Merge android16-6.12 into android16-6.12-lts

This merges the android16-6.12 branch into the -lts branch, catching
it up with the latest changes in there.

It contains the following commits:

* 5da94eacba ANDROID: rust_binder: print and avoid underflow with invalid refcnt
* 3744083cdb ANDROID: rust: miscdevice: fix formatting
* 4f7b72e026 FROMLIST: gendwarfksyms: order -T symtypes output by name
* 1493f0937f ANDROID: rust_binder: allow PollTable to be null
* 2a0e6416e5 FROMLIST: poll: rust: allow poll_table ptrs to be null
* 68f4f0b069 ANDROID: virt: gunyah: Fix ADDRSPACE_VMMIO_CONFIGURE hypercall handling
* 153a0a38e9 ANDROID: GKI: Update xiaomi symbol list.
* 111ee63e22 FROMGIT: dm: Check for forbidden splitting of zone write operations
* 177d6611a9 FROMGIT: dm: dm-crypt: Do not partially accept write BIOs with zoned targets
* 61e56b7483 FROMGIT: dm: Always split write BIOs to zoned device limits
* 5d1966b61d FROMGIT: block: Introduce bio_needs_zone_write_plugging()
* 5aca882326 BACKPORT: block: Make REQ_OP_ZONE_FINISH a write operation
* 1895d73589 ANDROID: "Revert: dm: Split zoned writes at the top of the stack"
* 8c8d462c9d ANDROID: GKI: add vendor_hook android_vh_health_report
* 1ccb219f8c ANDROID: GKI: update the pixel ABI symbol list
* 273b788415 FROMLIST: arm64: perf: Make exporting of pmu events configurable
* 22d634eb24 FROMLIST: KVM: arm64: Fix error path in init_hyp_mode()
* cd7cc3a247 ANDROID: rust_binder: do not take refcount during GET_NODE_INFO_FOR_REF
* d47caa0b28 ANDROID: rust_binder: reduce counts in NodeRef::absorb
* bc31899f7c ANDROID: KVM: arm64: Always unmap the pvmfw region at stage-2
* 453cb67e8f ANDROID: gki: Enable pkvm pviommu driver

Change-Id: If3192f301916efd2250924270d67b8ed0f82b0ef
Signed-off-by: Greg Kroah-Hartman <gregkh@google.com>

Documentation/admin-guide/kernel-parameters.txt

@@ -1541,6 +1541,10 @@
 			Format: { "fix" }
 			Permit 'security.evm' to be updated regardless of
 			current integrity status.
+	export_pmu_events
+			[KNL,ARM64] Sets the PMU export bit (PMCR_EL0.X), which enables
+			the exporting of events over an IMPLEMENTATION DEFINED PMU event
+			export bus to another device.
 
 	early_page_ext [KNL,EARLY] Enforces page_ext initialization to earlier
 			stages so cover more early boot allocations.

Documentation/admin-guide/sysctl/kernel.rst

@@ -286,6 +286,17 @@ domain names are in general different. For a detailed discussion
 see the ``hostname(1)`` man page.
 
 
+export_pmu_events (arm64 only)
+==============================
+
+Controls the PMU export bit (PMCR_EL0.X), which enables the exporting of
+events over an IMPLEMENTATION DEFINED PMU event export bus to another device.
+
+0: disables exporting of events (default).
+
+1: enables exporting of events.
+
+
 firmware_config
 ===============
 

arch/arm64/configs/gki_defconfig

@@ -626,6 +626,7 @@ CONFIG_HWSPINLOCK=y
 # CONFIG_SUN50I_ERRATUM_UNKNOWN1 is not set
 CONFIG_MAILBOX=y
 CONFIG_IOMMU_IO_PGTABLE_ARMV7S=y
+CONFIG_PKVM_PVIOMMU=y
 CONFIG_REMOTEPROC=y
 CONFIG_REMOTEPROC_CDEV=y
 CONFIG_RPMSG_CHAR=y

arch/arm64/kvm/arm.c

@@ -2391,7 +2391,9 @@ static void __init teardown_hyp_mode(void)
 	free_hyp_pgds();
 	for_each_possible_cpu(cpu) {
 		free_pages(per_cpu(kvm_arm_hyp_stack_base, cpu), NVHE_STACK_SHIFT - PAGE_SHIFT);
-		free_pages(kvm_nvhe_sym(kvm_arm_hyp_percpu_base)[cpu], nvhe_percpu_order());
+
+		if (!kvm_nvhe_sym(kvm_arm_hyp_percpu_base)[cpu])
+			continue;
 
