DEV Community: Takahiko Inayama

Do M1 Mac Dream of x86 Linux?

Takahiko Inayama — Fri, 03 Nov 2023 06:19:17 +0000

My primary machine is MacBook Pro 14” with M1 Max chip. It’s been great so far. It’s “scary” fast. But there is only one thing I can’t do with it. It doesn’t support to run Intel based VMs with Virtualization Framework due to the limitation of Rosetta translation layer.

About the Rosetta Translation Environment | Apple Developer Documentation

Rosetta can translate most Intel-based apps, including apps that contain just-in-time (JIT) compilers. However, Rosetta doesn’t translate the following executables:

Kernel extensions

Virtual Machine apps that virtualize x86_64 computer platforms

So I can’t run x86 VMs with VirtualBox like I often done with Intel Mac.

I’ve written another article about this topic. I used UTM at last time.

How to run x86 Linux on M1 MacBook?

Running Vagrant with QEMU backend

Let’s try Vagrant as a frontend at this time. Vagrant supports bunch of virtualization softwares as its backend.

vagrant-qemu is a great plugin for Apple Silicon Mac owners like me.

QEMU support to translate CPU instructions. Actually, Docker had been used QEMU to run x86 containers on Apple Silicon Mac before its transition to Rosetta.

The Magic Behind the Scenes of Docker Desktop | Docker

With support for Apple M1 ARM Mac and QEMU included in Docker Desktop, you are able to build and use multi-architecture images (Linux x86, ARM, Windows) on whatever platform you are working on out of the box.

There is an amazing gist which describe how to run VMs with QEMU and Vagrant.

Run x86 VM's on Mac M1 arm using vagrant with qemu hypervisor

Running x86 Linux VMs

Running Rocky Linux 9

Running Rocky Linux 9 is pretty simple.

You’ll need a Vagrantfile like this.

Vagrant.configure("2") do |config|
    config.vm.box = "rockylinux/9"

    config.vm.provider "qemu" do |qe|
      qe.arch = "x86_64"
      qe.machine = "q35"
      qe.cpu = "max"
      qe.net_device = "virtio-net-pci"
      qe.memory = "8G"
      qe.smp = "cpus=10,sockets=1,cores=10,threads=1"
    end
    config.vm.synced_folder '.', '/vagrant', disabled: true
    config.vm.network "forwarded_port", guest: 8080, host: 18080
  end

Then you can use it with no hustles.

Running Rocky Linux 8

Running Rocky Linux 8 is a bit tricky.(Including on Vagrant with VirtualBox backend)

You can’t run images on Vagrant directly.

Vagrant box rockylinux/8 - Vagrant Cloud

You’ll stuck on during vagrant up with a Vagrantfile like this.

Vagrant.configure("2") do |config|
    config.vm.box = "rockylinux/8"

    config.vm.provider "qemu" do |qe|
      qe.arch = "x86_64"
      qe.machine = "q35"
      qe.cpu = "max"
      qe.net_device = "virtio-net-pci"
      qe.memory = "8G"
      qe.smp = "cpus=10,sockets=1,cores=10,threads=1"
    end
    config.vm.synced_folder '.', '/vagrant', disabled: true
    config.vm.network "forwarded_port", guest: 8080, host: 18080
  end

You’ll find it failed to boot by using QEMU Monitor.

I’ve experienced the symptom like this before when I tried to run this OS with Vagrant and VirtualBox. All you have to do is make UEFI available while booting.

How to do it with QEMU? Fortunately, there is a package named Open Virtual Machine Firmware(OVMF). You can download its BIOS data from a package for Debian.

Debian -- Package Search Results -- ovmf

You’ll find data.tar.xz in it when you extracted it. So extract this too. Then you’ll find the BIOS image in usr/share/OVMF/OVMF.fd.

Then you need to tweak the Vagrantfile like this.

Vagrant.configure("2") do |config|
    config.vm.box = "rockylinux/8"

    config.vm.provider "qemu" do |qe|
      qe.arch = "x86_64"
      qe.machine = "q35"
      qe.cpu = "max"
      qe.net_device = "virtio-net-pci"
      qe.memory = "8G"
      qe.smp = "cpus=10,sockets=1,cores=10,threads=1"

      # Added
      qe.extra_qemu_args = %w(-bios /path/to/OVMF.fd)
    end
    config.vm.synced_folder '.', '/vagrant', disabled: true
  end

And….

Tada! It’s working!!

Conclusion

Vagrant made it simple to create and configure VMs. It’s nice to be able to use it to run x86 Linux on my M1 Mac.

