The year 2038 problem

All 32-bit kernels to date use a signed 32-bit time_t type, which can only represent time until January 2038. Since embedded systems running 32-bit Linux are going to survive beyond that date, we have to change all current uses, in a backwards compatible way.

User space interfaces

We will likely keep the 32-bit time_t in all user space interfaces that currently use it, but add new interfaces with a 64-bit timespec or another type that can represent later times. Most importantly that impacts system calls, but also specific ioctl commands and a few other interfaces. User space programs have to be recompiled to use the new interfaces, and the policy whether to use the old or the time time is left to the C library. While that policy is a complex topic itself, we don't cover it here.

System calls

System calls have been moved out "y2038/syscalls", that page is rather outdated, newer information is available at https://lwn.net/Articles/643234/

File systems

Each file system stores its file modification times in its own format on disk, and a lot of them have the same problem.

file system	time type	expiration year
9p (9P2000)	unsigned 32-bit seconds	2106
file system	time type	expiration year
9p (9P2000)	unsigned 32-bit seconds	2106
9p (9P2000.L)	signed 64-bit seconds, ns	never
adfs	40-bit cs since 1900	2248
affs	u32 days/mins/(secs/50)	11760870
afs	unsigned 32-bit seconds	2106
befs	unsigned 48-bit seconds	never
bfs	unsigned 32-bit seconds	2106
btrfs	signed 64-bit seconds, 32-bit ns	never
ceph	unsigned 32-bit second/ns	2106
cifs (smb)	7-bit years since 1980	2107
cifs (modern)	64-bit 100ns since 1601	30328
coda	timespec ioctl	2038
cramfs	fixed	1970
efs	unsigned 32-bit seconds	2106
exofs	signed 32-bit seconds	2038
ext2	signed 32-bit seconds	2038
ext3	signed 32-bit seconds	2038
ext4 (good old inodes)	signed 32-bit seconds	2038
ext4 (new inodes	34 bit seconds / 30-bit ns (but broken)	2038
f2fs	64-bit seconds / 32-bit ns	never
fat	7-bit years since 1980, 2s resolution	2107
freevxfs	unsigned 32-bit seconds/u32 microseconds	2106
fuse	64-bit second/32-bit ns	never
gfs2	u64 seconds/u32 ns	never
hfs	u32 seconds since 1904	2040
hfsplus	u32 seconds since 1904	2040
hostfs	timespec	2038
hpfs	unsigned 32-bit seconds	2106
isofs	'char' year since 1900 (fixable)	2028
jffs2	unsigned 32-bit seconds	2106
jfs	unsigned 32-bit seconds/ns	2106
logfs	signed 64-bit ns	2262
minix	unsigned 32-bit seconds	2106
ncpfs	7-bit year since 1980	2107
nfsv2,v3	unsigned 32-bit seconds/ns	2106
nfsv4	u64 seconds/u32 ns	never
nfsd	unsigned 32-bit seconds/ns	2106
nilfs2	u64 seconds/u32 ns	never
ntfs	64-bit 100ns since 1601	30828
ocfs2	34-bit seconds/30-bit ns	2514
omfs	64-bit milliseconds	never
pstore	ascii seconds	2106
qnx4	unsigned 32-bit seconds	2106
qnx6	unsigned 32-bit seconds	2106
reiserfs	unsigned 32-bit seconds	2106
romfs	fixed	1970
squashfs	unsigned 32-bit seconds	2106
sysv	unsigned 32-bit seconds	2106
ubifs	u64 second/u32 ns	never
udf	u16 year	2038
ufs1	unsigned 32-bit seconds	2106
ufs2	signed 64-bit seconds/u32 ns	never
xfs	signed 32-bit seconds/ns	2106

