From: "Christian König" <christian.koenig@amd.com>
To: Denys Vlasenko <dvlasenk@redhat.com>
Cc: Alex Deucher <alexander.deucher@amd.com>, <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] radeon: Deinline indirect register accessor functions
Date: Mon, 18 May 2015 20:06:22 +0200 [thread overview]
Message-ID: <555A2A1E.9000706@amd.com> (raw)
In-Reply-To: <1431971955-31231-1-git-send-email-dvlasenk@redhat.com>
I'm actually surprised how often people come along with that. The last
time we tried this it caused a noticeable performance drop.
Basic problem is that this line:
> + if ((reg < rdev->rmmio_size || reg < RADEON_MIN_MMIO_SIZE) && !always_indirect)
optimizes away in most of the cases which reduces the call to a readl
which is way faster than the spinlock path.
So this is a NAK,
Christian.
On 18.05.2015 19:59, Denys Vlasenko wrote:
> This patch deinlines indirect register accessor functions.
>
> These functions perform two mmio accesses, framed by spinlock/unlock.
> Spin lock/unlock by itself takes more than 50 cycles in ideal case.
>
> 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
> r100_mm_rreg: 112 bytes, 2083 callsites
> r100_mm_wreg: 116 bytes, 3570 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
>
> Reduction in code size is more than 80,000 bytes:
>
> text data bss dec hex filename
> 85740176 22294680 20627456 128662312 7ab3b28 vmlinux.before
> 85657104 22294872 20627456 128579432 7a9f768 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
> ---
> drivers/gpu/drm/radeon/r100.c | 34 +++++
> drivers/gpu/drm/radeon/radeon.h | 229 +++------------------------------
> drivers/gpu/drm/radeon/radeon_device.c | 179 ++++++++++++++++++++++++++
> 3 files changed, 233 insertions(+), 209 deletions(-)
>
> diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c
> index 04f2514..95868c7 100644
> --- a/drivers/gpu/drm/radeon/r100.c
> +++ b/drivers/gpu/drm/radeon/r100.c
> @@ -4090,6 +4090,40 @@ int r100_init(struct radeon_device *rdev)
> return 0;
> }
>
> +uint32_t r100_mm_rreg(struct radeon_device *rdev, uint32_t reg,
> + bool always_indirect)
> +{
> + /* The mmio size is 64kb at minimum. Allows the if to be optimized out. */
> + if ((reg < rdev->rmmio_size || reg < RADEON_MIN_MMIO_SIZE) && !always_indirect)
> + return readl(((void __iomem *)rdev->rmmio) + reg);
> + else {
> + unsigned long flags;
> + uint32_t ret;
> +
> + spin_lock_irqsave(&rdev->mmio_idx_lock, flags);
> + writel(reg, ((void __iomem *)rdev->rmmio) + RADEON_MM_INDEX);
> + ret = readl(((void __iomem *)rdev->rmmio) + RADEON_MM_DATA);
> + spin_unlock_irqrestore(&rdev->mmio_idx_lock, flags);
> +
> + return ret;
> + }
> +}
> +
> +void r100_mm_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v,
> + bool always_indirect)
> +{
> + if ((reg < rdev->rmmio_size || reg < RADEON_MIN_MMIO_SIZE) && !always_indirect)
> + writel(v, ((void __iomem *)rdev->rmmio) + reg);
> + else {
> + unsigned long flags;
> +
> + spin_lock_irqsave(&rdev->mmio_idx_lock, flags);
> + writel(reg, ((void __iomem *)rdev->rmmio) + RADEON_MM_INDEX);
> + writel(v, ((void __iomem *)rdev->rmmio) + RADEON_MM_DATA);