Of course UTM is still great for GUI usage. But vagrant-qemu definitely filled more lightweight VM usage.

References

This article is also available on Medium.

How to check Battery Health on iPadOS/macOS

Takahiko Inayama — Sat, 04 Feb 2023 01:16:34 +0000

This article is also available on Medium.

If you have iPhone, you probably have seen Battery Health. It’s useful to check how much your battery is damaged by age. Unfortunately, it’s not available on iPadOS and macOS.

But it’s technically not true. There are same statics recorded in analytics logs. I’ll tell you how to see them.

First, you have to enable equivalents of “Share iPhone & Watch Analytics” on your devices.

Share analytics, diagnostics, and usage information with Apple

Share analytics information from your Mac with Apple

Logs will be written on next 0:00 UTC once you’ve enabled these settings.

You can find logs from Settings app on iPad. (”Privacy & Security” ⇒ “Analytics & Improvements” ⇒ “Analytics Data”)

On Mac, they’ll be written on /Library/Logs/DiagnosticReports/(or will be moved to /Library/Logs/DiagnosticReports/Retired/).

There are JSON files start with Analytics-. You can view on your text editor. You’ll find rows which contains last_value_CycleCounton its values.

This line contains lots of statics. I’ll explain some of them I usually monitor.

last_value_CycleCount seems to be charge cycle counts.(Same as it on System Information app.)
last_value_MaximumCapacityPercentseems to be “Battery Health” which available on iPhone.
last_value_NominalChargeCapacity seems to be normalized battery capacity value which used to calculate “Battery Health’
last_value_DailyMaxSoc seems to be highest value of the battery percentage on the day.
last_value_DailyMinSoc seems to be lowest value of the battery percentage on the day.

Aside from above, last_value_AppleRawMaxCapacityis especially useful for me because I can see raw battery capacity values like before iOS 11.3.

For example, my iPad mini has 5124 mAh battery. And current last_value_AppleRawMaxCapacityis 4757. So it’s roughly 93% battery health. It’s lower compared to last_value_MaximumCapacityPercent (”Battery Health”) which has 97 percent. It’s interesting. 🤔

It’s also interesting to watch changes of these values.

Finally, I’ll show you an example Ruby code to read these stats.

require "json"

BATTERY_KEYS = [
  "last_value_CycleCount",
  # "last_value_AverageTemperature",
  # "last_value_DailyMaxSoc",
  # "last_value_DailyMinSoc",
  "last_value_MaximumCapacityPercent",
  "last_value_NominalChargeCapacity",
  "last_value_AppleRawMaxCapacity",
  # "last_value_batteryServiceFlags",
  # "last_value_ServiceOption",
  # "last_value_calibrationFlags",
]

Dir.glob("./*.{synced,core_analytics}") do |file_name|
  os_version = nil
  timestamp = nil
  messages = []

  File.readlines(file_name).each_with_index do |line, idx|
    data = JSON.parse(line)
    if idx == 0
      os_version = data["os_version"]
      timestamp = data["timestamp"]
    end

    message = data["message"]
    next unless message&.keys&.include?(BATTERY_KEYS[0])
    messages << message
  end

  puts os_version
  puts timestamp
  puts BATTERY_KEYS.join("\t")
  messages.each do |message|
    puts (BATTERY_KEYS.map do |k|
      message[k].nil? ? "null" : message[k]
    end).join("\t")
  end

  puts "\n"
end

Have fun!!

Does Unity keep your Mac staying awake? Maybe you can fix this.

Takahiko Inayama — Sat, 04 Dec 2021 09:36:06 +0000

This story is also available on Medium.

I use Unity sometimes on my MacBook. It's fine except one thing. It'll keep your Mac awake once you've hit "Play" button. Mac will turn off Display when idle. But won't fully sleep until you quit Unity.

Unity allocates audio session to play sounds. It's a usual thing. But the problem is, it won't release the audio session after you've stoped the game playing.

$ pmset -g assertion
...
pid 211(coreaudiod): [0x0000aef100019a93] 00:00:03 PreventUserIdleSystemSleep named: "com.apple.audio.AppleUSBAudioEngine:Generic:USB Audio:140000:2.context.preventuseridlesleep"

You'll find a config named "Enable Output Suspension" in Audio section of Project Settings. It should help to sleep from what it says in the description. But it's not working.

This confused me. 🤔 But I figured out.

Surprisingly, there were no config key for this toggle on your project files before you've changed.

You'll see "m_EnableOutputSuspension" is created like below.

