[Nouveau] [Bug 91523] [NVE7] driver cannot initialize gpu(failed to parse ramcfg data)

https://bugs.freedesktop.org/show_bug.cgi?id=91523 --- Comment #4 from Ilia Bozhinov --- I tested a kernel build with the relevant code commented(right after the comment I posted) and indeed the kernel module works! (it loads and DRI_PRIME=1 allows me to run

[Nouveau] [Bug 94990] New: Latest 4.6rc4 kernel, no booting on gtx970

https://bugs.freedesktop.org/show_bug.cgi?id=94990 Bug ID: 94990 Summary: Latest 4.6rc4 kernel, no booting on gtx970 Product: xorg Version: git Hardware: Other OS: All Status: NEW Severity: normal

[Nouveau] [Bug 94990] Latest 4.6rc4 kernel, no booting on gtx970

https://bugs.freedesktop.org/show_bug.cgi?id=94990 --- Comment #1 from Ilia Mirkin --- Does it work fine on a cold boot, but only fails on a warm boot? (Is that what you mean by "after rebooting"?) -- You are receiving this mail because: You are the assignee for the

[Nouveau] [PATCH v4 03/37] volt: save the voltage range we are able to set

We shouldn't set voltages below the min or above the max voltage the gpu is able to set, so save the range. Signed-off-by: Karol Herbst Reviewed-by: Martin Peres Tested-by: Pierre Moreau ---

[Nouveau] [PATCH v4 04/37] volt: add nvkm_volt_map_min function

this is a copy of nvkm_volt_map, which always returns the lowest possible voltage for a cstate. nvkm_volt_map will get a temperature parameter there later and also fix the voltage calculation, so that this functions will be completly different later. Signed-off-by: Karol Herbst

[Nouveau] [PATCH v4 12/37] clk: fixup cstate selection

now the cstatei parameter can be used of the nvkm_cstate_prog function to select a specific cstate. -1 is a magic value, which will always select the highest currently possible cstate Signed-off-by: Karol Herbst --- drm/nouveau/nvkm/subdev/clk/base.c | 12 ++--

[Nouveau] [PATCH v4 34/37] mc: fix NULL pointer access in libnouveau

Signed-off-by: Karol Herbst --- drm/nouveau/nvkm/subdev/mc/base.c | 7 ++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/drm/nouveau/nvkm/subdev/mc/base.c b/drm/nouveau/nvkm/subdev/mc/base.c index aa394af..88bc1cc 100644 ---

[Nouveau] [PATCH v4 14/37] bios: add parsing of BASE CLOCK table

this table contains three important clocks: base clock: this is the non boosted max clock tdp clock: the clock at wich the vbios guarentees the TDP won't ever be exceeded at max load (seems to be always the same as the base clock, but behaves differently). boost

[Nouveau] [PATCH v4 36/37] WIP volt/gk104: readout speedo

this gk104 volt implementation has to be reworked a little, because the speedo readout in maxwell doesn't need those strange 0 and 41 writes into 0x122634, but it needs this PWM thing. Maybe Maxwell is PWM only and we could just simplify it there, but without proper knowledge there has some

[Nouveau] [PATCH v4 19/37] volt: add gf100 subdev with speedo

Signed-off-by: Karol Herbst --- drm/nouveau/include/nvkm/subdev/volt.h | 1 + drm/nouveau/nvkm/engine/device/base.c | 17 +- drm/nouveau/nvkm/subdev/volt/Kbuild| 1 + drm/nouveau/nvkm/subdev/volt/gf100.c | 59 ++ 4 files

[Nouveau] [PATCH v4 37/37] volt: add NvVoltOffsetmV option

This option can be used to adjust the calculated voltage or the cstate voltage calculation Signed-off-by: Karol Herbst --- bin/nv_cmp_volt.c | 2 +- drm/nouveau/include/nvkm/subdev/volt.h | 4 +++- drm/nouveau/nvkm/subdev/clk/base.c | 8

[Nouveau] [PATCH v4 27/37] clk: make pstate a pointer to nvkm_pstate

we will access the current set cstate at least every second and this safes us some CPU cycles looking them up every second. Signed-off-by: Karol Herbst --- drm/nouveau/include/nvkm/subdev/clk.h | 2 +- drm/nouveau/nvkm/engine/device/ctrl.c | 5 -

[Nouveau] [PATCH v4 17/37] bios/vmap: unk0 field is the mode

this selects which formula is used to calculate the voltage. depending on the value, the entry maps to a different voltage and even enables if the temperature has any effect or not. This is easy to observe with the binary driver. Signed-off-by: Karol Herbst ---

[Nouveau] [PATCH v4 00/37] Volting/Clocking improvements for Fermi and newer

We are slowly getting there! v4 of the series with some realy good improvements, so I am sure this is like 95% done and only needs some proper polishing and proper Reviews! I also added the NvVoltOffsetmV module parameter, so that a user is able to over and !under!-volt the GPU. Overvolting

