BUG: soft lockup for a user mode linux image

Running trinity (1 process, no victim files, just "$>trinity -C1) for a longer time within a 32 bit user mode linux image with a recent git kernel (host: 3.11.3 guest 3.12-rc3-g...) yields into this konsole message :

* Starting local
net.core.warnings = 0 [ ok ]

BUG: soft lockup - CPU#0 stuck for 23s! [trinity-child0:2031]

and at the host t1 of the "linux"-processes eats all CPU cycles at 1 CPU core. 2 subsequent made back traces made with

$> sudo gdb /home/tfoerste/devel/linux/linux 28144 -n -batch -ex bt

shows nearly a similar position around __get_user_pages() - both are attached.

I'm not surprised that trinity harms a systems - I'm just wondering whether this particular picture is expected or if it points to an issue.

FWIW the last lines of trinity log were :

[2031] [94] setsid() = 2031
[2031] [95] setresgid(rgid=0xffff33e3, egid=0xffffff93, sgid=0x22000040) = -1 (Operation not permitted)
[2031] [96] vmsplice(fd=5, iov=0x85501e0, nr_segs=300, flags=9) = 0x3000
[2031] [97] setresuid(ruid=0x80549193, euid=0xc61041e0, suid=0xff19b6fa) = -1 (Operation not permitted)
[2031] [98] setpriority(which=0xff010000, who=0xf3737373, niceval=0x8088960c) = -1 (Invalid argument)
[2031] [99] socketcall(call=1, args=0x8550200) = -1 (Address family not supported by protocol)
[2031] [100] access(filename="�", mode=2017) = -1 (Invalid argument)
[2031] [101] getgroups(gidsetsize=0, grouplist=0x80d0000[page_rand]) = 3
[2031] [102] msync(start=0xc0100220, len=0, flags=3) = -1 (Invalid argument)
[2031] [103] sigpending(set=0x40025000) = 0
[2031] [104] signalfd4(ufd=383, user_mask=1, sizemask=0xa4200000, flags=0x80800) = -1 (Invalid argument)
[2031] [105] sendfile(out_fd=383, in_fd=382, offset=0, count=4096) = -1 (Invalid argument)
[2031] [106] fanotify_mark(fanotify_fd=382, flags=5, mask=0x8000023, dfd=382, pathname="/proc/1092/task/1092/fdinfo/68") = -1 (Invalid argument) [2031] [107] wait4(upid=1, stat_addr=4, options=0xd761979b, ru=8) = -1 (Invalid argument)
[2031] [108] sigpending(set=0x80ca000[page_zeros]) = 0
[2031] [109] setresuid(ruid=0xefffd6fc, euid=0x1bf4c92f, suid=0xffff2e33) = -1 (Operation not permitted)
[2031] [110] munlock(addr=0x40025000, len=34) = 0
[2031] [111] timer_delete(timer_id=0xffffffdc) = -1 (Invalid argument)
[2031] [112] sched_get_priority_max(policy=0x10000040) = -1 (Invalid argument)
[2031] [113] syslog(type=0xc1000000, buf=1, len=0x82a5) = -1 (Operation not permitted)
[2031] [114] setpriority(which=0xc4c806c6, who=0xffffff01, niceval=0xffff0682) = -1 (Invalid argument)
[2031] [115] getgroups16(gidsetsize=0xfffe, grouplist=1) = -1 (Bad address)
[2031] [116] rename(oldname=4, newname=8) = -1 (Bad address)
[2031] [117] inotify_init() = 654
[2031] [118] getgid() = 100
[2031] [119] fstatat64(dfd=382, filename="/sys/devices/virtual/net/sit0/duplex", statbuf=0, flag=0xb545d727) = -1 (Invalid argument) [2031] [120] unlinkat(dfd=382, pathname="/proc/sys/net/ipv4/neigh/default/retrans_time", flag=0xc00ef76) = -1 (Invalid argument) [2031] [121] timerfd_create(clockid=0, flags=0) = 655
[2031] [122] munlock(addr=4, len=0x3fff) = -1 (Cannot allocate memory)
[2031] [123] fremovexattr(fd=382, name=0) = -1 (Bad address)
[2031] [124] sched_get_priority_min(policy=0xff58bfef) = -1 (Invalid argument)
[2031] [125] mq_timedreceive(mqdes=397, u_msg_ptr=4, msg_len=5245, u_msg_prio=0xc0100220, u_abs_timeout=0xc0100220) = -1 (Bad address)

[2031] [126] chdir(filename="/proc/116/net/ptype") = -1 (Not a directory)
[2031] [127] ssetmask(newmask=0x88000092) = 0
[2031] [128] statfs(pathname="/proc/6/mounts", buf=0) = -1 (Bad address)
[2031] [129] fchown16(fd=397, user=104, group=0x94100000) = -1 (Operation not permitted)
[2031] [130] fchdir(fd=397) = -1 (Not a directory)
[2031] [131] mkdir(pathname="/proc/1092/task/1092/fdinfo/316", mode=525) = -1 (File exists)
[2031] [132] fsetxattr(fd=386, name=0x856f158, value=0x8571160, size=0, flags=0) = -1 (Numerical result out of range) [2031] [133] io_setup(nr_events=4095, ctxp=0x40266000) ^CKilled by signal 2.

Quoted 55 lines · Toralf Förster

Running trinity (1 process, no victim files, just "$>trinity -C1) for a longer time within a 32 bit user mode linux image with a recent git kernel (host: 3.11.3 guest 3.12-rc3-g...) yields into this konsole message :

* Starting local
net.core.warnings = 0 [ ok ]

BUG: soft lockup - CPU#0 stuck for 23s! [trinity-child0:2031]

and at the host t1 of the "linux"-processes eats all CPU cycles at 1 CPU core. 2 subsequent made back traces made with

$> sudo gdb /home/tfoerste/devel/linux/linux 28144 -n -batch -ex bt

shows nearly a similar position around __get_user_pages() - both are attached.

I'm not surprised that trinity harms a systems - I'm just wondering whether this particular picture is expected or if it points to an issue.

FWIW the last lines of trinity log were :

[2031] [94] setsid() = 2031
[2031] [95] setresgid(rgid=0xffff33e3, egid=0xffffff93, sgid=0x22000040) = -1 (Operation not permitted)
[2031] [96] vmsplice(fd=5, iov=0x85501e0, nr_segs=300, flags=9) = 0x3000
[2031] [97] setresuid(ruid=0x80549193, euid=0xc61041e0, suid=0xff19b6fa) = -1 (Operation not permitted)
[2031] [98] setpriority(which=0xff010000, who=0xf3737373, niceval=0x8088960c) = -1 (Invalid argument)
[2031] [99] socketcall(call=1, args=0x8550200) = -1 (Address family not supported by protocol)
[2031] [100] access(filename="�", mode=2017) = -1 (Invalid argument)
[2031] [101] getgroups(gidsetsize=0, grouplist=0x80d0000[page_rand]) = 3
[2031] [102] msync(start=0xc0100220, len=0, flags=3) = -1 (Invalid argument)
[2031] [103] sigpending(set=0x40025000) = 0
[2031] [104] signalfd4(ufd=383, user_mask=1, sizemask=0xa4200000, flags=0x80800) = -1 (Invalid argument)
[2031] [105] sendfile(out_fd=383, in_fd=382, offset=0, count=4096) = -1 (Invalid argument)
[2031] [106] fanotify_mark(fanotify_fd=382, flags=5, mask=0x8000023, dfd=382, pathname="/proc/1092/task/1092/fdinfo/68") = -1 (Invalid argument) [2031] [107] wait4(upid=1, stat_addr=4, options=0xd761979b, ru=8) = -1 (Invalid argument)
[2031] [108] sigpending(set=0x80ca000[page_zeros]) = 0
[2031] [109] setresuid(ruid=0xefffd6fc, euid=0x1bf4c92f, suid=0xffff2e33) = -1 (Operation not permitted)
[2031] [110] munlock(addr=0x40025000, len=34) = 0
[2031] [111] timer_delete(timer_id=0xffffffdc) = -1 (Invalid argument)
[2031] [112] sched_get_priority_max(policy=0x10000040) = -1 (Invalid argument)
[2031] [113] syslog(type=0xc1000000, buf=1, len=0x82a5) = -1 (Operation not permitted)
[2031] [114] setpriority(which=0xc4c806c6, who=0xffffff01, niceval=0xffff0682) = -1 (Invalid argument)
[2031] [115] getgroups16(gidsetsize=0xfffe, grouplist=1) = -1 (Bad address)
[2031] [116] rename(oldname=4, newname=8) = -1 (Bad address)
[2031] [117] inotify_init() = 654
[2031] [118] getgid() = 100
[2031] [119] fstatat64(dfd=382, filename="/sys/devices/virtual/net/sit0/duplex", statbuf=0, flag=0xb545d727) = -1 (Invalid argument) [2031] [120] unlinkat(dfd=382, pathname="/proc/sys/net/ipv4/neigh/default/retrans_time", flag=0xc00ef76) = -1 (Invalid argument) [2031] [121] timerfd_create(clockid=0, flags=0) = 655
[2031] [122] munlock(addr=4, len=0x3fff) = -1 (Cannot allocate memory)
[2031] [123] fremovexattr(fd=382, name=0) = -1 (Bad address)
[2031] [124] sched_get_priority_min(policy=0xff58bfef) = -1 (Invalid argument)
[2031] [125] mq_timedreceive(mqdes=397, u_msg_ptr=4, msg_len=5245, u_msg_prio=0xc0100220, u_abs_timeout=0xc0100220) = -1 (Bad address)

Quoted 8 lines

[2031] [126] chdir(filename="/proc/116/net/ptype") = -1 (Not a directory)
[2031] [127] ssetmask(newmask=0x88000092) = 0
[2031] [128] statfs(pathname="/proc/6/mounts", buf=0) = -1 (Bad address)
[2031] [129] fchown16(fd=397, user=104, group=0x94100000) = -1 (Operation not permitted)
[2031] [130] fchdir(fd=397) = -1 (Not a directory)
[2031] [131] mkdir(pathname="/proc/1092/task/1092/fdinfo/316", mode=525) = -1 (File exists)
[2031] [132] fsetxattr(fd=386, name=0x856f158, value=0x8571160, size=0, flags=0) = -1 (Numerical result out of range) [2031] [133] io_setup(nr_events=4095, ctxp=0x40266000) ^CKilled by signal 2.

Reading your gdb backtraces show that schedule_timeout() got called with a negative value.
Looks like an integer overflow.
The soft-lockup might also origin from that (very big integer which did not overflow jet)

Quoted 65 lines · Toralf Förster

Quoted 55 lines

Running trinity (1 process, no victim files, just "$>trinity -C1) for a longer time within a 32 bit user mode linux image with a recent git kernel (host: 3.11.3 guest 3.12-rc3-g...) yields into this konsole message :

* Starting local
net.core.warnings = 0 [ ok ]

BUG: soft lockup - CPU#0 stuck for 23s! [trinity-child0:2031]

and at the host t1 of the "linux"-processes eats all CPU cycles at 1 CPU core. 2 subsequent made back traces made with

$> sudo gdb /home/tfoerste/devel/linux/linux 28144 -n -batch -ex bt

shows nearly a similar position around __get_user_pages() - both are attached.

I'm not surprised that trinity harms a systems - I'm just wondering whether this particular picture is expected or if it points to an issue.

FWIW the last lines of trinity log were :

[2031] [94] setsid() = 2031
[2031] [95] setresgid(rgid=0xffff33e3, egid=0xffffff93, sgid=0x22000040) = -1 (Operation not permitted)
[2031] [96] vmsplice(fd=5, iov=0x85501e0, nr_segs=300, flags=9) = 0x3000
[2031] [97] setresuid(ruid=0x80549193, euid=0xc61041e0, suid=0xff19b6fa) = -1 (Operation not permitted)
[2031] [98] setpriority(which=0xff010000, who=0xf3737373, niceval=0x8088960c) = -1 (Invalid argument)
[2031] [99] socketcall(call=1, args=0x8550200) = -1 (Address family not supported by protocol)
[2031] [100] access(filename="�", mode=2017) = -1 (Invalid argument)
[2031] [101] getgroups(gidsetsize=0, grouplist=0x80d0000[page_rand]) = 3
[2031] [102] msync(start=0xc0100220, len=0, flags=3) = -1 (Invalid argument)
[2031] [103] sigpending(set=0x40025000) = 0
[2031] [104] signalfd4(ufd=383, user_mask=1, sizemask=0xa4200000, flags=0x80800) = -1 (Invalid argument)
[2031] [105] sendfile(out_fd=383, in_fd=382, offset=0, count=4096) = -1 (Invalid argument)
[2031] [106] fanotify_mark(fanotify_fd=382, flags=5, mask=0x8000023, dfd=382, pathname="/proc/1092/task/1092/fdinfo/68") = -1 (Invalid argument) [2031] [107] wait4(upid=1, stat_addr=4, options=0xd761979b, ru=8) = -1 (Invalid argument)
[2031] [108] sigpending(set=0x80ca000[page_zeros]) = 0
[2031] [109] setresuid(ruid=0xefffd6fc, euid=0x1bf4c92f, suid=0xffff2e33) = -1 (Operation not permitted)
[2031] [110] munlock(addr=0x40025000, len=34) = 0
[2031] [111] timer_delete(timer_id=0xffffffdc) = -1 (Invalid argument)
[2031] [112] sched_get_priority_max(policy=0x10000040) = -1 (Invalid argument)
[2031] [113] syslog(type=0xc1000000, buf=1, len=0x82a5) = -1 (Operation not permitted)
[2031] [114] setpriority(which=0xc4c806c6, who=0xffffff01, niceval=0xffff0682) = -1 (Invalid argument)
[2031] [115] getgroups16(gidsetsize=0xfffe, grouplist=1) = -1 (Bad address)
[2031] [116] rename(oldname=4, newname=8) = -1 (Bad address)
[2031] [117] inotify_init() = 654
[2031] [118] getgid() = 100
[2031] [119] fstatat64(dfd=382, filename="/sys/devices/virtual/net/sit0/duplex", statbuf=0, flag=0xb545d727) = -1 (Invalid argument) [2031] [120] unlinkat(dfd=382, pathname="/proc/sys/net/ipv4/neigh/default/retrans_time", flag=0xc00ef76) = -1 (Invalid argument) [2031] [121] timerfd_create(clockid=0, flags=0) = 655
[2031] [122] munlock(addr=4, len=0x3fff) = -1 (Cannot allocate memory)
[2031] [123] fremovexattr(fd=382, name=0) = -1 (Bad address)
[2031] [124] sched_get_priority_min(policy=0xff58bfef) = -1 (Invalid argument)
[2031] [125] mq_timedreceive(mqdes=397, u_msg_ptr=4, msg_len=5245, u_msg_prio=0xc0100220, u_abs_timeout=0xc0100220) = -1 (Bad address)

Quoted 8 lines

[2031] [126] chdir(filename="/proc/116/net/ptype") = -1 (Not a directory)
[2031] [127] ssetmask(newmask=0x88000092) = 0
[2031] [128] statfs(pathname="/proc/6/mounts", buf=0) = -1 (Bad address)
[2031] [129] fchown16(fd=397, user=104, group=0x94100000) = -1 (Operation not permitted)
[2031] [130] fchdir(fd=397) = -1 (Not a directory)
[2031] [131] mkdir(pathname="/proc/1092/task/1092/fdinfo/316", mode=525) = -1 (File exists)
[2031] [132] fsetxattr(fd=386, name=0x856f158, value=0x8571160, size=0, flags=0) = -1 (Numerical result out of range) [2031] [133] io_setup(nr_events=4095, ctxp=0x40266000) ^CKilled by signal 2.

