Y2038

y2038@lists.linaro.org

1132 discussions

[PATCH] ath10k: use 64-bit crash dump timestamps

by Arnd Bergmann

The dump format uses 64-bit timestamps already, but calling getnstimeofday() only returns a 32-bit number on 32-bit architectures, so that will overflow in y2038. This changes it to use ktime_get_real_ts64() instead. Signed-off-by: Arnd Bergmann <arnd(a)arndb.de> --- drivers/net/wireless/ath/ath10k/core.h | 2 +- drivers/net/wireless/ath/ath10k/debug.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wireless/ath/ath10k/core.h index 643041ef3271..dd3c6c63aae1 100644 --- a/drivers/net/wireless/ath/ath10k/core.h +++ b/drivers/net/wireless/ath/ath10k/core.h @@ -463,7 +463,7 @@ struct ath10k_fw_crash_data { bool crashed_since_read; guid_t guid; - struct timespec timestamp; + struct timespec64 timestamp; __le32 registers[REG_DUMP_COUNT_QCA988X]; struct ath10k_ce_crash_data ce_crash_data[CE_COUNT_MAX]; }; diff --git a/drivers/net/wireless/ath/ath10k/debug.c b/drivers/net/wireless/ath/ath10k/debug.c index df514507d3f1..66498e3c0f91 100644 --- a/drivers/net/wireless/ath/ath10k/debug.c +++ b/drivers/net/wireless/ath/ath10k/debug.c @@ -720,7 +720,7 @@ ath10k_debug_get_new_fw_crash_data(struct ath10k *ar) crash_data->crashed_since_read = true; guid_gen(&crash_data->guid); - getnstimeofday(&crash_data->timestamp); + ktime_get_real_ts64(&crash_data->timestamp); return crash_data; } -- 2.9.0

7 years, 4 months

[PATCH] wlcore: use boottime for fw time sync

by Arnd Bergmann

Using getnstimeofday()/timespec_to_ns() causes an overflow on 32-bit architectures in 2038, and may suffer from time jumps due to settimeofday() or leap seconds. I don't see a reason why this needs to be UTC, so either monotonic or boot time would be better here. Assuming that the fw time keeps running during suspend, boottime is better than monotonic, and ktime_get_boot_ns() will also save the additional conversion to nanoseconds. Signed-off-by: Arnd Bergmann <arnd(a)arndb.de> --- drivers/net/wireless/ti/wlcore/main.c | 4 +--- drivers/net/wireless/ti/wlcore/tx.c | 4 +--- 2 files changed, 2 insertions(+), 6 deletions(-) diff --git a/drivers/net/wireless/ti/wlcore/main.c b/drivers/net/wireless/ti/wlcore/main.c index d47921a84509..946a7124b09e 100644 --- a/drivers/net/wireless/ti/wlcore/main.c +++ b/drivers/net/wireless/ti/wlcore/main.c @@ -388,7 +388,6 @@ static void wl12xx_irq_update_links_status(struct wl1271 *wl, static int wlcore_fw_status(struct wl1271 *wl, struct wl_fw_status *status) { struct wl12xx_vif *wlvif; - struct timespec ts; u32 old_tx_blk_count = wl->tx_blocks_available; int avail, freed_blocks; int i; @@ -485,8 +484,7 @@ static int wlcore_fw_status(struct wl1271 *wl, struct wl_fw_status *status) } /* update the host-chipset time offset */ - getnstimeofday(&ts); - wl->time_offset = (timespec_to_ns(&ts) >> 10) - + wl->time_offset = (ktime_get_boot_ns() >> 10) - (s64)(status->fw_localtime); wl->fw_fast_lnk_map = status->link_fast_bitmap; diff --git a/drivers/net/wireless/ti/wlcore/tx.c b/drivers/net/wireless/ti/wlcore/tx.c index a3f5e9ca492a..00e9b4624dcf 100644 --- a/drivers/net/wireless/ti/wlcore/tx.c +++ b/drivers/net/wireless/ti/wlcore/tx.c @@ -264,7 +264,6 @@ static void wl1271_tx_fill_hdr(struct wl1271 *wl, struct wl12xx_vif *wlvif, struct sk_buff *skb, u32 extra, struct ieee80211_tx_info *control, u8 hlid) { - struct timespec ts; struct wl1271_tx_hw_descr *desc; int ac, rate_idx; s64 hosttime; @@ -287,8 +286,7 @@ static void wl1271_tx_fill_hdr(struct wl1271 *wl, struct wl12xx_vif *wlvif, } /* configure packet life time */ - getnstimeofday(&ts); - hosttime = (timespec_to_ns(&ts) >> 10); + hosttime = (ktime_get_boot_ns() >> 10); desc->start_time = cpu_to_le32(hosttime - wl->time_offset); is_dummy = wl12xx_is_dummy_packet(wl, skb); -- 2.9.0

