Multigrain Timestamps Try Again For Linux 6.13 - Now With Less Performance Impact

[phoronix]

Phoronix: Multigrain Timestamps Try Again For Linux 6.13 - Now With Less Performance Impact

Merged last year for Linux 6.6 was multi-grain(ed) timestamps to address the current coarse-grained timestamps when updating creation time and modification time that a lot of I/O activity can happen in the once-per-jiffy timestamp. Just a few weeks in the Linux 6.6 kernel, multi-grain timestamps were removed due to bugs. The multigrain code went back to be reworked and now just over one year later the code has been re-merged into the mainline Linux kernel...

https://www.phoronix.com/news/Linux-6.13-Multigrain-Timestamp

[chuckula]

Glad to see them going multi-grain, but we need gluten free timestamp options too.

[schmidtbag]

Glad to see them going multi-grain, but we need gluten free timestamp options too.

It's measured in jiffies. Jiffy peanut butter is naturally gluten-free, so I think we're alright.

[skeevy420]

Americans and your weird-ass measurement systems. Well, I wonder how many jars of peanut butter faster my CPU is gonna be able to monitor grain output.

[royce]

once-per-jiffy?

[cl333r]

Americans and your weird-ass measurement systems. Well, I wonder how many jars of peanut butter faster my CPU is gonna be able to monitor grain output.

I thought you're also American huh.
Anyway, what I'd like is inotify to send the rename of a file in ONE event, not like now which sends 2 events with no order guarantee or that you'll even receive the 2nd one cause it can be dropped to a different folder that you're not watching, which ultimately is a PITA to properly implement.
If I ever contribute to the Linux kernel that's what I'd try to tackle.

[skeevy420]

I thought you're also American huh.
Anyway, what I'd like is inotify to send the rename of a file in ONE event, not like now which sends 2 events with no order guarantee or that you'll even receive the 2nd one cause it can be dropped to a different folder that you're not watching, which ultimately is a PITA to properly implement.
If I ever contribute to the Linux kernel that's what I'd try to tackle.

That's why I'm so aware of our weird-ass measurement systems. About the only thing we get correct is our use of a duodecimal measurement system for building things...even if it is just a dumbed down sexagesimal.

How do y'all get thirds in metric decimal? Convert to sexagesimal (multiply by 6) and factor down?

[schmidtbag]

That's why I'm so aware of our weird-ass measurement systems. About the only thing we get correct is our use of a duodecimal measurement system for building things...even if it is just a dumbed down sexagesimal.

How do y'all get thirds in metric decimal? Convert to sexagesimal (multiply by 6) and factor down?

Unpopular opinion:
I think Fahrenheit is a better everyday unit for temperature than Celsius.
0C and 100C are only useful metrics regarding pure water at sea level on Earth. The scale means nothing outside of any other context.
Meanwhile, the F scale is pretty representative of comfort and safety levels of humans. Humans are the only ones that really care what temperature readings are, so, to me it makes sense to use the metric that applies to us regardless of where we are in the world or the universe.
0F at extended exposure is cold enough that you'll get frostbite without proper attire, and is also cold enough that salting the roads isn't going to help much with melting snow.
75F most would regard as comfortably warm.
100F at extended exposure is hot enough that you could get get heat stroke without proper attire. This of course depends on humidity levels too.

I think 100F would make more sense if it were equivalent to 40C (where you're very likely to get heat stroke) or 37C (normal human body temperature) but whatever, the measurements are close enough. I think most people can agree that you're going to want to bundle up at 0F and stay in the shade at 100F.

[skeevy420]

Unpopular opinion:
I think Fahrenheit is a better everyday unit for temperature than Celsius.
0C and 100C are only useful metrics regarding pure water at sea level on Earth. The scale means nothing outside of any other context.
Meanwhile, the F scale is pretty representative of comfort and safety levels of humans. Humans are the only ones that really care what temperature readings are, so, to me it makes sense to use the metric that applies to us regardless of where we are in the world or the universe.
0F at extended exposure is cold enough that you'll get frostbite without proper attire, and is also cold enough that salting the roads isn't going to help much with melting snow.
75F most would regard as comfortably warm.
100F at extended exposure is hot enough that you could get get heat stroke without proper attire. This of course depends on humidity levels too.

I think 100F would make more sense if it were equivalent to 40C (where you're very likely to get heat stroke) or 37C (normal human body temperature) but whatever, the measurements are close enough. I think most people can agree that you're going to want to bundle up at 0F and stay in the shade at 100F.

While I can't disagree with the generalization of Fahrenheit and Celsius, at the same time you'll just learn which temps suck and which don't as you gain experience to the measurement standard.

One thing I did disagree with is temperature scales not mattering outside of a human context. You have things like materials handling and creation, cooking and general food handling, and lots of other things where chemical reactions and state changes occur at temperatures that are in ranges well out of the human safely levels. We either have multiple temperature ranges or we adapt different things into the context of the one standard.

Take cooking. Butter and peanut oil smoke at wildly different temperatures. Around 325F for butter and 450F for peanut oil. Would 0B, butter degrees, be 90F since that's when butter changes state? Should we use the vegetable oil shortening standard, V degrees, with its melt point of 115F? 0B might end up being -5V. Should we just adapt other things to a single, common standard?

You'll eventually learn that -20 butter is comfortable and that -49 butter is when you should start to dress warm. If the butter sweats, so will you.

[billyswong]

Unpopular opinion:
I think Fahrenheit is a better everyday unit for temperature than Celsius.
0C and 100C are only useful metrics regarding pure water at sea level on Earth. The scale means nothing outside of any other context.
Meanwhile, the F scale is pretty representative of comfort and safety levels of humans. Humans are the only ones that really care what temperature readings are, so, to me it makes sense to use the metric that applies to us regardless of where we are in the world or the universe.
0F at extended exposure is cold enough that you'll get frostbite without proper attire, and is also cold enough that salting the roads isn't going to help much with melting snow.
75F most would regard as comfortably warm.
100F at extended exposure is hot enough that you could get get heat stroke without proper attire. This of course depends on humidity levels too.

I think 100F would make more sense if it were equivalent to 40C (where you're very likely to get heat stroke) or 37C (normal human body temperature) but whatever, the measurements are close enough. I think most people can agree that you're going to want to bundle up at 0F and stay in the shade at 100F.

1. Metric system units are all based on Earth scale, then redefined in terms of universal constants. Complaint of Celsius being only useful on Earth is nonsense. You may as well abandon the metric time unit Second.
2. What temperature is bearable / comfortable varies among people.
3. The water freezing point and boiling point metric depends not on the sea level on Earth, but the atmospheric pressure at sea level on Earth. We built the International Space Station with such atmospheric pressure inside. We are going to build space stations and space colonies with such atmospheric pressure in the future. So even if human beings forget why 1 "day" = 86400 "seconds" and why are they this long but not longer/shorter in the far sci-fi future, we will probably still be very familiar with 1 atm. Maybe we will reverse the rationale and say 1 atm is the pressure such that water freeze in 0C and boil in 100C. Just like the joke of floppy disk become the "physical save button".