Quoted 6 lines

Reading your gdb backtraces show that schedule_timeout() got called with a negative value.
Looks like an integer overflow.
The soft-lockup might also origin from that (very big integer which did not overflow jet)

If the culprit is solved by this patch I'd like to send it out. But I'm unsure whether it catches the culprit or if it just covers the root cause.

Am 03.10.2013 20:54, schrieb Toralf Förster:

Quoted 73 lines · Richard Weinberger

Quoted 71 lines · Toralf Förster

Quoted 55 lines

Running trinity (1 process, no victim files, just "$>trinity -C1) for a longer time within a 32 bit user mode linux image with a recent git kernel (host: 3.11.3 guest 3.12-rc3-g...) yields into this konsole message :

* Starting local
net.core.warnings = 0 [ ok ]

BUG: soft lockup - CPU#0 stuck for 23s! [trinity-child0:2031]

and at the host t1 of the "linux"-processes eats all CPU cycles at 1 CPU core. 2 subsequent made back traces made with

$> sudo gdb /home/tfoerste/devel/linux/linux 28144 -n -batch -ex bt

shows nearly a similar position around __get_user_pages() - both are attached.

I'm not surprised that trinity harms a systems - I'm just wondering whether this particular picture is expected or if it points to an issue.

FWIW the last lines of trinity log were :

[2031] [94] setsid() = 2031
[2031] [95] setresgid(rgid=0xffff33e3, egid=0xffffff93, sgid=0x22000040) = -1 (Operation not permitted)
[2031] [96] vmsplice(fd=5, iov=0x85501e0, nr_segs=300, flags=9) = 0x3000
[2031] [97] setresuid(ruid=0x80549193, euid=0xc61041e0, suid=0xff19b6fa) = -1 (Operation not permitted)
[2031] [98] setpriority(which=0xff010000, who=0xf3737373, niceval=0x8088960c) = -1 (Invalid argument)
[2031] [99] socketcall(call=1, args=0x8550200) = -1 (Address family not supported by protocol)
[2031] [100] access(filename="�", mode=2017) = -1 (Invalid argument)
[2031] [101] getgroups(gidsetsize=0, grouplist=0x80d0000[page_rand]) = 3
[2031] [102] msync(start=0xc0100220, len=0, flags=3) = -1 (Invalid argument)
[2031] [103] sigpending(set=0x40025000) = 0
[2031] [104] signalfd4(ufd=383, user_mask=1, sizemask=0xa4200000, flags=0x80800) = -1 (Invalid argument)
[2031] [105] sendfile(out_fd=383, in_fd=382, offset=0, count=4096) = -1 (Invalid argument)
[2031] [106] fanotify_mark(fanotify_fd=382, flags=5, mask=0x8000023, dfd=382, pathname="/proc/1092/task/1092/fdinfo/68") = -1 (Invalid argument) [2031] [107] wait4(upid=1, stat_addr=4, options=0xd761979b, ru=8) = -1 (Invalid argument)
[2031] [108] sigpending(set=0x80ca000[page_zeros]) = 0
[2031] [109] setresuid(ruid=0xefffd6fc, euid=0x1bf4c92f, suid=0xffff2e33) = -1 (Operation not permitted)
[2031] [110] munlock(addr=0x40025000, len=34) = 0
[2031] [111] timer_delete(timer_id=0xffffffdc) = -1 (Invalid argument)
[2031] [112] sched_get_priority_max(policy=0x10000040) = -1 (Invalid argument)
[2031] [113] syslog(type=0xc1000000, buf=1, len=0x82a5) = -1 (Operation not permitted)
[2031] [114] setpriority(which=0xc4c806c6, who=0xffffff01, niceval=0xffff0682) = -1 (Invalid argument)
[2031] [115] getgroups16(gidsetsize=0xfffe, grouplist=1) = -1 (Bad address)
[2031] [116] rename(oldname=4, newname=8) = -1 (Bad address)
[2031] [117] inotify_init() = 654
[2031] [118] getgid() = 100
[2031] [119] fstatat64(dfd=382, filename="/sys/devices/virtual/net/sit0/duplex", statbuf=0, flag=0xb545d727) = -1 (Invalid argument) [2031] [120] unlinkat(dfd=382, pathname="/proc/sys/net/ipv4/neigh/default/retrans_time", flag=0xc00ef76) = -1 (Invalid argument) [2031] [121] timerfd_create(clockid=0, flags=0) = 655
[2031] [122] munlock(addr=4, len=0x3fff) = -1 (Cannot allocate memory)
[2031] [123] fremovexattr(fd=382, name=0) = -1 (Bad address)
[2031] [124] sched_get_priority_min(policy=0xff58bfef) = -1 (Invalid argument)
[2031] [125] mq_timedreceive(mqdes=397, u_msg_ptr=4, msg_len=5245, u_msg_prio=0xc0100220, u_abs_timeout=0xc0100220) = -1 (Bad address)

Quoted 8 lines

[2031] [126] chdir(filename="/proc/116/net/ptype") = -1 (Not a directory)
[2031] [127] ssetmask(newmask=0x88000092) = 0
[2031] [128] statfs(pathname="/proc/6/mounts", buf=0) = -1 (Bad address)
[2031] [129] fchown16(fd=397, user=104, group=0x94100000) = -1 (Operation not permitted)
[2031] [130] fchdir(fd=397) = -1 (Not a directory)
[2031] [131] mkdir(pathname="/proc/1092/task/1092/fdinfo/316", mode=525) = -1 (File exists)
[2031] [132] fsetxattr(fd=386, name=0x856f158, value=0x8571160, size=0, flags=0) = -1 (Numerical result out of range) [2031] [133] io_setup(nr_events=4095, ctxp=0x40266000) ^CKilled by signal 2.

Reading your gdb backtraces show that schedule_timeout() got called with a negative value.
Looks like an integer overflow.
The soft-lockup might also origin from that (very big integer which did not overflow jet)

Quoted 2 lines

If the culprit is solved by this patch I'd like to send it out. But I'm unsure whether it catches the culprit or if it just covers the root cause.

I fear your Patch will not fix the issue.

Does the issue only trigger on 32bit UMLs?

How long does it take till trinity hits it?

Thanks,
//richard

Quoted 74 lines

Am 03.10.2013 20:54, schrieb Toralf Förster:

Quoted 73 lines · Richard Weinberger

Quoted 71 lines · Toralf Förster

Quoted 55 lines

Running trinity (1 process, no victim files, just "$>trinity -C1) for a longer time within a 32 bit user mode linux image with a recent git kernel (host: 3.11.3 guest 3.12-rc3-g...) yields into this konsole message :

* Starting local
net.core.warnings = 0 [ ok ]

BUG: soft lockup - CPU#0 stuck for 23s! [trinity-child0:2031]

and at the host t1 of the "linux"-processes eats all CPU cycles at 1 CPU core. 2 subsequent made back traces made with

$> sudo gdb /home/tfoerste/devel/linux/linux 28144 -n -batch -ex bt

shows nearly a similar position around __get_user_pages() - both are attached.

I'm not surprised that trinity harms a systems - I'm just wondering whether this particular picture is expected or if it points to an issue.

FWIW the last lines of trinity log were :

[2031] [94] setsid() = 2031
[2031] [95] setresgid(rgid=0xffff33e3, egid=0xffffff93, sgid=0x22000040) = -1 (Operation not permitted)
[2031] [96] vmsplice(fd=5, iov=0x85501e0, nr_segs=300, flags=9) = 0x3000
[2031] [97] setresuid(ruid=0x80549193, euid=0xc61041e0, suid=0xff19b6fa) = -1 (Operation not permitted)
[2031] [98] setpriority(which=0xff010000, who=0xf3737373, niceval=0x8088960c) = -1 (Invalid argument)
[2031] [99] socketcall(call=1, args=0x8550200) = -1 (Address family not supported by protocol)
[2031] [100] access(filename="�", mode=2017) = -1 (Invalid argument)
[2031] [101] getgroups(gidsetsize=0, grouplist=0x80d0000[page_rand]) = 3
[2031] [102] msync(start=0xc0100220, len=0, flags=3) = -1 (Invalid argument)
[2031] [103] sigpending(set=0x40025000) = 0
[2031] [104] signalfd4(ufd=383, user_mask=1, sizemask=0xa4200000, flags=0x80800) = -1 (Invalid argument)
[2031] [105] sendfile(out_fd=383, in_fd=382, offset=0, count=4096) = -1 (Invalid argument)
[2031] [106] fanotify_mark(fanotify_fd=382, flags=5, mask=0x8000023, dfd=382, pathname="/proc/1092/task/1092/fdinfo/68") = -1 (Invalid argument) [2031] [107] wait4(upid=1, stat_addr=4, options=0xd761979b, ru=8) = -1 (Invalid argument)
[2031] [108] sigpending(set=0x80ca000[page_zeros]) = 0
[2031] [109] setresuid(ruid=0xefffd6fc, euid=0x1bf4c92f, suid=0xffff2e33) = -1 (Operation not permitted)
[2031] [110] munlock(addr=0x40025000, len=34) = 0
[2031] [111] timer_delete(timer_id=0xffffffdc) = -1 (Invalid argument)
[2031] [112] sched_get_priority_max(policy=0x10000040) = -1 (Invalid argument)
[2031] [113] syslog(type=0xc1000000, buf=1, len=0x82a5) = -1 (Operation not permitted)
[2031] [114] setpriority(which=0xc4c806c6, who=0xffffff01, niceval=0xffff0682) = -1 (Invalid argument)
[2031] [115] getgroups16(gidsetsize=0xfffe, grouplist=1) = -1 (Bad address)
[2031] [116] rename(oldname=4, newname=8) = -1 (Bad address)
[2031] [117] inotify_init() = 654
[2031] [118] getgid() = 100
[2031] [119] fstatat64(dfd=382, filename="/sys/devices/virtual/net/sit0/duplex", statbuf=0, flag=0xb545d727) = -1 (Invalid argument) [2031] [120] unlinkat(dfd=382, pathname="/proc/sys/net/ipv4/neigh/default/retrans_time", flag=0xc00ef76) = -1 (Invalid argument) [2031] [121] timerfd_create(clockid=0, flags=0) = 655
[2031] [122] munlock(addr=4, len=0x3fff) = -1 (Cannot allocate memory)
[2031] [123] fremovexattr(fd=382, name=0) = -1 (Bad address)
[2031] [124] sched_get_priority_min(policy=0xff58bfef) = -1 (Invalid argument)
[2031] [125] mq_timedreceive(mqdes=397, u_msg_ptr=4, msg_len=5245, u_msg_prio=0xc0100220, u_abs_timeout=0xc0100220) = -1 (Bad address)

Quoted 8 lines

[2031] [126] chdir(filename="/proc/116/net/ptype") = -1 (Not a directory)
[2031] [127] ssetmask(newmask=0x88000092) = 0
[2031] [128] statfs(pathname="/proc/6/mounts", buf=0) = -1 (Bad address)
[2031] [129] fchown16(fd=397, user=104, group=0x94100000) = -1 (Operation not permitted)
[2031] [130] fchdir(fd=397) = -1 (Not a directory)
[2031] [131] mkdir(pathname="/proc/1092/task/1092/fdinfo/316", mode=525) = -1 (File exists)
[2031] [132] fsetxattr(fd=386, name=0x856f158, value=0x8571160, size=0, flags=0) = -1 (Numerical result out of range) [2031] [133] io_setup(nr_events=4095, ctxp=0x40266000) ^CKilled by signal 2.

Reading your gdb backtraces show that schedule_timeout() got called with a negative value.
Looks like an integer overflow.
The soft-lockup might also origin from that (very big integer which did not overflow jet)

Quoted 6 lines

Quoted 2 lines

If the culprit is solved by this patch I'd like to send it out. But I'm unsure whether it catches the culprit or if it just covers the root cause.

I fear your Patch will not fix the issue.

Does the issue only trigger on 32bit UMLs?

No diea, I do only have a 32 bit system here (both host and client).

Quoted 1 line

How long does it take till trinity hits it?

a command like

$> ssh tfoerste@trinity "rm -rf t3; mkdir t3; cd t3; trinity -C4"

usually needs 10 till 15 min to trigger the issue. With just 1 trinity task (-C1) however it needs often a hour or more.

Am 03.10.2013 21:16, schrieb Toralf Förster:

Quoted 88 lines · Richard Weinberger

Quoted 74 lines

Am 03.10.2013 20:54, schrieb Toralf Förster:

Quoted 73 lines · Richard Weinberger

Quoted 71 lines · Toralf Förster

Quoted 55 lines

Running trinity (1 process, no victim files, just "$>trinity -C1) for a longer time within a 32 bit user mode linux image with a recent git kernel (host: 3.11.3 guest 3.12-rc3-g...) yields into this konsole message :

* Starting local
net.core.warnings = 0 [ ok ]

BUG: soft lockup - CPU#0 stuck for 23s! [trinity-child0:2031]

and at the host t1 of the "linux"-processes eats all CPU cycles at 1 CPU core. 2 subsequent made back traces made with

$> sudo gdb /home/tfoerste/devel/linux/linux 28144 -n -batch -ex bt

shows nearly a similar position around __get_user_pages() - both are attached.

