On Tue, May 12, 2020 at 4:52 PM Roy Spliet <nouveau at spliet.org> wrote:> > Op 12-05-2020 om 14:36 schreef Alex Deucher: > > On Tue, May 12, 2020 at 4:16 AM Michel D?nzer <michel at daenzer.net> wrote: > >> > >> On 2020-05-11 10:12 p.m., Alex Deucher wrote: > >>> On Mon, May 11, 2020 at 1:17 PM Christian K?nig > >>> <ckoenig.leichtzumerken at gmail.com> wrote: > >>>> > >>>> AGP is deprecated for 10+ years now and not used any more on modern hardware. > >>>> > >>>> Old hardware should continue to work in PCI mode. > >>> > >>> Might want to clarify that there is no loss of functionality here. > >>> Something like: > >>> > >>> "There is no loss of functionality here. GPUs will continue to > >>> function. This just drops the use of the AGP MMU in the chipset in > >>> favor of the MMU on the device which has proven to be much more > >>> reliable. Due to its unreliability, AGP support has been disabled on > >>> PowerPC for years already so there is no change on PowerPC." > >> > >> There's a difference between something being disabled by default or not > >> being available at all. We may decide it's worth it anyway, but let's do > >> it based on facts. > >> > > > > I didn't mean to imply that AGP GART support was already removed. But > > for the vast majority of users the end result is the same. If you > > knew enough re-enable AGP GART, you probably wouldn't have been as > > confused about what this patch set does either. To reiterate, this > > patch set does not remove support for AGP cards, it only removes the > > support for AGP GART. The cards will still be functional using the > > device GART. There may be performance tradeoffs there in some cases. > > I'll volunteer to be the one asking: how big is this performance > difference? Have any benchmarks been run before and after removal of AGP > GART code on affected nouveau/radeon systems? Or is this code being > dropped _just_ because it's cumbersome, with no regard for metrics that > determine the value of AGP GART support? >I don't think anyone has any solid numbers, just anecdotal from memory. I certainly don't have any functional AGP systems at this point. It's mostly just cumbersome and would allow us to clean ttm and probably improve stability at the same time. At least on the radeon side, the only native AGP cards were r1xx, r2xx, and some of the early r3xx boards. Once we switched to pcie mid-way through r3xx, everything was native pcie and the AGP cards used a pcie to AGP bridge chip so they had a decent on chip MMU. Those older cards topped out at maybe 32 or 64 MB of vram, so they are going to be hard pressed to deal with modern desktops anyway. No idea what sort of GART capabilities NV AGP hardware at this time had. Alex> Roy > > > > > Alex > > _______________________________________________ > > Nouveau mailing list > > Nouveau at lists.freedesktop.org > > https://lists.freedesktop.org/mailman/listinfo/nouveau > > >
Christian König
2020-May-13 07:46 UTC
[Nouveau] [PATCH 1/3] drm/radeon: remove AGP support
Am 12.05.20 um 23:12 schrieb Alex Deucher:> On Tue, May 12, 2020 at 4:52 PM Roy Spliet <nouveau at spliet.org> wrote: >> Op 12-05-2020 om 14:36 schreef Alex Deucher: >>> On Tue, May 12, 2020 at 4:16 AM Michel D?nzer <michel at daenzer.net> wrote: >>>> On 2020-05-11 10:12 p.m., Alex Deucher wrote: >>>>> On Mon, May 11, 2020 at 1:17 PM Christian K?nig >>>>> <ckoenig.leichtzumerken at gmail.com> wrote: >>>>>> AGP is deprecated for 10+ years now and not used any more on modern hardware. >>>>>> >>>>>> Old hardware should continue to work in PCI mode. >>>>> Might want to clarify that there is no loss of functionality here. >>>>> Something like: >>>>> >>>>> "There is no loss of functionality here. GPUs will continue to >>>>> function. This just drops the use of the AGP MMU in the chipset in >>>>> favor of the MMU on the device which has proven to be much more >>>>> reliable. Due to its unreliability, AGP support has been disabled on >>>>> PowerPC for years already so there is no