Re: [PATCH 00/26] leon: various fixes and TN0018 errata workaround

[Daniel Hellstrom]

On 2021-03-08 16:43, Joel Sherrill wrote:



On Sun, Mar 7, 2021 at 9:51 AM Daniel Hellstrom > wrote:


On 2020-09-23 17:05, Gedare Bloom wrote:

On Wed, Sep 23, 2020 at 4:34 AM Daniel Hellstrom  
  wrote:

Hi Sebastian,

Thanks for asking and sorry for dropping the ball on these.

The status is that two needs updating (BSD license for new CAN files and
the last tn0018 patch needs some redesign based on feedback) and the
others are accepted for master. I've sent an response on the tn0018
errata patch just now. I would like to push them on the 5 and master
branches. To get them onto 5, should  I create a ticket for the whole
patch set? I will try getting this done next next couple of days, and
have a look at you patches too, thanks!


It would be good to separate them logically to the TN-0018 errata
fixes vs the CAN/grlib improvements. The concern for pushing them to 5
is that they touch core sparc files, but since you guys are releasing
them this way in RCC I'm also comfortable with it. I didn't see any
changes outside the sparc (since currently grlib is sparc-specific
too). We'll need those tickets to help us with the dot-release notes.


Sorry for my very late response. There were some more updates on a
few of the patches based on the review comments which has been
addressed. I have now created tickets for all of them which are
referenced from the patches, so I will go ahead and push them for
the 5-branch (the posted patches targeted 5).


I agree with Gedare on trusting the patches. My only concern is making 
sure proper tickets are filed. A couple of guidelines may help decide 
how many tickets and for what.


The first thing to remember is that tickets are automatically 
processed into release notes. If it is important enough to show up in 
a release note, file a ticket. I have been prodding Ryan to file 
tickets for the Coverity issues because I think they should be in 
release notes.


For 5, any changes should have tickets. This is a long standing rule 
for release branches.


Thanks for the comments, I will keep that in mind going forward. I made 
a couple of tickets for the RTEMS/master and tickets for all patches for 
5.2 milestone.




However, I will wait with the TN-0018 before I get an acknowledge
for that one. I updated the its ticket with links to the GCC patch
that has now been accepted into upstreams GCC (GCC-10 stable and
master). The TN0018 patch is not enabled if the GCC-patch is not
included in the toolchain, so I believe it should be ok to push,
even before RSB is updated?


It sounds like it will be ok.

Ok, thanks!


What happens with TN0018 on the 5 branch where we are using older tools?

https://devel.rtems.org/ticket/4155


I have submitted a RSB patch which as been acked by Chris and Sebastian, 
so I will proceed to push the tn0018 patch to 5 now.




Next step for me is to add some configurations for the new build
system before I can push them to RTEMS/master.

This has been done and pushed now. waf is really a speed improvement! 
Thanks!


Thanks for submitting all these. Is this going to clean your queue?


The queue is much smaller now! These were the most important when I 
started but got choked, I will follow up with a few important fixes done 
lately.


/Daniel



--joel

Thanks,
/Daniel



Kind Regards,
Daniel

On 2020-09-18 10:03, Sebastian Huber wrote:

Hallo Daniel,

what are your plans with respect to this patch set?

Please also have a look at:

https://lists.rtems.org/pipermail/devel/2020-September/062176.html  



___
devel mailing list
devel@rtems.org  
http://lists.rtems.org/mailman/listinfo/devel  


___
devel mailing list
devel@rtems.org 
http://lists.rtems.org/mailman/listinfo/devel


___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel

Re: [PATCH 00/26] leon: various fixes and TN0018 errata workaround

[Joel Sherrill]

On Sun, Mar 7, 2021 at 9:51 AM Daniel Hellstrom  wrote:

> On 2020-09-23 17:05, Gedare Bloom wrote:
>
> On Wed, Sep 23, 2020 at 4:34 AM Daniel Hellstrom  
>  wrote:
>
> Hi Sebastian,
>
> Thanks for asking and sorry for dropping the ball on these.
>
> The status is that two needs updating (BSD license for new CAN files and
> the last tn0018 patch needs some redesign based on feedback) and the
> others are accepted for master. I've sent an response on the tn0018
> errata patch just now. I would like to push them on the 5 and master
> branches. To get them onto 5, should  I create a ticket for the whole
> patch set? I will try getting this done next next couple of days, and
> have a look at you patches too, thanks!
>
>
> It would be good to separate them logically to the TN-0018 errata
> fixes vs the CAN/grlib improvements. The concern for pushing them to 5
> is that they touch core sparc files, but since you guys are releasing
> them this way in RCC I'm also comfortable with it. I didn't see any
> changes outside the sparc (since currently grlib is sparc-specific
> too). We'll need those tickets to help us with the dot-release notes.
>
> Sorry for my very late response. There were some more updates on a few of
> the patches based on the review comments which has been addressed. I have
> now created tickets for all of them which are referenced from the patches,
> so I will go ahead and push them for the 5-branch (the posted patches
> targeted 5).
>

I agree with Gedare on trusting the patches. My only concern is making sure
proper tickets are filed. A couple of guidelines may help decide how many
tickets and for what.

The first thing to remember is that tickets are automatically processed
into release notes. If it is important enough to show up in a release note,
file a ticket. I have been prodding Ryan to file tickets for the Coverity
issues because I think they should be in release notes.

For 5, any changes should have tickets. This is a long standing rule for
release branches.

> However, I will wait with the TN-0018 before I get an acknowledge for that
> one. I updated the its ticket with links to the GCC patch that has now been
> accepted into upstreams GCC (GCC-10 stable and master). The TN0018 patch is
> not enabled if the GCC-patch is not included in the toolchain, so I believe
> it should be ok to push, even before RSB is updated?
>

It sounds like it will be ok.

What happens with TN0018 on the 5 branch where we are using older tools?

> https://devel.rtems.org/ticket/4155
>
> Next step for me is to add some configurations for the new build system
> before I can push them to RTEMS/master.
>

Thanks for submitting all these. Is this going to clean your queue?

--joel

> Thanks,
> /Daniel
>
>
> Kind Regards,
> Daniel
>
> On 2020-09-18 10:03, Sebastian Huber wrote:
>
> Hallo Daniel,
>
> what are your plans with respect to this patch set?
>
> Please also have a look at:
> https://lists.rtems.org/pipermail/devel/2020-September/062176.html
>
> ___
> devel mailing listdevel@rtems.orghttp://lists.rtems.org/mailman/listinfo/devel
>
> ___
> devel mailing list
> devel@rtems.org
> http://lists.rtems.org/mailman/listinfo/devel
___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel

Re: [PATCH 00/26] leon: various fixes and TN0018 errata workaround

[Daniel Hellstrom]

On 2020-09-23 17:05, Gedare Bloom wrote:

On Wed, Sep 23, 2020 at 4:34 AM Daniel Hellstrom  wrote:

Hi Sebastian,

Thanks for asking and sorry for dropping the ball on these.

The status is that two needs updating (BSD license for new CAN files and
the last tn0018 patch needs some redesign based on feedback) and the
others are accepted for master. I've sent an response on the tn0018
errata patch just now. I would like to push them on the 5 and master
branches. To get them onto 5, should  I create a ticket for the whole
patch set? I will try getting this done next next couple of days, and
have a look at you patches too, thanks!


It would be good to separate them logically to the TN-0018 errata
fixes vs the CAN/grlib improvements. The concern for pushing them to 5
is that they touch core sparc files, but since you guys are releasing
them this way in RCC I'm also comfortable with it. I didn't see any
changes outside the sparc (since currently grlib is sparc-specific
too). We'll need those tickets to help us with the dot-release notes.


Sorry for my very late response. There were some more updates on a few 
of the patches based on the review comments which has been addressed. I 
have now created tickets for all of them which are referenced from the 
patches, so I will go ahead and push them for the 5-branch (the posted 
patches targeted 5).


However, I will wait with the TN-0018 before I get an acknowledge for 
that one. I updated the its ticket with links to the GCC patch that has 
now been accepted into upstreams GCC (GCC-10 stable and master). The 
TN0018 patch is not enabled if the GCC-patch is not included in the 
toolchain, so I believe it should be ok to push, even before RSB is updated?


    https://devel.rtems.org/ticket/4155

Next step for me is to add some configurations for the new build system 
before I can push them to RTEMS/master.


