Berkeley CSUA MOTD:Entry 37749
Berkeley CSUA MOTD
2017/10/18 [General] UID:1000 Activity:popular

2005/5/18-19 [Computer/HW/CPU, Computer/SW/OS/Windows] UID:37749 Activity:kinda low
5/18    Hyper-threading considered harmful:
                \- You mean it isn't fun and profitable?
        \_ I don't know about security problems, but lots of testing has
           led us to believe that in most instances it does not help
           performance. The number of cases it improves performance is
           small compared to the number where it degrades performance -
           especially on a multiuser machine. I always turn it off.
           \_ In my experience it helps responsiveness, at least in Windows.
              Windows 2K/XP threading just seems to work better when there are
              more than one processor.
           \_ I agree with pp.  Hyperthreading helps GUI responsiveness (Alt-
              Tabs, etc.), but does not decrease total execution time of a
              given process.  This is also the general wisdom.
              \_ Not only does not decrease, but can increase time spent
                 in CPU.
              \_ Well it *can* decrease the total time of two processes,
                 if they have certain properties.
                 \_ For most apps these days responsivenss under load is
                    a lot more important that some minor gain in raw speed.
                    I'd gladly sacrafice 1% speed (and we are really talking
                    less than that) for a noticable gain in UI responsiveness
                    when my machine is pegging.
                    \_ We are not talking about less than 1% in speed and
                       some of us actually do heavy work on our computers.
                       Depending on what you are doing, the UI response
                       might actually be *slower* because the OS thinks it
                       has 2 CPUs when it only has 1 (worse scenario is
                       when it thinks it has 4 and only has 2 and, I
                       imagine, gets worse the more CPUs you have). I suggest
                       always turning off hyperthreading.
        \_ Is the hyperthreading thing the same as bundling two CPUs together?
           I have a dual P3 machine, and it helps when my compiler compiles
           multiple modules in two parallel processes.
           \_ No.
2017/10/18 [General] UID:1000 Activity:popular

You may also be interested in these entries...
2009/8/6-14 [Computer/SW/OS/OsX] UID:53250 Activity:moderate
8/5     Why is Mac OS 10.6 $29 and 10.5.6 $129? Is it a typo?
        \_ $29 for existing users.
           \_ it doens't even support ppc does it.
              \_ who cares about ppc anymore? Everything is Intel based
                 \_ I have a PPC mini at home that I use.
                 \_ I have a quad core G5 ppc.
2009/4/16-6/29 [Computer/SW/OS/Linux] UID:52852 Activity:low
Linux http://soda.CSUA.Berkeley.EDU 2.6.26-2-amd64 #1 SMP Sun Jun 21 04:47:08 UTC 2009 x86_64
The programs included with the Debian GNU/Linux system are free software;
the exact distribution terms for each program are described in the
individual files in /usr/share/doc/*/copyright.
Debian GNU/Linux comes with ABSOLUTELY NO WARRANTY, to the extent
permitted by applicable law.
2009/6/1-3 [Computer/HW/CPU] UID:53068 Activity:high
5/31    History of winners and losers by *popularity*:
        VHS > Beta Max
        USB2 > Firewire
        x86 > PowerPC > Everything Else > DEC Alpha > Itanium
        BlueRay > HDDvd
        \_ It's too early to tell RE: "Blue"Ray. They may both turn out to be
2009/5/26-30 [Computer/HW/CPU] UID:53045 Activity:nil
5/26    Engineering is HOT man! Super hot co-inventor of USB at Intel:
2009/5/6 [Computer/HW/CPU] UID:52952 Activity:moderate
5/5     Any idea why the AMD only added 8 more registers with the AMD64 arch.?
        Why 16 registers and not, say, 32?
        \_ To double the accessible register count, you need one more bit
           in the instruction stream _per_ register field.  x86 instruction
           set is already packed like rat's nest as is.
