<div dir="auto"><div><br><br><div class="gmail_quote"><div dir="ltr">On Mon, Jul 30, 2018, 2:27 PM Amaan Cheval <<a href="mailto:amaan.cheval@gmail.com">amaan.cheval@gmail.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Quick status update: in working on the APIC timer, as a prerequisite,<br>
I've had to setup access to the Interrupt Descriptor Table (which is<br>
great because it helps us have the basic interrupt support we need at<br>
least).<br>
<br>
Another minor issue I've run into is the fact that the APIC is located<br>
at physical address 0xfee00000 by default on most x86 processors - per<br>
the FreeBSD bootloader's paging scheme, they map every GiB of virtual<br>
memory to the first 1 GiB of physical memory (0x40000000).<br>
<br>
As a workaround, I'll probably just move the APIC address to within<br>
this accessible range (through a Model Specific Register) with an<br>
"XXX" comment about creating a better paging scheme later.<br>
<br>
Let me know if anyone thinks better paging support should also come<br>
first! If not, what's next is initializing the PIT to calibrate the<br>
APIC timer, and then we should have a pretty nice and self-contained<br>
clock driver for the x86_64 port too.<br></blockquote></div></div><div dir="auto"><br></div><div dir="auto">Make the clock tick and timer driver work first. </div><div dir="auto"><br></div><div dir="auto">Will this have any impact on the amount of RAM accessible until this is fixed?</div><div dir="auto"><div class="gmail_quote"><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<br>
On Wed, Jul 18, 2018 at 7:53 PM, Gedare Bloom <<a href="mailto:gedare@rtems.org" target="_blank" rel="noreferrer">gedare@rtems.org</a>> wrote:<br>
> On Wed, Jul 18, 2018 at 10:17 AM, Joel Sherrill <<a href="mailto:joel@rtems.org" target="_blank" rel="noreferrer">joel@rtems.org</a>> wrote:<br>
>><br>
>><br>
>> On Wed, Jul 18, 2018 at 12:31 AM, Sebastian Huber<br>
>> <<a href="mailto:sebastian.huber@embedded-brains.de" target="_blank" rel="noreferrer">sebastian.huber@embedded-brains.de</a>> wrote:<br>
>>><br>
>>> Hello Amaan,<br>
>>><br>
>>> On 17/07/18 19:18, Amaan Cheval wrote:<br>
>>>><br>
>>>> Hi!<br>
>>>><br>
>>>> Now that I'm working on the clock driver, we need to pick what we<br>
>>>> support first. Our options in brief are:<br>
>>><br>
>>><br>
>>> The clock driver needs an interrupt. What is the status of the interrupt<br>
>>> controller support in the BSP?<br>
>>><br>
>>> For timekeeping we use a port of the FreeBSD timecounter in RTEMS. You may<br>
>>> have a look at the FreeBSD timecounter for this architecture, e.g.<br>
>>> sys/x86/x86/tsc.c. I looks quite complicated. I would not take to much care<br>
>>> about legacy support, e.g. ignore hardware which is older than five years?.<br>
>><br>
>><br>
>> That's not a good rule for PCs at all. The APIC was first introduced as an<br>
>> external controller with the i486,<br>
>> Based on your rule, we wouldn't support it even though it is the most likely<br>
>> choice.<br>
>><br>
><br>
> I believe he meant ignore hardware that is not available from products<br>
> in the last five years.<br>
><br>
><br>
>> Avoid things that are deemed legacy. The starting point for this is the old<br>
>> PC<br>
>> System Design Guide.<br>
>><br>
>> <a href="https://en.wikipedia.org/wiki/PC_System_Design_Guide" rel="noreferrer noreferrer" target="_blank">https://en.wikipedia.org/wiki/PC_System_Design_Guide</a><br>
>><br>
>> If it was deemed obsolete in PC2001, then you definitely want to avoid it.<br>
>> Those<br>
>> things are just now really disappearing.<br>
>><br>
><br>
> This is consistent with my interpretation.<br>
><br>
>> --joel<br>
>><br>
>>><br>
>>><br>
>>><br>
>>> --<br>
>>> Sebastian Huber, embedded brains GmbH<br>
>>><br>
>>> Address : Dornierstr. 4, D-82178 Puchheim, Germany<br>
>>> Phone : +49 89 189 47 41-16<br>
>>> Fax : +49 89 189 47 41-09<br>
>>> E-Mail : <a href="mailto:sebastian.huber@embedded-brains.de" target="_blank" rel="noreferrer">sebastian.huber@embedded-brains.de</a><br>
>>> PGP : Public key available on request.<br>
>>><br>
>>> Diese Nachricht ist keine geschäftliche Mitteilung im Sinne des EHUG.<br>
>>><br>
>><br>
</blockquote></div></div></div>