On Thu, Jan 4, 2018 at 11:23 PM, Оля Тележная <olyatelezhnaya@xxxxxxxxx> wrote: > > So for now 2 of my last commits fail, and I am tired of searching for the error. > I was also trying to leave cat_file_info variable and fill in both new > and old variables and then compare resulting values by printing them > into file. Everything is OK, but I find it dudpicious that the > resulting file is too small (fprintf was invoked only 3 times, it was > here: https://github.com/telezhnaya/git/commit/54a5b5a0167ad634c26e4fd7df234a46286ede0a#diff-2846189963e8aec1bcb559b69b7f20d0R1489) > > I have left few comments in github to simplify your understanding what > I was trying to achieve. Feel free to ask any questions if you find > the code strange, unclear or suspicious. Let me try to see how I can debug it. Running `./t1006-cat-file.sh -v -i` gives: --------------- expecting success: maybe_remove_timestamp "$batch_output" $no_ts >expect && maybe_remove_timestamp "$(echo $sha1 | git cat-file --batch)" $no_ts >actual && test_cmp expect actual Segmentation fault (core dumped) --- expect 2018-01-04 23:31:20.515114634 +0000 +++ actual 2018-01-04 23:31:20.635114274 +0000 @@ -1,2 +0,0 @@ -5e1c309dae7f45e0f39b1bf3ac3cd9db12e7d689 blob 11 -Hello World \ No newline at end of file not ok 9 - --batch output of blob is correct # # maybe_remove_timestamp "$batch_output" $no_ts >expect && # maybe_remove_timestamp "$(echo $sha1 | git cat-file --batch)" $no_ts >actual && # test_cmp expect actual # --------------- So there is a segfault probably when running $(echo $sha1 | git cat-file --batch). Let's try to run that manually. $ cd trash\ directory.t1006-cat-file/ $ echo 5e1c309dae7f45e0f39b1bf3ac3cd9db12e7d689 | git cat-file --batch Segmentation fault (core dumped) That's it. Now let's use gdb to see where it comes from: $ echo 5e1c309dae7f45e0f39b1bf3ac3cd9db12e7d689 > myarg.txt $ gdb git GNU gdb (Ubuntu 8.0.1-0ubuntu1) 8.0.1 Copyright (C) 2017 Free Software Foundation, Inc. ... (gdb) Let's run the cat-file command inside gdb: (gdb) run cat-file --batch < myarg.txt Starting program: /home/ubuntu/bin/git cat-file --batch < myarg.txt [Thread debugging using libthread_db enabled] Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1". Program received signal SIGSEGV, Segmentation fault. 0x00005555556e88e6 in populate_value (ref=0x7fffffffd430) at ref-filter.c:1496 1496 ref->disk_size = *obj_info.disk_sizep; (gdb) Let's get a backtrace: (gdb) bt #0 0x00005555556e88e6 in populate_value (ref=0x7fffffffd430) at ref-filter.c:1496 #1 0x00005555555783f1 in batch_object_write ( obj_name=0x555555a655f0 "5e1c309dae7f45e0f39b1bf3ac3cd9db12e7d689", opt=0x7fffffffd6e0, data=0x7fffffffd5e0) at builtin/cat-file.c:291 #2 0x0000555555578660 in batch_one_object ( obj_name=0x555555a655f0 "5e1c309dae7f45e0f39b1bf3ac3cd9db12e7d689", opt=0x7fffffffd6e0, data=0x7fffffffd5e0) at builtin/cat-file.c:346 Let's see what's the code that makes it segfault: (gdb) l 1491 fflush(stdout); 1492 return -1; 1493 } 1494 ref->type = *obj_info.typep; 1495 ref->size = *obj_info.sizep; 1496 ref->disk_size = *obj_info.disk_sizep; 1497 hashcpy(ref->delta_base_oid.hash, obj_info.delta_base_sha1); 1498 } 1499 1500 /* Fill in specials first */ Line 1496 has "ref->disk_size = *obj_info.disk_sizep;" so let's look at those variables: (gdb) p *ref $1 = {objectname = {hash = "^\034\060\235\256\177E\340\363\233\033\363\254<\331\333\022\347\326\211"}, flag = 0, kind = 4148386208, symref = 0x7ffff778b9e0 <_IO_2_1_stdin_> "\210 \255\373", commit = 0x7fffffffd510, values = 0x555555a66cb0, type = OBJ_BLOB, size = 11, disk_size = -7613955248136140544, rest = 0x0, delta_base_oid = { hash = "-\334qUUU\000\000\360\324\377\377\377\177\000\000\340\325\377\377"}, start_of_request = 0x555555a655f0 "5e1c309dae7f45e0f39b1bf3ac3cd9db12e7d689", refname = 0x7fffffffd4a8 ""} (gdb) p obj_info $2 = {typep = 0x555555a53df8 <o_type>, sizep = 0x555555a66c30, disk_sizep = 0x0, delta_base_sha1 = 0x0, typename = 0x0, contentp = 0x0, whence = OI_LOOSE, u = {packed = {pack = 0x0, offset = 0, is_delta = 0}}} Ok we can see that "disk_sizep = 0x0" which means that it segfault because line 1496 tries to read the value pointed to by disk_sizep which is NULL. I hope this will help you. Best, Christian.
