Re: ports -r484565 : qemu-arm-static fails with: (start < end): backtrace included; start+len arithmetic overflow (abi_ulong wrap) for TARGET_FREEBSD_NR_mmap use
On 2018-Nov-11, at 17:50, Mark Millard wrote: > On 2018-Nov-11, at 17:43, Kyle Evans wrote: > >> On Sun, Nov 11, 2018 at 5:24 AM Mark Millard wrote: >>> >>> I attached with gdb in order to stop at the assert and look around. >>> >>> >>> >>> The following is a backtrace with notes and prints mixed in: >>> >>> (gdb) bt >>> #0 thr_kill () at thr_kill.S:3 >>> #1 0x6028a21f in __raise (s=6) at /usr/src/lib/libc/gen/raise.c:52 >>> #2 0x60204949 in abort () at /usr/src/lib/libc/stdlib/abort.c:67 >>> #3 0x6027855a in __assert (func=, file=>> out>, line=, failedexpr=) at >>> /usr/src/lib/libc/gen/assert.c:51 >>> >>> Note end==37146624 below vs. start (37146624 will show up again in later >>> notes) >>> >>> #4 0x60036243 in page_set_flags (start=4143968256, end=37146624, >>> flags=9) at >>> /wrkdirs/usr/ports/emulators/qemu-user-static/work/qemu-bsd-user-2cb0cdd/accel/tcg/translate-all.c:2077 >>> >>> Note start and len below: >>> >>> #5 0x6003df2b in target_mmap (start=4143968256, len=188145664, >>> prot=, flags=, fd=, >>> offset=) >>> at >>> /wrkdirs/usr/ports/emulators/qemu-user-static/work/qemu-bsd-user-2cb0cdd/bsd-user/mmap.c:626 >>> >>> (gdb) print/x start >>> $5 = 0xf6fff000 >>> (gdb) print/x len >>> $6 = 0xb36e000 >>> >>> Note start+len for the above (without wrapping): >>> >>> (gdb) print/x (long long)start + (long long)len >>> $10 = 0x10236d000 >>> (gdb) print (long long)start + (long long)len >>> $11 = 4332113920 >>> >>> With wrapping: >>> >>> (gdb) print/x start+len >>> $8 = 0x236d000 >>> (gdb) print start+len >>> $9 = 37146624 >>> >>> And there is end's value again. >>> >> >> Hi, >> >> This should be fixed as of ports r484702; please do try this and let >> us know how it goes. > > I've updated ports to -r484783 and an amd64 -> armv7 > poudriere-devel/qemu-user-static cross-build is in > progress. devel/cmake completed fine, overall about > 63 ports have. The 1 port failure is not tied to > qemu-arm-static issues. > > It will be hours before lang/gcc8 would finish. There > are somewhat over 70 ports to go overall. > > So far so good. > lang/gcc8 (full bootstrap) and the other about 70 ports built fine. (There was one example of the lld hang-up, for which I used a gdb attach/detach sequence to cause the emulated lld to continue.) === Mark Millard marklmi at yahoo.com ( dsl-only.net went away in early 2018-Mar) ___ freebsd-ports@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to "freebsd-ports-unsubscr...@freebsd.org"
Re: ports -r484565 : qemu-arm-static fails with: (start < end): backtrace included; start+len arithmetic overflow (abi_ulong wrap) for TARGET_FREEBSD_NR_mmap use
