dri-devel June 2010

dri-devel@lists.freedesktop.org

71 participants
237 discussions

[PATCH] drm/radeon/kms: r600 dump last 64 dwords of ring.
by Jerome Glisse 04 Jun '10

04 Jun '10

Instead of dumping unprocessed dwords, dump the last 64 dwords of the ring. This make debugging of some case easier. Signed-off-by: Jerome Glisse <jglisse(a)redhat.com> --- drivers/gpu/drm/radeon/r600.c | 6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c index f68cc92..1537079 100644 --- a/drivers/gpu/drm/radeon/r600.c +++ b/drivers/gpu/drm/radeon/r600.c @@ -3371,9 +3371,9 @@ static int r600_debugfs_cp_ring_info(struct seq_file *m, void *data) seq_printf(m, "driver's copy of the CP_RB_RPTR 0x%08x\n", rdev->cp.rptr); seq_printf(m, "%u free dwords in ring\n", rdev->cp.ring_free_dw); seq_printf(m, "%u dwords in ring\n", count); - i = rdev->cp.rptr; - for (j = 0; j <= count; j++) { - seq_printf(m, "r[%04d]=0x%08x\n", i, rdev->cp.ring[i]); + i = (rdev->cp.wptr + (rdev->cp.ring_size / 4) - 64) & rdev->cp.ptr_mask; + for (j = 0; j <= 64; j++) { + seq_printf(m, "r[%04d]=0x%08X\n", i, rdev->cp.ring[i]); i = (i + 1) & rdev->cp.ptr_mask; } return 0; -- 1.7.0.1

3 3

[Bug 28383] New: Cloned screens with different res/refresh cause problems with mesa demos since new dri2 vsync.
by bugzilla-daemon＠freedesktop.org 04 Jun '10

04 Jun '10

https://bugs.freedesktop.org/show_bug.cgi?id=28383 Summary: Cloned screens with different res/refresh cause problems with mesa demos since new dri2 vsync. Product: Mesa Version: git Platform: x86 (IA32) OS/Version: Linux (All) Status: NEW Severity: normal Priority: medium Component: Drivers/DRI/R600 AssignedTo: dri-devel(a)lists.freedesktop.org ReportedBy: lists(a)andyfurniss.entadsl.com I think this is linked to the new dri2 vsync but mat be wrong... rv670 mesa git, ddx git xorg git (2 weeks ago) drt kernels. Not just latest drt and unrelated to tiling patches. I am starting with a TV connected @ 1024x768 + vga monitor @ 1920x1080 so the top left section of the monitor screen gets tv(50Hz) vsync - monitor is 60Hz. If I start a mesa demo and it starts in the TV area it will sync to 50HZ if I move it outside the top left area it will sync to 60Hz. The problem is if I then move it back to the TV area it will stop rendering and become unresponsive. I can <ctrl><c> it but the window will remain. If I the move the window around enough I can get xserver to segfault. This doesn't happen with UMS or a drt from 29th April (uses old vsync). It doesn't happen if I disable TV with xrandr. [ 489.682] 0: /home/andy/Src/Xorg-git/modular/bin/Xorg (xorg_backtrace+0x3b) [0x80da21b] [ 489.682] 1: /home/andy/Src/Xorg-git/modular/bin/Xorg (0x8048000+0x59ca5) [0x80a1ca5] [ 489.682] 2: (vdso) (__kernel_rt_sigreturn+0x0) [0xffffe40c] [ 489.682] 3: /home/andy/Src/Xorg-git/modular/bin/Xorg (LocalClient+0x41) [0x809d8d1] [ 489.682] 4: /home/andy/Src/Xorg-git/modular/lib/xorg/modules/extensions/libdri2.so (0xb73ce000+0x309b) [0xb73d109b] [ 489.682] 5: /home/andy/Src/Xorg-git/modular/bin/Xorg (0x8048000+0x25198) [0x806d198] [ 489.682] 6: /home/andy/Src/Xorg-git/modular/bin/Xorg (0x8048000+0x1a86a) [0x806286a] [ 489.682] 7: /lib/libc.so.6 (__libc_start_main+0xd0) [0xb7495380] [ 489.682] 8: /home/andy/Src/Xorg-git/modular/bin/Xorg (0x8048000+0x1a481) [0x8062481] [ 489.682] Segmentation fault at address 0x18 -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug.