7 years, 4 months

[PATCH] kdb: use __ktime_get_real_seconds instead of __current_kernel_time

by Arnd Bergmann

kdb is the only user of the __current_kernel_time() interface, which is not y2038 safe and should be removed at some point. The kdb code also goes to great lengths to print the time in a human-readable format from 'struct timespec', again using a non-y2038-safe re-implementation of the generic time_to_tm() code. Using __current_kernel_time() here is necessary since the regular accessors that require a sequence lock might hang when called during the xtime update. However, this is safe in the particular case since kdb is only interested in the tv_sec field that is updated atomically. In order to make this y2038-safe, I'm converting the code to the generic time64_to_tm helper, but that introduces the problem that we have no interface like __current_kernel_time() that provides a 64-bit timestamp in a lockless, safe and architecture-independent way. I have multiple ideas for how to solve that: - __ktime_get_real_seconds() is lockless, but can return incorrect results on 32-bit architectures in the special case that we are in the process of changing the time across the epoch, either during the timer tick that overflows the seconds in 2038, or while calling settimeofday. - ktime_get_real_fast_ns() would work in this context, but does require a call into the clocksource driver to return a high-resolution timestamp. This may have undesired side-effects in the debugger, since we want to limit the interactions with the rest of the kernel. - Adding a ktime_get_real_fast_seconds() based on tk_fast_mono plus tkr->base_real without the tk_clock_read() delta. Not sure about the value of adding yet another interface here. - Changing the existing ktime_get_real_seconds() to use tk_fast_mono on 32-bit architectures rather than xtime_sec. I think this could work, but am not entirely sure if this is an improvement. I picked the first of those for simplicity here. It's technically not correct but probably good enough as the time is only used for the debugging output and the race will likely never be hit in practice. Another downside is having to move the declaration into a public header file. Let me know if anyone has a different preference. Cc: Andy Shevchenko <andriy.shevchenko(a)linux.intel.com> Link: https://patchwork.kernel.org/patch/9775309/ Signed-off-by: Arnd Bergmann <arnd(a)arndb.de> --- include/linux/timekeeping.h | 1 + kernel/debug/kdb/kdb_main.c | 45 +++++--------------------------------- kernel/time/timekeeping_internal.h | 2 -- 3 files changed, 6 insertions(+), 42 deletions(-) diff --git a/include/linux/timekeeping.h b/include/linux/timekeeping.h index eb98cbdbb323..9b59473556fe 100644 --- a/include/linux/timekeeping.h +++ b/include/linux/timekeeping.h @@ -41,6 +41,7 @@ struct timespec64 get_monotonic_coarse64(void); extern void getrawmonotonic64(struct timespec64 *ts); extern void ktime_get_ts64(struct timespec64 *ts); extern time64_t ktime_get_seconds(void); +extern time64_t __ktime_get_real_seconds(void); extern time64_t ktime_get_real_seconds(void); extern int __getnstimeofday64(struct timespec64 *tv); diff --git a/kernel/debug/kdb/kdb_main.c b/kernel/debug/kdb/kdb_main.c index c8146d53ca67..69e70f4021fe 100644 --- a/kernel/debug/kdb/kdb_main.c +++ b/kernel/debug/kdb/kdb_main.c @@ -2479,41 +2479,6 @@ static int kdb_kill(int argc, const char **argv) return 0; } -struct kdb_tm { - int tm_sec; /* seconds */ - int tm_min; /* minutes */ - int tm_hour; /* hours */ - int tm_mday; /* day of the month */ - int tm_mon; /* month */ - int tm_year; /* year */ -}; - -static void kdb_gmtime(struct timespec *tv, struct kdb_tm *tm) -{ - /* This will work from 1970-2099, 2100 is not a leap year */ - static int mon_day[] = { 31, 29, 31, 30, 31, 30, 31, - 31, 30, 31, 30, 31 }; - memset(tm, 0, sizeof(*tm)); - tm->tm_sec = tv->tv_sec % (24 * 60 * 60); - tm->tm_mday = tv->tv_sec / (24 * 60 * 60) + - (2 * 365 + 1); /* shift base from 1970 to 1968 */ - tm->tm_min = tm->tm_sec / 60 % 60; - tm->tm_hour = tm->tm_sec / 60 / 60; - tm->tm_sec = tm->tm_sec % 60; - tm->tm_year = 68 + 4*(tm->tm_mday / (4*365+1)); - tm->tm_mday %= (4*365+1); - mon_day[1] = 29; - while (tm->tm_mday >= mon_day[tm->tm_mon]) { - tm->tm_mday -= mon_day[tm->tm_mon]; - if (++tm->tm_mon == 12) { - tm->tm_mon = 0; - ++tm->tm_year; - mon_day[1] = 28; - } - } - ++tm->tm_mday; -} - /* * Most of this code has been lifted from kernel/timer.c::sys_sysinfo(). * I cannot call that code directly from kdb, it has an unconditional @@ -2539,8 +2504,8 @@ static void kdb_sysinfo(struct sysinfo *val) */ static int kdb_summary(int argc, const char **argv) { - struct timespec now; - struct kdb_tm tm; + time64_t now; + struct tm tm; struct sysinfo val; if (argc) @@ -2554,9 +2519,9 @@ static int kdb_summary(int argc, const char **argv) kdb_printf("domainname %s\n", init_uts_ns.name.domainname); kdb_printf("ccversion %s\n", __stringify(CCVERSION)); - now = __current_kernel_time(); - kdb_gmtime(&now, &tm); - kdb_printf("date %04d-%02d-%02d %02d:%02d:%02d " + now = __ktime_get_real_seconds(); + time64_to_tm(now, 0, &tm); + kdb_printf("date %04ld-%02d-%02d %02d:%02d:%02d " "tz_minuteswest %d\n", 1900+tm.tm_year, tm.tm_mon+1, tm.tm_mday, tm.tm_hour, tm.tm_min, tm.tm_sec, diff --git a/kernel/time/timekeeping_internal.h b/kernel/time/timekeeping_internal.h index 9a18f121f399..58e79485de1b 100644 --- a/kernel/time/timekeeping_internal.h +++ b/kernel/time/timekeeping_internal.h @@ -30,6 +30,4 @@ static inline u64 clocksource_delta(u64 now, u64 last, u64 mask) } #endif -extern time64_t __ktime_get_real_seconds(void); - #endif /* _TIMEKEEPING_INTERNAL_H */ -- 2.9.0

7 years, 4 months

[PATCH v3 0/4] Make input drivers y2038 safe

by Deepa Dinamani

The series is aimed at making input events y2038 safe. It extends the lifetime of the realtime timestamps in the events to year 2106. The series is also a necessary update as glibc is set to provide 64 bit time_t support for 32 bit binaries. glibc plan is detailed at https://sourceware.org/glibc/wiki/Y2038ProofnessDesign . The series is a result of discussions with Arnd Bergmann and Dmitry Torokhov at last Plumbers. The plan is to deprecate realtime timestamps anyway as they are not appropriate for these timestamps as noted in the patch a80b83b7b8 by John Stultz. The design also updates the format of the input events read/ written to the device nodes. This breaks 32 bit interface to the input events at compile time as preferred by the maintainer. The userspace library changes to libevdev, libuinput and mtdev will be posted to the respective mailing groups for review. Changes from v2: * Updated the design to break 32 bit interfaces at compile time. Changes from v1: * Updated changes according to review comments. * Posted userspace library changes that go along with the series. Deepa Dinamani (4): uinput: Add ioctl for using monotonic/ boot times input: evdev: Replace timeval with timespec64 input: Deprecate real timestamps beyond year 2106 input: serio: Replace timeval by timespec64 drivers/input/evdev.c | 43 +++++++++++++++++++----------- drivers/input/input-compat.c | 11 ++++---- drivers/input/input-compat.h | 3 ++- drivers/input/misc/uinput.c | 57 +++++++++++++++++++++++++++++++++++++++- drivers/input/serio/hil_mlc.c | 37 +++++++++++++------------- drivers/input/serio/hp_sdc.c | 17 ++++++------ drivers/input/serio/hp_sdc_mlc.c | 10 +++---- include/linux/hil_mlc.h | 6 ++--- include/linux/hp_sdc.h | 6 ++--- include/uapi/linux/input.h | 12 ++++++++- include/uapi/linux/uinput.h | 3 +++ 11 files changed, 143 insertions(+), 62 deletions(-) base-commit: b0a84f19a5161418d4360cd57603e94ed489915e -- 2.14.1

7 years, 4 months

[PATCH] s390/dasd: remove 'struct timespec' usage

by Arnd Bergmann

getnstimeofday() and timespec are deprecated since they can overflow on 32-bit architectures. This simply changes to the explicitly typed timespec64 version that doesn't have that problem. It would be nice to also convert to monotonic timestamps and call ktime_get_ts64() rather than ktime_get_real_ts64(), but that would be a user-visible change. Signed-off-by: Arnd Bergmann <arnd(a)arndb.de> --- drivers/s390/block/dasd.c | 10 +++++----- drivers/s390/block/dasd_int.h | 2 +- 2 files changed, 6 insertions(+), 6 deletions(-) diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c index 0f1ff0813493..f48b3d5082cf 100644 --- a/drivers/s390/block/dasd.c +++ b/drivers/s390/block/dasd.c @@ -758,7 +758,7 @@ static void dasd_profile_end_add_data(struct dasd_profile_info *data, /* in case of an overflow, reset the whole profile */ if (data->dasd_io_reqs == UINT_MAX) { memset(data, 0, sizeof(*data)); - getnstimeofday(&data->starttod); + ktime_get_real_ts64(&data->starttod); } data->dasd_io_reqs++; data->dasd_io_sects += sectors; @@ -893,7 +893,7 @@ void dasd_profile_reset(struct dasd_profile *profile) return; } memset(data, 0, sizeof(*data)); - getnstimeofday(&data->starttod); + ktime_get_real_ts64(&data->starttod); spin_unlock_bh(&profile->lock); } @@ -910,7 +910,7 @@ int dasd_profile_on(struct dasd_profile *profile) kfree(data); return 0; } - getnstimeofday(&data->starttod); + ktime_get_real_ts64(&data->starttod); profile->data = data; spin_unlock_bh(&profile->lock); return 0; @@ -994,8 +994,8 @@ static void dasd_stats_array(struct seq_file *m, unsigned int *array) static void dasd_stats_seq_print(struct seq_file *m, struct dasd_profile_info *data) { - seq_printf(m, "start_time %ld.%09ld\n", - data->starttod.tv_sec, data->starttod.tv_nsec); + seq_printf(m, "start_time %lld.%09ld\n", + (s64)data->starttod.tv_sec, data->starttod.tv_nsec); seq_printf(m, "total_requests %u\n", data->dasd_io_reqs); seq_printf(m, "total_sectors %u\n", data->dasd_io_sects); seq_printf(m, "total_pav %u\n", data->dasd_io_alias); diff --git a/drivers/s390/block/dasd_int.h b/drivers/s390/block/dasd_int.h index b095a23bcc0c..96709b1a7bf8 100644 --- a/drivers/s390/block/dasd_int.h +++ b/drivers/s390/block/dasd_int.h @@ -441,7 +441,7 @@ struct dasd_profile_info { unsigned int dasd_io_nr_req[32]; /* hist. of # of requests in chanq */ /* new data */ - struct timespec starttod; /* time of start or last reset */ + struct timespec64 starttod; /* time of start or last reset */ unsigned int dasd_io_alias; /* requests using an alias */ unsigned int dasd_io_tpm; /* requests using transport mode */ unsigned int dasd_read_reqs; /* total number of read requests */ -- 2.9.0

