From 8c9d9211e241b97fa67d130436e7346bba28aeec Mon Sep 17 00:00:00 2001 From: Roman Kapl Date: Mon, 30 Oct 2017 11:56:13 +0100 Subject: [PATCH 2095/4131] drm/radeon: fix atombios on big endian The function for byteswapping the data send to/from atombios was buggy for num_bytes not divisible by four. The function must be aware of the fact that after byte-swapping the u32 units, valid bytes might end up after the num_bytes boundary. This patch was tested on kernel 3.12 and allowed us to sucesfully use DisplayPort on and Radeon SI card. Namely it fixed the link training and EDID readout. The function is patched both in radeon and amd drivers, since the functions and the fixes are identical. Signed-off-by: Roman Kapl Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org --- drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c | 26 ++++++++++---------------- 1 file changed, 10 insertions(+), 16 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c index 31a96a3..1f32110 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atombios.c @@ -1778,26 +1778,20 @@ bool amdgpu_atombios_scratch_need_asic_init(struct amdgpu_device *adev) void amdgpu_atombios_copy_swap(u8 *dst, u8 *src, u8 num_bytes, bool to_le) { #ifdef __BIG_ENDIAN - u8 src_tmp[20], dst_tmp[20]; /* used for byteswapping */ - u32 *dst32, *src32; + u32 src_tmp[5], dst_tmp[5]; int i; + u8 align_num_bytes = ALIGN(num_bytes, 4); - memcpy(src_tmp, src, num_bytes); - src32 = (u32 *)src_tmp; - dst32 = (u32 *)dst_tmp; if (to_le) { - for (i = 0; i < ((num_bytes + 3) / 4); i++) - dst32[i] = cpu_to_le32(src32[i]); - memcpy(dst, dst_tmp, num_bytes); + memcpy(src_tmp, src, num_bytes); + for (i = 0; i < align_num_bytes / 4; i++) + dst_tmp[i] = cpu_to_le32(src_tmp[i]); + memcpy(dst, dst_tmp, align_num_bytes); } else { - u8 dws = num_bytes & ~3; - for (i = 0; i < ((num_bytes + 3) / 4); i++) - dst32[i] = le32_to_cpu(src32[i]); - memcpy(dst, dst_tmp, dws); - if (num_bytes % 4) { - for (i = 0; i < (num_bytes % 4); i++) - dst[dws+i] = dst_tmp[dws+i]; - } + memcpy(src_tmp, src, align_num_bytes); + for (i = 0; i < align_num_bytes / 4; i++) + dst_tmp[i] = le32_to_cpu(src_tmp[i]); + memcpy(dst, dst_tmp, num_bytes); } #else memcpy(dst, src, num_bytes); -- 2.7.4