| Message ID | CADnq5_P8TOQ14NiWKgV-YUy0_8P+K842pZ4h7WuU6XBu4o9fyQ@mail.gmail.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
Hi! > >> >> >Any ideas? > >> >> > >> >> Alex probably knows more about this, but it sounds like problems with > >> >> switching the memory clocks on 3D load. > >> > > >> >> Try to disable power management completely with radeon.dpm=0 on the kernel > >> >> command line or nailing the hardware at a specific power level using > >> >> sysfs. > >> > > >> > I tried that, but it still flickers. > >> > >> It's probably pll stability. There seem to be a number of regressions > >> since the pll code was rewritten to support matching the hdmi clocks > >> more closely. Does this patch help? > >> > >> diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c > >> b/drivers/gpu/drm/radeon/atombios_crtc.c > >> index dac78ad..b86f06a 100644 > >> --- a/drivers/gpu/drm/radeon/atombios_crtc.c > >> +++ b/drivers/gpu/drm/radeon/atombios_crtc.c > >> @@ -569,6 +569,8 @@ static u32 atombios_adjust_pll(struct drm_crtc *crtc, > >> radeon_crtc->pll_flags = 0; > >> > >> if (ASIC_IS_AVIVO(rdev)) { > >> + radeon_crtc->pll_flags |= RADEON_PLL_PREFER_MINM_OVER_MAXP; > >> + > >> if ((rdev->family == CHIP_RS600) || > >> (rdev->family == CHIP_RS690) || > >> (rdev->family == CHIP_RS740)) > >> > > > > Help.. maybe... it is tricky to tell. It definitely does _not_ fix the > > issue completely. > > You could also try the old pll algorithm: I reverted the patch above, and switched to the old algorithm. The flicker is still there. (But maybe its less horrible, like with RADEON_PLL_PREFER_MINM_OVER_MAXP). Thanks, Pavel
Hi! > >> Unfortunately, it can't be applied as is because we had a similar > >> patch which was reverted because it regressed a bunch of other > >> systems. The actual pll limits probably need to be tweaked. > > > > Any ideas how to tweak the pll limits? > > Adjust the the algorithm in radeon_compute_pll_avivo() in radeon_display.c Hmm. Two values have diff = 0, I guess that leaves little room for improvement, as we already tried both with the PREFER_... setting. [ 1.236229] Linux agpgart interface v0.103 [ 1.236829] [drm] Initialized drm 1.1.0 20060810 [ 1.237013] [drm] radeon kernel modesetting enabled. [ 1.238284] [drm] initializing kernel modesetting (RV710 0x1002:0x9553 0x1025:0x0212) . [ 1.238362] [drm] register mmio base: 0xD6200000 [ 1.238417] [drm] register mmio size: 65536 [ 1.238622] ATOM BIOS: BR34582.001 [ 1.238789] radeon 0000:01:00.0: VRAM: 512M 0x0000000000000000 - 0x000000001FFFFFFF ( 512M used) [ 1.238856] radeon 0000:01:00.0: GTT: 1024M 0x0000000020000000 - 0x000000005FFFFFFF [ 1.238915] [drm] Detected VRAM RAM=512M, BAR=256M [ 1.238970] [drm] RAM width 64bits DDR [ 1.239266] [TTM] Zone kernel: Available graphics memory: 431276 kiB [ 1.239323] [TTM] Zone highmem: Available graphics memory: 1546602 kiB [ 1.239380] [TTM] Initializing pool allocator [ 1.240936] [TTM] Initializing DMA pool allocator [ 1.241174] [drm] radeon: 512M of VRAM memory ready [ 1.241231] [drm] radeon: 1024M of GTT memory ready. [ 1.241345] [drm] Loading RV710 Microcode [ 1.241483] radeon 0000:01:00.0: Direct firmware load for radeon/RV710_smc.bin failed with error -2 [ 1.241553] smc: error loading firmware "radeon/RV710_smc.bin" [ 1.241638] [drm] radeon: power management initialized [ 1.241754] radeon 0000:01:00.0: Direct firmware load for radeon/RV710_uvd.bin failed with error -2 [ 1.241823] radeon 0000:01:00.0: radeon_uvd: Can't load firmware "radeon/RV710_uvd.bi n" [ 1.241885] [drm] GART: num cpu pages 262144, num gpu pages 262144 [ 1.257273] [drm] PCIE GART of 1024M enabled (table at 0x0000000000040000). [ 1.257456] radeon 0000:01:00.0: WB enabled [ 1.257514] radeon 0000:01:00.0: fence driver on ring 0 use gpu addr 0x0000000020000c 00 and cpu addr 0xffc01c00 [ 1.257582] radeon 0000:01:00.0: fence driver on ring 3 use gpu addr 0x0000000020000c0c and cpu addr 0xffc01c0c [ 1.257655] [drm] Supports vblank timestamp caching Rev 2 (21.10.2013). [ 1.257713] [drm] Driver supports precise vblank timestamp query. [ 1.257770] radeon 0000:01:00.0: radeon: MSI limited to 32-bit [ 1.257921] [drm] radeon: irq initialized. [ 1.304343] [drm] ring test on 0 succeeded in 1 usecs [ 1.304403] [drm] ring test on 3 succeeded in 2 usecs [ 1.304835] [drm] ib test