Plan

1.Plan for VFS

Milestone	# of patches	Status	Targeted Kernel release
Infrastructure patches for vfs_time and granularity fields 1.4.1 and 1.4.2	2	Done	4.7- rc1
Test time ranges beyond y2038 support, in memory timestamps match with on disk -1.3.1 and 1.3.2	Xfs tests	In progress	5/28/16
Replace CURRENT_TIME and CURRENT_TIME_SEC – 1.2.1	~60	In Progress, waiting for 4.7 rc1 and repost previous.	4.8
Vfs_time patches – 1.4.3	~40	In Progress, waiting for 4.7-rc1	4.8
Fill in right granularities and ranges for FS – 1.2.2 and 1.2.3	~40	Yet to start	4.8
Expose FS granularities and ranges – 1.4.5 and 1.4.6	~5	Yet to start	4.8
Tests to check the granularities and ranges – 1.3.3 and 1.3.4	Xfs tests	Yet to start	6/10/16
Add policy for wrong granularities - 1.4.7	RFC for 4.9	Yet to start
Stat and utime changes – 1.4.8	RFC for 4.9	Yet to start
Tests for I_DIRTY_* flags and mount options -1.3.5 and 1.3.6	Xfs tests	Yet to start	6/30/16
Transition vfs to 64 bit – 1.4.4	2	Ready, wait for other patches to merge	4.10

1.2 Cleanup

1.2.1 Replace current_time and current_time_sec
1.2.2 Fill in right granularities
1.2.3 Fill in right ranges

1.3 Test

1.3.1 Test time written on disk and in memory are same
1.3.2 Test time ranges beyond y2038
1.3.3 Test time granularities are what the fs needs
1.3.4 Test time ranges supported are same as what fs reports
1.3.5 Perf tests
1.3.6 Test to check i_dirty_* flags are updating correctly
1.3.7 Test to check with different mount options like lazy atime etc.

1.4 Kernel tasks to transition vfs to use 64 bit times

1.4.1 Infrastructure patches for vfs_time approach
1.4.2 Infrastructure patch for vfs granularity fields
1.4.3 Change each fs to use vfs time patches
1.4.4 Change vfs to use timespec64
1.4.5 Expose fs granularities
1.4.6 Expose fs ranges
1.4.7 Add vfs policy for handling wrong ranges and granularity
1.4.8 Stat and utime api changes

1.5 Runtime library

1.5.1 Make changes in one library
1.5.2 Make it part of some userspace and run it

Tasks

The task list is for people that want to get involved, there will be many more tasks over time, so this is just a starting point. In the end, we should remove all instances of 'time_t', 'timespec' and 'timeval' from the kernel.

Trivial tasks

Find a driver using time_t/timespec/timeval internally and convert it to ktime_t/timespec64, examples:

drivers/firewire/core-cdev.c
drivers/firewire/nosy.c
drivers/hv/hv_util.c
drivers/infiniband/hw/nes/
drivers/misc/ibmasm/ibmasm.h
drivers/net/wireless/atmel.c
drivers/ide/pdc202xx_new.c
fs/fuse/dir.c
drivers/scsi/pmcraid.c

Small tasks

Find a driver using time_t/timespec/timeval internally and convert it to ktime_t/timespec64, examples:

drivers/block/sx8.c
drivers/char/ipmi/ipmi_ssif.c
drivers/char/mmtimer.c
drivers/cpufreq/speedstep-lib.c
drivers/hsi/clients/cmt_speech.c
drivers/media/platform/vivid/
drivers/net/hamradio/baycom_ser_fdx.c

Medium tasks

Modify an ioctl interface in a driver to support both 32- and 64-bit time interfaces, examples:
- drivers/usb/misc/usbtest.c
Convert the internal timekeeping in fs/nfsd
Convert all 'struct key' users (time_t)
fix all uses of timeval in drivers/scsi/bfa

Advanced tasks

Change the on-disk layout of a broken file system to optionally support longer time stamps
Port a small C library (uClibc, newlib, ...) to optionally use 64-bit time_t and build an embedded distribution (openembedded, openwrt, buildroot, ...) with this.
start migrating the times in 'struct inode' and 'struct 'kstat' to separate 64-bit second + 32-bit nanosecond fields.

Tasks later in the project

Hook up all 32-bit architectures to use the new system calls
Introduce a Kconfig symbol to disable all code that has not yet been converted at compile time.

