What kernel are you using? ali
On Jan 24, 2011, at 4:26 PM, Mario Donato Marino wrote: > Yes, Ali you were right, I was using tsb_osfpal and not tl_osfpal. > > With the a console and a tsb_osfpal version compiled by myself, using 2500MB > as the memory size (in Benchmarks.py), the simulation breaks here: > > linux output: > M5 console: m5AlphaAccess @ 0xFFFFFD0200000000 > Got Configuration 623 > memsize 9C400000 pages 4E200 > First free page after ROM 0xFFFFFC0000018000 > HWRPB 0xFFFFFC0000018000 l1pt 0xFFFFFC0000040000 l2pt 0xFFFFFC0000042000 > l3pt_rpb 0xFFFFFC0000044000 l3pt_kernel 0xFFFFFC0000048000 l2reserv > 0xFFFFFC0000046000 > kstart = 0xFFFFFC0000310000, kend = 0xFFFFFC0000899860, kentry = > 0xFFFFFC0000310000, numCPUs = 0x2 > CPU Clock at 2000 MHz IntrClockFrequency=1024 > Booting with 2 processor(s) > KSP: 0x20043FE8 PTBR 0x20 > KSP: 0x20043FE8 PTBR 0x20 > Console Callback at 0x0, fixup at 0x0, crb offset: 0x790 > Memory cluster 0 [0 - 392] > Memory cluster 1 [392 - 319608] > Initalizing mdt_bitmap addr 0xFFFFFC0000038000 mem_pages 4E200 > ConsoleDispatch at virt 100008D8 phys 188D8 val FFFFFC00000100A8 > Bootstraping CPU 1 with sp=0xFFFFFC0000076000 > unix_boot_mem ends at FFFFFC0000078000 > k_argc = 0 > jumping to kernel at 0xFFFFFC0000310000, (PCBB 0xFFFFFC0000018180 pfn 1101) > > Memory size seems right. > > Which M5 trace/code you suggest me to look at? > > Thanks! > > Mario > > On Wed, 19 Jan 2011, Ali Saidi wrote: > >> >> You should have a new tsb_osfpal and console. tl_osfpal is for a Alpha >> platform m5 no longer supports. How does it break? >> >> Ali >> >> >> On Wed, 19 Jan 2011 13:52:34 -0500 (EST), Mario Donato Marino >> <[email protected]> wrote: >>> Hey Ali, thanks for your reply! >>> I've re-compiled and generate a new console binary, however the >>> simulation breaks earlier than before (at least before, I was able to >>> use it perfectly under 2GB). Maybe I'm missing something: should I >>> also have tl_osfpal recompiled ? Are there any other binaries or >>> modifications required to use more than 2GB? >>> >>> Thanks, >>> >>> Mario >>> On Sun, 16 Jan 2011, Ali Saidi wrote: >>>> If you compile the console binary yourself from the alpha-system >>>> repository you shouldn't have that problem anymore. >>>> Ali >>>> On Jan 15, 2011, at 4:22 PM, Mario Donato Marino wrote: >>>>> Hi, >>>>> >>>>> I was wondering what is the maximum memory size I can use in >>>>> Benchmarks.py. For example, when using up to 2GB everything works fine, >>>>> however if I use 3GB, m5 simulation aborts and I get a couple messages >>>>> in the terminal file: >>>>> memsize FFFFFFFFC0000000 pages FFFFFFFFFFFE0000 >>>>> setup: forcing memory size to 33554432K >>>>> >>>>> Can anybody give any hints on how to increase the maximum memory size? >>>>> >>>>> Thanks, >>>>> >>>>> Mario >>>>> _______________________________________________ >>>>> m5-users mailing list >>>>> [email protected] >>>>> http://m5sim.org/cgi-bin/mailman/listinfo/m5-users >>>> _______________________________________________ >>>> m5-users mailing list >>>> [email protected] >>>> http://m5sim.org/cgi-bin/mailman/listinfo/m5-users >>> _______________________________________________ >>> m5-users mailing list >>> [email protected] >>> http://m5sim.org/cgi-bin/mailman/listinfo/m5-users >> >> _______________________________________________ >> m5-users mailing list >> [email protected] >> http://m5sim.org/cgi-bin/mailman/listinfo/m5-users >> > _______________________________________________ > m5-users mailing list > [email protected] > http://m5sim.org/cgi-bin/mailman/listinfo/m5-users > _______________________________________________ m5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