> + spin_unlock_irqrestore(&rdev->mmio_idx_lock, flags);
> + }
> +}
> +
> u32 r100_io_rreg(struct radeon_device *rdev, u32 reg)
> {
> if (reg < rdev->rio_mem_size)
> diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h
> index 5587603..bb6b25c 100644
> --- a/drivers/gpu/drm/radeon/radeon.h
> +++ b/drivers/gpu/drm/radeon/radeon.h
> @@ -2465,39 +2465,10 @@ int radeon_gpu_wait_for_idle(struct radeon_device *rdev);
>
> #define RADEON_MIN_MMIO_SIZE 0x10000
>
> -static inline uint32_t r100_mm_rreg(struct radeon_device *rdev, uint32_t reg,
> - bool always_indirect)
> -{
> - /* The mmio size is 64kb at minimum. Allows the if to be optimized out. */
> - if ((reg < rdev->rmmio_size || reg < RADEON_MIN_MMIO_SIZE) && !always_indirect)
> - return readl(((void __iomem *)rdev->rmmio) + reg);
> - else {
> - unsigned long flags;
> - uint32_t ret;
> -
> - spin_lock_irqsave(&rdev->mmio_idx_lock, flags);
> - writel(reg, ((void __iomem *)rdev->rmmio) + RADEON_MM_INDEX);
> - ret = readl(((void __iomem *)rdev->rmmio) + RADEON_MM_DATA);
> - spin_unlock_irqrestore(&rdev->mmio_idx_lock, flags);
> -
> - return ret;
> - }
> -}
> -
> -static inline void r100_mm_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v,
> - bool always_indirect)
> -{
> - if ((reg < rdev->rmmio_size || reg < RADEON_MIN_MMIO_SIZE) && !always_indirect)
> - writel(v, ((void __iomem *)rdev->rmmio) + reg);
> - else {
> - unsigned long flags;
> -
> - spin_lock_irqsave(&rdev->mmio_idx_lock, flags);
> - writel(reg, ((void __iomem *)rdev->rmmio) + RADEON_MM_INDEX);
> - writel(v, ((void __iomem *)rdev->rmmio) + RADEON_MM_DATA);
> - spin_unlock_irqrestore(&rdev->mmio_idx_lock, flags);
> - }
> -}
> +uint32_t r100_mm_rreg(struct radeon_device *rdev, uint32_t reg,
> + bool always_indirect);
> +void r100_mm_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v,
> + bool always_indirect);
>
> u32 r100_io_rreg(struct radeon_device *rdev, u32 reg);
> void r100_io_wreg(struct radeon_device *rdev, u32 reg, u32 v);
> @@ -2582,182 +2553,22 @@ static inline struct radeon_fence *to_radeon_fence(struct fence *f)
> /*
> * Indirect registers accessor
> */
> -static inline uint32_t rv370_pcie_rreg(struct radeon_device *rdev, uint32_t reg)
> -{
> - unsigned long flags;
> - uint32_t r;
> -
> - spin_lock_irqsave(&rdev->pcie_idx_lock, flags);
> - WREG32(RADEON_PCIE_INDEX, ((reg) & rdev->pcie_reg_mask));
> - r = RREG32(RADEON_PCIE_DATA);
> - spin_unlock_irqrestore(&rdev->pcie_idx_lock, flags);
> - return r;
> -}
> -
> -static inline void rv370_pcie_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v)
> -{
> - unsigned long flags;
> -
> - spin_lock_irqsave(&rdev->pcie_idx_lock, flags);
> - WREG32(RADEON_PCIE_INDEX, ((reg) & rdev->pcie_reg_mask));
> - WREG32(RADEON_PCIE_DATA, (v));
> - spin_unlock_irqrestore(&rdev->pcie_idx_lock, flags);
> -}
> -
> -static inline u32 tn_smc_rreg(struct radeon_device *rdev, u32 reg)
> -{
> - unsigned long flags;
> - u32 r;
> -
> - spin_lock_irqsave(&rdev->smc_idx_lock, flags);
> - WREG32(TN_SMC_IND_INDEX_0, (reg));
> - r = RREG32(TN_SMC_IND_DATA_0);
> - spin_unlock_irqrestore(&rdev->smc_idx_lock, flags);
