191–200 of 218 posts
Earlier quoted context omitted.
Does apple use psci too, hence that smc redirects to el2 extension they have?
I'm guessing that's so they can implement VMs that support PSCI but believe they're using the SMC mechanism instead of HVC.
Earlier quoted context omitted.
According to MBP 16" price, Apple charges $800 for 64 GB RAM and $800 for 2 TB SSD. IMO that's a reasonable price.
2 TB of NVME storage is little more than $200 off-the-shelf. 64 gigs of DDR4, high-bandwidth laptop memory costs ~$250. Even assuming Apple is springing for high-quality, high-speed RAM, $800 borders on insanity when other laptops offer similar configuration options at less than half the price. Apple's price gouging in this department is well-documented, I don't think I need to argue with HN users about that.
I still think the prices are inflated over raw hardware but not as exaggerated as finding the cheapest parts with the same capacities would make it seem.
Earlier quoted context omitted.
> who would spend 1.500 euros (in Europe) for a laptop that has a limit of 16gb of RAM? Let's say you use a laptop for 6 hours a day, five days a week, 48 weeks a year, and that it lasts for 3 years. That's 4,320 hours of usage. For many people, it's worth paying 35 cents per hour to use a machine that's even just a little bit better than one that costs half that. I'm writing this on a high-end Macbook Pro that's 8 y…
>35 cents per hour to use WOW that's kind allot TBH.
Does the method whereby they circumvent the T2 security enclave work on Intel-based Macs? I have a 2018 Mac mini which I would love to repurpose with Linux once the end-of-life support stage is reached.
I'm working on Wi-Fi for M1 Macs this week and that particular patch will be useful for T2 Macs too, so hopefully that will be upstream in the near future.
Earlier quoted context omitted.
ARMv8 does not require EL3 (nor EL2, for that matter). They are architecturally optional, and there's a whole bunch of the spec dedicated to how things interact with their presence or absence. There is definitely no EL3 on these machines. Does rPi4 really have these features without PSCI? The internet suggests it does not support sleep states at all, and I wouldn't be surprised if it doesn't do any fancy CPU idle sta…
The rpi4 has some power management via the mailbox interface to the videocore, but it's just hacked on to the side and is more of an example of how not do things IMO. Particularly how it's exposed to user space is real gross (basically raw mailbox calls). It's closer to the primitives that psci or acpi would be papering over.
Honestly, those Broadcom chips are crazier to support than th M1s, and the rPi foundation don't exactly have much of an upstream-first reputation... so I'm quite confident we can do a lot better.
> Amusingly, while implementing support for this in Linux, we ran into a bug in Linux’s ARM SMMU support that had been there ever since 52-bit address support was introduced. This was breaking systems with more than 256 TiB of RAM - I wonder why nobody noticed? Either way, Linux now correctly supports standard ARM systems with up to 4 PiB of RAM ;-). Interesting how working on new technologies helps improve existing…
> This was breaking systems with more than 256 TiB of RAM - I wonder why nobody noticed? Are there even ARM systems with that amount of RAM? I assume it's something in the order of what super computers use, but I don't think any of them are ARM. That doesn't sound that surprising to me that no one noticed if no one uses that much RAM.
Earlier quoted context omitted.
The rpi4 has some power management via the mailbox interface to the videocore, but it's just hacked on to the side and is more of an example of how not do things IMO. Particularly how it's exposed to user space is real gross (basically raw mailbox calls). It's closer to the primitives that psci or acpi would be papering over.
It's honestly a bit amusing how many people have been doubting how good Asahi can get without official support, but then you look at "officially supported" platforms like Raspberry Pi and so much stuff is barely held together with string and kernel forks... Honestly, those Broadcom chips are crazier to support than th M1s, and the rPi foundation don't exactly have much of an upstream-first reputation... so I'm quite…
AFAICT the way things are going y'all are setting the bar for support for a platform regardless of if you're first or third party.
Alder Lake, Intel's flagship dekstop class CPU, is only 10-15% faster than the M1 Pro/M1 Max in CPU performance but it requires an insane amount of cooling and the power consumption is frankly ridiculous. Seems like Apple Silion will be 5 years ahead of the competition - similar to how long it took for others to catch up with the first iPhone.
Performance per Watt only matters to Intel when they are leading. Otherwise they brute force to be "fastest". I think that the ARM in Macs is on a different trajectory than Intel. The battery life along in an ARM based Mac is second to none. That's more important to most folks, IMO.
These new macbooks are like a dream. By far the best laptops I have ever used.
Earlier quoted context omitted.
> My ThinkPad X1E has horrible battery life under Linux I ran the M1 Pro for a whole day during early testing without charging (because we hadn't initialized the USB-C port yet) and didn't even run out of battery. And I didn't even have the power management driver running yet. > has a boatload of issues with thunderbolt We'll see how that goes, but I get the feeling Apple's Thunderbolt controllers are going to be a l…
Love the work! One request though, could future progress reports include a little bit about the work being done for OpenBSD also? I know there is someone on the team who is working on the porting effort to Linux, but is also working on OpenBSD concurrently. Thank you!
Earlier quoted context omitted.
I'm curious, what don't you like about the X1E? I have a Gen 3 and it's super wonky with an external monitor, nvidia + linux seems to still be a mistake.
Battery life isn't great (I occasionally take it out for about 2-2.5 hours which it does manage but a longer duration would probably kill it). Once froze and refused to resume when I tried to put it on standby and I had to do multiple hard resets to somehow make it going again (could be related to S3 config in BIOS) - scared me off to ever try to standby again. Also, I think you are 100% correct on Nvidia + Linux poi…