1 1

[PATCH] drm/radeon/kms/evergreen: set accel_enabled
by Alex Deucher 04 Jun '10

04 Jun '10

This is needed to enable accel in the ddx. However, due to a bug in older versions of the ddx, it relies on accel being disabled in order to load properly on evergreen chips. To maintain compatility, we add a new get accel param and call that from the ddx. The old one always returns false for evergreen cards. Signed-off-by: Alex Deucher <alexdeucher(a)gmail.com> --- drivers/gpu/drm/radeon/evergreen.c | 2 +- drivers/gpu/drm/radeon/radeon_drv.c | 3 ++- drivers/gpu/drm/radeon/radeon_kms.c | 9 ++++++++- include/drm/radeon_drm.h | 1 + 4 files changed, 12 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c index 0440c09..49c94ae 100644 --- a/drivers/gpu/drm/radeon/evergreen.c +++ b/drivers/gpu/drm/radeon/evergreen.c @@ -2153,7 +2153,7 @@ int evergreen_init(struct radeon_device *rdev) if (r) return r; - rdev->accel_working = false; + rdev->accel_working = true; r = evergreen_startup(rdev); if (r) { dev_err(rdev->dev, "disabling GPU acceleration\n"); diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c index 902d173..e166fe4 100644 --- a/drivers/gpu/drm/radeon/radeon_drv.c +++ b/drivers/gpu/drm/radeon/radeon_drv.c @@ -45,9 +45,10 @@ * - 2.2.0 - add r6xx/r7xx const buffer support * - 2.3.0 - add MSPOS + 3D texture + r500 VAP regs * - 2.4.0 - add crtc id query + * - 2.5.0 - add get accel 2 to work around ddx breakage for evergreen */ #define KMS_DRIVER_MAJOR 2 -#define KMS_DRIVER_MINOR 4 +#define KMS_DRIVER_MINOR 5 #define KMS_DRIVER_PATCHLEVEL 0 int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags); int radeon_driver_unload_kms(struct drm_device *dev); diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c index 0406835..6a70c0d 100644 --- a/drivers/gpu/drm/radeon/radeon_kms.c +++ b/drivers/gpu/drm/radeon/radeon_kms.c @@ -118,7 +118,11 @@ int radeon_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) value = rdev->num_z_pipes; break; case RADEON_INFO_ACCEL_WORKING: - value = rdev->accel_working; + /* xf86-video-ati 6.13.0 relies on this being false for evergreen */ + if ((rdev->family >= CHIP_CEDAR) && (rdev->family <= CHIP_HEMLOCK)) + value = false; + else + value = rdev->accel_working; break; case RADEON_INFO_CRTC_FROM_ID: for (i = 0, found = 0; i < rdev->num_crtc; i++) { @@ -134,6 +138,9 @@ int radeon_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp) return -EINVAL; } break; + case RADEON_INFO_ACCEL_WORKING2: + value = rdev->accel_working; + break; default: DRM_DEBUG("Invalid request %d\n", info->request); return -EINVAL; diff --git a/include/drm/radeon_drm.h b/include/drm/radeon_drm.h index 3ff9fc0..5347063 100644 --- a/include/drm/radeon_drm.h +++ b/include/drm/radeon_drm.h @@ -903,6 +903,7 @@ struct drm_radeon_cs { #define RADEON_INFO_NUM_Z_PIPES 0x02 #define RADEON_INFO_ACCEL_WORKING 0x03 #define RADEON_INFO_CRTC_FROM_ID 0x04 +#define RADEON_INFO_ACCEL_WORKING2 0x05 struct drm_radeon_info { uint32_t request; -- 1.7.0.1

2 2

[PATCH] drm/radeon/kms: r600 reset GPU at module load time v2
by Jerome Glisse 04 Jun '10

