Looking at giving DF a go... advice?

[Mark Cullen]

----- Original Message ----- 
From: "Matthew Dillon" <dillon at xxxxxxxxxxxxxxxxxxxx>
To: "Markie" <mark.cullen at xxxxxxxxxxxxx>
Sent: Sunday, October 10, 2004 8:01 PM
Subject: Re: Looking at giving DF a go... advice?


| :I just patched it with the patch I had, it still panic'ed, however it was
| :missing the top part of the message where it says <AGP something
| :something>.
| :
| :However, I just now (think) I remembered what I had to do to fix it on
4.x.
| :The part that reads:
| :

agp_ali.c

| :sc->initial_aperture = AGP_GET_APERTURE(dev);
| :if (sc->inital_aperture == 0) {
| :    device_printf(dev, "bad initial aperture size, disabling\n");
| :    return ENXIO;
| :}
| :
| :Should read..
| :

agp_ali.c

| :sc->initial_aperture = AGP_GET_APERTURE(dev);
| :if (sc->inital_aperture == 0) {
| :    device_printf(dev, "bad initial aperture size, disabling\n");
| :    agp_generic_detach(dev);
| :    return ENXIO;
| :}
| :
| :Ok well I just now added that line into the original patch and it hasn't
| :paniced this time!
|
|     Excellent!

:-)

|
| :Should I just try adding the agp_deneric_detach(dev); to the original
| :agp_ali.c source?
|
|     The above patch is definitely the 'fix'.   The for()->while()
|     loop patch looks reasonable, too, but the original code should work
|     just as well with just the added detach so that is what I am going
|     to do.

I'll revert back to the original source and add the detach later on then.

|
|     p.s. any FreeBSD folks reading this: the fix should be applied to
|     FreeBSD-5.x, and HEAD as well, and FreeBSD-4.x could use both the
|     initial size check AND the above fix.

I actually sent the patch as a private email rather than back to the group,
possibly shouldn't have done that. Sorry!
I've replied to the message where you asked for the patch, so it should be
back on the list.

On a side note I have managed to get my desktop pretty much the way it is
under FreeBSD. However, the ndis driver didn't want to work, gives me some
error (i'll get that later if anyone is interested). smbfs seems to be
broken too, not sure if anyone knows about that? Just comes up with 'exec
format error' or something along those lines.

I've attached the patch I sent to Matt too, sorry for the screw up!

|
| -Matt
| Matthew Dillon
| <dillon at xxxxxxxxxxxxx>


begin 666 agp_ali.diff
M26YD97 at Z('-Y<R]P8VDO86=P7V%L:2YC#0H]/3T]/3T]/3T]/3T]/3T]/3T]
M/3T]/3T]/3T]/3T]/3T]/3T]/3T]/3T]/3T]/3T]/3T]/3T]/3T]/3T]/3T]
M/3T]#0I20U, at 9FEL93H@+VAO;64O;F-V<R]S<F,O<WES+W!C:2]A9W!?86QI
M+F,L=@T*<F5T<FEE=FEN9R!R979I<VEO;B Q+C at -"F1I9F8 at +74@+7(Q+C@@
M86=P7V%L:2YC#0HM+2T@<WES+W!C:2]A9W!?86QI+F,),C(@075G(#(P,#,@
M,#<Z,3,Z,C @+3 P,# ),2XX#0HK*RL@<WES+W!C:2]A9W!?86QI+F,)-2!/
M8W0@,C P,R Q.#HQ.3HP,B M,# P, T*0$ @+3$P,BPR,2 K,3 R+#(P($! 
M#0H@"0ER971U<FX at 97)R;W([#0H@#0H@"7-C+3YI;FET:6%L7V%P97)T=7)E
M(#T at 04=07T=%5%]!4$525%5212AD978I.PT**PEG871T(#T at 3E5,3#L-"B -
M"BT)9F]R("@[.RD@>PT**PEW:&EL92 H04=07T=%5%]!4$525%5212AD978I
M("$](# I('L-"B )"6=A='0@/2!A9W!?86QL;V-?9V%T="AD978I.PT*+0D)
M:68@*&=A='0I#0HK"0EI9B H9V%T=" A/2!.54Q,*0T*( D)"6)R96%K.PT*
M+0T*+0D)+RH-"BT)"2 J(%!R;V)A8FQY(&-O;G1I9VUA;&QO8R!F86EL=7)E
M+B!4<GD@<F5D=6-I;F<@=&AE#0HM"0D@*B!A<&5R='5R92!S;R!T:&%T('1H
M92!G871T('-I>F4@<F5D=6-E<RX-"BT)"2 J+PT*+0D):68@*$%'4%]3151?
M05!%4E154D4H9&5V+"!!1U!?1T547T%015)455)%*&1E=BD at +R R*2D@>PT*
M+0D)"6%G<%]G96YE<FEC7V1E=&%C:"AD978I.PT*+0D)"7)E='5R;B!%3D]-
M14T[#0HM"0E]#0HK"0E!1U!?4T547T%015)455)%*&1E=BP at 04=07T=%5%]!
M4$525%5212AD978I("\@,BD[#0HK"7T-"BL)"0T**PEI9B H9V%T=" ]/2!.
M54Q,*2![#0HK"0EA9W!?9V5N97)I8U]D971A8V at H9&5V*3L-"BL)"7)E='5R
M;B!%3D]-14T[#0H@"7T-"BL-"B )<V,M/F=A='0@/2!G871T.PT*( T*( DO
6*B!);G-T86QL('1H92!G871T+B J+P``
`
end