erlang (rabbitmq) generating core on Solaris SPARC

Pooja Desai pooja.desai10@REDACTED
Thu May 14 09:32:04 CEST 2020


Hi Mikael,

Please find flies you requested in attachment as erl_files.tar.gz
(compressed as facing issue with mail size)

Normal build option is:

# gcc  -Werror=undef -Werror=implicit -Werror=return-type  -m64 -g  -O3
-fomit-frame-pointer -Ierlang/src/solaris/otp/erts/sparc-sun-solaris2.10
 -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -fno-tree-copyrename
-DHAVE_CONFIG_H -Wall -Wstrict-prototypes -Wmissing-prototypes
-Wdeclaration-after-statement -DUSE_THREADS -D_THREAD_SAFE -D_REENTRANT
-DPOSIX_THREADS -D_POSIX_PTHREAD_SEMANTICS
-Isparc-sun-solaris2.10/opt/smp -Ibeam -Isys/unix -Isys/common
-Isparc-sun-solaris2.10 -Izlib  -Ipcre -Ihipe -I../include
-I../include/sparc-sun-solaris2.10 -I../include/internal
-I../include/internal/sparc-sun-solaris2.10 -c beam/erl_alloc_util.c -o
obj/sparc-sun-solaris2.10/opt/smp/erl_alloc_util.o

after your suggestion I updated it as below to generate erl_alloc_util file:

# gcc  -Werror=undef -Werror=implicit -Werror=return-type  -m64 -g  -O3
-fomit-frame-pointer -Ierlang/src/solaris/otp/erts/sparc-sun-solaris2.10
 -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -fno-tree-copyrename
-DHAVE_CONFIG_H -Wall -Wstrict-prototypes -Wmissing-prototypes
-Wdeclaration-after-statement -DUSE_THREADS -D_THREAD_SAFE -D_REENTRANT
-DPOSIX_THREADS -D_POSIX_PTHREAD_SEMANTICS
-Isparc-sun-solaris2.10/opt/smp -Ibeam -Isys/unix -Isys/common
-Isparc-sun-solaris2.10 -Izlib  -Ipcre -Ihipe -I../include
-I../include/sparc-sun-solaris2.10 -I../include/internal
-I../include/internal/sparc-sun-solaris2.10 -E beam/erl_alloc_util.c -o
obj/sparc-sun-solaris2.10/opt/smp/erl_alloc_util.i

Also one thing I missed to mention, we are using gcc version 4.9.2 (GCC)
for building on solaris SPARC as erlang doesn't support Sun's native
compiler.

Thanks & Regards,
Pooja

On Tue, May 12, 2020 at 10:44 PM Mikael Pettersson <mikpelinux@REDACTED>
wrote:

> On Tue, May 12, 2020 at 4:18 PM Pooja Desai <pooja.desai10@REDACTED>
> wrote:
> >
> > Hi,
> >
> >
> >
> > Thanks for response Mikael
> >
> > As per your suggestion I am trying to write similar code to conclude if
> there is some issue with Solaris SPARC compiler.
> >
> >
> >
> > But I have some doubts,
> >
> > 1.     If there is problem with compiler then we should be able to see
> this crash everywhere else also, any idea why its only reproduced here?
> >
> > 2.     As I understand your explanation it reads 64 bits by assembling
> two adjacent 32 bits fields. Will it really cause problem in multi-threaded
> program? Considering while context switching to another thread, OS will
> save current context of the thread (and hence registers) and will bring
> back when thread is active again.
> >
> >
>
> Breaking up a 64-bit load into two 32-bit loads loses atomicity with
> any concurrent store into that location, meaning the read may end up
> observing a result composed of 32 bit from the old value and 32 bit
> from the newly stored value, whereas the code expects to see either
> the old or the new, but never this mixture.  This can happen also on a
> single-threaded CPU with preemptive multitasking.
>
> To move forward on the issue, I think you need to recreate the
> pre-processed source for erl_alloc_util.c.  To do that:
> 1. Compile Erlang/OTP as usual, starting from a pristine source
> directory (no left-overs from a previous build, best is to start fresh
> somewhere), but pass "V=1" to make.  Save the output from "make" in a
> file.
> 2. Note the step where it compiles erl_alloc_util.c.
> 3. Reexecute that step, but replace any "-c" with "-E" and "-o
> erl_alloc_util.o" with "-o erl_alloc_util.i".
> 4. Please send this ".i" file, together with the exact build steps and
> configuration options you used, and
> "erts/sparc-sun-solaris11/config.h" (I'm guessing the file name here)
> to me.
>
> My theory is that Erlang/OTP selects the wrong low-level primitives
> for this platform.
>
>
> >
> >
> > Thanks & Regards,
> >
> > Pooja
> >
> >
> > On Mon, May 11, 2020 at 10:36 PM Mikael Pettersson <mikpelinux@REDACTED>
> wrote:
> >>
> >> Hello Pooja,
> >>
> >> On Mon, May 11, 2020 at 8:10 AM Pooja Desai <pooja.desai10@REDACTED>
> wrote:
> >> >
> >> > Hi,
> >> >
> >> > Facing erlang core issue on solaris SPARC setup while running RabbitMQ
> >>
> >> This looks like a 64-bit build, but the code doesn't look similar to
> >> what I get with gcc-9.3, so I'm assuming you used Sun's compiler?
> >>
> >>
> >> > (dbx) where
> >> >
> >> > =>[1] cpool_insert(0x1004efd40, 0xffffffff75600000, 0x61850,
> 0xffffffff75600018, 0x90f, 0x1004effd0), at 0x10006db14
> >> >
> >> >   [2] abandon_carrier(0x1004efd40, 0xffffffff75600000,
> 0xffffffff75645ec0, 0xffffffff77d03818, 0x0, 0x6), at 0x10006de3c
> >> >
> >> >   [3] 17(0x1004efd40, 0xcb3, 0x2, 0xffffffff75645e60, 0x0,
> 0x1004efd40), at 0x10006e958
> >> >
> >> >   [4] erts_alcu_check_delayed_dealloc(0x1004efd40, 0x1,
> 0xffffffff77d03a40, 0xffffffff77d03a48, 0xffffffff77d03a44,
> 0x8000000000000007), at 0x100075244
> >> >
> >> >   [5] erts_alloc_scheduler_handle_delayed_dealloc(0xffffffff3a82a620,
> 0xffffffff77d03a40, 0xffffffff77d03a48, 0xffffffff77d03a44, 0x100464,
> 0xffffffff3a82a5d0),
> >> >
> >> > at 0x1000622c0
> >> >
> >> >   [6] handle_aux_work(0xffffffff3a8204a0, 0x2, 0x1, 0x2, 0x100400,
> 0x4e5ce123), at 0x1002a6044
> >> >
> >> >   [7] erts_schedule(0xffffffff3a820380, 0x9, 0x9, 0xffffffff3a81fc80,
> 0x2, 0x2), at 0x1002a3040
> >> >
> >> >   [8] process_main(0x100469, 0xffffffff3a302240, 0xfa0, 0x802a,
> 0xffffffff38f00438, 0x3), at 0x1002901bc
> >> >
> >> >   [9] sched_thread_func(0xffffffff3a820380, 0x0, 0x0,