--- a/ProjectSettings/AudioManager.asset
+++ b/ProjectSettings/AudioManager.asset
@@ -12,6 +12,7 @@ AudioManager:
   m_DSPBufferSize: 1024
   m_VirtualVoiceCount: 512
   m_RealVoiceCount: 32
+  m_EnableOutputSuspension: 1
   m_SpatializerPlugin:
   m_AmbisonicDecoderPlugin:
   m_DisableAudio: 0

And your Mac will sleep after doing this. I also found that Unity releases the audio session after a while since you've stopped the playing.

I think this may be some sort of bugs on Unity.

"Compression Algorithm" for iPhone 13's product page

Takahiko Inayama — Wed, 03 Nov 2021 08:24:27 +0000

This article is also available on Medium.

If you've ever seen Apple's products page, you may seen lots of gimmicks triggered by scrolling.

And I initially thought that this animation was implemented with video tag. It's common to use videos to make a scroll animation by changing seek position of it. (Spoiler alert: This isn't a video tag!)

If you glanced a network activity, you'll find a few images which used for animation.

But you'll find something strange. Not all frames are complete image. Almost all frames are like an image below. It's shredded fragments.

What's going on? Let'd dive into it!!

Background: How video compression works?

I talked about using video for animation. So what actually video (and its compression algorithm) does? Please take a look at this article.

How Modern Video Compression Algorithms Actually Work - Make Tech Easier

I don't discuss much in this article. It's basically generate difference data between each frames so you can use previous pixels if they weren't changed.

Think Different: Apple's way to compress animation

I think you guessed what's going on. Yes. That strange images are difference data and Apple implemented video compression algorithm its own way!!

A manifest file

You can find a file named flow_manifest.json from network activity. The content is like this.

{
  "version": 5,
  "frameCount": 149,
  "blockSize": 8,
  "imagesRequired": 143,
  "reversible": true,
  "superframeFrequency": 4,
  "frames": [
    {
      "type": "diff",
      "startImageIndex": 0,
      "startBlockIndex": 0,
      "value": "ACLAKs..."
    },
    {
      "type": "diff",
      "startImageIndex": 1,
      "startBlockIndex": 0,
      "value": "ACLALA..."
    },
    {
      "type": "diff",
      "startImageIndex": 2,
      "startBlockIndex": 0,
      "value": "ACKAMA..."
    },
    ...
    {
      "type": "chapter",
      "x": 0,
      "y": 8,
      "width": 760,
      "height": 568
    },
    {
      "type": "diff",
      "startImageIndex": 48,
      "startBlockIndex": 0,
      "value": "ACFAJAD..."
    },
    ...
  ]
}

I think blockSize is same as block sizes of video compressing. It represents a size of pixels treated as minimal unit. So you need to treat 8x8 pixels as a block.

Thevalue on framesarray are pretty long. It seems to be encoded in some way.

I found a curious function from bundled JavaScript file.

_parseFrame(t) {
    var e = this.canvasElement.width / 8
      , i = this.canvasElement.height / 8
      , s = t.value
      , a = t.startImageIndex
      , n = t.startBlockIndex;
    if ("keyframe" === t.type || "chapter" === t.type)
        return {
            type: t.type,
            width: t.width,
            height: t.height,
            x: t.x,
            y: t.y
        };
    let o = []
      , h = []
      , l = n;
    for (let t = 0; t < i; t++)
        o.push(new r(e)),
        h.push({});
    for (let t = 0; t < s.length; t += 5) {
        let i = this._createNumberFromBase64Range(s, t, 3)
          , r = this._createNumberFromBase64Range(s, t + 3, 2);
        for (let t = i; t < i + r; t++) {
            let i = Math.floor(t / e)
              , s = t % e;
            o[i].set(s);
            try {
                h[i]["" + s] = l
            } catch (t) {}
            l++
        }
    }
    return {
        type: "diff",
        startImageIndex: a,
        rowMasks: o,
        rowDiffLocations: h
    }
}

_valueForCharAt(t, e) {
    var i = t.charCodeAt(e);
    return i > 64 && i < 91 ? i - 65 : i > 96 && i < 123 ? i - 71 : i > 47 && i < 58 ? i + 4 : 43 === i ? 62 : 47 === i ? 63 : void 0
}
_createNumberFromBase64Range(t, e, i) {
    for (var s = 0; i--; )
        s += this._valueForCharAt(t, e++) << 6 * i;
    return s
}

It seemingly reads value data as combined Base64 encoded numbers. And this generates an object with type, startImageIndex, rowMasks and rowDiffLocations.

You can try this function on your machine with a bit of modification. You'll get a sequence of numbers as rowMasks and sequence of key-value data as rowDiffLocations.

How to use this data? There is a function for that.