on ring 0 succeeded in 0 usecs [ 1.304912] [drm] ib test on ring 3 succeeded in 0 usecs [ 1.307453] [drm] Radeon Display Connectors [ 1.307511] [drm] Connector 0: [ 1.307565] [drm] VGA-1 [ 1.307619] [drm] DDC: 0x7e40 0x7e40 0x7e44 0x7e44 0x7e48 0x7e48 0x7e4c 0x7e4c [ 1.307677] [drm] Encoders: [ 1.307730] [drm] CRT1: INTERNAL_KLDSCP_DAC1 [ 1.336107] ACPI: Deprecated procfs I/F for battery is loaded, please retry with CONFIG_ACPI_PROCFS_POWER cleared [ 1.336209] ACPI: Battery Slot [BAT0] (battery absent) [ 1.342090] [drm] fb mappable at 0xC0241000 [ 1.342146] [drm] vram apper at 0xC0000000 [ 1.342201] [drm] size 8294400 [ 1.342254] [drm] fb depth is 24 [ 1.342307] [drm] pitch is 7680 [ 1.342777] fbcon: radeondrmfb (fb0) is primary device [ 1.344374] post_div = 5, diff = 270 [ 1.344375] post_div = 6, diff = 0 [ 1.344375] post_div = 7, diff = 192 [ 1.344376] post_div = 8, diff = 0 [ 1.344377] Selected post_div = 8 [ 1.344378] 148500 - 148500, pll dividers - fb: 88.0 ref: 2, post 8 [ 1.381561] Console: switching to colour frame buffer device 192x60 [ 1.391454] radeon 0000:01:00.0: fb0: radeondrmfb frame buffer device [ 1.404499] [drm] Initialized radeon 2.43.0 20080528 for 0000:01:00.0 on minor 0 [ 1.404669] [drm] amdgpu kernel modesetting enabled. [ 1.416219] loop: module loaded [ 1.418413] nbd: registered device at major 43 The "error loading firmware" messages confuse me a bit, but I do have some firmware built into kernel, and 3D acceleration seems to work. pavel@half:/data/l/linux$ grep FIRMWARE .config CONFIG_PREVENT_FIRMWARE_BUILD=y CONFIG_FIRMWARE_IN_KERNEL=y CONFIG_EXTRA_FIRMWARE="radeon/R700_rlc.bin" CONFIG_EXTRA_FIRMWARE_DIR="/lib/firmware" # CONFIG_CYPRESS_FIRMWARE is not set # CONFIG_DRM_LOAD_EDID_FIRMWARE is not set CONFIG_FIRMWARE_EDID=y CONFIG_FIRMWARE_MEMMAP=y # CONFIG_GOOGLE_FIRMWARE is not set # CONFIG_TEST_FIRMWARE is not set pavel@half:/data/l/linux$ Best regards, Pavel
On 04.11.2015 00:03, Pavel Machek wrote: > Hi! > > >>>>>>> Any ideas? >>>>>> Alex probably knows more about this, but it sounds like problems with >>>>>> switching the memory clocks on 3D load. >>>>>> Try to disable power management completely with radeon.dpm=0 on the kernel >>>>>> command line or nailing the hardware at a specific power level using >>>>>> sysfs. >>>>> I tried that, but it still flickers. >>>> It's probably pll stability. There seem to be a number of regressions >>>> since the pll code was rewritten to support matching the hdmi clocks >>>> more closely. Does this patch help? >>>> >>>> diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c >>>> b/drivers/gpu/drm/radeon/atombios_crtc.c >>>> index dac78ad..b86f06a 100644 >>>> --- a/drivers/gpu/drm/radeon/atombios_crtc.c >>>> +++ b/drivers/gpu/drm/radeon/atombios_crtc.c >>>> @@ -569,6 +569,8 @@ static u32 atombios_adjust_pll(struct drm_crtc *crtc, >>>> radeon_crtc->pll_flags = 0; >>>> >>>> if (ASIC_IS_AVIVO(rdev)) { >>>> + radeon_crtc->pll_flags |= RADEON_PLL_PREFER_MINM_OVER_MAXP; >>>> + >>>> if ((rdev->family == CHIP_RS600) || >>>> (rdev->family == CHIP_RS690) || >>>> (rdev->family == CHIP_RS740)) >>>> >>> Help.. maybe... it is tricky to tell. It definitely does _not_ fix the >>> issue completely. >> You could also try the old pll algorithm: > I reverted the patch above, and switched to the old algorithm. > > The flicker is still there. (But maybe its less horrible, like with > RADEON_PLL_PREFER_MINM_OVER_MAXP). The flickering would vanish completely if that's the reason for the issue you are seeing. Try setting ref_div_min and ref_div_max to 2 in radeon_compute_pll_avivo(). But I'm not 100% convinced that this is actually a PLL problem, try to compile the firmware it complains about into the kernel as well. Regards, Christian. > > Thanks, > Pavel >
diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c index dac78ad..8c6e8fa 100644 --- a/drivers/gpu/drm/radeon/atombios_crtc.c +++ b/drivers/gpu/drm/radeon/atombios_crtc.c @@ -1094,8 +1094,8 @@ static void atombios_crtc_set_pll(struct drm_crtc *crtc, struct drm_display_mode radeon_compute_pll_legacy(pll, radeon_crtc->adjusted_clock, &pll_clock, &fb_div, &frac_fb_div, &ref_div, &post_div); else if (ASIC_IS_AVIVO(rdev)) - radeon_compute_pll_avivo(pll, radeon_crtc->adjusted_clock, &pll_clock, - &fb_div, &frac_fb_div, &ref_div, &post_div);