Segmentation fault when running on sqlite testcases

[dp1]

Hi,
I tried running jigsaw on the testcases linked in the README, but I found that it crashes on sqlite.

The following reproduces the crash (on ubuntu 22.04.2):

$ git clone git@github.com:R-Fuzz/jigsaw.git && cd jigsaw
$ mkdir build && cd build && cmake .. && make
$ wget https://jigsaw.cs.ucr.edu/sqlite.txz
$ tar -xvf sqlite.txz
$ ./rgd 8 0 sqlite_reload/
 number of threads 8
 pin_core_start 0
Loading in progress 951
all count is 1220814, drop count is 0 record number is 769548 allexpr is 5650888
processed 10000 constraints
elapsed time is 165788 cons cache lookup 9698 iter 2189226 solved 7741
processed 20000 constraints
elapsed time is 281916 cons cache lookup 18968 iter 4337370 solved 15464
processed 30000 constraints
elapsed time is 384794 cons cache lookup 28227 iter 6064549 solved 23388
processed 40000 constraints
elapsed time is 508458 cons cache lookup 38667 iter 8321184 solved 31064
processed 50000 constraints
elapsed time is 628742 cons cache lookup 49061 iter 10530927 solved 38695
processed 60000 constraints
elapsed time is 742013 cons cache lookup 58231 iter 12581507 solved 46138
processed 70000 constraints
elapsed time is 844249 cons cache lookup 67040 iter 14339973 solved 54202
processed 80001 constraints
elapsed time is 962371 cons cache lookup 76075 iter 16398217 solved 62015
processed 90000 constraints
elapsed time is 1074447 cons cache lookup 85189 iter 18438857 solved 69883
processed 100000 constraints
elapsed time is 1204198 cons cache lookup 97389 iter 20862894 solved 77363
processed 110000 constraints
elapsed time is 1327262 cons cache lookup 107408 iter 23091567 solved 84991
processed 120000 constraints
elapsed time is 1444137 cons cache lookup 116952 iter 25306721 solved 92622
Segmentation fault (core dumped)

Am I making a mistake somewhere, or is there a specific version I should be running? I also tried running in Docker, but it unfortunately fails to build.
If you need any more information please let me know.

[ChengyuSong]

no, it triggers an assertion error. the toDNF process should make sure every expression to be jitted has a comparison as the root of the AST and no additional comparison in the expression. when this assertion is triggered, it means the toNDF process has a bug. will look into it further.