search for: data_valid

The power sensors can be configured to sample the readout values over time. Nvidia does this too, so nouveau should probably do that too. Karol Herbst (4): iccsense: remove read function iccsense: convert to linked list iccsense: split sensor into own struct iccsense: configure sensors like nvidia does drm/nouveau/include/nvkm/subdev/iccsense.h | 6 +- drm/nouveau/nouveau_hwmon.c

...> > > > > > > > Fixes commit bb91accf2733 ("virtio-net: XDP support for small buffers") > > > > > > > > > Signed-off-by: Jason Wang<jasowang at redhat.com> > > > > > > > > The issue would be VIRTIO_NET_HDR_F_DATA_VALID might be set. > > > > > > > > What do you think? > > > > > > > I think it's safe. For XDP_PASS, it work like in the past. > > > > > > That's the part I don't get. With DATA_VALID csum in packet is wrong, XDP > > > &...

...> > > > > > > > Fixes commit bb91accf2733 ("virtio-net: XDP support for small buffers") > > > > > > > > > Signed-off-by: Jason Wang<jasowang at redhat.com> > > > > > > > > The issue would be VIRTIO_NET_HDR_F_DATA_VALID might be set. > > > > > > > > What do you think? > > > > > > > I think it's safe. For XDP_PASS, it work like in the past. > > > > > > That's the part I don't get. With DATA_VALID csum in packet is wrong, XDP > > > &...

The power sensors can be configured to sample the readout values over time. Nvidia does this too, so nouveau should probably do that too. v2: use list_del and rework an error message Karol Herbst (4): iccsense: remove read function iccsense: convert to linked list iccsense: split sensor into own struct iccsense: configure sensors like nvidia does

...v/iccsense.h index c3defcd..a4c0da0 100644 --- a/drm/nouveau/include/nvkm/subdev/iccsense.h +++ b/drm/nouveau/include/nvkm/subdev/iccsense.h @@ -3,12 +3,10 @@ #include <core/subdev.h> -struct nkvm_iccsense_rail; struct nvkm_iccsense { struct nvkm_subdev subdev; - u8 rail_count; bool data_valid; - struct nvkm_iccsense_rail *rails; + struct list_head rails; }; int gf100_iccsense_new(struct nvkm_device *, int index, struct nvkm_iccsense **); diff --git a/drm/nouveau/nouveau_hwmon.c b/drm/nouveau/nouveau_hwmon.c index 67edd2f..74f237b 100644 --- a/drm/nouveau/nouveau_hwmon.c +++ b/drm/no...

...--git a/drm/nouveau/include/nvkm/subdev/iccsense.h b/drm/nouveau/include/nvkm/subdev/iccsense.h index a4c0da0..3c2ddd9 100644 --- a/drm/nouveau/include/nvkm/subdev/iccsense.h +++ b/drm/nouveau/include/nvkm/subdev/iccsense.h @@ -6,6 +6,7 @@ struct nvkm_iccsense { struct nvkm_subdev subdev; bool data_valid; + struct list_head sensors; struct list_head rails; }; diff --git a/drm/nouveau/nvkm/subdev/iccsense/base.c b/drm/nouveau/nvkm/subdev/iccsense/base.c index 6fde68d..b6f6222 100644 --- a/drm/nouveau/nvkm/subdev/iccsense/base.c +++ b/drm/nouveau/nvkm/subdev/iccsense/base.c @@ -30,15 +30,14 @@...

...t;> >>>>>>>>>> Fixes commit bb91accf2733 ("virtio-net: XDP support for small buffers") >>>>>>>>>> Signed-off-by: Jason Wang<jasowang at redhat.com> >>>>>>>>> The issue would be VIRTIO_NET_HDR_F_DATA_VALID might be set. >>>>>>>>> What do you think? >>>>>>>> I think it's safe. For XDP_PASS, it work like in the past. >>>>>>> That's the part I don't get. With DATA_VALID csum in packet is wrong, XDP >>>>>&g...

...switch (attr) { > + case hwmon_power_input: > + if (iccsense && move the pointer check before the switch, because you need to check it in every case (or move it to every case, but doing it once is simplier/cleaner) > + iccsense->data_valid && > + !list_empty(&iccsense->rails)) > + return 0444; add return due to fallthrough > + case hwmon_power_max: > + if (iccsense->power_w_max) > + return 0444; add return due to...

...tions(+), 1 deletion(-) diff --git a/drm/nouveau/include/nvkm/subdev/iccsense.h b/drm/nouveau/include/nvkm/subdev/iccsense.h index 3c2ddd9..b7a9b04 100644 --- a/drm/nouveau/include/nvkm/subdev/iccsense.h +++ b/drm/nouveau/include/nvkm/subdev/iccsense.h @@ -8,6 +8,9 @@ struct nvkm_iccsense { bool data_valid; struct list_head sensors; struct list_head rails; + + u32 power_w_max; + u32 power_w_crit; }; int gf100_iccsense_new(struct nvkm_device *, int index, struct nvkm_iccsense **); diff --git a/drm/nouveau/nvkm/subdev/iccsense/base.c b/drm/nouveau/nvkm/subdev/iccsense/base.c index 658355f..4e2b...

