|
Size: 2554
Comment: start out with a syscall problem statement
|
Size: 8242
Comment:
|
| Deletions are marked like this. | Additions are marked like this. |
| Line 1: | 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 5: | Line 7: |
| == User space interfaces == |
=== User space interfaces === |
| Line 9: | Line 10: |
| === System calls === | ==== System calls ==== |
| Line 11: | Line 12: |
| ==== interfaces that uses relative time_t/timespec/timeval ==== | System calls have been moved out [[:y2038/syscalls:]], that page is rather outdated, newer information is available at https://lwn.net/Articles/643234/ |
| Line 13: | Line 14: |
| These can stay compatible, but we'd have to use a different type if we change time_t. |
=== 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. |
| Line 16: | Line 17: |
| * nanosleep * select/pselect/poll/ppoll/epoll * getrusage * sched_rr_get_interval * sigtimedwait * alarm |
|| 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 || |
| Line 23: | Line 74: |
| ==== 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. |
|
| Line 29: | Line 75: |
| * getitimer/setitimer * timer_settime/timer_gettime * gettimeofday/settimeofday * adjtimex * clock_gettime/clock_settime/clock_adjtime/clock_nanosleep * time/stime * sysv ipc (msg, sem, shm) * sysinfo |
=== Plan === || 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 || |
| Line 38: | Line 89: |
| ==== 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. |
|
| Line 45: | Line 90: |
| * stat/lstat/fstatat/ * utime/utimes/futimesat |
=== Cleanup === * Replace current_time and current_time_sec |
| Line 48: | Line 93: |
| === ioctl === | * Fill in right granularities |
| Line 50: | Line 95: |
| There are numerous ioctl commands using a time argument. This list is incomplete | * Fill in right ranges |
| Line 52: | Line 97: |
| * audio time stamps * v4l time stamps * input event time stamps * socket time stamps * ... |
=== Test === * Test time written on disk and in memory are same |
| Line 58: | Line 100: |
| === memory mapped packet sockets === | * Test time ranges beyond y2038 |
| Line 60: | Line 102: |
| 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. |
* Test time granularities are what the fs needs * Test time ranges supported are same as what fs reports * Perf tests * Test to check i_dirty_* flags are updating correctly * Test to check with different mount options like lazy atime etc. === Kernel tasks to transition vfs to use 64 bit times === * Infrastructure patches for vfs_time approach * Infrastructure patch for vfs granularity fields * Change each fs to use vfs time patches * Change vfs to use timespec64 * Expose fs granularities * Expose fs ranges * Add vfs policy for handling wrong ranges and granularity * Stat and utime api changes === Runtime library === * Make changes in one library * 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. |
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.
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.
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 have been moved out :y2038/syscalls:, that page is rather outdated, newer information is available at https://lwn.net/Articles/643234/
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
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
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.
Find a driver using time_t/timespec/timeval internally and convert it to ktime_t/timespec64, examples:
Find a driver using time_t/timespec/timeval internally and convert it to ktime_t/timespec64, examples:
The year 2038 problem
User space interfaces
System calls
File systems
Plan
Cleanup
Test
Kernel tasks to transition vfs to use 64 bit times
Runtime library
Tasks
Trivial tasks
Small tasks
Medium tasks
Advanced tasks
Tasks later in the project