The problems described below still exist in -2.6.0-test1.
>Date: Wed, 18 Jun 2003 00:57:12 -0700
>From: "B. D. Elliott" <[email protected]>
>To: [email protected]
>Subject: Sparc64-2.5.72: A Serious Time Problem
>Mail-Followup-To: [email protected]
>Mime-Version: 1.0
>Content-Type: text/plain; charset=us-ascii
>Content-Disposition: inline
>User-Agent: Mutt/1.5.4i
>Message-Id: <[email protected]>
>Status: RO
>Content-Length: 1651
>Lines: 47
>
>There is a serious bug in setting time on 64-bit sparcs (and probably other
>64-bit systems). The symptom is that ntpdate or date set the time back to
>1969 or 1970. The underlying problems are that stime is broken, and any
>settimeofday call fails with a bad fractional value. Ntpdate falls back to
>stime when settimeofday fails.
>
>The settimeofday problem is that the timeval and timespec structures are not
>the same size. In particular, the fractional part is an int in timeval, and
>a long in timespec. The stime problem is that the argument is not an int,
>but a time_t, which is long on at least some 64-bit systems.
>
>The following patch appears to fix this on my sparc64.
>
>===================================================================
>--- ./kernel/time.c.orig 2003-06-16 22:36:04.000000000 -0700
>+++ ./kernel/time.c 2003-06-18 00:00:43.000000000 -0700
>@@ -66,7 +66,7 @@
> * architectures that need it).
> */
>
>-asmlinkage long sys_stime(int * tptr)
>+asmlinkage long sys_stime(time_t * tptr)
> {
> struct timespec tv;
>
>@@ -162,13 +162,15 @@
>
> asmlinkage long sys_settimeofday(struct timeval __user *tv, struct timezone __user *tz)
> {
>+ struct timeval user_tv;
> struct timespec new_tv;
> struct timezone new_tz;
>
> if (tv) {
>- if (copy_from_user(&new_tv, tv, sizeof(*tv)))
>+ if (copy_from_user(&user_tv, tv, sizeof(*tv)))
> return -EFAULT;
>- new_tv.tv_nsec *= NSEC_PER_USEC;
>+ new_tv.tv_sec = user_tv.tv_sec;
>+ new_tv.tv_nsec = user_tv.tv_usec * NSEC_PER_USEC;
> }
> if (tz) {
> if (copy_from_user(&new_tz, tz, sizeof(*tz)))
>===================================================================
--
B. D. Elliott [email protected]
On Tue, 15 Jul 2003 00:40:49 -0700
"B. D. Elliott" <[email protected]> wrote:
> The problems described below still exist in -2.6.0-test1.
Not surprising, since you didn't even CC: the sparc
development list :-)
On Tue, 15 Jul 2003 00:40:49 -0700
"B. D. Elliott" <[email protected]> wrote:
> The problems described below still exist in -2.6.0-test1.
Ignore my previous reply, not a Sparc problem obviously
but a generic one, sorry.
Andrew, he's right, we need to fix this and his patch looks
perfectly fine.
> >Date: Wed, 18 Jun 2003 00:57:12 -0700
> >From: "B. D. Elliott" <[email protected]>
> >To: [email protected]
> >Subject: Sparc64-2.5.72: A Serious Time Problem
> >Mail-Followup-To: [email protected]
> >Mime-Version: 1.0
> >Content-Type: text/plain; charset=us-ascii
> >Content-Disposition: inline
> >User-Agent: Mutt/1.5.4i
> >Message-Id: <[email protected]>
> >Status: RO
> >Content-Length: 1651
> >Lines: 47
> >
> >There is a serious bug in setting time on 64-bit sparcs (and probably other
> >64-bit systems). The symptom is that ntpdate or date set the time back to
> >1969 or 1970. The underlying problems are that stime is broken, and any
> >settimeofday call fails with a bad fractional value. Ntpdate falls back to
> >stime when settimeofday fails.
> >
> >The settimeofday problem is that the timeval and timespec structures are not
> >the same size. In particular, the fractional part is an int in timeval, and
> >a long in timespec. The stime problem is that the argument is not an int,
> >but a time_t, which is long on at least some 64-bit systems.
> >
> >The following patch appears to fix this on my sparc64.
> >
> >===================================================================
> >--- ./kernel/time.c.orig 2003-06-16 22:36:04.000000000 -0700
> >+++ ./kernel/time.c 2003-06-18 00:00:43.000000000 -0700
> >@@ -66,7 +66,7 @@
> > * architectures that need it).