Hallo, at first, sorry for my bad english. I have a big problem. I have xen 3.0.1 running. I have 4 domU. One domU is hosting a dns-server. All runns fine. Yesterday i have updated to 3.0.2. Sorce from the tar file. Build a new kernel and reboot the system. It''s running. But i have one problem. The dns resolving runns not fine. Is there any changes for the dns between 3.0.1 and

On 2017?01?21? 00:45, Michael S. Tsirkin wrote: > On Fri, Jan 20, 2017 at 02:32:42PM +0800, Jason Wang wrote: >> Commit 501db511397f ("virtio: don't set VIRTIO_NET_HDR_F_DATA_VALID on >> xmit") in fact disables VIRTIO_HDR_F_DATA_VALID on receiving path too, >> fixing this by adding a hint (has_data_valid) and set it only on the >> receiving path. >> >> Cc: Rolf Neugebauer<rolf.neugebauer at docker.com> >> Signed-off-by: Jason Wan...

On 2017?01?21? 00:45, Michael S. Tsirkin wrote: > On Fri, Jan 20, 2017 at 02:32:42PM +0800, Jason Wang wrote: >> Commit 501db511397f ("virtio: don't set VIRTIO_NET_HDR_F_DATA_VALID on >> xmit") in fact disables VIRTIO_HDR_F_DATA_VALID on receiving path too, >> fixing this by adding a hint (has_data_valid) and set it only on the >> receiving path. >> >> Cc: Rolf Neugebauer<rolf.neugebauer at docker.com> >> Signed-off-by: Jason Wan...

...mean with return due to fallthrough. I mean, I know what is it, but I don't see how I can do it there. Moving the check before the switch looks like that: if (!iccsense) return 0; switch (attr) { case hwmon_power_input: if (iccsense->data_valid && !list_empty(&iccsense->rails)) return 0444; case hwmon_power_max: if (iccsense->power_w_max) return 0444; case hwmon_power_crit: if (iccsense->power_w_crit...

There is an optinal header field in the power budget table we can use to read out the power cap of the GPU. The meaning of this header field was confirmed with nvidia-smi -q: The rows "Min Power Limit", "Power Limit" and "Max Power Limit" are set to the "min", "avg" and "max" values from the referenced power budget entry respectively.

...ouveau_power_is_visible(const void *data, u32 attr, int channel) +{ + struct nouveau_drm *drm = nouveau_drm((struct drm_device *)data); + struct nvkm_iccsense *iccsense = nvxx_iccsense(&drm->client.device); + + switch (attr) { + case hwmon_power_input: + if (iccsense && iccsense->data_valid && + !list_empty(&iccsense->rails)) + return 0444; + case hwmon_power_max: + case hwmon_power_crit: + if (iccsense->power_w_max && iccsense->power_w_crit) + return 0444; + default: + return 0; + } +} + +static umode_t +nouveau_temp_is_visible(const void *data, u...

...wmon_power_attributes, > }; > +static const struct attribute_group hwmon_power_caps_attrgroup = { > + .attrs = hwmon_power_caps_attributes, > +}; > #endif > > int > @@ -728,8 +763,16 @@ nouveau_hwmon_init(struct drm_device *dev) > if (iccsense && iccsense->data_valid && !list_empty(&iccsense->rails)) { > ret = sysfs_create_group(&hwmon_dev->kobj, > &hwmon_power_attrgroup); > + > if (ret) > goto error; > + > + if (iccsense->power_w_max && iccsense->power_w_crit) { > + ret = sysfs...

There is an optinal header field in the power budget table we can use to read out the power cap of the GPU. Sadly it is optional and if that field isn't sad, things beome complicated. Anyhow, this is good enough for most cards and we can use it later for capping the power consumption of the GPUs, but first, just export those values through hwmon. First design, will change stuff, want

There is an optinal header field in the power budget table we can use to read out the power cap of the GPU. We should have this in the kernel before actually using it to catch errors and see how reliable this is, but as it seems it works on all GPUs as expected on Kepler und Maxwells with the power cap field set in the vbios. This series keeps things really simple for now until we figure out

There is an optinal header field in the power budget table we can use to read out the power cap of the GPU. We should have this in the kernel before actually using it to catch errors and see how reliable this is, but as it seems it works on all GPUs as expected on Kepler und Maxwells with the power cap field set in the vbios. This series keeps things really simple for now until we figure out

...l) > +{ > + struct nouveau_drm *drm = nouveau_drm((struct drm_device *)data); > + struct nvkm_iccsense *iccsense = nvxx_iccsense(&drm->client.device); > + > + if (!iccsense) > + return 0; > + > + switch (attr) { > + case hwmon_power_input: > + if (iccsense->data_valid && > + !list_empty(&iccsense->rails)) Not sure if the kernel coding style would mandate !list_empty to be aligned with the iccsense in the above line, but this is our coding style at least. Anyway, this condition has to be moved before the switch as we should not expose power_...