Skip to contentSkip to conversation

drm/radeon/kms: implement reading active PCIE lanes on R600+

PATCH In review Rafał Miłecki 2 participants 5 messages Feb 18, 2010

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH v3 1/1 drm/radeon/kms: implement reading active PCIE lanes on R600+

---
V2: use already implemented functions for reading/writing PCIE PORT
V3: fix rreg/wreg typo in macro

Thanks Alex for reviewing!

---
 drivers/gpu/drm/radeon/r300.c        |    5 ++++-
 drivers/gpu/drm/radeon/radeon.h      |    2 ++
 drivers/gpu/drm/radeon/radeon_asic.h |    4 ++--
 drivers/gpu/drm/radeon/radeon_pm.c   |    2 ++
 4 files changed, 10 insertions(+), 3 deletions(-)
drivers/gpu/drm/radeon/r300.c +4 -1
index 7e9f956..29ef9d8 100644
@@ -549,7 +549,10 @@ int rv370_get_pcie_lanes(struct radeon_device *rdev)
 
 	/* FIXME wait for idle */
 
-	link_width_cntl = RREG32_PCIE(RADEON_PCIE_LC_LINK_WIDTH_CNTL);
+	if (rdev->family < CHIP_R600)
+		link_width_cntl = RREG32_PCIE(RADEON_PCIE_LC_LINK_WIDTH_CNTL);
+	else
+		link_width_cntl = RREG32_PCIE_P(RADEON_PCIE_LC_LINK_WIDTH_CNTL);
 
 	switch ((link_width_cntl & RADEON_PCIE_LC_LINK_WIDTH_RD_MASK) >> RADEON_PCIE_LC_LINK_WIDTH_RD_SHIFT) {
 	case RADEON_PCIE_LC_LINK_WIDTH_X0:
drivers/gpu/drm/radeon/radeon.h +2 -0
index b110994..ef55955 100644
@@ -1014,6 +1014,8 @@ static inline void r100_mm_wreg(struct radeon_device *rdev, uint32_t reg, uint32
 #define WREG32_MC(reg, v) rdev->mc_wreg(rdev, (reg), (v))
 #define RREG32_PCIE(reg) rv370_pcie_rreg(rdev, (reg))
 #define WREG32_PCIE(reg, v) rv370_pcie_wreg(rdev, (reg), (v))
+#define RREG32_PCIE_P(reg) rdev->pciep_rreg(rdev, (reg))
+#define WREG32_PCIE_P(reg, v) rdev->pciep_wreg(rdev, (reg), (v))
 #define WREG32_P(reg, val, mask)				\
 	do {							\
 		uint32_t tmp_ = RREG32(reg);			\
drivers/gpu/drm/radeon/radeon_asic.h +2 -2
index b7030d7..4572a66 100644
@@ -626,7 +626,7 @@ static struct radeon_asic r600_asic = {
 	.set_engine_clock = &radeon_atom_set_engine_clock,
 	.get_memory_clock = &radeon_atom_get_memory_clock,
 	.set_memory_clock = &radeon_atom_set_memory_clock,
-	.get_pcie_lanes = NULL,
+	.get_pcie_lanes = &rv370_get_pcie_lanes,
 	.set_pcie_lanes = NULL,
 	.set_clock_gating = NULL,
 	.set_surface_reg = r600_set_surface_reg,
@@ -672,7 +672,7 @@ static struct radeon_asic rv770_asic = {
 	.set_engine_clock = &radeon_atom_set_engine_clock,
 	.get_memory_clock = &radeon_atom_get_memory_clock,
 	.set_memory_clock = &radeon_atom_set_memory_clock,
-	.get_pcie_lanes = NULL,
+	.get_pcie_lanes = &rv370_get_pcie_lanes,
 	.set_pcie_lanes = NULL,
 	.set_clock_gating = &radeon_atom_set_clock_gating,
 	.set_surface_reg = r600_set_surface_reg,
drivers/gpu/drm/radeon/radeon_pm.c +2 -0
index f46d574..0602844 100644
@@ -442,6 +442,8 @@ static int radeon_debugfs_pm_info(struct seq_file *m, void *data)
 	seq_printf(m, "default memory clock: %u0 kHz\n", rdev->clock.default_mclk);
 	if (rdev->asic->get_memory_clock)
 		seq_printf(m, "current memory clock: %u0 kHz\n", radeon_get_memory_clock(rdev));
+	if (rdev->asic->get_pcie_lanes)
+		seq_printf(m, "PCIE lanes: %d\n", radeon_get_pcie_lanes(rdev));
 
 	return 0;
 }
Signature · 14 lines
1.6.4.2


------------------------------------------------------------------------------
Download Intel&#174; Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
--
_______________________________________________
Dri-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/dri-devel
#1 lore raw