[Nouveau] [PATCH v4 18/37] volt: add speedo

Signed-off-by: Karol Herbst --- bin/nv_cmp_volt.c | 2 +- drm/nouveau/include/nvkm/subdev/volt.h | 2 ++ drm/nouveau/nvkm/subdev/volt/base.c| 12 drm/nouveau/nvkm/subdev/volt/priv.h| 1 + 4 files changed, 16 insertions(+), 1

[Nouveau] [PATCH v4 07/37] volt: add min_id parameter to nvkm_volt_set_id

Each pstate has its own voltage map entry like each cstate has. The voltages of those entries act as a floor value for the currently selected pstate and nvidia never sets a voltage below them. Signed-off-by: Karol Herbst Reviewed-by: Martin Peres

[Nouveau] [PATCH v4 32/37] clk: only do partial reclocks as required

we don't want to reclock to the same pstate or cstate over and over again, so only do things we actually have to do. v4: move into gf100 Signed-off-by: Karol Herbst --- drm/nouveau/nvkm/subdev/clk/base.c | 11 +-- drm/nouveau/nvkm/subdev/clk/gf100.c | 62

[Nouveau] [PATCH v4 10/37] add daemon to compare nouveau with blob voltage

this tool can be run alongside the nvidia driver to print information about the current p/cstate, which voltage was set by nvidia and what nouveau would set in the same situation. v4: parse default options Signed-off-by: Karol Herbst --- bin/nv_cmp_volt.c | 139

[Nouveau] [PATCH v4 31/37] clk: split out update code to nv40

this code will change for gf100 and newer Signed-off-by: Karol Herbst --- drm/nouveau/nvkm/subdev/clk/base.c | 14 ++ drm/nouveau/nvkm/subdev/clk/g84.c | 1 + drm/nouveau/nvkm/subdev/clk/gf100.c | 1 + drm/nouveau/nvkm/subdev/clk/gk104.c | 1 +

[Nouveau] [PATCH v4 15/37] clk: allow boosting only when NvBoost is set

0: base clock from the vbios is max clock 1: boost only to boost clock from the vbios (default) 2: boost to max clock available v2: moved into nvkm_cstate_valid v4: check the existence of the clocks before limiting Signed-off-by: Karol Herbst Reviewed-by: Martin Peres

[Nouveau] [PATCH v4 28/37] clk: hold information about the current cstate status

Signed-off-by: Karol Herbst --- drm/nouveau/include/nvkm/subdev/clk.h | 5 + drm/nouveau/nvkm/subdev/clk/base.c| 32 +--- 2 files changed, 30 insertions(+), 7 deletions(-) diff --git a/drm/nouveau/include/nvkm/subdev/clk.h

[Nouveau] [PATCH v4 08/37] clk: export nvkm_volt_map

before clocking to a cstate, we have to check if the voltage is within the allowed range. Signed-off-by: Karol Herbst --- drm/nouveau/include/nvkm/subdev/volt.h | 1 + drm/nouveau/nvkm/subdev/volt/base.c| 2 +- 2 files changed, 2 insertions(+), 1 deletion(-) diff

[Nouveau] [PATCH v4 20/37] volt: add coefficients

I am sure that those are a bit different, but while testing the biggest error compared to nvidia was -1.5%. Without this change we are most of the time around 10% below nvidias voltage, so this change causes no harm and improves the situation a lot already. These coefficients were REed by

[Nouveau] [PATCH v4 25/37] clk: remove dstate and tstate

we won't need them, because we will adjust the clocks depending on engine loads later on anyway. It also simplifies the clocking logic. Signed-off-by: Karol Herbst --- drm/nouveau/include/nvkm/subdev/clk.h | 4 drm/nouveau/nvkm/subdev/clk/base.c| 28

[Nouveau] [PATCH v4 13/37] clk: respect voltage limits in nvkm_cstate_prog

we should never allow to select a cstate which current voltage (depending on the temperature) is higher than 1. the max volt entries in the voltage map table 2. what tha gpu actually can volt to. this resolves all remaining volting errors on fermi and newer. v3: use find_best for all cstates

[Nouveau] [PATCH v4 26/37] therm: don't cancel the timer

we will need a always running therm daemon to adjust the voltage/clocks on the fly. Signed-off-by: Karol Herbst --- drm/nouveau/nvkm/subdev/therm/base.c | 9 ++--- 1 file changed, 2 insertions(+), 7 deletions(-) diff --git a/drm/nouveau/nvkm/subdev/therm/base.c