I'm not surprised that trinity harms a systems - I'm just wondering whether this particular picture is expected or if it points to an issue.

FWIW the last lines of trinity log were :

[2031] [94] setsid() = 2031
[2031] [95] setresgid(rgid=0xffff33e3, egid=0xffffff93, sgid=0x22000040) = -1 (Operation not permitted)
[2031] [96] vmsplice(fd=5, iov=0x85501e0, nr_segs=300, flags=9) = 0x3000
[2031] [97] setresuid(ruid=0x80549193, euid=0xc61041e0, suid=0xff19b6fa) = -1 (Operation not permitted)
[2031] [98] setpriority(which=0xff010000, who=0xf3737373, niceval=0x8088960c) = -1 (Invalid argument)
[2031] [99] socketcall(call=1, args=0x8550200) = -1 (Address family not supported by protocol)
[2031] [100] access(filename="�", mode=2017) = -1 (Invalid argument)
[2031] [101] getgroups(gidsetsize=0, grouplist=0x80d0000[page_rand]) = 3
[2031] [102] msync(start=0xc0100220, len=0, flags=3) = -1 (Invalid argument)
[2031] [103] sigpending(set=0x40025000) = 0
[2031] [104] signalfd4(ufd=383, user_mask=1, sizemask=0xa4200000, flags=0x80800) = -1 (Invalid argument)
[2031] [105] sendfile(out_fd=383, in_fd=382, offset=0, count=4096) = -1 (Invalid argument)
[2031] [106] fanotify_mark(fanotify_fd=382, flags=5, mask=0x8000023, dfd=382, pathname="/proc/1092/task/1092/fdinfo/68") = -1 (Invalid argument) [2031] [107] wait4(upid=1, stat_addr=4, options=0xd761979b, ru=8) = -1 (Invalid argument)
[2031] [108] sigpending(set=0x80ca000[page_zeros]) = 0
[2031] [109] setresuid(ruid=0xefffd6fc, euid=0x1bf4c92f, suid=0xffff2e33) = -1 (Operation not permitted)
[2031] [110] munlock(addr=0x40025000, len=34) = 0
[2031] [111] timer_delete(timer_id=0xffffffdc) = -1 (Invalid argument)
[2031] [112] sched_get_priority_max(policy=0x10000040) = -1 (Invalid argument)
[2031] [113] syslog(type=0xc1000000, buf=1, len=0x82a5) = -1 (Operation not permitted)
[2031] [114] setpriority(which=0xc4c806c6, who=0xffffff01, niceval=0xffff0682) = -1 (Invalid argument)
[2031] [115] getgroups16(gidsetsize=0xfffe, grouplist=1) = -1 (Bad address)
[2031] [116] rename(oldname=4, newname=8) = -1 (Bad address)
[2031] [117] inotify_init() = 654
[2031] [118] getgid() = 100
[2031] [119] fstatat64(dfd=382, filename="/sys/devices/virtual/net/sit0/duplex", statbuf=0, flag=0xb545d727) = -1 (Invalid argument) [2031] [120] unlinkat(dfd=382, pathname="/proc/sys/net/ipv4/neigh/default/retrans_time", flag=0xc00ef76) = -1 (Invalid argument) [2031] [121] timerfd_create(clockid=0, flags=0) = 655
[2031] [122] munlock(addr=4, len=0x3fff) = -1 (Cannot allocate memory)
[2031] [123] fremovexattr(fd=382, name=0) = -1 (Bad address)
[2031] [124] sched_get_priority_min(policy=0xff58bfef) = -1 (Invalid argument)
[2031] [125] mq_timedreceive(mqdes=397, u_msg_ptr=4, msg_len=5245, u_msg_prio=0xc0100220, u_abs_timeout=0xc0100220) = -1 (Bad address)

Quoted 8 lines

[2031] [126] chdir(filename="/proc/116/net/ptype") = -1 (Not a directory)
[2031] [127] ssetmask(newmask=0x88000092) = 0
[2031] [128] statfs(pathname="/proc/6/mounts", buf=0) = -1 (Bad address)
[2031] [129] fchown16(fd=397, user=104, group=0x94100000) = -1 (Operation not permitted)
[2031] [130] fchdir(fd=397) = -1 (Not a directory)
[2031] [131] mkdir(pathname="/proc/1092/task/1092/fdinfo/316", mode=525) = -1 (File exists)
[2031] [132] fsetxattr(fd=386, name=0x856f158, value=0x8571160, size=0, flags=0) = -1 (Numerical result out of range) [2031] [133] io_setup(nr_events=4095, ctxp=0x40266000) ^CKilled by signal 2.

Reading your gdb backtraces show that schedule_timeout() got called with a negative value.
Looks like an integer overflow.
The soft-lockup might also origin from that (very big integer which did not overflow jet)

Quoted 6 lines

Quoted 2 lines

If the culprit is solved by this patch I'd like to send it out. But I'm unsure whether it catches the culprit or if it just covers the root cause.

I fear your Patch will not fix the issue.

Does the issue only trigger on 32bit UMLs?

No diea, I do only have a 32 bit system here (both host and client).

Quoted 1 line

How long does it take till trinity hits it?

a command like

$> ssh tfoerste@trinity "rm -rf t3; mkdir t3; cd t3; trinity -C4"

Quoted 2 lines

usually needs 10 till 15 min to trigger the issue. With just 1 trinity task (-C1) however it needs often a hour or more.

That's good. :-)
You can place some printk()s into balance_dirty_pages() and observe the values of period, max_pause, min_pause, etc...
Maybe this will give us a clue.

So far the issue looks not really UML specific.
But maybe it is more likely to happen on UML because of the slow page faults...

Thanks,
//richard

Quoted 89 lines

Am 03.10.2013 21:16, schrieb Toralf Förster:

Quoted 88 lines · Richard Weinberger

Quoted 74 lines

Am 03.10.2013 20:54, schrieb Toralf Förster:

Quoted 73 lines · Richard Weinberger

Quoted 71 lines · Toralf Förster

Quoted 55 lines

Running trinity (1 process, no victim files, just "$>trinity -C1) for a longer time within a 32 bit user mode linux image with a recent git kernel (host: 3.11.3 guest 3.12-rc3-g...) yields into this konsole message :

* Starting local
net.core.warnings = 0 [ ok ]

BUG: soft lockup - CPU#0 stuck for 23s! [trinity-child0:2031]

and at the host t1 of the "linux"-processes eats all CPU cycles at 1 CPU core. 2 subsequent made back traces made with

$> sudo gdb /home/tfoerste/devel/linux/linux 28144 -n -batch -ex bt

shows nearly a similar position around __get_user_pages() - both are attached.

I'm not surprised that trinity harms a systems - I'm just wondering whether this particular picture is expected or if it points to an issue.

FWIW the last lines of trinity log were :

[2031] [94] setsid() = 2031
[2031] [95] setresgid(rgid=0xffff33e3, egid=0xffffff93, sgid=0x22000040) = -1 (Operation not permitted)
[2031] [96] vmsplice(fd=5, iov=0x85501e0, nr_segs=300, flags=9) = 0x3000
[2031] [97] setresuid(ruid=0x80549193, euid=0xc61041e0, suid=0xff19b6fa) = -1 (Operation not permitted)
[2031] [98] setpriority(which=0xff010000, who=0xf3737373, niceval=0x8088960c) = -1 (Invalid argument)
[2031] [99] socketcall(call=1, args=0x8550200) = -1 (Address family not supported by protocol)
[2031] [100] access(filename="�", mode=2017) = -1 (Invalid argument)
[2031] [101] getgroups(gidsetsize=0, grouplist=0x80d0000[page_rand]) = 3
[2031] [102] msync(start=0xc0100220, len=0, flags=3) = -1 (Invalid argument)
[2031] [103] sigpending(set=0x40025000) = 0
[2031] [104] signalfd4(ufd=383, user_mask=1, sizemask=0xa4200000, flags=0x80800) = -1 (Invalid argument)
[2031] [105] sendfile(out_fd=383, in_fd=382, offset=0, count=4096) = -1 (Invalid argument)
[2031] [106] fanotify_mark(fanotify_fd=382, flags=5, mask=0x8000023, dfd=382, pathname="/proc/1092/task/1092/fdinfo/68") = -1 (Invalid argument) [2031] [107] wait4(upid=1, stat_addr=4, options=0xd761979b, ru=8) = -1 (Invalid argument)
[2031] [108] sigpending(set=0x80ca000[page_zeros]) = 0
[2031] [109] setresuid(ruid=0xefffd6fc, euid=0x1bf4c92f, suid=0xffff2e33) = -1 (Operation not permitted)
[2031] [110] munlock(addr=0x40025000, len=34) = 0
[2031] [111] timer_delete(timer_id=0xffffffdc) = -1 (Invalid argument)
[2031] [112] sched_get_priority_max(policy=0x10000040) = -1 (Invalid argument)
[2031] [113] syslog(type=0xc1000000, buf=1, len=0x82a5) = -1 (Operation not permitted)
[2031] [114] setpriority(which=0xc4c806c6, who=0xffffff01, niceval=0xffff0682) = -1 (Invalid argument)
[2031] [115] getgroups16(gidsetsize=0xfffe, grouplist=1) = -1 (Bad address)
[2031] [116] rename(oldname=4, newname=8) = -1 (Bad address)
[2031] [117] inotify_init() = 654
[2031] [118] getgid() = 100
[2031] [119] fstatat64(dfd=382, filename="/sys/devices/virtual/net/sit0/duplex", statbuf=0, flag=0xb545d727) = -1 (Invalid argument) [2031] [120] unlinkat(dfd=382, pathname="/proc/sys/net/ipv4/neigh/default/retrans_time", flag=0xc00ef76) = -1 (Invalid argument) [2031] [121] timerfd_create(clockid=0, flags=0) = 655
[2031] [122] munlock(addr=4, len=0x3fff) = -1 (Cannot allocate memory)
[2031] [123] fremovexattr(fd=382, name=0) = -1 (Bad address)
[2031] [124] sched_get_priority_min(policy=0xff58bfef) = -1 (Invalid argument)
[2031] [125] mq_timedreceive(mqdes=397, u_msg_ptr=4, msg_len=5245, u_msg_prio=0xc0100220, u_abs_timeout=0xc0100220) = -1 (Bad address)

Quoted 8 lines

[2031] [126] chdir(filename="/proc/116/net/ptype") = -1 (Not a directory)
[2031] [127] ssetmask(newmask=0x88000092) = 0
[2031] [128] statfs(pathname="/proc/6/mounts", buf=0) = -1 (Bad address)
[2031] [129] fchown16(fd=397, user=104, group=0x94100000) = -1 (Operation not permitted)
[2031] [130] fchdir(fd=397) = -1 (Not a directory)
[2031] [131] mkdir(pathname="/proc/1092/task/1092/fdinfo/316", mode=525) = -1 (File exists)
[2031] [132] fsetxattr(fd=386, name=0x856f158, value=0x8571160, size=0, flags=0) = -1 (Numerical result out of range) [2031] [133] io_setup(nr_events=4095, ctxp=0x40266000) ^CKilled by signal 2.

Reading your gdb backtraces show that schedule_timeout() got called with a negative value.
Looks like an integer overflow.
The soft-lockup might also origin from that (very big integer which did not overflow jet)

Quoted 6 lines

Quoted 2 lines

If the culprit is solved by this patch I'd like to send it out. But I'm unsure whether it catches the culprit or if it just covers the root cause.

I fear your Patch will not fix the issue.

Does the issue only trigger on 32bit UMLs?

No diea, I do only have a 32 bit system here (both host and client).

Quoted 1 line

How long does it take till trinity hits it?

a command like

$> ssh tfoerste@trinity "rm -rf t3; mkdir t3; cd t3; trinity -C4"

Quoted 7 lines

Quoted 2 lines

usually needs 10 till 15 min to trigger the issue. With just 1 trinity task (-C1) however it needs often a hour or more.

That's good. :-)
You can place some printk()s into balance_dirty_pages() and observe the values of period, max_pause, min_pause, etc...
Maybe this will give us a clue.

Will do it again. I did it before already but period was the only thing which becames otherwise negative -
and because pages_dirtied was often in the range of 20 - 30 and perioad around -3000

I just wondered if HZ is casts as a short int in the calculation ? BTW printk doesn't showed anything in the syslog (ok, I used printk(KERN_ERR ...)

Am 03.10.2013 21:33, schrieb Toralf Förster:

Quoted 98 lines · Richard Weinberger

Quoted 89 lines

Am 03.10.2013 21:16, schrieb Toralf Förster:

Quoted 88 lines · Richard Weinberger

Quoted 74 lines

Am 03.10.2013 20:54, schrieb Toralf Förster:

Quoted 73 lines · Richard Weinberger

Quoted 71 lines · Toralf Förster

Quoted 55 lines

Running trinity (1 process, no victim files, just "$>trinity -C1) for a longer time within a 32 bit user mode linux image with a recent git kernel (host: 3.11.3 guest 3.12-rc3-g...) yields into this konsole message :

* Starting local
net.core.warnings = 0 [ ok ]

BUG: soft lockup - CPU#0 stuck for 23s! [trinity-child0:2031]

and at the host t1 of the "linux"-processes eats all CPU cycles at 1 CPU core. 2 subsequent made back traces made with

$> sudo gdb /home/tfoerste/devel/linux/linux 28144 -n -batch -ex bt

shows nearly a similar position around __get_user_pages() - both are attached.

I'm not surprised that trinity harms a systems - I'm just wondering whether this particular picture is expected or if it points to an issue.

