Date: Mon, 27 Mar 2017 19:49:05 +0300 From: Konstantin Belousov <kostikbel@gmail.com> To: Steven Hartland <killing@multiplay.co.uk> Cc: "K. Macy" <kmacy@freebsd.org>, "freebsd-hackers@freebsd.org" <freebsd-hackers@freebsd.org> Subject: Re: Help needed to identify golang fork / memory corruption issue on FreeBSD Message-ID: <20170327164905.GN43712@kib.kiev.ua> In-Reply-To: <3ec35a46-ae70-35cd-29f8-82e7cebb0eb6@multiplay.co.uk> References: <20161206143532.GR54029@kib.kiev.ua> <e160381c-9935-6edf-04a9-1ff78e95d818@multiplay.co.uk> <CAHM0Q_Mg662u9D0KJ9knEWWqi9Ydy38qKDnjLt6XaS0ks%2B9-iw@mail.gmail.com> <18b40a69-4460-faf2-c0ce-7491eca92782@multiplay.co.uk> <20170317082333.GP16105@kib.kiev.ua> <180a601b-5481-bb41-f7fc-67976aabe451@multiplay.co.uk> <20170317124437.GR16105@kib.kiev.ua> <5ba92447-945e-6fea-ad4f-f58ac2a0012e@multiplay.co.uk> <20170327161833.GL43712@kib.kiev.ua> <3ec35a46-ae70-35cd-29f8-82e7cebb0eb6@multiplay.co.uk>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, Mar 27, 2017 at 05:33:49PM +0100, Steven Hartland wrote: > On 27/03/2017 17:18, Konstantin Belousov wrote: > > On Mon, Mar 27, 2017 at 12:47:11PM +0100, Steven Hartland wrote: > >> OK now the similar but unrelated issue with signal stacks is solved I've > >> moved back to the initial issue. > >> > >> I've made some progress with a reproduction case as detailed here: > >> https://github.com/golang/go/issues/15658#issuecomment-288747812 > >> > >> In short it seems that having a running child, while the parent runs GC, > >> is some how responsible for memory corruption in the parent. > >> > >> The reason I believe this is if I run the same GC in the parent after > >> the child exits instead of while its running, I've been unable to > >> reproduce the issue. > >> > >> As the memory segments are COW then the issue might be in VM subsystem. > > Well, it might be, but it is a strange corruption mode to believe. > Indeed, but would you agree the evidence seems to indicate that this may > be the case, as otherwise I would have expected that running the GC > after the child process has exited would have zero impact on the issue. > > > >> In order to confirm / deny this I was wondering if there was a way to > >> force a full copy of all segments for the child instead of using the COW > >> optimisation. > > No, there is no. By design, copying only occurs on faults, when VM > > detects that the map entry needs copying. Doing the actual copy at fork > > time would require writing a lot of new code. > I noticed in vm_map_copy_entry the following: > /* > * We don't want to make writeable wired pages > copy-on-write. > * Immediately copy these pages into the new map by > simulating > * page faults. The new pages are pageable. > */ > vm_fault_copy_entry(dst_map, src_map, dst_entry, src_entry, > fork_charge); > > I wondered if I could use vm_fault_copy_entry to force the copy on fork? No, the vm_fault_copy_entry() only works with wired entries, e.g. it cannot page in not yet touched page, and the result is also wired. > > Does go have FreeBSD/i386 port ? If yes, is the issue reproducable there ? > Yes it does, I don't currently have i386 machine to test with, I'm > assuming testing i386 on amd64 kernel, would likely not have any effect. Only if the bug is in kernel and not in the go runtime. I am still not convinced that the kernel is the culprit. > > Another blind experiment to try is to comment out call to > > vm_object_collapse() in sys/vm/vm_map.c:vm_map_copy_entry() and see if > > it changes anything. > I'll do that shortly. > > What could be quite interesting is to look at the parent and possibly > > child address map after the error occured, using procstat -v. At > > least for parent, this should be relatively easy to set up, just make > > go runtime spin or pause on panic, instead of exiting, and then use > > procstat. > I've been looking at the output from procstat -v I have seen the parent > FLAGS ping ping between C--- and CN--, not sure if that's relevant e.g. C means that the entry is COW, N means that COW was not yet applied after the last fork, i.e. there were no write access. I am interested in the procstat -v output after the failure. I understand that there should be no surprising data for normally