Re: [Nouveau] [PATCH v2 1/2] drm/nouveau: Fix ida leak

2021-05-12 Thread Pierre Moreau
Why remove the const modifier on `nb`? On 2021-05-11 — 16:28, Zhen Lei wrote: > When the 'nb' value allocated from 'bl_ida' is greater than or equal to > 100, it will not be released. In fact, we can simplify operations by > limiting the range of idas that can be applied for. > > By the way,

Re: [Nouveau] [PATCH v2 2/2] drm/nouveau: Fix error return code in nouveau_backlight_init()

2021-05-12 Thread Pierre Moreau
Reviewed-by: Pierre Moreau On 2021-05-11 — 16:28, Zhen Lei wrote: > Fix to return a negative error code from the error handling case instead > of 0, as done elsewhere in this function. > > Fixes: db1a0ae21461 ("drm/nouveau/bl: Assign different names to interfaces") > Su

Re: [Nouveau] [PATCH 1/1] drm/nouveau: fix error return code in nouveau_backlight_init()

2021-05-08 Thread Pierre Moreau
Hello Zhen, There was a similar patch sent in last month, though which does not seem to have been merged yet; see https://lists.freedesktop.org/archives/nouveau/2021-April/038451.html. Whether `ret` should be `-ENOSPC` or `-ENOMEM` is hard to say as `nouveau_get_backlight_name()` could fail due

Re: [Nouveau] [PATCH] drm/nouveau: fix an error code in nouveau_backlight_init()

2021-04-19 Thread Pierre Moreau
I can not remember why the original code did return 0 rather than an error, but -ENOMEM seems indeed way more fitting. Reviewed-by: Pierre Moreau On 2021-04-14 — 08:58, Dan Carpenter wrote: > If nouveau_get_backlight_name() fails then this should return -ENOMEM > but currently it r

Re: [Nouveau] [PATCH] drm/nouveau/mmu: Remove unneeded semicolon

2019-12-27 Thread Pierre Moreau
With the comment further below addressed, this patch is Reviewed-by: Pierre Moreau On 2019-12-16 — 11:48, zhengbin wrote: > Fixes coccicheck warning: > > drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.c:583:2-3: Unneeded semicolon > drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.h:307:2

Re: [Nouveau] [PATCH v2] drm/nouveau/mmu: Remove unneeded semicolon

2019-12-18 Thread Pierre Moreau
Reviewed-by: Pierre Moreau On 2019-12-18 — 09:28, zhengbin wrote: > Fixes coccicheck warning: > > drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.c:583:2-3: Unneeded semicolon > drivers/gpu/drm/nouveau/nvkm/subdev/mmu/vmm.h:307:2-3: Unneeded semicolon > > Reported-by: Hulk R

Re: [Nouveau] [PATCH -next] drm/nouveau: fix copy-paste error in nouveau_fence_wait_uevent_handler

2019-02-01 Thread Pierre Moreau
Friendly ping, Ben. I see that in `nouveau_fence_done()` there is a check on `chan` not being NULL prior to passing it to `nouveau_fence_update()`. Would something similar be needed here? Pierre On 2018-11-15 — 12:14, YueHaibing wrote: > Fixes gcc '-Wunused-but-set-variable' warning: > >

Re: [Nouveau] [PATCH] drm/nouveau/bl: Fix oops on driver unbind

2018-03-18 Thread Pierre Moreau
Hey Lukas, > Sorry Pierre, I missed this question and am seeing it only now on > saving the message away: No worries, and thank you for the great explanation! I’ll definitely make a mental note to try unloading Nouveau, whenever writing/testing similar patches. Thanks, Pierre signature.asc

Re: [Nouveau] [PATCH v2] bl: fix backlight regression

2018-02-19 Thread Pierre Moreau
A few words about why this change is needed would be nice, as well as what changed since the v1 (well, looking back to your original patch, absolutely everything changed :-D). On 2018-02-19 — 17:09, Karol Herbst wrote: > fixes d9c0aadc5aa241df26ce8301d34a8418919fb5ae The formatting does not

Re: [Nouveau] [PATCH] drm/nouveau/bl: Fix oops on driver unbind

2018-02-19 Thread Pierre Moreau
;drm/nouveau/bl: Do not register interface if Apple GMUX > detected") > Cc: sta...@vger.kernel.org # v4.10+ > Cc: Pierre Moreau <pierre.mor...@free.fr> > Signed-off-by: Lukas Wunner <lu...@wunner.de> > --- > I reviewed the patch causing the oops but unfortunat

Re: [Nouveau] 4.16-rc1: UBSAN warning in nouveau/nvkm/subdev/therm/base.c + oops in nvkm_therm_clkgate_fini

2018-02-14 Thread Pierre Moreau
On 2018-02-14 — 09:36, Ilia Mirkin wrote: > On Wed, Feb 14, 2018 at 9:35 AM, Ilia Mirkin wrote: > > On Wed, Feb 14, 2018 at 9:29 AM, Meelis Roos wrote: > >>> This is 4.16-rc1+todays git on a lowly P4 with NV5, worked fine in 4.15: > >> > >> NV5 in another PC

Re: [Nouveau] [PATCH] ramgf100: Fix small typo.

