Y2038

y2038@lists.linaro.org

1132 discussions

[PATCH 0/7] scsi: bfa: do_gettimeofday removal

by Arnd Bergmann

The bfa driver is one of the main users of do_gettimeofday(), a function that I'm trying to remove as part of the y2038 cleanup. The timestamps are all uses in slightly different ways, so this has turned into a rather longish series for doing something that should be simple. The last patch in the series ("scsi: bfa: use 64-bit times in bfa_aen_entry_s ABI") is one that needs to be reviewed very carefully, and it can be skipped if the maintainers prefer to leave the 32-bit ABI unchanged, the rest are hopefully fairly straightforward. Arnd Arnd Bergmann (7): scsi: bfa: use ktime_get_real_ts64 for firmware timestamp scsi: bfa: use proper time accessor for stats_reset_time scsi: bfa: improve bfa_ioc_send_enable/disable data scsi: bfa: document overflow of io_profile_start_time scsi: bfa: replace bfa_get_log_time() with ktime_get_real_seconds() scsi: bfa: try to sanitize vendor netlink events scsi: bfa: use 64-bit times in bfa_aen_entry_s ABI drivers/scsi/bfa/bfa_cs.h | 6 +++--- drivers/scsi/bfa/bfa_defs_svc.h | 3 ++- drivers/scsi/bfa/bfa_fcpim.c | 3 ++- drivers/scsi/bfa/bfa_fcpim.h | 4 ++-- drivers/scsi/bfa/bfa_ioc.c | 8 ++++--- drivers/scsi/bfa/bfa_port.c | 15 +++---------- drivers/scsi/bfa/bfa_port.h | 2 +- drivers/scsi/bfa/bfa_svc.c | 47 ++++++++++++----------------------------- drivers/scsi/bfa/bfa_svc.h | 2 +- drivers/scsi/bfa/bfad_bsg.c | 4 +--- drivers/scsi/bfa/bfad_im.h | 32 +++++++++++++++++++--------- 11 files changed, 56 insertions(+), 70 deletions(-) -- 2.9.0

7 years, 5 months

[PATCH v2 1/2] alpha: osf_sys.c: fix put_tv32 regression

by Arnd Bergmann

There was a typo in the new version of put_tv32() that caused an unguarded access of a user space pointer, and failed to return the correct result in gettimeofday(), wait4(), usleep_thread() and old_adjtimex(). This fixes it to give the correct behavior again. Cc: stable(a)vger.kernel.org Fixes: 1cc6c4635e9f ("osf_sys.c: switch handling of timeval32/itimerval32 to copy_{to,from}_user()") Signed-off-by: Arnd Bergmann <arnd(a)arndb.de> --- v2: fix incorrect changelog description --- arch/alpha/kernel/osf_sys.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/alpha/kernel/osf_sys.c b/arch/alpha/kernel/osf_sys.c index ce3a675c0c4b..75a5c35a2067 100644 --- a/arch/alpha/kernel/osf_sys.c +++ b/arch/alpha/kernel/osf_sys.c @@ -964,8 +964,8 @@ static inline long put_tv32(struct timeval32 __user *o, struct timeval *i) { return copy_to_user(o, &(struct timeval32){ - .tv_sec = o->tv_sec, - .tv_usec = o->tv_usec}, + .tv_sec = i->tv_sec, + .tv_usec = i->tv_usec}, sizeof(struct timeval32)); } -- 2.9.0

7 years, 5 months

[PATCH 1/2] alpha: osf_sys.c: fix put_tv32 regression

by Arnd Bergmann

There was a typo in the new version of put_tv32() that caused uninitialized stack data to be written back to user space, rather than writing the actual timeval for the emulation of gettimeofday(), wait4(), usleep_thread() and old_adjtimex(). This fixes it to write the correct data again. Cc: stable(a)vger.kernel.org Fixes: 1cc6c4635e9f ("osf_sys.c: switch handling of timeval32/itimerval32 to copy_{to,from}_user()") Signed-off-by: Arnd Bergmann <arnd(a)arndb.de> --- arch/alpha/kernel/osf_sys.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/arch/alpha/kernel/osf_sys.c b/arch/alpha/kernel/osf_sys.c index ce3a675c0c4b..75a5c35a2067 100644 --- a/arch/alpha/kernel/osf_sys.c +++ b/arch/alpha/kernel/osf_sys.c @@ -964,8 +964,8 @@ static inline long put_tv32(struct timeval32 __user *o, struct timeval *i) { return copy_to_user(o, &(struct timeval32){ - .tv_sec = o->tv_sec, - .tv_usec = o->tv_usec}, + .tv_sec = i->tv_sec, + .tv_usec = i->tv_usec}, sizeof(struct timeval32)); } -- 2.9.0

7 years, 5 months

[PATCH] powerpc: eeh: stop using do_gettimeofday()

by Arnd Bergmann

This interface is inefficient and deprecated because of the y2038 overflow. ktime_get_seconds() is an appropriate replacement here, since it has sufficient granularity but is more efficient and uses monotonic time. Signed-off-by: Arnd Bergmann <arnd(a)arndb.de> --- arch/powerpc/include/asm/eeh.h | 2 +- arch/powerpc/kernel/eeh_driver.c | 2 +- arch/powerpc/kernel/eeh_pe.c | 8 ++++---- 3 files changed, 6 insertions(+), 6 deletions(-) diff --git a/arch/powerpc/include/asm/eeh.h b/arch/powerpc/include/asm/eeh.h index cd1df96335e5..5161c37dd039 100644 --- a/arch/powerpc/include/asm/eeh.h +++ b/arch/powerpc/include/asm/eeh.h @@ -93,7 +93,7 @@ struct eeh_pe { struct pci_bus *bus; /* Top PCI bus for bus PE */ int check_count; /* Times of ignored error */ int freeze_count; /* Times of froze up */ - struct timeval tstamp; /* Time on first-time freeze */ + time64_t tstamp; /* Time on first-time freeze */ int false_positives; /* Times of reported #ff's */ atomic_t pass_dev_cnt; /* Count of passed through devs */ struct eeh_pe *parent; /* Parent PE */ diff --git a/arch/powerpc/kernel/eeh_driver.c b/arch/powerpc/kernel/eeh_driver.c index 4e1b433f6cb5..4f71e4c9beb7 100644 --- a/arch/powerpc/kernel/eeh_driver.c +++ b/arch/powerpc/kernel/eeh_driver.c @@ -623,7 +623,7 @@ static int eeh_reset_device(struct eeh_pe *pe, struct pci_bus *bus, struct eeh_rmv_data *rmv_data) { struct pci_bus *frozen_bus = eeh_pe_bus_get(pe); - struct timeval tstamp; + time64_t tstamp; int cnt, rc; struct eeh_dev *edev; diff --git a/arch/powerpc/kernel/eeh_pe.c b/arch/powerpc/kernel/eeh_pe.c index 2e8d1b2b5af4..2d4956e97aa9 100644 --- a/arch/powerpc/kernel/eeh_pe.c +++ b/arch/powerpc/kernel/eeh_pe.c @@ -526,16 +526,16 @@ int eeh_rmv_from_parent_pe(struct eeh_dev *edev) */ void eeh_pe_update_time_stamp(struct eeh_pe *pe) { - struct timeval tstamp; + time64_t tstamp; if (!pe) return; if (pe->freeze_count <= 0) { pe->freeze_count = 0; - do_gettimeofday(&pe->tstamp); + pe->tstamp = ktime_get_seconds(); } else { - do_gettimeofday(&tstamp); - if (tstamp.tv_sec - pe->tstamp.tv_sec > 3600) { + tstamp = ktime_get_seconds(); + if (tstamp - pe->tstamp > 3600) { pe->tstamp = tstamp; pe->freeze_count = 0; } -- 2.9.0

7 years, 5 months

Y2038 meeting at ELCE

by Ben Hutchings

There was a meeting at ELCE about the status and development of Y2038- safe APIs in Linux and GNU libc. This included several developers from members of CIP, plus Arnd Bergmann, Mark Brown and Albert Aribaud. Below are my notes from the meeting. Ben. ## Kernel Arnd Bergmann started working on Y2038 about 6 years ago, initially looking at file-systems and VFS. File-systems are mostly ready but VFS hasn't been switched over yet. The largest missing piece is the syscall interface. "We know what we want to do." There was a complete patch set for an older version, but it has not been completely rebased onto current mainline. On 32-bit systems about 35 syscalls use 32-bit time, but half of those are already obsolete. Something like 22 new syscalls will be needed. Network, sound, media, key management, input have not been dealt with yet. Patches are available for some of these but they can be invasive and hard to review. This is a low priority for some subsystem maintainers. About 100 device drivers need changes, ranging from an obvious 1 line change to a week's work. About 10% of the changes are needed for Y2038 safety on both 32-bit and 64-bit architectures, the rest only for 32-bit. Arnd wants to include a kconfig option to disable the 32-bit time APIs, so that any remaining users are easy to detect. ## GNU libc Albert Aribaut talked about the status of glibc. It will need to support both 32-bit `time_t` and 64-bit `time_t` independently of the kernel. A draft specification for this exists at <https://sourceware.org/glibc/wiki/Y2038ProofnessDesign>. About 60 APIs are affected, using `time_t` or derived type. Ideally source can be rebuilt to use 64-bit `time_t` just by defining the feature macro to enable it. The implementation is not complete, partly because syscalls haven't yet been defined. ## Other C libraries Arnd says some other C libraries will support 64-bit `time_t` but as a build-time option. I.e. libc and all applications must be built for either 32-bit or 64-bit `time_t`. ## Application compatibility issues If Unix timestamps are used in binary file formats or network protocols, these will need a new version. In some cases switching to unsigned 32-bit values is easy and will work for long enough. If `time_t` is used in library APIs then an ABI change is required. cppcheck(?) can find instances of this. Some libraries may use their own time types, so changing `time_t` won't be an ABI change but they will need to be updated anyway. Printing a value of type `time_t` with `printf()` and similar functions requires casting as there's no format specifier for it. It will be necessary to cast to `long long`, whereas previously `long` would work. The sparse static checker is supposed to be able to check for truncating conversions of `time_t`. ## Ongoing work in kernel and glibc A few people are working part time on this. Kernel patches are 60% done after 5 years, GNU libc about 75% (but only some of those changes have been applied). More people may be needed to speed this up and get it finished. The kernel side is coordinated through the y2038 mailing list: <https://lists.linaro.org/mailman/listinfo/y2038>. Patches are all sent to this mailing list. There is currently no git tree collecting them all. Help is wanted to: * Update device drivers * Review sound patches * Collect patches into single git tree The glibc side is coordinated through the general development mailing list: <https://www.gnu.org/software/libc/involved.html>, <https://sourceware.org/ml/libc-alpha/>. -- Ben Hutchings Software Developer, Codethink Ltd.

