-after-free in ax25_send_control+0x33/0x210
[  102.297448] Read of size 8 at addr ffff888009e6e408 by task ax25_close/602
[  102.297448] Call Trace:
[  102.303751]  ax25_send_control+0x33/0x210
[  102.303751]  ax25_release+0x356/0x450
[  102.305431]  __sock_release+0x6d/0x120
[  102.305431]  sock_close+0xf/0x20
[  102.305431]  __fput+0x11f/0x420
[  102.305431]  task_work_run+0x86/0xd0
[  102.307130]  get_signal+0x1075/0x1220
[  102.308253]  arch_do_signal_or_restart+0x1df/0xc00
[  102.308253]  exit_to_user_mode_prepare+0x150/0x1e0
[  102.308253]  syscall_exit_to_user_mode+0x19/0x50
[  102.308253]  do_syscall_64+0x48/0x90
[  102.308253]  entry_SYSCALL_64_after_hwframe+0x44/0xae
[  102.308253] RIP: 0033:0x405ae7

This patch defers the free operation of ax25_dev and net_device after all corresponding dereference sites in ax25_release() to avoid UAF.

Fixes 9fd75b66b8f6 ("ax25: Fix refcount leaks caused by ax25_cb_del()")
---
 net/ax25/af_ax25.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
net/ax25/af_ax25.c +4 -4
@@ -991,10 +991,6 @@ static int ax25_release(struct socket *s
 	sock_orphan(sk);
 	ax25 = sk_to_ax25(sk);
 	ax25_dev = ax25->ax25_dev;
-	if (ax25_dev) {
-		dev_put_track(ax25_dev->dev, &ax25_dev->dev_tracker);
-		ax25_dev_put(ax25_dev);
-	}
 
 	if (sk->sk_type == SOCK_SEQPACKET) {
 		switch (ax25->state) {
@@ -1056,6 +1052,10 @@ static int ax25_release(struct socket *s
 		sk->sk_state_change(sk);
 		ax25_destroy_socket(ax25);
 	}
+	if (ax25_dev) {
+		dev_put_track(ax25_dev->dev, &ax25_dev->dev_tracker);
+		ax25_dev_put(ax25_dev);
+	}
 
 	sock->sk   = NULL;
 	release_sock(sk);
author maintainer
#1079 lore raw

PATCH 1075/1126 Reinstate some of "swiotlb: rework "fix info leak with DMA_FROM_DEVICE""

In reply to Greg Kroah-Hartman

From: Linus Torvalds <[email protected]>

commit 901c7280ca0d5e2b4a8929fbe0bfb007ac2a6544 upstream.

Halil Pasic points out [1] that the full revert of that commit (revert in bddac7c1e02b), and that a partial revert that only reverts the problematic case, but still keeps some of the cleanups is probably better. 

And that partial revert [2] had already been verified by Oleksandr Natalenko to also fix the issue, I had just missed that in the long discussion.

So let's reinstate the cleanups from commit aa6f8dcbab47 ("swiotlb:
rework "fix info leak with DMA_FROM_DEVICE""), and effectively only revert the part that caused problems.

---
 Documentation/core-api/dma-attributes.rst |    8 --------
 include/linux/dma-mapping.h               |    8 --------
 kernel/dma/swiotlb.c                      |   12 ++++++++----
 3 files changed, 8 insertions(+), 20 deletions(-)
Documentation/core-api/dma-attributes.rst +0 -8
@@ -130,11 +130,3 @@ accesses to DMA buffers in both privileg
 subsystem that the buffer is fully accessible at the elevated privilege
 level (and ideally inaccessible or at least read-only at the
 lesser-privileged levels).
-
-DMA_ATTR_OVERWRITE
-------------------
-
-This is a hint to the DMA-mapping subsystem that the device is expected to
-overwrite the entire mapped size, thus the caller does not require any of the
-previous buffer contents to be preserved. This allows bounce-buffering
-implementations to optimise DMA_FROM_DEVICE transfers.
include/linux/dma-mapping.h +0 -8
@@ -62,14 +62,6 @@
 #define DMA_ATTR_PRIVILEGED		(1UL << 9)
 
 /*
- * This is a hint to the DMA-mapping subsystem that the device is expected
- * to overwrite the entire mapped size, thus the caller does not require any
- * of the previous buffer contents to be preserved. This allows
- * bounce-buffering implementations to optimise DMA_FROM_DEVICE transfers.
- */
-#define DMA_ATTR_OVERWRITE		(1UL << 10)
-
-/*
  * A dma_addr_t can hold any valid DMA or bus address for the platform.  It can
  * be given to a device to use as a DMA source or target.  It is specific to a
  * given device and there may be a translation between the CPU physical address
kernel/dma/swiotlb.c +8 -4
@@ -627,10 +627,14 @@ phys_addr_t swiotlb_tbl_map_single(struc
 	for (i = 0; i < nr_slots(alloc_size + offset); i++)
 		mem->slots[index + i].orig_addr = slot_addr(orig_addr, i);
 	tlb_addr = slot_addr(mem->start, index) + offset;
-	if (!(attrs & DMA_ATTR_SKIP_CPU_SYNC) &&
-	    (!(attrs & DMA_ATTR_OVERWRITE) || dir == DMA_TO_DEVICE ||
-	    dir == DMA_BIDIRECTIONAL))
-		swiotlb_bounce(dev, tlb_addr, mapping_size, DMA_TO_DEVICE);
+	/*
+	 * When dir == DMA_FROM_DEVICE we could omit the copy from the orig
+	 * to the tlb buffer, if we knew for sure the device will
+	 * overwirte the entire current content. But we don't. Thus
+	 * unconditional bounce may prevent leaking swiotlb content (i.e.
+	 * kernel memory) to user-space.
+	 */
+	swiotlb_bounce(dev, tlb_addr, mapping_size, DMA_TO_DEVICE);
 	return tlb_addr;
 }
 