2010/2/18 Rafał Miłecki <[email protected]>:

Quoted 18 lines

---
V2: use already implemented functions for reading/writing PCIE PORT
V3: fix rreg/wreg typo in macro

Thanks Alex for reviewing!
---
 drivers/gpu/drm/radeon/r300.c        |    5 ++++- drivers/gpu/drm/radeon/radeon.h      |    2 ++ drivers/gpu/drm/radeon/radeon_asic.h |    4 ++-- drivers/gpu/drm/radeon/radeon_pm.c   |    2 ++

 4 files changed, 10 insertions(+), 3 deletions(-)
drivers/gpu/drm/radeon/r300.c +0 -0
index 7e9f956..29ef9d8 100644

drivers/gpu/drm/radeon/r300.c int rv370_get_pcie_lanes(struct radeon_device *rdev) · line 555

Quoted 8 lines
drivers/gpu/drm/radeon/r300.c +4 -1
@@ -549,7 +549,10 @@ int rv370_get_pcie_lanes(struct radeon_device *rdev)
 
        /* FIXME wait for idle */
 
-       link_width_cntl = RREG32_PCIE(RADEON_PCIE_LC_LINK_WIDTH_CNTL);
+       if (rdev->family < CHIP_R600)
+               link_width_cntl = RREG32_PCIE(RADEON_PCIE_LC_LINK_WIDTH_CNTL);
+       else
+               link_width_cntl = RREG32_PCIE_P(RADEON_PCIE_LC_LINK_WIDTH_CNTL);

Since you also have a patch to implement setting the number of lanes, it would be better to create a new r600 version of get_pcie_lanes() since we may need other r6xx+ specific stuff in it.

Alex

Quoted 55 lines
drivers/gpu/drm/radeon/r300.c +0 -0
        switch ((link_width_cntl & RADEON_PCIE_LC_LINK_WIDTH_RD_MASK) >> RADEON_PCIE_LC_LINK_WIDTH_RD_SHIFT) {

       case RADEON_PCIE_LC_LINK_WIDTH_X0:

drivers/gpu/drm/radeon/radeon.h +2 -0
index b110994..ef55955 100644
@@ -1014,6 +1014,8 @@ static inline void r100_mm_wreg(struct radeon_device *rdev, uint32_t reg, uint32
  #define WREG32_MC(reg, v) rdev->mc_wreg(rdev, (reg), (v))
  #define RREG32_PCIE(reg) rv370_pcie_rreg(rdev, (reg))
  #define WREG32_PCIE(reg, v) rv370_pcie_wreg(rdev, (reg), (v))
+#define RREG32_PCIE_P(reg) rdev->pciep_rreg(rdev, (reg))
+#define WREG32_PCIE_P(reg, v) rdev->pciep_wreg(rdev, (reg), (v))
  #define WREG32_P(reg, val, mask)                               \
        do {                                                    \
                uint32_t tmp_ = RREG32(reg);                    \
drivers/gpu/drm/radeon/radeon_asic.h +2 -2
index b7030d7..4572a66 100644
@@ -626,7 +626,7 @@ static struct radeon_asic r600_asic = {
        .set_engine_clock = &radeon_atom_set_engine_clock,
        .get_memory_clock = &radeon_atom_get_memory_clock,
        .set_memory_clock = &radeon_atom_set_memory_clock,
-       .get_pcie_lanes = NULL,
+       .get_pcie_lanes = &rv370_get_pcie_lanes,
        .set_pcie_lanes = NULL,
        .set_clock_gating = NULL,
        .set_surface_reg = r600_set_surface_reg,
@@ -672,7 +672,7 @@ static struct radeon_asic rv770_asic = {
        .set_engine_clock = &radeon_atom_set_engine_clock,
        .get_memory_clock = &radeon_atom_get_memory_clock,
        .set_memory_clock = &radeon_atom_set_memory_clock,
-       .get_pcie_lanes = NULL,
+       .get_pcie_lanes = &rv370_get_pcie_lanes,
        .set_pcie_lanes = NULL,
        .set_clock_gating = &radeon_atom_set_clock_gating,
        .set_surface_reg = r600_set_surface_reg,
drivers/gpu/drm/radeon/radeon_pm.c +2 -1
index f46d574..0602844 100644
@@ -442,6 +442,8 @@ static int radeon_debugfs_pm_info(struct seq_file *m, void *data)
        seq_printf(m, "default memory clock: %u0 kHz\n", rdev->clock.default_mclk);
        if (rdev->asic->get_memory_clock)
                seq_printf(m, "current memory clock: %u0 kHz\n", radeon_get_memory_clock(rdev));
+       if (rdev->asic->get_pcie_lanes)
+               seq_printf(m, "PCIE lanes: %d\n", radeon_get_pcie_lanes(rdev));
 
