[Intel-gfx] [PATCH 21/22] drm/i915/tgl: Add and use new DC5 and DC6 residency counter registers
Lucas De Marchi
lucas.demarchi at intel.com
Fri Jul 19 18:39:54 UTC 2019
On Thu, Jul 18, 2019 at 11:17:03AM +0530, Anshuman Gupta wrote:
>On 2019-07-12 at 18:09:39 -0700, Lucas De Marchi wrote:
>> From: José Roberto de Souza <jose.souza at intel.com>
>>
>> Tiger Lask has a new register offset for DC5 and DC6 residency counters.
>>
>> Signed-off-by: José Roberto de Souza <jose.souza at intel.com>
>> Signed-off-by: Lucas De Marchi <lucas.demarchi at intel.com>
>> ---
>> drivers/gpu/drm/i915/i915_debugfs.c | 21 +++++++++++++--------
>> drivers/gpu/drm/i915/i915_reg.h | 2 ++
>> 2 files changed, 15 insertions(+), 8 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
>> index a1843e3de6d7..4b12b5588dd6 100644
>> --- a/drivers/gpu/drm/i915/i915_debugfs.c
>> +++ b/drivers/gpu/drm/i915/i915_debugfs.c
>> @@ -2465,6 +2465,7 @@ static int i915_dmc_info(struct seq_file *m, void *unused)
>> struct drm_i915_private *dev_priv = node_to_i915(m->private);
>> intel_wakeref_t wakeref;
>> struct intel_csr *csr;
>> + i915_reg_t dc5_reg, dc6_reg = {};
>>
>> if (!HAS_CSR(dev_priv))
>> return -ENODEV;
>> @@ -2482,15 +2483,19 @@ static int i915_dmc_info(struct seq_file *m, void *unused)
>> seq_printf(m, "version: %d.%d\n", CSR_VERSION_MAJOR(csr->version),
>> CSR_VERSION_MINOR(csr->version));
>>
>> - if (WARN_ON(INTEL_GEN(dev_priv) > 11))
>> - goto out;
>> + if (INTEL_GEN(dev_priv) >= 12) {
>> + dc5_reg = TGL_CSR_DC5_RES_COUNT;
>> + dc6_reg = TGL_CSR_DC6_RES_COUNT;
>> + } else {
>> + dc5_reg = IS_BROXTON(dev_priv) ? BXT_CSR_DC3_DC5_COUNT :
>> + SKL_CSR_DC3_DC5_COUNT;
>> + if (!IS_GEN9_LP(dev_priv))
>> + dc6_reg = SKL_CSR_DC5_DC6_COUNT;
>> + }
>>
>> - seq_printf(m, "DC3 -> DC5 count: %d\n",
>> - I915_READ(IS_BROXTON(dev_priv) ? BXT_CSR_DC3_DC5_COUNT :
>> - SKL_CSR_DC3_DC5_COUNT));
>> - if (!IS_GEN9_LP(dev_priv))
>> - seq_printf(m, "DC5 -> DC6 count: %d\n",
>> - I915_READ(SKL_CSR_DC5_DC6_COUNT));
>> + seq_printf(m, "DC3 -> DC5 count: %d\n", I915_READ(dc5_reg));
>> + if (dc6_reg.reg)
>> + seq_printf(m, "DC5 -> DC6 count: %d\n", I915_READ(dc6_reg));
>>
>> out:
>> seq_printf(m, "program base: 0x%08x\n", I915_READ(CSR_PROGRAM(0)));
>> diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
>> index 6cfcdf6bb1bb..5e3434cdd1e8 100644
>> --- a/drivers/gpu/drm/i915/i915_reg.h
>> +++ b/drivers/gpu/drm/i915/i915_reg.h
>> @@ -7269,6 +7269,8 @@ enum {
>> #define SKL_CSR_DC3_DC5_COUNT _MMIO(0x80030)
>> #define SKL_CSR_DC5_DC6_COUNT _MMIO(0x8002C)
>> #define BXT_CSR_DC3_DC5_COUNT _MMIO(0x80038)
>> +#define TGL_CSR_DC5_RES_COUNT _MMIO(0x101084)
>> +#define TGL_CSR_DC6_RES_COUNT _MMIO(0x101088)
>CSR memory ranges from 0x80000 till 0x8ffff
>These DMC_DEBUG regisers are not are not CSR register.
>It would make sense to rename the macro.
True.
TGL_DMC_DEBUG_DC5_COUNT and TGL_DMC_DEBUG_DC6_COUNT sounds good?
Lucas De Marchi
>>
>> /* interrupts */
>> #define DE_MASTER_IRQ_CONTROL (1 << 31)
>> --
>> 2.21.0
>>
>> _______________________________________________
>> Intel-gfx mailing list
>> Intel-gfx at lists.freedesktop.org
>> https://lists.freedesktop.org/mailman/listinfo/intel-gfx
More information about the Intel-gfx
mailing list