7 years, 5 months

[PATCH] atm: mpoa: remove 32-bit timekeeping

by Arnd Bergmann

From: Tina Ruchandani <ruchandani.tina(a)gmail.com> net/atm/mpoa_* files use 'struct timeval' to store event timestamps. struct timeval uses a 32-bit seconds field which will overflow in the year 2038 and beyond. Morever, the timestamps are being compared only to get seconds elapsed, so struct timeval which stores a seconds and microseconds field is an overkill. This patch replaces the use of struct timeval with time64_t to store a 64-bit seconds field. Signed-off-by: Tina Ruchandani <ruchandani.tina(a)gmail.com> Signed-off-by: Arnd Bergmann <arnd(a)arndb.de> --- net/atm/common.c | 2 +- net/atm/mpc.c | 9 +++++---- net/atm/mpoa_caches.c | 48 +++++++++++++++++++++++------------------------- net/atm/mpoa_caches.h | 9 +++++---- net/atm/mpoa_proc.c | 15 +++++++++------ 5 files changed, 43 insertions(+), 40 deletions(-) diff --git a/net/atm/common.c b/net/atm/common.c index 8a4f99114cd2..5763fd241dc3 100644 --- a/net/atm/common.c +++ b/net/atm/common.c @@ -14,7 +14,7 @@ #include <linux/capability.h> #include <linux/mm.h> #include <linux/sched/signal.h> -#include <linux/time.h> /* struct timeval */ +#include <linux/time64.h> /* 64-bit time for seconds */ #include <linux/skbuff.h> #include <linux/bitops.h> #include <linux/init.h> diff --git a/net/atm/mpc.c b/net/atm/mpc.c index 7c6a1cc760a2..31e0dcb970f8 100644 --- a/net/atm/mpc.c +++ b/net/atm/mpc.c @@ -1089,7 +1089,7 @@ static void MPOA_trigger_rcvd(struct k_message *msg, struct mpoa_client *mpc) msg->type = SND_MPOA_RES_RQST; msg->content.in_info = entry->ctrl_info; msg_to_mpoad(msg, mpc); - do_gettimeofday(&(entry->reply_wait)); + entry->reply_wait = ktime_get_seconds(); mpc->in_ops->put(entry); return; } @@ -1099,7 +1099,7 @@ static void MPOA_trigger_rcvd(struct k_message *msg, struct mpoa_client *mpc) msg->type = SND_MPOA_RES_RQST; msg->content.in_info = entry->ctrl_info; msg_to_mpoad(msg, mpc); - do_gettimeofday(&(entry->reply_wait)); + entry->reply_wait = ktime_get_seconds(); mpc->in_ops->put(entry); return; } @@ -1175,8 +1175,9 @@ static void MPOA_res_reply_rcvd(struct k_message *msg, struct mpoa_client *mpc) } entry->ctrl_info = msg->content.in_info; - do_gettimeofday(&(entry->tv)); - do_gettimeofday(&(entry->reply_wait)); /* Used in refreshing func from now on */ + entry->time = ktime_get_seconds(); + /* Used in refreshing func from now on */ + entry->reply_wait = ktime_get_seconds(); entry->refresh_time = 0; ddprintk_cont("entry->shortcut = %p\n", entry->shortcut); diff --git a/net/atm/mpoa_caches.c b/net/atm/mpoa_caches.c index e01450bb32d6..4bb418313720 100644 --- a/net/atm/mpoa_caches.c +++ b/net/atm/mpoa_caches.c @@ -117,7 +117,7 @@ static in_cache_entry *in_cache_add_entry(__be32 dst_ip, memcpy(entry->MPS_ctrl_ATM_addr, client->mps_ctrl_addr, ATM_ESA_LEN); entry->ctrl_info.in_dst_ip = dst_ip; - do_gettimeofday(&(entry->tv)); + entry->time = ktime_get_seconds(); entry->retry_time = client->parameters.mpc_p4; entry->count = 1; entry->entry_state = INGRESS_INVALID; @@ -148,7 +148,7 @@ static int cache_hit(in_cache_entry *entry, struct mpoa_client *mpc) if (qos != NULL) msg.qos = qos->qos; msg_to_mpoad(&msg, mpc); - do_gettimeofday(&(entry->reply_wait)); + entry->reply_wait = ktime_get_seconds(); entry->entry_state = INGRESS_RESOLVING; } if (entry->shortcut != NULL) @@ -171,7 +171,7 @@ static int cache_hit(in_cache_entry *entry, struct mpoa_client *mpc) if (qos != NULL) msg.qos = qos->qos; msg_to_mpoad(&msg, mpc); - do_gettimeofday(&(entry->reply_wait)); + entry->reply_wait = ktime_get_seconds(); } return CLOSED; @@ -227,17 +227,16 @@ static void in_cache_remove_entry(in_cache_entry *entry, static void clear_count_and_expired(struct mpoa_client *client) { in_cache_entry *entry, *next_entry; - struct timeval now; + time64_t now; - do_gettimeofday(&now); + now = ktime_get_seconds(); write_lock_bh(&client->ingress_lock); entry = client->in_cache; while (entry != NULL) { entry->count = 0; next_entry = entry->next; - if ((now.tv_sec - entry->tv.tv_sec) - > entry->ctrl_info.holding_time) { + if ((now - entry->time) > entry->ctrl_info.holding_time) { dprintk("holding time expired, ip = %pI4\n", &entry->ctrl_info.in_dst_ip); client->in_ops->remove_entry(entry, client); @@ -253,35 +252,35 @@ static void check_resolving_entries(struct mpoa_client *client) struct atm_mpoa_qos *qos; in_cache_entry *entry; - struct timeval now; + time64_t now; struct k_message msg; - do_gettimeofday(&now); + now = ktime_get_seconds(); read_lock_bh(&client->ingress_lock); entry = client->in_cache; while (entry != NULL) { if (entry->entry_state == INGRESS_RESOLVING) { - if ((now.tv_sec - entry->hold_down.tv_sec) < - client->parameters.mpc_p6) { + + if ((now - entry->hold_down) + < client->parameters.mpc_p6) { entry = entry->next; /* Entry in hold down */ continue; } - if ((now.tv_sec - entry->reply_wait.tv_sec) > - entry->retry_time) { + if ((now - entry->reply_wait) > entry->retry_time) { entry->retry_time = MPC_C1 * (entry->retry_time); /* * Retry time maximum exceeded, * put entry in hold down. */ if (entry->retry_time > client->parameters.mpc_p5) { - do_gettimeofday(&(entry->hold_down)); + entry->hold_down = ktime_get_seconds(); entry->retry_time = client->parameters.mpc_p4; entry = entry->next; continue; } /* Ask daemon to send a resolution request. */ - memset(&(entry->hold_down), 0, sizeof(struct timeval)); + memset(&entry->hold_down, 0, sizeof(time64_t)); msg.type = SND_MPOA_RES_RTRY; memcpy(msg.MPS_ctrl, client->mps_ctrl_addr, ATM_ESA_LEN); msg.content.in_info = entry->ctrl_info; @@ -289,7 +288,7 @@ static void check_resolving_entries(struct mpoa_client *client) if (qos != NULL) msg.qos = qos->qos; msg_to_mpoad(&msg, client); - do_gettimeofday(&(entry->reply_wait)); + entry->reply_wait = ktime_get_seconds(); } } entry = entry->next; @@ -300,18 +299,18 @@ static void check_resolving_entries(struct mpoa_client *client) /* Call this every MPC-p5 seconds. */ static void refresh_entries(struct mpoa_client *client) { - struct timeval now; + time64_t now; struct in_cache_entry *entry = client->in_cache; ddprintk("refresh_entries\n"); - do_gettimeofday(&now); + now = ktime_get_seconds(); read_lock_bh(&client->ingress_lock); while (entry != NULL) { if (entry->entry_state == INGRESS_RESOLVED) { if (!