 		if (free_sve) {
 			struct cpu_sve_state *sve_state;
@@ -2399,6 +2401,9 @@ static void __init teardown_hyp_mode(void)
 			sve_state = per_cpu_ptr_nvhe_sym(kvm_host_data, cpu)->sve_state;
 			free_pages((unsigned long) sve_state, pkvm_host_sve_state_order());
 		}
+
+		free_pages(kvm_nvhe_sym(kvm_arm_hyp_percpu_base)[cpu], nvhe_percpu_order());
+
 	}
 }
 

arch/arm64/kvm/pkvm.c

@@ -85,6 +85,7 @@ static void __init sort_memblock_regions(void)
 static int __init register_memblock_regions(void)
 {
 	struct memblock_region *reg;
+	bool pvmfw_in_mem = false;
 
 	for_each_mem_region(reg) {
 		if (*hyp_memblock_nr_ptr >= HYP_MEMBLOCK_REGIONS)
@@ -92,6 +93,26 @@ static int __init register_memblock_regions(void)
 
 		hyp_memory[*hyp_memblock_nr_ptr] = *reg;
 		(*hyp_memblock_nr_ptr)++;
+
+		if (!pvmfw_size || pvmfw_in_mem ||
+			!memblock_addrs_overlap(reg->base, reg->size, pvmfw_base, pvmfw_size))
+			continue;
+		/* If the pvmfw region overlaps a memblock, it must be a subset */
+		if (pvmfw_base < reg->base || (pvmfw_base + pvmfw_size) > (reg->base + reg->size))
+			return -EINVAL;
+		pvmfw_in_mem = true;
+	}
+
+	if (pvmfw_size && !pvmfw_in_mem) {
+		if (*hyp_memblock_nr_ptr >= HYP_MEMBLOCK_REGIONS)
+			return -ENOMEM;
+
+		hyp_memory[*hyp_memblock_nr_ptr] = (struct memblock_region) {
+			.base   = pvmfw_base,
+			.size   = pvmfw_size,
+			.flags  = MEMBLOCK_NOMAP,
+		};
+		(*hyp_memblock_nr_ptr)++;
 	}
 	sort_memblock_regions();
 

block/blk-mq.c

@@ -3139,8 +3139,10 @@ void blk_mq_submit_bio(struct bio *bio)
 	if (blk_mq_attempt_bio_merge(q, bio, nr_segs))
 		goto queue_exit;
 
-	if (blk_queue_is_zoned(q) && blk_zone_plug_bio(bio, nr_segs))
+	if (bio_needs_zone_write_plugging(bio)) {
-		goto queue_exit;
+		if (blk_zone_plug_bio(bio, nr_segs))
+			goto queue_exit;
+	}
 
 new_request:
 	if (!rq) {

block/blk-zoned.c

@@ -1131,25 +1131,7 @@ bool blk_zone_plug_bio(struct bio *bio, unsigned int nr_segs)
 {
 	struct block_device *bdev = bio->bi_bdev;
 
-	if (!bdev->bd_disk->zone_wplugs_hash)
+	if (WARN_ON_ONCE(!bdev->bd_disk->zone_wplugs_hash))
-		return false;
-
-	/*
-	 * If the BIO already has the plugging flag set, then it was already
-	 * handled through this path and this is a submission from the zone
-	 * plug bio submit work.
-	 */
-	if (bio_flagged(bio, BIO_ZONE_WRITE_PLUGGING))
-		return false;
-
-	/*
-	 * We do not need to do anything special for empty flush BIOs, e.g
-	 * BIOs such as issued by blkdev_issue_flush(). The is because it is
-	 * the responsibility of the user to first wait for the completion of
-	 * write operations for flush to have any effect on the persistence of
-	 * the written data.
-	 */
-	if (op_is_flush(bio->bi_opf) && !bio_sectors(bio))
 		return false;
 