2009/1/16-23 [Computer/HW/CPU] UID:52404 Activity:nil
1/16    AMD to layoff 9%, suspend 401(k) match, cut engineer salaries 10%
        \_ Awwww, too bad                                       -Intel
           \_ My heart bleeds for you. --transmeta.
              \_ Wait, another sodan worked there? --ex-transmeta
                 \_ Hello transmeta-coward, meet another transmeta-coward.
2008/12/4-10 [Computer/HW/CPU, Computer/HW/Drives] UID:52163 Activity:nil
12/4    A question to you old crufy alumni: So lately we've suggested
        VMs, and been asked why it's necessary. We've suggested top-of-the-line
        hardware and been told we don't need that much power. So I'd like to
        ask -- what exactly do you think the CSUA is supposed to _be_?
        \_ Noone said VMs weren't needed.  They suggested you use the
        \_ No one said VMs weren't needed.  They suggested you use the
2008/12/4-10 [Computer/HW/Memory, Computer/HW/Drives] UID:52172 Activity:nil
12/5    What would you guys think of this?
        TYAN Tank barebone
        2x Intel Quad Xeon E5420
        16GB ram
2008/12/3-8 [Computer/HW/CPU] UID:52157 Activity:kinda low
12/3    Are any of you CSUA alums working at Intel?  Is it possible that we
        might be able to hit Intel up for donated/partially donated (reduced
        price) Core i7 Xeons when they come out?  Who would be a good person
        to contact about something like that?  We're of course willing to put
        out for them - perhaps we'll tattoo an Intel logo on toulouse if that's
        what they want :-p  -- steven
2013/3/25-5/18 [Computer/SW/OS/Windows] UID:54639 Activity:nil
3/25    It's strange that only every other Windows version is a hit.
        NT 4.0: hit
        2000:   flop
        XP:     hit
        Vista:  flop
        7:      hit
2013/2/19-3/26 [Computer/SW/OS/OsX] UID:54611 Activity:nil
2/19    I program a lot by sshing to a Linux cluster.  So I'm used to using
        Xemacs to code.  This works fine from a Linux or Windows workstation,
        but sometimes I have to use a Mac.  On Mac, the meta is usually
        bound to option, but that often doesn't work over ssh for some reason.
        This makes using emacs a real pain.  Any suggestions on how to fix it?
        (Other than "use vi")
2012/3/15-6/1 [Computer/SW/Languages, Computer/SW/OS/Windows] UID:54340 Activity:nil
3/15    Why does MS put double-quotes around the '8' in Windows Server 8, like
        the following?
        - Windows 8
        - Windows Server "8"
        \_ Because when they didn't do it, code didn't see the '\0'
           and went over?  Looks better than '8','\0' *shrug*
2012/2/23-3/26 [Computer/SW/OS/Windows] UID:54312 Activity:nil
2/23    fixboot wrote FAT boot sector to my WinXP hard drive.  How can I convert
        the drive back to NTFS?
        \_ Does C:\WINDOWS\system32\convert.exe work?
2011/12/21-2012/2/6 [Computer/SW/Editors/Emacs, Computer/HW/Laptop] UID:54269 Activity:nil
12/21   In Emacs, how do I make it so that it will resize its screen
        when the maximum resolution of my monitor changes? When I
        use my laptop, my emacs is too big (and I can't resize it
        because the bottom-right corner is not accessible).
        \_ Which OS?  Can't you drag the top border (not a corner) to resize
           the height to be smaller?
2011/11/8-30 [Computer/SW/Security, Computer/SW/OS/Windows] UID:54218 Activity:nil
11/8    ObM$Sucks
        \_ How is this different from the hundreds of other M$ security
           vulnerabilities that people have been finding?
           \_ "The vulnerability could allow remote code execution if an
               attacker sends a continuous flow of specially crafted UDP
2011/5/19-7/30 [Computer/SW/Security] UID:54110 Activity:nil
5/19    Uh, is anyone still using this? Please mark here if you post and
        haven't added this yet. I'll start:
        \_ person k
        \_ ausman, I check in about once a week.
