| 2013-09-06 17:41:49 |
Colin Ian King |
description |
On a Samsung Galaxy Nexus I observed that systemd-udev is handling events (from /devices/platform/omapfb ?) quite frequently. I see on average epoll_wait() being called once every 4-8 seconds to handle these.
epoll_wait(11, {{EPOLLIN, {u32=7, u64=7}}}, 8, -1) = 1
clock_gettime(CLOCK_MONOTONIC, {14518, 144413196}) = 0
stat64("/etc/udev/rules.d", {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0
stat64("/run/udev/rules.d", 0xbe9db540) = -1 ENOENT (No such file or directory)
stat64("/lib/udev/rules.d", {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0
fstat64(10, {st_mode=S_IFREG|0444, st_size=5658459, ...}) = 0
stat64("/etc/modprobe.d", {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0
stat64("/lib/modprobe.d", {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0
stat64("/lib/modules/3.0.0-3-maguro/modules.dep.bin", {st_mode=S_IFREG|0644, st_size=25755, ...}) = 0
stat64("/lib/modules/3.0.0-3-maguro/modules.alias.bin", {st_mode=S_IFREG|0644, st_size=8363, ...}) = 0
stat64("/lib/modules/3.0.0-3-maguro/modules.symbols.bin", {st_mode=S_IFREG|0644, st_size=14023, ...}) = 0
stat64("/lib/modules/3.0.0-3-maguro/modules.builtin.bin", {st_mode=S_IFREG|0644, st_size=12735, ...}) = 0
clock_gettime(CLOCK_MONOTONIC, {14518, 159977159}) = 0
read(7, "\5\0\0\0\0\0\0\0\200\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 128) = 128
epoll_wait(11, {{EPOLLIN, {u32=4, u64=4}}}, 8, -1) = 1
clock_gettime(CLOCK_MONOTONIC, {14535, 253147325}) = 0
stat64("/etc/udev/rules.d", {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0
stat64("/run/udev/rules.d", 0xbe9db540) = -1 ENOENT (No such file or directory)
stat64("/lib/udev/rules.d", {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0
fstat64(10, {st_mode=S_IFREG|0444, st_size=5658459, ...}) = 0
stat64("/etc/modprobe.d", {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0
stat64("/lib/modprobe.d", {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0
stat64("/lib/modules/3.0.0-3-maguro/modules.dep.bin", {st_mode=S_IFREG|0644, st_size=25755, ...}) = 0
stat64("/lib/modules/3.0.0-3-maguro/modules.alias.bin", {st_mode=S_IFREG|0644, st_size=8363, ...}) = 0
stat64("/lib/modules/3.0.0-3-maguro/modules.symbols.bin", {st_mode=S_IFREG|0644, st_size=14023, ...}) = 0
stat64("/lib/modules/3.0.0-3-maguro/modules.builtin.bin", {st_mode=S_IFREG|0644, st_size=12735, ...}) = 0
clock_gettime(CLOCK_MONOTONIC, {14535, 260898790}) = 0
recvmsg(4, {msg_name(12)={sa_family=AF_NETLINK, pid=0, groups=00000001}, msg_iov(1)=[{"change@/devices/platform/omapfb\0"..., 8192}], msg_controllen=24, {cmsg_len=24, cmsg_level=SOL_SOCKET, cmsg_type=SCM_CREDENTIALS{pid=0, uid=0, gid=0}}, msg_flags=0}, 0) = 170
clock_gettime(CLOCK_MONOTONIC, {14535, 262150011}) = 0
_llseek(5, 0, [8], SEEK_CUR) = 0
write(5, "\335\f\0\0\0\0\0\0\30\0/devices/platform/omap"..., 34) = 34
socket(PF_NETLINK, SOCK_RAW|SOCK_CLOEXEC|SOCK_NONBLOCK, 15) = 12
bind(12, {sa_family=AF_NETLINK, pid=0, groups=00000000}, 12) = 0
getsockname(12, {sa_family=AF_NETLINK, pid=541, groups=00000000}, [12]) = 0
setsockopt(12, SOL_SOCKET, SO_PASSCRED, [1], 4) = 0
clone(child_stack=0, flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLD, child_tidptr=0x400b5068) = 4896
close(12) = 0
clock_gettime(CLOCK_MONOTONIC, {14535, 266971788}) = 0
epoll_wait(11, {{EPOLLIN, {u32=8, u64=8}}}, 8, 3000) = 1
clock_gettime(CLOCK_MONOTONIC, {14535, 270206651}) = 0
recv(8, " \23\0\0\0\0\0\0", 8, MSG_DONTWAIT) = 8
close(5) = 0
munmap(0x40054000, 4096) = 0
open("/run/udev/queue.tmp", O_RDWR|O_CREAT|O_TRUNC|O_LARGEFILE|O_CLOEXEC, 0666) = 5
fstat64(5, {st_mode=S_IFREG|0644, st_size=0, ...}) = 0
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x400d5000
write(5, "\335\f\0\0\0\0\0\0", 8) = 8
rename("/run/udev/queue.tmp", "/run/udev/queue.bin") = 0
recv(8, 0xbe9db828, 8, MSG_DONTWAIT) = -1 EAGAIN (Resource temporarily unavailable)
I see the same activity but very frequently if I touch the touch screen. Is that expected behaviour? |
On a Samsung Galaxy Nexus I observed that systemd-udev is handling events (from /devices/platform/omapfb ?) quite frequently even on an idle phone. I see on average epoll_wait() being called once every 4-8 seconds to handle these.
epoll_wait(11, {{EPOLLIN, {u32=7, u64=7}}}, 8, -1) = 1
clock_gettime(CLOCK_MONOTONIC, {14518, 144413196}) = 0
stat64("/etc/udev/rules.d", {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0
stat64("/run/udev/rules.d", 0xbe9db540) = -1 ENOENT (No such file or directory)
stat64("/lib/udev/rules.d", {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0
fstat64(10, {st_mode=S_IFREG|0444, st_size=5658459, ...}) = 0
stat64("/etc/modprobe.d", {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0
stat64("/lib/modprobe.d", {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0
stat64("/lib/modules/3.0.0-3-maguro/modules.dep.bin", {st_mode=S_IFREG|0644, st_size=25755, ...}) = 0
stat64("/lib/modules/3.0.0-3-maguro/modules.alias.bin", {st_mode=S_IFREG|0644, st_size=8363, ...}) = 0
stat64("/lib/modules/3.0.0-3-maguro/modules.symbols.bin", {st_mode=S_IFREG|0644, st_size=14023, ...}) = 0
stat64("/lib/modules/3.0.0-3-maguro/modules.builtin.bin", {st_mode=S_IFREG|0644, st_size=12735, ...}) = 0
clock_gettime(CLOCK_MONOTONIC, {14518, 159977159}) = 0
read(7, "\5\0\0\0\0\0\0\0\200\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 128) = 128
epoll_wait(11, {{EPOLLIN, {u32=4, u64=4}}}, 8, -1) = 1
clock_gettime(CLOCK_MONOTONIC, {14535, 253147325}) = 0
stat64("/etc/udev/rules.d", {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0
stat64("/run/udev/rules.d", 0xbe9db540) = -1 ENOENT (No such file or directory)
stat64("/lib/udev/rules.d", {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0
fstat64(10, {st_mode=S_IFREG|0444, st_size=5658459, ...}) = 0
stat64("/etc/modprobe.d", {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0
stat64("/lib/modprobe.d", {st_mode=S_IFDIR|0755, st_size=4096, ...}) = 0
stat64("/lib/modules/3.0.0-3-maguro/modules.dep.bin", {st_mode=S_IFREG|0644, st_size=25755, ...}) = 0
stat64("/lib/modules/3.0.0-3-maguro/modules.alias.bin", {st_mode=S_IFREG|0644, st_size=8363, ...}) = 0
stat64("/lib/modules/3.0.0-3-maguro/modules.symbols.bin", {st_mode=S_IFREG|0644, st_size=14023, ...}) = 0
stat64("/lib/modules/3.0.0-3-maguro/modules.builtin.bin", {st_mode=S_IFREG|0644, st_size=12735, ...}) = 0
clock_gettime(CLOCK_MONOTONIC, {14535, 260898790}) = 0
recvmsg(4, {msg_name(12)={sa_family=AF_NETLINK, pid=0, groups=00000001}, msg_iov(1)=[{"change@/devices/platform/omapfb\0"..., 8192}], msg_controllen=24, {cmsg_len=24, cmsg_level=SOL_SOCKET, cmsg_type=SCM_CREDENTIALS{pid=0, uid=0, gid=0}}, msg_flags=0}, 0) = 170
clock_gettime(CLOCK_MONOTONIC, {14535, 262150011}) = 0
_llseek(5, 0, [8], SEEK_CUR) = 0
write(5, "\335\f\0\0\0\0\0\0\30\0/devices/platform/omap"..., 34) = 34
socket(PF_NETLINK, SOCK_RAW|SOCK_CLOEXEC|SOCK_NONBLOCK, 15) = 12
bind(12, {sa_family=AF_NETLINK, pid=0, groups=00000000}, 12) = 0
getsockname(12, {sa_family=AF_NETLINK, pid=541, groups=00000000}, [12]) = 0
setsockopt(12, SOL_SOCKET, SO_PASSCRED, [1], 4) = 0
clone(child_stack=0, flags=CLONE_CHILD_CLEARTID|CLONE_CHILD_SETTID|SIGCHLD, child_tidptr=0x400b5068) = 4896
close(12) = 0
clock_gettime(CLOCK_MONOTONIC, {14535, 266971788}) = 0
epoll_wait(11, {{EPOLLIN, {u32=8, u64=8}}}, 8, 3000) = 1
clock_gettime(CLOCK_MONOTONIC, {14535, 270206651}) = 0
recv(8, " \23\0\0\0\0\0\0", 8, MSG_DONTWAIT) = 8
close(5) = 0
munmap(0x40054000, 4096) = 0
open("/run/udev/queue.tmp", O_RDWR|O_CREAT|O_TRUNC|O_LARGEFILE|O_CLOEXEC, 0666) = 5
fstat64(5, {st_mode=S_IFREG|0644, st_size=0, ...}) = 0
mmap2(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x400d5000
write(5, "\335\f\0\0\0\0\0\0", 8) = 8
rename("/run/udev/queue.tmp", "/run/udev/queue.bin") = 0
recv(8, 0xbe9db828, 8, MSG_DONTWAIT) = -1 EAGAIN (Resource temporarily unavailable)
I see the same activity but very frequently if I touch the touch screen. Is that expected behaviour? |
|
