Overview

There appears to be some unexplained change in behaviour in the recent version of macos 15.6.1 which is causing the BSD socket sendto() syscall to no longer send the data when the source socket is bound to a IPv4-mapped IPv6 address. I have attached a trivial native code which reproduces the issue. What this reproducer does is explained as a comment on that code's main() function: // Creates a AF_INET6 datagram socket, marks it as dual socket (i.e. IPV6_V6ONLY = 0), // then binds the socket to a IPv4-mapped IPv6 address (chosen on the host where this test runs). // // The test then uses sendto() to send some bytes. For the sake of this test, it uses the same IPv4-mapped // IPv6 address as the destination address to sendto(). The test then waits for (a maximum of) 15 seconds to // receive that sent message by calling recvfrom(). // // The test passes on macos (x64 and aarch64) hosts of versions 12.x, 13.x, 14.x and 15.x upto 15.5. // Only on macos 15.6.1 and the recent macos 26, the test fails. Specifically, the first message that is // sent using sendto() is never sent (and thus the recvfrom()) times out. sendto() however returns 0, // incorrectly indicating a successful send. Interesting, if you repeat sendto() a second message from the // same bound socket to the exact same destination address, the send message is indeed correctly sent and // received immediately by the recvfrom(). It's only the first message which goes missing (the test uses // unique content in each message to be sure which exact message was received and it has been observed that // only the second message is received and the first one lost). // // Logs collected using "sudo log collect --last 2m" (after the test program returns) shows the following log // message, which seem relevant: // ... // default kernel cfil_hash_entry_log:6088 <CFIL: Error: sosend_reinject() failed>: // [86868 a.out] <UDP(17) out so 59faaa5dbbcef55d 127846646561221313 127846646561221313 age 0> // lport 65051 fport 65051 laddr 192.168.1.2 faddr 192.168.1.2 hash 201AAC1 // default kernel cfil_service_inject_queue:4472 CFIL: sosend() failed 22 // ... // As noted, this test passes without issues on various macosx version (12 through 15.5), both x64 and aarch64 but always fails against 15.6.1. I have been told that it also fails on the recently released macos 26 but I don't have access to such host to verify it myself. The release notes don't usually contain this level of detail, so it's hard to tell if something changed intentionally or if this is a bug. Should I report this through the feedback assistant? Attached is the source of the reproducer, run it as: clang dgramsend.c ./a.out On macos 15.6.1, you will see that it will fail to send (and thus receive) the message on first attempt but the second one passes: ... created and bound a datagram dual socket to ::ffff:192.168.1.2:65055 ::ffff:192.168.1.2:65055 sendto() ::ffff:192.168.1.2:65055 ---- Attempt 1 ---- sending greeting "hello 1" sendto() succeeded, sent 8 bytes calling recvfrom() receive timed out --------------------- ---- Attempt 2 ---- sending greeting "hello 2" sendto() succeeded, sent 8 bytes calling recvfrom() received 8 bytes: "hello 2" --------------------- TEST FAILED ... The output "log collect --last 2m" contains a related error (and this log message consistently shows up every time you run that reproducer): ... default kernel cfil_hash_entry_log:6088 <CFIL: Error: sosend_reinject() failed>: [86248 a.out] <UDP(17) out so 59faaa5dbbcef55d 127846646561221313 127846646561221313 age 0> lport 65055 fport 65055 laddr 192.168.1.2 faddr 192.168.1.2 hash 201AAC1 default kernel cfil_service_inject_queue:4472 CFIL: sosend() failed 22 ... I don't know what it means though. dgramsend.c

The following code worked as expected on iOS 26 RC, but it no longer works on the official release of iOS 26. Is there something I need to change in order to make it work on the official version? Registration BGTaskScheduler.shared.register( forTaskWithIdentifier: taskIdentifier, using: nil ) { task in ////////////////////////////////////////////////////////////////////// // This closure is not called on the official release of iOS 26 ////////////////////////////////////////////////////////////////////// let task = task as! BGContinuedProcessingTask var shouldContinue = true task.expirationHandler = { shouldContinue = false } task.progress.totalUnitCount = 100 task.progress.completedUnitCount = 0 while shouldContinue { sleep(1) task.progress.completedUnitCount += 1 task.updateTitle("\(task.progress.completedUnitCount) / \(task.progress.totalUnitCount)", subtitle: "any subtitle") if task.progress.completedUnitCount == task.progress.totalUnitCount { break } } let completed = task.progress.completedUnitCount >= task.progress.totalUnitCount if completed { task.updateTitle("Completed", subtitle: "") } task.setTaskCompleted(success: completed) } Request let request = BGContinuedProcessingTaskRequest( identifier: taskIdentifier, title: "any title", subtitle: "any subtitle", ) request.strategy = .queue try BGTaskScheduler.shared.submit(request) Sample project code: https://github.com/HikaruSato/ExampleBackgroundProcess

For years, my app has been receiving XLSX files from other apps using the share command. For example, in an email, I use the share command on an xlsx attachment and send it to my app. From my app, I go to the Documents/Inbox folder and find the file. This mechanism has broken! And I'm not talking about an app compiled with XCode26, but simply installing my app, still compiled with XCode16, on iPadOS26. It seems that the operating system no longer puts files in the Inbox. Is this true?

With macOS Tahoe, Launchpad has been replaced by an App Library–style mode within Spotlight. While the alleged intention is UX consistency across the Apple ecosystem, the result is both a catastrophic usability regression and a radical break in consistency with iOS and iPadOS. Predefined App Library categorization is functionally incoherent: On iOS and now macOS, Apple’s predefined App Library categories place apps with seemingly identical functionality into unrelated groups—for example, 3D scanning tools scattered across Education, Utilities, and Productivity. Instead of making apps easier to find, this effectively creates a labyrinth that users must traverse to locate apps whose names and icons they may not recall. However Apple defines its app categories, they are not only inconsistent but also hopelessly inadequate for the long tail of real-world applications and user workflows. Loss of user control: Launchpad enabled users to group and organize applications according to their workflows. This aligns with Apple’s own Human Interface Guidelines, which emphasize user control, discoverability, and predictable behavior. The new Spotlight interface removes that flexibility, locking users into predefined categories that both impede and mislead—and cannot be overridden. Consistency across platforms is broken: If the goal was to unify iOS, iPadOS, and macOS, this approach actually undermines consistency. On iOS and iPadOS, users can still rely on a customizable Home Screen—a Launchpad-like experience—as their primary way of launching apps. In Tahoe, that option has been removed. macOS now forces users to depend exclusively on Spotlight with App Library categories, while eliminating the very feature that was consistent across platforms. Catastrophic impact on my workflow: As an interdisciplinary artist working in 2D, 3D, and time-based media, as well as coding, I make extensive use of a constantly changing array of AI tools and experiment with many new apps and web services, which I often turn into Web Apps. I cannot possibly recall the names of every native and web app on my system. I need predictable access to groups of related tools. Tahoe’s new auto-categories split those apps apart arbitrarily, slowing me down and interrupting established workflows, forcing me to navigate the aforementioned labyrinth just to find what I need. Proposal: A constructive way forward High-level objective: Simply restore Launchpad—or restore the ability to customize app categories/folders and manually assign apps to them, overriding or augmenting the predefined categories. This ensures users can launch apps according to their workflow, without needing to remember exact names or icons. Possible solutions: Allow manual subfolders within Applications, represented hierarchically in Spotlight. Provide a fullscreen Launchpad-like organizer (with uninstall via long-click, etc.), either as a replacement or toggleable option. Retain Apple’s auto-categories for those who prefer them, but let users override or augment them with their own. In summary: Tahoe eliminates a working, consistent paradigm (Launchpad/Home Screen) and forces reliance on an App Library system that categorizes poorly and cannot be customized. This is both a step backwards in functionality and a break in cross-platform consistency. A constructive solution is to restore Launchpad—or at least restore the ability for users to organize apps in ways that fit their workflows.