(entry->refresh_time)) entry->refresh_time = (2 * (entry->ctrl_info.holding_time))/3; - if ((now.tv_sec - entry->reply_wait.tv_sec) > + if ((now - entry->reply_wait) > entry->refresh_time) { dprintk("refreshing an entry.\n"); entry->entry_state = INGRESS_REFRESHING; @@ -480,7 +479,7 @@ static eg_cache_entry *eg_cache_add_entry(struct k_message *msg, memcpy(entry->MPS_ctrl_ATM_addr, client->mps_ctrl_addr, ATM_ESA_LEN); entry->ctrl_info = msg->content.eg_info; - do_gettimeofday(&(entry->tv)); + entry->time = ktime_get_seconds(); entry->entry_state = EGRESS_RESOLVED; dprintk("new_eg_cache_entry cache_id %u\n", ntohl(entry->ctrl_info.cache_id)); @@ -495,7 +494,7 @@ static eg_cache_entry *eg_cache_add_entry(struct k_message *msg, static void update_eg_cache_entry(eg_cache_entry *entry, uint16_t holding_time) { - do_gettimeofday(&(entry->tv)); + entry->time = ktime_get_seconds(); entry->entry_state = EGRESS_RESOLVED; entry->ctrl_info.holding_time = holding_time; } @@ -503,17 +502,16 @@ static void update_eg_cache_entry(eg_cache_entry *entry, uint16_t holding_time) static void clear_expired(struct mpoa_client *client) { eg_cache_entry *entry, *next_entry; - struct timeval now; + time64_t now; struct k_message msg; - do_gettimeofday(&now); + now = ktime_get_seconds(); write_lock_irq(&client->egress_lock); entry = client->eg_cache; while (entry != NULL) { next_entry = entry->next; - if ((now.tv_sec - entry->tv.tv_sec) - > entry->ctrl_info.holding_time) { + if ((now - entry->time) > entry->ctrl_info.holding_time) { msg.type = SND_EGRESS_PURGE; msg.content.eg_info = entry->ctrl_info; dprintk("egress_cache: holding time expired, cache_id = %u.\n", diff --git a/net/atm/mpoa_caches.h b/net/atm/mpoa_caches.h index 6a266669ebf4..464c4c7f8d1f 100644 --- a/net/atm/mpoa_caches.h +++ b/net/atm/mpoa_caches.h @@ -2,6 +2,7 @@ #ifndef MPOA_CACHES_H #define MPOA_CACHES_H +#include <linux/time64.h> #include <linux/netdevice.h> #include <linux/types.h> #include <linux/atm.h> @@ -16,9 +17,9 @@ void atm_mpoa_init_cache(struct mpoa_client *mpc); typedef struct in_cache_entry { struct in_cache_entry *next; struct in_cache_entry *prev; - struct timeval tv; - struct timeval reply_wait; - struct timeval hold_down; + time64_t time; + time64_t reply_wait; + time64_t hold_down; uint32_t packets_fwded; uint16_t entry_state; uint32_t retry_time; @@ -53,7 +54,7 @@ struct in_cache_ops{ typedef struct eg_cache_entry{ struct eg_cache_entry *next; struct eg_cache_entry *prev; - struct timeval tv; + time64_t time; uint8_t MPS_ctrl_ATM_addr[ATM_ESA_LEN]; struct atm_vcc *shortcut; uint32_t packets_rcvd; diff --git a/net/atm/mpoa_proc.c b/net/atm/mpoa_proc.c index 8a0c17e1c203..2212da9c2da2 100644 --- a/net/atm/mpoa_proc.c +++ b/net/atm/mpoa_proc.c @@ -8,7 +8,7 @@ #include <linux/mm.h> #include <linux/module.h> #include <linux/proc_fs.h> -#include <linux/time.h> +#include <linux/ktime.h> #include <linux/seq_file.h> #include <linux/uaccess.h> #include <linux/atmmpc.h> @@ -138,7 +138,7 @@ static int mpc_show(struct seq_file *m, void *v) int i; in_cache_entry *in_entry; eg_cache_entry *eg_entry; - struct timeval now; + time64_t now; unsigned char ip_string[16]; if (v == SEQ_START_TOKEN) { @@ -148,15 +148,17 @@ static int mpc_show(struct seq_file *m, void *v) seq_printf(m, "\nInterface %d:\n\n", mpc->dev_num); seq_printf(m, "Ingress Entries:\nIP address State Holding time Packets fwded VPI VCI\n"); - do_gettimeofday(&now); + now = ktime_get_seconds(); for (in_entry = mpc->in_cache; in_entry; in_entry = in_entry->next) { + unsigned long seconds_delta = now - in_entry->time; + sprintf(ip_string, "%pI4", &in_entry->ctrl_info.in_dst_ip); seq_printf(m, "%-16s%s%-14lu%-12u", ip_string, ingress_state_string(in_entry->entry_state), in_entry->ctrl_info.holding_time - - (now.tv_sec-in_entry->tv.tv_sec), + seconds_delta, in_entry->packets_fwded); if (in_entry->shortcut) seq_printf(m, " %-3d %-3d", @@ -169,13 +171,14 @@ static int mpc_show(struct seq_file *m, void *v) seq_printf(m, "Egress Entries:\nIngress MPC ATM addr\nCache-id State Holding time Packets recvd Latest IP addr VPI VCI\n"); for (eg_entry = mpc->eg_cache; eg_entry; eg_entry = eg_entry->next) { unsigned char *p = eg_entry->ctrl_info.in_MPC_data_ATM_addr; + unsigned long seconds_delta = now - eg_entry->time; + for (i = 0; i < ATM_ESA_LEN; i++) seq_printf(m, "%02x", p[i]); seq_printf(m, "\n%-16lu%s%-14lu%-15u", (unsigned long)ntohl(eg_entry->ctrl_info.cache_id), egress_state_string(eg_entry->entry_state), - (eg_entry->ctrl_info.holding_time - - (now.tv_sec-eg_entry->tv.tv_sec)), + (eg_entry->ctrl_info.holding_time - seconds_delta), eg_entry->packets_rcvd); /* latest IP address */ -- 2.9.0