        return 0;
  }
--

1.6.4.2

Quoted 12 lines

------------------------------------------------------------------------------
Download Intel&#174; Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
--
_______________________________________________
Dri-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/dri-devel

------------------------------------------------------------------------------
Download Intel&#174; Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
--
_______________________________________________
Dri-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/dri-devel

#2 lore raw

W dniu 18 lutego 2010 22:04 użytkownik Alex Deucher

Quoted 22 lines · <[email protected]> napisał:

2010/2/18 Rafał Miłecki <[email protected]>:

Quoted 21 lines

---
V2: use already implemented functions for reading/writing PCIE PORT
V3: fix rreg/wreg typo in macro

Thanks Alex for reviewing!
---
 drivers/gpu/drm/radeon/r300.c        |    5 ++++- drivers/gpu/drm/radeon/radeon.h      |    2 ++ drivers/gpu/drm/radeon/radeon_asic.h |    4 ++-- drivers/gpu/drm/radeon/radeon_pm.c   |    2 ++

 4 files changed, 10 insertions(+), 3 deletions(-)
drivers/gpu/drm/radeon/r300.c +0 -0
index 7e9f956..29ef9d8 100644
@@ -549,7 +549,10 @@ int rv370_get_pcie_lanes(struct radeon_device *rdev)
 
        /* FIXME wait for idle */
 
Quoted 9 lines

drivers/gpu/drm/radeon/r300.c int rv370_get_pcie_lanes(struct radeon_device *rdev) · line 555

Quoted 5 lines
drivers/gpu/drm/radeon/r300.c +4 -1
-       link_width_cntl = RREG32_PCIE(RADEON_PCIE_LC_LINK_WIDTH_CNTL);
+       if (rdev->family < CHIP_R600)
+               link_width_cntl = RREG32_PCIE(RADEON_PCIE_LC_LINK_WIDTH_CNTL);
+       else
+               link_width_cntl = RREG32_PCIE_P(RADEON_PCIE_LC_LINK_WIDTH_CNTL);

Since you also have a patch to implement setting the number of lanes, it would be better to create a new r600 version of get_pcie_lanes() since we may need other r6xx+ specific stuff in it.

If we get to point where r6xx+ code will differ more, I'm ready to implement that. But if that's matter of just one condition, is that worth splitting it into two functions?

Signature · 13 lines
Rafał

------------------------------------------------------------------------------
Download Intel&#174; Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
--
_______________________________________________
Dri-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/dri-devel
#3 lore raw

2010/2/18 Rafał Miłecki <[email protected]>:

Quoted 34 lines

W dniu 18 lutego 2010 22:04 użytkownik Alex Deucher

Quoted 22 lines · <[email protected]> napisał:

2010/2/18 Rafał Miłecki <[email protected]>:

Quoted 21 lines

---
V2: use already implemented functions for reading/writing PCIE PORT
V3: fix rreg/wreg typo in macro

Thanks Alex for reviewing!
---
 drivers/gpu/drm/radeon/r300.c        |    5 ++++- drivers/gpu/drm/radeon/radeon.h      |    2 ++ drivers/gpu/drm/radeon/radeon_asic.h |    4 ++-- drivers/gpu/drm/radeon/radeon_pm.c   |    2 ++

