Y2038 October 2015

y2038@lists.linaro.org

39 participants
64 discussions

[PATCH v3] Staging: olpc_dcon: Replace timespec with ktime_t

by Ksenija Stanojevic

Struct timespec will overflow in year 2038, here it will not cause an overflow because it is used with timespec_sub, but still has to be removed as part of y2038 changes. Replace it with ktime_t. Also use monotonic instead of real-time by replacing functions getnstimeofday with ktime_get. Signed-off-by: Ksenija Stanojevic <ksenija.stanojevic(a)gmail.com> --- Changes in v3: -fix commit message (ktime_sub -> ktime_get) Changes in v2: -use ktime_t instead timespec64 -use ktime_sub instead timespec64_sub -use monotonic instead real-time. drivers/staging/olpc_dcon/olpc_dcon.c | 27 +++++++++++++-------------- drivers/staging/olpc_dcon/olpc_dcon.h | 4 ++-- 2 files changed, 15 insertions(+), 16 deletions(-) diff --git a/drivers/staging/olpc_dcon/olpc_dcon.c b/drivers/staging/olpc_dcon/olpc_dcon.c index d115f5c..2dc5d36 100644 --- a/drivers/staging/olpc_dcon/olpc_dcon.c +++ b/drivers/staging/olpc_dcon/olpc_dcon.c @@ -238,13 +238,12 @@ static void dcon_sleep(struct dcon_priv *dcon, bool sleep) */ static void dcon_load_holdoff(struct dcon_priv *dcon) { - struct timespec delta_t, now; + ktime_t delta_t, now; while (1) { - getnstimeofday(&now); - delta_t = timespec_sub(now, dcon->load_time); - if (delta_t.tv_sec != 0 || - delta_t.tv_nsec > NSEC_PER_MSEC * 20) { + now = ktime_get(); + delta_t = ktime_sub(now, dcon->load_time); + if (ktime_to_ns(delta_t) > NSEC_PER_MSEC * 20) { break; } mdelay(4); @@ -325,19 +324,19 @@ static void dcon_source_switch(struct work_struct *work) /* And turn off the DCON */ pdata->set_dconload(1); - getnstimeofday(&dcon->load_time); + dcon->load_time = ktime_get(); pr_info("The CPU has control\n"); break; case DCON_SOURCE_DCON: { - struct timespec delta_t; + ktime_t delta_t; pr_info("dcon_source_switch to DCON\n"); /* Clear DCONLOAD - this implies that the DCON is in control */ pdata->set_dconload(0); - getnstimeofday(&dcon->load_time); + dcon->load_time = ktime_get(); wait_event_timeout(dcon->waitq, dcon->switched, HZ/2); @@ -355,14 +354,14 @@ static void dcon_source_switch(struct work_struct *work) * deassert and reassert, and hope for the best. * see http://dev.laptop.org/ticket/9664 */ - delta_t = timespec_sub(dcon->irq_time, dcon->load_time); - if (dcon->switched && delta_t.tv_sec == 0 && - delta_t.tv_nsec < NSEC_PER_MSEC * 20) { + delta_t = ktime_sub(dcon->irq_time, dcon->load_time); + if (dcon->switched && + ktime_to_ns(delta_t) < NSEC_PER_MSEC * 20) { pr_err("missed loading, retrying\n"); pdata->set_dconload(1); mdelay(41); pdata->set_dconload(0); - getnstimeofday(&dcon->load_time); + dcon->load_time = ktime_get(); mdelay(41); } } @@ -742,7 +741,7 @@ irqreturn_t dcon_interrupt(int irq, void *id) case 2: /* switch to DCON mode */ case 1: /* switch to CPU mode */ dcon->switched = true; - getnstimeofday(&dcon->irq_time); + dcon->irq_time = ktime_get(); wake_up(&dcon->waitq); break; @@ -756,7 +755,7 @@ irqreturn_t dcon_interrupt(int irq, void *id) */ if (dcon->curr_src != dcon->pending_src && !dcon->switched) { dcon->switched = true; - getnstimeofday(&dcon->irq_time); + dcon->irq_time = ktime_get(); wake_up(&dcon->waitq); pr_debug("switching w/ status 0/0\n"); } else { diff --git a/drivers/staging/olpc_dcon/olpc_dcon.h b/drivers/staging/olpc_dcon/olpc_dcon.h index d06e19d..215e7ec 100644 --- a/drivers/staging/olpc_dcon/olpc_dcon.h +++ b/drivers/staging/olpc_dcon/olpc_dcon.h @@ -79,8 +79,8 @@ struct dcon_priv { /* Variables used during switches */ bool switched; - struct timespec irq_time; - struct timespec load_time; + ktime_t irq_time; + ktime_t load_time; /* Current output type; true == mono, false == color */ bool mono; -- 1.9.1

8 years, 6 months

[PATCH 0/5] y2038 conversion for ntp/pps and sfc driver

by Arnd Bergmann

When trying to build a kernel with time_t commented out, I found that the ntp subsystem still relies on timespec for its pps handling. This series addresses this and converts all the code to use timespec64 instead, step by step. There is one device driver that interacts with this code directly (rather than only through the ptp subsystem), so I have to convert that driver at the same time. The patches should ideally stay together as a series, but they do span multiple subsystems, so I'm also looking for the right person to merge them. Please review. Thanks, Arnd Arnd Bergmann (5): ntp/pps: use timespec64 for hardpps() ntp/pps: replace getnstime_raw_and_real with 64-bit version ntp: use timespec64 in sync_cmos_clock ntp/pps: use y2038 safe types in pps_event_time net: sfc: avoid using timespec drivers/net/ethernet/sfc/ptp.c | 30 +++++++++++++++--------------- drivers/pps/kapi.c | 4 ++-- include/linux/pps_kernel.h | 16 ++++++++-------- include/linux/timekeeping.h | 4 ++-- include/linux/timex.h | 2 +- kernel/time/ntp.c | 16 ++++++++-------- kernel/time/ntp_internal.h | 2 +- kernel/time/timekeeping.c | 14 +++++++------- 8 files changed, 44 insertions(+), 44 deletions(-) -- 2.1.0.rc2

8 years, 6 months

[PATCH] ceph: fix message length computation

by Arnd Bergmann

create_request_message() computes the maximum length of a message, but uses the wrong type for the time stamp: sizeof(struct timespec) may be 8 or 16 depending on the architecture, while sizeof(struct ceph_timespec) is always 8, and that is what gets put into the message. Found while auditing the uses of timespec for y2038 problems. Signed-off-by: Arnd Bergmann <arnd(a)arndb.de> Fixes: b8e69066d8af ("ceph: include time stamp in every MDS request") --- diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index 51cb02da75d9..fe2c982764e7 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -1935,7 +1935,7 @@ static struct ceph_msg *create_request_message(struct ceph_mds_client *mdsc, len = sizeof(*head) + pathlen1 + pathlen2 + 2*(1 + sizeof(u32) + sizeof(u64)) + - sizeof(struct timespec); + sizeof(struct ceph_timespec); /* calculate (max) length for cap releases */ len += sizeof(struct ceph_mds_request_release) *

8 years, 6 months

Courier was unable to deliver the parcel, ID00000967235

by FedEx 2Day

Dear Customer, We could not deliver your parcel. Shipment Label is attached to this email. Yours sincerely, Greg Weiss, Sr. Delivery Manager.

8 years, 6 months

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

Y2038 October 2015