2018-01-09 Thread Pierre Moreau
Indeed, it looks like the devel-clk branch does not have that typo. On 2017-12-29 — 13:39, Rosen Penev wrote: > This is fixed in a different nouveau branch. from Ben Skeggs last I remember. > > On Fri, Dec 29, 2017 at 1:37 PM, Pierre Moreau <pierre.mor...@free.fr> wrote: > &

Re: [Nouveau] [PATCH v2] nv50/ir: Initialize all members of GCRA (trivial)

2017-12-30 Thread Pierre Moreau
On 2017-12-30 — 15:21, Tobias Klausmann wrote: > v2: use initialization list (Pierre) > > Signed-off-by: Tobias Klausmann <tobias.johannes.klausm...@mni.thm.de> > Reviewed-by: Pierre Moreau <pierre.mor...@free.fr> > --- > src/gallium/drivers/nouveau/codegen/nv5

Re: [Nouveau] [PATCH] drm/nouveau/hwmon: Uninitialized variables in sysfs

2017-12-29 Thread Pierre Moreau
Ping: this can still be an issue today as kstrtol() & co haven’t changed their possible return value since the patch was written. On 2017-07-22 — 14:13, Pierre Moreau wrote: > Reviewed-by: Pierre Moreau <pierre.mor...@free.fr> > > On 2017-07-17 — 11:17, Dan Carpenter

Re: [Nouveau] drm/nouveau: Remove no longer needed local ACPI_VIDEO_NOTIFY_PROBE define

2017-12-29 Thread Pierre Moreau
Ping On 2017-07-22 — 13:54, Pierre Moreau wrote: > Reviewed-by: Pierre Moreau <pierre.mor...@free.fr> > > On 2017-06-21 — 15:19, Hans de Goede wrote: > > Hi, > > > > On 26-05-17 09:35, Hans De Goede wrote: > > > The local #define of ACPI_VID

Re: [Nouveau] [PATCH] ramgf100: Fix small typo.

2017-12-29 Thread Pierre Moreau
The typo is still present in master. Not sure whether it will help you Ben, regarding Fermi reclocking, but it shouldn’t harm in reducing possible errors. :-) Regards, Pierre On 2017-07-27 — 18:20, Rosen Penev wrote: > Signed-off by: Rosen Penev > --- >

Re: [Nouveau] [PATCH] nv50/ir: Initialize all members of GCRA (trivial)

2017-12-29 Thread Pierre Moreau
It looks like this patch was never merged. You could initialise “nodeCount” and “nodes” directly in the member initialisation list. With that changed, this patch is Reviewed-by: Pierre Moreau <pierre.mor...@free.fr> On 2017-08-12 — 01:45, Tobias Klausmann wrote: > Signed-off-b

Re: [Nouveau] [PATCH] drm/nouveau/imem/nv50: fix incorrect use of refcount API

2017-12-18 Thread Pierre Moreau
Hey Ard, It seems that Ben already committed a similar patch to his tree (see [0]). I do not know whether he is planning to have it part of a pull request of fixes for 4.15. Best regards, Pierre [0]: https://github.com/skeggsb/nouveau/commit/9068f1df2394f0e4ab2b2a28cac06b462fe0a0aa On

Re: [Nouveau] [PATCH v3] nouveau/compiler: Allow to omit line numbers when printing instructions

2017-11-24 Thread Pierre Moreau
_info to forward information to the >print pass (Pierre Moreau) > > Signed-off-by: Tobias Klausmann <tobias.johannes.klausm...@mni.thm.de> > --- > src/gallium/drivers/nouveau/codegen/nv50_ir_driver.h | 1 + > src/gallium/drivers/nouveau/codegen/nv50_ir_print.cpp | 12 +-

Re: [Nouveau] [PATCH v2] nouveau/compiler: Allow to omit line numbers when printing instructions

2017-11-20 Thread Pierre Moreau
The “Program” class keeps a pointer to the “nv50_ir_prog_info” structure, so you could just use that in the “Program::print()” function, rather than passing the flag as an argument to “print”. Pierre On 2017-11-17 — 17:21, Tobias Klausmann wrote: > This comes in handy when checking

Re: [Nouveau] [PATCH] nouveau/codegen: dump tgsi floats as hex values

2017-11-15 Thread Pierre Moreau
This looks like the saner approach, compared to changing tgsi_dump.c to display more fractional digits. Maybe there could be a second option to display as both float and hex? Reviewed-by: Pierre Moreau <pierre.mor...@free.fr> On 2017-11-14 — 15:11, Tobias Klausmann wrote: > Printin

Re: [Nouveau] [RFC PATCH 17/29] clk: thermal throttling

2017-10-08 Thread Pierre Moreau
Reviewed-by: Pierre Moreau <pierre.mor...@free.fr> On 2017-09-15 — 17:11, Karol Herbst wrote: > v2: make message about relaxed throttling an info > rework reporting about current clk state > > Signed-off-by: Karol Herbst <karolher...@gmail.com> > --- > drm/nou

Re: [Nouveau] [RFC PATCH 15/29] bios: add thermal policies table

2017-10-08 Thread Pierre Moreau
Looking good. It would be nice to have some defines/enums for the different modes. Some comments about t0, t1 and t2 would be nice. I saw you are using t0 in patch 16, but I have no idea why use t0 rather than t1 or t2. Otherwise, Acked-by: Pierre Moreau <pierre.mor...@free.fr> On 2017

Re: [Nouveau] [RFC PATCH 16/29] clk: parse thermal policies for throttling thresholds

2017-10-08 Thread Pierre Moreau
On 2017-09-15 — 17:11, Karol Herbst wrote: > v2: use min_t > > Signed-off-by: Karol Herbst > --- > drm/nouveau/include/nvkm/subdev/clk.h | 2 ++ > drm/nouveau/nvkm/subdev/clk/base.c| 42 > +++ > 2 files changed, 44 insertions(+) > >