-  ⇤ ← Revision 15 as of 2014-10-09 16:10:07 → 
  Size: 10725
  Editor: arnd
  Comment: add some more staging tasks
+   ← Revision 29 as of 2016-05-20 02:46:19 → ⇥
  Size: 8458
  Editor: DeepaDinamani
  Comment:
-Deletions are marked like this.
+Additions are marked like this.
 Line 1:
-= The year 2038 problem =
+''unfortunately this page got deleted by accident, this version is restored from https://web.archive.org/web/20150318062024/http://kernelnewbies.org/y2038, but most of the formatting is still missing. ''

== The year 2038 problem ==
-Line 4:
+Line 6:
-== User space interfaces ==
+=== User space interfaces ===
-Line 7:
+Line 9:
+==== System calls ====
System calls have been moved out [["y2038/syscalls"]], that page is rather outdated, newer information is available at https://lwn.net/Articles/643234/
-Line 8:
+Line 12:
-=== System calls ===
==== interfaces that uses relative time_t/timespec/timeval ====
These can stay compatible, but we'd have to use a different type if we change time_t.

 * nanosleep
 * select/pselect/poll/ppoll/epoll
 * getrusage (+ wait4?)

 * sched_rr_get_interval
 * sigtimedwait
 * alarm
 * Also: futex, recvmmsg, io_getevents?

==== interfaces that don't make sense for times in the past ====
Here, we are relatively free to change the start of the epoch in the kernel but convert to something else on the user space boundary. One possibility is to scale them to boot time and use ktime_t in the kernel.

[Does checkpoint/restore have any implications here wrt to how freely we can change the start of the epoch? E.g., when freezing/restoring processed from different systems that have timer_settime() timers?]

 * getitimer/setitimer
 * timer_settime/timer_gettime(+timerfd_settime, timerfd_gettime?)

 * gettimeofday/settimeofday
 * adjtimex
 * clock_gettime/clock_settime/clock_adjtime/clock_nanosleep(? what about clock_nanosleep TIMER_ABSTIME?)

 * time/stime
 * sysv ipc (msg, sem, shm)
 * sysinfo

==== interfaces that require absolute times ====
These absolutely have to use something better than time_t both in user space and in the kernel so we can deal with old files. A lot of file systems need to be fixed as well so we can actually store the times, regardless of whether we are running a 32 or 64 bit kernel.

 * stat/lstat/fstatat/
 * utime/utimes/futimesa/utimensat

=== ioctl ===
There are numerous ioctl commands using a time argument. This list is incomplete

 * audio time stamps
 * v4l time stamps
 * input event time stamps
 * socket time stamps
 * ...

=== memory mapped packet sockets ===
Socket timestamps are exported to user space using a memory mapped interface defined in include/uapi/linux/if_packet.h. There are currently three versions of this interface, all use a 32-bit time type. We will likely need a version 4 to solve this.


=== Audit of include/uapi for time_t impact ===

Structure and IOCTL dependency:
{{{
time_t
 struct msqid64_ds (has 2038 padding!)
 struct semid64_ds (has 2038 padding!)
 struct cyclades_idle_stats
 struct video_event
  VIDEO_GET_EVENT
 struct msqid_ds
 struct ppp_idle
  PPPIOCGIDLE
 struct semid_ds
  union semun
 struct timespec
  SIOCGSTAMPNS
  struct coda_vattr
   ...
  struct scm_timestamping
  struct som_hdr
  struct itimerspec
  struct v4l2_event
   VIDIOC_DQEVENT
  struct snd_pcm_status
   SNDRV_PCM_IOCTL_STATUS
  struct snd_pcm_mmap_status
   struct snd_pcm_sync_ptr
    SNDRV_PCM_IOCTL_SYNC_PTR
  struct snd_rawmidi_status
   SNDRV_RAWMIDI_IOCTL_STATUS
  struct snd_timer_status
   SNDRV_TIMER_IOCTL_STATUS
  struct snd_timer_tread
  struct snd_ctl_elem_value
   SNDRV_CTL_IOCTL_ELEM_READ
   SNDRV_CTL_IOCTL_ELEM_WRITE
 struct timeval
  SIOCGSTAMP
  struct zatm_t_hist
  struct bcm_msg_head
  struct elf_prstatus
  struct input_event
  struct omap3isp_stat_data
   VIDIOC_OMAP3ISP_STAT_REQ
  PPGETTIME
  PPSETTIME
  struct rusage
  struct itimerval
  struct timex
  struct v4l2_buffer
   VIDIOC_QUERYBUF
   VIDIOC_QBUF
   VIDIOC_DQBUF
   VIDIOC_PREPARE_BUF
 struct utimbuf
}}}

