search for: nouveau_agp

...uld work with. See https://bugs.freedesktop.org/show_bug.cgi?id=20341 Reported-by: Jason Detring <detringj at gmail.com> Signed-off-by: Ilia Mirkin <imirkin at alum.mit.edu> --- I didn't go as far as subdevice matching... IMO that's overkill for now. drivers/gpu/drm/nouveau/nouveau_agp.c | 44 +++++++++++++++++++++++++++++++---- 1 file changed, 39 insertions(+), 5 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_agp.c b/drivers/gpu/drm/nouveau/nouveau_agp.c index 6e7a55f..2953c4e 100644 --- a/drivers/gpu/drm/nouveau/nouveau_agp.c +++ b/drivers/gpu/drm/nouveau/nouveau_ag...

...m like it would be a bad idea. > > Reported-by: Jason Detring <detringj at gmail.com> > Signed-off-by: Ilia Mirkin <imirkin at alum.mit.edu> > --- > > I didn't go as far as subdevice matching... IMO that's overkill for now. > > drivers/gpu/drm/nouveau/nouveau_agp.c | 44 +++++++++++++++++++++++++++++++---- > 1 file changed, 39 insertions(+), 5 deletions(-) > > diff --git a/drivers/gpu/drm/nouveau/nouveau_agp.c b/drivers/gpu/drm/nouveau/nouveau_agp.c > index 6e7a55f..2953c4e 100644 > --- a/drivers/gpu/drm/nouveau/nouveau_agp.c > +++ b/driv...

...+339,8 @@ nouveau_drm_load(struct drm_device *dev, unsigned long flags) INIT_LIST_HEAD(&drm->clients); spin_lock_init(&drm->tile.lock); + nouveau_get_hdmi_dev(dev); + /* make sure AGP controller is in a consistent state before we * (possibly) execute vbios init tables (see nouveau_agp.h) */ @@ -400,6 +427,9 @@ fail_ttm: nouveau_agp_fini(drm); nouveau_vga_fini(drm); fail_device: + if (drm->hdmi_device) + pci_dev_put(drm->hdmi_device); + nouveau_cli_destroy(&drm->client); return ret; } @@ -424,6 +454,8 @@ nouveau_drm_unload(struct drm_device *dev) nouv...

...> drivers/gpu/drm/nouveau/core/subdev/instmem/nv40.c | 7 +- > drivers/gpu/drm/nouveau/core/subdev/mc/base.c | 39 ++++++---- > drivers/gpu/drm/nouveau/core/subdev/mxm/base.c | 2 +- > drivers/gpu/drm/nouveau/nouveau_abi16.c | 13 +++- > drivers/gpu/drm/nouveau/nouveau_agp.c | 2 +- > drivers/gpu/drm/nouveau/nouveau_bios.c | 4 ++ > drivers/gpu/drm/nouveau/nouveau_bo.c | 22 +++--- > drivers/gpu/drm/nouveau/nouveau_chan.c | 2 +- > drivers/gpu/drm/nouveau/nouveau_display.c | 3 +- > driver...

...v/i2c/base.c | 2 +- drivers/gpu/drm/nouveau/core/subdev/instmem/nv40.c | 7 +- drivers/gpu/drm/nouveau/core/subdev/mc/base.c | 39 ++++++---- drivers/gpu/drm/nouveau/core/subdev/mxm/base.c | 2 +- drivers/gpu/drm/nouveau/nouveau_abi16.c | 13 +++- drivers/gpu/drm/nouveau/nouveau_agp.c | 2 +- drivers/gpu/drm/nouveau/nouveau_bios.c | 4 ++ drivers/gpu/drm/nouveau/nouveau_bo.c | 22 +++--- drivers/gpu/drm/nouveau/nouveau_chan.c | 2 +- drivers/gpu/drm/nouveau/nouveau_display.c | 3 +- drivers/gpu/drm/nouveau/nouveau...

...uveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c > [...] >> @@ -345,7 +368,7 @@ nouveau_drm_load(struct drm_device *dev, unsigned long flags) >> /* make sure AGP controller is in a consistent state before we >> * (possibly) execute vbios init tables (see nouveau_agp.h) >> */ >> - if (drm_pci_device_is_agp(dev) && dev->agp) { >> + if (pdev && drm_pci_device_is_agp(dev) && dev->agp) { > > Perhaps move the check for valid PCI device into drm_pci_device_is_agp() > rather than requiring callers...

...truct drm_device *dev, unsigned long flags) INIT_LIST_HEAD(&drm->clients); spin_lock_init(&drm->tile.lock); - nouveau_get_hdmi_dev(dev); + nouveau_get_hdmi_dev(drm); /* make sure AGP controller is in a consistent state before we * (possibly) execute vbios init tables (see nouveau_agp.h) -- 1.8.3.2

...drm_device *dev, unsigned long flags) > INIT_LIST_HEAD(&drm->clients); > spin_lock_init(&drm->tile.lock); > > + nouveau_get_hdmi_dev(dev); > + > /* make sure AGP controller is in a consistent state before we > * (possibly) execute vbios init tables (see nouveau_agp.h) > */ > @@ -400,6 +427,9 @@ fail_ttm: > nouveau_agp_fini(drm); > nouveau_vga_fini(drm); > fail_device: > + if (drm->hdmi_device) > + pci_dev_put(drm->hdmi_device); > + > nouveau_cli_destroy(&drm->client); > return ret; > } > @@ -424,6...

...m/nouveau/core/subdev/instmem/nv40.c | 7 +- >> drivers/gpu/drm/nouveau/core/subdev/mc/base.c | 39 ++++++---- >> drivers/gpu/drm/nouveau/core/subdev/mxm/base.c | 2 +- >> drivers/gpu/drm/nouveau/nouveau_abi16.c | 13 +++- >> drivers/gpu/drm/nouveau/nouveau_agp.c | 2 +- >> drivers/gpu/drm/nouveau/nouveau_bios.c | 4 ++ >> drivers/gpu/drm/nouveau/nouveau_bo.c | 22 +++--- >> drivers/gpu/drm/nouveau/nouveau_chan.c | 2 +- >> drivers/gpu/drm/nouveau/nouveau_display.c...

...-git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c [...] > @@ -345,7 +368,7 @@ nouveau_drm_load(struct drm_device *dev, unsigned long flags) > /* make sure AGP controller is in a consistent state before we > * (possibly) execute vbios init tables (see nouveau_agp.h) > */ > - if (drm_pci_device_is_agp(dev) && dev->agp) { > + if (pdev && drm_pci_device_is_agp(dev) && dev->agp) { Perhaps move the check for valid PCI device into drm_pci_device_is_agp() rather than requiring callers to check explicitly? > @@ -384,8...

...e(nouveau_name(dev), "DRM", sizeof(*drm), + (void **)&drm); if (ret) return ret; @@ -345,7 +368,7 @@ nouveau_drm_load(struct drm_device *dev, unsigned long flags) /* make sure AGP controller is in a consistent state before we * (possibly) execute vbios init tables (see nouveau_agp.h) */ - if (drm_pci_device_is_agp(dev) && dev->agp) { + if (pdev && drm_pci_device_is_agp(dev) && dev->agp) { /* dummy device object, doesn't init anything, but allows * agp code access to registers */ @@ -384,8 +407,10 @@ nouveau_drm_load(struct drm_...

...ice *dev, unsigned long flags) INIT_LIST_HEAD(&drm->clients); spin_lock_init(&drm->tile.lock); - nouveau_get_hdmi_dev(dev); + if (pdev) + nouveau_get_hdmi_dev(dev); /* make sure AGP controller is in a consistent state before we * (possibly) execute vbios init tables (see nouveau_agp.h) */ - if (drm_pci_device_is_agp(dev) && dev->agp) { + if (pdev && drm_pci_device_is_agp(dev) && dev->agp) { /* dummy device object, doesn't init anything, but allows * agp code access to registers */ @@ -384,8 +402,10 @@ nouveau_drm_load(struct drm_...