7 years, 5 months

[PATCH 1/7] timekeeping: consolidate timekeeping_inject_offset code

by Arnd Bergmann

The code to check the adjtimex() or clock_adjtime() arguments is spread out across multiple files for presumably only historic reasons. As a preparatation for a rework to get rid of the use of 'struct timeval' and 'struct timespec' in there, this moves all the portions into kernel/time/timekeeping.c and marks them as 'static'. The warp_clock() function here is not as closely related as the others, but I feel it still makes sense to move it here in order to consolidate all callers of timekeeping_inject_offset(). Signed-off-by: Arnd Bergmann <arnd(a)arndb.de> --- include/linux/time.h | 26 ---------- kernel/time/ntp.c | 61 ---------------------- kernel/time/ntp_internal.h | 1 - kernel/time/time.c | 36 +------------ kernel/time/timekeeping.c | 123 ++++++++++++++++++++++++++++++++++++++++++++- kernel/time/timekeeping.h | 2 +- 6 files changed, 123 insertions(+), 126 deletions(-) diff --git a/include/linux/time.h b/include/linux/time.h index 9bc1f945777c..c0fbad08448f 100644 --- a/include/linux/time.h +++ b/include/linux/time.h @@ -134,32 +134,6 @@ static inline bool timeval_valid(const struct timeval *tv) extern struct timespec timespec_trunc(struct timespec t, unsigned gran); -/* - * Validates if a timespec/timeval used to inject a time offset is valid. - * Offsets can be postive or negative. The value of the timeval/timespec - * is the sum of its fields, but *NOTE*: the field tv_usec/tv_nsec must - * always be non-negative. - */ -static inline bool timeval_inject_offset_valid(const struct timeval *tv) -{ - /* We don't check the tv_sec as it can be positive or negative */ - - /* Can't have more microseconds then a second */ - if (tv->tv_usec < 0 || tv->tv_usec >= USEC_PER_SEC) - return false; - return true; -} - -static inline bool timespec_inject_offset_valid(const struct timespec *ts) -{ - /* We don't check the tv_sec as it can be positive or negative */ - - /* Can't have more nanoseconds then a second */ - if (ts->tv_nsec < 0 || ts->tv_nsec >= NSEC_PER_SEC) - return false; - return true; -} - /* Some architectures do not supply their own clocksource. * This is mainly the case in architectures that get their * inter-tick times by reading the counter on their interval diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c index edf19cc53140..a5e702669d84 100644 --- a/kernel/time/ntp.c +++ b/kernel/time/ntp.c @@ -653,67 +653,6 @@ static inline void process_adjtimex_modes(struct timex *txc, } - -/** - * ntp_validate_timex - Ensures the timex is ok for use in do_adjtimex - */ -int ntp_validate_timex(struct timex *txc) -{ - if (txc->modes & ADJ_ADJTIME) { - /* singleshot must not be used with any other mode bits */ - if (!(txc->modes & ADJ_OFFSET_SINGLESHOT)) - return -EINVAL; - if (!(txc->modes & ADJ_OFFSET_READONLY) && - !capable(CAP_SYS_TIME)) - return -EPERM; - } else { - /* In order to modify anything, you gotta be super-user! */ - if (txc->modes && !capable(CAP_SYS_TIME)) - return -EPERM; - /* - * if the quartz is off by more than 10% then - * something is VERY wrong! - */ - if (txc->modes & ADJ_TICK && - (txc->tick < 900000/USER_HZ || - txc->tick > 1100000/USER_HZ)) - return -EINVAL; - } - - if (txc->modes & ADJ_SETOFFSET) { - /* In order to inject time, you gotta be super-user! */ - if (!capable(CAP_SYS_TIME)) - return -EPERM; - - if (txc->modes & ADJ_NANO) { - struct timespec ts; - - ts.tv_sec = txc->time.tv_sec; - ts.tv_nsec = txc->time.tv_usec; - if (!timespec_inject_offset_valid(&ts)) - return -EINVAL; - - } else { - if (!timeval_inject_offset_valid(&txc->time)) - return -EINVAL; - } - } - - /* - * Check for potential multiplication overflows that can - * only happen on 64-bit systems: - */ - if ((txc->modes & ADJ_FREQUENCY) && (BITS_PER_LONG == 64)) { - if (LLONG_MIN / PPM_SCALE > txc->freq) - return -EINVAL; - if (LLONG_MAX / PPM_SCALE < txc->freq) - return -EINVAL; - } - - return 0; -} - - /* * adjtimex mainly allows reading (and writing, if superuser) of * kernel time-keeping variables. used by xntpd. diff --git a/kernel/time/ntp_internal.h b/kernel/time/ntp_internal.h index d8a7c11fa71a..74b52cd48209 100644 --- a/kernel/time/ntp_internal.h +++ b/kernel/time/ntp_internal.h @@ -7,7 +7,6 @@ extern void ntp_clear(void); extern u64 ntp_tick_length(void); extern ktime_t ntp_get_next_leap(void); extern int second_overflow(time64_t secs); -extern int ntp_validate_timex(struct timex *); extern int __do_adjtimex(struct timex *, struct timespec64 *, s32 *); extern void __hardpps(const struct timespec64 *, const struct timespec64 *); #endif /* _LINUX_NTP_INTERNAL_H */ diff --git a/kernel/time/time.c b/kernel/time/time.c index 44a8c1402133..04684e294f00 100644 --- a/kernel/time/time.c +++ b/kernel/time/time.c @@ -158,40 +158,6 @@ SYSCALL_DEFINE2(gettimeofday, struct timeval __user *, tv, } /* - * Indicates if there is an offset between the system clock and the hardware - * clock/persistent clock/rtc. - */ -int persistent_clock_is_local; - -/* - * Adjust the time obtained from the CMOS to be UTC time instead of - * local time. - * - * This is ugly, but preferable to the alternatives. Otherwise we - * would either need to write a program to do it in /etc/rc (and risk - * confusion if the program gets run more than once; it would also be - * hard to make the program warp the clock precisely n hours) or - * compile in the timezone information into the kernel. Bad, bad.... - * - * - TYT, 1992-01-01 - * - * The best thing to do is to keep the CMOS clock in universal time (UTC) - * as real UNIX machines always do it. This avoids all headaches about - * daylight saving times and warping kernel clocks. - */ -static inline void warp_clock(void) -{ - if (sys_tz.tz_minuteswest != 0) { - struct timespec adjust; - - persistent_clock_is_local = 1; - adjust.tv_sec = sys_tz.tz_minuteswest * 60; - adjust.tv_nsec = 0; - timekeeping_inject_offset(&adjust); - } -} - -/* * In case for some reason the CMOS clock has not already been running * in UTC, but in some local time: The first time we set the timezone, * we will warp the clock so that it is ticking UTC time instead of @@ -224,7 +190,7 @@ int do_sys_settimeofday64(const struct timespec64 *tv, const struct timezone *tz if (firsttime) { firsttime = 0; if (!tv) - warp_clock(); + timekeeping_warp_clock(); } } if (tv) diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index 8af77006e937..679dbfbea419 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -1300,13 +1300,39 @@ int do_settimeofday64(const struct timespec64 *ts) } EXPORT_SYMBOL(do_settimeofday64); +/* + * Validates if a timespec/timeval used to inject a time offset is valid. + * Offsets can be postive or negative. The value of the timeval/timespec + * is the sum of its fields, but *NOTE*: the field tv_usec/tv_nsec must + * always be non-negative. + */ +static inline bool timeval_inject_offset_valid(const struct timeval *tv) +{ + /* We don't check the tv_sec as it can be positive or negative */ + + /* Can't have more microseconds then a second */ + if (tv->tv_usec < 0 || tv->tv_usec >= USEC_PER_SEC) + return false; + return true; +} + +static inline bool timespec_inject_offset_valid(const struct timespec *ts) +{ + /* We don't check the tv_sec as it can be positive or negative */ + + /* Can't have more nanoseconds then a second */ + if (ts->tv_nsec < 0 || ts->tv_nsec >= NSEC_PER_SEC) + return false; + return true; +} + /** * timekeeping_inject_offset - Adds or subtracts from the current time. * @tv: pointer to the timespec variable containing the offset * * Adds or subtracts an offset value from the current time. */ -int timekeeping_inject_offset(struct timespec *ts) +static int timekeeping_inject_offset(struct timespec *ts) { struct timekeeper *tk = &tk_core.timekeeper; unsigned long flags; @@ -1345,7 +1371,40 @@ int timekeeping_inject_offset(struct timespec *ts) return ret; } -EXPORT_SYMBOL(timekeeping_inject_offset); + +/* + * Indicates if there is an offset between the system clock and the hardware + * clock/persistent clock/rtc. + */ +int persistent_clock_is_local; + +/* + * Adjust the time obtained from the CMOS to be UTC time instead of + * local time. + * + * This is ugly, but preferable to the alternatives. Otherwise we + * would either need to write a program to do it in /etc/rc (and risk + * confusion if the program gets run more than once; it would also be + * hard to make the program warp the clock precisely n hours) or + * compile in the timezone information into the kernel. Bad, bad.... + * + * - TYT, 1992-01-01 + * + * The best thing to do is to keep the CMOS clock in universal time (UTC) + * as real UNIX machines always do it. This avoids all headaches about + * daylight saving times and warping kernel clocks. + */ +void timekeeping_warp_clock(void) +{ + if (sys_tz.tz_minuteswest != 0) { + struct timespec adjust; + + persistent_clock_is_local = 1; + adjust.tv_sec = sys_tz.tz_minuteswest * 60; + adjust.tv_nsec = 0; + timekeeping_inject_offset(&adjust); + } +} /** * __timekeeping_set_tai_offset - Sets the TAI offset from UTC and monotonic @@ -2290,6 +2349,66 @@ ktime_t ktime_get_update_offsets_now(unsigned int *cwsseq, ktime_t *offs_real, } /** + * ntp_validate_timex - Ensures the timex is ok for use in do_adjtimex + */ +static int ntp_validate_timex(struct timex *txc) +{ + if (txc->modes & ADJ_ADJTIME) { + /* singleshot must not be used with any other mode bits */ + if (!(txc->modes & ADJ_OFFSET_SINGLESHOT)) + return -EINVAL; + if (!(txc->modes & ADJ_OFFSET_READONLY) && + !capable(CAP_SYS_TIME)) + return -EPERM; + } else { + /* In order to modify anything, you gotta be super-user! */ + if (txc->modes && !capable(CAP_SYS_TIME)) + return -EPERM; + /* + * if the quartz is off by more than 10% then + * something is VERY wrong! + */ + if (txc->modes & ADJ_TICK && + (txc->tick < 900000/USER_HZ || + txc->tick > 1100000/USER_HZ)) + return -EINVAL; + } + + if (txc->modes & ADJ_SETOFFSET) { + /* In order to inject time, you gotta be super-user! */ + if (!capable(CAP_SYS_TIME)) + return -EPERM; + + if (txc->modes & ADJ_NANO) { + struct timespec ts; + + ts.tv_sec = txc->time.tv_sec; + ts.tv_nsec = txc->time.tv_usec; + if (!timespec_inject_offset_valid(&ts)) + return -EINVAL; + + } else { + if (!timeval_inject_offset_valid(&txc->time)) + return -EINVAL; + } + } + + /* + * Check for potential multiplication overflows that can + * only happen on 64-bit systems: + */ + if ((txc->modes & ADJ_FREQUENCY) && (BITS_PER_LONG == 64)) { + if (LLONG_MIN / PPM_SCALE > txc->freq) + return -EINVAL; + if (LLONG_MAX / PPM_SCALE < txc->freq) + return -EINVAL; + } + + return 0; +} + + +/** * do_adjtimex() - Accessor function to NTP __do_adjtimex function */ int do_adjtimex(struct timex *txc) diff --git a/kernel/time/timekeeping.h b/kernel/time/timekeeping.h index d0914676d4c5..44aec7893cdd 100644 --- a/kernel/time/timekeeping.h +++ b/kernel/time/timekeeping.h @@ -10,7 +10,7 @@ extern ktime_t ktime_get_update_offsets_now(unsigned int *cwsseq, extern int timekeeping_valid_for_hres(void); extern u64 timekeeping_max_deferment(void); -extern int timekeeping_inject_offset(struct timespec *ts); +extern void timekeeping_warp_clock(void); extern int timekeeping_suspend(void); extern void timekeeping_resume(void); -- 2.9.0

