All of lore.kernel.org
 help / color / mirror / Atom feed
From: Stefan Roese <sr@denx.de>
To: Joel Stanley <joel@jms.id.au>
Cc: u-boot@lists.denx.de, trini@konsulko.com, sjg@chromium.org,
	rasmus.villemoes@prevas.dk,
	Chia-Wei Wang <chiawei_wang@aspeedtech.com>
Subject: Re: [RFC PATCH 8/8] WIP: .azure-pipelines.yml: Remove evb-ast2600
Date: Fri, 2 Sep 2022 09:44:18 +0200	[thread overview]
Message-ID: <4cfd48da-3376-62d4-b092-6b2f477f3766@denx.de> (raw)
In-Reply-To: <914db124-f920-0f4d-b701-fcc280bf61e5@denx.de>

Hi Joel,

On 02.09.22 08:14, Stefan Roese wrote:
> Hi Joel,
> 
> On 02.09.22 08:00, Joel Stanley wrote:
>> On Fri, 2 Sept 2022 at 04:09, Joel Stanley <joel@jms.id.au> wrote:
>>>
>>> On Mon, 29 Aug 2022 at 06:23, Stefan Roese <sr@denx.de> wrote:
>>>>
>>>> The evb-ast2600 target always runs into an timeout error when run via
>>>> Azure CI. For test purpose only, this patch removes this CI build
>>>> target so that the world build can be run.
>>>>
>>>> Joel, Chia-Wei, could you perhaps take a look at this and help to fix
>>>> the problems that I'm seeing with this build target here?
>>>
>>> Where does it hang? Can you attach the logs?
>>
>> The hang appears to be in the timer code, where the spl hangs due to a
>> div by zero. When inspecting, gd->arch.timer_rate_hz is zero at this
>> point.
>>
>> However I also noticed the cycling code introduces a symbol that ends
>> up in BSS. If I give cyclic_running a non-zero value it moves back
>> into the data section, and the SPL and u-boot proper make it all the
>> way through.
> 
> I might rework the code a bit, so that cyclic does not introduce new
> static variables but uses a new data pointer in struct global_data
> instead.
> 
>> This suggests there's still something wrong with the BSS handling for
>> the 2600 SPL (not resolved by f6810b749f2e ("aspeed/ast2600: Fix SPL
>> linker script") or the follow up fix). If I use the common arm spl
>> linker script, the system boots.
> 
> Thanks for looking into this.

Using the GD for the cyclic data and instead of new static variables in
this cyclic code seems to fix this CI build issue AFAICT. I'll send an
updates patchset later today.

Thanks,
Stefan

> 
>> Chai Wei, I think we need a different solution for placing the BSS
>> outside of the flash.
> 
> Thanks,
> Stefan
> 
>> Cheers,
>>
>> Joel
>>
>>>
>>> What kind of debugging have you tried to date?
>>>
>>> Cheers,
>>>
>>> Joel
>>>
>>>>
>>>> Signed-off-by: Stefan Roese <sr@denx.de>
>>>> Cc: Joel Stanley <joel@jms.id.au>
>>>> Cc: Chia-Wei Wang <chiawei_wang@aspeedtech.com>
>>>> ---
>>>>   .azure-pipelines.yml | 3 ---
>>>>   1 file changed, 3 deletions(-)
>>>>
>>>> diff --git a/.azure-pipelines.yml b/.azure-pipelines.yml
>>>> index 0fa92479b4c4..e48f1667f8de 100644
>>>> --- a/.azure-pipelines.yml
>>>> +++ b/.azure-pipelines.yml
>>>> @@ -261,9 +261,6 @@ stages:
>>>>           evb_ast2500:
>>>>             TEST_PY_BD: "evb-ast2500"
>>>>             TEST_PY_ID: "--id qemu"
>>>> -        evb_ast2600:
>>>> -          TEST_PY_BD: "evb-ast2600"
>>>> -          TEST_PY_ID: "--id qemu"
>>>>           vexpress_ca9x4:
>>>>             TEST_PY_BD: "vexpress_ca9x4"
>>>>             TEST_PY_ID: "--id qemu"
>>>> -- 
>>>> 2.37.2
>>>>
> 
> Viele Grüße,
> Stefan Roese
> 

Viele Grüße,
Stefan Roese

-- 
DENX Software Engineering GmbH,      Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-51 Fax: (+49)-8142-66989-80 Email: sr@denx.de

      reply	other threads:[~2022-09-02  7:44 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-08-29  6:23 [RFC PATCH 0/8] Migrate watchdog reset to cyclic infrastructure Stefan Roese
2022-08-29  6:23 ` [RFC PATCH 1/8] cmd/cyclic: Use div64 macros for division and remainder Stefan Roese
2022-08-30  2:31   ` Simon Glass
2022-08-29  6:23 ` [RFC PATCH 2/8] watchdog: Integrate watchdog triggering into the cyclic framework Stefan Roese
2022-08-29  7:38   ` Rasmus Villemoes
2022-08-29  8:09     ` Stefan Roese
2022-08-29  6:23 ` [RFC PATCH 3/8] cyclic: Introduce schedule() function Stefan Roese
2022-08-30  2:31   ` Simon Glass
2022-08-29  6:23 ` [RFC PATCH 4/8] cyclic: Use schedule() instead of WATCHDOG_RESET() Stefan Roese
2022-08-30  2:31   ` Simon Glass
2022-08-29  6:23 ` [RFC PATCH 5/8] watchdog: Get rid of ASSEMBLY hacks Stefan Roese
2022-08-29  7:50   ` Rasmus Villemoes
2022-08-29  8:31     ` Stefan Roese
2022-08-29  6:23 ` [RFC PATCH 6/8] watchdog: Remove WATCHDOG_RESET macro Stefan Roese
2022-08-30  2:31   ` Simon Glass
2022-08-29  6:23 ` [RFC PATCH 7/8] watchdog: Further cleanup Stefan Roese
2022-08-30  2:31   ` Simon Glass
2022-08-29  6:23 ` [RFC PATCH 8/8] WIP: .azure-pipelines.yml: Remove evb-ast2600 Stefan Roese
2022-09-02  4:09   ` Joel Stanley
2022-09-02  6:00     ` Joel Stanley
2022-09-02  6:14       ` Stefan Roese
2022-09-02  7:44         ` Stefan Roese [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4cfd48da-3376-62d4-b092-6b2f477f3766@denx.de \
    --to=sr@denx.de \
    --cc=chiawei_wang@aspeedtech.com \
    --cc=joel@jms.id.au \
    --cc=rasmus.villemoes@prevas.dk \
    --cc=sjg@chromium.org \
    --cc=trini@konsulko.com \
    --cc=u-boot@lists.denx.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.