7 years, 5 months

[PATCH] openvswitch: use ktime_get_ts64() instead of ktime_get_ts()

by Arnd Bergmann

timespec is deprecated because of the y2038 overflow, so let's convert this one to ktime_get_ts64(). The code is already safe even on 32-bit architectures, since it uses monotonic times. On 64-bit architectures, nothing changes, while on 32-bit architectures this avoids one type conversion. Signed-off-by: Arnd Bergmann <arnd(a)arndb.de> --- net/openvswitch/flow.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/net/openvswitch/flow.c b/net/openvswitch/flow.c index dbe2379329c5..76d050aba7a4 100644 --- a/net/openvswitch/flow.c +++ b/net/openvswitch/flow.c @@ -56,12 +56,12 @@ u64 ovs_flow_used_time(unsigned long flow_jiffies) { - struct timespec cur_ts; + struct timespec64 cur_ts; u64 cur_ms, idle_ms; - ktime_get_ts(&cur_ts); + ktime_get_ts64(&cur_ts); idle_ms = jiffies_to_msecs(jiffies - flow_jiffies); - cur_ms = (u64)cur_ts.tv_sec * MSEC_PER_SEC + + cur_ms = (u64)(u32)cur_ts.tv_sec * MSEC_PER_SEC + cur_ts.tv_nsec / NSEC_PER_MSEC; return cur_ms - idle_ms; -- 2.9.0

7 years, 5 months

[PATCH] netxen: remove timespec usage

by Arnd Bergmann

netxen_collect_minidump() evidently just wants to get a monotonic timestamp. Using jiffies_to_timespec(jiffies, &ts) is not appropriate here, since it will overflow after 2^32 jiffies, which may be as short as 49 days of uptime. ktime_get_seconds() is the correct interface here. Signed-off-by: Arnd Bergmann <arnd(a)arndb.de> --- drivers/net/ethernet/qlogic/netxen/netxen_nic_hw.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/drivers/net/ethernet/qlogic/netxen/netxen_nic_hw.c b/drivers/net/ethernet/qlogic/netxen/netxen_nic_hw.c index 0a66389c06c2..1cd39c9a0345 100644 --- a/drivers/net/ethernet/qlogic/netxen/netxen_nic_hw.c +++ b/drivers/net/ethernet/qlogic/netxen/netxen_nic_hw.c @@ -2502,12 +2502,10 @@ netxen_collect_minidump(struct netxen_adapter *adapter) { int ret = 0; struct netxen_minidump_template_hdr *hdr; - struct timespec val; hdr = (struct netxen_minidump_template_hdr *) adapter->mdump.md_template; hdr->driver_capture_mask = adapter->mdump.md_capture_mask; - jiffies_to_timespec(jiffies, &val); - hdr->driver_timestamp = (u32) val.tv_sec; + hdr->driver_timestamp = ktime_get_seconds(); hdr->driver_info_word2 = adapter->fw_version; hdr->driver_info_word3 = NXRD32(adapter, CRB_DRIVER_VERSION); ret = netxen_parse_md_template(adapter); -- 2.9.0