[Nouveau] [PATCH v4 09/37] clk: add index field to nvkm_cstate

Signed-off-by: Karol Herbst --- drm/nouveau/include/nvkm/subdev/clk.h | 1 + drm/nouveau/nvkm/subdev/clk/base.c| 1 + 2 files changed, 2 insertions(+) diff --git a/drm/nouveau/include/nvkm/subdev/clk.h b/drm/nouveau/include/nvkm/subdev/clk.h index fb54417..6226f0d

[Nouveau] [PATCH v4 22/37] clk: rename nvkm_pstate_calc to nvkm_clk_update

this function will be used to update the current clock state. This will happen for various reasons: * temperature changes (may change cstate and/or voltage) * user changes boost mode * load changes v2: add wait parameter Signed-off-by: Karol Herbst ---

[Nouveau] [PATCH v4 06/37] volt: parse the max voltage map entries

There are at least three "max" entries, which specify the max voltage. Because they are actually normal voltage map entries, they can also be affected by the temperature. Nvidia respects those entries and if they get changed, nvidia uses the lower voltage from both. We shouldn't exceed those

[Nouveau] [PATCH v4 02/37] volt: properly detect entry based voltage tables

there is a field in the voltage table which tells us if the VIDs are taken from the entries or calculated through the header. v2: don't break older versions Signed-off-by: Karol Herbst Reviewed-by: Martin Peres Tested-by: Pierre Moreau

[Nouveau] [PATCH v4 05/37] clk: don't create cstates whit voltages higher than what the gpu can do

Signed-off-by: Karol Herbst Reviewed-by: Martin Peres Tested-by: Pierre Moreau --- drm/nouveau/nvkm/subdev/clk/base.c | 4 1 file changed, 4 insertions(+) diff --git a/drm/nouveau/nvkm/subdev/clk/base.c

[Nouveau] [PATCH v4 11/37] volt: add temperature parameter to nvkm_volt_map

the voltage entries actually may map to a different voltage depending on the current temperature. v2: only read the temperatue when actually needed Signed-off-by: Karol Herbst --- bin/nv_cmp_volt.c | 2 +- drm/nouveau/include/nvkm/subdev/volt.h |

[Nouveau] [PATCH v4 23/37] nvif: add boost info and set operations

Signed-off-by: Karol Herbst Reviewed-by: Martin Peres --- drm/nouveau/include/nvif/if0001.h | 15 ++ drm/nouveau/nvkm/engine/device/ctrl.c | 55 +++ 2 files changed, 70 insertions(+) diff --git

[Nouveau] [PATCH v4 21/37] clk: save the max clock we can set

Signed-off-by: Karol Herbst --- drm/nouveau/include/nvkm/subdev/clk.h | 1 + drm/nouveau/nvkm/subdev/clk/base.c| 2 ++ 2 files changed, 3 insertions(+) diff --git a/drm/nouveau/include/nvkm/subdev/clk.h b/drm/nouveau/include/nvkm/subdev/clk.h index 99ee05c..61d99fd

[Nouveau] [PATCH v4 24/37] debugfs: add boost interface to change the boost_mode

Signed-off-by: Karol Herbst Reviewed-by: Martin Peres --- drm/nouveau/nouveau_debugfs.c | 76 +++ 1 file changed, 76 insertions(+) diff --git a/drm/nouveau/nouveau_debugfs.c b/drm/nouveau/nouveau_debugfs.c

[Nouveau] [PATCH v4 33/37] therm: trigger reclock in temperature daemon

depending on the temperature, cstates might become unreachable or the maped voltage of a cstate changes. We want to adjust to that. Signed-off-by: Karol Herbst --- drm/nouveau/nvkm/subdev/therm/base.c | 5 + 1 file changed, 5 insertions(+) diff --git

[Nouveau] [PATCH v4 29/37] clk: we should pass the pstate id around not the index in the list

this makes the code easier, because we can compare the id with pstate->pstate and safe us the trouble iterating over the entire pstate list Signed-off-by: Karol Herbst --- drm/nouveau/nouveau_debugfs.c | 6 ++--- drm/nouveau/nvkm/subdev/clk/base.c | 49

[Nouveau] [Bug 92126] [NVE4] GTX 760 Auto fan speed stays at 10% until temp hits 90C

https://bugs.freedesktop.org/show_bug.cgi?id=92126 --- Comment #6 from daemon32 --- (In reply to Karol Herbst from comment #5) > Is this bug fixed in recent kernels? (4.5 I think?) Well, sort of, the fan spins up, but it takes off like a jet engine at a mere 50° C. (I'm

Re: [Nouveau] GM108GLM?

Is there anything I can do in this regard? /smo On 2016-03-31 23:12, Pierre Moreau wrote: > Oh, hum, GM108 is NV118 and not NV108 which is Gk208… My bad! > > SMF from the bug report seemed to have some working setup, and since he didn't > pinged back, I guess it's working nicely. But some more