SoylentNews is people
SoylentNews
Greg Kroah-Hartman, the stable Linux kernel maintainer, could have prefaced his Open Source Summit Europe keynote speech, MDS, Fallout, Zombieland, and Linux, by paraphrasing Winston Churchill: I have nothing to offer but blood sweat and tears for dealing with Intel CPU's security problems.
Or as a Chinese developer told him recently about these problems: "This is a sad talk." The sadness is that the same Intel CPU speculative execution problems, which led to Meltdown and Spectre security issues, are alive and well and causing more trouble.
The problem with how Intel designed speculative execution is that, while anticipating the next action for the CPU to take does indeed speed things up, it also exposes data along the way. That's bad enough on your own server, but when it breaks down the barriers between virtual machines (VM)s in cloud computing environments, it's a security nightmare.
(Score: 1, Interesting) by Anonymous Coward on Wednesday October 30 2019, @01:21PM
SPARC Processors date back to the 1980s. Sparc32(v7) ended in ~1996-1996, when Sun stabbed most of its chip partners in the back by migrating to Sparc64(v8) and deprecating OS support needed for people still producing Sparc32 compatible chips in order to pull a chipzilla and take the whole Sparc market for themselves (what Intel did with the Pentium era chipsets leading to a number of company's bankruptcy and then the Invention of Super Socket 7 when Intel completed the move with the Pentium 2 Slot-1 interface, leading to the split between Intel and AMD bus interface and sockets and the collapse of all but two other players in the market (ALi and VIA, the former becoming ATI's chipset branch for both Intel and AMD during the late socket 478/early 775/939/940 era. VIA moved away from sockets altogether as part of their settlement with Intel over violating it's IP related to the 370 and 478 socket and bus interfaces.)
Point being: Java didn't come out until... 1995? And if I remember correctly the originally released implementation was for the Pentium with support for the SPARC virtual machine not coming until a year or two later. The project goal was to get most software for SPARC, which was losing out at that point to Wintel, by producing fully portable binaries that would run on any system. Unfortunately the initial GUI libraries for it all sucked, there were a number of other issues with performance and security cornercases, and it wasn't until a few years later (1.2 or 1.3?) that it really came into its own as people had servers beefy enough to run it and it turned out quite well for business logic, web interfaces to COBOL, etc. The move to JIT compilation made another big leap, but the shaky legal groud for second source implementations, and Sun's will I won't I IP handling made it into a Second Microsoft situation for many in the internet community, fresh off the heels of the anti-trust suit, and looking to just add another layer of lockdown to an already shaking foundation of computing. Java 1.0-1.4 library and code changes didn't help either.
The real reason SPARC died was an inability to bring in new operators of the technology due to not providing an entry level system that people could actually afford. The only serious attempt at this was the Sun Blade 100/150 systems back in 1999-2001 era and those required expensive unregistered ECC SDRAM, only had one PCI expansion bus for videocards and any other peripherals, usb 1.0, and a clock speed that was 200-400 mhz less for 400+ dollars more than you could build a PC for at the same time.