author maintainer
#1080 lore raw

PATCH 1076/1126 tracing: Have type enum modifications copy the strings

In reply to Greg Kroah-Hartman

From: Steven Rostedt (Google) <[email protected]>

commit 795301d3c28996219d555023ac6863401b6076bc upstream.

When an enum is used in the visible parts of a trace event that is exported to user space, the user space applications like perf and trace-cmd do not have a way to know what the value of the enum is. To solve this, at boot up (or module load) the printk formats are modified to replace the enum with their numeric value in the string output.

Array fields of the event are defined by [<nr-elements>] in the type portion of the format file so that the user space parsers can correctly parse the array into the appropriate size chunks. But in some trace events, an enum is used in defining the size of the array, which once again breaks the parsing of user space tooling.

This was solved the same way as the print formats were, but it modified the type strings of the trace event. This caused crashes in some architectures because, as supposed to the print string, is a const string value. This was not detected on x86, as it appears that const strings are still writable (at least in boot up), but other architectures this is not the case, and writing to a const string will cause a kernel fault.

To fix this, use kstrdup() to copy the type before modifying it. If the trace event is for the core kernel there's no need to free it because the string will be in use for the life of the machine being on line. For modules, create a link list to store all the strings being allocated for modules and when the module is removed, free them.

Fixes b3bc8547d3be ("tracing: Have TRACE_DEFINE_ENUM affect trace event types as well")
---
 kernel/trace/trace_events.c |   62 +++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 61 insertions(+), 1 deletion(-)
kernel/trace/trace_events.c +61 -1
@@ -40,6 +40,14 @@ static LIST_HEAD(ftrace_generic_fields);
 static LIST_HEAD(ftrace_common_fields);
 static bool eventdir_initialized;
 
+static LIST_HEAD(module_strings);
+
+struct module_string {
+	struct list_head	next;
+	struct module		*module;
+	char			*str;
+};
+
 #define GFP_TRACE (GFP_KERNEL | __GFP_ZERO)
 
 static struct kmem_cache *field_cachep;
@@ -2637,14 +2645,40 @@ static void update_event_printk(struct t
 	}
 }
 
+static void add_str_to_module(struct module *module, char *str)
+{
+	struct module_string *modstr;
+
+	modstr = kmalloc(sizeof(*modstr), GFP_KERNEL);
+
+	/*
+	 * If we failed to allocate memory here, then we'll just
+	 * let the str memory leak when the module is removed.
+	 * If this fails to allocate, there's worse problems than
+	 * a leaked string on module removal.
+	 */
+	if (WARN_ON_ONCE(!modstr))
+		return;
+
+	modstr->module = module;
+	modstr->str = str;
+
+	list_add(&modstr->next, &module_strings);
+}
+
 static void update_event_fields(struct trace_event_call *call,
 				struct trace_eval_map *map)
 {
 	struct ftrace_event_field *field;
 	struct list_head *head;
 	char *ptr;
+	char *str;
 	int len = strlen(map->eval_string);
 
+	/* Dynamic events should never have field maps */
+	if (WARN_ON_ONCE(call->flags & TRACE_EVENT_FL_DYNAMIC))
+		return;
+
 	head = trace_get_fields(call);
 	list_for_each_entry(field, head, link) {
 		ptr = strchr(field->type, '[');
@@ -2658,9 +2692,26 @@ static void update_event_fields(struct t
 		if (strncmp(map->eval_string, ptr, len) != 0)
 			continue;
 
+		str = kstrdup(field->type, GFP_KERNEL);
+		if (WARN_ON_ONCE(!str))
+			return;
+		ptr = str + (ptr - field->type);
 		ptr = eval_replace(ptr, map, len);
 		/* enum/sizeof string smaller than value */
-		WARN_ON_ONCE(!ptr);
+		if (WARN_ON_ONCE(!ptr)) {
+			kfree(str);
+			continue;
+		}
+
+		/*
+		 * If the event is part of a module, then we need to free the string
+		 * when the module is removed. Otherwise, it will stay allocated
+		 * until a reboot.
+		 */
+		if (call->module)
+			add_str_to_module(call->module, str);
+
+		field->type = str;
 	}
 }
 