Syscalls affected by the above types:
{{{
 sys_time
 sys_stime
 sys_nanosleep
 sys_clock_settime
 sys_clock_gettime
 sys_clock_getres
 sys_clock_nanosleep
 sys_sched_rr_get_interval
 sys_futex
 sys_rt_sigtimedwait
 sys_io_getevents
 sys_recvmmsg
 sys_semtimedop
 sys_mq_timedsend
 sys_mq_timedreceive
 sys_utimensat
 sys_pselect6
 sys_ppoll
 sys_gettimeofday
 sys_settimeofday
 sys_utimes
 sys_select
 sys_futimesat
 sys_utime
 sys_timer_gettime
 sys_timer_settime
 sys_timerfd_settime
 sys_timerfd_gettime
 sys_wait4
 sys_waitid
 sys_getrusage
 sys_getitimer
 sys_setitimer
 sys_adjtimex
 sys_clock_adjtime
}}}


== File systems ==
+=== File systems ===
-Line 158:
+Line 15:
-|| '''file system''' || '''time type''' || '''expiration year''' ||
+|| file system || time type || expiration year ||
|| 9p (9P2000) || unsigned 32-bit seconds || 2106 ||
|| file system || time type || expiration year ||
-Line 161:
+Line 20:
-|| adfs || 40-bit cs since 1900 || 2248 ||
+|| adfs || 40-bit  cs since 1900 || 2248 ||
-Line 163:
+Line 22:
-|| afs ||  unsigned 32-bit seconds  || 2106 ||
|| befs || unsigned 48-bit seconds ||  never ||
|| bfs ||  unsigned 32-bit seconds ||  2106 ||
+|| afs || unsigned 32-bit seconds || 2106 ||
|| befs || unsigned 48-bit seconds || never ||
|| bfs || unsigned 32-bit seconds || 2106 ||
-Line 180:
+Line 39:
-|| freevxfs ||  unsigned 32-bit seconds/u32 microseconds ||  2106 ||
+|| freevxfs || unsigned 32-bit seconds/u32 microseconds || 2106 ||
-Line 182:
+Line 41:
-|| gfs2 ||  u64 seconds/u32 ns ||  never ||
+|| gfs2 || u64 seconds/u32 ns || never ||
-Line 185:
+Line 44:
-|| hostfs ||  timespec ||  2038 ||
|| hpfs ||  unsigned 32-bit seconds ||  2106 ||
|| isofs || 'char' year since 1900 (fixable) ||  2028 (!) ||
+|| hostfs || timespec || 2038 ||
|| hpfs || unsigned 32-bit seconds || 2106 ||
|| isofs || 'char' year since 1900 (fixable) || 2028 ||
-Line 189:
+Line 48:
-|| jfs ||  unsigned 32-bit seconds/ns ||  2106 ||
+|| jfs || unsigned 32-bit seconds/ns || 2106 ||
-Line 191:
+Line 50:
-|| minix ||  unsigned 32-bit seconds ||  2106 ||
|| ncpfs ||  7-bit year since 1980 ||  2107 ||
+|| minix || unsigned 32-bit seconds || 2106 ||
|| ncpfs || 7-bit year since 1980 || 2107 ||
-Line 196:
+Line 55:
-|| nilfs2 ||  u64 seconds/u32 ns ||  never ||
+|| nilfs2 || u64 seconds/u32 ns || never ||
-Line 199:
+Line 58:
-|| omfs ||  64-bit milliseconds ||  never ||
+|| omfs || 64-bit milliseconds || never ||
-Line 201:
+Line 60:
-|| qnx4 ||  unsigned 32-bit seconds ||  2106 ||
|| qnx6 ||  unsigned 32-bit seconds ||  2106 ||
+|| qnx4 || unsigned 32-bit seconds || 2106 ||
|| qnx6 || unsigned 32-bit seconds || 2106 ||
-Line 206:
+Line 65:
-|| sysv ||  unsigned 32-bit seconds ||  2106 ||
+|| sysv || unsigned 32-bit seconds || 2106 ||
-Line 210:
+Line 69:
-|| ufs2 || signed 64-bit seconds/u32 ns ||never ||
+|| ufs2 || signed 64-bit seconds/u32 ns || never ||
-Line 213:
+Line 72:
-== Tasks ==
-Line 215:
+Line 73:
+= Plan =
=== 1.Plan for VFS ===
|| Milestone || # of patches || Status || Targeted Kernel release ||
|| Infrastructure patches for vfs_time and granularity fields 1.4.1 and 1.4.2 || 2 || Done || 4.7- rc1 ||
|| Test time ranges beyond y2038 support, in memory timestamps match with on disk -1.3.1 and 1.3.2 || Xfs tests || In progress || 5/28/16 ||
|| Replace CURRENT_TIME and CURRENT_TIME_SEC – 1.2.1 || ~60 || In Progress, waiting for 4.7 rc1 and repost previous. || 4.8 ||
|| Vfs_time patches – 1.4.3 || ~40 || In Progress, waiting for 4.7-rc1 || 4.8 ||
|| Fill in right granularities and ranges for FS – 1.2.2 and 1.2.3 || ~40 || Yet to start || 4.8 ||
|| Expose FS granularities and ranges – 1.4.5 and 1.4.6 || ~5 || Yet to start || 4.8 ||
|| Tests to check the granularities and ranges – 1.3.3 and 1.3.4 || Xfs tests || Yet to start || 6/10/16 ||
|| Add policy for wrong granularities - 1.4.7 || RFC for 4.9 || Yet to start ||  ||
|| Stat and utime changes – 1.4.8 || RFC for 4.9 || Yet to start ||  ||
|| Tests for I_DIRTY_* flags and mount options -1.3.5 and 1.3.6 || Xfs tests || Yet to start || 6/30/16 ||
|| Transition vfs to 64 bit – 1.4.4 || 2 || Ready, wait for other patches to merge || 4.10 ||