04 Jun '10

When loading/unloading several time the driver on r600 we end up with the GPU in broken state and loading fail to initialize acceleration. Reset the GPU at load time so acceleration keep working over load/unload. v2 Move status print after reset in asic_reset to avoid cluttering log with this at module load time. Signed-off-by: Jerome Glisse <jglisse(a)redhat.com> --- drivers/gpu/drm/radeon/r600.c | 37 +++++++++++++++++++++++-------------- 1 files changed, 23 insertions(+), 14 deletions(-) diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c index 1537079..b932f48 100644 --- a/drivers/gpu/drm/radeon/r600.c +++ b/drivers/gpu/drm/radeon/r600.c @@ -1163,13 +1163,6 @@ int r600_gpu_soft_reset(struct radeon_device *rdev) S_008014_CB2_BUSY(1) | S_008014_CB3_BUSY(1); u32 tmp; - dev_info(rdev->dev, "GPU softreset \n"); - dev_info(rdev->dev, " R_008010_GRBM_STATUS=0x%08X\n", - RREG32(R_008010_GRBM_STATUS)); - dev_info(rdev->dev, " R_008014_GRBM_STATUS2=0x%08X\n", - RREG32(R_008014_GRBM_STATUS2)); - dev_info(rdev->dev, " R_000E50_SRBM_STATUS=0x%08X\n", - RREG32(R_000E50_SRBM_STATUS)); rv515_mc_stop(rdev, &save); if (r600_mc_wait_for_idle(rdev)) { dev_warn(rdev->dev, "Wait for MC idle timedout !\n"); @@ -1207,12 +1200,6 @@ int r600_gpu_soft_reset(struct radeon_device *rdev) WREG32(R_008020_GRBM_SOFT_RESET, 0); /* Wait a little for things to settle down */ mdelay(1); - dev_info(rdev->dev, " R_008010_GRBM_STATUS=0x%08X\n", - RREG32(R_008010_GRBM_STATUS)); - dev_info(rdev->dev, " R_008014_GRBM_STATUS2=0x%08X\n", - RREG32(R_008014_GRBM_STATUS2)); - dev_info(rdev->dev, " R_000E50_SRBM_STATUS=0x%08X\n", - RREG32(R_000E50_SRBM_STATUS)); rv515_mc_resume(rdev, &save); return 0; } @@ -1245,7 +1232,23 @@ bool r600_gpu_is_lockup(struct radeon_device *rdev) int r600_asic_reset(struct radeon_device *rdev) { - return r600_gpu_soft_reset(rdev); + int r; + + dev_info(rdev->dev, "GPU softreset \n"); + dev_info(rdev->dev, " R_008010_GRBM_STATUS=0x%08X\n", + RREG32(R_008010_GRBM_STATUS)); + dev_info(rdev->dev, " R_008014_GRBM_STATUS2=0x%08X\n", + RREG32(R_008014_GRBM_STATUS2)); + dev_info(rdev->dev, " R_000E50_SRBM_STATUS=0x%08X\n", + RREG32(R_000E50_SRBM_STATUS)); + r = r600_gpu_soft_reset(rdev); + dev_info(rdev->dev, " R_008010_GRBM_STATUS=0x%08X\n", + RREG32(R_008010_GRBM_STATUS)); + dev_info(rdev->dev, " R_008014_GRBM_STATUS2=0x%08X\n", + RREG32(R_008014_GRBM_STATUS2)); + dev_info(rdev->dev, " R_000E50_SRBM_STATUS=0x%08X\n", + RREG32(R_000E50_SRBM_STATUS)); + return r; } static u32 r600_get_tile_pipe_to_backend_map(u32 num_tile_pipes, @@ -2460,6 +2463,12 @@ int r600_init(struct radeon_device *rdev) DRM_INFO("GPU not posted. posting now...\n"); atom_asic_init(rdev->mode_info.atom_context); } + + /* GPU is sometimes in broken state (especialy after loading/unloading + * the driver several time. Reset the GPU. + */ + r600_gpu_soft_reset(rdev); + /* Initialize scratch registers */ r600_scratch_init(rdev); /* Initialize surface registers */ -- 1.7.0.1

1 0

[PATCH] drm/radeon/kms: r600 reset GPU at module load time
by Jerome Glisse 04 Jun '10

04 Jun '10

When loading/unloading several time the driver on r600 we end up with the GPU in broken state and loading fail to initialize acceleration. Reset the GPU at load time so acceleration keep working over load/unload. Signed-off-by: Jerome Glisse <jglisse(a)redhat.com> --- drivers/gpu/drm/radeon/r600.c | 20 +++++++++++++------- 1 files changed, 13 insertions(+), 7 deletions(-) diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c index 1537079..5fc0525 100644 --- a/drivers/gpu/drm/radeon/r600.c +++ b/drivers/gpu/drm/radeon/r600.c @@ -1163,13 +1163,6 @@ int r600_gpu_soft_reset(struct radeon_device *rdev) S_008014_CB2_BUSY(1) | S_008014_CB3_BUSY(1); u32 tmp; - dev_info(rdev->dev, "GPU softreset \n"); - dev_info(rdev->dev, " R_008010_GRBM_STATUS=0x%08X\n", - RREG32(R_008010_GRBM_STATUS)); - dev_info(rdev->dev, " R_008014_GRBM_STATUS2=0x%08X\n", - RREG32(R_008014_GRBM_STATUS2)); - dev_info(rdev->dev, " R_000E50_SRBM_STATUS=0x%08X\n", - RREG32(R_000E50_SRBM_STATUS)); rv515_mc_stop(rdev, &save); if (r600_mc_wait_for_idle(rdev)) { dev_warn(rdev->dev, "Wait for MC idle timedout !\n"); @@ -1245,6 +1238,13 @@ bool r600_gpu_is_lockup(struct radeon_device *rdev) int r600_asic_reset(struct radeon_device *rdev) { + dev_info(rdev->dev, "GPU softreset \n"); + dev_info(rdev->dev, " R_008010_GRBM_STATUS=0x%08X\n", + RREG32(R_008010_GRBM_STATUS)); + dev_info(rdev->dev, " R_008014_GRBM_STATUS2=0x%08X\n", + RREG32(R_008014_GRBM_STATUS2)); + dev_info(rdev->dev, " R_000E50_SRBM_STATUS=0x%08X\n", + RREG32(R_000E50_SRBM_STATUS)); return r600_gpu_soft_reset(rdev); } @@ -2460,6 +2460,12 @@ int r600_init(struct radeon_device *rdev) DRM_INFO("GPU not posted. posting now...\n"); atom_asic_init(rdev->mode_info.atom_context); } + + /* GPU is sometimes in broken state (especialy after loading/unloading + * the driver several time. Reset the GPU. + */ + r600_gpu_soft_reset(rdev); + /* Initialize scratch registers */ r600_scratch_init(rdev); /* Initialize surface registers */ -- 1.7.0.1

