| Message ID | 20180525155030.3667352-2-arnd@arndb.de (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Fri, May 25, 2018 at 6:50 PM, Arnd Bergmann <arnd@arndb.de> wrote: > Casting a pointer to a 64-bit type causes a warning on 32-bit targets: > > drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c:473:24: error: cast from pointer to integer of different size [-Werror=pointer-to-int-cast] > lower_32_bits((uint64_t)wptr)); > ^ > drivers/gpu/drm/amd/amdgpu/amdgpu.h:1701:53: note: in definition of macro 'WREG32' > #define WREG32(reg, v) amdgpu_mm_wreg(adev, (reg), (v), 0) > ^ > drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c:473:10: note: in expansion of macro 'lower_32_bits' > lower_32_bits((uint64_t)wptr)); > ^~~~~~~~~~~~~ > > The correct method is to cast to 'uintptr_t'. > > Fixes: d5a114a6c5f7 ("drm/amdgpu: Add GFXv9 kfd2kgd interface functions") > Signed-off-by: Arnd Bergmann <arnd@arndb.de> > --- > drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c > index 8f37991df61b..f0c0d3953f69 100644 > --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c > +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c > @@ -470,9 +470,9 @@ static int kgd_hqd_load(struct kgd_dev *kgd, void *mqd, uint32_t pipe_id, > WREG32(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_WPTR_HI), > upper_32_bits(guessed_wptr)); > WREG32(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_WPTR_POLL_ADDR), > - lower_32_bits((uint64_t)wptr)); > + lower_32_bits((uintptr_t)wptr)); > WREG32(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_WPTR_POLL_ADDR_HI), > - upper_32_bits((uint64_t)wptr)); > + upper_32_bits((uintptr_t)wptr)); > WREG32(SOC15_REG_OFFSET(GC, 0, mmCP_PQ_WPTR_POLL_CNTL1), > get_queue_mask(adev, pipe_id, queue_id)); > } > -- > 2.9.0 > There is a change scheduled for the next merge window that will cause this file to not build anymore on 32-bit targets (because the amdkfd driver is not supported on 32 bit targets). Never the less I'm taking the patch for good measure. This patch is: Reviewed-by: Oded Gabbay <oded.gabbay@gmail.com>
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c index 8f37991df61b..f0c0d3953f69 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c @@ -470,9 +470,9 @@ static int kgd_hqd_load(struct kgd_dev *kgd, void *mqd, uint32_t pipe_id, WREG32(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_WPTR_HI), upper_32_bits(guessed_wptr)); WREG32(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_WPTR_POLL_ADDR), - lower_32_bits((uint64_t)wptr)); + lower_32_bits((uintptr_t)wptr)); WREG32(SOC15_REG_OFFSET(GC, 0, mmCP_HQD_PQ_WPTR_POLL_ADDR_HI), - upper_32_bits((uint64_t)wptr)); + upper_32_bits((uintptr_t)wptr)); WREG32(SOC15_REG_OFFSET(GC, 0, mmCP_PQ_WPTR_POLL_CNTL1), get_queue_mask(adev, pipe_id, queue_id)); }
Casting a pointer to a 64-bit type causes a warning on 32-bit targets: drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c:473:24: error: cast from pointer to integer of different size [-Werror=pointer-to-int-cast] lower_32_bits((uint64_t)wptr)); ^ drivers/gpu/drm/amd/amdgpu/amdgpu.h:1701:53: note: in definition of macro 'WREG32' #define WREG32(reg, v) amdgpu_mm_wreg(adev, (reg), (v), 0) ^ drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c:473:10: note: in expansion of macro 'lower_32_bits' lower_32_bits((uint64_t)wptr)); ^~~~~~~~~~~~~ The correct method is to cast to 'uintptr_t'. Fixes: d5a114a6c5f7 ("drm/amdgpu: Add GFXv9 kfd2kgd interface functions") Signed-off-by: Arnd Bergmann <arnd@arndb.de> --- drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gfx_v9.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)