=== 1.2 Cleanup ===
 '''1.2.1 '''Replace current_time and current_time_sec

 '''1.2.2''' Fill in right granularities

 '''1.2.3''' Fill in right ranges

=== 1.3 Test ===
 '''1.3.1''' Test time written on disk and in memory are same

 '''1.3.2 '''Test time ranges beyond y2038

 '''1.3.3''' Test time granularities are what the fs needs

 '''1.3.4''' Test time ranges supported are same as what fs reports

 '''1.3.5''' Perf tests

 '''1.3.6''' Test to check i_dirty_* flags are updating correctly

 '''1.3.7''' Test to check with different mount options like lazy atime etc.

=== 1.4 Kernel tasks to transition vfs to use 64 bit times ===
 '''1.4.1''' Infrastructure patches for vfs_time approach

 '''1.4.2''' Infrastructure patch for vfs granularity fields

 '''1.4.3''' Change each fs to use vfs time patches

 '''1.4.4''' Change vfs to use timespec64

 '''1.4.5''' Expose fs granularities

 '''1.4.6''' Expose fs ranges

 '''1.4.7''' Add vfs policy for handling wrong ranges and granularity

 '''1.4.8''' Stat and utime api changes

=== 1.5 Runtime library ===
 '''1.5.1''' Make changes in one library

 '''1.5.2''' Make it part of some userspace and run it


=== Tasks ===
-Line 217:
+Line 137:
-=== Small tasks ===
+==== Trivial tasks ====
Find a driver using time_t/timespec/timeval internally and convert it to ktime_t/timespec64, examples:
-Line 219:
+Line 140:
- * Find a driver using time_t/timespec/timeval internally and convert it to ktime_t/timespec64, examples:
  * drivers/staging/media/lirc/lirc_imon.c (timeval, trivial)
  * drivers/staging/ft1000/ (time_t and timeval)
  * drivers/staging/android/sync_debug.c (timeval, very easy)
  * drivers/staging/android/timed_gpio.c (timeval, easy)
  * drivers/staging/bcm/LeakyBucket.c (timeval, slightly tricky)
  * drivers/staging/bcm/Bcmchar.c (timeval, very easy)
  * drivers/staging/comedi/drivers/comedi_test.c (timeval)
  * drivers/staging/comedi/drivers/serial2002.c (timeval, easy)
  * drivers/staging/dgnc/dgnc_tty.c (timeval, very easy)
  * drivers/staging/gdm72xx/ (timeval, easy)
  * drivers/staging/media/lirc/lirc_igorplugusb.c (timeval)
  * drivers/staging/media/lirc/lirc_parallel.c (timeval, easy)
  * drivers/staging/media/lirc/lirc_sasem.c (timeval, very easy)
  * drivers/staging/media/lirc/lirc_serial.c (timeval, easy)
  * drivers/staging/media/lirc/lirc_sir.c (timeval)