Re: [Nouveau] [RFC PATCH 14/29] therm: Trigger reclock in temperature daemon

2017-10-08 Thread Pierre Moreau
Reviewed-by: Pierre Moreau <pierre.mor...@free.fr> On 2017-09-15 — 17:11, Karol Herbst wrote: > 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 <karolher...@

Re: [Nouveau] [RFC PATCH 09/29] clk: Set clocks to pre suspend state after suspend

2017-10-08 Thread Pierre Moreau
Reviewed-by: Pierre Moreau <pierre.mor...@free.fr> On 2017-09-15 — 17:11, Karol Herbst wrote: > The idea is to clear out the saved state, because after a resume we can't > know what the GPU is clocked to. The reclock is triggered by the call to > nvkm_clk_update later in nvkm_cl

Re: [Nouveau] [RFC PATCH 08/29] clk: We should pass the pstate id around not the index in the list

2017-10-08 Thread Pierre Moreau
On 2017-09-15 — 17:11, Karol Herbst wrote: > This makes the code easier, because we can compare the id with > pstate->pstate and saves us from the trouble of iterating over the pstates > to match the index. I don’t remember whether I have already done this comment before, but I am not sure where

Re: [Nouveau] [RFC PATCH 07/29] clk: Hold information about the current cstate status

2017-10-08 Thread Pierre Moreau
Acked-by: Pierre Moreau <pierre.mor...@free.fr> On 2017-09-15 — 17:11, Karol Herbst wrote: > Later we will have situations where the expected and the current state > isn't the same. > > Signed-off-by: Karol Herbst <karolher...@gmail.com> > Reviewed-by: Martin

Re: [Nouveau] [RFC PATCH 06/29] clk: Make pstate a pointer to nvkm_pstate

2017-10-08 Thread Pierre Moreau
The patch seems fine but I found it super confusing that sometimes `pstate` is a pointer (for example `clk->pstate`), sometimes it is an int (for example `args->v0.pstate`). On 2017-09-15 — 17:11, Karol Herbst wrote: > We will access the current cstate at least every second and this saves us >

Re: [Nouveau] [RFC PATCH 05/29] clk: Remove dstate

2017-10-08 Thread Pierre Moreau
This patch could be applied before patch 04, that way patch 04 does not need to update a function (`nvkm_clk_dstate()`) that will be removed in the next patch. Reviewed-by: Pierre Moreau <pierre.mor...@free.fr> On 2017-09-15 — 17:11, Karol Herbst wrote: > We won't need it now, b

Re: [Nouveau] [RFC PATCH 04/29] clk: Rename nvkm_pstate_calc to nvkm_clk_update and export it

2017-10-08 Thread Pierre Moreau
On 2017-09-15 — 17:11, Karol Herbst wrote: > This function will be used to update the current clock state. > > This will happen for various reasons: > * Temperature changes > * User changes clocking state > * Load changes > > v2: remove parameter name > > Signed-off-by: Karol Herbst

Re: [Nouveau] [RFC PATCH 03/29] subdev/volt/gk104: return error when read fails

2017-10-08 Thread Pierre Moreau
en “that” and “sensors”. Otherwise, Reviewed-by: Pierre Moreau <pierre.mor...@free.fr> > > Signed-off-by: Karol Herbst <karolher...@gmail.com> > --- > drm/nouveau/nvkm/subdev/volt/gk104.c | 7 ++- > 1 file changed, 6 insertions(+), 1 deletion(-) > > diff --git a/drm

Re: [Nouveau] [RFC PATCH 02/29] hwmon: properly check for errors

2017-10-08 Thread Pierre Moreau
Reviewed-by: Pierre Moreau <pierre.mor...@free.fr> On 2017-09-15 — 17:11, Karol Herbst wrote: > Otherwise hwmon interprets error codes as real values. > > Signed-off-by: Karol Herbst <karolher...@gmail.com> > --- > drm/nouveau/nouveau_hwmon.c | 33

Re: [Nouveau] [RFC PATCH 01/29] therm: split return code and value in nvkm_get_temp

2017-10-08 Thread Pierre Moreau
As you changed the return value of `temp_get()` to solely be the error code, or absence of an error, I would change all those tests that checked whether the returned value was strictly less, or greater than, 0 to now only compare against 0 (no error). For example, if (therm && therm->attr_get

Re: [Nouveau] [PATCH] drm/nouveau: Document nouveau support for Tegra in DRIVER_DESC

2017-10-01 Thread Pierre Moreau
You are right, Tegra should probably be part of the description, even if we do not support pre-K1 Tegras. Acked-by: Pierre Moreau <pierre.mor...@free.fr> On 2017-08-11 — 10:56, Rhys Kidd wrote: > nouveau supports the Tegra K1 and higher after the SoC-based GPUs converged > with the

Re: [Nouveau] Power Management under Nouveau

2017-09-04 Thread Pierre Moreau
Hello Paulo, Power management is made of different things: * reclocking the memory and the engines: this is supported on some Tesla cards (>G92), all Kepler ones and on the first generation of Maxwell (GM10x). In that regard, the Arch Linux wiki is right, as Pascal cards do not support

[Nouveau] [PATCH] drm/Makefile: Support custom output dir with O

2017-08-15 Thread Pierre Moreau
having to keep a custom patch around. Signed-off-by: Pierre Moreau <pierre.mor...@free.fr> --- drm/Makefile | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drm/Makefile b/drm/Makefile index d5e23738..1f4ef45e 100644 --- a/drm/Makefile +++ b/drm/Makefile @@ -11,6 +11,9 @@ endif KBUIL

