I met a gpu addr bug recently and the kernel log tells me the pc is memcpy/memset and link register is radeon_uvd_resume.
As we know, in some architectures, optimized memcpy/memset may not work well on device memory. Trival memcpy_toio/memset_io can fix this problem.
BTW, amdgpu has already done it in: commit ba0b2275a678 ("drm/amdgpu: use memcpy_to/fromio for UVD fw upload"), that's why it has no this issue on the same gpu and platform.
Also fix some coding issues reported from sparse.
Signed-off-by: Chen Li chenli@uniontech.com --- drivers/gpu/drm/radeon/radeon_uvd.c | 30 ++++++++++++++++------------- 1 file changed, 17 insertions(+), 13 deletions(-)
diff --git a/drivers/gpu/drm/radeon/radeon_uvd.c b/drivers/gpu/drm/radeon/radeon_uvd.c index dfa9fdbe98da..0d6a5cfa2abf 100644 --- a/drivers/gpu/drm/radeon/radeon_uvd.c +++ b/drivers/gpu/drm/radeon/radeon_uvd.c @@ -152,9 +152,11 @@ int radeon_uvd_init(struct radeon_device *rdev)
rdev->uvd.fw_header_present = true;
- family_id = le32_to_cpu(hdr->ucode_version) & 0xff; - version_major = (le32_to_cpu(hdr->ucode_version) >> 24) & 0xff; - version_minor = (le32_to_cpu(hdr->ucode_version) >> 8) & 0xff; + family_id = (__force u32)(hdr->ucode_version) & 0xff; + version_major = (le32_to_cpu((__force __le32)(hdr->ucode_version)) + >> 24) & 0xff; + version_minor = (le32_to_cpu((__force __le32)(hdr->ucode_version)) + >> 8) & 0xff; DRM_INFO("Found UVD firmware Version: %u.%u Family ID: %u\n", version_major, version_minor, family_id);
@@ -286,7 +288,9 @@ int radeon_uvd_resume(struct radeon_device *rdev) if (rdev->uvd.vcpu_bo == NULL) return -EINVAL;
- memcpy(rdev->uvd.cpu_addr, rdev->uvd_fw->data, rdev->uvd_fw->size); + memcpy_toio((void __iomem *)rdev->uvd.cpu_addr, + rdev->uvd_fw->data, + le32_to_cpu((__force __le32)rdev->uvd_fw->size));
size = radeon_bo_size(rdev->uvd.vcpu_bo); size -= rdev->uvd_fw->size; @@ -294,7 +298,7 @@ int radeon_uvd_resume(struct radeon_device *rdev) ptr = rdev->uvd.cpu_addr; ptr += rdev->uvd_fw->size;
- memset(ptr, 0, size); + memset_io((void __iomem *)ptr, 0, size);
return 0; } @@ -791,17 +795,17 @@ int radeon_uvd_get_create_msg(struct radeon_device *rdev, int ring, return r;
/* stitch together an UVD create msg */ - writel(cpu_to_le32(0x00000de4), &msg[0]); + writel((__force u32)cpu_to_le32(0x00000de4), &msg[0]); writel(0x0, (void __iomem *)&msg[1]); - writel(cpu_to_le32(handle), &msg[2]); + writel((__force u32)cpu_to_le32(handle), &msg[2]); writel(0x0, &msg[3]); writel(0x0, &msg[4]); writel(0x0, &msg[5]); writel(0x0, &msg[6]); - writel(cpu_to_le32(0x00000780), &msg[7]); - writel(cpu_to_le32(0x00000440), &msg[8]); + writel((__force u32)cpu_to_le32(0x00000780), &msg[7]); + writel((__force u32)cpu_to_le32(0x00000440), &msg[8]); writel(0x0, &msg[9]); - writel(cpu_to_le32(0x01b37000), &msg[10]); + writel((__force u32)cpu_to_le32(0x01b37000), &msg[10]); for (i = 11; i < 1024; ++i) writel(0x0, &msg[i]);
@@ -827,9 +831,9 @@ int radeon_uvd_get_destroy_msg(struct radeon_device *rdev, int ring, return r;
/* stitch together an UVD destroy msg */ - writel(cpu_to_le32(0x00000de4), &msg[0]); - writel(cpu_to_le32(0x00000002), &msg[1]); - writel(cpu_to_le32(handle), &msg[2]); + writel((__force u32)cpu_to_le32(0x00000de4), &msg[0]); + writel((__force u32)cpu_to_le32(0x00000002), &msg[1]); + writel((__force u32)cpu_to_le32(handle), &msg[2]); writel(0x0, &msg[3]); for (i = 4; i < 1024; ++i) writel(0x0, &msg[i]);