terminate called after throwing an instance of 'std::out_of_range'

[Pocketkid2]

When running nextpnr-xilinx targeting a basys3 board like so:
nextpnr-xilinx --chipdb ../xc7a35t.bin --xdc ../basys3_nextpnr.xdc --json pong.json --write pong.routed.json --fasm pong.fasm --log nextpnr.log

I get this error:
terminate called after throwing an instance of 'std::out_of_range' what(): vector::_M_range_check: __n (which is 2) >= this->size() (which is 0)

This seems like the same thing that is seen in this issue, but I'm not sure how to locate a stack trace and figure out what's going on.

My constraints file (basys3_nextpnr.xdc) is here

[marzoul]

I have the same issue.
Using Archlinux, a debug package is created when compiling from sources. So I installed it and got the stacktrace :

Info: Packing DSPs..
Info:     Created 35 DSP48E1_DSP48E1 cells from:
Info:             35x DSP48E1
terminate called after throwing an instance of 'std::out_of_range'
  what():  vector::_M_range_check: __n (which is 1) >= this->size() (which is 1)

Program received signal SIGABRT, Aborted.
0x00007ffff79b63f4 in ?? () from /usr/lib/libc.so.6
(gdb) bt
#0  0x00007ffff79b63f4 in ?? () from /usr/lib/libc.so.6
#1  0x00007ffff795d120 in raise () from /usr/lib/libc.so.6
#2  0x00007ffff79444c3 in abort () from /usr/lib/libc.so.6
#3  0x00007ffff7c97b2c in __gnu_cxx::__verbose_terminate_handler () at /usr/src/debug/gcc/gcc/libstdc++-v3/libsupc++/vterminate.cc:95
#4  0x00007ffff7cadffa in __cxxabiv1::__terminate (handler=<optimized out>) at /usr/src/debug/gcc/gcc/libstdc++-v3/libsupc++/eh_terminate.cc:48
#5  0x00007ffff7c9752a in std::terminate () at /usr/src/debug/gcc/gcc/libstdc++-v3/libsupc++/eh_terminate.cc:58
#6  0x00007ffff7cae2b6 in __cxxabiv1::__cxa_throw (obj=<optimized out>, tinfo=0x7ffff7e76dc0 <typeinfo for std::out_of_range>, dest=0x7ffff7cc5810 <std::out_of_range::~out_of_range()>) at /usr/src/debug/gcc/gcc/libstdc++-v3/libsupc++/eh_throw.cc:98
#7  0x00007ffff7c9b423 in std::__throw_out_of_range_fmt (__fmt=__fmt@entry=0x5555556f9180 "vector::_M_range_check: __n (which is %zu) >= this->size() (which is %zu)") at /usr/src/debug/gcc/gcc/libstdc++-v3/src/c++11/functexcept.cc:101
#8  0x000055555557ddf0 in std::vector<nextpnr_xilinx::indexed_store<nextpnr_xilinx::PortRef>::slot, std::allocator<nextpnr_xilinx::indexed_store<nextpnr_xilinx::PortRef>::slot> >::_M_range_check (this=0x55556139d4b0, __n=<optimized out>)
    at /usr/include/c++/14.2.1/bits/stl_vector.h:1160
#9  std::vector<nextpnr_xilinx::indexed_store<nextpnr_xilinx::PortRef>::slot, std::allocator<nextpnr_xilinx::indexed_store<nextpnr_xilinx::PortRef>::slot> >::at (this=0x55556139d4b0, __n=<optimized out>) at /usr/include/c++/14.2.1/bits/stl_vector.h:1182
#10 nextpnr_xilinx::indexed_store<nextpnr_xilinx::PortRef>::at (this=0x55556139d4b0, idx=...) at /usr/src/debug/nextpnr-xilinx-git/nextpnr-xilinx/common/kernel/indexed_store.h:165
#11 nextpnr_xilinx::indexed_store<nextpnr_xilinx::PortRef>::iterator::operator* (this=<optimized out>, this=<optimized out>) at /usr/src/debug/nextpnr-xilinx-git/nextpnr-xilinx/common/kernel/indexed_store.h:202
#12 operator() (__closure=<synthetic pointer>, ni=0x55556139d490, port="PCOUT47") at /usr/src/debug/nextpnr-xilinx-git/nextpnr-xilinx/xilinx/pack_dsp_xc7.cc:34
#13 nextpnr_xilinx::XC7Packer::walk_dsp (this=this@entry=0x7fffffffbb90, root=<optimized out>, current_cell=<optimized out>, constr_z=constr_z@entry=25) at /usr/src/debug/nextpnr-xilinx-git/nextpnr-xilinx/xilinx/pack_dsp_xc7.cc:47
#14 0x00005555556a0502 in nextpnr_xilinx::XC7Packer::pack_dsps (this=0x7fffffffbb90) at /usr/src/debug/nextpnr-xilinx-git/nextpnr-xilinx/xilinx/pack_dsp_xc7.cc:158
#15 0x0000555555691e32 in nextpnr_xilinx::Arch::pack (this=0x55555576ba30) at /usr/src/debug/nextpnr-xilinx-git/nextpnr-xilinx/xilinx/pack.cc:904
#16 0x00005555556d402b in nextpnr_xilinx::CommandHandler::executeMain(std::unique_ptr<nextpnr_xilinx::Context, std::default_delete<nextpnr_xilinx::Context> >) [clone .constprop.0] (this=this@entry=0x7fffffffc5c0, ctx=std::unique_ptr<nextpnr_xilinx::Context> = {...})
    at /usr/src/debug/nextpnr-xilinx-git/nextpnr-xilinx/common/kernel/command.cc:452
#17 0x000055555558940f in nextpnr_xilinx::CommandHandler::exec (this=0x7fffffffc5c0) at /usr/src/debug/nextpnr-xilinx-git/nextpnr-xilinx/common/kernel/command.cc:549
#18 main (argc=<optimized out>, argv=<optimized out>) at /usr/src/debug/nextpnr-xilinx-git/nextpnr-xilinx/xilinx/main.cc:91
(gdb)

The design is non-trivial, around 10k LUTs + BRAMs + DSPs.
In case the above stacktrace is not enough to find the issue I'll provide it as an archive somewhere.

[marzoul]

The code is weird.
I looked at involved source files pack_dsp_xc7.cc and indexed_store.h.
I don't think that line (pack_dsp_xc7.cc:34) is supposed to work in the first place (and/or extremely misleading coding style) :
PortRef& user = *ni->users.end();

[marzoul]

Experiment : I added a rbegin() method to class indexed_store and replaced calls from pack_dsp_xc7.cc accordingly. The crash seems gone (but no guarantee I've done the right thing).

However I get a crash later in mapping LUTRAM-related cells :
ERROR: Unable to place cell 'user_inst.i_norm3.gen_par_n1_gen_mem_ram.0.9.genblk1.genblk1[1].genblk1.slice', no BELs remaining to implement cell type 'RAM128X1S'
(unrelated issue, just to say that the proposed fix would not make every design pass)

[marzoul]

To track progress/discussions, I opened the pull request to fix this for nextpnr-himbaechel:
YosysHQ/nextpnr#1361