Re: [Nouveau] [RFC PATCH 13/13] clk: thermal throttling

2017-07-22 Thread Pierre Moreau
On 2017-07-21 — 23:55, Karol Herbst wrote: > Signed-off-by: Karol Herbst > --- > drm/nouveau/include/nvkm/subdev/clk.h | 1 + > drm/nouveau/nvkm/subdev/clk/base.c| 35 > +++ > 2 files changed, 32 insertions(+), 4 deletions(-) > > diff

Re: [Nouveau] [PATCH] drm/nouveau/hwmon: Uninitialized variables in sysfs

2017-07-22 Thread Pierre Moreau
Reviewed-by: Pierre Moreau <pierre.mor...@free.fr> On 2017-07-17 — 11:17, Dan Carpenter wrote: > kstrtol() and friends can return -EINVAL or -ERANGE. We have to test > for both, otherwise the value is possibly uninitialized. Also in some > of these files we accidentally return &

Re: [Nouveau] GTX 1080 Ti (NV132/GP102)

2017-05-25 Thread Pierre Moreau
> Another issue is that in Google Chrome I seem have no hardware acceleration > for anything other than "Multiple Raster Threads". See attached. Two things: you are using nomodeset, and hardware acceleration for Pascal cards was only merged in 4.12 (which has not been released yet), as pointed

Re: [Nouveau] NV130 - gtx 1050 ti

2017-05-04 Thread Pierre Moreau
Hello, Acceleration for Pascal cards is coming in Linux 4.12, support for Pascal cards in xorg-video-nouveau is in 1.0.15, and if I remember correctly, on the Mesa-side, you will need >=17.0. Pierre On 02:04 am - May 04 2017, Karl Schmidt wrote: > This is probably the problem > May 3 16:44:45

[Nouveau] [PATCH REBASED 2/2] Do not register interface if Apple GMUX detected

2016-12-19 Thread Pierre Moreau
From: Pierre Moreau <pierre.mor...@free.fr> The Apple GMUX is the one managing the backlight, so there is no need for Nouveau to register its own backlight interface. v2: Do not split information message on two lines as it prevents from grepping it, as pointed out by Lukas Wunner Sign

[Nouveau] [PATCH v3 1/2] nouveau/bl: Assign different names to interfaces

2016-12-19 Thread Pierre Moreau
From: Pierre Moreau <pierre.mor...@free.fr> Currently, every backlight interface created by Nouveau uses the same name, nv_backlight. This leads to a sysfs warning as it tries to create an already existing folder. This patch adds a incremented number to the name, but keeps the initia

[Nouveau] [PATCH v3 2/2] Do not register interface if Apple GMUX detected

2016-12-07 Thread Pierre Moreau
message Signed-off-by: Pierre Moreau <pierre.mor...@free.fr> --- drm/nouveau/nouveau_backlight.c | 6 ++ 1 file changed, 6 insertions(+) diff --git a/drm/nouveau/nouveau_backlight.c b/drm/nouveau/nouveau_backlight.c index a34cd35..8b1ca4a 100644 --- a/drm/nouveau/nouveau_backlight.c +++

[Nouveau] [PATCH v4 1/2] nouveau/bl: Assign different names to interfaces

2016-12-07 Thread Pierre Moreau
not create similarly named interfaces when reaching the maximum amount of unique names, but fail instead, as pointed out by Lukas Wunner Signed-off-by: Pierre Moreau <pierre.mor...@free.fr> --- drm/nouveau/nouveau_backlight.c | 74 ++--- drm/nouveau/nouveau_dis

Re: [Nouveau] [PATCH v3 1/2] nouveau/bl: Assign different names to interfaces

2016-11-14 Thread Pierre Moreau
On 11:17 am - Nov 14 2016, Lukas Wunner wrote: > On Sun, Nov 13, 2016 at 08:57:06PM +0100, Pierre Moreau wrote: > > From: Pierre Moreau <pierre.mor...@free.fr> > > > > Currently, every backlight interface created by Nouveau uses the same name, > > nv_backlig

Re: [Nouveau] Atomic modesetting + DisplayPort MST

2016-11-13 Thread Pierre Moreau
Hello Ben, You can find below my results; there is a slight over-representation of 1st gen Tesla cards, but… that’s what I have. :-D No regressions observed, apart from: * screen rotation on G80, MCP79 * resuming on G86 I’ll retest those cards with your linux-4.10 branch but without the atomic

Re: [Nouveau] [PATCH 1/3] nvbios/power_budget: Add basic power budget parsing

2016-11-12 Thread Pierre Moreau
On 11:55 am - Nov 12 2016, Karol Herbst wrote: > v2: Set entry to 0xff if not found > Add cap entry for ver 0x30 tables > Rework to fix memory leak > > Signed-off-by: Karol Herbst > --- > .../include/nvkm/subdev/bios/power_budget.h| 24 >

Re: [Nouveau] NVAC "No Signal"

2016-11-09 Thread Pierre Moreau
and bisecting those issues… Pierre On 06:12 am - Nov 08 2016, poma wrote: > On 21.10.2016 10:56, Pierre Moreau wrote: > > On 01:15 am - Oct 21 2016, Lukas Wunner wrote: > >> On Thu, Oct 20, 2016 at 10:08:28AM +0200, Lukas Wunner wrote: > >>> On Wed, Oct 19, 2016 at 07:58:0

Re: [Nouveau] GM108GLM?