From what I’ve seen, this issue has been around since macOS 13 and can be reproduced reliably. It happens with some apps like Music, Notes, and Google Chrome. Here’s how to see it: 1.Make sure “Minimize windows into application” is enabled in System Settings, or just open a minimized app later directly from its application icon. 2.Open one of the apps mentioned above. 3.Minimize it. 4.Click the minimized app in the Dock to restore it. You’ll notice the GUI flashes for a moment and the minimize animation plays again. Some additional info here: https://forums.macrumors.com/threads/weird-glitches-during-restore-from-minimalization-of-any-app.2370260/ A video clipped from another GitHub issue: https://private-user-images.githubusercontent.com/13177224/445474477-36d8c784-9588-4186-8b6a-875c4077ce1c.mp4?jwt=eyJ0eXAiOiJKV1QiLCJhbGciOiJIUzI1NiJ9.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.xbAxdTgxadCVCZPsnZkhx9HnVbjP-D5w1GfPTBatIWQ

For difficult reasons I won’t get into, I ended up manually downloading the latest iOS 26 simulator runtime. I now have a file named 78756498-8AB4-4E5A-986C-7AA435758657.aar copied to my Mac. How do I get this archive installed so Xcode 26 recognizes it as a proper simulator runtime component? All searching I‘ve done for manually installing simulators references dmg files and older versions of Xcode. There’s no mention of aar files. When I tried the command: sudo xcrun simctl runtime add ./78756498-8AB4-4E5A-986C-7AA435758657.aar I get the result: An error was encountered processing the command (domain=NSPOSIXErrorDomain, code=22): Error while creating AEA backend Invalid argument I tried to use Archive Utility to open the file but that just says it is unable to expand the file. I even tried renaming the file with a dmg extension and then tried mounting the file and I get the same “AEA backend” error. My Mac doesn’t have sufficient Internet access to let me download and install this normally through Xcode. I need to find a way to get this file installed manually.

Dear Apple Support, We are encountering an issue with deep linking on iOS 26 when using Safari and App Store redirection. Below are the detailed steps to reproduce the problem: The app is not installed on the device. User clicks on a Branch link: For example:- https://qewed.app.link/99u88ef9f?uuid=88dbwh5ubd4b Safari opens and displays the fallback page. User taps on “Get the App”, which navigates to the App Store. User installs and opens the app. Expected Behavior:
The app should receive the Branch link parameters (in this case, uuid=88dbwh5ubd4b) upon first open. Actual Behavior:
The app opens successfully, but the uuid parameter is not passed to the app. This issue seems specific to the Safari → App Store → App flow on iOS 26, as other flows behave correctly. Could you please advise whether this is a known issue or if there are recommended adjustments on our side to ensure parameters are consistently delivered after App Store redirection? Note: We are using Branch SDK version 3.11.0 (Cocoapods dependency) Thank you for your assistance.

Hi, I want to understand how Per-App VPN can be configured on iOS. Specifically: Is Per-App VPN something that can be controlled directly from code (inside an app), or does it always require MDM / configuration profiles (.mobileconfig)? If it requires a configuration profile, how can I specify which apps (by bundle identifier) should use the VPN? Is it possible to let users dynamically choose apps for Per-App VPN, or is this strictly managed by MDM? My goal is to make sure that only apps selected by user dynamically use the VPN connection, while other apps continue to use the normal network. Any official Apple documentation, examples, or guidance would be greatly appreciated.

I have 14 total devices, from way back. I am currently in a financial bind and can't renew just yet. BUT I am at past my time to reset the device list back to zero. But the screen to do that is behind the paid account. Catch 22 Can we fix it? As it stands I must email tech support, but this is a bug so I posted

I’m building a document-based macOS app using SwiftUI with an AppKit NSDocument. By default, when the window has no toolbar, the document title and proxy icon (with the edited state dot and standard saving controls) appear nicely centered in the title bar. However, as soon as I attach a toolbar - even an empty one - the document proxy moves to the leading edge of the title bar. Is there a way to keep the document proxy/title centered in a document-based SwiftUI app while also using a toolbar? Or is the left-alignment now the only supported behavior when a toolbar is present? Thanks in advance for any guidance.