change on PowerPC." >>>> There's a difference between something being disabled by default or not >>>> being available at all. We may decide it's worth it anyway, but let's do >>>> it based on facts. >>>> >>> I didn't mean to imply that AGP GART support was already removed. But >>> for the vast majority of users the end result is the same. If you >>> knew enough re-enable AGP GART, you probably wouldn't have been as >>> confused about what this patch set does either. To reiterate, this >>> patch set does not remove support for AGP cards, it only removes the >>> support for AGP GART. The cards will still be functional using the >>> device GART. There may be performance tradeoffs there in some cases. >> I'll volunteer to be the one asking: how big is this performance >> difference? Have any benchmarks been run before and after removal of AGP >> GART code on affected nouveau/radeon systems? Or is this code being >> dropped _just_ because it's cumbersome, with no regard for metrics that >> determine the value of AGP GART support? >> > I don't think anyone has any solid numbers, just anecdotal from > memory. I certainly don't have any functional AGP systems at this > point. It's mostly just cumbersome and would allow us to clean ttm > and probably improve stability at the same time. At least on the > radeon side, the only native AGP cards were r1xx, r2xx, and some of > the early r3xx boards. Once we switched to pcie mid-way through r3xx, > everything was native pcie and the AGP cards used a pcie to AGP bridge > chip so they had a decent on chip MMU. Those older cards topped out > at maybe 32 or 64 MB of vram, so they are going to be hard pressed to > deal with modern desktops anyway. No idea what sort of GART > capabilities NV AGP hardware at this time had.I could only test with an old x86 Mac and an r3xx generation hw and in this case making the switch didn't had any noticeable effect at all. But I didn't do more than playing around with the desktop effects and playing a video. I do have a PC x86 AGP board lying around somewhere here, going top give that one a try a well. Christian.> > Alex > >> Roy >> >>> Alex >>> _______________________________________________ >>> Nouveau mailing list >>> Nouveau at lists.freedesktop.org >>> https://lists.freedesktop.org/mailman/listinfo/nouveau >>>
On 2020-05-13 9:46 a.m., Christian K?nig wrote:> Am 12.05.20 um 23:12 schrieb Alex Deucher: >> On Tue, May 12, 2020 at 4:52 PM Roy Spliet <nouveau at spliet.org> wrote: >>> >>> I'll volunteer to be the one asking: how big is this performance >>> difference? Have any benchmarks been run before and after removal of AGP >>> GART code on affected nouveau/radeon systems? Or is this code being >>> dropped _just_ because it's cumbersome, with no regard for metrics that >>> determine the value of AGP GART support? >>> >> I don't think anyone has any solid numbers, just anecdotal from >> memory.? I certainly don't have any functional AGP systems at this >> point.? It's mostly just cumbersome and would allow us to clean ttm >> and probably improve stability at the same time.? At least on the >> radeon side, the only native AGP cards were r1xx, r2xx, and some of >> the early r3xx boards.? Once we switched to pcie mid-way through r3xx, >> everything was native pcie and the AGP cards used a pcie to AGP bridge >> chip so they had a decent on chip MMU.? Those older cards topped out >> at maybe 32 or 64 MB of vram, so they are going to be hard pressed to >> deal with modern desktops anyway.? No idea what sort of GART >> capabilities NV AGP hardware at this time had. > > I could only test with an old x86 Mac and an r3xx generation hw and in > this case making the switch didn't had any noticeable effect at all. > > But I didn't do more than playing around with the desktop effects and > playing a video.Yeah, that's not enough to see a difference. Try an OpenGL game, or even just glxgears. -- Earthling Michel D?nzer | https://redhat.com Libre software enthusiast | Mesa and X developer