> - return r;
> -}
> -
> -static inline void tn_smc_wreg(struct radeon_device *rdev, u32 reg, u32 v)
> -{
> - unsigned long flags;
> -
> - spin_lock_irqsave(&rdev->smc_idx_lock, flags);
> - WREG32(TN_SMC_IND_INDEX_0, (reg));
> - WREG32(TN_SMC_IND_DATA_0, (v));
> - spin_unlock_irqrestore(&rdev->smc_idx_lock, flags);
> -}
> -
> -static inline 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;
> -}
> -
> -static inline 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);
> -}
> -
> -static inline 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;
> -}
> -
> -static inline 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);
> -}
> -
> -static inline 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;
> -}
> -
> -static inline 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);
> -}
> -
> -static inline 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;
> -}
> -
> -static inline 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 inline 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;
> -}
> -
> -static inline 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);
> -}
> -
> -
> -static inline u32 cik_didt_rreg(struct radeon_device *rdev, u32 reg)
> -{
> - unsigned long flags;
> - u32 r;
> -
> - spin_lock_irqsave(&rdev->didt_idx_lock, flags);
> - WREG32(CIK_DIDT_IND_INDEX, (reg));
> - r = RREG32(CIK_DIDT_IND_DATA);
> - spin_unlock_irqrestore(&rdev->didt_idx_lock, flags);
> - return r;
> -}
> -
> -static inline void cik_didt_wreg(struct radeon_device *rdev, u32 reg, u32 v)
> -{
> - unsigned long flags;
> -
> - spin_lock_irqsave(&rdev->didt_idx_lock, flags);
> - WREG32(CIK_DIDT_IND_INDEX, (reg));
> - WREG32(CIK_DIDT_IND_DATA, (v));
> - spin_unlock_irqrestore(&rdev->didt_idx_lock, flags);
> -}
> +uint32_t rv370_pcie_rreg(struct radeon_device *rdev, uint32_t reg);
> +void rv370_pcie_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v);
> +u32 tn_smc_rreg(struct radeon_device *rdev, u32 reg);
> +void tn_smc_wreg(struct radeon_device *rdev, u32 reg, u32 v);
> +u32 r600_rcu_rreg(struct radeon_device *rdev, u32 reg);
> +void r600_rcu_wreg(struct radeon_device *rdev, u32 reg, u32 v);
> +u32 eg_cg_rreg(struct radeon_device *rdev, u32 reg);
> +void eg_cg_wreg(struct radeon_device *rdev, u32 reg, u32 v);
> +u32 eg_pif_phy0_rreg(struct radeon_device *rdev, u32 reg);
> +void eg_pif_phy0_wreg(struct radeon_device *rdev, u32 reg, u32 v);
> +u32 eg_pif_phy1_rreg(struct radeon_device *rdev, u32 reg);
> +void eg_pif_phy1_wreg(struct radeon_device *rdev, u32 reg, u32 v);
> +u32 r600_uvd_ctx_rreg(struct radeon_device *rdev, u32 reg);
> +void r600_uvd_ctx_wreg(struct radeon_device *rdev, u32 reg, u32 v);
> +u32 cik_didt_rreg(struct radeon_device *rdev, u32 reg);
> +void cik_didt_wreg(struct radeon_device *rdev, u32 reg, u32 v);
>
> void r100_pll_errata_after_index(struct radeon_device *rdev);
>
> diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
> index bd7519f..6712505 100644
> --- a/drivers/gpu/drm/radeon/radeon_device.c
> +++ b/drivers/gpu/drm/radeon/radeon_device.c
> @@ -161,6 +161,185 @@ static void radeon_device_handle_px_quirks(struct radeon_device *rdev)