I noticed that when I have a fullscreen window in macOS 26, sidebars look like they are cut off at the top: they suddenly stop where the title bar/toolbar would appear when moving the mouse to the top of the screen, leaving a wide empty gap. Am I the only one who finds this ugly? Is this intended, or is there a workaround? This is how it looks in fullscreen (the sidebar borders are not easy to distinguish, look for the drop shadow): And this when moving the mouse to the top screen border to show the menu bar: I created FB20291636. @main class AppDelegate: NSObject, NSApplicationDelegate, NSWindowDelegate { func applicationDidFinishLaunching(_ aNotification: Notification) { let splitViewController = NSSplitViewController() splitViewController.addSplitViewItem(NSSplitViewItem(sidebarWithViewController: ViewController())) splitViewController.addSplitViewItem(NSSplitViewItem(viewController: ViewController())) let window = NSWindow(contentViewController: splitViewController) window.styleMask = [.titled, .closable, .miniaturizable, .resizable, .fullSizeContentView] window.toolbar = NSToolbar() window.delegate = self window.makeKeyAndOrderFront(nil) } func window(_ window: NSWindow, willUseFullScreenPresentationOptions proposedOptions: NSApplication.PresentationOptions = []) -> NSApplication.PresentationOptions { return [.autoHideToolbar, .autoHideMenuBar, .fullScreen] } } class ViewController: NSViewController { override func loadView() { let stack = NSStackView(views: [ NSTextField(labelWithString: "asdf") ]) stack.orientation = .vertical stack.alignment = .leading view = stack view.frame = CGRect(x: 0, y: 0, width: 300, height: 300) } }

When using LazyVGrid within a ScrollView with the .searchable modifier, scrolling is impossible on iPhone 15 (I’m assuming other older devices are affected too). This behavior does not occur on iPhone 16 and newer. This also only happens, when the search bar is placed at the top, for example if the ScrollView is within a TabView. Here's a short screen recording of the issue: And this is a minimal example causing the issue: import SwiftUI struct ContentView: View { var body: some View { TabView { Tab("Text", systemImage: "gear") { ExampleTab() } } } } struct ExampleTab: View { @State private var searchText: String = "" var body: some View { NavigationStack { ScrollView { LazyVGrid( columns: [GridItem( .adaptive(minimum: 120) )], spacing: 20 ) { ForEach(1..<100) { index in Text("Test \(index)") } } } .searchable(text: self.$searchText) } } }

Pinned 2 homes address for the same contact Steps Initial check in Apple Maps No saved places or pinned addresses appear. Open Personal Contacts You have two addresses stored in your contact card: Main and Home. Pin & Edit “Main” You pinned the Main address in Maps. Refined the location on the map. Renamed it (but still saved under the type “My Home”). Open “Home” Address in Contacts Refined the location again. Changed the type to “My Home.” Attempted to rename, but no option to change the label. Final Saved Places View Shows two entries both called “Main.” Opening either of them displays the same details for the Home address. Saved Places list only shows the full address text, without the ability to rename them inside Maps. Results Both addresses appear duplicated with the same name (“Main”), even though they point to different underlying addresses. When selecting either entry, Apple Maps incorrectly shows the same Home address details. The Saved Places section does not allow renaming; it defaults to showing the full address string. Issues Identified Sync Conflict Between Contacts & Maps Apple Maps pulls labels/types from Contacts, but the edits don’t update consistently across apps. Duplicate Naming Bug Both “Main” and “Home” collapse into “Main” in Saved Places, making them indistinguishable. One-to-One Mapping Failure Regardless of which saved place you open, Maps shows the same Home entry, meaning the system isn’t correctly binding each saved place to its respective contact address. Renaming Limitation Apple Maps doesn’t allow renaming saved addresses directly — it relies on Contacts. Since Contacts only supports preset labels (Home, Work, School, etc.), custom naming is blocked.