7 years, 5 months

[PATCH] mmc_test: use ktime_get_ts64 for timestamps

by Arnd Bergmann

Calling getnstimeofday() can suffer from time jumps and from the y2038 overflow, so it is not appropriate here. Using ktime_get_ts64() solves both problems. Using ktime_get() with ktime_t timestamps would also work, but it seems that we mainly want to print the times as seconds+nanoseconds, so it would require an extra division in the output. Signed-off-by: Arnd Bergmann <arnd(a)arndb.de> --- drivers/mmc/core/mmc_test.c | 88 ++++++++++++++++++++++----------------------- 1 file changed, 42 insertions(+), 46 deletions(-) diff --git a/drivers/mmc/core/mmc_test.c b/drivers/mmc/core/mmc_test.c index 478869805b96..eae8810d22d9 100644 --- a/drivers/mmc/core/mmc_test.c +++ b/drivers/mmc/core/mmc_test.c @@ -101,7 +101,7 @@ struct mmc_test_transfer_result { struct list_head link; unsigned int count; unsigned int sectors; - struct timespec ts; + struct timespec64 ts; unsigned int rate; unsigned int iops; }; @@ -515,14 +515,11 @@ static int mmc_test_map_sg_max_scatter(struct mmc_test_mem *mem, /* * Calculate transfer rate in bytes per second. */ -static unsigned int mmc_test_rate(uint64_t bytes, struct timespec *ts) +static unsigned int mmc_test_rate(uint64_t bytes, struct timespec64 *ts) { uint64_t ns; - ns = ts->tv_sec; - ns *= 1000000000; - ns += ts->tv_nsec; - + ns = timespec64_to_ns(ts); bytes *= 1000000000; while (ns > UINT_MAX) { @@ -542,7 +539,7 @@ static unsigned int mmc_test_rate(uint64_t bytes, struct timespec *ts) * Save transfer results for future usage */ static void mmc_test_save_transfer_result(struct mmc_test_card *test, - unsigned int count, unsigned int sectors, struct timespec ts, + unsigned int count, unsigned int sectors, struct timespec64 ts, unsigned int rate, unsigned int iops) { struct mmc_test_transfer_result *tr; @@ -567,21 +564,21 @@ static void mmc_test_save_transfer_result(struct mmc_test_card *test, * Print the transfer rate. */ static void mmc_test_print_rate(struct mmc_test_card *test, uint64_t bytes, - struct timespec *ts1, struct timespec *ts2) + struct timespec64 *ts1, struct timespec64 *ts2) { unsigned int rate, iops, sectors = bytes >> 9; - struct timespec ts; + struct timespec64 ts; - ts = timespec_sub(*ts2, *ts1); + ts = timespec64_sub(*ts2, *ts1); rate = mmc_test_rate(bytes, &ts); iops = mmc_test_rate(100, &ts); /* I/O ops per sec x 100 */ - pr_info("%s: Transfer of %u sectors (%u%s KiB) took %lu.%09lu " + pr_info("%s: Transfer of %u sectors (%u%s KiB) took %llu.%09u " "seconds (%u kB/s, %u KiB/s, %u.%02u IOPS)\n", mmc_hostname(test->card->host), sectors, sectors >> 1, - (sectors & 1 ? ".5" : ""), (unsigned long)ts.tv_sec, - (unsigned long)ts.tv_nsec, rate / 1000, rate / 1024, + (sectors & 1 ? ".5" : ""), (u64)ts.tv_sec, + (u32)ts.tv_nsec, rate / 1000, rate / 1024, iops / 100, iops % 100); mmc_test_save_transfer_result(test, 1, sectors, ts, rate, iops); @@ -591,24 +588,24 @@ static void mmc_test_print_rate(struct mmc_test_card *test, uint64_t bytes, * Print the average transfer rate. */ static void mmc_test_print_avg_rate(struct mmc_test_card *test, uint64_t bytes, - unsigned int count, struct timespec *ts1, - struct timespec *ts2) + unsigned int count, struct timespec64 *ts1, + struct timespec64 *ts2) { unsigned int rate, iops, sectors = bytes >> 9; uint64_t tot = bytes * count; - struct timespec ts; + struct timespec64 ts; - ts = timespec_sub(*ts2, *ts1); + ts = timespec64_sub(*ts2, *ts1); rate = mmc_test_rate(tot, &ts); iops = mmc_test_rate(count * 100, &ts); /* I/O ops per sec x 100 */ pr_info("%s: Transfer of %u x %u sectors (%u x %u%s KiB) took " - "%lu.%09lu seconds (%u kB/s, %u KiB/s, " + "%llu.%09u seconds (%u kB/s, %u KiB/s, " "%u.%02u IOPS, sg_len %d)\n", mmc_hostname(test->card->host), count, sectors, count, sectors >> 1, (sectors & 1 ? ".5" : ""), - (unsigned long)ts.tv_sec, (unsigned long)ts.tv_nsec, + (u64)ts.tv_sec, (u32)ts.tv_nsec, rate / 1000, rate / 1024, iops / 100, iops % 100, test->area.sg_len); @@ -1449,7 +1446,7 @@ static int mmc_test_area_io_seq(struct mmc_test_card *test, unsigned long sz, int max_scatter, int timed, int count, bool nonblock, int min_sg_len) { - struct timespec ts1, ts2; + struct timespec64 ts1, ts2; int ret = 0; int i; struct mmc_test_area *t = &test->area; @@ -1475,7 +1472,7 @@ static int mmc_test_area_io_seq(struct mmc_test_card *test, unsigned long sz, return ret; if (timed) - getnstimeofday(&ts1); + ktime_get_ts64(&ts1); if (nonblock) ret = mmc_test_nonblock_transfer(test, t->sg, t->sg_len, dev_addr, t->blocks, 512, write, count); @@ -1489,7 +1486,7 @@ static int mmc_test_area_io_seq(struct mmc_test_card *test, unsigned long sz, return ret; if (timed) - getnstimeofday(&ts2); + ktime_get_ts64(&ts2); if (timed) mmc_test_print_avg_rate(test, sz, count, &ts1, &ts2); @@ -1747,7 +1744,7 @@ static int mmc_test_profile_trim_perf(struct mmc_test_card *test) struct mmc_test_area *t = &test->area; unsigned long sz; unsigned int dev_addr; - struct timespec ts1, ts2; + struct timespec64 ts1, ts2; int ret; if (!mmc_can_trim(test->card)) @@ -1758,19 +1755,19 @@ static int mmc_test_profile_trim_perf(struct mmc_test_card *test) for (sz = 512; sz < t->max_sz; sz <<= 1) { dev_addr = t->dev_addr + (sz >> 9); - getnstimeofday(&ts1); + ktime_get_ts64(&ts1); ret = mmc_erase(test->card, dev_addr, sz >> 9, MMC_TRIM_ARG); if (ret) return ret; - getnstimeofday(&ts2); + ktime_get_ts64(&ts2); mmc_test_print_rate(test, sz, &ts1, &ts2); } dev_addr = t->dev_addr; - getnstimeofday(&ts1); + ktime_get_ts64(&ts1); ret = mmc_erase(test->card, dev_addr, sz >> 9, MMC_TRIM_ARG); if (ret) return ret; - getnstimeofday(&ts2); + ktime_get_ts64(&ts2); mmc_test_print_rate(test, sz, &ts1, &ts2); return 0; } @@ -1779,19 +1776,19 @@ static int mmc_test_seq_read_perf(struct mmc_test_card *test, unsigned long sz) { struct mmc_test_area *t = &test->area; unsigned int dev_addr, i, cnt; - struct timespec ts1, ts2; + struct timespec64 ts1, ts2; int ret; cnt = t->max_sz / sz; dev_addr = t->dev_addr; - getnstimeofday(&ts1); + ktime_get_ts64(&ts1); for (i = 0; i < cnt; i++) { ret = mmc_test_area_io(test, sz, dev_addr, 0, 0, 0); if (ret) return ret; dev_addr += (sz >> 9); } - getnstimeofday(&ts2); + ktime_get_ts64(&ts2); mmc_test_print_avg_rate(test, sz, cnt, &ts1, &ts2); return 0; } @@ -1818,7 +1815,7 @@ static int mmc_test_seq_write_perf(struct mmc_test_card *test, unsigned long sz) { struct mmc_test_area *t = &test->area; unsigned int dev_addr, i, cnt; - struct timespec ts1, ts2; + struct timespec64 ts1, ts2; int ret; ret = mmc_test_area_erase(test); @@ -1826,14 +1823,14 @@ static int mmc_test_seq_write_perf(struct mmc_test_card *test, unsigned long sz) return ret; cnt = t->max_sz / sz; dev_addr = t->dev_addr; - getnstimeofday(&ts1); + ktime_get_ts64(&ts1); for (i = 0; i < cnt; i++) { ret = mmc_test_area_io(test, sz, dev_addr, 1, 0, 0); if (ret) return ret; dev_addr += (sz >> 9); } - getnstimeofday(&ts2); + ktime_get_ts64(&ts2); mmc_test_print_avg_rate(test, sz, cnt, &ts1, &ts2); return 0; } @@ -1864,7 +1861,7 @@ static int mmc_test_profile_seq_trim_perf(struct mmc_test_card *test) struct mmc_test_area *t = &test->area; unsigned long sz; unsigned int dev_addr, i, cnt; - struct timespec ts1, ts2; + struct timespec64 ts1, ts2; int ret; if (!mmc_can_trim(test->card)) @@ -1882,7 +1879,7 @@ static int mmc_test_profile_seq_trim_perf(struct mmc_test_card *test) return ret; cnt = t->max_sz / sz; dev_addr = t->dev_addr; - getnstimeofday(&ts1); + ktime_get_ts64(&ts1); for (i = 0; i < cnt; i++) { ret = mmc_erase(test->card, dev_addr, sz >> 9, MMC_TRIM_ARG); @@ -1890,7 +1887,7 @@ static int mmc_test_profile_seq_trim_perf(struct mmc_test_card *test) return ret; dev_addr += (sz >> 9); } - getnstimeofday(&ts2); + ktime_get_ts64(&ts2); mmc_test_print_avg_rate(test, sz, cnt, &ts1, &ts2); } return 0; @@ -1912,7 +1909,7 @@ static int mmc_test_rnd_perf(struct mmc_test_card *test, int write, int print, { unsigned int dev_addr, cnt, rnd_addr, range1, range2, last_ea = 0, ea; unsigned int ssz; - struct timespec ts1, ts2, ts; + struct timespec64 ts1, ts2, ts; int ret; ssz = sz >> 9; @@ -1921,10 +1918,10 @@ static int mmc_test_rnd_perf(struct mmc_test_card *test, int write, int print, range1 = rnd_addr / test->card->pref_erase; range2 = range1 / ssz; - getnstimeofday(&ts1); + ktime_get_ts64(&ts1); for (cnt = 0; cnt < UINT_MAX; cnt++) { - getnstimeofday(&ts2); - ts = timespec_sub(ts2, ts1); + ktime_get_ts64(&ts2); + ts = timespec64_sub(ts2, ts1); if (ts.tv_sec >= 10) break; ea = mmc_test_rnd_num(range1); @@ -1998,7 +1995,7 @@ static int mmc_test_seq_perf(struct mmc_test_card *test, int write, { struct mmc_test_area *t = &test->area; unsigned int dev_addr, i, cnt, sz, ssz; - struct timespec ts1, ts2; + struct timespec64 ts1, ts2; int ret; sz = t->max_tfr; @@ -2025,7 +2022,7 @@ static int mmc_test_seq_perf(struct mmc_test_card *test, int write, cnt = tot_sz / sz; dev_addr &= 0xffff0000; /* Round to 64MiB boundary */ - getnstimeofday(&ts1); + ktime_get_ts64(&ts1); for (i = 0; i < cnt; i++) { ret = mmc_test_area_io(test, sz, dev_addr, write, max_scatter, 0); @@ -2033,7 +2030,7 @@ static int mmc_test_seq_perf(struct mmc_test_card *test, int write, return ret; dev_addr += ssz; } - getnstimeofday(&ts2); + ktime_get_ts64(&ts2); mmc_test_print_avg_rate(test, sz, cnt, &ts1, &ts2); @@ -3066,10 +3063,9 @@ static int mtf_test_show(struct seq_file *sf, void *data) seq_printf(sf, "Test %d: %d\n", gr->testcase + 1, gr->result); list_for_each_entry(tr, &gr->tr_lst, link) { - seq_printf(sf, "%u %d %lu.%09lu %u %u.%02u\n", + seq_printf(sf, "%u %d %llu.%09u %u %u.%02u\n", tr->count, tr->sectors, - (unsigned long)tr->ts.tv_sec, - (unsigned long)tr->ts.tv_nsec, + (u64)tr->ts.tv_sec, (u32)tr->ts.tv_nsec, tr->rate, tr->iops / 100, tr->iops % 100); } } -- 2.9.0

