Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit b9cc70e3 authored by Alyssa Rosenzweig's avatar Alyssa Rosenzweig Committed by Greg Kroah-Hartman
Browse files

drm/panfrost: Clamp lock region to Bifrost minimum 



commit bd7ffbc3ca12629aeb66fb9e28cf42b7f37e3e3b upstream.

When locking a region, we currently clamp to a PAGE_SIZE as the minimum
lock region. While this is valid for Midgard, it is invalid for Bifrost,
where the minimum locking size is 8x larger than the 4k page size. Add a
hardware definition for the minimum lock region size (corresponding to
KBASE_LOCK_REGION_MIN_SIZE_LOG2 in kbase) and respect it.

Signed-off-by: default avatarAlyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Tested-by: default avatarChris Morgan <macromorgan@hotmail.com>
Reviewed-by: default avatarSteven Price <steven.price@arm.com>
Reviewed-by: default avatarRob Herring <robh@kernel.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: default avatarSteven Price <steven.price@arm.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210824173028.7528-4-alyssa.rosenzweig@collabora.com


Signed-off-by: default avatarGreg Kroah-Hartman <gregkh@linuxfoundation.org>
parent 9a6c8854

drivers/gpu/drm/panfrost/panfrost_mmu.c

+1 −1
Original line number Diff line number Diff line
@@ -60,7 +60,7 @@ static void lock_region(struct panfrost_device *pfdev, u32 as_nr, 
	/* The size is encoded as ceil(log2) minus(1), which may be calculated

	 * with fls. The size must be clamped to hardware bounds.

	 */

	size = max_t(u64, size, PAGE_SIZE);

	size = max_t(u64, size, AS_LOCK_REGION_MIN_SIZE);

	region_width = fls64(size - 1) - 1;

	region |= region_width;

drivers/gpu/drm/panfrost/panfrost_regs.h

+2 −0
Original line number Diff line number Diff line
@@ -318,6 +318,8 @@ 
#define AS_FAULTSTATUS_ACCESS_TYPE_READ		(0x2 << 8)

#define AS_FAULTSTATUS_ACCESS_TYPE_WRITE	(0x3 << 8)



#define AS_LOCK_REGION_MIN_SIZE                 (1ULL << 15)



#define gpu_write(dev, reg, data) writel(data, dev->iomem + reg)

#define gpu_read(dev, reg) readl(dev->iomem + reg)