> 0xffffffff7a911240, 0x100000, 0x1), at 0x100038f08
> >> >
> >> >   [10] thr_wrapper(0xffffffff7fffc278, 0x0, 0x0, 0x100289d48,
> 0xffffffff3a820380, 0x100038da0), at 0x100289dc8
> >> >
> >> >
> >> >
> >> > This issue is extremely intermittent so I am not able to reproduce it
> with debug build. But on our test setup I have seen this core twice only
> for solaris Sparc server for other servers (RHEL, Suse linux, Solarisx86,
> Windows etc.) with similar test environment things are working fine.
> >> >
> >> > In two instances when I faced this issue we are restarting Rabbitmq
> server. i.e. stop RabbitMQ and epmd then run startup script for rabbitmq.
> This performs 2 operations,
> >> >
> >> > First ping rabbitmq using "rabbitmqctl ping" to confirm rabbitmq is
> not already running ( I guess in background this will also start epmd) and
> then start rabbitmq-server in detached mode.
> >> >
> >> > Core is generated while starting this demon.
> >> >
> >> >
> >> > I checked code around abandon_carrier("
> https://github.com/erlang/otp/blame/master/erts/emulator/beam/erl_alloc_util.c")
> but nothing changed in that area recently. So I am really clueless
> situation.
> >> >
> >> > Please le me know if anyone faced similar issue in past or have any
> idea around this. Using OTP version 22.2 and RabbitMQ version 3.7.23.
> >> >
> >> > Let me know any further information is required, pasting full core
> dump information below:
> >> >
> >> > debugging core file of beam.smp (64-bit) from hostname01
> >> > file: temp_dir/erlang/erts-10.6/bin/beam.smp
> >> > initial argv:
> >> > /temp_dir/erlang/erts-10.6/bin/beam.smp -- -root /temp_dir/
> >> > threading model: native threads
> >> > status: process terminated by SIGSEGV (Segmentation Fault), addr=
> >> > ffffffff004631b0
> >>
> >> Ok, this tells us the address was unmapped.  (It's not an alignment
> >> fault, another common issue on SPARC.)
> >>
> >>
> >> >
> >> > C++ symbol demangling enabled
> >> >
> >> > # stack
> >> >
> >> > cpool_insert+0xd0(10051c500, ffffffff7a400000, ffffffff7a441de8,
> ffffffff7c903818, 0, 23)
> >> > dealloc_block.part.17+0x1c0(10051c500, cb3, 2, ffffffff7a441d88, 0,
> 10051c500)
> >> > erts_alcu_check_delayed_dealloc+0xe4(10051c500, 1, ffffffff7c903a40,
> ffffffff7c903a48, ffffffff7c903a44, 8000000000000007)
> >> > erts_alloc_scheduler_handle_delayed_dealloc+0x34(ffffffff3b729c20,
> ffffffff7c903a40, ffffffff7c903a48, ffffffff7c903a44, 100464,
> ffffffff3b729bd0)
> >> > handle_aux_work+0xa50(ffffffff3b71faa0, 402, 1, 402, 100400, 42da0c68)
> >> > erts_schedule+0x192c(ffffffff3b71f980, 9, 9, ffffffff3b71f280, 402, 2)
> >> > process_main+0xc4(100469, ffffffff3b202240, fa0, ffffffff3b71f980,
> 241, 100294204)
> >> > sched_thread_func+0x168(ffffffff3b71f980, 0, 0, ffffffff39401a40,
> 100000, 1)
> >> > thr_wrapper+0x80(ffffffff7fffb318, 0, 0, 100289d48, ffffffff3b71f980,
> 100038da0)
> >> > libc.so.1`_lwp_start(0, 0, 0, 0, 0, 0)
> >> >
> >> >
> #############################################################################
> >> >
> >> > # registers
> >> >
> >> > %g0 = 0x0000000000000000                 %l0 = 0xffffffff7a4307a0
> >> > %g1 = 0xffffffff004631a1                 %l1 = 0x0000000000000000
> >> > %g2 = 0x0000000000000000                 %l2 = 0x0000000000000000
> >> > %g3 = 0x000000010051c798                 %l3 = 0x0000000000000000
> >> > %g4 = 0xffffffff004631a0                 %l4 = 0x0000000000000000
> >> > %g5 = 0x00000001004631a0 beam.smp`firstfit_carrier_pool %l5 =
> 0x0000000000000000
> >>
> >> This is interesting.  Notice how the low 32-bits 004631a0 show up in
> >> three variations:
> >> 1. 00000001004631a0 beam.smp`firstfit_carrier_pool (the address of the