> rdev->flags &= ~RADEON_IS_PX;
> }
>
> +/*
> + * Indirect registers accessor
> + */
> +uint32_t rv370_pcie_rreg(struct radeon_device *rdev, uint32_t reg)
> +{
> + unsigned long flags;
> + uint32_t r;
> +
> + spin_lock_irqsave(&rdev->pcie_idx_lock, flags);
> + WREG32(RADEON_PCIE_INDEX, ((reg) & rdev->pcie_reg_mask));
> + r = RREG32(RADEON_PCIE_DATA);
> + spin_unlock_irqrestore(&rdev->pcie_idx_lock, flags);
> + return r;
> +}
> +
> +void rv370_pcie_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v)
> +{
> + unsigned long flags;
> +
> + spin_lock_irqsave(&rdev->pcie_idx_lock, flags);
> + WREG32(RADEON_PCIE_INDEX, ((reg) & rdev->pcie_reg_mask));
> + WREG32(RADEON_PCIE_DATA, (v));
> + spin_unlock_irqrestore(&rdev->pcie_idx_lock, flags);
> +}
> +
> +u32 tn_smc_rreg(struct radeon_device *rdev, u32 reg)
> +{
> + unsigned long flags;
> + u32 r;
> +
> + spin_lock_irqsave(&rdev->smc_idx_lock, flags);
> + WREG32(TN_SMC_IND_INDEX_0, (reg));
> + r = RREG32(TN_SMC_IND_DATA_0);
> + spin_unlock_irqrestore(&rdev->smc_idx_lock, flags);
> + return r;
> +}
> +
> +void tn_smc_wreg(struct radeon_device *rdev, u32 reg, u32 v)
> +{
> + unsigned long flags;
> +
> + spin_lock_irqsave(&rdev->smc_idx_lock, flags);
> + WREG32(TN_SMC_IND_INDEX_0, (reg));
> + WREG32(TN_SMC_IND_DATA_0, (v));
> + spin_unlock_irqrestore(&rdev->smc_idx_lock, flags);
> +}
> +
> +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 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);
> +}
> +
> +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);
> +}
> +
> +u32 cik_didt_rreg(struct radeon_device *rdev, u32 reg)
> +{
> + unsigned long flags;
> + u32 r;
> +
> + spin_lock_irqsave(&rdev->didt_idx_lock, flags);
> + WREG32(CIK_DIDT_IND_INDEX, (reg));
> + r = RREG32(CIK_DIDT_IND_DATA);
> + spin_unlock_irqrestore(&rdev->didt_idx_lock, flags);
> + return r;
> +}
> +
> +void cik_didt_wreg(struct radeon_device *rdev, u32 reg, u32 v)
> +{
> + unsigned long flags;
> +
> + spin_lock_irqsave(&rdev->didt_idx_lock, flags);
> + WREG32(CIK_DIDT_IND_INDEX, (reg));
> + WREG32(CIK_DIDT_IND_DATA, (v));
> + spin_unlock_irqrestore(&rdev->didt_idx_lock, flags);
> +}
> +
> /**
> * radeon_program_register_sequence - program an array of registers.
> *
next prev parent reply other threads:[~2015-05-18 19:39 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-05-18 17:59 [PATCH] radeon: Deinline indirect register accessor functions Denys Vlasenko
2015-05-18 17:59 ` [PATCH] radeon: Shrink radeon_ring_write() Denys Vlasenko
2015-05-18 18:11 ` Christian König
2015-05-18 18:25 ` Denys Vlasenko
2015-05-18 19:01 ` Christian König
2015-05-18 18:06 ` Christian König [this message]
2015-05-18 18:50 ` [PATCH] radeon: Deinline indirect register accessor functions Denys Vlasenko
2015-05-18 19:04 ` Christian König
2015-05-18 19:22 ` Ilia Mirkin
2015-05-18 19:22 ` Ilia Mirkin
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=555A2A1E.9000706@amd.com \
--to=christian.koenig@amd.com \
--cc=alexander.deucher@amd.com \
--cc=dvlasenk@redhat.com \
--cc=linux-kernel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.