From: Michal Simek <michal.simek-gjFFaj9aHVfQT0dZR+AlfA@public.gmane.org>
To: Anatolij Gustschin <agust-ynQEQJNshbs@public.gmane.org>,
Michal Simek
<michal.simek-gjFFaj9aHVfQT0dZR+AlfA@public.gmane.org>
Cc: linux-fpga-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
Alan Tull
<atull-yzvPICuk2ABMcg4IHK0kFoH6Mc4MB0Vx@public.gmane.org>,
Moritz Fischer
<moritz.fischer-+aYTwkv1SeIAvxtiuMwx3w@public.gmane.org>,
Rob Herring <robh+dt-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>,
Mark Rutland <mark.rutland-5wv7dgnIgG8@public.gmane.org>
Subject: Re: [PATCH v3 2/2] fpga manager: Add Xilinx slave serial SPI driver
Date: Thu, 23 Feb 2017 10:33:13 +0100 [thread overview]
Message-ID: <8ef97075-a9a3-07cc-83df-f89860c8f15a@xilinx.com> (raw)
In-Reply-To: <20170223102852.47d5bf5d@crub>
On 23.2.2017 10:28, Anatolij Gustschin wrote:
> On Wed, 22 Feb 2017 16:04:47 +0100
> Michal Simek michal.simek-gjFFaj9aHVfQT0dZR+AlfA@public.gmane.org wrote:
> ...
>>> + /* program latency */
>>> + usleep_range(prog_latency_1000us, prog_latency_1000us + 500);
>>
>> Based on my colleague for spartan6 this should be 5ms and other xilinx
>> fpgas can take up to 7.5ms.
>>
>> In general, it is better to poll for INIT_B pin==High (but that takes
>> another GPIO) and wait until at least 10 ms before timeout.
>
> checked in the ug380.pdf, it depends on the speed grade, for some
> spartan6 it can be max. 4ms, for others 5ms. It is a max. value,
> INIT_B can go high earlier, but in my hw setup INIT_B is not
> connected. Someone with a hw setup with INIT_B connected can
> add optional support for INIT_P polling.
>
> ...
>>> +static int xilinx_spi_write_complete(struct fpga_manager *mgr,
>>> + struct fpga_image_info *info)
>>> +{
>>> + struct xilinx_spi_conf *conf = mgr->priv;
>>> + int wait;
>>> +
>>> + if (gpiod_get_value(conf->done))
>>> + return 0;
>>> +
>>> + wait = info->config_complete_timeout_us / FPGA_MIN_CFG_WAIT_US;
>>> + if (info->config_complete_timeout_us % FPGA_MIN_CFG_WAIT_US)
>>> + wait += 1;
>>> +
>>> + while (wait--) {
>>> + usleep_range(FPGA_MIN_CFG_WAIT_US, FPGA_MIN_CFG_WAIT_US + 5);
>>> +
>>> + if (gpiod_get_value(conf->done))
>>> + return 0;
>>> + }
>>> +
>>
>> Based on my colleague:
>> IMHO, for step 4 it is better to do these steps to cover more possible
>> configuration DONE scenarios:
>>
>> While ((DONE==0) and (NOT TIMEOUT)) {
>> Continue to apply CCLK cycles;
>> Check DONE;
>> timeout++;
>> }
>>
>> Apply at least 8 more CCLK cycles.
>>
>> The timeout should be equivalent to at least 10 ms of time.
>
> You mead Step 8 I think, not step 4. Will see what I can change.
This was just c&p from his reply with different steps numbers. :-)
Thanks,
Michal
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
prev parent reply other threads:[~2017-02-23 9:33 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-02-22 10:39 [PATCH v3 0/2] Xilinx Slave Serial FPGA Manager Anatolij Gustschin
[not found] ` <1487759951-20711-1-git-send-email-agust-ynQEQJNshbs@public.gmane.org>
2017-02-22 10:39 ` [PATCH v3 1/2] dt: bindings: fpga: add xilinx slave-serial binding description Anatolij Gustschin
[not found] ` <1487759951-20711-2-git-send-email-agust-ynQEQJNshbs@public.gmane.org>
2017-02-27 22:39 ` Rob Herring
2017-02-22 10:39 ` [PATCH v3 2/2] fpga manager: Add Xilinx slave serial SPI driver Anatolij Gustschin
[not found] ` <1487759951-20711-3-git-send-email-agust-ynQEQJNshbs@public.gmane.org>
2017-02-22 15:04 ` Michal Simek
[not found] ` <e9620f07-48bd-1f83-a686-6874becbba86-gjFFaj9aHVfQT0dZR+AlfA@public.gmane.org>
2017-02-23 9:28 ` Anatolij Gustschin
2017-02-23 9:33 ` Michal Simek [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=8ef97075-a9a3-07cc-83df-f89860c8f15a@xilinx.com \
--to=michal.simek-gjffaj9ahvfqt0dzr+alfa@public.gmane.org \
--cc=agust-ynQEQJNshbs@public.gmane.org \
--cc=atull-yzvPICuk2ABMcg4IHK0kFoH6Mc4MB0Vx@public.gmane.org \
--cc=devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-fpga-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=mark.rutland-5wv7dgnIgG8@public.gmane.org \
--cc=moritz.fischer-+aYTwkv1SeIAvxtiuMwx3w@public.gmane.org \
--cc=robh+dt-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).