@@ -2885,6 +2936,7 @@ static void trace_module_add_events(stru
 static void trace_module_remove_events(struct module *mod)
 {
 	struct trace_event_call *call, *p;
+	struct module_string *modstr, *m;
 
 	down_write(&trace_event_sem);
 	list_for_each_entry_safe(call, p, &ftrace_events, list) {
@@ -2893,6 +2945,14 @@ static void trace_module_remove_events(s
 		if (call->module == mod)
 			__trace_remove_event_call(call);
 	}
+	/* Check for any strings allocade for this module */
+	list_for_each_entry_safe(modstr, m, &module_strings, next) {
+		if (modstr->module != mod)
+			continue;
+		list_del(&modstr->next);
+		kfree(modstr->str);
+		kfree(modstr);
+	}
 	up_write(&trace_event_sem);
 
 	/*
author maintainer
#1081 lore raw

PATCH 1077/1126 mips: Enable KCSAN - take 2

In reply to Greg Kroah-Hartman

From: Nemanja Rakovic <[email protected]>

commit d1ca45f93c3f95d1590d60b012cf8fcf6db633ee upstream.

Replaces KASAN_SANITIZE with KCSAN_SANITIZE in
boot/compressed/Makefile.

Fixes e0a8b93efa23 mips: Enable KCSAN
---
 arch/mips/boot/compressed/Makefile |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/mips/boot/compressed/Makefile +1 -1
@@ -38,7 +38,7 @@ KBUILD_AFLAGS := $(KBUILD_AFLAGS) -D__AS
 KCOV_INSTRUMENT		:= n
 GCOV_PROFILE := n
 UBSAN_SANITIZE := n
-KASAN_SANITIZE			:= n
+KCSAN_SANITIZE			:= n
 
 # decompressor objects (linked with vmlinuz)
 vmlinuzobjs-y := $(obj)/head.o $(obj)/decompress.o $(obj)/string.o $(obj)/bswapsi.o
author maintainer
#1082 lore raw

PATCH 1078/1126 net: add skb_set_end_offset() helper

In reply to Greg Kroah-Hartman

From: Eric Dumazet <[email protected]>

commit 763087dab97547230a6807c865a6a5ae53a59247 upstream.

We have multiple places where this helper is convenient, and plan using it in the following patch.

---
 include/linux/skbuff.h |   10 ++++++++++
 net/core/skbuff.c      |   19 +++++--------------
 2 files changed, 15 insertions(+), 14 deletions(-)
include/linux/skbuff.h +10 -0
@@ -1475,6 +1475,11 @@ static inline unsigned int skb_end_offse
 {
 	return skb->end;
 }
+
+static inline void skb_set_end_offset(struct sk_buff *skb, unsigned int offset)
+{
+	skb->end = offset;
+}
 #else
 static inline unsigned char *skb_end_pointer(const struct sk_buff *skb)
 {
@@ -1485,6 +1490,11 @@ static inline unsigned int skb_end_offse
 {
 	return skb->end - skb->head;
 }
+
+static inline void skb_set_end_offset(struct sk_buff *skb, unsigned int offset)
+{
+	skb->end = skb->head + offset;
+}
 #endif
 
 /* Internal */
net/core/skbuff.c +5 -14
@@ -201,7 +201,7 @@ static void __build_skb_around(struct sk
 	skb->head = data;
 	skb->data = data;
 	skb_reset_tail_pointer(skb);
-	skb->end = skb->tail + size;
+	skb_set_end_offset(skb, size);
 	skb->mac_header = (typeof(skb->mac_header))~0U;
 	skb->transport_header = (typeof(skb->transport_header))~0U;
 
@@ -1736,11 +1736,10 @@ int pskb_expand_head(struct sk_buff *skb
 	skb->head     = data;
 	skb->head_frag = 0;
 	skb->data    += off;
+
+	skb_set_end_offset(skb, size);
 #ifdef NET_SKBUFF_DATA_USES_OFFSET
-	skb->end      = size;
 	off           = nhead;
-#else
-	skb->end      = skb->head + size;
 #endif
 	skb->tail	      += off;
 	skb_headers_offset_update(skb, nhead);
@@ -6044,11 +6043,7 @@ static int pskb_carve_inside_header(stru
 	skb->head = data;
 	skb->data = data;
 	skb->head_frag = 0;
-#ifdef NET_SKBUFF_DATA_USES_OFFSET
-	skb->end = size;
-#else
-	skb->end = skb->head + size;
-#endif
+	skb_set_end_offset(skb, size);
 	skb_set_tail_pointer(skb, skb_headlen(skb));
 	skb_headers_offset_update(skb, 0);
 	skb->cloned = 0;
@@ -6186,11 +6181,7 @@ static int pskb_carve_inside_nonlinear(s
 	skb->head = data;
 	skb->head_frag = 0;
 	skb->data = data;
-#ifdef NET_SKBUFF_DATA_USES_OFFSET
-	skb->end = size;
-#else
-	skb->end = skb->head + size;
-#endif
+	skb_set_end_offset(skb, size);
 	skb_reset_tail_pointer(skb);
 	skb_headers_offset_update(skb, 0);
 	skb->cloned   = 0;