I'm encountering an issue with system accessories in UICollectionViewCells after overriding the trait collection. Specifically, the accessories are misaligned and shifted downwards. This issue occurs when using setOverrideTraitCollection (other trait override methods produce the same result). Interestingly, this doesn't happen with all accessories; for example, the disclosureIndicator is scaled correctly. If I don't use setOverrideTraitCollection (or other trait override methods), the system accessories scale as expected. Here's a code snippet demonstrating the issue. I have a "Fix Size" button that overrides the trait collection to a UITraitCollection with a UIContentSizeCategory of large. The "Follow Settings" button resets the trait collection, allowing the views to scale according to the system settings. import UIKit class ViewController: UIViewController { let button: UIButton = { let button = UIButton(type: .system) button.setTitle("Fix Size", for: .normal) return button }() let buttonRe: UIButton = { let button = UIButton(type: .system) button.setTitle("Follow Settings", for: .normal) return button }() var listItems: [Int] = [1, 2, 3, 4, 5] var collectionView: UICollectionView? override func viewDidLoad() { super.viewDidLoad() button.addTarget( self, action: #selector(ViewController.buttonTapped), for: .touchUpInside ) buttonRe.addTarget( self, action: #selector(ViewController.buttonTappedToRe), for: .touchUpInside ) view.backgroundColor = .white view.addSubview(button) view.addSubview(buttonRe) setupCollectionView() if let collectionView = collectionView { view.addSubview(collectionView) } } override func viewDidLayoutSubviews() { super.viewDidLayoutSubviews() collectionView?.frame = CGRect( x: 0, y: 0, width: view.bounds.width, height: view.bounds.height - 100 ) button.frame = CGRect( x: 0, y: view.bounds.height - 100, width: view.bounds.width / 2, height: 50 ) buttonRe.frame = CGRect( x: view.bounds.width / 2, y: view.bounds.height - 100, width: view.bounds.width / 2, height: 50 ) } @objc func buttonTapped() { setOverrideTraitCollection( UITraitCollection(preferredContentSizeCategory: .large), forChild: self ) } @objc func buttonTappedToRe() { setOverrideTraitCollection(nil,forChild: self) } private func updateCollectionViewLayout() { guard let collectionView = collectionView else { return } collectionView.collectionViewLayout.invalidateLayout() collectionView.performBatchUpdates(nil) collectionView.reloadData() } private func setupCollectionView() { var config = UICollectionLayoutListConfiguration(appearance: .insetGrouped) config.trailingSwipeActionsConfigurationProvider = { [weak self] indexPath in let deleteAction = UIContextualAction(style: .destructive, title: "Delete") { [weak self] _, _, completion in self?.deleteItem(at: indexPath) completion(true) } return UISwipeActionsConfiguration(actions: [deleteAction]) } let layout = UICollectionViewCompositionalLayout.list(using: config) let collectionView = UICollectionView(frame: view.bounds, collectionViewLayout: layout) collectionView.delegate = self collectionView.dataSource = self collectionView.register(UICollectionViewListCell.self, forCellWithReuseIdentifier: "cell") collectionView.isEditing = true self.collectionView = collectionView } private func deleteItem(at indexPath: IndexPath) { listItems.remove(at: indexPath.item) guard let collectionView = collectionView else { return } collectionView.deleteItems(at: [indexPath]) } override func traitCollectionDidChange(_ previousTraitCollection: UITraitCollection?) { super.traitCollectionDidChange(previousTraitCollection) if previousTraitCollection?.preferredContentSizeCategory != traitCollection.preferredContentSizeCategory { updateCollectionViewLayout() } } } extension ViewController: UICollectionViewDataSource { func numberOfSections(in collectionView: UICollectionView) -> Int { return 1 } func collectionView(_ collectionView: UICollectionView, numberOfItemsInSection section: Int) -> Int { return listItems.count } func collectionView(_ collectionView: UICollectionView, cellForItemAt indexPath: IndexPath) -> UICollectionViewCell { let cell = collectionView.dequeueReusableCell(withReuseIdentifier: "cell", for: indexPath) as! UICollectionViewListCell var content = UIListContentConfiguration.valueCell() content.text = "Item \(listItems[indexPath.item])" cell.contentConfiguration = content cell.accessories = [.delete( options: UICellAccessory.DeleteOptions( reservedLayoutWidth: .custom(50) ) )] return cell } } extension ViewController: UICollectionViewDelegate { func collectionView(_ collectionView: UICollectionView, didSelectItemAt indexPath: IndexPath) { collectionView.deselectItem(at: indexPath, animated: true) } } The attached screenshot illustrates the misalignment that occurs after tapping the 'Fix Size' button, with the system accessibility text size set to accessibilityExtraExtraExtraLarge setting Has anyone else experienced this issue or have suggestions on how to resolve it? Any help would be greatly appreciated!