 	/*

drivers/android/binder/node.rs

@@ -827,6 +827,22 @@ impl NodeRef {
         other.weak_count = 0;
         other.strong_node_count = 0;
         other.weak_node_count = 0;
+
+        if self.strong_node_count >= 2 || self.weak_node_count >= 2 {
+            let mut guard = self.node.owner.inner.lock();
+            let inner = self.node.inner.access_mut(&mut guard);
+
+            if self.strong_node_count >= 2 {
+                inner.strong.count -= self.strong_node_count - 1;
+                self.strong_node_count = 1;
+                assert_ne!(inner.strong.count, 0);
+            }
+            if self.weak_node_count >= 2 {
+                inner.weak.count -= self.weak_node_count - 1;
+                self.weak_node_count = 1;
+                assert_ne!(inner.weak.count, 0);
+            }
+        }
     }
 
     pub(crate) fn get_count(&self) -> (usize, usize) {
@@ -874,6 +890,13 @@ impl NodeRef {
             }
             *count += 1;
         } else {
+            if *count == 0 {
+                pr_warn!(
+                    "pid {} performed invalid decrement on ref\n",
+                    kernel::current!().pid()
+                );
+                return false;
+            }
             *count -= 1;
             if *count == 0 {
                 self.node.update_refcount(false, *node_count, strong);

drivers/android/binder/process.rs

@@ -926,6 +926,8 @@ impl Process {
                 refs.by_handle.remove(&handle);
                 refs.by_node.remove(&id);
             }
+        } else {
+            pr_warn!("{}: no such ref {handle}\n", kernel::current!().pid());
         }
         Ok(())
     }
@@ -1141,11 +1143,10 @@ impl Process {
             return Err(EPERM);
         }
 
-        let node_ref = self
-            .get_node_from_handle(out.handle, true)
-            .or(Err(EINVAL))?;
-        // Get the counts from the node.
         {
+            let mut node_refs = self.node_refs.lock();
+            let node_info = node_refs.by_handle.get_mut(&out.handle).ok_or(ENOENT)?;
+            let node_ref = node_info.node_ref();
             let owner_inner = node_ref.node.owner.inner.lock();
             node_ref.node.populate_counts(&mut out, &owner_inner);
         }
@@ -1634,7 +1635,7 @@ impl Process {
     pub(crate) fn poll(
         this: ArcBorrow<'_, Process>,
         file: &File,
-        table: &mut PollTable,
+        table: PollTable<'_>,
     ) -> Result<u32> {
         let thread = this.get_current_thread()?;
         let (from_proc, mut mask) = thread.poll(file, table);

drivers/android/binder/rust_binder.rs

@@ -472,7 +472,7 @@ unsafe extern "C" fn rust_binder_poll(
     // SAFETY: The caller ensures that the file is valid.
     let fileref = unsafe { File::from_raw_file(file) };
     // SAFETY: The caller ensures that the `PollTable` is valid.
-    match Process::poll(f, fileref, unsafe { PollTable::from_ptr(wait) }) {
+    match Process::poll(f, fileref, unsafe { PollTable::from_raw(wait) }) {
         Ok(v) => v,
         Err(_) => bindings::POLLERR,
     }

drivers/android/binder/thread.rs

@@ -1614,7 +1614,7 @@ impl Thread {
         ret
     }
 
-    pub(crate) fn poll(&self, file: &File, table: &mut PollTable) -> (bool, u32) {
+    pub(crate) fn poll(&self, file: &File, table: PollTable<'_>) -> (bool, u32) {
         table.register_wait(file, &self.work_condvar);
         let mut inner = self.inner.lock();
         (inner.should_use_process_work_queue(), inner.poll())

drivers/android/vendor_hooks.c

@@ -33,6 +33,7 @@
 #include <trace/hooks/pm_domain.h>
 #include <trace/hooks/cpuidle_psci.h>
 #include <trace/hooks/vmscan.h>
+#include <trace/hooks/health_report.h>
 #include <trace/hooks/avc.h>
 #include <trace/hooks/creds.h>
 #include <trace/hooks/selinux.h>
@@ -203,6 +204,7 @@ EXPORT_TRACEPOINT_SYMBOL_GPL(android_vh_tune_mmap_readaround);
 EXPORT_TRACEPOINT_SYMBOL_GPL(android_rvh_hw_protection_shutdown);
 EXPORT_TRACEPOINT_SYMBOL_GPL(android_vh_shrink_slab_bypass);
 EXPORT_TRACEPOINT_SYMBOL_GPL(android_vh_do_shrink_slab_ex);
+EXPORT_TRACEPOINT_SYMBOL_GPL(android_vh_health_report);
 EXPORT_TRACEPOINT_SYMBOL_GPL(android_vh_drain_all_pages_bypass);
 EXPORT_TRACEPOINT_SYMBOL_GPL(android_vh_security_audit_log_setid);
 EXPORT_TRACEPOINT_SYMBOL_GPL(android_rvh_selinux_avc_insert);