7 years, 5 months

[PATCH] drm: via: use ktime_get() instead of do_gettimeofday

by Arnd Bergmann

We want to remove uses of do_gettimeofday() from the kernel since the resulting timeval structure overflows in 2038. This is not a problem for this particular use, but do_gettimeofday() is also not an appropriate method for measuring time intervals, since it requires a conversion into microseconds and is complicated to work with. ktime_get() is a better replacement, as it works with the monontonic kernel timebase and requires a minimum of computation. I'm slightly changing the output from microseconds to nanoseconds here, to avoid introducing a new division operation. This should be fine since the value is only used for debugging. Signed-off-by: Arnd Bergmann <arnd(a)arndb.de> --- drivers/gpu/drm/via/via_drv.h | 4 ++-- drivers/gpu/drm/via/via_irq.c | 21 +++++++-------------- 2 files changed, 9 insertions(+), 16 deletions(-) diff --git a/drivers/gpu/drm/via/via_drv.h b/drivers/gpu/drm/via/via_drv.h index 9873942ca8f4..6d1ae834484c 100644 --- a/drivers/gpu/drm/via/via_drv.h +++ b/drivers/gpu/drm/via/via_drv.h @@ -74,9 +74,9 @@ typedef struct drm_via_private { volatile uint32_t *last_pause_ptr; volatile uint32_t *hw_addr_ptr; drm_via_ring_buffer_t ring; - struct timeval last_vblank; + ktime_t last_vblank; int last_vblank_valid; - unsigned usec_per_vblank; + ktime_t nsec_per_vblank; atomic_t vbl_received; drm_via_state_t hc_state; char pci_buf[VIA_PCI_BUF_SIZE]; diff --git a/drivers/gpu/drm/via/via_irq.c b/drivers/gpu/drm/via/via_irq.c index ea8172c747a2..24e71578af4d 100644 --- a/drivers/gpu/drm/via/via_irq.c +++ b/drivers/gpu/drm/via/via_irq.c @@ -88,13 +88,6 @@ static int via_num_unichrome = ARRAY_SIZE(via_unichrome_irqs); static int via_irqmap_unichrome[] = {-1, -1, -1, 0, -1, 1}; -static unsigned time_diff(struct timeval *now, struct timeval *then) -{ - return (now->tv_usec >= then->tv_usec) ? - now->tv_usec - then->tv_usec : - 1000000 - (then->tv_usec - now->tv_usec); -} - u32 via_get_vblank_counter(struct drm_device *dev, unsigned int pipe) { drm_via_private_t *dev_priv = dev->dev_private; @@ -111,7 +104,7 @@ irqreturn_t via_driver_irq_handler(int irq, void *arg) drm_via_private_t *dev_priv = (drm_via_private_t *) dev->dev_private; u32 status; int handled = 0; - struct timeval cur_vblank; + ktime_t cur_vblank; drm_via_irq_t *cur_irq = dev_priv->via_irqs; int i; @@ -119,18 +112,18 @@ irqreturn_t via_driver_irq_handler(int irq, void *arg) if (status & VIA_IRQ_VBLANK_PENDING) { atomic_inc(&dev_priv->vbl_received); if (!(atomic_read(&dev_priv->vbl_received) & 0x0F)) { - do_gettimeofday(&cur_vblank); + cur_vblank = ktime_get(); if (dev_priv->last_vblank_valid) { - dev_priv->usec_per_vblank = - time_diff(&cur_vblank, - &dev_priv->last_vblank) >> 4; + dev_priv->nsec_per_vblank = + ktime_sub(cur_vblank, + dev_priv->last_vblank) >> 4; } dev_priv->last_vblank = cur_vblank; dev_priv->last_vblank_valid = 1; } if (!(atomic_read(&dev_priv->vbl_received) & 0xFF)) { - DRM_DEBUG("US per vblank is: %u\n", - dev_priv->usec_per_vblank); + DRM_DEBUG("nsec per vblank is: %llu\n", + ktime_to_ns(dev_priv->nsec_per_vblank)); } drm_handle_vblank(dev, 0); handled = 1; -- 2.9.0