  * drivers/staging/rts5208/rtsx.h (timeval)
  * drivers/staging/olpc_dcon/olpc_dcon.c (timespec, rather broken)
  * drivers/staging/ozwpan/ozhcd.c (timespec)
  * drivers/staging/ozwpan/ozproto.c (timespec)
  * kernel/cpuset.c (time_t)
  * fs/reiserfs/journal.c (time_t)
  * drivers/scsi/ips.c (time_t)
  * sound/pci/es1968.c (timeval)
  * kernel/power/hibernate.c (timeval)
  * drivers/s390/net/ctcm_fsms.c (timespec)
  * drivers/power/ab8500_fg.c (timespec)
+ * drivers/firewire/core-cdev.c
 * drivers/firewire/nosy.c
 * drivers/hv/hv_util.c
 * drivers/infiniband/hw/nes/
 * drivers/misc/ibmasm/ibmasm.h
 * drivers/net/wireless/atmel.c
 * drivers/ide/pdc202xx_new.c
 * fs/fuse/dir.c
 * drivers/scsi/pmcraid.c
-Line 247:
+Line 150:
-=== Medium tasks ===
+==== Small tasks ====
Find a driver using time_t/timespec/timeval internally and convert it to ktime_t/timespec64, examples:
-Line 249:
+Line 153:
+ * drivers/block/sx8.c
 * drivers/char/ipmi/ipmi_ssif.c
 * drivers/char/mmtimer.c
 * drivers/cpufreq/speedstep-lib.c
 * drivers/hsi/clients/cmt_speech.c
 * drivers/media/platform/vivid/
 * drivers/net/hamradio/baycom_ser_fdx.c

==== Medium tasks ====
-Line 250:
+Line 163:
-  * drivers/staging/comedi/comedi_fops.c (INSN_GTOD, timeval)
  * drivers/staging/android/alarm-dev.c (timespec)
  * include/uapi/linux/atm_zatm.h (zatm_t_hist/timeval)
  * include/uapi/linux/videodev2.h (v4l2_buffer/timespec)
 * Fix the android logger time format (drivers/staging/android/logger.c)
+  * drivers/usb/misc/usbtest.c
-Line 256:
+Line 166:
- * Convert all 'ptp' users in the kernel
-Line 258:
+Line 168:
- * Introduce known unsafe types (possibly like __kernel_time32_t, __kernel_compat_time_t etc) so we can annotate interfaces that are known to use a fixed size and cannot be changed to new types.
 * Fix all time issues in drivers/staging/lustre (maybe advanced task)
-Line 261:
+Line 169:
-=== Advanced tasks ===
+ * fix all uses of timeval in drivers/scsi/bfa
-Line 263:
+Line 171:
- * Introduce a new system call family to replace one or more of the problematic calls listed above.
+==== Advanced tasks ====
-Line 265:
+Line 173:
-Line 267:
+Line 176:
-=== Tasks later in the project ===
+ * start migrating the times in 'struct inode' and 'struct 'kstat' to separate 64-bit second + 32-bit nanosecond fields.

==== Tasks later in the project ====
-Line 269:
+Line 180: