Please create an account to participate in the Slashdot moderation system

 



Forgot your password?
typodupeerror

Comment Re:Fragmentation (Score 2) 100

ARMv8 is not eliminating them, it's reducing the number of instructions that have them. Conditional instructions are useful because you can eliminate branches and so keep the pipeline full. For example, consider this contrived example:

if (a < b)
  a++;

On ARMv7 and earlier, this would be a conditional add. The pipeline would always be full, the add would always be executed, but the result would only be retired if the condition is true. On MIPS, it would be a branch (complete with the insanity known as branch delay slots, which if you look at the diassembly of most MIPS code typically means with a nop, so you get to waste some i-cache as well) and if it's mispredicted then you get a pipeline stall.

On ARMv8, you don't have a conditional add, but you do have a conditional register-register move and you have twice as many registers. The compiler would still issue the add instruction and then would do a conditional move to put it in the result register. From the compiler perspective, this means that you can lower PHI nodes from your SSA representation directly to conditional moves in a lot of cases.

Basically, 32-bit ARM is designed for assembly writers, ARMv8 is designed for compilers. As a compiler writer, it's hands-down the best ISA I've worked with, although I would prefer to write assembly by hand for ARMv7. I wouldn't want to do either with MIPS, although I currently am working on MIPS-based CPU with some extra extensions.

Actually, ARM's reasoning is that modern branch predictors on high end AP's can do a good enough job of following a test and branch and keeping the pipeline(s) full that there is very little value in conditional instructions on future chips. It's hard to cause a pipeline stall or bubble by branching a few instructions forward or back on these CPUs since they are decoding well in advance of the execution pipelines. Added to that, there is an energy cost in executing an instruction and throwing away the result. Obviously, not all cases are wins. In the example you noted, a register to register mov on a register-renaming system is basically a 0-cycle operation (never makes it out of the instruction decoder), so it's hard to do better than that.

Security

Israel Faces Escalating Cyberwar 200

New submitter 9re9 writes "The NY Times describes what may be the beginning of an actual cyberwar between a pro-Palestinian group and Israeli companies, specifically El Al and the Tel Aviv stock exchange. From the article: 'A hacker identifying himself as oxOmar, already notorious for posting the details of more than 20,000 Israeli credit cards, sent an overnight warning to Israel's Ynet news outlet that a group of pro-Palestinian cyberattackers called Nightmare planned to bring down the sites in the morning.' Though the article is skimpy on technical details, the group appears to have engaged merely in a DDOS attack. Hamas praised the attack as opening 'a new resistance front against Israel.' Is this the first acknowledged cyberwar?"

Comment Re:Iran never called for Israel's destruction (Score 1) 164

This is a reference to an older quote by Khomeini, but in Khomeini's days Iran was an ally of Israel. The alliance was broken off after the 1990 gulf war, because with Iraq gone and the USSR gone Israel had to change its strategy and Iran became a regional competitor.

I stopped reading after this sentence since it is so wrong, I assume the rest of it is also drivel. Israel and Iran were allies up until the Islamic revolution of 1979 in which Khomeini took power. You may have heard about the Iran hostage crisis, which happened at the same time when Iran's friendly relationship with the US was similarly dissolved. The Islamic council of Iran, which still rules today, has always been hostile to the US and Israel

Handhelds

When Should I Buy an Android Tablet? 396

jpyeck writes "I've deliberately avoided the smartphone craze, due to the fact I've never utilized any phone (landline or otherwise) enough to justify the monthly fees. But the geek in me craves the 'smart' part of the equation, especially since I got a bonus this year-end that is burning a hole in my pocket. The iPad is out of the question because I need a bit more hack-ability in my gadgets. I am drooling over the Android Honeycomb demo from the CES. I've had my eye on the Galaxy Tab, though it sounds like it won't support Honeycomb. The Xoom looks great, but who knows when it will come out? The consensus seems to be 'wait a few months for Honeycomb.' If you were me, with limited patience, would you buy an Android tablet now? If so, which?"

Comment Re:"Other minor changes"? (Score 1) 212

This is a huge deal for developers since the iPad and iPhone/touch are now running the same software release. This means that developers no longer needs to target iOS 3.2 as the lowest common denominator for all devices, and can instead start using the new 4.x features (assuming they don't care about the very first generation iPhone/touch).