 4 files changed, 10 insertions(+), 3 deletions(-)
drivers/gpu/drm/radeon/r300.c +0 -0
index 7e9f956..29ef9d8 100644
@@ -549,7 +549,10 @@ int rv370_get_pcie_lanes(struct radeon_device *rdev)
 
        /* FIXME wait for idle */
 
Quoted 9 lines

drivers/gpu/drm/radeon/r300.c int rv370_get_pcie_lanes(struct radeon_device *rdev) · line 555

Quoted 5 lines
drivers/gpu/drm/radeon/r300.c +4 -1
-       link_width_cntl = RREG32_PCIE(RADEON_PCIE_LC_LINK_WIDTH_CNTL);
+       if (rdev->family < CHIP_R600)
+               link_width_cntl = RREG32_PCIE(RADEON_PCIE_LC_LINK_WIDTH_CNTL);
+       else
+               link_width_cntl = RREG32_PCIE_P(RADEON_PCIE_LC_LINK_WIDTH_CNTL);

Since you also have a patch to implement setting the number of lanes, it would be better to create a new r600 version of get_pcie_lanes() since we may need other r6xx+ specific stuff in it.

Quoted 3 lines

If we get to point where r6xx+ code will differ more, I'm ready to implement that. But if that's matter of just one condition, is that worth splitting it into two functions?

I guess it's ok for now. I just don't want to get make it too confusing later on. BTW, does it actually work? I found that generally it always reads back 0 regardless of how many lanes are enabled.

Alex

------------------------------------------------------------------------------
Download Intel&#174; Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
--

#4 lore raw

W dniu 18 lutego 2010 23:36 użytkownik Alex Deucher

Quoted 35 lines · <[email protected]> napisał:

2010/2/18 Rafał Miłecki <[email protected]>:

Quoted 34 lines

W dniu 18 lutego 2010 22:04 użytkownik Alex Deucher

Quoted 31 lines · <[email protected]> napisał:

2010/2/18 Rafał Miłecki <[email protected]>:

Quoted 18 lines

---
V2: use already implemented functions for reading/writing PCIE PORT
V3: fix rreg/wreg typo in macro

Thanks Alex for reviewing!
---
 drivers/gpu/drm/radeon/r300.c        |    5 ++++- drivers/gpu/drm/radeon/radeon.h      |    2 ++ drivers/gpu/drm/radeon/radeon_asic.h |    4 ++-- drivers/gpu/drm/radeon/radeon_pm.c   |    2 ++

 4 files changed, 10 insertions(+), 3 deletions(-)
drivers/gpu/drm/radeon/r300.c +0 -0
index 7e9f956..29ef9d8 100644

drivers/gpu/drm/radeon/r300.c int rv370_get_pcie_lanes(struct radeon_device *rdev) · line 555

Quoted 8 lines
drivers/gpu/drm/radeon/r300.c +4 -1
@@ -549,7 +549,10 @@ int rv370_get_pcie_lanes(struct radeon_device *rdev)
 
        /* FIXME wait for idle */
 
-       link_width_cntl = RREG32_PCIE(RADEON_PCIE_LC_LINK_WIDTH_CNTL);
+       if (rdev->family < CHIP_R600)
+               link_width_cntl = RREG32_PCIE(RADEON_PCIE_LC_LINK_WIDTH_CNTL);
+       else
+               link_width_cntl = RREG32_PCIE_P(RADEON_PCIE_LC_LINK_WIDTH_CNTL);

Since you also have a patch to implement setting the number of lanes, it would be better to create a new r600 version of get_pcie_lanes() since we may need other r6xx+ specific stuff in it.

Quoted 8 lines
Quoted 3 lines

If we get to point where r6xx+ code will differ more, I'm ready to implement that. But if that's matter of just one condition, is that worth splitting it into two functions?

I guess it's ok for now.  I just don't want to get make it too confusing later on.  BTW, does it actually work?  I found that generally it always reads back 0 regardless of how many lanes are enabled.

It reads fine for me... 16 for default :| Not sure about changing to lower values.

Signature · 13 lines
Rafał

------------------------------------------------------------------------------
Download Intel&#174; Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
--
_______________________________________________
Dri-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/dri-devel