Thanks,
/Daniel





Kind Regards,
Daniel

On 2020-09-18 10:03, Sebastian Huber wrote:

Hallo Daniel,

what are your plans with respect to this patch set?

Please also have a look at:

https://lists.rtems.org/pipermail/devel/2020-September/062176.html


___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel
___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel

Re: [PATCH 00/26] leon: various fixes and TN0018 errata workaround

[Sebastian Huber]

Hello,

I added two tickets for the errata fixes:

https://devel.rtems.org/ticket/4155

https://devel.rtems.org/ticket/4156

--
Sebastian Huber, embedded brains GmbH

Address : Dornierstr. 4, D-82178 Puchheim, Germany
Phone   : +49 89 189 47 41-16
Fax : +49 89 189 47 41-09
E-Mail  : sebastian.hu...@embedded-brains.de
PGP : Public key available on request.

Diese Nachricht ist keine geschäftliche Mitteilung im Sinne des EHUG.
___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel

Re: [PATCH 00/26] leon: various fixes and TN0018 errata workaround

[Gedare Bloom]

On Wed, Sep 23, 2020 at 4:34 AM Daniel Hellstrom  wrote:
>
> Hi Sebastian,
>
> Thanks for asking and sorry for dropping the ball on these.
>
> The status is that two needs updating (BSD license for new CAN files and
> the last tn0018 patch needs some redesign based on feedback) and the
> others are accepted for master. I've sent an response on the tn0018
> errata patch just now. I would like to push them on the 5 and master
> branches. To get them onto 5, should  I create a ticket for the whole
> patch set? I will try getting this done next next couple of days, and
> have a look at you patches too, thanks!
>

It would be good to separate them logically to the TN-0018 errata
fixes vs the CAN/grlib improvements. The concern for pushing them to 5
is that they touch core sparc files, but since you guys are releasing
them this way in RCC I'm also comfortable with it. I didn't see any
changes outside the sparc (since currently grlib is sparc-specific
too). We'll need those tickets to help us with the dot-release notes.

> Kind Regards,
> Daniel
>
> On 2020-09-18 10:03, Sebastian Huber wrote:
> > Hallo Daniel,
> >
> > what are your plans with respect to this patch set?
> >
> > Please also have a look at:
> >
> > https://lists.rtems.org/pipermail/devel/2020-September/062176.html
> >
> ___
> devel mailing list
> devel@rtems.org
> http://lists.rtems.org/mailman/listinfo/devel
___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel

Re: [PATCH 00/26] leon: various fixes and TN0018 errata workaround

[Daniel Hellstrom]

Hi Sebastian,

Thanks for asking and sorry for dropping the ball on these.

The status is that two needs updating (BSD license for new CAN files and 
the last tn0018 patch needs some redesign based on feedback) and the 
others are accepted for master. I've sent an response on the tn0018 
errata patch just now. I would like to push them on the 5 and master 
branches. To get them onto 5, should  I create a ticket for the whole 
patch set? I will try getting this done next next couple of days, and 
have a look at you patches too, thanks!


Kind Regards,
Daniel

On 2020-09-18 10:03, Sebastian Huber wrote:

Hallo Daniel,

what are your plans with respect to this patch set?

Please also have a look at:

https://lists.rtems.org/pipermail/devel/2020-September/062176.html


___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel

Re: [PATCH 00/26] leon: various fixes and TN0018 errata workaround

[Sebastian Huber]

Hallo Daniel,

what are your plans with respect to this patch set?

Please also have a look at:

https://lists.rtems.org/pipermail/devel/2020-September/062176.html

___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel

Re: [PATCH 00/26] leon: various fixes and TN0018 errata workaround

[Sebastian Huber]

Hello Daniel,

thanks for the fixes. The patch set looks good, except patch 26/26.

___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel

[PATCH 00/26] leon: various fixes and TN0018 errata workaround

[Daniel Hellstrom]

Hi,

The GRLIB-TN-0018 errata (gaisler.com/notes) was recently published and
affecting the LEON3FT devices like UT699, UT700 and GR712RC. RTEMS SW
workarounds were released in May/June in the RCC dists so now I would
like to request to upstream that work and other differences between the
RTEMS5 code base for RCC-1.3.

