May be it is just getting hot ? Are you using at least the biggest meanest CPU cooler you could find ? I run a Cyrix 166 and 233 and don't have any problems.
Peter Barbera Eric Jacoboni wrote: > > Hi, > > I've a Cyrix P166+ with 96 EDO Ram on an Asus motherboard > PI55T2P4C and a Matrox Millenium MGA video card. My system runs under > Debian Hamm. > > I often use Xemacs and Gnus (original deb packages) under X11 > (3.3.2.3-1) and Wmaker (0.19 from slink). > > Too often, the system freeze when under X11 and the _only_ solution is > to press the reset button :-( No way to have a console. > > I've read docs about set6x86 and i've removed it from rc.boot to be > sure the problem doesn't come from a bad configuration of it. In both > cases (with or without set6x86) the problem arises... > > 6x86_reg gives now : > > =-=-=-=-=-= > > 6x86 (Classic/L/MX) Register Dump utility > > 6x86 DIR0: 0x31 2X core/bus clock ratio > DIR1: 0x16 6x86 Rev. 2.6 > > Wait a moment... > Calculated BogoMIPS: 130.00 > Kernel BogoMIPS: 106.09 > > 6x86 CCR0: 0x2 NC1 set (address region 640Kb-1Mb non-cacheable) > CCR1: 0x82 NO_LOCK reset > CCR2: 0x80 SUSP_HLT reset (low power suspend mode disabled) > CCR3: 0x10 > CCR4: 0x17 DTE cache enabled, no I/O recovery time > CCR5: 0x21 slow LOOP disabled, allocate cache lines on write misses > > 6x86 Address Region Register dump: > ARR0: address = 0xA0000 , size = 128 KB > RCR = 0x9 : not cached, write gathering > ARR1: address = 0xC0000 , size = 256 KB > RCR = 0x1 : not cached > ARR2: disabled > ARR3: address = 0xA8000 , size = 32 KB > RCR = 0x9 : not cached, write gathering > ARR4: disabled > ARR5: disabled > ARR6: address = 0x6000000 , size = 32 MB > RCR = 0x1 : not cached > ARR7: address = 0x0 , size = 128 MB > RCR = 0xB : cached, weak write ordering, write gathering > =-=-=-=-= > > I've read the doc about the Cyrix 'coma' bug and i've try to use > set6x86 to set the NO_LOCK bit : no change... > > (BTW : the doc speaks about a little source code to detect this coma > bug but i've found nowhere...) > > I've read the various log files : nothing... Is there a way to have a > trace of the crash (a core dump or something like that ?). > > Thanks for any advice... > -- > ----------------------------------------------------------------------- > Éric Jacoboni « J'ai épuisé le tout-venant, je vais me risquer dans > le bizarre » (M. Audiard) > ----------------------------------------------------------------------- > > -- > Unsubscribe? mail -s unsubscribe [EMAIL PROTECTED] < /dev/null