FWIW the last lines of trinity log were :

[2031] [94] setsid() = 2031
[2031] [95] setresgid(rgid=0xffff33e3, egid=0xffffff93, sgid=0x22000040) = -1 (Operation not permitted)
[2031] [96] vmsplice(fd=5, iov=0x85501e0, nr_segs=300, flags=9) = 0x3000
[2031] [97] setresuid(ruid=0x80549193, euid=0xc61041e0, suid=0xff19b6fa) = -1 (Operation not permitted)
[2031] [98] setpriority(which=0xff010000, who=0xf3737373, niceval=0x8088960c) = -1 (Invalid argument)
[2031] [99] socketcall(call=1, args=0x8550200) = -1 (Address family not supported by protocol)
[2031] [100] access(filename="�", mode=2017) = -1 (Invalid argument)
[2031] [101] getgroups(gidsetsize=0, grouplist=0x80d0000[page_rand]) = 3
[2031] [102] msync(start=0xc0100220, len=0, flags=3) = -1 (Invalid argument)
[2031] [103] sigpending(set=0x40025000) = 0
[2031] [104] signalfd4(ufd=383, user_mask=1, sizemask=0xa4200000, flags=0x80800) = -1 (Invalid argument)
[2031] [105] sendfile(out_fd=383, in_fd=382, offset=0, count=4096) = -1 (Invalid argument)
[2031] [106] fanotify_mark(fanotify_fd=382, flags=5, mask=0x8000023, dfd=382, pathname="/proc/1092/task/1092/fdinfo/68") = -1 (Invalid argument) [2031] [107] wait4(upid=1, stat_addr=4, options=0xd761979b, ru=8) = -1 (Invalid argument)
[2031] [108] sigpending(set=0x80ca000[page_zeros]) = 0
[2031] [109] setresuid(ruid=0xefffd6fc, euid=0x1bf4c92f, suid=0xffff2e33) = -1 (Operation not permitted)
[2031] [110] munlock(addr=0x40025000, len=34) = 0
[2031] [111] timer_delete(timer_id=0xffffffdc) = -1 (Invalid argument)
[2031] [112] sched_get_priority_max(policy=0x10000040) = -1 (Invalid argument)
[2031] [113] syslog(type=0xc1000000, buf=1, len=0x82a5) = -1 (Operation not permitted)
[2031] [114] setpriority(which=0xc4c806c6, who=0xffffff01, niceval=0xffff0682) = -1 (Invalid argument)
[2031] [115] getgroups16(gidsetsize=0xfffe, grouplist=1) = -1 (Bad address)
[2031] [116] rename(oldname=4, newname=8) = -1 (Bad address)
[2031] [117] inotify_init() = 654
[2031] [118] getgid() = 100
[2031] [119] fstatat64(dfd=382, filename="/sys/devices/virtual/net/sit0/duplex", statbuf=0, flag=0xb545d727) = -1 (Invalid argument) [2031] [120] unlinkat(dfd=382, pathname="/proc/sys/net/ipv4/neigh/default/retrans_time", flag=0xc00ef76) = -1 (Invalid argument) [2031] [121] timerfd_create(clockid=0, flags=0) = 655
[2031] [122] munlock(addr=4, len=0x3fff) = -1 (Cannot allocate memory)
[2031] [123] fremovexattr(fd=382, name=0) = -1 (Bad address)
[2031] [124] sched_get_priority_min(policy=0xff58bfef) = -1 (Invalid argument)
[2031] [125] mq_timedreceive(mqdes=397, u_msg_ptr=4, msg_len=5245, u_msg_prio=0xc0100220, u_abs_timeout=0xc0100220) = -1 (Bad address)

Quoted 8 lines

[2031] [126] chdir(filename="/proc/116/net/ptype") = -1 (Not a directory)
[2031] [127] ssetmask(newmask=0x88000092) = 0
[2031] [128] statfs(pathname="/proc/6/mounts", buf=0) = -1 (Bad address)
[2031] [129] fchown16(fd=397, user=104, group=0x94100000) = -1 (Operation not permitted)
[2031] [130] fchdir(fd=397) = -1 (Not a directory)
[2031] [131] mkdir(pathname="/proc/1092/task/1092/fdinfo/316", mode=525) = -1 (File exists)
[2031] [132] fsetxattr(fd=386, name=0x856f158, value=0x8571160, size=0, flags=0) = -1 (Numerical result out of range) [2031] [133] io_setup(nr_events=4095, ctxp=0x40266000) ^CKilled by signal 2.

Reading your gdb backtraces show that schedule_timeout() got called with a negative value.
Looks like an integer overflow.
The soft-lockup might also origin from that (very big integer which did not overflow jet)

Quoted 6 lines

Quoted 2 lines

If the culprit is solved by this patch I'd like to send it out. But I'm unsure whether it catches the culprit or if it just covers the root cause.

I fear your Patch will not fix the issue.

Does the issue only trigger on 32bit UMLs?

No diea, I do only have a 32 bit system here (both host and client).

Quoted 1 line

How long does it take till trinity hits it?

a command like

$> ssh tfoerste@trinity "rm -rf t3; mkdir t3; cd t3; trinity -C4"

Quoted 7 lines

Quoted 2 lines

usually needs 10 till 15 min to trigger the issue. With just 1 trinity task (-C1) however it needs often a hour or more.

That's good. :-)
You can place some printk()s into balance_dirty_pages() and observe the values of period, max_pause, min_pause, etc...
Maybe this will give us a clue.

Quoted 4 lines

Will do it again. I did it before already but period was the only thing which becames otherwise negative -
and because pages_dirtied was often in the range of 20 - 30 and perioad around -3000

Did you print it before or after

                if (current->dirty_paused_when)
                        pause -= now - current->dirty_paused_when;

I'm also wondering why

		if (pause < min_pause) {

did not trigger or did it?

Quoted 3 lines

I just wondered if HZ is casts as a short int in the calculation ? BTW printk doesn't showed anything in the syslog (ok, I used printk(KERN_ERR ...)

syslog or dmesg?

Thanks,
//richard

The UML stopped here :
...

                if (unlikely(task_ratelimit == 0)) {
                        period = max_pause;
                        pause = max_pause;
                        BUG_ON(pause < 0);
                        goto pause;
                }
                BUG_ON(pages_dirtied < 0);
                BUG_ON(task_ratelimit < 0);
                period = HZ * pages_dirtied / task_ratelimit;
                BUG_ON(period < 0);         <----------------------here

The back trace is :

#9 0x08411c64 in balance_dirty_pages (pages_dirtied=9, mapping=<optimized out>) at mm/page-writeback.c:1471

Quoted 12 lines · Toralf Förster

The UML stopped here :
...

                if (unlikely(task_ratelimit == 0)) {
                        period = max_pause;
                        pause = max_pause;
                        BUG_ON(pause < 0);
                        goto pause;
                }
                BUG_ON(pages_dirtied < 0);
                BUG_ON(task_ratelimit < 0);
                period = HZ * pages_dirtied / task_ratelimit;
                BUG_ON(period < 0);         <----------------------here

So pages_dirtied becomes that big compared to task_ratelimit (both are "unsigned long"), that period (which is "long", just like "pause") overflows into a negative number.

This is indeed much more likely to happen on 32-bit.

Quoted 3 lines

The back trace is :

#9 0x08411c64 in balance_dirty_pages (pages_dirtied=9, mapping=<optimized out>) at mm/page-writeback.c:1471

But here pages_dirtied is only 9??

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds

Quoted 20 lines · Geert Uytterhoeven

Quoted 12 lines · Toralf Förster

The UML stopped here :
...

                if (unlikely(task_ratelimit == 0)) {
                        period = max_pause;
                        pause = max_pause;
                        BUG_ON(pause < 0);
                        goto pause;
                }
                BUG_ON(pages_dirtied < 0);
                BUG_ON(task_ratelimit < 0);
                period = HZ * pages_dirtied / task_ratelimit;
                BUG_ON(period < 0);         <----------------------here

So pages_dirtied becomes that big compared to task_ratelimit (both are "unsigned long"), that period (which is "long", just like "pause") overflows into a negative number.

This is indeed much more likely to happen on 32-bit.

Quoted 5 lines

Quoted 3 lines

The back trace is :

#9 0x08411c64 in balance_dirty_pages (pages_dirtied=9, mapping=<optimized out>) at mm/page-writeback.c:1471

But here pages_dirtied is only 9??

Well, this points to an overflow or ? :

tfoerste@n22 ~/devel/linux $ nl -ba mm/page-writeback.c | grep -A 5 -B 5 1468
1463 BUG_ON(pause < 0);
1464 goto pause;

  1465                  }
  1466                  period = HZ * pages_dirtied / task_ratelimit;
  1467                  pause = period;
  1468                  BUG_ON(pause < 0 && pages_dirtied > 0 && task_ratelimit > 0);
  1469                  if (current->dirty_paused_when)

1470 pause -= now - current->dirty_paused_when;
1471 /*
1472 * For less than 1s think time (ext3/4 may block the dirtier
1473 * for up to 800ms from time to time on 1-HDD; so does xfs,

and the back trace is :

#9 0x08411c6c in balance_dirty_pages (pages_dirtied=0, mapping=<optimized out>) at mm/page-writeback.c:1468

Quoted 21 lines · Toralf Förster

Quoted 20 lines · Geert Uytterhoeven

Quoted 12 lines · Toralf Förster

The UML stopped here :
...

                if (unlikely(task_ratelimit == 0)) {
                        period = max_pause;
                        pause = max_pause;
                        BUG_ON(pause < 0);
                        goto pause;
                }
                BUG_ON(pages_dirtied < 0);
                BUG_ON(task_ratelimit < 0);
                period = HZ * pages_dirtied / task_ratelimit;
                BUG_ON(period < 0);         <----------------------here

So pages_dirtied becomes that big compared to task_ratelimit (both are "unsigned long"), that period (which is "long", just like "pause") overflows into a negative number.

This is indeed much more likely to happen on 32-bit.

Quoted 7 lines

Quoted 5 lines

Quoted 3 lines

The back trace is :

#9 0x08411c64 in balance_dirty_pages (pages_dirtied=9, mapping=<optimized out>) at mm/page-writeback.c:1471

But here pages_dirtied is only 9??

Well, this points to an overflow or ? :

Negative indicates an overflow, but pages_dirtied doesn't.

Quoted 14 lines

tfoerste@n22 ~/devel/linux $ nl -ba mm/page-writeback.c | grep -A 5 -B 5 1468
1463 BUG_ON(pause < 0);
1464 goto pause;

  1465                  }
  1466                  period = HZ * pages_dirtied / task_ratelimit;
  1467                  pause = period;
  1468                  BUG_ON(pause < 0 && pages_dirtied > 0 && task_ratelimit > 0);
  1469                  if (current->dirty_paused_when)

1470 pause -= now - current->dirty_paused_when;
1471 /*
1472 * For less than 1s think time (ext3/4 may block the dirtier
1473 * for up to 800ms from time to time on 1-HDD; so does xfs,

Quoted 3 lines

and the back trace is :

#9 0x08411c6c in balance_dirty_pages (pages_dirtied=0, mapping=<optimized out>) at mm/page-writeback.c:1468

Hmm, now pages_dirtied is zero, according to the backtrace, but the BUG_ON() asserts its strict positive?!?

Can you please try the following instead of the BUG_ON():

if (pause < 0) {
        printk("pages_dirtied = %lu\n", pages_dirtied);
        printk("task_ratelimit = %lu\n", task_ratelimit);
        printk("pause = %ld\n", pause);
}

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds

Quoted 21 lines · Toralf Förster

Quoted 20 lines · Geert Uytterhoeven

Quoted 12 lines · Toralf Förster

The UML stopped here :
...

                if (unlikely(task_ratelimit == 0)) {
                        period = max_pause;
                        pause = max_pause;
                        BUG_ON(pause < 0);
                        goto pause;
                }
                BUG_ON(pages_dirtied < 0);
                BUG_ON(task_ratelimit < 0);
                period = HZ * pages_dirtied / task_ratelimit;
                BUG_ON(period < 0);         <----------------------here

So pages_dirtied becomes that big compared to task_ratelimit (both are "unsigned long"), that period (which is "long", just like "pause") overflows into a negative number.

This is indeed much more likely to happen on 32-bit.

Quoted 7 lines

Quoted 5 lines

Quoted 3 lines

The back trace is :

#9 0x08411c64 in balance_dirty_pages (pages_dirtied=9, mapping=<optimized out>) at mm/page-writeback.c:1471

But here pages_dirtied is only 9??

Well, this points to an overflow or ? :

Negative indicates an overflow, but pages_dirtied doesn't.

Quoted 14 lines

tfoerste@n22 ~/devel/linux $ nl -ba mm/page-writeback.c | grep -A 5 -B 5 1468
1463 BUG_ON(pause < 0);
1464 goto pause;

  1465                  }
  1466                  period = HZ * pages_dirtied / task_ratelimit;
  1467                  pause = period;
  1468                  BUG_ON(pause < 0 && pages_dirtied > 0 && task_ratelimit > 0);
  1469                  if (current->dirty_paused_when)

1470 pause -= now - current->dirty_paused_when;
1471 /*
1472 * For less than 1s think time (ext3/4 may block the dirtier
1473 * for up to 800ms from time to time on 1-HDD; so does xfs,

Quoted 3 lines

and the back trace is :

#9 0x08411c6c in balance_dirty_pages (pages_dirtied=0, mapping=<optimized out>) at mm/page-writeback.c:1468

Hmm, now pages_dirtied is zero, according to the backtrace, but the BUG_ON() asserts its strict positive?!?

Can you please try the following instead of the BUG_ON():

if (pause < 0) {
        printk("pages_dirtied = %lu\n", pages_dirtied);
        printk("task_ratelimit = %lu\n", task_ratelimit);
        printk("pause = %ld\n", pause);
}

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds

Quoted 13 lines

Hmm, now pages_dirtied is zero, according to the backtrace, but the BUG_ON() asserts its strict positive?!?

Can you please try the following instead of the BUG_ON():

if (pause < 0) {
        printk("pages_dirtied = %lu\n", pages_dirtied);
        printk("task_ratelimit = %lu\n", task_ratelimit);
        printk("pause = %ld\n", pause);
}

Gr{oetje,eeting}s,

Quoted 1 line

Geert

I tried it in different ways already - I'm completely unsuccessful in getting any printk output. As soon as the issue happens I do have a

BUG: soft lockup - CPU#0 stuck for 22s! [trinity-child0:1521]

at stderr of the UML and then no further input is accepted. With uml_mconsole I'm however able to run very basic commands like a crash dump, sysrq ond so on.

Well, the quick&dirty hack below at least works for the moment to overcome the soft lookup and the hang/unresponsiveness of the 32 bit user mode linux guest :

mm/page-writeback.c +0 -0

index f5236f8..7e9483c 100644

mm/page-writeback.c +0 -0

 > if (pause < 0) {
 >         printk("pages_dirtied = %lu\n", pages_dirtied);

Quoted 2 lines

        printk("task_ratelimit = %lu\n", task_ratelimit);
        printk("pause = %ld\n", pause);

Quoted 1 line

}

Quoted 2 lines

Quoted 2 lines

Quoted 2 lines

Quoted 2 lines

                BUG_ON(pages_dirtied < 0);
                BUG_ON(task_ratelimit < 0);

Quoted 13 lines · Toralf Förster

Quoted 13 lines

Hmm, now pages_dirtied is zero, according to the backtrace, but the BUG_ON() asserts its strict positive?!?

Can you please try the following instead of the BUG_ON():

if (pause < 0) {
        printk("pages_dirtied = %lu\n", pages_dirtied);
        printk("task_ratelimit = %lu\n", task_ratelimit);
        printk("pause = %ld\n", pause);
}

Gr{oetje,eeting}s,

Quoted 8 lines

Quoted 1 line

Geert

I tried it in different ways already - I'm completely unsuccessful in getting any printk output. As soon as the issue happens I do have a

BUG: soft lockup - CPU#0 stuck for 22s! [trinity-child0:1521]

at stderr of the UML and then no further input is accepted. With uml_mconsole I'm however able to run very basic commands like a crash dump, sysrq ond so on.

You may get an idea of the magnitude of pages_dirtied by using a chain of BUG_ON()s, like:

BUG_ON(pages_dirtied > 2000000000);
BUG_ON(pages_dirtied > 1000000000);
BUG_ON(pages_dirtied > 100000000);
BUG_ON(pages_dirtied > 10000000);
BUG_ON(pages_dirtied > 1000000);

Probably 1 million is already too much for normal operation?

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds

Quoted 32 lines · Toralf Förster

Quoted 13 lines

Quoted 13 lines

Hmm, now pages_dirtied is zero, according to the backtrace, but the BUG_ON() asserts its strict positive?!?

Can you please try the following instead of the BUG_ON():

if (pause < 0) {
        printk("pages_dirtied = %lu\n", pages_dirtied);
        printk("task_ratelimit = %lu\n", task_ratelimit);
        printk("pause = %ld\n", pause);
}

Gr{oetje,eeting}s,

Quoted 8 lines

Quoted 1 line

Geert

I tried it in different ways already - I'm completely unsuccessful in getting any printk output. As soon as the issue happens I do have a

BUG: soft lockup - CPU#0 stuck for 22s! [trinity-child0:1521]

at stderr of the UML and then no further input is accepted. With uml_mconsole I'm however able to run very basic commands like a crash dump, sysrq ond so on.

You may get an idea of the magnitude of pages_dirtied by using a chain of BUG_ON()s, like:

BUG_ON(pages_dirtied > 2000000000);
BUG_ON(pages_dirtied > 1000000000);
BUG_ON(pages_dirtied > 100000000);
BUG_ON(pages_dirtied > 10000000);
BUG_ON(pages_dirtied > 1000000);

Quoted 2 lines

Probably 1 million is already too much for normal operation?

period = HZ * pages_dirtied / task_ratelimit;
		BUG_ON(pages_dirtied > 2000000000);
		BUG_ON(pages_dirtied > 1000000000);      <-------------- this is line 1467

the back trace is :

tfoerste@n22 ~/devel/linux $ gdb --core=/mnt/ramdisk/core /home/tfoerste/devel/linux/linux -batch -ex bt [New LWP 6911]
Core was generated by `/home/tfoerste/devel/linux/linux earlyprintk ubda=/home/tfoerste/virtual/uml/tr'. Program terminated with signal 6, Aborted.

#0  0xb77a7424 in __kernel_vsyscall ()
#0  0xb77a7424 in __kernel_vsyscall ()
#1  0x083bdf35 in kill ()
#2  0x0807296d in uml_abort () at arch/um/os-Linux/util.c:93
#3  0x08072ca5 in os_dump_core () at arch/um/os-Linux/util.c:148
#4  0x080623c4 in panic_exit (self=0x85c1558 <panic_exit_notifier>, unused1=0, unused2=0x85f76e0 <buf.16221>) at arch/um/kernel/um_arch.c:240
#5  0x0809ba86 in notifier_call_chain (nl=0x0, val=0, v=0x85f76e0 <buf.16221>, nr_to_call=-2, nr_calls=0x0) at kernel/notifier.c:93
#6  0x0809bba1 in __atomic_notifier_call_chain (nh=0x85f76c4 <panic_notifier_list>, val=0, v=0x85f76e0 <buf.16221>, nr_to_call=0, nr_calls=0x0) at kernel/notifier.c:182
#7  0x0809bbdf in atomic_notifier_call_chain (nh=0x0, val=0, v=0x0) at kernel/notifier.c:191
#8  0x0841b5bc in panic (fmt=0x0) at kernel/panic.c:130
#9  0x0841c470 in balance_dirty_pages (pages_dirtied=23, mapping=<optimized out>) at mm/page-writeback.c:1467

#10 0x080d3595 in balance_dirty_pages_ratelimited (mapping=0x6) at mm/page-writeback.c:1661
#11 0x080e4a3f in __do_fault (mm=0x45ba3600, vma=0x48777b90, address=1084678144, pmd=0x0, pgoff=0, flags=0, orig_pte=<incomplete type>) at mm/memory.c:3452
#12 0x080e6e0f in do_linear_fault (orig_pte=..., flags=<optimized out>, pmd=<optimized out>, address=<optimized out>, vma=<optimized out>, mm=<optimized out>, page_table=<optimized out>) at mm/memory.c:3486 #13 handle_pte_fault (flags=<optimized out>, pmd=<optimized out>, pte=<optimized out>, address=<optimized out>, vma=<optimized out>, mm=<optimized out>) at mm/memory.c:3710 #14 __handle_mm_fault (flags=<optimized out>, address=<optimized out>, vma=<optimized out>, mm=<optimized out>) at mm/memory.c:3845
#15 handle_mm_fault (mm=0x45ba3600, vma=0x487034c8, address=1084678144, flags=1) at mm/memory.c:3868
#16 0x080e7817 in __get_user_pages (tsk=0x48705800, mm=0x45ba3600, start=1084678144, nr_pages=1025, gup_flags=519, pages=0x48558000, vmas=0x0, nonblocking=0x0) at mm/memory.c:1822 #17 0x080e7ae3 in get_user_pages (tsk=0x0, mm=0x0, start=0, nr_pages=0, write=1, force=0, pages=0x48777b90, vmas=0x6) at mm/memory.c:2019
#18 0x08143dc6 in aio_setup_ring (ctx=<optimized out>) at fs/aio.c:340
#19 ioctx_alloc (nr_events=<optimized out>) at fs/aio.c:605
#20 SYSC_io_setup (ctxp=<optimized out>, nr_events=<optimized out>) at fs/aio.c:1122
#21 SyS_io_setup (nr_events=65535, ctxp=135081984) at fs/aio.c:1105
#22 0x08062984 in handle_syscall (r=0x487059d4) at arch/um/kernel/skas/syscall.c:35
#23 0x08074fb5 in handle_trap (local_using_sysemu=<optimized out>, regs=<optimized out>, pid=<optimized out>) at arch/um/os-Linux/skas/process.c:198
#24 userspace (regs=0x487059d4) at arch/um/os-Linux/skas/process.c:431
#25 0x0805f750 in fork_handler () at arch/um/kernel/process.c:160
#26 0x00000000 in ?? ()

Quoted 11 lines

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds

CC'ing mm folks.
Please see below.

Am 09.10.2013 19:26, schrieb Toralf Förster:

Quoted 33 lines · Geert Uytterhoeven

Quoted 32 lines · Toralf Förster

Quoted 13 lines

Quoted 13 lines

Hmm, now pages_dirtied is zero, according to the backtrace, but the BUG_ON() asserts its strict positive?!?

Can you please try the following instead of the BUG_ON():

if (pause < 0) {
        printk("pages_dirtied = %lu\n", pages_dirtied);
        printk("task_ratelimit = %lu\n", task_ratelimit);
        printk("pause = %ld\n", pause);
}

Gr{oetje,eeting}s,

Quoted 8 lines

Quoted 1 line

Geert

I tried it in different ways already - I'm completely unsuccessful in getting any printk output. As soon as the issue happens I do have a

BUG: soft lockup - CPU#0 stuck for 22s! [trinity-child0:1521]

at stderr of the UML and then no further input is accepted. With uml_mconsole I'm however able to run very basic commands like a crash dump, sysrq ond so on.

You may get an idea of the magnitude of pages_dirtied by using a chain of BUG_ON()s, like:

BUG_ON(pages_dirtied > 2000000000);
BUG_ON(pages_dirtied > 1000000000);
BUG_ON(pages_dirtied > 100000000);
BUG_ON(pages_dirtied > 10000000);
BUG_ON(pages_dirtied > 1000000);

Quoted 5 lines

Quoted 2 lines

Probably 1 million is already too much for normal operation?

period = HZ * pages_dirtied / task_ratelimit;
		BUG_ON(pages_dirtied > 2000000000);
		BUG_ON(pages_dirtied > 1000000000);      <-------------- this is line 1467

Summary for mm people:

Toralf runs trinty on UML/i386.
After some time pages_dirtied becomes very large.
More than 1000000000 pages in this case.

Thus, period = HZ * pages_dirtied / task_ratelimit overflows and period/pause becomes extremely large.

It looks like io_schedule_timeout() get's called with a very large timeout. I don't know why "if (unlikely(pause > max_pause)) {" does not help.

Quoted 34 lines

the back trace is :

tfoerste@n22 ~/devel/linux $ gdb --core=/mnt/ramdisk/core /home/tfoerste/devel/linux/linux -batch -ex bt [New LWP 6911]
Core was generated by `/home/tfoerste/devel/linux/linux earlyprintk ubda=/home/tfoerste/virtual/uml/tr'. Program terminated with signal 6, Aborted.

#0  0xb77a7424 in __kernel_vsyscall ()
#0  0xb77a7424 in __kernel_vsyscall ()
#1  0x083bdf35 in kill ()
#2  0x0807296d in uml_abort () at arch/um/os-Linux/util.c:93
#3  0x08072ca5 in os_dump_core () at arch/um/os-Linux/util.c:148
#4  0x080623c4 in panic_exit (self=0x85c1558 <panic_exit_notifier>, unused1=0, unused2=0x85f76e0 <buf.16221>) at arch/um/kernel/um_arch.c:240
#5  0x0809ba86 in notifier_call_chain (nl=0x0, val=0, v=0x85f76e0 <buf.16221>, nr_to_call=-2, nr_calls=0x0) at kernel/notifier.c:93
#6  0x0809bba1 in __atomic_notifier_call_chain (nh=0x85f76c4 <panic_notifier_list>, val=0, v=0x85f76e0 <buf.16221>, nr_to_call=0, nr_calls=0x0) at kernel/notifier.c:182
#7  0x0809bbdf in atomic_notifier_call_chain (nh=0x0, val=0, v=0x0) at kernel/notifier.c:191
#8  0x0841b5bc in panic (fmt=0x0) at kernel/panic.c:130
#9  0x0841c470 in balance_dirty_pages (pages_dirtied=23, mapping=<optimized out>) at mm/page-writeback.c:1467

#10 0x080d3595 in balance_dirty_pages_ratelimited (mapping=0x6) at mm/page-writeback.c:1661
#11 0x080e4a3f in __do_fault (mm=0x45ba3600, vma=0x48777b90, address=1084678144, pmd=0x0, pgoff=0, flags=0, orig_pte=<incomplete type>) at mm/memory.c:3452
#12 0x080e6e0f in do_linear_fault (orig_pte=..., flags=<optimized out>, pmd=<optimized out>, address=<optimized out>, vma=<optimized out>, mm=<optimized out>, page_table=<optimized out>) at mm/memory.c:3486 #13 handle_pte_fault (flags=<optimized out>, pmd=<optimized out>, pte=<optimized out>, address=<optimized out>, vma=<optimized out>, mm=<optimized out>) at mm/memory.c:3710 #14 __handle_mm_fault (flags=<optimized out>, address=<optimized out>, vma=<optimized out>, mm=<optimized out>) at mm/memory.c:3845
#15 handle_mm_fault (mm=0x45ba3600, vma=0x487034c8, address=1084678144, flags=1) at mm/memory.c:3868
#16 0x080e7817 in __get_user_pages (tsk=0x48705800, mm=0x45ba3600, start=1084678144, nr_pages=1025, gup_flags=519, pages=0x48558000, vmas=0x0, nonblocking=0x0) at mm/memory.c:1822 #17 0x080e7ae3 in get_user_pages (tsk=0x0, mm=0x0, start=0, nr_pages=0, write=1, force=0, pages=0x48777b90, vmas=0x6) at mm/memory.c:2019
#18 0x08143dc6 in aio_setup_ring (ctx=<optimized out>) at fs/aio.c:340
#19 ioctx_alloc (nr_events=<optimized out>) at fs/aio.c:605
#20 SYSC_io_setup (ctxp=<optimized out>, nr_events=<optimized out>) at fs/aio.c:1122
#21 SyS_io_setup (nr_events=65535, ctxp=135081984) at fs/aio.c:1105
#22 0x08062984 in handle_syscall (r=0x487059d4) at arch/um/kernel/skas/syscall.c:35
#23 0x08074fb5 in handle_trap (local_using_sysemu=<optimized out>, regs=<optimized out>, pid=<optimized out>) at arch/um/os-Linux/skas/process.c:198
#24 userspace (regs=0x487059d4) at arch/um/os-Linux/skas/process.c:431
#25 0x0805f750 in fork_handler () at arch/um/kernel/process.c:160
#26 0x00000000 in ?? ()

Quoted 2 lines

CC'ing mm folks.
Please see below.

Added Fenguang to CC since he is the author of this code.