...vc0.o -nouveau-y += core/engine/vp/nve0.o +nouveau-$(CONFIG_DRM_NOUVEAU_NVC0) += core/engine/vp/nvc0.o +nouveau-$(CONFIG_DRM_NOUVEAU_NVC0) += core/engine/vp/nve0.o # drm/core nouveau-y += nouveau_drm.o nouveau_chan.o nouveau_dma.o nouveau_fence.o @@ -299,12 +299,14 @@ nouveau-y += nouveau_vga.o nouveau_agp.o nouveau-y += nouveau_ttm.o nouveau_sgdma.o nouveau_bo.o nouveau_gem.o nouveau-y += nouveau_prime.o nouveau_abi16.o nouveau-y += nv04_fence.o nv10_fence.o nv17_fence.o -nouveau-y += nv50_fence.o nv84_fence.o nvc0_fence.o +nouveau-y += nv50_fence.o nv84_fence.o +nouveau-$(CONFIG_DRM_NOUVEAU_NVC0...

...ine/vp/nve0.o > +nouveau-$(CONFIG_DRM_NOUVEAU_NVC0) += core/engine/vp/nvc0.o > +nouveau-$(CONFIG_DRM_NOUVEAU_NVC0) += core/engine/vp/nve0.o > > # drm/core > nouveau-y += nouveau_drm.o nouveau_chan.o nouveau_dma.o nouveau_fence.o > @@ -299,12 +299,14 @@ nouveau-y += nouveau_vga.o nouveau_agp.o > nouveau-y += nouveau_ttm.o nouveau_sgdma.o nouveau_bo.o nouveau_gem.o > nouveau-y += nouveau_prime.o nouveau_abi16.o > nouveau-y += nv04_fence.o nv10_fence.o nv17_fence.o > -nouveau-y += nv50_fence.o nv84_fence.o nvc0_fence.o > +nouveau-y += nv50_fence.o nv84_fence.o > +nou...

...o +nouveau-$(CONFIG_DRM_NOUVEAU_NV50) += core/engine/vp/nv98.o +nouveau-$(CONFIG_DRM_NOUVEAU_NVC0) += core/engine/vp/nvc0.o +nouveau-$(CONFIG_DRM_NOUVEAU_NVC0) += core/engine/vp/nve0.o # drm/core nouveau-y += nouveau_drm.o nouveau_chan.o nouveau_dma.o nouveau_fence.o nouveau-y += nouveau_vga.o nouveau_agp.o nouveau-y += nouveau_ttm.o nouveau_sgdma.o nouveau_bo.o nouveau_gem.o nouveau-y += nouveau_prime.o nouveau_abi16.o -nouveau-y += nv04_fence.o nv10_fence.o nv17_fence.o -nouveau-y += nv50_fence.o nv84_fence.o nvc0_fence.o +nouveau-$(CONFIG_DRM_NOUVEAU_NV04) += nv04_fence.o nv10_fence.o nv17_fenc...

Overall idea: Detect lockups by watching for timeouts (vm flush / fence), return -EIOs, handle them at ioctl level, reset the GPU and repeat last ioctl. GPU reset is done by doing suspend / resume cycle with few tweaks: - CPU-only bo eviction - ignoring vm flush / fence timeouts - shortening waits Signed-off-by: Marcin Slusarz <marcin.slusarz at gmail.com> ---

Hello everyone, GK20A is the Kepler-based GPU used in the upcoming Tegra K1 chips. The following patches perform architectural changes to Nouveau that are necessary to support non-PCI GPUs and add initial support for GK20A. Although the support is still very basic and more user-space changes will be needed to make the full graphics stack run on top of it, we were able to successfully open