On 4/24/19 4:04 AM, Bruce Ashfield wrote:
Since this is still being discussed on the -rt mailing list, I'm going to wait to see how that shakes out before merging the change.


Yes, but it seems like that upstream won't fix the issue in a short time.

--Hongzhi




Bruce

On Mon, Apr 22, 2019 at 11:20 PM Hongzhi, Song <hongzhi.s...@windriver.com <mailto:hongzhi.s...@windriver.com>> wrote:

    Please merge to v5.0/standard/preempt-rt/base and
    v5.0/standard/preempt-rt/intel-x86


    --Hongzhi


    On 4/23/19 11:14 AM, Hongzhi.Song wrote:
    > Two commits from rt-kernel cause the following disorder log at the
    > early stage of boot with qemux86-64. Maybe it has other disorder
    logs
    > which I omit.
    >
    > Commits:
    > 7128a815: printk: introduce emergency messages
    > b9d460e4: serial: 8250: implement write_atomic
    >
    > [    0.401166] 000: 6200.04 BogoMIPS (lpj=3100023)
    > [    0.877953] 000: Intel(R) Core(TM)2 Duo CPU     T7700 @ 2.40GHz
    > [    0.877953] 000:  (family: 0x6, model: 0xf
    > [    0.877953] 000: , stepping: 0xb)
    > [    1.255009] 000:  5
    > [    1.255167] 000:  *10
    > [    1.255295] 000:  11
    > [    1.255406] 000: )
    >
    > Obviously, the printk emergency mechanism doesn't work as expected.
    > Removing partial code of printk emergency mechanism can
    > fix the issue temporarily.
    >
    > Signed-off-by: Hongzhi.Song <hongzhi.s...@windriver.com
    <mailto:hongzhi.s...@windriver.com>>
    > ---
    >   kernel/printk/printk.c | 4 ++++
    >   1 file changed, 4 insertions(+)
    >
    > diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c
    > index 997d07b..99d4a94 100644
    > --- a/kernel/printk/printk.c
    > +++ b/kernel/printk/printk.c
    > @@ -1726,6 +1726,7 @@ static void call_console_drivers(u64 seq,
    const char *ext_text, size_t ext_len,
    >                       con->wrote_history = 1;
    >                       con->printk_seq = seq - 1;
    >               }
    > +#if 0
    >               if (con->write_atomic && level <
    emergency_console_loglevel &&
    >                   facility == 0) {
    >                       /* skip emergency messages, already printed */
    > @@ -1733,6 +1734,7 @@ static void call_console_drivers(u64 seq,
    const char *ext_text, size_t ext_len,
    >                               con->printk_seq = seq;
    >                       continue;
    >               }
    > +#endif
    >               if (con->flags & CON_BOOT && facility == 0) {
    >                       /* skip emergency messages, already printed */
    >                       if (con->printk_seq < seq)
    > @@ -3134,10 +3136,12 @@ static void
    call_emergency_console_drivers(int level, const char *text,
    >       for_each_console(con) {
    >               if (!(con->flags & CON_ENABLED))
    >                       continue;
    > +#if 0
    >               if (con->write_atomic && level <
    emergency_console_loglevel) {
    >                       con->write_atomic(con, text, text_len);
    >                       continue;
    >               }
    > +#endif
    >               if (con->write && (con->flags & CON_BOOT)) {
    >                       con->write(con, text, text_len);
    >                       continue;



--
- Thou shalt not follow the NULL pointer, for chaos and madness await thee at its end
- "Use the force Harry" - Gandalf, Star Trek II

-- 
_______________________________________________
linux-yocto mailing list
linux-yocto@yoctoproject.org
https://lists.yoctoproject.org/listinfo/linux-yocto

Reply via email to