Quoted 40 lines

Am 09.10.2013 19:26, schrieb Toralf F�rster:

Quoted 33 lines · Geert Uytterhoeven

Quoted 32 lines · Toralf F�rster

Quoted 13 lines

Quoted 13 lines

Hmm, now pages_dirtied is zero, according to the backtrace, but the BUG_ON() asserts its strict positive?!?

Can you please try the following instead of the BUG_ON():

if (pause < 0) {
        printk("pages_dirtied = %lu\n", pages_dirtied);
        printk("task_ratelimit = %lu\n", task_ratelimit);
        printk("pause = %ld\n", pause);
}

Gr{oetje,eeting}s,

Quoted 8 lines

Quoted 1 line

Geert

I tried it in different ways already - I'm completely unsuccessful in getting any printk output. As soon as the issue happens I do have a

BUG: soft lockup - CPU#0 stuck for 22s! [trinity-child0:1521]

at stderr of the UML and then no further input is accepted. With uml_mconsole I'm however able to run very basic commands like a crash dump, sysrq ond so on.

You may get an idea of the magnitude of pages_dirtied by using a chain of BUG_ON()s, like:

BUG_ON(pages_dirtied > 2000000000);
BUG_ON(pages_dirtied > 1000000000);
BUG_ON(pages_dirtied > 100000000);
BUG_ON(pages_dirtied > 10000000);
BUG_ON(pages_dirtied > 1000000);

Quoted 5 lines

Quoted 2 lines

Probably 1 million is already too much for normal operation?

period = HZ * pages_dirtied / task_ratelimit;
		BUG_ON(pages_dirtied > 2000000000);
		BUG_ON(pages_dirtied > 1000000000);      <-------------- this is line 1467

Quoted 5 lines

Summary for mm people:

Toralf runs trinty on UML/i386.
After some time pages_dirtied becomes very large.
More than 1000000000 pages in this case.

Huh, this is really strange. pages_dirtied is passed into
balance_dirty_pages() from current->nr_dirtied. So I wonder how a value over 10^9 can get there. After all that is over 4TB so I somewhat doubt the task was ever able to dirty that much during its lifetime (but correct me if I'm wrong here, with UML and memory backed disks it is not totally impossible)... I went through the logic of handling ->nr_dirtied but I didn't find any obvious problem there. Hum, maybe one thing - what 'task_ratelimit' values do you see in balance_dirty_pages? If that one was huge, we could possibly accumulate huge current->nr_dirtied.

Quoted 2 lines · Richard Weinberger

CC'ing mm folks.
Please see below.

Added Fenguang to CC since he is the author of this code.

Quoted 40 lines

Am 09.10.2013 19:26, schrieb Toralf Förster:

Quoted 38 lines · Geert Uytterhoeven

Quoted 26 lines · Toralf Förster

Quoted 13 lines

Quoted 13 lines

Hmm, now pages_dirtied is zero, according to the backtrace, but the BUG_ON() asserts its strict positive?!?

Can you please try the following instead of the BUG_ON():

if (pause < 0) {
        printk("pages_dirtied = %lu\n", pages_dirtied);
        printk("task_ratelimit = %lu\n", task_ratelimit);
        printk("pause = %ld\n", pause);
}

Gr{oetje,eeting}s,

Quoted 8 lines

Quoted 1 line

Geert

I tried it in different ways already - I'm completely unsuccessful in getting any printk output. As soon as the issue happens I do have a

BUG: soft lockup - CPU#0 stuck for 22s! [trinity-child0:1521]

at stderr of the UML and then no further input is accepted. With uml_mconsole I'm however able to run very basic commands like a crash dump, sysrq ond so on.

You may get an idea of the magnitude of pages_dirtied by using a chain of BUG_ON()s, like:

Quoted 8 lines

BUG_ON(pages_dirtied > 2000000000);
BUG_ON(pages_dirtied > 1000000000);
BUG_ON(pages_dirtied > 100000000);
BUG_ON(pages_dirtied > 10000000);
BUG_ON(pages_dirtied > 1000000);

Probably 1 million is already too much for normal operation?

period = HZ * pages_dirtied / task_ratelimit;
		BUG_ON(pages_dirtied > 2000000000);
		BUG_ON(pages_dirtied > 1000000000);      <-------------- this is line 1467

Quoted 5 lines

Summary for mm people:

Toralf runs trinty on UML/i386.
After some time pages_dirtied becomes very large.
More than 1000000000 pages in this case.

Huh, this is really strange. pages_dirtied is passed into
balance_dirty_pages() from current->nr_dirtied. So I wonder how a value over 10^9 can get there.

After all that is over 4TB so I somewhat doubt the
task was ever able to dirty that much during its lifetime (but correct me if I'm wrong here, with UML and memory backed disks it is not totally impossible)... I went through the logic of handling ->nr_dirtied but I didn't find any obvious problem there. Hum, maybe one thing - what 'task_ratelimit' values do you see in balance_dirty_pages? If that one was huge, we could possibly accumulate huge current->nr_dirtied.

Quoted 2 lines

Thus, period = HZ * pages_dirtied / task_ratelimit overflows and period/pause becomes extremely large.

Quoted 2 lines

It looks like io_schedule_timeout() get's called with a very large timeout. I don't know why "if (unlikely(pause > max_pause)) {" does not help.

Quoted 27 lines

Quoted 27 lines

the back trace is :

tfoerste@n22 ~/devel/linux $ gdb --core=/mnt/ramdisk/core /home/tfoerste/devel/linux/linux -batch -ex bt [New LWP 6911]
Core was generated by `/home/tfoerste/devel/linux/linux earlyprintk ubda=/home/tfoerste/virtual/uml/tr'. Program terminated with signal 6, Aborted.

#0  0xb77a7424 in __kernel_vsyscall ()
#0  0xb77a7424 in __kernel_vsyscall ()
#1  0x083bdf35 in kill ()
#2  0x0807296d in uml_abort () at arch/um/os-Linux/util.c:93
#3  0x08072ca5 in os_dump_core () at arch/um/os-Linux/util.c:148
#4  0x080623c4 in panic_exit (self=0x85c1558 <panic_exit_notifier>, unused1=0, unused2=0x85f76e0 <buf.16221>) at arch/um/kernel/um_arch.c:240
#5  0x0809ba86 in notifier_call_chain (nl=0x0, val=0, v=0x85f76e0 <buf.16221>, nr_to_call=-2, nr_calls=0x0) at kernel/notifier.c:93
#6  0x0809bba1 in __atomic_notifier_call_chain (nh=0x85f76c4 <panic_notifier_list>, val=0, v=0x85f76e0 <buf.16221>, nr_to_call=0, nr_calls=0x0) at kernel/notifier.c:182
#7  0x0809bbdf in atomic_notifier_call_chain (nh=0x0, val=0, v=0x0) at kernel/notifier.c:191
#8  0x0841b5bc in panic (fmt=0x0) at kernel/panic.c:130
#9  0x0841c470 in balance_dirty_pages (pages_dirtied=23, mapping=<optimized out>) at mm/page-writeback.c:1467

#10 0x080d3595 in balance_dirty_pages_ratelimited (mapping=0x6) at mm/page-writeback.c:1661
#11 0x080e4a3f in __do_fault (mm=0x45ba3600, vma=0x48777b90, address=1084678144, pmd=0x0, pgoff=0, flags=0, orig_pte=<incomplete type>) at mm/memory.c:3452
#12 0x080e6e0f in do_linear_fault (orig_pte=..., flags=<optimized out>, pmd=<optimized out>, address=<optimized out>, vma=<optimized out>, mm=<optimized out>, page_table=<optimized out>) at mm/memory.c:3486 #13 handle_pte_fault (flags=<optimized out>, pmd=<optimized out>, pte=<optimized out>, address=<optimized out>, vma=<optimized out>, mm=<optimized out>) at mm/memory.c:3710 #14 __handle_mm_fault (flags=<optimized out>, address=<optimized out>, vma=<optimized out>, mm=<optimized out>) at mm/memory.c:3845
#15 handle_mm_fault (mm=0x45ba3600, vma=0x487034c8, address=1084678144, flags=1) at mm/memory.c:3868
#16 0x080e7817 in __get_user_pages (tsk=0x48705800, mm=0x45ba3600, start=1084678144, nr_pages=1025, gup_flags=519, pages=0x48558000, vmas=0x0, nonblocking=0x0) at mm/memory.c:1822 #17 0x080e7ae3 in get_user_pages (tsk=0x0, mm=0x0, start=0, nr_pages=0, write=1, force=0, pages=0x48777b90, vmas=0x6) at mm/memory.c:2019
#18 0x08143dc6 in aio_setup_ring (ctx=<optimized out>) at fs/aio.c:340

Quoted 8 lines

Quoted 8 lines

#19 ioctx_alloc (nr_events=<optimized out>) at fs/aio.c:605
#20 SYSC_io_setup (ctxp=<optimized out>, nr_events=<optimized out>) at fs/aio.c:1122
#21 SyS_io_setup (nr_events=65535, ctxp=135081984) at fs/aio.c:1105
#22 0x08062984 in handle_syscall (r=0x487059d4) at arch/um/kernel/skas/syscall.c:35
#23 0x08074fb5 in handle_trap (local_using_sysemu=<optimized out>, regs=<optimized out>, pid=<optimized out>) at arch/um/os-Linux/skas/process.c:198 #24 userspace (regs=0x487059d4) at arch/um/os-Linux/skas/process.c:431
#25 0x0805f750 in fork_handler () at arch/um/kernel/process.c:160
#26 0x00000000 in ?? ()

Honza

Signature · 2 lines

Jan Kara <[email protected]>
SUSE Labs, CR

Quoted 42 lines · Jan Kara

Quoted 41 lines · Richard Weinberger

Am 09.10.2013 19:26, schrieb Toralf Förster:

Quoted 29 lines · Geert Uytterhoeven

Quoted 28 lines · Toralf Förster

Quoted 10 lines

Quoted 9 lines

Hmm, now pages_dirtied is zero, according to the backtrace, but the BUG_ON() asserts its strict positive?!?

Can you please try the following instead of the BUG_ON():

if (pause < 0) {
        printk("pages_dirtied = %lu\n", pages_dirtied);
        printk("task_ratelimit = %lu\n", task_ratelimit);
        printk("pause = %ld\n", pause);

Quoted 7 lines

I tried it in different ways already - I'm completely unsuccessful in getting any printk output. As soon as the issue happens I do have a

BUG: soft lockup - CPU#0 stuck for 22s! [trinity-child0:1521]

at stderr of the UML and then no further input is accepted. With uml_mconsole I'm however able to run very basic commands like a crash dump, sysrq ond so on.

You may get an idea of the magnitude of pages_dirtied by using a chain of BUG_ON()s, like:

BUG_ON(pages_dirtied > 2000000000);
BUG_ON(pages_dirtied > 1000000000);
BUG_ON(pages_dirtied > 100000000);
BUG_ON(pages_dirtied > 10000000);
BUG_ON(pages_dirtied > 1000000);

Quoted 5 lines

Quoted 2 lines

Probably 1 million is already too much for normal operation?

period = HZ * pages_dirtied / task_ratelimit;
          BUG_ON(pages_dirtied > 2000000000);
          BUG_ON(pages_dirtied > 1000000000);      <-------------- this is line 1467

Summary for mm people:

Toralf runs trinty on UML/i386.
After some time pages_dirtied becomes very large.
More than 1000000000 pages in this case.

Quoted 3 lines

Huh, this is really strange. pages_dirtied is passed into
balance_dirty_pages() from current->nr_dirtied. So I wonder how a value over 10^9 can get there.

I noticed aio_setup_ring() in the call trace and find it recently added a SetPageDirty() call in a loop by commit 36bc08cc01 ("fs/aio:
Add support to aio ring pages migration"). So added CC to its authors.

Quoted 10 lines

After all that is over 4TB so I somewhat doubt the
task was ever able to dirty that much during its lifetime (but correct me if I'm wrong here, with UML and memory backed disks it is not totally impossible)... I went through the logic of handling ->nr_dirtied but I didn't find any obvious problem there. Hum, maybe one thing - what 'task_ratelimit' values do you see in balance_dirty_pages? If that one was huge, we could possibly accumulate huge current->nr_dirtied.

Quoted 2 lines

Thus, period = HZ * pages_dirtied / task_ratelimit overflows and period/pause becomes extremely large.

Quoted 2 lines

Quoted 2 lines

It looks like io_schedule_timeout() get's called with a very large timeout. I don't know why "if (unlikely(pause > max_pause)) {" does not help.

Quoted 59 lines · Fengguang Wu

Quoted 37 lines · Jan Kara

Quoted 36 lines · Richard Weinberger

Am 09.10.2013 19:26, schrieb Toralf Förster:

Quoted 34 lines · Geert Uytterhoeven

Quoted 22 lines · Toralf Förster

Quoted 10 lines

Quoted 9 lines

Hmm, now pages_dirtied is zero, according to the backtrace, but the BUG_ON() asserts its strict positive?!?

Can you please try the following instead of the BUG_ON():

if (pause < 0) {
        printk("pages_dirtied = %lu\n", pages_dirtied);
        printk("task_ratelimit = %lu\n", task_ratelimit);
        printk("pause = %ld\n", pause);

Quoted 7 lines

I tried it in different ways already - I'm completely unsuccessful in getting any printk output. As soon as the issue happens I do have a

BUG: soft lockup - CPU#0 stuck for 22s! [trinity-child0:1521]

at stderr of the UML and then no further input is accepted. With uml_mconsole I'm however able to run very basic commands like a crash dump, sysrq ond so on.

You may get an idea of the magnitude of pages_dirtied by using a chain of BUG_ON()s, like:

Quoted 8 lines

BUG_ON(pages_dirtied > 2000000000);
BUG_ON(pages_dirtied > 1000000000);
BUG_ON(pages_dirtied > 100000000);
BUG_ON(pages_dirtied > 10000000);
BUG_ON(pages_dirtied > 1000000);

Probably 1 million is already too much for normal operation?

period = HZ * pages_dirtied / task_ratelimit;
          BUG_ON(pages_dirtied > 2000000000);
          BUG_ON(pages_dirtied > 1000000000);      <-------------- this is line 1467

Quoted 8 lines

Quoted 5 lines

Summary for mm people:

Toralf runs trinty on UML/i386.
After some time pages_dirtied becomes very large.
More than 1000000000 pages in this case.

