* [PATCH] gpu: drm: radeon: r600: Replace mdelay() and udelay() with msleep() and usleep_range()
@ 2018-08-04 0:12 Jia-Ju Bai
0 siblings, 0 replies; only message in thread
From: Jia-Ju Bai @ 2018-08-04 0:12 UTC (permalink / raw)
To: alexander.deucher, christian.koenig, David1.Zhou, airlied
Cc: amd-gfx, dri-devel, linux-kernel, Jia-Ju Bai
r600_gpu_soft_reset() and r600_gpu_pci_config_reset() are never
called in atomic context.
They call mdelay() and udelay() to busily wait, which is not necessary.
mdelay() and udelay() can be replaced with msleep() and usleep_range().
This is found by a static analysis tool named DCNS written by myself.
Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com>
---
drivers/gpu/drm/radeon/r600.c | 16 ++++++++--------
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index e06e2d8feab3..de5f6d9f251e 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -1705,7 +1705,7 @@ static void r600_gpu_soft_reset(struct radeon_device *rdev, u32 reset_mask)
WREG32(DMA_RB_CNTL, tmp);
}
- mdelay(50);
+ msleep(50);
rv515_mc_stop(rdev, &save);
if (r600_mc_wait_for_idle(rdev)) {
@@ -1782,7 +1782,7 @@ static void r600_gpu_soft_reset(struct radeon_device *rdev, u32 reset_mask)
WREG32(R_008020_GRBM_SOFT_RESET, tmp);
tmp = RREG32(R_008020_GRBM_SOFT_RESET);
- udelay(50);
+ usleep_range(50, 100);
tmp &= ~grbm_soft_reset;
WREG32(R_008020_GRBM_SOFT_RESET, tmp);
@@ -1796,7 +1796,7 @@ static void r600_gpu_soft_reset(struct radeon_device *rdev, u32 reset_mask)
WREG32(SRBM_SOFT_RESET, tmp);
tmp = RREG32(SRBM_SOFT_RESET);
- udelay(50);
+ usleep_range(50, 100);
tmp &= ~srbm_soft_reset;
WREG32(SRBM_SOFT_RESET, tmp);
@@ -1804,10 +1804,10 @@ static void r600_gpu_soft_reset(struct radeon_device *rdev, u32 reset_mask)
}
/* Wait a little for things to settle down */
- mdelay(1);
+ usleep_range(1000, 2000);
rv515_mc_resume(rdev, &save);
- udelay(50);
+ usleep_range(50, 100);
r600_print_gpu_status_regs(rdev);
}
@@ -1835,7 +1835,7 @@ static void r600_gpu_pci_config_reset(struct radeon_device *rdev)
tmp &= ~DMA_RB_ENABLE;
WREG32(DMA_RB_CNTL, tmp);
- mdelay(50);
+ msleep(50);
/* set mclk/sclk to bypass */
if (rdev->family >= CHIP_RV770)
@@ -1857,12 +1857,12 @@ static void r600_gpu_pci_config_reset(struct radeon_device *rdev)
/* reset */
radeon_pci_config_reset(rdev);
- mdelay(1);
+ usleep_range(1000, 2000);
/* BIF reset workaround. Not sure if this is needed on 6xx */
tmp = SOFT_RESET_BIF;
WREG32(SRBM_SOFT_RESET, tmp);
- mdelay(1);
+ usleep_range(1000, 2000);
WREG32(SRBM_SOFT_RESET, 0);
/* wait for asic to come out of reset */
--
2.17.0
^ permalink raw reply related [flat|nested] only message in thread
only message in thread, other threads:[~2018-08-04 0:15 UTC | newest]
Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-08-04 0:12 [PATCH] gpu: drm: radeon: r600: Replace mdelay() and udelay() with msleep() and usleep_range() Jia-Ju Bai
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).