7 years, 5 months

[PATCH] [RFC v4] packet: experimental support for 64-bit timestamps

by Arnd Bergmann

As I noticed in my previous patch to remove the 'timespec' usage in the packet socket, the timestamps in the packet socket are slightly inefficient as they convert a nanosecond value into seconds/nanoseconds or seconds/microseconds. This adds two new socket options for the timestamp to resolve that: PACKET_SKIPTIMESTAMP sets a flag to indicate whether to generate timestamps at all. When this is set, all timestamps are hardcoded to zero, which saves a few cycles for the conversion and the access of the hardware clocksource. The idea was taken from pktgen, which has an F_NO_TIMESTAMP option for the same purpose. PACKET_TIMESTAMP_NS64 changes the interpretation of the time stamp fields: instead of having 32 bits for seconds plus 32 bits for nanoseconds or microseconds, we now always send down 64 bits worth of nanoseconds when this flag is set. Link: https://patchwork.kernel.org/patch/10077199/ Suggested-by: Willem de Bruijn <willemdebruijn.kernel(a)gmail.com> Signed-off-by: Arnd Bergmann <arnd(a)arndb.de> --- I still have not done any runtime testing on this patch, only implemented the suggestions from the previous versions. While I don't think anyone is actively looking for this feature, I don't think there are any reasons left against merging it either, and it might come in handy for someone. v4: address minor comments from Willem v3: rework to use setsockopt v2: use new tstamp flags instead of a new version v1: original implementation using TPACKET_V4 --- include/uapi/linux/if_packet.h | 2 + net/packet/af_packet.c | 160 ++++++++++++++++++++++++++++------------- net/packet/internal.h | 2 + 3 files changed, 116 insertions(+), 48 deletions(-) diff --git a/include/uapi/linux/if_packet.h b/include/uapi/linux/if_packet.h index 67b61d91d89b..2eba54770e6b 100644 --- a/include/uapi/linux/if_packet.h +++ b/include/uapi/linux/if_packet.h @@ -57,6 +57,8 @@ struct sockaddr_ll { #define PACKET_QDISC_BYPASS 20 #define PACKET_ROLLOVER_STATS 21 #define PACKET_FANOUT_DATA 22 +#define PACKET_SKIPTIMESTAMP 23 +#define PACKET_TIMESTAMP_NS64 24 #define PACKET_FANOUT_HASH 0 #define PACKET_FANOUT_LB 1 diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c index 7432c6699818..f55f330ab547 100644 --- a/net/packet/af_packet.c +++ b/net/packet/af_packet.c @@ -200,7 +200,7 @@ static void prb_retire_current_block(struct tpacket_kbdq_core *, struct packet_sock *, unsigned int status); static int prb_queue_frozen(struct tpacket_kbdq_core *); static void prb_open_block(struct tpacket_kbdq_core *, - struct tpacket_block_desc *); + struct tpacket_block_desc *, struct packet_sock *); static void prb_retire_rx_blk_timer_expired(struct timer_list *); static void _prb_refresh_rx_retire_blk_timer(struct tpacket_kbdq_core *); static void prb_fill_rxhash(struct tpacket_kbdq_core *, struct tpacket3_hdr *); @@ -439,52 +439,91 @@ static int __packet_get_status(struct packet_sock *po, void *frame) } } -static __u32 tpacket_get_timestamp(struct sk_buff *skb, struct timespec64 *ts, - unsigned int flags) +static __u32 tpacket_get_timestamp(struct sk_buff *skb, __u32 *hi, __u32 *lo) { + struct packet_sock *po = pkt_sk(skb->sk); struct skb_shared_hwtstamps *shhwtstamps = skb_hwtstamps(skb); + ktime_t stamp; + u32 type; + + if (po->tp_skiptstamp) + return 0; if (shhwtstamps && - (flags & SOF_TIMESTAMPING_RAW_HARDWARE) && - ktime_to_timespec64_cond(shhwtstamps->hwtstamp, ts)) - return TP_STATUS_TS_RAW_HARDWARE; + (po->tp_tstamp & SOF_TIMESTAMPING_RAW_HARDWARE) && + shhwtstamps->hwtstamp) { + stamp = shhwtstamps->hwtstamp; + type = TP_STATUS_TS_RAW_HARDWARE; + } else if (skb->tstamp) { + stamp = skb->tstamp; + type = TP_STATUS_TS_SOFTWARE; + } else { + return 0; + } - if (ktime_to_timespec64_cond(skb->tstamp, ts)) - return TP_STATUS_TS_SOFTWARE; + if (po->tp_tstamp_ns64) { + __u64 ns = ktime_to_ns(stamp); - return 0; + *hi = upper_32_bits(ns); + *lo = lower_32_bits(ns); + } else { + struct timespec64 ts = ktime_to_timespec64(stamp); + + *hi = ts.tv_sec; + if (po->tp_version > TPACKET_V1) + *lo = ts.tv_nsec; + else + *lo = ts.tv_nsec / NSEC_PER_USEC; + } + + return type; +} + +static void packet_get_time(struct packet_sock *po, __u32 *hi, __u32 *lo) +{ + if (po->tp_skiptstamp) { + *hi = 0; + *lo = 0; + } else if (po->tp_tstamp_ns64) { + __u64 ns = ktime_get_real_ns(); + + *hi = upper_32_bits(ns); + *hi = lower_32_bits(ns); + } else { + struct timespec64 ts; + + ktime_get_real_ts64(&ts); + /* unsigned seconds overflow in y2106 here */ + *hi = ts.tv_sec; + if (po->tp_version > TPACKET_V1) + *lo = ts.tv_nsec; + else + *lo = ts.tv_nsec / NSEC_PER_USEC; + } } static __u32 __packet_set_timestamp(struct packet_sock *po, void *frame, struct sk_buff *skb) { union tpacket_uhdr h; - struct timespec64 ts; - __u32 ts_status; + __u32 ts_status, hi, lo; - if (!(ts_status = tpacket_get_timestamp(skb, &ts, po->tp_tstamp))) + if (!(ts_status = tpacket_get_timestamp(skb, &hi, &lo))) return 0; h.raw = frame; - /* - * versions 1 through 3 overflow the timestamps in y2106, since they - * all store the seconds in a 32-bit unsigned integer. - * If we create a version 4, that should have a 64-bit timestamp, - * either 64-bit seconds + 32-bit nanoseconds, or just 64-bit - * nanoseconds. - */ switch (po->tp_version) { case TPACKET_V1: - h.h1->tp_sec = ts.tv_sec; - h.h1->tp_usec = ts.tv_nsec / NSEC_PER_USEC; + h.h1->tp_sec = hi; + h.h1->tp_usec = lo; break; case TPACKET_V2: - h.h2->tp_sec = ts.tv_sec; - h.h2->tp_nsec = ts.tv_nsec; + h.h2->tp_sec = hi; + h.h2->tp_nsec = lo; break; case TPACKET_V3: - h.h3->tp_sec = ts.tv_sec; - h.h3->tp_nsec = ts.tv_nsec; + h.h3->tp_sec = hi; + h.h3->tp_nsec = lo; break; default: WARN(1, "TPACKET version not supported.