Huh, this is really strange. pages_dirtied is passed into
balance_dirty_pages() from current->nr_dirtied. So I wonder how a value over 10^9 can get there.

I noticed aio_setup_ring() in the call trace and find it recently added a SetPageDirty() call in a loop by commit 36bc08cc01 ("fs/aio:
Add support to aio ring pages migration"). So added CC to its authors.

Quoted 8 lines

After all that is over 4TB so I somewhat doubt the
task was ever able to dirty that much during its lifetime (but correct me if I'm wrong here, with UML and memory backed disks it is not totally impossible)... I went through the logic of handling ->nr_dirtied but I didn't find any obvious problem there. Hum, maybe one thing - what 'task_ratelimit' values do you see in balance_dirty_pages? If that one was huge, we could possibly accumulate huge current->nr_dirtied.

Quoted 2 lines

Quoted 2 lines

Quoted 2 lines

Thus, period = HZ * pages_dirtied / task_ratelimit overflows and period/pause becomes extremely large.

period/pause are signed long, so they become negative instead of extremely large when overflowing.

Quoted 2 lines

Quoted 2 lines

Quoted 2 lines

It looks like io_schedule_timeout() get's called with a very large timeout. I don't know why "if (unlikely(pause > max_pause)) {" does not help.

Because pause is now negative.

Am 09.10.2013 23:47, schrieb Jan Kara:

Quoted 4 lines · Richard Weinberger

Quoted 2 lines

CC'ing mm folks.
Please see below.

Added Fenguang to CC since he is the author of this code.

Thx, get_maintainer.pl didn't list him.

Quoted 45 lines

Quoted 40 lines

Am 09.10.2013 19:26, schrieb Toralf Förster:

Quoted 33 lines · Geert Uytterhoeven

Quoted 32 lines · Toralf Förster

Quoted 13 lines

Quoted 13 lines

Hmm, now pages_dirtied is zero, according to the backtrace, but the BUG_ON() asserts its strict positive?!?

Can you please try the following instead of the BUG_ON():

if (pause < 0) {
        printk("pages_dirtied = %lu\n", pages_dirtied);
        printk("task_ratelimit = %lu\n", task_ratelimit);
        printk("pause = %ld\n", pause);
}

Gr{oetje,eeting}s,

Quoted 8 lines

Quoted 1 line

Geert

I tried it in different ways already - I'm completely unsuccessful in getting any printk output. As soon as the issue happens I do have a

BUG: soft lockup - CPU#0 stuck for 22s! [trinity-child0:1521]

at stderr of the UML and then no further input is accepted. With uml_mconsole I'm however able to run very basic commands like a crash dump, sysrq ond so on.

You may get an idea of the magnitude of pages_dirtied by using a chain of BUG_ON()s, like:

BUG_ON(pages_dirtied > 2000000000);
BUG_ON(pages_dirtied > 1000000000);
BUG_ON(pages_dirtied > 100000000);
BUG_ON(pages_dirtied > 10000000);
BUG_ON(pages_dirtied > 1000000);

Quoted 5 lines

Quoted 2 lines

Probably 1 million is already too much for normal operation?

period = HZ * pages_dirtied / task_ratelimit;
		BUG_ON(pages_dirtied > 2000000000);
		BUG_ON(pages_dirtied > 1000000000);      <-------------- this is line 1467

Quoted 5 lines

Summary for mm people:

Toralf runs trinty on UML/i386.
After some time pages_dirtied becomes very large.
More than 1000000000 pages in this case.

Quoted 9 lines

Huh, this is really strange. pages_dirtied is passed into
balance_dirty_pages() from current->nr_dirtied. So I wonder how a value over 10^9 can get there. After all that is over 4TB so I somewhat doubt the task was ever able to dirty that much during its lifetime (but correct me if I'm wrong here, with UML and memory backed disks it is not totally impossible)... I went through the logic of handling ->nr_dirtied but I didn't find any obvious problem there. Hum, maybe one thing - what 'task_ratelimit' values do you see in balance_dirty_pages? If that one was huge, we could possibly accumulate huge current->nr_dirtied.

Toralf, you can try a snipplet like this one to get the values printed out:

mm/page-writeback.c +6 -0

index f5236f8..a80e520 100644
@@ -1463,6 +1463,12 @@ static void balance_dirty_pages(struct address_space *mapping,
                        goto pause;
                }
                period = HZ * pages_dirtied / task_ratelimit;
+
+               {
+                       extern int printf(char *, ...);
+                       printf("---> task_ratelimit: %lu\n", task_ratelimit);
+               }
+
                pause = period;
                if (current->dirty_paused_when)
                        pause -= now - current->dirty_paused_when;

Yes, printf(), not printk().
Using this hack we print directly to host's stdout. :)

Thanks,
//richard

Quoted 83 lines · Richard Weinberger

Am 09.10.2013 23:47, schrieb Jan Kara:

Quoted 4 lines · Richard Weinberger

Quoted 2 lines

CC'ing mm folks.
Please see below.

Added Fenguang to CC since he is the author of this code.

Thx, get_maintainer.pl didn't list him.

Quoted 45 lines

Quoted 40 lines

Am 09.10.2013 19:26, schrieb Toralf Förster:

Quoted 33 lines · Geert Uytterhoeven

Quoted 32 lines · Toralf Förster

Quoted 13 lines

Quoted 13 lines

Hmm, now pages_dirtied is zero, according to the backtrace, but the BUG_ON() asserts its strict positive?!?

Can you please try the following instead of the BUG_ON():

if (pause < 0) {
        printk("pages_dirtied = %lu\n", pages_dirtied);
        printk("task_ratelimit = %lu\n", task_ratelimit);
        printk("pause = %ld\n", pause);
}

Gr{oetje,eeting}s,

Quoted 8 lines

Quoted 1 line

Geert

I tried it in different ways already - I'm completely unsuccessful in getting any printk output. As soon as the issue happens I do have a

BUG: soft lockup - CPU#0 stuck for 22s! [trinity-child0:1521]

at stderr of the UML and then no further input is accepted. With uml_mconsole I'm however able to run very basic commands like a crash dump, sysrq ond so on.

You may get an idea of the magnitude of pages_dirtied by using a chain of BUG_ON()s, like:

BUG_ON(pages_dirtied > 2000000000);
BUG_ON(pages_dirtied > 1000000000);
BUG_ON(pages_dirtied > 100000000);
BUG_ON(pages_dirtied > 10000000);
BUG_ON(pages_dirtied > 1000000);

Quoted 5 lines

Quoted 2 lines

Probably 1 million is already too much for normal operation?

period = HZ * pages_dirtied / task_ratelimit;
		BUG_ON(pages_dirtied > 2000000000);
		BUG_ON(pages_dirtied > 1000000000);      <-------------- this is line 1467

Quoted 5 lines

Summary for mm people:

Toralf runs trinty on UML/i386.
After some time pages_dirtied becomes very large.
More than 1000000000 pages in this case.

Quoted 9 lines

Huh, this is really strange. pages_dirtied is passed into
balance_dirty_pages() from current->nr_dirtied. So I wonder how a value over 10^9 can get there. After all that is over 4TB so I somewhat doubt the task was ever able to dirty that much during its lifetime (but correct me if I'm wrong here, with UML and memory backed disks it is not totally impossible)... I went through the logic of handling ->nr_dirtied but I didn't find any obvious problem there. Hum, maybe one thing - what 'task_ratelimit' values do you see in balance_dirty_pages? If that one was huge, we could possibly accumulate huge current->nr_dirtied.

Toralf, you can try a snipplet like this one to get the values printed out:

mm/page-writeback.c +6 -0

index f5236f8..a80e520 100644
@@ -1463,6 +1463,12 @@ static void balance_dirty_pages(struct address_space *mapping,
                        goto pause;
                }
                period = HZ * pages_dirtied / task_ratelimit;
+
+               {
+                       extern int printf(char *, ...);
+                       printf("---> task_ratelimit: %lu\n", task_ratelimit);
+               }
+
                pause = period;
                if (current->dirty_paused_when)
                        pause -= now - current->dirty_paused_when;

Quoted 3 lines

Yes, printf(), not printk().
Using this hack we print directly to host's stdout. :)

*head smack* ofc - works fine.
So given this diff :

iff --git a/mm/page-writeback.c b/mm/page-writeback.c index f5236f8..5a2c337 100644

mm/page-writeback.c +7 -0

@@ -1464,6 +1464,13 @@ static void balance_dirty_pages(struct address_space *mapping,
                }
                period = HZ * pages_dirtied / task_ratelimit;
                pause = period;
+               if (pause < 0)  {
+                       extern int printf(char *, ...);
+                       printf("overflow : pause : %li\n", pause);
+                       printf("overflow : pages_dirtied : %lu\n", pages_dirtied);
+                       printf("overflow :  task_ratelimit: %lu\n", task_ratelimit);
+                       BUG_ON(1);
+               }
                if (current->dirty_paused_when)
                        pause -= now - current->dirty_paused_when;
                /*

mm/page-writeback.c +6 -0

+                       extern int printf(char *, ...);
+                       printf("ick : pause : %li\n", pause);
+                       printf("ick: pages_dirtied : %lu\n", pages_dirtied);
+                       printf("ick: task_ratelimit: %lu\n", task_ratelimit);
+                       BUG_ON(1);
+               }
                trace_balance_dirty_pages(bdi,
                                          dirty_thresh,
                                          background_thresh,

I got this :

* Starting local
net.core.warnings = 0 [ ok ]
ick : pause : -984
ick: pages_dirtied : 0
ick: task_ratelimit: 0
Kernel panic - not syncing: BUG!

CPU: 0 PID: 1434 Comm: trinity-child2 Not tainted 3.12.0-rc4-00029-g0e7a3ed-dirty #12
47397c84 47397cb0 0841b5a0 084c30e8 085f76e0 084b4745 47397cbc 00000000 
       fffffc28 01fff278 085cb4a0 47397d2c 0841c5a0 084b4745 084c5398 000005ee 
       08432cf0 43e47600 471757f8 47397cf0 ffffff0c 47397cdc 5256d8b9 3417ec18 47397c5c:  [<08060b2c>] show_stack+0x7c/0xd0
47397c7c:  [<0841e34d>] dump_stack+0x26/0x28
47397c8c:  [<0841b5a0>] panic+0x7a/0x180
47397cb4:  [<0841c5a0>] balance_dirty_pages.isra.32+0x4e3/0x5ad
47397d30:  [<080d3595>] balance_dirty_pages_ratelimited+0xf5/0x100
47397d44:  [<080e4a3f>] __do_fault+0x3cf/0x440
47397d9c:  [<080e6e0f>] handle_mm_fault+0xef/0x7c0
47397dec:  [<080e7817>] __get_user_pages+0x227/0x420
47397e24:  [<080e7ae3>] get_user_pages+0x63/0x70
47397e4c:  [<08143dc6>] SyS_io_setup+0x3c6/0x760
47397eb0:  [<08062984>] handle_syscall+0x64/0x80
47397ef0:  [<08074fb5>] userspace+0x475/0x5f0
47397fec:  [<0805f750>] fork_handler+0x60/0x70
47397ffc:  [<00000000>] 0x0

EIP: 0073:[<40001282>] CPU: 0 Not tainted ESP: 007b:bfb348f8 EFLAGS: 00000246
Not tainted
EAX: ffffffda EBX: 00001000 ECX: 080d0000 EDX: 80000048
ESI: 80fbff1f EDI: ffe02f77 EBP: 90f6e2a3 DS: 007b ES: 007b

47397c0c:  [<0807947f>] show_regs+0x10f/0x120
47397c28:  [<080623a9>] panic_exit+0x29/0x50
47397c38:  [<0809ba86>] notifier_call_chain+0x36/0x60
47397c60:  [<0809bba1>] __atomic_notifier_call_chain+0x21/0x30
47397c70:  [<0809bbdf>] atomic_notifier_call_chain+0x2f/0x40
47397c8c:  [<0841b5bc>] panic+0x96/0x180
47397cb4:  [<0841c5a0>] balance_dirty_pages.isra.32+0x4e3/0x5ad
47397d30:  [<080d3595>] balance_dirty_pages_ratelimited+0xf5/0x100
47397d44:  [<080e4a3f>] __do_fault+0x3cf/0x440
47397d9c:  [<080e6e0f>] handle_mm_fault+0xef/0x7c0
47397dec:  [<080e7817>] __get_user_pages+0x227/0x420
47397e24:  [<080e7ae3>] get_user_pages+0x63/0x70
47397e4c:  [<08143dc6>] SyS_io_setup+0x3c6/0x760
47397eb0:  [<08062984>] handle_syscall+0x64/0x80
47397ef0:  [<08074fb5>] userspace+0x475/0x5f0
47397fec:  [<0805f750>] fork_handler+0x60/0x70
47397ffc:  [<00000000>] 0x0

/home/tfoerste/workspace/bin/start_uml.sh: line 115: 18718 Aborted (core dumped) $LINUX earlyprintk ubda=$ROOTFS ubdb=$SWAP eth0=$NET mem=$MEM $TTY umid=uml_$NAME rootfstype=ext4 "$ARGS"

Quoted 5 lines

From what I see there are 2 different types of issues - and this is an example of the other of both

Thanks,
//richard

Signature · 3 lines

MfG/Sincerely
Toralf Förster
pgp finger print: 7B1A 07F4 EC82 0F90 D4C2 8936 872A E508 7DB6 9DA3

Quoted 109 lines · Toralf Förster

Quoted 83 lines · Richard Weinberger

Am 09.10.2013 23:47, schrieb Jan Kara:

Quoted 4 lines · Richard Weinberger

Quoted 2 lines

CC'ing mm folks.
Please see below.

Added Fenguang to CC since he is the author of this code.

Thx, get_maintainer.pl didn't list him.

Quoted 45 lines

Quoted 40 lines

Am 09.10.2013 19:26, schrieb Toralf Förster:

Quoted 33 lines · Geert Uytterhoeven

Quoted 32 lines · Toralf Förster

Quoted 13 lines

Quoted 13 lines

Hmm, now pages_dirtied is zero, according to the backtrace, but the BUG_ON() asserts its strict positive?!?