2016-10-27 Thread Pierre Moreau
Hello, The idea was to use the modesetting DDX instead of Nouveau’s one for Maxwell+ as EXA was [broken][1]. But you can give a try at Ilia’s [patches][2], which fix the Nouveau DDX for GM10x and GM20x (I don’t think it has been tested on a GM108 yet). Best regards, Pierre Moreau [1]: https

Re: [Nouveau] NVAC "No Signal"

2016-10-21 Thread Pierre Moreau
On 01:15 am - Oct 21 2016, Lukas Wunner wrote: > On Thu, Oct 20, 2016 at 10:08:28AM +0200, Lukas Wunner wrote: > > On Wed, Oct 19, 2016 at 07:58:06PM +0200, Pierre Moreau wrote: > > > For example, my laptop (which also has an NVAC) has been triggering the > > > no

Re: [Nouveau] NVAC "No Signal"

2016-10-19 Thread Pierre Moreau
Hello, On 07:37 pm - Oct 19 2016, poma wrote: > On 19.10.2016 17:03, Karol Herbst wrote: > > > You don't get why I try to say. We have to actually find out when to > > apply this workaround, not to create some silly whitelist/blacklist. > > It's the last option, we never want to actually use. >

Re: [Nouveau] [PATCH] drm/nouveau: add Maxwell to backlight initialization

2016-08-15 Thread Pierre Moreau
@Ben: ping I would guess that the same can be done for Pascal as well. Cheers, Pierre On 02:41 PM - May 21 2016, Faris Alsalama wrote: > Signed-off-by: Faris Alsalama > --- > drivers/gpu/drm/nouveau/nouveau_backlight.c | 1 + > 1 file changed, 1 insertion(+) > > diff

Re: [Nouveau] XDC2016

2016-08-15 Thread Pierre Moreau
Hello everyone, Same place as FOSDEM? XDC is in Helsinki and FOSDEM was in Brussels… (IIRC, FOSDEM is always located in Brussels.) I would prefer picking an hotel in Helsinki rather than traveling every day between Helsinki and Brussels. :-p Or was that what you meant by "more or less"? :-D I’ve

Re: [Nouveau] [PATCH] drm/nouveau: add Maxwell to backlight initialization

2016-05-22 Thread Pierre Moreau
Acked-by: Pierre Moreau <pierre.mor...@free.fr> On 02:41 PM - May 21 2016, Faris Alsalama wrote: > Signed-off-by: Faris Alsalama <farisbenbra...@gmail.com> > --- > drivers/gpu/drm/nouveau/nouveau_backlight.c | 1 + > 1 file changed, 1 insertion(+) > > diff -

Re: [Nouveau] [PATCH] gpu/nouveau/nouveau_acpi.c: Fix Type Mismatch ACPI warning

2016-05-19 Thread Pierre Moreau
Hello Marcos, I sent a serie a year ago to fix some of the ACPI handling in Nouveau and add runtime pm support for laptops with an Apple GMUX (see [0], and especially [1] and [2]). I was told that a more generic work for the runtime pm was in the work, so I let the whole serie slip away. I was

Re: [Nouveau] [PATCH v2 2/2] nouveau/bl: Do not register interface if Apple GMUX detected

2016-05-01 Thread Pierre Moreau
On 02:32 PM - May 01 2016, Pierre Moreau wrote: > The Apple GMUX is the one managing the backlight, so there is no need for > Nouveau to register its own backlight interface. > > v2: Do not split information message on two lines as it prevents from grepping > it, as pointed

[Nouveau] [PATCH v2 2/2] nouveau/bl: Do not register interface if Apple GMUX detected

2016-05-01 Thread Pierre Moreau
The Apple GMUX is the one managing the backlight, so there is no need for Nouveau to register its own backlight interface. v2: Do not split information message on two lines as it prevents from grepping it, as pointed out by Hans de Goede Signed-off-by: Pierre Moreau <pierre.mor...@free

Re: [Nouveau] GM108GLM?

2016-04-22 Thread Pierre Moreau
> /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

[Nouveau] [PATCH v2 1/2] nouveau/bl: Assign different names to interfaces

2016-04-17 Thread Pierre Moreau
`nouveau_get_backlight_name()` to avoid forward declaration, as suggested by Ilia Mirkin; * Fix reference to bug report formatting, as reported by Nick Tenney. Signed-off-by: Pierre Moreau <pierre.mor...@free.fr> --- drm/nouveau/nouveau_backlight.c | 64 ++--- drm/n

[Nouveau] [PATCH REBASED 2/2] nouveau/bl: Do not register interface if Apple GMUX detected

2016-04-17 Thread Pierre Moreau
The Apple GMUX is the one managing the backlight, so there is no need for Nouveau to register its own backlight interface. Signed-off-by: Pierre Moreau <pierre.mor...@free.fr> --- drm/nouveau/nouveau_backlight.c | 7 +++ 1 file changed, 7 insertions(+) diff --git a/drm/n

Re: [Nouveau] [PATCH 1/2] nouveau/bl: Assign different names to interfaces

2016-04-17 Thread Pierre Moreau
[…] > > > + const int nb = atomic_inc_return(_interfaces_nb) - 1; > > This kinda sucks if you reload nouveau a bunch. How about using an > "ida". Have a look in drivers/gpu/drm/drm_crtc.c for how I use that > one. I had a quick look at drm_crtc.c. This seems to be exactly what I need,

Re: [Nouveau] [PATCH 1/2] nouveau/bl: Assign different names to interfaces

2016-04-16 Thread Pierre Moreau
On 02:40 PM - Apr 15 2016, Nick Tenney wrote: > On Fri, Apr 15, 2016 at 11:25 AM, Ilia Mirkin <imir...@alum.mit.edu> wrote: > > > On Fri, Apr 15, 2016 at 11:22 AM, Pierre Moreau <pierre.mor...@free.fr> > > wrote: > > > On 11:06 AM - Apr 15 2016, Ilia Mi

Re: [Nouveau] [PATCH 1/2] nouveau/bl: Assign different names to interfaces

2016-04-15 Thread Pierre Moreau
On 11:06 AM - Apr 15 2016, Ilia Mirkin wrote: > On Fri, Apr 15, 2016 at 10:57 AM, Pierre Moreau <pierre.mor...@free.fr> wrote: > > Currently, every backlight interface created by Nouveau uses the same name, > > nv_backlight. This leads to a sysfs warning as it tries

[Nouveau] [PATCH 2/2] nouveau/bl: Do not register interface if Apple GMUX detected

2016-04-15 Thread Pierre Moreau
The Apple GMUX is the one managing the backlight, so there is no need for Nouveau to register its own backlight interface. Signed-off-by: Pierre Moreau <pierre.mor...@free.fr> --- drm/nouveau/nouveau_backlight.c | 7 +++ 1 file changed, 7 insertions(+) diff --git a/drm/n

[Nouveau] [PATCH 1/2] nouveau/bl: Assign different names to interfaces

2016-04-15 Thread Pierre Moreau
userspace; the second interface will be named nv_backlight1, and so on. Fixes: fdo#86539 Signed-off-by: Pierre Moreau <pierre.mor...@free.fr> --- drm/nouveau/nouveau_backlight.c | 35 +-- 1 file changed, 33 insertions(+), 2 deletions(-) diff --git a/drm/n

Re: [Nouveau] GM108GLM?

2016-03-31 Thread Pierre Moreau
108 is not one of the recognized chips. Someone needs to go > through and check that its goldens didn't change. Nobody's done that. > > See https://bugs.freedesktop.org/show_bug.cgi?id=89558 > > On Thu, Mar 31, 2016 at 4:57 PM, Pierre Moreau <pierre.mor...@free.fr> wrote: > &g

Re: [Nouveau] GM108GLM?

2016-03-31 Thread Pierre Moreau
Hello, Acceleration support for GM107 was merged in kernel 4.1, and modesetting support was added to 3.15. Which kernel version did you try? The GM108 chipset seems to be recognised since at least 2015/08/20. Regards, Pierre Moreau On 02:55 PM - Mar 29 2016, Sune Mølgaard wrote: > H

Re: [Nouveau] [Mesa-dev] [PATCH mesa 3/3] nouveau: Add support for clover / OpenCL kernel input parameters

2016-03-10 Thread Pierre Moreau
On 11:05 AM - Mar 10 2016, Ilia Mirkin wrote: > On Thu, Mar 10, 2016 at 11:03 AM, Pierre Moreau <pierre.mor...@free.fr> wrote: > > You might want to increment the address by at least > > `info->prop.cp.inputOffset`, and if inputs still end up in shared on Tesla, > >

Re: [Nouveau] [PATCH mesa 3/3] nouveau: Add support for clover / OpenCL kernel input parameters

2016-03-10 Thread Pierre Moreau
On 04:27 PM - Mar 10 2016, Samuel Pitoiset wrote: > > > On 03/10/2016 04:23 PM, Ilia Mirkin wrote: > >On Thu, Mar 10, 2016 at 10:14 AM, Hans de Goede wrote: > >>Add support for clover / OpenCL kernel input parameters. > >> > >>Signed-off-by: Hans de Goede

Re: [Nouveau] RFC: [PATCH] x86/kmmio: fix mmiotrace for hugepages

2016-03-03 Thread Pierre Moreau
The secondary hit exception thrown while MMIOtracing NVIDIA's driver is gone with this patch. Tested-by: Pierre Moreau <pierre.mor...@free.fr> On 02:03 AM - Mar 03 2016, Karol Herbst wrote: > Because Linux might use bigger pages than the 4K pages to handle those mmio > ioremaps, th

Re: [Nouveau] Dealing with opencl kernel parameters in nouveau now that RES support is gone

2016-02-23 Thread Pierre Moreau
On 11:43 AM - Feb 23 2016, Hans de Goede wrote: > Hi, > [snip] > > >You may have to add LOAD64/STORE64 for 64-bit > >addresses though. Or we could decree that all addressing on global > >memory shall be 64-bit (and thus read the .xy components of the > >address source). > > I would prefer to

Re: [Nouveau] Dealing with opencl kernel parameters in nouveau now that RES support is gone

2016-02-22 Thread Pierre Moreau
- Mail original - > On Mon, Feb 22, 2016 at 10:50 AM, Hans de Goede > wrote: > >> But assuming I'm right, what I'm proposing is that instead of > >> passing > >> the input in as a global buffer, to instead pass it in as a const > >> buffer. As such instead of

Re: [Nouveau] Dealing with opencl kernel parameters in nouveau now that RES support is gone

2016-02-22 Thread Pierre Moreau
Hello, > On 22 Feb 2016, at 15:22, Ilia Mirkin wrote: > >> On Mon, Feb 22, 2016 at 9:17 AM, Hans de Goede wrote: >> Hi, >> >>> On 22-02-16 14:47, Ilia Mirkin wrote: >>> On Mon, Feb 22, 2016 at 8:45 AM, Ilia Mirkin wrote:

Re: [Nouveau] [PATCH v5 00/12] Enable GPU switching on pre-retina MacBook Pro

2016-02-01 Thread Pierre Moreau
reviews or acks unfortunately. > Any takers? > > Merging this would allow fdo #61115 to be closed > (currently assigned to intel-gfx). > > FWIW this series has in the meantime been tested by more folks: > > Tested-by: Pierre Moreau <pierre.mor...@free.fr> >

Re: [Nouveau] nouveau contributor dinner Friday evening before FOSDEM?