drivers/md/dm-crypt.c

@@ -253,17 +253,35 @@ MODULE_PARM_DESC(max_read_size, "Maximum size of a read request");
 static unsigned int max_write_size = 0;
 module_param(max_write_size, uint, 0644);
 MODULE_PARM_DESC(max_write_size, "Maximum size of a write request");
-static unsigned get_max_request_size(struct crypt_config *cc, bool wrt)
+
+static unsigned get_max_request_sectors(struct dm_target *ti, struct bio *bio)
 {
+	struct crypt_config *cc = ti->private;
 	unsigned val, sector_align;
-	val = !wrt ? READ_ONCE(max_read_size) : READ_ONCE(max_write_size);
+	bool wrt = op_is_write(bio_op(bio));
-	if (likely(!val))
+
-		val = !wrt ? DM_CRYPT_DEFAULT_MAX_READ_SIZE : DM_CRYPT_DEFAULT_MAX_WRITE_SIZE;
+	if (wrt) {
-	if (wrt || cc->used_tag_size) {
+		/*
-		if (unlikely(val > BIO_MAX_VECS << PAGE_SHIFT))
+		 * For zoned devices, splitting write operations creates the
-			val = BIO_MAX_VECS << PAGE_SHIFT;
+		 * risk of deadlocking queue freeze operations with zone write
+		 * plugging BIO work when the reminder of a split BIO is
+		 * issued. So always allow the entire BIO to proceed.
+		 */
+		if (ti->emulate_zone_append)
+			return bio_sectors(bio);
+
+		val = min_not_zero(READ_ONCE(max_write_size),
+				   DM_CRYPT_DEFAULT_MAX_WRITE_SIZE);
+	} else {
+		val = min_not_zero(READ_ONCE(max_read_size),
+				   DM_CRYPT_DEFAULT_MAX_READ_SIZE);
 	}
-	sector_align = max(bdev_logical_block_size(cc->dev->bdev), (unsigned)cc->sector_size);
+
+	if (wrt || cc->used_tag_size)
+		val = min(val, BIO_MAX_VECS << PAGE_SHIFT);
+
+	sector_align = max(bdev_logical_block_size(cc->dev->bdev),
+			   (unsigned)cc->sector_size);
 	val = round_down(val, sector_align);
 	if (unlikely(!val))
 		val = sector_align;
@@ -3517,7 +3535,7 @@ static int crypt_map(struct dm_target *ti, struct bio *bio)
 	/*
 	 * Check if bio is too large, split as needed.
 	 */
-	max_sectors = get_max_request_size(cc, bio_data_dir(bio) == WRITE);
+	max_sectors = get_max_request_sectors(ti, bio);
 	if (unlikely(bio_sectors(bio) > max_sectors))
 		dm_accept_partial_bio(bio, max_sectors);
 
@@ -3754,6 +3772,17 @@ static void crypt_io_hints(struct dm_target *ti, struct queue_limits *limits)
 		max_t(unsigned int, limits->physical_block_size, cc->sector_size);
 	limits->io_min = max_t(unsigned int, limits->io_min, cc->sector_size);
 	limits->dma_alignment = limits->logical_block_size - 1;
+
+	/*
+	 * For zoned dm-crypt targets, there will be no internal splitting of
+	 * write BIOs to avoid exceeding BIO_MAX_VECS vectors per BIO. But
+	 * without respecting this limit, crypt_alloc_buffer() will trigger a
+	 * BUG(). Avoid this by forcing DM core to split write BIOs to this
+	 * limit.
+	 */
+	if (ti->emulate_zone_append)
+		limits->max_hw_sectors = min(limits->max_hw_sectors,
+					     BIO_MAX_VECS << PAGE_SECTORS_SHIFT);
 }
 
 static struct target_type crypt_target = {

drivers/md/dm.c

@@ -1307,8 +1307,9 @@ out:
 /*
  * A target may call dm_accept_partial_bio only from the map routine.  It is
  * allowed for all bio types except REQ_PREFLUSH, REQ_OP_ZONE_* zone management
- * operations, REQ_OP_ZONE_APPEND (zone append writes) and any bio serviced by
+ * operations, zone append writes (native with REQ_OP_ZONE_APPEND or emulated
- * __send_duplicate_bios().
+ * with write BIOs flagged with BIO_EMULATES_ZONE_APPEND) and any bio serviced
+ * by __send_duplicate_bios().
  *
  * dm_accept_partial_bio informs the dm that the target only wants to process
  * additional n_sectors sectors of the bio and the rest of the data should be
@@ -1341,11 +1342,19 @@ void dm_accept_partial_bio(struct bio *bio, unsigned int n_sectors)
 	unsigned int bio_sectors = bio_sectors(bio);
 