On 2018-Nov-11, at 17:43, Kyle Evans wrote: > On Sun, Nov 11, 2018 at 5:24 AM Mark Millard wrote: >> >> I attached with gdb in order to stop at the assert and look around. >> >> >> >> The following is a backtrace with notes and prints mixed in: >> >> (gdb) bt >> #0 thr_kill () at thr_kill.S:3 >> #1 0x6028a21f in __raise (s=6) at /usr/src/lib/libc/gen/raise.c:52 >> #2 0x60204949 in abort () at /usr/src/lib/libc/stdlib/abort.c:67 >> #3 0x6027855a in __assert (func=, file=> out>, line=, failedexpr=) at >> /usr/src/lib/libc/gen/assert.c:51 >> >> Note end==37146624 below vs. start (37146624 will show up again in later >> notes) >> >> #4 0x60036243 in page_set_flags (start=4143968256, end=37146624, >> flags=9) at >> /wrkdirs/usr/ports/emulators/qemu-user-static/work/qemu-bsd-user-2cb0cdd/accel/tcg/translate-all.c:2077 >> >> Note start and len below: >> >> #5 0x6003df2b in target_mmap (start=4143968256, len=188145664, >> prot=, flags=, fd=, >> offset=) >>at >> /wrkdirs/usr/ports/emulators/qemu-user-static/work/qemu-bsd-user-2cb0cdd/bsd-user/mmap.c:626 >> >> (gdb) print/x start >> $5 = 0xf6fff000 >> (gdb) print/x len >> $6 = 0xb36e000 >> >> Note start+len for the above (without wrapping): >> >> (gdb) print/x (long long)start + (long long)len >> $10 = 0x10236d000 >> (gdb) print (long long)start + (long long)len >> $11 = 4332113920 >> >> With wrapping: >> >> (gdb) print/x start+len >> $8 = 0x236d000 >> (gdb) print start+len >> $9 = 37146624 >> >> And there is end's value again. >> > > Hi, > > This should be fixed as of ports r484702; please do try this and let > us know how it goes. I've updated ports to -r484783 and an amd64 -> armv7 poudriere-devel/qemu-user-static cross-build is in progress. devel/cmake completed fine, overall about 63 ports have. The 1 port failure is not tied to qemu-arm-static issues. It will be hours before lang/gcc8 would finish. There are somewhat over 70 ports to go overall. So far so good. === Mark Millard marklmi at yahoo.com ( dsl-only.net went away in early 2018-Mar) ___ freebsd-ports@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to "freebsd-ports-unsubscr...@freebsd.org"
Re: ports -r484565 : qemu-arm-static fails with: (start < end): backtrace included; start+len arithmetic overflow (abi_ulong wrap) for TARGET_FREEBSD_NR_mmap use
On Sun, Nov 11, 2018 at 5:24 AM Mark Millard wrote: > > I attached with gdb in order to stop at the assert and look around. > > > > The following is a backtrace with notes and prints mixed in: > > (gdb) bt > #0 thr_kill () at thr_kill.S:3 > #1 0x6028a21f in __raise (s=6) at /usr/src/lib/libc/gen/raise.c:52 > #2 0x60204949 in abort () at /usr/src/lib/libc/stdlib/abort.c:67 > #3 0x6027855a in __assert (func=, file= out>, line=, failedexpr=) at > /usr/src/lib/libc/gen/assert.c:51 > > Note end==37146624 below vs. start (37146624 will show up again in later > notes) > > #4 0x60036243 in page_set_flags (start=4143968256, end=37146624, > flags=9) at > /wrkdirs/usr/ports/emulators/qemu-user-static/work/qemu-bsd-user-2cb0cdd/accel/tcg/translate-all.c:2077 > > Note start and len below: > > #5 0x6003df2b in target_mmap (start=4143968256, len=188145664, > prot=, flags=, fd=, > offset=) > at > /wrkdirs/usr/ports/emulators/qemu-user-static/work/qemu-bsd-user-2cb0cdd/bsd-user/mmap.c:626 > > (gdb) print/x start > $5 = 0xf6fff000 > (gdb) print/x len > $6 = 0xb36e000 > > Note start+len for the above (without wrapping): > > (gdb) print/x (long long)start + (long long)len > $10 = 0x10236d000 > (gdb) print (long long)start + (long long)len > $11 = 4332113920 > > With wrapping: > > (gdb) print/x start+len > $8 = 0x236d000 > (gdb) print start+len > $9 = 37146624 > > And there is end's value again. > Hi, This should be fixed as of ports r484702; please do try this and let us know how it goes. Thanks, Kyle Evans ___ freebsd-ports@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to "freebsd-ports-unsubscr...@freebsd.org"