2016-01-29 Thread Pierre Moreau
at: > >>> > >>> > >>>http://www.tripadvisor.com/Restaurant_Review-g188644-d802232-Reviews-Le_Falstaff-Brussels.html > >>> > >>>Which is not as bad as the reviews there make it out to be, > >>>note I'm open to better suggestions, esp. fro

Re: [Nouveau] nouveau contributor dinner Friday evening before FOSDEM?

2016-01-21 Thread Pierre Moreau
Hello everyone, I'm in for the dinner! As for the restaurant, since I never went to Brussels, I'll follow the group's decision. Some time around 19:30–20:00? Most of us should be in Brussels' centre by then I think. Cheers, Pierre On 02:09 PM - Jan 21 2016, Hans de Goede wrote: > Hi All, > >

Re: [Nouveau] [RFC PATCH v2 0/7] stabilize kepler reclocking

2015-12-25 Thread Pierre Moreau
Hello, Maybe my e-mail client is messing with me, but I couldn't find any dmesg output attached to your e-mail. Could you please try to attach it again? By the way, since you have a Kepler, you should try booting with "nouveau.War00C800_0=1". That workaround is enabled by default in 4.4-rc5

Re: [Nouveau] [RFC PATCH 4/5] subdev/clk: print the base clocks

2015-12-01 Thread Pierre Moreau
Hi, On 05:42 PM - Dec 01 2015, Karol Herbst wrote: > this is just a nice thing to know and there is no harm in printing them > --- > drm/nouveau/nvkm/subdev/clk/base.c | 16 > 1 file changed, 16 insertions(+) > > diff --git a/drm/nouveau/nvkm/subdev/clk/base.c >

Re: [Nouveau] [RFC PATCH 5/5] clk: allow boosting only when NvBoost is set

2015-12-01 Thread Pierre Moreau
Hi Karol, I have some comments below. On 05:42 PM - Dec 01 2015, Karol Herbst wrote: > 0: disable boosting (cap to base clock from the vbios) > 1: boost only to boost clock from the vbios > 2: boost to max clock available > --- > drm/nouveau/include/nvkm/subdev/clk.h | 10 +- >

Re: [Nouveau] NV50 compute support questions

2015-11-20 Thread Pierre Moreau
eda118232fabd53965410afd > you write: "This compute support has been tested by > Pierre Moreau and myself with some compute kernels." > > Can you provide testing instructions (and the > necessary files) so that I can try to reproduce > your tests ? > > And once I've

Re: [Nouveau] Nouveau for FreeBSD

2015-11-04 Thread Pierre Moreau
Wasn't there some work being done by François Tigeot or Jean-Sébastien Pédron to port Nouveau to DragonFlyBSD/FreeBSD? Or at least they were talking about porting it at XDC 2014, no idea what the current status is. Pierre PS: Looking at François' slides from this year XDC, it seems there isn't

Re: [Nouveau] Chipset & Family

2015-10-06 Thread Pierre Moreau
Hello poma, The chipset didn't disappear and is still displayed: it is the G98 you get on the "[2.483843] nouveau :02:00.0: NVIDIA G98 (098200a2)" line. The "NV98" was the "Nouveau" chipset, but the switch was made to use the same naming as NVIDIA. So rather than displaying both the

Re: [Nouveau] [PATCH v3 0/6] Enable gpu switching on the MacBook Pro

2015-10-05 Thread Pierre Moreau
The repo is a fork of Linus' tree, with the patches on top of it. So you just need to build that modified version of the kernel and boot it. :-) Pierre > On 05 Oct 2015, at 16:15, Evan Foss wrote: > >> On Mon, Oct 5, 2015 at 9:23 AM, Lukas Wunner wrote:

[Nouveau] [PATCH v5] pci: Handle 5-bit and 8-bit tag field

2015-10-03 Thread Pierre Moreau
rom nvkm_pci_init to g84_pci_init and remove PCIe and chipset checks v5: * Rebase code on latest PCI structure * Restore PCIe check * Fix namings in nvkm_pci_mask * Rephrase part of the commit message Signed-off-by: Pierre Moreau <pierre.mor...@free.fr> --- drm/nouveau/include/nv

[Nouveau] [PATCH v4] pci: Handle 5-bit and 8-bit tag field

2015-09-30 Thread Pierre Moreau
nit to g84_pci_init and remove PCIe and chipset checks Signed-off-by: Pierre Moreau <pierre.mor...@free.fr> --- drm/nouveau/include/nvkm/subdev/pci.h | 4 +++ drm/nouveau/nvkm/engine/device/base.c | 64 +-- drm/nouveau/nvkm/subdev/pci/Kbuild| 1 + dr

Re: [Nouveau] [PATCH v3] pci: Handle 5-bit and 8-bit tag field

2015-09-26 Thread Pierre Moreau
sible values to 32. Apparently bits 7:0 of 0x8841c > stores some > number of outstanding requests, so cap it to 32 if extended tag is > unsupported. > > Fixes: fdo#86537 > > v2: Restrict changes to chipsets >= 0x84 > v3: > * Add nvkm_pci_mask to pci.h > * Mask

[Nouveau] [PATCH v2] pci: Handle 5-bit and 8-bit tag field

2015-09-16 Thread Pierre Moreau
some number of outstanding requests, so cap it to 32 if extended tag is unsupported. Fixes: fdo#86537 v2: Restrict changes to chipsets >= 0x84 Signed-off-by: Pierre Moreau <pierre.mor...@free.fr> --- drm/nouveau/nvkm/subdev/pci/base.c | 25 + 1 file changed, 25 i

