On 08.03.21 19:11, Bezdeka, Florian (T RDA IOT SES-DE) wrote:
> On Mon, 2021-03-08 at 18:12 +0100, Jan Kiszka wrote:
>> On 08.03.21 18:02, Florian Bezdeka wrote:
>>> On systems using 32 bit for time_t the sem_timedwait syscall was broken
>>> because the function used for copying the timeout value from userspace
>>> to kernel (=sem_fetch_timeout()) was always copying
>>> sizeof(struct timespec64).
>>>
>>> A 32 bit application (or more specific an application with 4 byte
>>> time_t) would only provide sizeof(struct old_timespec32).
>>>
>>> Notable changes:
>>> - The copy operation from userspace to kernel is now already done in
>>> the syscall handler. So it is always done. Previously it was copied
>>> over and validated before the first use (when used at all).
>>> So we have some additional instructions now that may be
>>> unnecessary, but that simplifies the code.
>>>
>>> - Validation: Switched to timespec64_valid() instead of our own
>>> check.
>>>
>>> Fixes: 8043eccd232d ("cobalt/kernel: y2038: convert struct timespec to
>>> timespec64")
>>
>> This patch was dropped from next again (and that sha was never applied).
>> We should go a for fixed version of that series (patches 2..5, only
>> patch 1 is now applied).
>
> That's one of the things that may happen when working with branches
> that are getting rebased... The correct refspec is maybe 4d31feb5 now,
> which is part of wip/dovetail.
All these are development branches, nothing considered stable. And, no,
even wip/dovetail is the wrong target. The right one is the patch that
needs to be fixed in v2 of Philippe's series.
>
> Please note the RFC tag. Would be nice to get some feedback before
> going ahead, otherwise the loops with all the necessary synchronization
> and rebases is getting endless...
For this particular patch, Philippe should comment first.
I will have a look at the rest now.
Jan
--
Siemens AG, T RDA IOT
Corporate Competence Center Embedded Linux