Can you please try the following instead of the BUG_ON():

if (pause < 0) {
        printk("pages_dirtied = %lu\n", pages_dirtied);
        printk("task_ratelimit = %lu\n", task_ratelimit);
        printk("pause = %ld\n", pause);
}

Gr{oetje,eeting}s,

Quoted 8 lines

Quoted 1 line

Geert

I tried it in different ways already - I'm completely unsuccessful in getting any printk output. As soon as the issue happens I do have a

BUG: soft lockup - CPU#0 stuck for 22s! [trinity-child0:1521]

at stderr of the UML and then no further input is accepted. With uml_mconsole I'm however able to run very basic commands like a crash dump, sysrq ond so on.

You may get an idea of the magnitude of pages_dirtied by using a chain of BUG_ON()s, like:

BUG_ON(pages_dirtied > 2000000000);
BUG_ON(pages_dirtied > 1000000000);
BUG_ON(pages_dirtied > 100000000);
BUG_ON(pages_dirtied > 10000000);
BUG_ON(pages_dirtied > 1000000);

Quoted 5 lines

Quoted 2 lines

Probably 1 million is already too much for normal operation?

period = HZ * pages_dirtied / task_ratelimit;
		BUG_ON(pages_dirtied > 2000000000);
		BUG_ON(pages_dirtied > 1000000000);      <-------------- this is line 1467

Quoted 5 lines

Summary for mm people:

Toralf runs trinty on UML/i386.
After some time pages_dirtied becomes very large.
More than 1000000000 pages in this case.

Quoted 9 lines

Huh, this is really strange. pages_dirtied is passed into
balance_dirty_pages() from current->nr_dirtied. So I wonder how a value over 10^9 can get there. After all that is over 4TB so I somewhat doubt the task was ever able to dirty that much during its lifetime (but correct me if I'm wrong here, with UML and memory backed disks it is not totally impossible)... I went through the logic of handling ->nr_dirtied but I didn't find any obvious problem there. Hum, maybe one thing - what 'task_ratelimit' values do you see in balance_dirty_pages? If that one was huge, we could possibly accumulate huge current->nr_dirtied.

Toralf, you can try a snipplet like this one to get the values printed out:

mm/page-writeback.c +6 -0

index f5236f8..a80e520 100644
@@ -1463,6 +1463,12 @@ static void balance_dirty_pages(struct address_space *mapping,
                        goto pause;
                }
                period = HZ * pages_dirtied / task_ratelimit;
+
+               {
+                       extern int printf(char *, ...);
+                       printf("---> task_ratelimit: %lu\n", task_ratelimit);
+               }
+
                pause = period;
                if (current->dirty_paused_when)
                        pause -= now - current->dirty_paused_when;

Quoted 3 lines

Yes, printf(), not printk().
Using this hack we print directly to host's stdout. :)

*head smack* ofc - works fine.
So given this diff :

iff --git a/mm/page-writeback.c b/mm/page-writeback.c index f5236f8..5a2c337 100644

mm/page-writeback.c +7 -0

@@ -1464,6 +1464,13 @@ static void balance_dirty_pages(struct address_space *mapping,
                }
                period = HZ * pages_dirtied / task_ratelimit;
                pause = period;
+               if (pause < 0)  {
+                       extern int printf(char *, ...);
+                       printf("overflow : pause : %li\n", pause);
+                       printf("overflow : pages_dirtied : %lu\n", pages_dirtied);
+                       printf("overflow :  task_ratelimit: %lu\n", task_ratelimit);
+                       BUG_ON(1);
+               }
                if (current->dirty_paused_when)
                        pause -= now - current->dirty_paused_when;
                /*

mm/page-writeback.c static void balance_dirty_pages(struct address_space *mapping, · line 1513

Quoted 4 lines

mm/page-writeback.c +1 -0

@@ -1503,6 +1510,13 @@ static void balance_dirty_pages(struct address_space *mapping,
                }
 
 pause:
+               if (pause < 0)  {

Oops, we got

        ick : pause : -984

here! Since pause is bounded by [min_pause, max_pause], it means something goes wrong with the bounds. Would you help print min_pause and max_pause as well?

It seems there are "long <=> unsigned long" conversion problems in bdi_max_pause(), which might make max_pause a negative number. And the min_pause calculation is based on max_pause to some degree..

Thanks,
Fengguang

Quoted 77 lines

mm/page-writeback.c +6 -0

+                       extern int printf(char *, ...);
+                       printf("ick : pause : %li\n", pause);
+                       printf("ick: pages_dirtied : %lu\n", pages_dirtied);
+                       printf("ick: task_ratelimit: %lu\n", task_ratelimit);
+                       BUG_ON(1);
+               }
                trace_balance_dirty_pages(bdi,
                                          dirty_thresh,
                                          background_thresh,

I got this :

* Starting local
net.core.warnings = 0 [ ok ]
ick : pause : -984
ick: pages_dirtied : 0
ick: task_ratelimit: 0
Kernel panic - not syncing: BUG!

CPU: 0 PID: 1434 Comm: trinity-child2 Not tainted 3.12.0-rc4-00029-g0e7a3ed-dirty #12
47397c84 47397cb0 0841b5a0 084c30e8 085f76e0 084b4745 47397cbc 00000000 
       fffffc28 01fff278 085cb4a0 47397d2c 0841c5a0 084b4745 084c5398 000005ee 
       08432cf0 43e47600 471757f8 47397cf0 ffffff0c 47397cdc 5256d8b9 3417ec18 47397c5c:  [<08060b2c>] show_stack+0x7c/0xd0
47397c7c:  [<0841e34d>] dump_stack+0x26/0x28
47397c8c:  [<0841b5a0>] panic+0x7a/0x180
47397cb4:  [<0841c5a0>] balance_dirty_pages.isra.32+0x4e3/0x5ad
47397d30:  [<080d3595>] balance_dirty_pages_ratelimited+0xf5/0x100
47397d44:  [<080e4a3f>] __do_fault+0x3cf/0x440
47397d9c:  [<080e6e0f>] handle_mm_fault+0xef/0x7c0
47397dec:  [<080e7817>] __get_user_pages+0x227/0x420
47397e24:  [<080e7ae3>] get_user_pages+0x63/0x70
47397e4c:  [<08143dc6>] SyS_io_setup+0x3c6/0x760
47397eb0:  [<08062984>] handle_syscall+0x64/0x80
47397ef0:  [<08074fb5>] userspace+0x475/0x5f0
47397fec:  [<0805f750>] fork_handler+0x60/0x70
47397ffc:  [<00000000>] 0x0

EIP: 0073:[<40001282>] CPU: 0 Not tainted ESP: 007b:bfb348f8 EFLAGS: 00000246
Not tainted
EAX: ffffffda EBX: 00001000 ECX: 080d0000 EDX: 80000048
ESI: 80fbff1f EDI: ffe02f77 EBP: 90f6e2a3 DS: 007b ES: 007b

47397c0c:  [<0807947f>] show_regs+0x10f/0x120
47397c28:  [<080623a9>] panic_exit+0x29/0x50
47397c38:  [<0809ba86>] notifier_call_chain+0x36/0x60
47397c60:  [<0809bba1>] __atomic_notifier_call_chain+0x21/0x30
47397c70:  [<0809bbdf>] atomic_notifier_call_chain+0x2f/0x40
47397c8c:  [<0841b5bc>] panic+0x96/0x180
47397cb4:  [<0841c5a0>] balance_dirty_pages.isra.32+0x4e3/0x5ad
47397d30:  [<080d3595>] balance_dirty_pages_ratelimited+0xf5/0x100
47397d44:  [<080e4a3f>] __do_fault+0x3cf/0x440
47397d9c:  [<080e6e0f>] handle_mm_fault+0xef/0x7c0
47397dec:  [<080e7817>] __get_user_pages+0x227/0x420
47397e24:  [<080e7ae3>] get_user_pages+0x63/0x70
47397e4c:  [<08143dc6>] SyS_io_setup+0x3c6/0x760
47397eb0:  [<08062984>] handle_syscall+0x64/0x80
47397ef0:  [<08074fb5>] userspace+0x475/0x5f0
47397fec:  [<0805f750>] fork_handler+0x60/0x70
47397ffc:  [<00000000>] 0x0

/home/tfoerste/workspace/bin/start_uml.sh: line 115: 18718 Aborted (core dumped) $LINUX earlyprintk ubda=$ROOTFS ubdb=$SWAP eth0=$NET mem=$MEM $TTY umid=uml_$NAME rootfstype=ext4 "$ARGS"

Quoted 5 lines

From what I see there are 2 different types of issues - and this is an example of the other of both

Thanks,
//richard

Signature · 3 lines

MfG/Sincerely
Toralf Förster
pgp finger print: 7B1A 07F4 EC82 0F90 D4C2 8936 872A E508 7DB6 9DA3

yeah, now the picture becomes more clear
...
net.core.warnings = 0 [ ok ]
ick: pause : -717
ick : min_pause : -177
ick : max_pause : -717
ick: pages_dirtied : 14
ick: task_ratelimit: 0

Quoted 8 lines · Toralf Förster

yeah, now the picture becomes more clear
...
net.core.warnings = 0 [ ok ]
ick: pause : -717
ick : min_pause : -177
ick : max_pause : -717
ick: pages_dirtied : 14
ick: task_ratelimit: 0

Great and thanks! So it's the max pause calculation went wrong.

Quoted 8 lines · Toralf Förster

yeah, now the picture becomes more clear
...
net.core.warnings = 0 [ ok ]
ick: pause : -717
ick : min_pause : -177
ick : max_pause : -717
ick: pages_dirtied : 14
ick: task_ratelimit: 0

Great and thanks! So it's the max pause calculation went wrong. Would help you try the below patch?

Quoted 1 line

Quoted 1 line

From 5420b9bbe42dd0a366d7615e9f3d3724cee725c4 Mon Sep 17 00:00:00 2001

From: Fengguang Wu <[email protected]>
Date: Fri, 11 Oct 2013 16:53:26 +0800
Subject: [PATCH] fix bdi max pause calculation

Signed-off-by Fengguang Wu <[email protected]>

---
 mm/page-writeback.c |   10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

mm/page-writeback.c +5 -5

index 3f0c895..241a746 100644
@@ -1104,11 +1104,11 @@ static unsigned long dirty_poll_interval(unsigned long dirty,
 	return 1;
 }
 
-static long bdi_max_pause(struct backing_dev_info *bdi,
-			  unsigned long bdi_dirty)
+static unsigned long bdi_max_pause(struct backing_dev_info *bdi,
+				   unsigned long bdi_dirty)
 {
-	long bw = bdi->avg_write_bandwidth;
-	long t;
+	unsigned long bw = bdi->avg_write_bandwidth;
+	unsigned long t;
 
 	/*
 	 * Limit pause time for small memory systems. If sleeping for too long

@@ -1120,7 +1120,7 @@ static long bdi_max_pause(struct backing_dev_info *bdi,
 	t = bdi_dirty / (1 + bw / roundup_pow_of_two(1 + HZ / 8));
 	t++;
 
-	return min_t(long, t, MAX_PAUSE);
+	return min_t(unsigned long, t, MAX_PAUSE);
 }
 
 static long bdi_min_pause(struct backing_dev_info *bdi,

Toralf runs trinity on UML/i386.
After some time it hangs and the last message line is

	BUG: soft lockup - CPU#0 stuck for 22s! [trinity-child0:1521]

It's found that pages_dirtied becomes very large.
More than 1000000000 pages in this case:

	period = HZ * pages_dirtied / task_ratelimit;
	BUG_ON(pages_dirtied > 2000000000);
	BUG_ON(pages_dirtied > 1000000000);      <---------

UML debug printf shows that we got negative pause here:

ick: pause : -984
ick: pages_dirtied : 0
ick: task_ratelimit: 0

pause:
+ if (pause < 0) {
+ extern int printf(char *, ...);
+ printf("ick : pause : %li\n", pause);
+ printf("ick: pages_dirtied : %lu\n", pages_dirtied);
+ printf("ick: task_ratelimit: %lu\n", task_ratelimit);
+ BUG_ON(1);
+ }
trace_balance_dirty_pages(bdi,

Since pause is bounded by [min_pause, max_pause] where min_pause is also bounded by max_pause. It's suspected and demonstrated that the max_pause calculation goes wrong:

	ick: pause : -717
	ick: min_pause : -177
	ick: max_pause : -717
	ick: pages_dirtied : 14
	ick: task_ratelimit: 0

The problem lies in the two "long = unsigned long" assignments in bdi_max_pause() which might go negative if the highest bit is 1, and the min_t(long, ...) check failed to protect it falling under 0. Fix all of them by using "unsigned long" throughout the function.

Reported-by Toralf Förster <[email protected]>

Tested-by Toralf Förster <[email protected]>

Cc <[email protected]>

Cc Jan Kara <[email protected]>

Cc Richard Weinberger <[email protected]>

Cc Geert Uytterhoeven <[email protected]>

Signed-off-by Fengguang Wu <[email protected]>

---
 mm/page-writeback.c |   10 +++++-----
 mm/readahead.c      |    2 +-
 2 files changed, 6 insertions(+), 6 deletions(-)

Changes since v1: Add CC list.

mm/page-writeback.c +5 -5

index 3f0c895..241a746 100644
@@ -1104,11 +1104,11 @@ static unsigned long dirty_poll_interval(unsigned long dirty,
 	return 1;
 }
 
-static long bdi_max_pause(struct backing_dev_info *bdi,
-			  unsigned long bdi_dirty)
+static unsigned long bdi_max_pause(struct backing_dev_info *bdi,
+				   unsigned long bdi_dirty)
 {
-	long bw = bdi->avg_write_bandwidth;
-	long t;
+	unsigned long bw = bdi->avg_write_bandwidth;
+	unsigned long t;
 
 	/*
 	 * Limit pause time for small memory systems. If sleeping for too long

@@ -1120,7 +1120,7 @@ static long bdi_max_pause(struct backing_dev_info *bdi,
 	t = bdi_dirty / (1 + bw / roundup_pow_of_two(1 + HZ / 8));
 	t++;
 
-	return min_t(long, t, MAX_PAUSE);
+	return min_t(unsigned long, t, MAX_PAUSE);
 }
 
 static long bdi_min_pause(struct backing_dev_info *bdi,