Comment What EFI is and isn't (Score 5, Informative) 216

Every time Slashdot has a story of EFI, we get a thousand uninformed posts about what it is, isn't, what it should do and why it sucks. As someone who has worked on EFI for years, let me clear them up for you:

1) EFI was designed by Intel as a replacement for BIOS. UEFI (edk2) is the second generation EFI, and is open source (see: http://tianocore.sourceforge.net). Intel delivers all of their boot support code as EFI drivers these days.

2) EFI is NOT a graphical interface. Some Chinese motherboard makers created terrible graphical configuration applications for it. Apple created a pretty nice boot selector. It can just as easily "post" and give you a console-mode menu like you're used to.

3) EFI is very common. All Apple computers use it. Most PC (Windows / Linux) laptops use it (your laptop probably does if it was made in the last 4 years). EFI drops into BIOS emulation mode after boot because Windows doesn't support it.

4) EFI machines generally allocate a small EFI partition on the hard drive, particularly if they use GPT. All Intel-based machines boot from flash memory and would successfully boot without this partition. This partition is for additional EFI firmware volumes or drivers that can be dynamically loaded.

5) EFI is much better than BIOS. It runs in full 32/64 bit mode. It can dynamically load drivers built into the ROM of your hardware (like a video card) and therefore doesn't have to rely on ancient backward-compatibility modes. It can run "apps", like a safe firmware updater so you don't have to boot your PC with a DOS boot disk to update the firmware. It can communicate a lot of configuration information to the OS and even provide hooks for some low level hardware-specific drivers. It can do things like boot from a network-shared CD-ROM drive or from a disk image stored on a USB stick (without resorting to making bootable partitions and jumping through a bunch of hoops like your average Linux USB stick). EFI can read FAT, NTFS, EXT2, HFS+ filesystems and boot the kernel directly from there (and the initrd image) without involving grub or other second stage boot loader. It can boot your GPT-tagged disks in your chosen order no matter what order you changed them around (take your boot drive, move it to a USB enclosure, boot from it).

Booting Intel machines is really fucking complicated, and EFI makes it much simpler.

Comment Re:Windows Phone 7 is great (Score 2, Informative) 319

For example, side by side, the Android browser in Froyo as running on a Nexus One has been demonstrated to be faster than the iPhone 3GS and the iPad despite the fact that they are both based on Webkit.

The Nexus One has a 1 GHz processor and 512 MB of RAM, whereas the 3GS has a 600 MHz (both ARM Cortex A8) and 256 MB of RAM. The fact that up until Froyo the browser didn't run faster should be quite embarrassing.

Comment Re:Transparent is no lie (Score 1) 114

Google needs your data, just like how you need Google. Search is their core business, after all. What we need to make sure is that those TOS and agreements are not just some legal stuff to make the whiners go away, and it's in their enlightened self-interest to make sure we can verify it. Google is not Microsoft: it won't cost you days of work and months of learning to move away from their products if they piss you off, and they know it.

You are mistaken, advertising is their core business. Search, email, maps, mobile operating systems, these are just hooks to lure in the product (users) to sell to the advertisers. Protecting your privacy is contrary to their entire business model, which is to sell targeted advertising based on collected data. They will change some wording around to placate loud complaints, but at their core they will continuously encroach into your personal information as best as they can.

Microsoft

Microsoft "Courier" Pictures 230

tekgoblin writes to let us know that Gizmodo has some early shots of the new prototype "Courier" booklet (foldable tablet) on the way from Microsoft. "Courier is a real device, and we've heard that it's in the 'late prototype' stage of development. It's not a tablet, it's a booklet. The dual 7-inch (or so) screens are multitouch, and designed for writing, flicking and drawing with a stylus, in addition to fingers. They're connected by a hinge that holds a single iPhone-esque home button. Statuses, like wireless signal and battery life, are displayed along the rim of one of the screens. On the back cover is a camera, and it might charge through an inductive pad, like the Palm Touchstone charging dock for Pre."

Slashdot Top Deals

grep me no patterns and I'll tell you no lines.

Working...