On 10/25/07, James Cornell <sparcdr at sparcdr.com> wrote: > Tony Reeves wrote: > > On 10/25/07, James Cornell <sparcdr at sparcdr.com> wrote: > > > >> Tony Reeves wrote: > >> > >>> On 10/24/07, James Cornell <sparcdr at sparcdr.com> wrote: > >>> > >>> > >>>> I heard that the AMD Opteron 1218 was the first to get support. > >>>> (Ultra-20 M2) Are you telling me I'm out in the cold? It supports > >>>> clock reduction, has several modes. Just not in Solaris because it's > >>>> older than dirt and wasn't engineered with x86 as a priority, as it's > >>>> obvious to me from your bit about TSC reliance. > >>>> > >>>> > >>> Hmmm, I see the same sort of clock/timing issues with Linux and > >>> Windows XP/Vista on my Turion X2 laptop. And I am aware that at least > >>> one of the *BSD groups have been working on the issue. So I don't > >>> think the issue is unique to OS's that are older than dirt (whatever > >>> that means) or engineered with or without x86 as a priority. > >>> > >>> > >> Means exactly what it means. Sun didn't innovate much until Solaris 9, > >> and Solaris 10. Casper can vouch at the time it was a matter of > >> "hanging fruit" and the priority of money and manpower was not in the > >> x86 area. It's older than dirt so to speak because for years it > >> architecturally remained the same, in some parts it didn't improve when > >> it would had been in their best interest. > >> > >> > > > > Quite frankly, bollocks. Maybe I am too old, but I recall innovation > > in SunOS 4, 4.1 (support for SMP for example), SunOS 5 with a new SVR4 > > based kernel (painfull at the time but we reap the benefits now), > > SunOS 5.2 (when binary compatibility really kicked in) and so on. We > > see innovation now, because it is on our face, but I believe there > > truly has been some form of real innovation from Sun every year. And > > much of it has been in the OS. A lot of the innovation has been > > behind the scenes and until recently largely invisible to users except > > for improved performance and security, but it has happened. > > > Now you're talking about the SunOS days. I'm referring to in comparison > to the high level of innovation now to the lack of innovation abroad > between Solaris 2.6-8. 9 is when they started doing new and important > changes. I do respect the innovation done with the original SVR > reimplementation, and I do enjoy the fruits of a mature ABI. It is > second to none, most backwards and mostly forwards OS I've ever had > experience with. One good innovation during the dark ages was being > essentially the second vendor to have full 64-bit capability, this was > back in 1997. The only other vendor to have mostly stable kernel API, > and stable ABI was DEC's Alpha, actually I believe they were the first > to engineer and deliver 64-bit microprocessors.
Ahhhh, Solaris IS SunOS, plus other applications and GUI etc.. Right now we are at SunOS 5.10, production and SUNOS 5.11, beta/nevada. And between Solaris 2.6-8 there was massive innovation around kernel threads, locks etc.. But it was largely invisible to the outside world, except for the fact that every version went faster on the same hardware. It needed to, due to the slow rate of progress with hardware. I guess this is what got me most with your post, it was the OS engineers, plus the DB tuning gurus, that kept Sun hardware in many accounts when the hardware was slipping further ad further behind.
