I think I figured out the problem, given Henry's hints and what I'm seeing.
No need for anyone else to spend time on this. Thanks, -- Raul On Fri, Nov 24, 2023 at 10:12 PM Bill Heagy <[email protected]> wrote: > > Probably not helpful. This produced the right answer, but complained a > lot. I just recompiled with gcc, which produced the wrong answer, but > didn't complain. > > Bill H. > > On 11/24/23 22:01, Bill Heagy wrote: > > Does this help: > > > > I've compiled with sanitize: > > CC='clang -g -fsanitize=address -fno-omit-frame-pointer > > -fsanitize-recover=address' jplatform=linux j64x=j64avx2 make2/build_all.sh > > > > > > $ USE_OPENMPI=2 ASAN_OPTIONS=halt_on_error=0 ./jlibrary/bin/jconsole > > [various complaints starting up] > > ....... > > 0j30 ": (*%) 11 c. 665142606648569600281099799288x > > ================================================================= > > ==136587==ERROR: AddressSanitizer: global-buffer-overflow on address > > 0x7eff18c54180 at pc 0x7eff18611c52 bp 0x7ffc21b2e3f0 sp 0x7ffc21b2e3e8 > > READ of size 8 at 0x7eff18c54180 thread T0 > > #0 0x7eff18611c51 in mvc > > /home/wheagy/git/jsource/make2/obj/linux/j64avx2/../../../../jsrc/u.c:345:8 > > #1 0x7eff18571535 in jtfmte > > /home/wheagy/git/jsource/make2/obj/linux/j64avx2/../../../../jsrc/f2.c:224:94 > > #2 0x7eff18571535 in jtfmt1 > > /home/wheagy/git/jsource/make2/obj/linux/j64avx2/../../../../jsrc/f2.c:244:13 > > #3 0x7eff1856d943 in jtth2a > > /home/wheagy/git/jsource/make2/obj/linux/j64avx2/../../../../jsrc/f2.c:287:3 > > #4 0x7eff1856d943 in jtthorn2 > > /home/wheagy/git/jsource/make2/obj/linux/j64avx2/../../../../jsrc/f2.c:390:3 > > #5 0x7eff185c964f in jtparsea > > /home/wheagy/git/jsource/make2/obj/linux/j64avx2/../../../../jsrc/p.c:751:10 > > #6 0x7eff185c817b in jtparse > > /home/wheagy/git/jsource/make2/obj/linux/j64avx2/../../../../jsrc/p.c:290:4 > > #7 0x7eff185d6afb in jtimmex > > /home/wheagy/git/jsource/make2/obj/linux/j64avx2/../../../../jsrc/px.c:54:28 > > #8 0x7eff1858f47c in jtimmexexecct > > /home/wheagy/git/jsource/make2/obj/linux/j64avx2/../../../../jsrc/io.c:386:2 > > #9 0x7eff1858f47c in jdo > > /home/wheagy/git/jsource/make2/obj/linux/j64avx2/../../../../jsrc/io.c:422:111 > > #10 0x7eff1858ee62 in JDo > > /home/wheagy/git/jsource/make2/obj/linux/j64avx2/../../../../jsrc/io.c:527:9 > > #11 0x563c3aa02bff in main > > /home/wheagy/git/jsource/make2/obj/linux/j64avx2/../../../../jsrc/jconsole.c:393:28 > > #12 0x7eff1fe286c9 in __libc_start_call_main > > csu/../sysdeps/nptl/libc_start_call_main.h:58:16 > > #13 0x7eff1fe28784 in __libc_start_main csu/../csu/libc-start.c:360:3 > > #14 0x563c3a92b440 in _start > > (/home/wheagy/git/jsource/jlibrary/bin/jconsole+0x23440) (BuildId: > > 5cef6dfdd9af34a072711de8a792780461088bc1) > > > > 0x7eff18c54180 is located 32 bytes before global variable '.str.13' > > defined in > > '/home/wheagy/git/jsource/make2/obj/linux/j64avx2/../../../../jsrc/f2.c:104' > > (0x7eff18c541a0) of size 2 > > '.str.13' is ascii string '*' > > 0x7eff18c54182 is located 0 bytes after global variable '.str.12' > > defined in > > '/home/wheagy/git/jsource/make2/obj/linux/j64avx2/../../../../jsrc/f2.c:98' > > (0x7eff18c54180) of size 2 > > '.str.12' is ascii string '0' > > SUMMARY: AddressSanitizer: global-buffer-overflow > > /home/wheagy/git/jsource/make2/obj/linux/j64avx2/../../../../jsrc/u.c:345:8 > > in mvc > > Shadow bytes around the buggy address: > > 0x7eff18c53f00: 00 00 00 02 f9 f9 f9 f9 00 00 00 00 00 00 00 00 > > 0x7eff18c53f80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 > > 0x7eff18c54000: 00 05 f9 f9 00 07 f9 f9 00 06 f9 f9 04 f9 f9 f9 > > 0x7eff18c54080: 03 f9 f9 f9 02 f9 f9 f9 04 f9 f9 f9 03 f9 f9 f9 > > 0x7eff18c54100: 05 f9 f9 f9 04 f9 f9 f9 03 f9 f9 f9 02 f9 f9 f9 > > =>0x7eff18c54180:[02]f9 f9 f9 02 f9 f9 f9 00 00 00 f9 f9 f9 f9 f9 > > 0x7eff18c54200: 00 00 06 f9 f9 f9 f9 f9 03 f9 f9 f9 00 00 00 00 > > 0x7eff18c54280: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 > > 0x7eff18c54300: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 > > 0x7eff18c54380: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 > > 0x7eff18c54400: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 > > Shadow byte legend (one shadow byte represents 8 application bytes): > > Addressable: 00 > > Partially addressable: 01 02 03 04 05 06 07 > > Heap left redzone: fa > > Freed heap region: fd > > Stack left redzone: f1 > > Stack mid redzone: f2 > > Stack right redzone: f3 > > Stack after return: f5 > > Stack use after scope: f8 > > Global redzone: f9 > > Global init order: f6 > > Poisoned by user: f7 > > Container overflow: fc > > Array cookie: ac > > Intra object redzone: bb > > ASan internal: fe > > Left alloca redzone: ca > > Right alloca redzone: cb > > 1.000000000000000000000000000000 > > > > > > On 11/24/23 21:41, Raul Miller wrote: > >> So... near as I can tell, the problem occurs inside plusXX which > >> resides between jtva2 and jtxplus. > >> > >> Specifically, an audittstack in jtva2 like this: > >> {if(MEMAUDIT&2)audittstack(jt);} > >> {I lrc=((AHDR2FN*)aadocv->f)(n,m,av,wv,zv,jt); // run one > >> section. Result of 0 means error > >> {if(MEMAUDIT&2)audittstack(jt);} > >> > >> and in jtplusx like this: > >> > >> XF2(jtxplus){ // a+w > >> {if(MEMAUDIT&2)audittstack(jt);} > >> > >> gives me a segfault with a stack trace on that first line of jtxplus, > >> with plusXX between jtva2 and jtxplus, and of course with the above > >> lrc= line on the stack for jtva2 (but only after the 150 seconds of > >> scripting to trigger the problem). > >> > > ---------------------------------------------------------------------- > > For information about J forums see http://www.jsoftware.com/forums.htm > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