Re: ports -r484565 : qemu-arm-static fails with: (start < end): backtrace included; start+len arithmetic overflow (abi_ulong wrap) for TARGET_FREEBSD_NR_mmap use
On Sun, Nov 11, 2018 at 5:24 AM Mark Millard wrote: > > I attached with gdb in order to stop at the assert and look around. > > > > The following is a backtrace with notes and prints mixed in: > > (gdb) bt > #0 thr_kill () at thr_kill.S:3 > #1 0x6028a21f in __raise (s=6) at /usr/src/lib/libc/gen/raise.c:52 > #2 0x60204949 in abort () at /usr/src/lib/libc/stdlib/abort.c:67 > #3 0x6027855a in __assert (func=, file= out>, line=, failedexpr=) at > /usr/src/lib/libc/gen/assert.c:51 > > Note end==37146624 below vs. start (37146624 will show up again in later > notes) > > #4 0x60036243 in page_set_flags (start=4143968256, end=37146624, > flags=9) at > /wrkdirs/usr/ports/emulators/qemu-user-static/work/qemu-bsd-user-2cb0cdd/accel/tcg/translate-all.c:2077 > > Note start and len below: > > #5 0x6003df2b in target_mmap (start=4143968256, len=188145664, > prot=, flags=, fd=, > offset=) > at > /wrkdirs/usr/ports/emulators/qemu-user-static/work/qemu-bsd-user-2cb0cdd/bsd-user/mmap.c:626 > > (gdb) print/x start > $5 = 0xf6fff000 > (gdb) print/x len > $6 = 0xb36e000 > Ahhh, I see where I messed up. I will whip up a patch ASAP to fix. Thanks! ___ freebsd-ports@freebsd.org mailing list https://lists.freebsd.org/mailman/listinfo/freebsd-ports To unsubscribe, send any mail to "freebsd-ports-unsubscr...@freebsd.org"
Re: ports -r484565 : qemu-arm-static fails with: (start < end): backtrace included; start+len arithmetic overflow (abi_ulong wrap) for TARGET_FREEBSD_NR_mmap use
I attached with gdb in order to stop at the assert and look around.
The following is a backtrace with notes and prints mixed in:
(gdb) bt
#0 thr_kill () at thr_kill.S:3
#1 0x6028a21f in __raise (s=6) at /usr/src/lib/libc/gen/raise.c:52
#2 0x60204949 in abort () at /usr/src/lib/libc/stdlib/abort.c:67
#3 0x6027855a in __assert (func=, file=,
line=, failedexpr=) at
/usr/src/lib/libc/gen/assert.c:51
Note end==37146624 below vs. start (37146624 will show up again in later notes)
#4 0x60036243 in page_set_flags (start=4143968256, end=37146624,
flags=9) at
/wrkdirs/usr/ports/emulators/qemu-user-static/work/qemu-bsd-user-2cb0cdd/accel/tcg/translate-all.c:2077
Note start and len below:
#5 0x6003df2b in target_mmap (start=4143968256, len=188145664,
prot=, flags=, fd=,
offset=)
at
/wrkdirs/usr/ports/emulators/qemu-user-static/work/qemu-bsd-user-2cb0cdd/bsd-user/mmap.c:626
(gdb) print/x start
$5 = 0xf6fff000
(gdb) print/x len
$6 = 0xb36e000
Note start+len for the above (without wrapping):
(gdb) print/x (long long)start + (long long)len
$10 = 0x10236d000
(gdb) print (long long)start + (long long)len
$11 = 4332113920
With wrapping:
(gdb) print/x start+len
$8 = 0x236d000
(gdb) print start+len
$9 = 37146624
And there is end's value again.
