From mboxrd@z Thu Jan  1 00:00:00 1970
From: erik quanstrom <quanstro@quanstro.net>
Date: Wed, 14 Jan 2015 21:21:58 -0800
To: 9fans@9fans.net
Message-ID: <ede2d0798558d31dc68c90626ec58989@brasstown>
In-Reply-To: <581cb553c4f0794103d1444a810602cd@felloff.net>
References: <581cb553c4f0794103d1444a810602cd@felloff.net>
MIME-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit
Subject: Re: [9fans] aux/vga hsync/hblank confustion
Topicbox-Message-UUID: 39fdb522-ead9-11e9-9d60-3106f5b1d025

> there are some vgadb entries that specify shs *smaller* than shb and the
> edid parser does the following:
>
> 	m->x = ha;
> 	m->ht = ha+hb;
> 	m->shs = ha;				<- *WHAT*
> 	m->shb = ha+hso;			<- *WHAT*
> 	m->ehb = ha+hso+hspw;		<- *WHAT*
> 	m->ehs = ha+hb;				<- *WHAT*
>
> where:
>
> ha = horizontal active
> hb = horizontal blanking
> hso = horizontal sync pulse offset (relative to start of blank)
> hspw = horisontal sync pulse width
>
> the hardcoded vesa standard modes in vesadb.c seem to use the field
> in the right way i think.
>
> interestinly, shs and ehb are rarely given in vgadb and we default
> to: shs = shb and ehs = ehb when omited, so this might explain why
> stuff works at all.
>
> so what is going on here?

this might explain some frustration i had years ago with getting timings correct.

- erik