executing code. > procstat -v 27099 > PID START END PRT RES PRES REF SHD FLAG > TP PATH > 27099 0x400000 0x70d000 r-x 309 635 3 1 CN-- vn > /root/golang/src/test5/test5 > 27099 0x70d000 0x94e000 r-- 270 635 3 1 CN-- vn > /root/golang/src/test5/test5 > 27099 0x94e000 0x985000 rw- 55 0 1 0 C--- vn > /root/golang/src/test5/test5 > 27099 0x985000 0x9a8000 rw- 18 18 1 0 C--- df > 27099 0x80094e000 0x800b4e000 rw- 38 38 1 0 C--- df > 27099 0x800b4e000 0x800c1e000 rw- 28 28 1 0 C--- df > 27099 0x800c1e000 0x800c6e000 rw- 18 18 1 0 C--- df > 27099 0x800c6e000 0x800cae000 rw- 2 2 1 0 C--- df > 27099 0x800cae000 0x800cee000 rw- 2 2 1 0 C--- df > 27099 0x800cee000 0x800dae000 rw- 5 5 1 0 C--- df > 27099 0x800dae000 0x800dee000 rw- 1 1 1 0 C--- df > 27099 0x800dee000 0x800e2e000 rw- 1 1 1 0 C--- df > 27099 0x800e2e000 0x800e6e000 rw- 1 1 1 0 C--- df > 27099 0x800e6e000 0x800eae000 rw- 1 1 1 0 C--- df > 27099 0xc000000000 0xc000001000 rw- 1 1 1 0 CN-- df > 27099 0xc41fff0000 0xc41fff8000 rw- 3 3 1 0 CN-- df > 27099 0xc41fff8000 0xc420200000 rw- 255 255 1 0 C--- df > 27099 0x7ffffffdf000 0x7ffffffff000 rwx 2 2 1 0 C--D df > 27099 0x7ffffffff000 0x800000000000 r-x 1 1 37 0 ---- ph > > procstat -v 27099 > PID START END PRT RES PRES REF SHD FLAG > TP PATH > 27099 0x400000 0x70d000 r-x 309 635 5 1 CN-- vn > /root/golang/src/test5/test5 > 27099 0x70d000 0x94e000 r-- 270 635 5 1 CN-- vn > /root/golang/src/test5/test5 > 27099 0x94e000 0x985000 rw- 55 0 1 0 C--- vn > /root/golang/src/test5/test5 > 27099 0x985000 0x9a8000 rw- 18 0 1 0 C--- df > 27099 0x80094e000 0x800b4e000 rw- 38 38 2 0 CN-- df > 27099 0x800b4e000 0x800c1e000 rw- 28 28 2 0 CN-- df > 27099 0x800c1e000 0x800c6e000 rw- 18 18 2 0 CN-- df > 27099 0x800c6e000 0x800cae000 rw- 2 2 2 0 CN-- df > 27099 0x800cae000 0x800cee000 rw- 2 2 2 0 CN-- df > 27099 0x800cee000 0x800dae000 rw- 5 5 2 0 CN-- df > 27099 0x800dae000 0x800dee000 rw- 1 1 2 0 CN-- df > 27099 0x800dee000 0x800e2e000 rw- 1 1 2 0 CN-- df > 27099 0x800e2e000 0x800e6e000 rw- 1 1 2 0 CN-- df > 27099 0x800e6e000 0x800eae000 rw- 1 1 2 0 CN-- df > 27099 0xc000000000 0xc000001000 rw- 1 1 2 0 CN-- df > 27099 0xc41fff0000 0xc41fff8000 rw- 3 3 2 0 CN-- df > 27099 0xc41fff8000 0xc420200000 rw- 255 255 1 0 C--- df > 27099 0x7ffffffdf000 0x7ffffffff000 rwx 2 2 1 0 CN-D df > 27099 0x7ffffffff000 0x800000000000 r-x 1 1 38 0 ---- ph > > procstat -v 27099 > PID START END PRT RES PRES REF SHD FLAG > TP PATH > 27099 0x400000 0x70d000 r-x 309 635 5 1 CN-- vn > /root/golang/src/test5/test5 > 27099 0x70d000 0x94e000 r-- 270 635 5 1 CN-- vn > /root/golang/src/test5/test5 > 27099 0x94e000 0x985000 rw- 55 0 1 0 C--- vn > /root/golang/src/test5/test5 > 27099 0x985000 0x9a8000 rw- 18 0 1 0 C--- df > 27099 0x80094e000 0x800b4e000 rw- 38 0 1 0 C--- df > 27099 0x800b4e000 0x800c1e000 rw- 28 28 2 0 CN-- df > 27099 0x800c1e000 0x800c6e000 rw- 18 0 1 0 C--- df > 27099 0x800c6e000 0x800cae000 rw- 2 2 2 0 CN-- df > 27099 0x800cae000 0x800cee000 rw- 2 2 2 0 CN-- df > 27099 0x800cee000 0x800dae000 rw- 5 5 2 0 CN-- df > 27099 0x800dae000 0x800dee000 rw- 1 1 2 0 CN-- df > 27099 0x800dee000 0x800e2e000 rw- 1 0 1 0 C--- df > 27099 0x800e2e000 0x800e6e000 rw- 1 1 2 0 CN-- df > 27099 0x800e6e000 0x800eae000 rw- 1 1 2 0 CN-- df > 27099 0xc000000000 0xc000001000 rw- 1 1 2 0 CN-- df > 27099 0xc41fff0000 0xc41fff8000 rw- 3 3 2 0 CN-- df > 27099 0xc41fff8000 0xc420200000 rw- 255 0 1 0 C--- df > 27099 0x7ffffffdf000 0x7ffffffff000 rwx 2 2 1 0 C--D df > 27099 0x7ffffffff000 0x800000000000 r-x 1 1 38 0 ---- ph > > I'll definitely try capturing the output on fault, see what that looks like > > > >> Is this something that would be relatively easy to hack into the kernel, > >> and if so pointers would be appreciated. > > BTW, I looked some more at the go code, and I noted that > > runtime<stupid UTF-8 char>mmap() implementation looks very strange. > > It ignores %rflags.C bit to identify error, and instead callers > > of mmap() compare the return value with 4096, assuming Linux-style > > error reporting. This would certainly break if mmap(2) syscall > > returns ERESTART one day. > I'll look at this too, thanks for the heads up. > > Regards > Steve
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20170327164905.GN43712>