 	BUG_ON(dm_tio_flagged(tio, DM_TIO_IS_DUPLICATE_BIO));
-	BUG_ON(op_is_zone_mgmt(bio_op(bio)));
-	BUG_ON(bio_op(bio) == REQ_OP_ZONE_APPEND);
 	BUG_ON(bio_sectors > *tio->len_ptr);
 	BUG_ON(n_sectors > bio_sectors);
 
+	if (static_branch_unlikely(&zoned_enabled) &&
+	    unlikely(bdev_is_zoned(bio->bi_bdev))) {
+		enum req_op op = bio_op(bio);
+
+		BUG_ON(op_is_zone_mgmt(op));
+		BUG_ON(op == REQ_OP_WRITE);
+		BUG_ON(op == REQ_OP_WRITE_ZEROES);
+		BUG_ON(op == REQ_OP_ZONE_APPEND);
+	}
+
 	*tio->len_ptr -= bio_sectors - n_sectors;
 	bio->bi_iter.bi_size = n_sectors << SECTOR_SHIFT;
 
@@ -1788,18 +1797,34 @@ static inline bool dm_zone_bio_needs_split(struct mapped_device *md,
 					   struct bio *bio)
 {
 	/*
-	 * For a mapped device that needs zone append emulation, we must
+	 * Special case the zone operations that cannot or should not be split.
-	 * split any large BIO that straddles zone boundaries. Additionally,
-	 * split sequential zoned writes to prevent that splitting lower in the
-	 * stack causes bio reordering.
 	 */
-	return ((dm_emulate_zone_append(md) && bio_straddles_zones(bio)) ||
+	switch (bio_op(bio)) {
-		(bio_op(bio) == REQ_OP_WRITE && bdev_is_zoned(bio->bi_bdev))) &&
+	case REQ_OP_ZONE_APPEND:
-	       !bio_flagged(bio, BIO_ZONE_WRITE_PLUGGING);
+	case REQ_OP_ZONE_FINISH:
+	case REQ_OP_ZONE_RESET:
+	case REQ_OP_ZONE_RESET_ALL:
+		return false;
+	default:
+		break;
+	}
+
+	/*
+	 * Mapped devices that require zone append emulation will use the block
+	 * layer zone write plugging. In such case, we must split any large BIO
+	 * to the mapped device limits to avoid potential deadlocks with queue
+	 * freeze operations.
+	 */
+	if (!dm_emulate_zone_append(md))
+		return false;
+	return bio_needs_zone_write_plugging(bio) || bio_straddles_zones(bio);
 }
+
 static inline bool dm_zone_plug_bio(struct mapped_device *md, struct bio *bio)
 {
-	return dm_emulate_zone_append(md) && blk_zone_plug_bio(bio, 0);
+	if (!bio_needs_zone_write_plugging(bio))
+		return false;
+	return blk_zone_plug_bio(bio, 0);
 }
 
 static blk_status_t __send_zone_reset_all_emulated(struct clone_info *ci,
@@ -1943,9 +1968,7 @@ static void dm_split_and_process_bio(struct mapped_device *md,
 
 	is_abnormal = is_abnormal_io(bio);
 	if (static_branch_unlikely(&zoned_enabled)) {
-		/* Special case REQ_OP_ZONE_RESET_ALL as it cannot be split. */
+		need_split = is_abnormal || dm_zone_bio_needs_split(md, bio);
-		need_split = (bio_op(bio) != REQ_OP_ZONE_RESET_ALL) &&
-			(is_abnormal || dm_zone_bio_needs_split(md, bio));
 	} else {
 		need_split = is_abnormal;
 	}

drivers/perf/arm_pmuv3.c

@@ -325,6 +325,7 @@ GEN_PMU_FORMAT_ATTR(threshold_compare);
 GEN_PMU_FORMAT_ATTR(threshold);
 
 static int sysctl_perf_user_access __read_mostly;
+static int sysctl_export_pmu_events __read_mostly;
 
 static bool armv8pmu_event_is_64bit(struct perf_event *event)
 {
@@ -1053,6 +1054,17 @@ static int armv8pmu_set_event_filter(struct hw_perf_event *event,
 	return 0;
 }
 