We are currently using Live Activities in our app and supporting both of the following use cases: Starting a Live Activity directly from the app using ActivityKit APIs. Starting a Live Activity from the backend using the start token. In the first case (initiated from the app), the OS generates an update token, and we are able to continuously update the Live Activity via our backend—even if the user has not explicitly provided "Allow" or "Always Allow" consent from the lock screen. This works as expected. In the second case (initiated from the backend), if the user does provide consent ("Allow" or "Always Allow") from the lock screen, we receive the update token and can continue updating the Live Activity. However, if the user does not provide consent, the OS does not provide the update token, and we are unable to send further updates. Question: Is it possible to receive the update token from the OS when the Live Activity is started from the backend, without the user explicitly providing "Allow" or "Always Allow" consent from the lock screen? We would appreciate any clarification or official documentation related to this behavior. Thank you!

I have several games on the app store which are setup as "For Kids" which means these games cannot have any way to access the outside world including the App Store. No problem. These games have worked fine for years until iOS 26. Now, all my updates are being rejected because the new version of Game Center running on iPadOS 26 has a way for people to exit the game and go to the App Store. I have no control over this since it's built into Game Center, and the app review folks want me to put a "parental gate" on it, but there's no way to do that because... well... it's in Game Center, not my code. So, I'm unable to update my apps because of this. Presumably, the existing versions on the app store still do this exact same thing, so my update isn't going to make any difference. Does anyone know of a way to make that crap at the top go away so this isn't an issue?

Xcode build done. 0.7s Failed to build iOS app Uncategorized (Xcode): Unable to find a destination matching the provided destination specifier: { id:42969747-3560-448B-8EB3-CB5ED88D75C1 } Available destinations for the "Runner" scheme: { platform:macOS, arch:arm64, variant:Designed for [iPad,iPhone], id:00008132-0004256C34A1801C, name:My Mac } { platform:iOS, id:dvtdevice-DVTiPhonePlaceholder-iphoneos:placeholder, name:Any iOS Device } { platform:iOS Simulator, id:dvtdevice-DVTiOSDeviceSimulatorPlaceholder-iphonesimulator:placeholder, name:Any iOS Simulator Device } Could not build the application for the simulator. Error launching application on iPhone 17.

When preparing the app for the new iOS 26, I came across an unpleasant design decision. Specifically, in the new design, the keyboard has rounded corners, under which the system background is visible. And here we have only two options, a light/dark background, which breaks all keyboard calls in the application. Can you tell me if there is any way around this problem?

I have a TabView (no modifiers) as the top-level view in my app. Starting with iOS 26 it starts off partially "under" the Status Bar, and then repositions if I switch between apps. Starting Point After Switching To/From Another App In the simulator, pressing "Home" and then reopening the app will fix it. Anyone else seeing something similar? Is there a modifier I'm missing on TabView that might prevent this behaviour? Thanks!

Hello, I am currently testing an Adyen integration with Sylius and need to verify Apple Pay with Cartes Bancaires in the sandbox environment. Could you please advise how Cartes Bancaires can be tested in Apple Pay Sandbox (e.g. cards details)? Thank you in advance for your guidance. Best regards, Grzegorz