MrZipf

JSC is barely ever mentioned in the v8 team. They gave up wanting to be benchmark kings in the interest of having decent performance in Android markets (low-end and high).

Comments around cheating are likely aimed at OEMS who ship Chrome derived browsers and spin up the clock frequency when they detect certain benchmarks are running. These look great in online reviews.

This was stupid comment, long day. Totally looking at this the wrong way.

How does the hardware visible context get suitably updated if the confidential data in the other SIP isn't touched by execution (speculative or otherwise)? Doesn't something need to be pulling that state into the visible context?

In Singularity, the system gets to decide what happens at SIP boundaries (and the kernel ABI boundary).

HIP is the full address space change, but any mitigation steps can be introduced in the IPC hand-off between SIPs (or at the kernel ABI) compiled into the untrusted process. This code is compiler controlled. The system is in ring-0 so IPC code gets full access to available instructions (such as mitigation is possible there).

Of course doing this makes channel communication in both directions slow for untrusted processes, but that is the cost of doing business with Spectre. And if someone wrote a browser that didn't use channels for talking to the JS engine, then all bets would be off.

Preliminary VarHandle support in Android start in P. The API is hidden, but it had to start ahead of the Java 9 libraries because it's a lot of work.

All code needs to be capable of being interpreted as that's the fallback when compilation invariants are violated (and also when stepping through code with the debugger).

A large number of operations combined with different vartypes and backing store types (fields, static fields, arrays, Buffers, etc) makes for a lot of work, even though this is just the slow path.

There is no compiler support yet for VarHandles, or MethodHandles, on Android, but there are bugs on these.

Hey Joe, can't you just stop nagging users to use your browser and stop overriding the users choice of browser? No need to repackage anything, just move on.

Completely agree.

My home network is built with Mikrotik kit which is priced where it's affordable to have spares. I have yet to encounter a failure, but could drop in a new router in a couple of minutes with the saved configs.

I have SNMP monitoring feeding from telegraf into influxdb on an RPI. Dashboard rendered with Grafana on PC. Also have telegraf pinging to all 24x7 devices and collecting data from electricity meter, smartplugs, and Nests. It's been fun to do.

I love what DeepMind is doing, but find their publication choices bizarre. Sure Nature has the occasional ML / NN paper, but it's not a top journal for AI / ML / CS and it makes getting hold of papers awkward and it doesn't seem in the spirit of most CS research.

Microsoft does not preinstall third party bloatware, that's OEMs running on thin margins. There have been Microsoft efforts to share with OEMs the effect of bloatware on boot times and incentivizing boot time reduction.

It doesn't take 10 minutes for a LaTeX or Word user to get frustrated with docs. What doesn't it do:

1. user defined style definitions

2. citations / bibliography

3. anchoring

There are plugins that can help, but at least where I work using these is often banned to avoid the risk of leaking corp info.

Why is this HN material? It's a Slack customer advisory.