+static int __init export_pmu_events(char *str)
+{
+	/* Enable exporting of pmu events at early bootup with kernel
+	 * arguments.
+	 */
+	sysctl_export_pmu_events = 1;
+	return 0;
+}
+
+early_param("export_pmu_events", export_pmu_events);
+
 static void armv8pmu_reset(void *info)
 {
 	struct arm_pmu *cpu_pmu = (struct arm_pmu *)info;
@@ -1077,6 +1089,9 @@ static void armv8pmu_reset(void *info)
 	if (armv8pmu_has_long_event(cpu_pmu))
 		pmcr |= ARMV8_PMU_PMCR_LP;
 
+	if (sysctl_export_pmu_events)
+		pmcr |= ARMV8_PMU_PMCR_X;
+
 	armv8pmu_pmcr_write(pmcr);
 }
 
@@ -1280,6 +1295,15 @@ static struct ctl_table armv8_pmu_sysctl_table[] = {
 		.extra1		= SYSCTL_ZERO,
 		.extra2		= SYSCTL_ONE,
 	},
+	{
+		.procname       = "export_pmu_events",
+		.data           = &sysctl_export_pmu_events,
+		.maxlen         = sizeof(unsigned int),
+		.mode           = 0644,
+		.proc_handler   = proc_dointvec_minmax,
+		.extra1         = SYSCTL_ZERO,
+		.extra2         = SYSCTL_ONE,
+	},
 };
 
 static void armv8_pmu_register_sysctl_table(void)

drivers/virt/coco/gunyah-guest/gunyah-guest.c

@@ -33,7 +33,8 @@ static int gunyah_mmio_guard_ioremap_hook(phys_addr_t phys, size_t size, pgprot_
 	ret = gunyah_hypercall_addrspc_configure_vmmio_range(our_addrspace_capid,
 				phys, size, GUNYAH_ADDRSPACE_VMMIO_CONFIGURE_OP_ADD_RANGE);
 
-	if (ret == GUNYAH_ERROR_UNIMPLEMENTED || ret == GUNYAH_ERROR_BUSY)
+	if (ret == GUNYAH_ERROR_UNIMPLEMENTED || ret == GUNYAH_ERROR_BUSY
+			|| ret == GUNYAH_ERROR_CSPACE_INSUF_RIGHTS)
 		/* Gunyah would have configured VMMIO via DT */
 		ret = GUNYAH_ERROR_OK;
 

gki/aarch64/symbols/pixel

@@ -1432,6 +1432,7 @@
   kthread_stop_put
   kthread_unpark
   kthread_worker_fn
+  ktime_add_safe
   ktime_get
   ktime_get_coarse_ts64
   ktime_get_coarse_with_offset

gki/aarch64/symbols/xiaomi

@@ -89,6 +89,7 @@
 
 #required by stability
   sock_from_file
+  get_dmabuf_debugfs_data
   __traceiter_android_vh_cma_alloc_fail
   __tracepoint_android_vh_cma_alloc_fail
   __traceiter_android_vh_warn_alloc_show_mem_bypass

include/linux/blk_types.h

@@ -443,7 +443,7 @@ static inline enum req_op bio_op(const struct bio *bio)
 
 static inline bool op_is_write(blk_opf_t op)
 {
-	return !!(op & (__force blk_opf_t)1);
+	return (op & (__force blk_opf_t)1) || op == REQ_OP_ZONE_FINISH;
 }
 
 /*

include/linux/blkdev.h

@@ -840,6 +840,61 @@ static inline u64 sb_bdev_nr_blocks(struct super_block *sb)
 		(sb->s_blocksize_bits - SECTOR_SHIFT);
 }
 
+#ifdef CONFIG_BLK_DEV_ZONED
+/**
+ * bio_needs_zone_write_plugging - Check if a BIO needs to be handled with zone
+ *				   write plugging
+ * @bio: The BIO being submitted
+ *
+ * Return true whenever @bio execution needs to be handled through zone
+ * write plugging (using blk_zone_plug_bio()). Return false otherwise.
+ */
+static inline bool bio_needs_zone_write_plugging(struct bio *bio)
+{
+	enum req_op op = bio_op(bio);
+
+	/*
+	 * Only zoned block devices have a zone write plug hash table. But not
+	 * all of them have one (e.g. DM devices may not need one).
+	 */
+	if (!bio->bi_bdev->bd_disk->zone_wplugs_hash)
+		return false;
+
+	/* Only write operations need zone write plugging. */
+	if (!op_is_write(op))
+		return false;
+
+	/* Ignore empty flush */
+	if (op_is_flush(bio->bi_opf) && !bio_sectors(bio))
+		return false;
+
+	/* Ignore BIOs that already have been handled by zone write plugging. */
+	if (bio_flagged(bio, BIO_ZONE_WRITE_PLUGGING))
+		return false;
+
+	/*
+	 * All zone write operations must be handled through zone write plugging
+	 * using blk_zone_plug_bio().
+	 */
+	switch (op) {
+	case REQ_OP_ZONE_APPEND:
+	case REQ_OP_WRITE:
+	case REQ_OP_WRITE_ZEROES:
+	case REQ_OP_ZONE_FINISH:
+	case REQ_OP_ZONE_RESET:
+	case REQ_OP_ZONE_RESET_ALL:
+		return true;
+	default:
+		return false;
+	}
+}
+#else /* CONFIG_BLK_DEV_ZONED */
+static inline bool bio_needs_zone_write_plugging(struct bio *bio)
+{
+	return false;
+}
+#endif
+
 int bdev_disk_changed(struct gendisk *disk, bool invalidate);
 