\n"); @@ -633,7 +672,7 @@ static void init_prb_bdqc(struct packet_sock *po, p1->max_frame_len = p1->kblk_size - BLK_PLUS_PRIV(p1->blk_sizeof_priv); prb_init_ft_ops(p1, req_u); prb_setup_retire_blk_timer(po); - prb_open_block(p1, pbd); + prb_open_block(p1, pbd, po); } /* Do NOT update the last_blk_num first. @@ -730,7 +769,7 @@ static void prb_retire_rx_blk_timer_expired(struct timer_list *t) * opening a block thaws the queue,restarts timer * Thawing/timer-refresh is a side effect. */ - prb_open_block(pkc, pbd); + prb_open_block(pkc, pbd, po); goto out; } } @@ -812,10 +851,8 @@ static void prb_close_block(struct tpacket_kbdq_core *pkc1, * It shouldn't really happen as we don't close empty * blocks. See prb_retire_rx_blk_timer_expired(). */ - struct timespec64 ts; - ktime_get_real_ts64(&ts); - h1->ts_last_pkt.ts_sec = ts.tv_sec; - h1->ts_last_pkt.ts_nsec = ts.tv_nsec; + packet_get_time(po, &h1->ts_last_pkt.ts_sec, + &h1->ts_last_pkt.ts_nsec); } smp_wmb(); @@ -841,9 +878,8 @@ static void prb_thaw_queue(struct tpacket_kbdq_core *pkc) * */ static void prb_open_block(struct tpacket_kbdq_core *pkc1, - struct tpacket_block_desc *pbd1) + struct tpacket_block_desc *pbd1, struct packet_sock *po) { - struct timespec64 ts; struct tpacket_hdr_v1 *h1 = &pbd1->hdr.bh1; smp_rmb(); @@ -856,10 +892,8 @@ static void prb_open_block(struct tpacket_kbdq_core *pkc1, BLOCK_NUM_PKTS(pbd1) = 0; BLOCK_LEN(pbd1) = BLK_PLUS_PRIV(pkc1->blk_sizeof_priv); - ktime_get_real_ts64(&ts); - - h1->ts_first_pkt.ts_sec = ts.tv_sec; - h1->ts_first_pkt.ts_nsec = ts.tv_nsec; + packet_get_time(po, &h1->ts_first_pkt.ts_sec, + &h1->ts_first_pkt.ts_nsec); pkc1->pkblk_start = (char *)pbd1; pkc1->nxt_offset = pkc1->pkblk_start + BLK_PLUS_PRIV(pkc1->blk_sizeof_priv); @@ -936,7 +970,7 @@ static void *prb_dispatch_next_block(struct tpacket_kbdq_core *pkc, * open this block and return the offset where the first packet * needs to get stored. */ - prb_open_block(pkc, pbd); + prb_open_block(pkc, pbd, po); return (void *)pkc->nxt_offset; } @@ -1068,7 +1102,7 @@ static void *__packet_lookup_frame_in_block(struct packet_sock *po, * opening a block also thaws the queue. * Thawing is a side effect. */ - prb_open_block(pkc, pbd); + prb_open_block(pkc, pbd, po); } } @@ -2191,8 +2225,8 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, unsigned long status = TP_STATUS_USER; unsigned short macoff, netoff, hdrlen; struct sk_buff *copy_skb = NULL; - struct timespec64 ts; __u32 ts_status; + __u32 tstamp_hi, tstamp_lo; bool is_drop_n_account = false; bool do_vnet = false; @@ -2318,8 +2352,8 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, skb_copy_bits(skb, 0, h.raw + macoff, snaplen); - if (!(ts_status = tpacket_get_timestamp(skb, &ts, po->tp_tstamp))) - ktime_get_real_ts64(&ts); + if (!(ts_status = tpacket_get_timestamp(skb, &tstamp_hi, &tstamp_lo))) + packet_get_time(po, &tstamp_hi, &tstamp_lo); status |= ts_status; @@ -2329,8 +2363,8 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, h.h1->tp_snaplen = snaplen; h.h1->tp_mac = macoff; h.h1->tp_net = netoff; - h.h1->tp_sec = ts.tv_sec; - h.h1->tp_usec = ts.tv_nsec / NSEC_PER_USEC; + h.h1->tp_sec = tstamp_hi; + h.h1->tp_usec = tstamp_lo; hdrlen = sizeof(*h.h1); break; case TPACKET_V2: @@ -2338,8 +2372,8 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, h.h2->tp_snaplen = snaplen; h.h2->tp_mac = macoff; h.h2->tp_net = netoff; - h.h2->tp_sec = ts.tv_sec; - h.h2->tp_nsec = ts.tv_nsec; + h.h2->tp_sec = tstamp_hi; + h.h2->tp_nsec = tstamp_lo; if (skb_vlan_tag_present(skb)) { h.h2->tp_vlan_tci = skb_vlan_tag_get(skb); h.h2->tp_vlan_tpid = ntohs(skb->vlan_proto); @@ -2360,8 +2394,8 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, h.h3->tp_snaplen = snaplen; h.h3->tp_mac = macoff; h.h3->tp_net = netoff; - h.h3->tp_sec = ts.tv_sec; - h.h3->tp_nsec = ts.tv_nsec; + h.h3->tp_sec = tstamp_hi; + h.h3->tp_nsec = tstamp_lo; memset(h.h3->tp_padding, 0, sizeof(h.h3->tp_padding)); hdrlen = sizeof(*h.h3); break; @@ -3792,6 +3826,30 @@ packet_setsockopt(struct socket *sock, int level, int optname, char __user *optv po->tp_tstamp = val; return 0; } + case PACKET_SKIPTIMESTAMP: + { + int val; + + if (optlen != sizeof(val)) + return -EINVAL; + if (copy_from_user(&val, optval, sizeof(val))) + return -EFAULT; + + po->tp_skiptstamp = val; + return 0; + } + case PACKET_TIMESTAMP_NS64: + { + int val; + + if (optlen != sizeof(val)) + return -EINVAL; + if (copy_from_user(&val, optval, sizeof(val))) + return -EFAULT; + + po->tp_tstamp_ns64 = val; + return 0; + } case PACKET_FANOUT: { int val; @@ -3921,6 +3979,12 @@ static int packet_getsockopt(struct socket *sock, int level, int optname, case PACKET_TIMESTAMP: val = po->tp_tstamp; break; + case PACKET_SKIPTIMESTAMP: + val = po->tp_skiptstamp; + break; + case PACKET_TIMESTAMP_NS64: + val = po->tp_tstamp_ns64; + break; case PACKET_FANOUT: val = (po->fanout ? ((u32)po->fanout->id | diff --git a/net/packet/internal.h b/net/packet/internal.h index 562fbc155006..20b69512210f 100644 --- a/net/packet/internal.h +++ b/net/packet/internal.h @@ -128,6 +128,8 @@ struct packet_sock { unsigned int tp_reserve; unsigned int tp_loss:1; unsigned int tp_tx_has_off:1; + unsigned int tp_skiptstamp:1; + unsigned int tp_tstamp_ns64:1; unsigned int tp_tstamp; struct net_device __rcu *cached_dev; int (*xmit)(struct sk_buff *skb); -- 2.9.0

7 years, 5 months

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

Y2038