forge/linux-pinenote
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

radeon: Deinline indirect register accessor functions

Browse source 
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>
This commit is contained in:
Denys Vlasenko 2015-05-20 13:02:37 +02:00 • committed by Alex Deucher
parent b0b9bb4dd5
commit 9e5acbc213
7 changed files with 241 additions and 197 deletions
Download patch file Download diff file Expand all files Collapse all files

47
drivers/gpu/drm/radeon/r600.c
View file

@ -108,6 +108,53 @@ static void r600_pcie_gen2_enable(struct radeon_device *rdev);
extern int evergreen_rlc_resume(struct radeon_device *rdev);
extern void rv770_set_clk_bypass_mode(struct radeon_device *rdev);
/*
* Indirect registers accessor
*/
u32 r600_rcu_rreg(struct radeon_device *rdev, u32 reg)
{
unsigned long flags;
u32 r;
spin_lock_irqsave(&rdev->rcu_idx_lock, flags);
WREG32(R600_RCU_INDEX, ((reg) & 0x1fff));
r = RREG32(R600_RCU_DATA);
spin_unlock_irqrestore(&rdev->rcu_idx_lock, flags);
return r;
}
void r600_rcu_wreg(struct radeon_device *rdev, u32 reg, u32 v)
{
unsigned long flags;
spin_lock_irqsave(&rdev->rcu_idx_lock, flags);
WREG32(R600_RCU_INDEX, ((reg) & 0x1fff));
WREG32(R600_RCU_DATA, (v));
spin_unlock_irqrestore(&rdev->rcu_idx_lock, flags);
}
u32 r600_uvd_ctx_rreg(struct radeon_device *rdev, u32 reg)
{
unsigned long flags;
u32 r;
spin_lock_irqsave(&rdev->uvd_idx_lock, flags);
WREG32(R600_UVD_CTX_INDEX, ((reg) & 0x1ff));
r = RREG32(R600_UVD_CTX_DATA);
spin_unlock_irqrestore(&rdev->uvd_idx_lock, flags);
return r;
}
void r600_uvd_ctx_wreg(struct radeon_device *rdev, u32 reg, u32 v)
{
unsigned long flags;
spin_lock_irqsave(&rdev->uvd_idx_lock, flags);
WREG32(R600_UVD_CTX_INDEX, ((reg) & 0x1ff));
WREG32(R600_UVD_CTX_DATA, (v));
spin_unlock_irqrestore(&rdev->uvd_idx_lock, flags);
}
/**
* r600_get_allowed_info_register - fetch the register for the info ioctl
*