 void put_disk(struct gendisk *disk);

include/trace/hooks/health_report.h

@@ -0,0 +1,20 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM health_report
+
+#define TRACE_INCLUDE_PATH trace/hooks
+
+#if !defined(_TRACE_HOOK_HEALTH_REPORT_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_HOOK_HEALTH_REPORT_H
+
+#include <trace/hooks/vendor_hooks.h>
+
+DECLARE_HOOK(android_vh_health_report,
+	TP_PROTO(unsigned int err_code, const char *func, unsigned int line,
+		unsigned long *entries, unsigned int nr_entries),
+	TP_ARGS(err_code, func, line, entries, nr_entries));
+
+
+#endif /* _TRACE_HOOK_HEALTH_REPORT_H */
+/* This part must be outside protection */
+#include <trace/define_trace.h>

rust/helpers/helpers.c

@@ -21,6 +21,7 @@
 #include "mman.c"
 #include "mutex.c"
 #include "page.c"
+#include "poll.c"
 #include "rbtree.c"
 #include "refcount.c"
 #include "security.c"

rust/helpers/poll.c

@@ -0,0 +1,10 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <linux/export.h>
+#include <linux/poll.h>
+
+void rust_helper_poll_wait(struct file *filp, wait_queue_head_t *wait_address,
+			   poll_table *p)
+{
+	poll_wait(filp, wait_address, p);
+}

rust/kernel/miscdevice.rs

@@ -20,12 +20,7 @@ use crate::{
     str::CStr,
     types::{AsBytes, ForeignOwnable, Opaque},
 };
-use core::{
+use core::{marker::PhantomData, mem::MaybeUninit, pin::Pin, ptr::NonNull};
-    marker::PhantomData,
-    mem::MaybeUninit,
-    pin::Pin,
-    ptr::NonNull,
-};
 
 /// The kernel `loff_t` type.
 #[allow(non_camel_case_types)]

rust/kernel/sync/poll.rs

@@ -9,9 +9,8 @@ use crate::{
     fs::File,
     prelude::*,
     sync::{CondVar, LockClassKey},
-    types::Opaque,
 };
-use core::ops::Deref;
+use core::{marker::PhantomData, ops::Deref};
 
 /// Creates a [`PollCondVar`] initialiser with the given name and a newly-created lock class.
 #[macro_export]
@@ -23,58 +22,43 @@ macro_rules! new_poll_condvar {
     };
 }
 
-/// Wraps the kernel's `struct poll_table`.
+/// Wraps the kernel's `poll_table`.
 ///
 /// # Invariants
 ///
-/// This struct contains a valid `struct poll_table`.
+/// The pointer must be null or reference a valid `poll_table`.
-///
-/// For a `struct poll_table` to be valid, its `_qproc` function must follow the safety
-/// requirements of `_qproc` functions:
-///
-/// * The `_qproc` function is given permission to enqueue a waiter to the provided `poll_table`
-///   during the call. Once the waiter is removed and an rcu grace period has passed, it must no
-///   longer access the `wait_queue_head`.
 #[repr(transparent)]
-pub struct PollTable(Opaque<bindings::poll_table>);
+pub struct PollTable<'a> {
+    table: *mut bindings::poll_table,
+    _lifetime: PhantomData<&'a bindings::poll_table>,
+}
 