7 years, 5 months

[PATCH v2 1/2] isofs: fix timestamps beyond 2027

by Arnd Bergmann

isofs uses a 'char' variable to load the number of years since 1900 for an inode timestamp. On architectures that use a signed char type by default, this results in an invalid date for anything beyond 2027. This changes the function argument to a 'u8' array, which is defined the same way on all architectures, and unambiguously lets us use years until 2155. This should be backported to all kernels that might still be in use by that date. Cc: stable(a)vger.kernel.org Signed-off-by: Arnd Bergmann <arnd(a)arndb.de> --- v2: change function prototype instead of adding a cast --- fs/isofs/isofs.h | 2 +- fs/isofs/rock.h | 2 +- fs/isofs/util.c | 4 ++-- 3 files changed, 4 insertions(+), 4 deletions(-) diff --git a/fs/isofs/isofs.h b/fs/isofs/isofs.h index 133a456b0425..bd4047585431 100644 --- a/fs/isofs/isofs.h +++ b/fs/isofs/isofs.h @@ -106,7 +106,7 @@ static inline unsigned int isonum_733(char *p) /* Ignore bigendian datum due to broken mastering programs */ return get_unaligned_le32(p); } -extern int iso_date(char *, int); +extern int iso_date(u8 *, int); struct inode; /* To make gcc happy */ diff --git a/fs/isofs/rock.h b/fs/isofs/rock.h index ed09e2b08637..f835976ce033 100644 --- a/fs/isofs/rock.h +++ b/fs/isofs/rock.h @@ -65,7 +65,7 @@ struct RR_PL_s { }; struct stamp { - char time[7]; + __u8 time[7]; /* actually 6 unsigned, 1 signed */ } __attribute__ ((packed)); struct RR_TF_s { diff --git a/fs/isofs/util.c b/fs/isofs/util.c index 005a15cfd30a..335f62db0b53 100644 --- a/fs/isofs/util.c +++ b/fs/isofs/util.c @@ -15,12 +15,12 @@ * to GMT. Thus we should always be correct. */ -int iso_date(char * p, int flag) +int iso_date(u8 *p, int flag) { int year, month, day, hour, minute, second, tz; int crtime; - year = p[0]; + year = (int)(u8)p[0]; month = p[1]; day = p[2]; hour = p[3]; -- 2.9.0

7 years, 6 months

[PATCH] f2fs: avoid using timespec

by Arnd Bergmann

All uses of timespec are deprecated, and this one is not particularly useful, as the documented method for converting seconds to jiffies is to multiply by 'HZ'. Signed-off-by: Arnd Bergmann <arnd(a)arndb.de> --- fs/f2fs/f2fs.h | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index 4b4a72f392be..b9a4a5db426c 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -1178,8 +1178,7 @@ static inline void f2fs_update_time(struct f2fs_sb_info *sbi, int type) static inline bool f2fs_time_over(struct f2fs_sb_info *sbi, int type) { - struct timespec ts = {sbi->interval_time[type], 0}; - unsigned long interval = timespec_to_jiffies(&ts); + unsigned long interval = sbi->interval_time[type] * HZ; return time_after(jiffies, sbi->last_time[type] + interval); } -- 2.9.0