        \_ erikred, twice a week or so.
        \_ mehlhaff, I login when I actually own my home directory instead of
2011/2/14-4/20 [Computer/SW/Unix] UID:54039 Activity:nil
2/14    You sure soda isn't running windows in disguise?  It would explain the
        \_ hardly, My winbox stays up longer.
        \_ Nobody cares about uptime anymore brother, that's what web2.0 has
           taught us.  Everything is "stateless".
           \_ You;d think gamers would care more about uptime.
2011/4/6-20 [Industry/Startup, Computer/SW/Unix, Computer/SW/Mail] UID:54078 Activity:nil
4/6     My company is evaluating version control systems. Our two candidates
        are Perforce and Subversion. Anyone worked with both and have good
        arguments one way or the other? (These are the only two options we
        have.) We're most interested in client performance, ease of use, and
        reasonable branching.
        \_ I'll be 'that guy'. If perforce and subversion are optins, why isn't
Cache (3529 bytes) ->
Hyper-Threading Considered Harmful Hyper-Threading, as currently implemented on Intel Pentium Extreme Editio n, Pentium 4, Mobile Pentium 4, and Xeon processors, suffers from a seri ous security flaw. This flaw permits local information disclosure, inclu ding allowing an unprivileged user to steal an RSA private key being use d on the same machine. Administrators of multi-user systems are strongly advised to take action to disable Hyper-Threading immediately; single-u ser systems (ie, desktop computers) are not affected. Cache Missing for Fun and Profit, discussing this flaw and related problems, both realized and theoretica l Vendor statements The following statements have been provided to me by vendors: CVE: The Common Vulnerabilities and Exposures (CVE) project has assigned the name CAN-2005-0109 to the problem of information disclosure resultin g from cache evictions in simultaneous multithreading processors. CVE list, which standardizes names for security problems. NetBSD: The NetBSD Security-Officer Team believes that workarounds will b e suitable for the majority of our users. Since this issue is a complex one, the 'right' solution will require a larger discussion which is only possible once this issue is public. This issue will be addressed in adv isory NetBSD-SA2005-001, which will provide a list of workarounds for us e until the 'final' conclusion is reached. OpenBSD: OpenBSD does not directly support hyperthreading at this time, t herefore no patch is available. Affected users may disable hyperthreadin g in their system BIOS. We will revisit this issue when hyperthreading s upport is improved. UPDATE: Linus Torvalds has commented on this on the linux-kernel mailing list. He clearly doesn't understand the problem -- but in his defence, h e's not a security guy. Hopefully the Linux developers who do understand security will talk some sense into him. As far as I know, this flaw only exists on Intel processors. This may be because they're too busy fixing the problem, or it may be due to corporate policies which forbid such disclosures. Either way, if there isn't a statement above, it's because I haven't received one. AMD64 is the name given to the 64-bit version of the x86 architecture developed by AMD Recent Intel processors with "EM64T" use this architecture. For the past three months, I've spent almost all of my time working on this security flaw -- investigating how serious it was, contacting all of the affected vendors, explaining how this should be fixed, et cetera. I simply haven't had time to go out and get a job -- and I decided that making sure that this issue was properly reported and fixed was far more important than earning some money. In all seriousness, I'd like to spend a few months writing a completely new cryptgraphic library which is designed from the ground up to be immune to this attack, as well as all other timing attacks. I can't guarantee that I'll be able to reply to everyone -- I have no idea how many emails I'll get -- but I will make an effort to address every serious question I receive either via personal email or on this web page. Disclosure timeline Late October, 2004: Initial discovery. December 2004: Proof-of-concept exploit written and tested. December 31, 2004: FreeBSD Security Officer Team notified of upcoming sec urity issue. February 27, 2005 - March 18, 2005: Other security teams and vendors (inc luding Intel) contacted. May 13, 2005 @ 00:00 UTC: Official public disclosure that a security flaw exists in Hyper-Threading.