> > */
> >
> >-asmlinkage long sys_stime(int * tptr)
> >+asmlinkage long sys_stime(time_t * tptr)
> > {
> > struct timespec tv;
> >
> >@@ -162,13 +162,15 @@
> >
> > asmlinkage long sys_settimeofday(struct timeval __user *tv, struct timezone __user *tz)
> > {
> >+ struct timeval user_tv;
> > struct timespec new_tv;
> > struct timezone new_tz;
> >
> > if (tv) {
> >- if (copy_from_user(&new_tv, tv, sizeof(*tv)))
> >+ if (copy_from_user(&user_tv, tv, sizeof(*tv)))
> > return -EFAULT;
> >- new_tv.tv_nsec *= NSEC_PER_USEC;
> >+ new_tv.tv_sec = user_tv.tv_sec;
> >+ new_tv.tv_nsec = user_tv.tv_usec * NSEC_PER_USEC;
> > }
> > if (tz) {
> > if (copy_from_user(&new_tz, tz, sizeof(*tz)))
> >===================================================================
>
> --
> B. D. Elliott [email protected]
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
David S. Miller wrote:
> On Tue, 15 Jul 2003 00:40:49 -0700
> "B. D. Elliott" <[email protected]> wrote:
>
>
>>The problems described below still exist in -2.6.0-test1.
>
>
> Ignore my previous reply, not a Sparc problem obviously
> but a generic one, sorry.
>
> Andrew, he's right, we need to fix this and his patch looks
> perfectly fine.
Looks good to me.
-g
>
>
>>>Date: Wed, 18 Jun 2003 00:57:12 -0700
>>>From: "B. D. Elliott" <[email protected]>
>>>To: [email protected]
>>>Subject: Sparc64-2.5.72: A Serious Time Problem
>>>Mail-Followup-To: [email protected]
>>>Mime-Version: 1.0
>>>Content-Type: text/plain; charset=us-ascii
>>>Content-Disposition: inline
>>>User-Agent: Mutt/1.5.4i
>>>Message-Id: <[email protected]>
>>>Status: RO
>>>Content-Length: 1651
>>>Lines: 47
>>>
>>>There is a serious bug in setting time on 64-bit sparcs (and probably other
>>>64-bit systems). The symptom is that ntpdate or date set the time back to
>>>1969 or 1970. The underlying problems are that stime is broken, and any
>>>settimeofday call fails with a bad fractional value. Ntpdate falls back to
>>>stime when settimeofday fails.
>>>
>>>The settimeofday problem is that the timeval and timespec structures are not
>>>the same size. In particular, the fractional part is an int in timeval, and
>>>a long in timespec. The stime problem is that the argument is not an int,
>>>but a time_t, which is long on at least some 64-bit systems.
>>>
>>>The following patch appears to fix this on my sparc64.
>>>
>>>===================================================================
>>>--- ./kernel/time.c.orig 2003-06-16 22:36:04.000000000 -0700
>>>+++ ./kernel/time.c 2003-06-18 00:00:43.000000000 -0700
>>>@@ -66,7 +66,7 @@
>>> * architectures that need it).
>>> */
>>>
>>>-asmlinkage long sys_stime(int * tptr)
>>>+asmlinkage long sys_stime(time_t * tptr)
>>>{
>>> struct timespec tv;
>>>
>>>@@ -162,13 +162,15 @@
>>>
>>>asmlinkage long sys_settimeofday(struct timeval __user *tv, struct timezone __user *tz)
>>>{
>>>+ struct timeval user_tv;
>>> struct timespec new_tv;
>>> struct timezone new_tz;
>>>
>>> if (tv) {
>>>- if (copy_from_user(&new_tv, tv, sizeof(*tv)))
>>>+ if (copy_from_user(&user_tv, tv, sizeof(*tv)))
>>> return -EFAULT;
>>>- new_tv.tv_nsec *= NSEC_PER_USEC;
>>>+ new_tv.tv_sec = user_tv.tv_sec;
>>>+ new_tv.tv_nsec = user_tv.tv_usec * NSEC_PER_USEC;
>>> }
>>> if (tz) {
>>> if (copy_from_user(&new_tz, tz, sizeof(*tz)))
>>>===================================================================
>>
>>--
>>B. D. Elliott [email protected]
>>-
>>To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
>>the body of a message to [email protected]
>>More majordomo info at http://vger.kernel.org/majordomo-info.html
>>Please read the FAQ at http://www.tux.org/lkml/
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
>
--
George Anzinger [email protected]
High-res-timers: http://sourceforge.net/projects/high-res-timers/
Preemption patch: http://www.kernel.org/pub/linux/kernel/people/rml