radeon: Deinline indirect register accessor functions
This patch deinlines indirect register accessor functions.
These functions perform two mmio accesses, framed by spin lock/unlock.
Spin lock/unlock by itself takes more than 50 cycles in ideal case
(if lock is exclusively cached on current CPU).
With this .config: http://busybox.net/~vda/kernel_config,
after uninlining these functions have sizes and callsite counts
as follows:
r600_uvd_ctx_rreg: 111 bytes, 4 callsites
r600_uvd_ctx_wreg: 113 bytes, 5 callsites
eg_pif_phy0_rreg: 106 bytes, 13 callsites
eg_pif_phy0_wreg: 108 bytes, 13 callsites
eg_pif_phy1_rreg: 107 bytes, 13 callsites
eg_pif_phy1_wreg: 108 bytes, 13 callsites
rv370_pcie_rreg: 111 bytes, 21 callsites
rv370_pcie_wreg: 113 bytes, 24 callsites
r600_rcu_rreg: 111 bytes, 16 callsites
r600_rcu_wreg: 113 bytes, 25 callsites
cik_didt_rreg: 106 bytes, 10 callsites
cik_didt_wreg: 107 bytes, 10 callsites
tn_smc_rreg: 106 bytes, 126 callsites
tn_smc_wreg: 107 bytes, 116 callsites
eg_cg_rreg: 107 bytes, 20 callsites
eg_cg_wreg: 108 bytes, 52 callsites
Functions r100_mm_rreg() and r100_mm_rreg() have a fast path and
a locked (slow) path. This patch deinlines only slow path.
r100_mm_rreg_slow: 78 bytes, 2083 callsites
r100_mm_wreg_slow: 81 bytes, 3570 callsites
Reduction in code size is more than 65,000 bytes:
text data bss dec hex filename
85740176 22294680 20627456 128662312 7ab3b28 vmlinux.before
85674192 22294776 20627456 128598664 7aa4288 vmlinux
Signed-off-by: Denys Vlasenko <dvlasenk@redhat.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index 05e6d6e..5397bed 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -35,6 +35,75 @@
#include "evergreen_blit_shaders.h"
#include "radeon_ucode.h"
+/*
+ * Indirect registers accessor
+ */
+u32 eg_cg_rreg(struct radeon_device *rdev, u32 reg)
+{
+ unsigned long flags;
+ u32 r;
+
+ spin_lock_irqsave(&rdev->cg_idx_lock, flags);
+ WREG32(EVERGREEN_CG_IND_ADDR, ((reg) & 0xffff));
+ r = RREG32(EVERGREEN_CG_IND_DATA);
+ spin_unlock_irqrestore(&rdev->cg_idx_lock, flags);
+ return r;
+}
+
+void eg_cg_wreg(struct radeon_device *rdev, u32 reg, u32 v)
+{
+ unsigned long flags;
+
+ spin_lock_irqsave(&rdev->cg_idx_lock, flags);
+ WREG32(EVERGREEN_CG_IND_ADDR, ((reg) & 0xffff));
+ WREG32(EVERGREEN_CG_IND_DATA, (v));
+ spin_unlock_irqrestore(&rdev->cg_idx_lock, flags);
+}
+
+u32 eg_pif_phy0_rreg(struct radeon_device *rdev, u32 reg)
+{
+ unsigned long flags;
+ u32 r;
+
+ spin_lock_irqsave(&rdev->pif_idx_lock, flags);
+ WREG32(EVERGREEN_PIF_PHY0_INDEX, ((reg) & 0xffff));
+ r = RREG32(EVERGREEN_PIF_PHY0_DATA);
+ spin_unlock_irqrestore(&rdev->pif_idx_lock, flags);
+ return r;
+}
+
+void eg_pif_phy0_wreg(struct radeon_device *rdev, u32 reg, u32 v)
+{
+ unsigned long flags;
+
+ spin_lock_irqsave(&rdev->pif_idx_lock, flags);
+ WREG32(EVERGREEN_PIF_PHY0_INDEX, ((reg) & 0xffff));
+ WREG32(EVERGREEN_PIF_PHY0_DATA, (v));
+ spin_unlock_irqrestore(&rdev->pif_idx_lock, flags);
+}
+
+u32 eg_pif_phy1_rreg(struct radeon_device *rdev, u32 reg)
+{
+ unsigned long flags;
+ u32 r;
+
+ spin_lock_irqsave(&rdev->pif_idx_lock, flags);
+ WREG32(EVERGREEN_PIF_PHY1_INDEX, ((reg) & 0xffff));
+ r = RREG32(EVERGREEN_PIF_PHY1_DATA);
+ spin_unlock_irqrestore(&rdev->pif_idx_lock, flags);
+ return r;
+}
+
+void eg_pif_phy1_wreg(struct radeon_device *rdev, u32 reg, u32 v)
+{
+ unsigned long flags;
+
+ spin_lock_irqsave(&rdev->pif_idx_lock, flags);
+ WREG32(EVERGREEN_PIF_PHY1_INDEX, ((reg) & 0xffff));
+ WREG32(EVERGREEN_PIF_PHY1_DATA, (v));
+ spin_unlock_irqrestore(&rdev->pif_idx_lock, flags);
+}
+
static const u32 crtc_offsets[6] =
{
EVERGREEN_CRTC0_REGISTER_OFFSET,