The code doing the wrapping is (with more context):
621 if (p == MAP_FAILED)
622 goto fail;
623 }
624 }
625 the_end1:
626 page_set_flags(start, start + len, prot | PAGE_VALID);
627 the_end:
628 #ifdef DEBUG_MMAP
629 printf("ret=0x" TARGET_ABI_FMT_lx "\n", start);
630 page_dump(stdout);
#6 0x6004219c in do_bsd_mmap (arg1=, arg2=, arg3=, arg4=2, arg5=, arg6=,
arg7=, arg8=0,
cpu_env=) at
/wrkdirs/usr/ports/emulators/qemu-user-static/work/qemu-bsd-user-2cb0cdd/bsd-user/bsd-mem.h:75
The code for the above is:
if (regpairs_aligned(cpu_env) != 0) {
arg6 = arg7;
arg7 = arg8;
}
return get_errno(target_mmap(arg1, arg2, arg3,
target_to_host_bitmask(arg4, mmap_flags_tbl), arg5,
target_arg64(arg6, arg7)));
#7 do_freebsd_syscall (cpu_env=0x860c08318, num=477, arg1=,
arg2=, arg3=, arg4=2, arg5=9, arg6=0, arg7=0,
arg8=0)
at
/wrkdirs/usr/ports/emulators/qemu-user-static/work/qemu-bsd-user-2cb0cdd/bsd-user/syscall.c:946
The code above is (with some context):
break;
/*
* Memory management system calls.
*/
case TARGET_FREEBSD_NR_mmap: /* mmap(2) */
ret = do_bsd_mmap(cpu_env, arg1, arg2, arg3, arg4, arg5, arg6, arg7,
arg8);
break;
#8 0x60038be3 in target_cpu_loop (env=0x860c08318) at
/wrkdirs/usr/ports/emulators/qemu-user-static/work/qemu-bsd-user-2cb0cdd/bsd-user/arm/target_arch_cpu.h:207
The code and its context for the above is:
break;
case EXCP_SWI:
case EXCP_BKPT:
. . .
/*
* system call
* See arm/arm/trap.c cpu_fetch_syscall_args()
*/
. . .
DEBUG_PRINTF("AVANT CALL %d\n", n);
if (bsd_type == target_freebsd) {
int ret;
abi_ulong params = get_sp_from_cpustate(env);
int32_t syscall_nr = n;
int32_t arg1, arg2, arg3, arg4, arg5, arg6, arg7, arg8;
if (syscall_nr == TARGET_FREEBSD_NR_syscall) {
. . .
} else if (syscall_nr == TARGET_FREEBSD_NR___syscall) {
. . .
} else {
arg1 = env->regs[0];
arg2 = env->regs[1];
arg3 = env->regs[2];
arg4 = env->regs[3];
get_user_s32(arg5, params);
params += sizeof(int32_t);
get_user_s32(arg6, params);
params += sizeof(int32_t);
get_user_s32(arg7, params);
params += sizeof(int32_t);
get_user_s32(arg8, params);
}
ret = do_freebsd_syscall(env, syscall_nr, arg1, arg2, arg3,
arg4, arg5, arg6, arg7, arg8);
#9 0x60038589 in cpu_loop (env=0x18b2f) at
/wrkdirs/usr/ports/emulators/qemu-user-static/work/qemu-bsd-user-2cb0cdd/bsd-user/main.c:121
#10 0x60039802 in main (argc=-10089, argv=0x7fffd4e0) at
/wrkdirs/usr/ports/emulators/qemu-user-static/work/qemu-bsd-user-2cb0cdd/bsd-user/main.c:513
===
Mark Millard
marklmi at yahoo.com
( dsl-only.net went
away in early 2018-Mar)
___
freebsd-ports@freebsd.org mailing list
https://lists.freebsd.org/mailman/listinfo/freebsd-ports
To unsubscribe, send any mail to "freebsd-ports-unsubscr...@freebsd.org"
