Re: Need help debugging hedgewars

2017-12-17 Thread Richard Shaw
On Sun, Dec 17, 2017 at 4:31 PM, John Reiser wrote: > Then there's the subtle point that filing bugzilla encourages > the exhibit of a reproducible test case: the actual command line, > the actual .rpms for the program and libraries, > the actual environment variables,

Re: Need help debugging hedgewars

2017-12-17 Thread John Reiser
On 12/17/2017 13:57 UTC, Tony Nelson wrote: On 17-12-17 14:33:43, John Reiser wrote:  ... At this point I suggest to file a bugzilla report against the compiler for not keeping %rsp 16-byte aligned.  Include the traceback and the register info from the gdb session at SIGSEGV, and the

Re: Need help debugging hedgewars

2017-12-17 Thread Tony Nelson
On 17-12-17 14:33:43, John Reiser wrote: ... At this point I suggest to file a bugzilla report against the compiler for not keeping %rsp 16-byte aligned. Include the traceback and the register info from the gdb session at SIGSEGV, and the identities (.so name, build-id, .rpm name) of the

Re: Need help debugging hedgewars

2017-12-17 Thread Richard Shaw
On Sun, Dec 17, 2017 at 1:33 PM, John Reiser wrote: > > Anyway, the focus now shifts to the backtrace from > https://pastebin.com/LCDQpB5d : > #0 __GI__dl_catch_error > #1 0x763b76a5 in _dlerror_run at dlerror.c:163 > #2 0x763b701f in __dlclose at

Re: Need help debugging hedgewars

2017-12-17 Thread John Reiser
At this point ... You could also plant a breakpoint at the <+0> of each routine named in the traceback, and print %rsp. It should be (8 mod 16) because the dynamically-preceding 'call' should have pushed 8 bytes when the %rsp was 16-byte aligned. (gdb) b * # use a numerical address, NOT a

Re: Need help debugging hedgewars

2017-12-17 Thread John Reiser
Dump of assembler code for function __GI__dl_catch_error:    0x76136100 <+0>:push   %rbx    0x76136101 <+1>:sub    $0x140,%rsp    0x76136108 <+8>:lea    0x4c(%rsp),%rax    0x7613610d <+13>:mov    %rdi,0x10(%rsp)    0x76136112 <+18>:mov   

Re: Need help debugging hedgewars

2017-12-17 Thread Richard Shaw
Dump of assembler code for function __GI__dl_catch_error: 0x76136100 <+0>: push %rbx 0x76136101 <+1>: sub$0x140,%rsp 0x76136108 <+8>: lea0x4c(%rsp),%rax 0x7613610d <+13>: mov%rdi,0x10(%rsp) 0x76136112 <+18>: mov%rsi,0x8(%rsp)

Re: Need help debugging hedgewars

2017-12-17 Thread John Reiser
https://pastebin.com/r03HxDrX rsp0x7fffd6e8 0x7fffd6e8 BAD: not 16-byte aligned pc=> 0x7613616c <__GI__dl_catch_error+108>:movaps %xmm0,0x50(%rsp) The problem is that the effective address 0x...738 is not 16-byte aligned, and this generates SIGSEGV

Re: Need help debugging hedgewars

2017-12-17 Thread Richard Shaw
On Sun, Dec 17, 2017 at 8:04 AM, John Reiser wrote: > > SIGSEGV is a bad memory reference, but none of those URLs gives any info > about the memory reference itself: the address, the size, > Read/Write/Execute, > the program counter value, the contents of the instruction

Re: Need help debugging hedgewars

2017-12-17 Thread John Reiser
On 12/17/2017 12:58 UTC, Richard Shaw wrote: Ok, so I built the latest version of hedgewars (0.9.23) and submitted an update and quickly found out that when the game exits (hwengine, not the hedgewars binary) it get's a segmentation fault. http://hedgewars.org/node/6879?page=0

Re: Need help debugging hedgewars

2017-12-17 Thread Artur Iwicki
Hello, Richard. I've happened to stumble across a very similar issue recently. I've been working on an update to Colorful - which, like Hedgewars, is a game written in Pascal and compiled with FPC - and the game gives me an Access Violation upon exiting. This happens after all of "my" code has

Need help debugging hedgewars

2017-12-17 Thread Richard Shaw
Ok, so I built the latest version of hedgewars (0.9.23) and submitted an update and quickly found out that when the game exits (hwengine, not the hedgewars binary) it get's a segmentation fault. https://bodhi.fedoraproject.org/updates/FEDORA-2017-f9a419a9e9 I have posted to the hedgewars support