Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 1 Mar 2009 10:32:37 -0800 (PST)
From:      Jakub Lach <jakub_lach@mailplus.pl>
To:        freebsd-current@freebsd.org
Subject:   Re: Interrupt stom on cardbus device
Message-ID:  <22276147.post@talk.nabble.com>
In-Reply-To: <1235930810.1236.5.camel@widget.2hip.net>
References:  <200902271730.07660.snasonov@bcc.ru> <200902271208.48212.jhb@freebsd.org> <1235760629.1289.51.camel@widget.2hip.net> <200902271403.46828.jhb@freebsd.org> <597436B54F5FF74FA8EA7F3224FE251F12C69213@mail.bcc> <1235778229.1289.60.camel@widget.2hip.net> <597436B54F5FF74FA8EA7F3224FE251F12C69214@mail.bcc> <1235839108.1289.72.camel@widget.2hip.net> <597436B54F5FF74FA8EA7F3224FE251F12C69215@mail.bcc> <1235843413.1289.92.camel@widget.2hip.net> <20090228215536.GA1625@wep4035.physik.uni-wuerzburg.de> <22271062.post@talk.nabble.com> <1235930810.1236.5.camel@widget.2hip.net>

next in thread | previous in thread | raw e-mail | index | archive | help

By stuttering I mean that animated gears stop, and FPS count is greatly
reduced 900FPS->20FPS
However, in real 3D accelerated apliccations FPS count is the same as
before, so I'm happy.

best regards,=20
-Jakub Lach
=20

Robert Noland wrote:
>=20
> On Sun, 2009-03-01 at 01:15 -0800, Jakub Lach wrote:
>> Hello.
>>=20
>> I also had such problems with interrupt storm @ IRQ 16. Hardware is
>> Lenovo
>> T400 (X4500,GM45) Patch greatly helped, although it  introduces other
>> problem. As long as output from glxgears is at ttyv0, it is working fine=
.
>> I
>> output is @ xterm window, it stutters while writing the FPS count. If
>> output
>> is at ttyv0 and I switch terminal from X11 to ttyv0 while glxgears is
>> working glxgears stutters too.=20
>=20
> Ok, I can try to look into this at some point, but right now it is not
> high priority.  When you VT switch, the irq handler gets uninstalled,
> and the vblank counters get reset in hardware.  We attempt to deal with
> that case, but it's not optimal.  As for stutter when printing while in
> X, it could be text rendering performance ( which kinda sucks right now,
> at least on intel ), it could also be that the G45 handles vblank
> counters differently.
>=20
> But I'm glad that the patch resolved the interrupt storm.  It's likely
> that it will help other devices than graphics cards as well...
>=20
> robert.
>=20
>> dmesg
>>=20
>> drm0: <Mobile Intel=C3=82=C5=BD GM45 Express Chipset> on vgapci0
>> info: [drm] MSI enabled 1 message(s)
>> vgapci0: child drm0 requested pci_enable_busmaster
>> info: [drm] AGP at 0xd0000000 256MB
>> info: [drm] Initialized i915 1.6.0 20080730
>> drm0: [ITHREAD]
>> error: [drm:pid26643:gm45_get_vblank_counter] *ERROR* trying to get
>> vblank
>> count for disabled pipe 0
>> drm0: [ITHREAD]
>> drm0: [ITHREAD]
>> drm0: [ITHREAD]
>>=20
>> best regards,=20
>> Jakub Lach
> --=20
> Robert Noland <rnoland@FreeBSD.org>
> FreeBSD
>=20
> =20
>=20

--=20
View this message in context: http://www.nabble.com/Interrupt-stom-on-cardb=
us-device-tp22247443p22276147.html
Sent from the freebsd-current mailing list archive at Nabble.com.




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?22276147.post>