Re: [PATCH 4/6] drm/vc4: Fix some failure to track __iomem decorations on pointers.

From: Eric Anholt
Date: Fri Oct 23 2015 - 12:35:08 EST


Ilia Mirkin <imirkin@xxxxxxxxxxxx> writes:

> On Oct 23, 2015 6:32 AM, "Eric Anholt" <eric@xxxxxxxxxx> wrote:
>>
>> Caught by the kbuild test robot.
>>
>> Signed-off-by: Eric Anholt <eric@xxxxxxxxxx>
>> ---
>> drivers/gpu/drm/vc4/vc4_crtc.c | 3 ++-
>> drivers/gpu/drm/vc4/vc4_hvs.c | 8 ++++----
>> 2 files changed, 6 insertions(+), 5 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/vc4/vc4_crtc.c
> b/drivers/gpu/drm/vc4/vc4_crtc.c
>> index a3a77dd..32c0348 100644
>> --- a/drivers/gpu/drm/vc4/vc4_crtc.c
>> +++ b/drivers/gpu/drm/vc4/vc4_crtc.c
>> @@ -401,7 +401,8 @@ static void vc4_crtc_atomic_flush(struct drm_crtc
> *crtc,
>> dlist_next++;
>>
>> HVS_WRITE(SCALER_DISPLISTX(vc4_crtc->channel),
>> - (u32 *)vc4_crtc->dlist - (u32
> *)vc4->hvs->dlist);
>> + (u32 __iomem *)vc4_crtc->dlist -
>> + (u32 __iomem *)vc4->hvs->dlist);
>>
>> /* Make the next display list start after ours. */
>> vc4_crtc->dlist_size -= (dlist_next - vc4_crtc->dlist);
>> diff --git a/drivers/gpu/drm/vc4/vc4_hvs.c b/drivers/gpu/drm/vc4/vc4_hvs.c
>> index ab1673f..4d81d43 100644
>> --- a/drivers/gpu/drm/vc4/vc4_hvs.c
>> +++ b/drivers/gpu/drm/vc4/vc4_hvs.c
>> @@ -75,10 +75,10 @@ void vc4_hvs_dump_state(struct drm_device *dev)
>> for (i = 0; i < 64; i += 4) {
>> DRM_INFO("0x%08x (%s): 0x%08x 0x%08x 0x%08x 0x%08x\n",
>> i * 4, i < HVS_BOOTLOADER_DLIST_END ? "B" : "D",
>> - ((uint32_t *)vc4->hvs->dlist)[i + 0],
>> - ((uint32_t *)vc4->hvs->dlist)[i + 1],
>> - ((uint32_t *)vc4->hvs->dlist)[i + 2],
>> - ((uint32_t *)vc4->hvs->dlist)[i + 3]);
>> + readl((u32 __iomem *)vc4->hvs->dlist + 0),
>> + readl((u32 __iomem *)vc4->hvs->dlist + 1),
>> + readl((u32 __iomem *)vc4->hvs->dlist + 2),
>> + readl((u32 __iomem *)vc4->hvs->dlist + 3));
>
> Looks like you lost the +i here, no?

Indeed. Thanks!

Attachment: signature.asc
Description: PGP signature