-impl PollTable {
+impl<'a> PollTable<'a> {
-    /// Creates a reference to a [`PollTable`] from a valid pointer.
+    /// Creates a [`PollTable`] from a valid pointer.
     ///
     /// # Safety
     ///
-    /// The caller must ensure that for the duration of 'a, the pointer will point at a valid poll
+    /// The pointer must be null or reference a valid `poll_table` for the duration of `'a`.
-    /// table (as defined in the type invariants).
+    pub unsafe fn from_raw(table: *mut bindings::poll_table) -> Self {
-    ///
+        // INVARIANTS: The safety requirements are the same as the struct invariants.
-    /// The caller must also ensure that the `poll_table` is only accessed via the returned
+        PollTable {
-    /// reference for the duration of 'a.
+            table,
-    pub unsafe fn from_ptr<'a>(ptr: *mut bindings::poll_table) -> &'a mut PollTable {
+            _lifetime: PhantomData,
-        // SAFETY: The safety requirements guarantee the validity of the dereference, while the
+        }
-        // `PollTable` type being transparent makes the cast ok.
-        unsafe { &mut *ptr.cast() }
-    }
-
-    fn get_qproc(&self) -> bindings::poll_queue_proc {
-        let ptr = self.0.get();
-        // SAFETY: The `ptr` is valid because it originates from a reference, and the `_qproc`
-        // field is not modified concurrently with this call since we have an immutable reference.
-        unsafe { (*ptr)._qproc }
     }
 
     /// Register this [`PollTable`] with the provided [`PollCondVar`], so that it can be notified
     /// using the condition variable.
-    pub fn register_wait(&mut self, file: &File, cv: &PollCondVar) {
+    pub fn register_wait(&self, file: &File, cv: &PollCondVar) {
-        if let Some(qproc) = self.get_qproc() {
+        // SAFETY:
-            // SAFETY: The pointers to `file` and `self` need to be valid for the duration of this
+        // * `file.as_ptr()` references a valid file for the duration of this call.
-            // call to `qproc`, which they are because they are references.
+        // * `self.table` is null or references a valid poll_table for the duration of this call.
-            //
+        // * Since `PollCondVar` is pinned, its destructor is guaranteed to run before the memory
-            // The `cv.wait_queue_head` pointer must be valid until an rcu grace period after the
+        //   containing `cv.wait_queue_head` is invalidated. Since the destructor clears all
-            // waiter is removed. The `PollCondVar` is pinned, so before `cv.wait_queue_head` can
+        //   waiters and then waits for an rcu grace period, it's guaranteed that
-            // be destroyed, the destructor must run. That destructor first removes all waiters,
+        //   `cv.wait_queue_head` remains valid for at least an rcu grace period after the removal
-            // and then waits for an rcu grace period. Therefore, `cv.wait_queue_head` is valid for
+        //   of the last waiter.
-            // long enough.
+        unsafe { bindings::poll_wait(file.as_ptr(), cv.wait_queue_head.get(), self.table) }
-            unsafe { qproc(file.as_ptr() as _, cv.wait_queue_head.get(), self.0.get()) };
-        }
     }
 }
 

scripts/gendwarfksyms/types.c

@@ -6,6 +6,8 @@
 #define _GNU_SOURCE
 #include <inttypes.h>
 #include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
 #include <zlib.h>
 
 #include "gendwarfksyms.h"
@@ -179,20 +181,41 @@ static int type_map_get(const char *name, struct type_expansion **res)
 	return -1;
 }
 
+static int cmp_expansion_name(const void *p1, const void *p2)
+{
+	struct type_expansion *const *e1 = p1;
+	struct type_expansion *const *e2 = p2;
+
+	return strcmp((*e1)->name, (*e2)->name);
+}
+
 static void type_map_write(FILE *file)
 {
 	struct type_expansion *e;
 	struct hlist_node *tmp;
+	struct type_expansion **es;
+	size_t count = 0;
+	size_t i = 0;
 
 	if (!file)
 		return;
 
-	hash_for_each_safe(type_map, e, tmp, hash) {
+	hash_for_each_safe(type_map, e, tmp, hash)
-		checkp(fputs(e->name, file));
+		++count;
+	es = xmalloc(count * sizeof(struct type_expansion *));
+	hash_for_each_safe(type_map, e, tmp, hash)
+		es[i++] = e;
+
+	qsort(es, count, sizeof(struct type_expansion *), cmp_expansion_name);
+
+	for (i = 0; i < count; ++i) {
+		checkp(fputs(es[i]->name, file));
 		checkp(fputs(" ", file));
-		type_list_write(&e->expanded, file);
+		type_list_write(&es[i]->expanded, file);
 		checkp(fputs("\n", file));
 	}
+
+	free(es);
 }
 
 static void type_map_free(void)