1 0

[patch] drm/vmwgfx: return -EFAULT for copy_to_user errors
by Dan Carpenter 04 Jun '10

04 Jun '10

copy_to/from_user() returns the number of bytes remaining to be copied but we want to return a negative error code here. This gets returned to userspace. Signed-off-by: Dan Carpenter <error27(a)gmail.com> diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c index f8fbbc6..8612378 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_resource.c @@ -597,8 +597,10 @@ int vmw_surface_define_ioctl(struct drm_device *dev, void *data, ret = copy_from_user(srf->sizes, user_sizes, srf->num_sizes * sizeof(*srf->sizes)); - if (unlikely(ret != 0)) + if (unlikely(ret != 0)) { + ret = -EFAULT; goto out_err1; + } if (srf->scanout && srf->num_sizes == 1 && @@ -697,9 +699,11 @@ int vmw_surface_reference_ioctl(struct drm_device *dev, void *data, if (user_sizes) ret = copy_to_user(user_sizes, srf->sizes, srf->num_sizes * sizeof(*srf->sizes)); - if (unlikely(ret != 0)) + if (unlikely(ret != 0)) { DRM_ERROR("copy_to_user failed %p %u\n", user_sizes, srf->num_sizes); + ret = -EFAULT; + } out_bad_resource: out_no_reference: ttm_base_object_unref(&base); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c index dbd36b8..ba15038 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c @@ -644,6 +644,7 @@ int vmw_execbuf_ioctl(struct drm_device *dev, void *data, ret = copy_from_user(cmd, user_cmd, arg->command_size); if (unlikely(ret != 0)) { + ret = -EFAULT; DRM_ERROR("Failed copying commands.\n"); goto out_commit; }

1 0

[patch] drm/drm_crtc: return -EFAULT on copy_to_user errors
by Dan Carpenter 04 Jun '10

04 Jun '10

copy_from_user() returns the number of bytes left to be copied but we want to return a negative error code here. This is in the ioctl handler so the error code get returned to userspace. Signed-off-by: Dan Carpenter <error27(a)gmail.com> diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c index 994d23b..57cea01 100644 --- a/drivers/gpu/drm/drm_crtc.c +++ b/drivers/gpu/drm/drm_crtc.c @@ -1840,8 +1840,10 @@ int drm_mode_dirtyfb_ioctl(struct drm_device *dev, ret = copy_from_user(clips, clips_ptr, num_clips * sizeof(*clips)); - if (ret) + if (ret) { + ret = -EFAULT; goto out_err2; + } } if (fb->funcs->dirty) {

1 0

[PATCH] drm/radeon/kms/pm: patch default power state with default clocks/voltages on r6xx+
by Alex Deucher 04 Jun '10

