On 28.01.2016 03:25, Huang Rui wrote:
On Wed, Jan 27, 2016 at 12:12:56PM +0100, Borislav Petkov wrote:
On Wed, Jan 27, 2016 at 12:02:09PM +0100, Gioh Kim wrote:
Add bit masking to read ApmTdpLimit precisely
Signed-off-by: Gioh Kim <[email protected]>
---
drivers/hwmon/fam15h_power.c | 10 +++++++++-
1 file changed, 9 insertions(+), 1 deletion(-)
diff --git a/drivers/hwmon/fam15h_power.c b/drivers/hwmon/fam15h_power.c
index f77eb97..4f695d8 100644
--- a/drivers/hwmon/fam15h_power.c
+++ b/drivers/hwmon/fam15h_power.c
@@ -90,7 +90,15 @@ static ssize_t show_power(struct device *dev,
pci_bus_read_config_dword(f4->bus, PCI_DEVFN(PCI_SLOT(f4->devfn), 5),
REG_TDP_LIMIT3, &val);
- tdp_limit = val >> 16;
+ /*
+ * On Carrizo and later platforms, ApmTdpLimit bit field
+ * is extended to 16:31 from 16:28.
+ */
+ if (boot_cpu_data.x86 == 0x15 && boot_cpu_data.x86_model >= 0x60)
+ tdp_limit = val >> 16;
+ else
+ tdp_limit = (val >> 16) & 0x1fff;
+
curr_pwr_watts = ((u64)(tdp_limit +
data->base_tdp)) << running_avg_range;
curr_pwr_watts -= running_avg_capture;
--
Acked-by: Borislav Petkov <[email protected]>
Btw, Rui, you could consider unifying the code under a single
if (boot_cpu_data.x86 == 0x15 && boot_cpu_data.x86_model >= 0x60) {
...
else
...
as with this patch you'll have two of those checks. Unified might be
better readable but that is for another patch.
Make sence, I will do that. :-)
Thanks,
Rui
Agree. It'll be better.
Thank you.
--
Best regards,
Gi-Oh Kim