> >> firstfit_carrier_pool global variable)
> >> 2. ffffffff004631a0 (the above, but with the high 32 bits replaced
> >> with all-bits-one)
> >> 3. ffffffff004631a1 (the above, but with a tag in the low bit)
> >>
> >> > %g6 = 0x0000000000000000                 %l6 = 0x0000000000000000
> >> > %g7 = 0xffffffff39401a40                 %l7 = 0x0000000000000000
> >> > %o0 = 0x000000010051c500                 %i0 = 0x000000010051c500
> >> > %o1 = 0xffffffff7a400000                 %i1 = 0xffffffff7a400000
> >> > %o2 = 0x00000000000676c0                 %i2 = 0xffffffff7a441de8
> >> > %o3 = 0xffffffff7a400018                 %i3 = 0xffffffff7c903818
> >> > %o4 = 0x00000000000007b9                 %i4 = 0x0000000000000000
> >> > %o5 = 0x000000010051c790                 %i5 = 0x0000000000000023
> >> > %o6 = 0xffffffff7c902eb1                 %i6 = 0xffffffff7c902f61
> >> > %o7 = 0x000000010006de3c abandon_carrier+0x118 %i7 =
> 0x000000010006e958 dealloc_block.part.17+0x1c0
> >> >
> >> >  %ccr = 0x44 xcc=nZvc icc=nZvc
> >> >    %y = 0x0000000000000000
> >> >   %pc = 0x000000010006db14 cpool_insert+0xd0
> >> >  %npc = 0x000000010006db18 cpool_insert+0xd4
> >> >   %sp = 0xffffffff7c902eb1
> >> >   %fp = 0xffffffff7c902f61
> >> >
> >> >  %asi = 0x82
> >> > %fprs = 0x00
> >> >
> >> > # dissassembly around pc
> >> >
> >> > cpool_insert+0xa8:              mov       %g1, %g2
> >> > cpool_insert+0xac:              ldx       [%g5 + 0x10], %g1
> >> > cpool_insert+0xb0:              membar    #LoadLoad|#LoadStore
> >> > cpool_insert+0xb4:              ba,pt     %xcc, +0x1c
>  <cpool_insert+0xd0>
> >> > cpool_insert+0xb8:              and       %g1, -0x4, %g4
> >>
> >> > cpool_insert+0xbc:              membar    #LoadLoad|#LoadStore
> >> > cpool_insert+0xc0:              and       %g2, 0x3, %g3
> >> > cpool_insert+0xc4:              brz,pn    %g3, +0x1ec
>  <cpool_insert+0x2b0>
> >> > cpool_insert+0xc8:              mov       %g2, %g1
> >> > cpool_insert+0xcc:              and       %g1, -0x4, %g4
> >> > cpool_insert+0xd0:              ld        [%g4 + 0x10], %g1
> >>
> >> This is the faulting instruction. We're in the /* Find a predecessor
> >> to be, and set mod marker on its next ptr */ loop.
> >>
> >> > cpool_insert+0xd4:              ld        [%g4 + 0x14], %g2
> >> > cpool_insert+0xd8:              sllx      %g1, 0x20, %g1
> >> > cpool_insert+0xdc:              cmp       %g5, %g4
> >> > cpool_insert+0xe0:              bne,pt    %xcc, -0x24
>  <cpool_insert+0xbc>
> >> > cpool_insert+0xe4:              or        %g2, %g1, %g2
> >>
> >> The above reads a 64-bit "->next" pointer by assembling two adjacent
> >> 32-bit fields.  Weird, but arithmetically Ok.
> >>
> >> Two things strike me:
> >> 1. The compiler implements "atomic load of 64-bits" as "load 32 bits,
> >> load another 32 bits, combine", which isn't correct in a multithreaded
> >> program.  The error could be in the compiler, or in the source code.
> >> 2. In the register dump it was obvious that the high bits of an
> >> address had been clobbered.
> >>
> >> My suspicion is that either Sun's compiler is buggy, or Erlang is
> >> selecting non thread-safe code in this case.
> >>
> >> On SPARC64 Linux w/ GCC I get very different code that uses "ldx" for
> >> those 64-bit loads, as expected.
> >>
> >> /Mikael
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://erlang.org/pipermail/erlang-questions/attachments/20200514/36b5eccd/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: erl_files.tar.gz
Type: application/x-gzip
Size: 134475 bytes
Desc: not available
URL: <http://erlang.org/pipermail/erlang-questions/attachments/20200514/36b5eccd/attachment-0001.bin>


More information about the erlang-questions mailing list