![]() |
Bug in 64-bit lasieve
When I used the 64-bit lasieve compiled on my Core2 Duo E8200 PC, it crashed due to segmentation fault.
The debug message is given below: [code]math@linux-0rdr:~/Desktop/application/nfs/ggnfs/snfs> gdb /home/math/Desktop/other/test/ggnfs/src/experimental/lasieve4_64/gnfs-lasieve4I11e GNU gdb (GDB) SUSE (6.8.91.20090930-2.4) Copyright (C) 2009 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html> This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. Type "show copying" and "show warranty" for details. This GDB was configured as "x86_64-suse-linux". For bug reporting instructions, please see: <http://www.gnu.org/software/gdb/bugs/>... Reading symbols from /home/math/Desktop/other/test/ggnfs/src/experimental/lasieve4_64/gnfs-lasieve4I11e...done. (gdb) run -R -k -o spairs.out.T1 -v -n0 -r snfs.job.T1 Starting program: /home/math/Desktop/other/test/ggnfs/src/experimental/lasieve4_64/gnfs-lasieve4I11e -R -k -o spairs.out.T1 -v -n0 -r snfs.job.T1 Resuming with -f 0 -c 1 FBsize 29823+0 (deg 4), 13847+0 (deg 1) SCHED_PATHOLOGY q0=150001 k=5 excess=100 Detaching after fork from child process 23965. SCHED_PATHOLOGY q0=150011 k=4 excess=34 Detaching after fork from child process 23966. SCHED_PATHOLOGY q0=150041 k=1 excess=234 Detaching after fork from child process 23967. SCHED_PATHOLOGY q0=150053 k=4 excess=98 Detaching after fork from child process 23968. SCHED_PATHOLOGY q0=150061 k=6 excess=182 Detaching after fork from child process 23969. SCHED_PATHOLOGY q0=150067 k=4 excess=22 Detaching after fork from child process 23970. SCHED_PATHOLOGY q0=150077 k=1 excess=318 Detaching after fork from child process 23971. SCHED_PATHOLOGY q0=150083 k=2 excess=32 Detaching after fork from child process 23972. SCHED_PATHOLOGY q0=150089 k=3 excess=148 Detaching after fork from child process 23973. SCHED_PATHOLOGY q0=150091 k=2 excess=148 Detaching after fork from child process 23974. SCHED_PATHOLOGY q0=150097 k=1 excess=12 Detaching after fork from child process 23975. SCHED_PATHOLOGY q0=150107 k=4 excess=132 Detaching after fork from child process 23976. Program received signal SIGSEGV, Segmentation fault. 0x00000000004086cf in main (argc=<value optimized out>, argv=<value optimized out>) at gnfs-lasieve4e.c:2403 2403 schedule_ptr= schedules[s][j].schedule[l][k]+SCHED_SI_OFFS; (gdb) bt full #0 0x00000000004086cf in main (argc=<value optimized out>, argv=<value optimized out>) at gnfs-lasieve4e.c:2403 x = 37 '%' schedule_ptr = <value optimized out> l = 0 j = 0 new_clock = <value optimized out> clock_diff = <value optimized out> s = <value optimized out> subsieve_nr = 0 nr = <value optimized out> r = 0xb147f0 zip_output = <value optimized out> force_aFBcalc = 1849 catch_signals = 1 all_spq_done = 1 n_spq = 0 n_spq_discard = 0 tNow = <value optimized out> lastReport = 0.040000000000000001 (gdb) [/code] And there is the content of the snfs.poly file: [code]n: 270270270270270270270270270270270270270270270270270270270270270270270270270270270270270270270270271 # 86^64-1, difficulty: 125.44, skewness: 1.22, alpha: 0.00 # cost: 8.42734e+13, est. time: 0.04 GHz days (not accurate yet!) skew: 1 c4: 1 c0: 27 Y1: -1 Y0: 10000000000000000000000000 m: 10000000000000000000000000 type: snfs [/code] My lasieve is compiled with "-g -O3 -march=nocona -mtune=nocona -funroll-loops" flags, the version of gcc is 4.4.1, the version of gmp is 4.3.1, and my OS is OpenSUSE 11.2. |
Did you read all instructions in the INSTALL file?
_____ Here's what it says: These should build and run on Athlon64, or Core2 cpus. ****************************** NOTE: on Core2 replace in athlon64/ls-defs.asm - define(l1_bits,16)dnl + define(l1_bits,15)dnl ****************************** |
[QUOTE=Batalov;196255]Did you read all instructions in the INSTALL file?
_____ Here's what it says: These should build and run on Athlon64, or Core2 cpus. ****************************** NOTE: on Core2 replace in athlon64/ls-defs.asm - define(l1_bits,16)dnl + define(l1_bits,15)dnl ******************************[/QUOTE] I followed the illustration of the INSTALL file, recompiled it and reran it, it didn`t crash. |
Very good. I repeated it here, too, on a Core2 and on k8.
|
| All times are UTC. The time now is 20:32. |
Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.