04 Jun '10

The default power state does not always match the default clocks and voltage for a particular card. The information in the firmware info table is correct and should be used in preference to the info the default power state. Signed-off-by: Alex Deucher <alexdeucher(a)gmail.com> --- drivers/gpu/drm/radeon/radeon_atombios.c | 25 ++++++++++++++++++++++++- 1 files changed, 24 insertions(+), 1 deletions(-) diff --git a/drivers/gpu/drm/radeon/radeon_atombios.c b/drivers/gpu/drm/radeon/radeon_atombios.c index f858be0..ebfe8fd 100644 --- a/drivers/gpu/drm/radeon/radeon_atombios.c +++ b/drivers/gpu/drm/radeon/radeon_atombios.c @@ -1746,9 +1746,22 @@ void radeon_atombios_get_power_modes(struct radeon_device *rdev) rdev->pm.power_state[state_index].misc2 = 0; } } else { + int fw_index = GetIndexIntoMasterTable(DATA, FirmwareInfo); + uint8_t fw_frev, fw_crev; + uint16_t fw_data_offset, vddc = 0; + union firmware_info *firmware_info; + ATOM_PPLIB_THERMALCONTROLLER *controller = &power_info->info_4.sThermalController; + + if (atom_parse_data_header(mode_info->atom_context, fw_index, NULL, + &fw_frev, &fw_crev, &fw_data_offset)) { + firmware_info = + (union firmware_info *)(mode_info->atom_context->bios + + fw_data_offset); + vddc = firmware_info->info_14.usBootUpVDDCVoltage; + } + /* add the i2c bus for thermal/fan chip */ /* no support for internal controller yet */ - ATOM_PPLIB_THERMALCONTROLLER *controller = &power_info->info_4.sThermalController; if (controller->ucType > 0) { if ((controller->ucType == ATOM_PP_THERMALCONTROLLER_RV6xx) || (controller->ucType == ATOM_PP_THERMALCONTROLLER_RV770) || @@ -1895,6 +1908,16 @@ void radeon_atombios_get_power_modes(struct radeon_device *rdev) rdev->pm.default_power_state_index = state_index; rdev->pm.power_state[state_index].default_clock_mode = &rdev->pm.power_state[state_index].clock_info[mode_index - 1]; + /* patch the table values with the default slck/mclk from firmware info */ + for (j = 0; j < mode_index; j++) { + rdev->pm.power_state[state_index].clock_info[j].mclk = + rdev->clock.default_mclk; + rdev->pm.power_state[state_index].clock_info[j].sclk = + rdev->clock.default_sclk; + if (vddc) + rdev->pm.power_state[state_index].clock_info[j].voltage.voltage = + vddc; + } } state_index++; } -- 1.5.6.3

2 9

[Bug 28363] New: [r600c, kms, tiling] shadows wrong in doom3
by bugzilla-daemon＠freedesktop.org 03 Jun '10

03 Jun '10

https://bugs.freedesktop.org/show_bug.cgi?id=28363 Summary: [r600c, kms, tiling] shadows wrong in doom3 Product: DRI Version: XOrg CVS Platform: x86-64 (AMD64) OS/Version: Linux (All) Status: NEW Severity: normal Priority: medium Component: DRM/Radeon AssignedTo: dri-devel(a)lists.freedesktop.org ReportedBy: liquid.acid(a)gmx.net Hi there, I'm using the most recent drm-radeon-testing kernel which incorporates the r6xx+ tiling patchset. Both the ddx and mesa are patched so that tiling is used (ColorTiling=true in xorg.conf). Both ddx and mesa are also git master (plus the patches, like stated before). Just tried the doom3 demo and noticed that the shadows are not rendered correctly, probably some issue with stencil buffering. Should be reproducable with this version of the demo: doom3-linux-1.3.1.1304.x86.run Played around a bit with the ingame graphics settings, but ramping the quality up and down doesn't have any effect on the issue (when not completly disabling the shadows). Looks like it always uses the same approach, I think it was called Carmack's reverse. Performance is quite good and I don't see any other obvious rendering issues, so fixing the stencil stuff would make d3 a 'perfectly playable' game on the OSS driver :) Greets, Tobias -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are the assignee for the bug.

1 8

[PATCH] drm/radeon/kms: Allow interruptible wait when testing IB
by Jerome Glisse 03 Jun '10

03 Jun '10

On single CPU waiting without allowing interruption will lead to infinite wait. Signed-off-by: Jerome Glisse <jglisse(a)redhat.com> --- drivers/gpu/drm/radeon/r600.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c index f68cc92..6db9d11 100644 --- a/drivers/gpu/drm/radeon/r600.c +++ b/drivers/gpu/drm/radeon/r600.c @@ -2607,7 +2607,7 @@ int r600_ib_test(struct radeon_device *rdev) DRM_ERROR("radeon: failed to schedule ib (%d).\n", r); return r; } - r = radeon_fence_wait(ib->fence, false); + r = radeon_fence_wait(ib->fence, true); if (r) { DRM_ERROR("radeon: fence wait failed (%d).\n", r); return r; -- 1.7.0.1

1 0

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

dri-devel June 2010