Apart from fixes there is also driver support for the new CAN-FD in
GRLIB: GRCANFD. The restructuring of the CAN code is also an improvement
for GR712RC and GR740 by sharing more common code.

The patches are also found in the submit/20200629 branch of rcc.git:
  https://git.rtems.org/danielh/rcc.git/log/?h=submit/20200629

Please let me know if I should separate the patches in smaller chunks
and resend or any other comments. As most are fixes I didn't do that
initially.

I've just learn that RTEMS5 branched! It would be great if the patches
could go on both master and 5.

Kind Regards,
Daniel Hellstrom

---
 
 create mode 100644 bsps/include/grlib/canbtrs.h
 create mode 100644 bsps/shared/grlib/can/canbtrs.c
 create mode 100644 bsps/shared/grlib/can/grcan_internal.h
 create mode 100644 bsps/shared/grlib/can/grcanfd.c
 create mode 100644 bsps/shared/grlib/can/grcanstd.c
 create mode 100644 bsps/sparc/leon3/start/drvmgr_def_drivers.c
 create mode 100644 cpukit/score/cpu/sparc/include/libcpu/grlib-tn-0018.h
 bsps/headers.am|   1 +
 bsps/include/grlib/ahbstat.h   |   2 +
 bsps/include/grlib/ambapp_ids.h|  19 +
 bsps/include/grlib/canbtrs.h   |  62 ++
 bsps/include/grlib/grcan.h | 129 +++-
 bsps/include/grlib/greth.h |   1 +
 bsps/include/grlib/grlib_impl.h|  10 +
 bsps/include/grlib/grspw_pkt.h |   4 +-
 bsps/include/grlib/grspw_router.h  |   1 +
 bsps/shared/grlib-sources.am   |   3 +
 bsps/shared/grlib/1553/gr1553b.c   |   4 +
 bsps/shared/grlib/1553/gr1553bc.c  |  76 ++-
 bsps/shared/grlib/1553/gr1553bm.c  |  49 +-
 bsps/shared/grlib/1553/gr1553rt.c  | 123 ++--
 bsps/shared/grlib/amba/ahbstat.c   |   8 +
 bsps/shared/grlib/amba/ambapp_names.c  |  21 +-
 bsps/shared/grlib/btimer/tlib_ckinit.c |   7 +-
 bsps/shared/grlib/can/canbtrs.c| 116 
 bsps/shared/grlib/can/grcan.c  | 719 +++--
 bsps/shared/grlib/can/grcan_internal.h | 140 
 bsps/shared/grlib/can/grcanfd.c| 535 +++
 bsps/shared/grlib/can/grcanstd.c   | 435 +
 bsps/shared/grlib/can/occan.c  | 151 ++---
 bsps/shared/grlib/l2c/l2c.c|   4 +-
 bsps/shared/grlib/net/greth.c  |  22 +-
 bsps/shared/grlib/spw/grspw.c  |   7 +
 bsps/shared/grlib/spw/grspw_router.c   |   5 +
 bsps/sparc/leon3/config/gr712rc.cfg|   5 +-
 bsps/sparc/leon3/config/ut699.cfg  |   3 +-
 bsps/sparc/leon3/config/ut700.cfg  |   3 +-
 bsps/sparc/leon3/start/amba.c  |  12 -
 bsps/sparc/leon3/start/cpucounter.c|   5 +-
 bsps/sparc/leon3/start/drvmgr_def_drivers.c|  28 +
 c/src/lib/libbsp/sparc/leon3/Makefile.am   |   1 +
 cpukit/score/cpu/sparc/cpu_asm.S   |  14 +-
 cpukit/score/cpu/sparc/headers.am  |   1 +
 .../score/cpu/sparc/include/libcpu/grlib-tn-0018.h |  58 ++
 cpukit/score/cpu/sparc/include/rtems/score/sparc.h |   4 +
 cpukit/score/cpu/sparc/sparc-access.S  |   4 +-
 cpukit/score/cpu/sparc/sparc-counter-asm.S |   7 +-
 cpukit/score/cpu/sparc/syscall.S   |   3 +-
 cpukit/score/cpu/sparc/window.S|   9 +-
 42 files changed, 1927 insertions(+), 884 deletions(-)

--
2.7.4
___
devel mailing list
devel@rtems.org
http://lists.rtems.org/mailman/listinfo/devel