[Nouveau] [PATCH v3] pci: Handle 5-bit and 8-bit tag field

2015-09-16 Thread Pierre Moreau
some number of outstanding requests, so cap it to 32 if extended tag is unsupported. Fixes: fdo#86537 v2: Restrict changes to chipsets >= 0x84 v3: * Add nvkm_pci_mask to pci.h * Mask bit 8 before setting it Signed-off-by: Pierre Moreau <pierre.mor...@free.fr> --- drm/nouveau/include/nv

[Nouveau] [PATCH] pci: Handle 5-bit and 8-bit tag field

2015-09-06 Thread Pierre Moreau
some number of outstanding requests, so cap it to 32 if extended tag is unsupported. Fixes: fdo#86537 Signed-off-by: Pierre Moreau <d...@pmoreau.org> --- drm/nouveau/nvkm/subdev/pci/base.c | 25 + 1 file changed, 25 insertions(+) diff --git a/drm/nouveau/nvkm/subd

Re: [Nouveau] Will nouveau support for cuda?

2015-07-05 Thread Pierre Moreau
Hello, Nouveau will support Cuda and OpenCL, someday. :) As to say when, I don't know. I am starting to work on translating between different intermediate representation (from SPIR-V to NV50 IR), and will later work on finishing compute support for Tesla cards (I only have Tesla cards, apart

Re: [Nouveau] [PATCH v2 8/9] acpi: Add support for Apple Gmux _DMS

2015-05-30 Thread Pierre Moreau
Hi Lukas - Mail original - Hi Pierre, On Thu, May 28, 2015 at 10:52:56AM +0200, Pierre Moreau wrote: How is it supposed to work, by simply echo'ing ON or OFF to vga_switcheroo/switch? Then I probably forgot some stuff as it doesn't want to work on my laptop. What exactly

Re: [Nouveau] [PATCH NEW 5/9] acpi: Use a single type of errors in evaluate functions

2015-05-28 Thread Pierre Moreau
I wasn't sure how to insert this one into the flow, if I should make it a patch 4.5 or not. I went with patch 5, so all the following patches are incremented by 1 compared to the v1, sorry about that. What should be the way to go? Pierre On 28 May 2015, at 08:40, Pierre Moreau pierre.mor

[Nouveau] [PATCH v2 8/9] acpi: Add support for Apple Gmux _DMS

2015-05-28 Thread Pierre Moreau
-by: Pierre Moreau pierre.mor...@free.fr --- drm/nouveau/nouveau_acpi.c | 57 -- drm/nouveau/nouveau_acpi.h | 4 drm/nouveau/nouveau_drm.c | 12 +- drm/nouveau/nouveau_vga.c | 10 4 files changed, 70 insertions(+), 13 deletions(-) diff

[Nouveau] [PATCH v2 2/9] acpi: Add evaluate to name of functions evaluating _DSM

2015-05-28 Thread Pierre Moreau
This makes it clearer when reading the function name, as well as following the names of related ACPI functions. Changes since v1: * Fix typo in commit message * Slightly reformulate the commit message to be clearer Signed-off-by: Pierre Moreau pierre.mor...@free.fr --- drm/nouveau

[Nouveau] [PATCH v2 1/9] acpi: Rename v1 DSM to mux to avoid ambiguity

2015-05-28 Thread Pierre Moreau
This is especially true when variables or functions are just called dsm without specifying the v1. Changes since v1: * Fix typo in commit explanation * Change has_dsm to has_mux in nouveau_dsm_detect Signed-off-by: Pierre Moreau pierre.mor...@free.fr --- drm/nouveau/nouveau_acpi.c | 72

[Nouveau] [PATCH 3/9] acpi: Define static variables for revision ids

2015-05-28 Thread Pierre Moreau
Signed-off-by: Pierre Moreau pierre.mor...@free.fr --- drm/nouveau/nouveau_acpi.c | 8 +--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/drm/nouveau/nouveau_acpi.c b/drm/nouveau/nouveau_acpi.c index e18fbeb..3f01067 100644 --- a/drm/nouveau/nouveau_acpi.c +++ b/drm/nouveau

[Nouveau] [PATCH v2 9/9] acpi: Move detecting available _DSM outside of while loops

2015-05-28 Thread Pierre Moreau
Changes since v1: * Re-write the whole patch to detect available _DSMs when we enable them rather than on each loop run Signed-off-by: Pierre Moreau pierre.mor...@free.fr --- drm/nouveau/nouveau_acpi.c | 27 ++- 1 file changed, 6 insertions(+), 21 deletions(-) diff

[Nouveau] [PATCH v2 4/9] acpi: Allow evaluate_optimus_dsm to take NULL results

2015-05-28 Thread Pierre Moreau
Changes since v1: * Fix check for result being not NULL in nouveau_evaluate_optimus_dsm Signed-off-by: Pierre Moreau pierre.mor...@free.fr --- drm/nouveau/nouveau_acpi.c | 10 -- 1 file changed, 4 insertions(+), 6 deletions(-) diff --git a/drm/nouveau/nouveau_acpi.c b/drm/nouveau

[Nouveau] [PATCH v2 6/9] acpi: Check returned object type by Optimus _DSM locally

2015-05-28 Thread Pierre Moreau
: * Replace ACPI errors with standard kernel ones Signed-off-by: Pierre Moreau pierre.mor...@free.fr --- drm/nouveau/nouveau_acpi.c | 15 --- 1 file changed, 12 insertions(+), 3 deletions(-) diff --git a/drm/nouveau/nouveau_acpi.c b/drm/nouveau/nouveau_acpi.c index b837ab1..44a7328

  1   2   >