7 years, 6 months

[PATCH] nfds: avoid gettimeofday for nfssvc_boot time

by Arnd Bergmann

do_gettimeofday() is deprecated and we should generally use time64_t based functions instead. In case of nfsd, all three users of nfssvc_boot only use the initial time as a unique token, and are not affected by it overflowing, so they are not affected by the y2038 overflow. This converts the structure to timespec64 anyway and adds comments to all uses, to document that we have thought about it and avoid having to look at it again. Signed-off-by: Arnd Bergmann <arnd(a)arndb.de> --- fs/nfsd/netns.h | 2 +- fs/nfsd/nfs3xdr.c | 10 ++++++---- fs/nfsd/nfs4proc.c | 5 +++-- fs/nfsd/nfssvc.c | 2 +- 4 files changed, 11 insertions(+), 8 deletions(-) diff --git a/fs/nfsd/netns.h b/fs/nfsd/netns.h index 3714231a9d0f..1c91391f4805 100644 --- a/fs/nfsd/netns.h +++ b/fs/nfsd/netns.h @@ -107,7 +107,7 @@ struct nfsd_net { bool lockd_up; /* Time of server startup */ - struct timeval nfssvc_boot; + struct timespec64 nfssvc_boot; /* * Max number of connections this nfsd container will allow. Defaults diff --git a/fs/nfsd/nfs3xdr.c b/fs/nfsd/nfs3xdr.c index bf444b664011..3579e0ae1131 100644 --- a/fs/nfsd/nfs3xdr.c +++ b/fs/nfsd/nfs3xdr.c @@ -747,8 +747,9 @@ nfs3svc_encode_writeres(struct svc_rqst *rqstp, __be32 *p) if (resp->status == 0) { *p++ = htonl(resp->count); *p++ = htonl(resp->committed); - *p++ = htonl(nn->nfssvc_boot.tv_sec); - *p++ = htonl(nn->nfssvc_boot.tv_usec); + /* unique identifier, y2038 overflow can be ignored */ + *p++ = htonl((u32)nn->nfssvc_boot.tv_sec); + *p++ = htonl(nn->nfssvc_boot.tv_nsec); } return xdr_ressize_check(rqstp, p); } @@ -1118,8 +1119,9 @@ nfs3svc_encode_commitres(struct svc_rqst *rqstp, __be32 *p) p = encode_wcc_data(rqstp, p, &resp->fh); /* Write verifier */ if (resp->status == 0) { - *p++ = htonl(nn->nfssvc_boot.tv_sec); - *p++ = htonl(nn->nfssvc_boot.tv_usec); + /* unique identifier, y2038 overflow can be ignored */ + *p++ = htonl((u32)nn->nfssvc_boot.tv_sec); + *p++ = htonl(nn->nfssvc_boot.tv_nsec); } return xdr_ressize_check(rqstp, p); } diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index 7896f841482e..008ea0b627d0 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -564,10 +564,11 @@ static void gen_boot_verifier(nfs4_verifier *verifier, struct net *net) /* * This is opaque to client, so no need to byte-swap. Use - * __force to keep sparse happy + * __force to keep sparse happy. y2038 time_t overflow is + * irrelevant in this usage. */ verf[0] = (__force __be32)nn->nfssvc_boot.tv_sec; - verf[1] = (__force __be32)nn->nfssvc_boot.tv_usec; + verf[1] = (__force __be32)nn->nfssvc_boot.tv_nsec; memcpy(verifier->data, verf, sizeof(verifier->data)); } diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c index 6bbc717f40f2..28ff3e078af6 100644 --- a/fs/nfsd/nfssvc.c +++ b/fs/nfsd/nfssvc.c @@ -516,7 +516,7 @@ int nfsd_create_serv(struct net *net) register_inet6addr_notifier(&nfsd_inet6addr_notifier); #endif } - do_gettimeofday(&nn->nfssvc_boot); /* record boot time */ + ktime_get_real_ts64(&nn->nfssvc_boot); /* record boot time */ return 0; } -- 2.9.0

7 years, 6 months

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

Y2038