_applyDiffFrame(t, e=[]) {
    a(t >= 0 && t < this.manifest.frames.length, "Frame out of range."),
    a("diff" === this.manifest.frames[t].type, "Invalid attempt to play a keyframe as a diff.");
    const i = this.canvasElement.width / 8
      , s = this.manifest.frames[t].image
      , r = s.width / 8;
    for (let a = 0; a < this.manifest.frames[t].rowMasks.length; a++) {
        let n = this.manifest.frames[t].rowMasks[a]
          , o = this.manifest.frames[t].rowDiffLocations[a]
          , h = this._rangesFromRowMask(n, o, a, i, r);
        if (h.length < 1)
            continue;
        let l = e.map((t=>this.manifest.frames[t].rowMasks[a]))
          , c = n;
        for (; l.length > 0; ) {
            let t = l.pop();
            c = c.and(t.xor(this._bitsetForNAND))
        }
        let u = this._rangesFromRowMask(c, o, a, i, r);
        u.length > h.length && (u = h);
        for (let t = 0; t < u.length; t++)
            this._applyDiffRange(u[t], s, r, i)
    }
}

_rangesFromRowMask(t, e, i, s, a) {
  let n = []
    , o = null;
  if (t.isEqual(new r(s)))
      return [];
  let h = s;
  for (let r = 0; r < h; r++)
      if (t.get(r)) {
          for (o = {
              location: i * s + r,
              length: 1,
              block: e[r]
          },
          o.block; ++r < h && t.get(r); ) {
              0 === (o.block + o.length) % a ? (n.push(o),
              o = {
                  location: o.location + o.length,
                  length: 1,
                  block: e[r]
              }) : o.length++
          }
          n.push(o)
      }
  return n
}
_applyDiffRange(t, e, i, s) {
  var r = t.block
    , a = r % i * 8
    , n = 8 * Math.floor(r / i)
    , o = 8 * t.length
    , h = t.location % s * 8
    , l = 8 * Math.floor(t.location / s);
  this.glCanvas.drawImage(e, a, n, o, 8, h, l, o, 8)
}

It's pretty complicated. I guess it first determine the rows which should be modified by using rowMasks. All other rows excluded by them will remain same.

Then find a horizontal location to be modified by using rowDiffLocations. rowDiffLocations seems to be a relation data of index on compressed images and which on canvas.

Actual redrawing were executed on _applyDiffRange function.

But why?

Why does Apple implemented in this way? It's seems reimplementing a wheel again. Why not simply using video tag?

I think there are tow reasons.

Loading performance

You can prioritize to load specific frames at first with this algorithm.

Imagine if you tried to scroll really fast right after page loading, or simply when you're using slow internet connection. All frames may not be loaded on at that point. But it's possible to load key frames first with this algorithm. So you can see at least low frame rate version of the animation. It's resembles to loading low quality images at first then swap them for high quality ones.

I think actual loading order proves this theory.

Chroma format

Have you ever heard of chroma subsampling? It's a traditionally way to reduce amount of data.

Chroma subsampling - Wikipedia

If you chose to use lower than 4:4:4 chroma, partial color information would be dropped. Not all pixels have original color information.

So you need to use 4:4:4 chroma if you want better picture quality.

By the way, it's common to use 4:2:0 chroma for video encoding. You need to use Hi444PP profile on AVC if you want to encode with 4:4:4 chroma. Besides devices of these days have hardware video decoder for AVC/HEVC, I couldn't find a device with hardware acceleration for Hi444PP profile which available for consumers.

So it's may be difficult to achieve smooth animation with 4:4:4 chroma picture if you're using videos.

How about the chrome format of Apple's animation data?

I examined some frames with exiftool. And of course it's 4:4:4 chroma!!

Y Cb Cr Sub Sampling            : YCbCr4:4:4 (1 1)

I initially thought Apple's algorithm wasn't efficient but it's not really inefficient compared to software decoded videos. It may be reasonable to use this algorithm if you need to show 4:4:4 chroma pictures. It really Apple to obsessed with quality. 😁

Active Storage: How to retain uploaded files on form resubmission?

Takahiko Inayama — Sat, 09 Oct 2021 13:22:58 +0000

This article is also available on Medium.

If you've ever used Active Storage, I bet you must have encountered the same problem I was confused by.

Imagine if your app has a simple form with file upload tags along with other fields. Then one of the fields has unmet with validation rules on submission. So your app displays the form again. Finally, the user corrected the error and submit again. But files that should have saved were simply disappeared...

This kind of story isn't uncommon if you're using Active Storage. I had faced this too. But I found a solution after all.

Note:

If you can use direct_upload in your application, you don't need to read this article. You can simply use this solution.

When using activestorage in Rails 6, how do I retain a file when redisplaying a form?

TL;DR

You need to upload files and save ActiveStorage::Blob objects manually before form reappearance.
You need to submit blob's key using hidden fields.

Background

Active Storage uses 2 models to store attachment information to database. First one is ActiveStorage::Blob. This is a model to store file's metadata. And another is ActiveStorage::Attachment. This is an intermediate table to combine ActiveStorage::Blog to your model.

Thehas_one_attached method makes your model to manage these 2 models automatically. I'll talk about has-one case but most logics are same as has-many one.

Source code of has_one_attached is here. Take a look at these 2 lines.

after_save { attachment_changes[name.to_s]&.save }

after_commit(on: %i[ create update ]) { attachment_changes.delete(name.to_s).try(:upload) }

If you made changes to your model's attachment, those actions are stored as object in attachment_changes . For example, ActiveStorage::Attached::Changes::CreateOne, ActiveStorage::Attached::Changes::DeleteOne, and so on.

Your model will be configured to call these objects's save method right before database access. (after_save hook will be called before actual changes to the database.) Objects of ActiveStorage::Blob and ActiveStorage::Attachmentwill be created and referenced to your model on the save method.

And attached files will be uploaded on after_commit hook.

What is the root cause of problem?

Attachment persistence is handled when you save your model, as I mentioned in the previous section. This is a problem because your model won't be saved before form resubmission.

Solution

Problems are identified. What should we do?

Step.1

You need to add this hidden fields to your form. This makes file's signed_id to sent on form resubmission.

= f.file_field :doc
= f.hidden_field :doc, value: f.object.doc.signed_id if f.object.doc.attached?

I used this page as a reference.

When using activestorage in Rails 6, how do I retain a file when redisplaying a form?

Step.2

You need to write these codes on your app. This will let you to upload files and save ActiveStorage::Blob object manually.

# In your app's controller
def update
  ...
  # `obj` is your model that using `has_one_attached`.
  if(obj.update)
    redirect_to ...
  else
    obj.attachment_changes.each do |_, change|
      if change.is_a?(ActiveStorage::Attached::Changes::CreateOne)
        change.upload
        change.blob.save
      end
      # TODO: ActiveStorage::Attached::Changes::CreateMany
    end
    ...
  end
end

That's it!

Use InputSystem package for tvOS in Unity 2019.

Takahiko Inayama — Sun, 03 May 2020 12:22:01 +0000

Use InputSystem package for tvOS in Unity 2019.

If you are trying to build games for tvOS with Unity 2019 with new InputSystem package. You'll see compile error and fails to build.
I found a solution about this.

TL;DR;

Drop a line with com.unity.modules.vr from Packages/manifest.json on your project.

Details

First, create a 3D game project with Unity 2019.3.

Then add InputSystem package.

Switch target platform to tvOS.

You'll see error messages in console when you hit a play button to build.

Library/PackageCache/com.unity.inputsystem@1.0.0/InputSystem/Plugins/XR/TrackedPoseDriver.cs(153,17):
error CS1069: The type name 'XRDevice' could not be found in the namespace 'UnityEngine.XR'.

It means UnityEngine.XR.XRDevice couldn't be found in XR namespace.

TrackedPoseDriver.cs:153 is looks like this.

        protected virtual void OnDestroy()
        {
#if UNITY_INPUT_SYSTEM_ENABLE_VR
            if (HasStereoCamera())
            {
                UnityEngine.XR.XRDevice.DisableAutoXRCameraTracking(GetComponent<Camera>(), false);
            }
#endif
        }

You may found that UNITY_INPUT_SYSTEM_ENABLE_VR is used in a preprocessor directive in TrackedPoseDriver.cs

And let's take a look at Assembly Definitions of this package.

Packages/com.unity.inputsystem/InputSystem/Unity.InputSystem.asmdef#L45-L47

    "versionDefines": [
        {
            "name": "com.unity.modules.vr",
            "expression": "1.0.0",
            "define": "UNITY_INPUT_SYSTEM_ENABLE_VR"
        },

This says UNITY_INPUT_SYSTEM_ENABLE_VR is set when com.unity.modules.vr:1.0.0 available.

There is no VR support for tvOS yet. So I dropped this without hesitation.

Tada!!

Now I can use InputSystem package.

It's a bit weird solution but works.
This error only happens on tvOS. Maybe available classes are different between platforms even if same com.unity.modules.vr packages are listed in manifest.json.

This article is originally posted on medium.