v3dv: avoid TFU reading unmapped pages beyond the end of the buffers

TFU units is doing a readahead of 64 bytes. This is causing invalid read
MMU errors that can be observed at the nightly full Vulkan runs on
Broadcom devices.

04:13:59.969: [   85.623205] v3d 1002000000.v3d: MMU error from client TLB (3) at 0x4869000, pte invalid
04:14:05.408: [   91.019321] v3d 1002000000.v3d: MMU error from client TLB (3) at 0x5209000, pte invalid
04:14:05.413: [   91.031662] v3d 1002000000.v3d: MMU error from client TLB (3) at 0x7521000, pte invalid

Although the log reports the TLB the real culprit is the TFU. A fix
to the kernel was submitted to fix AXI ID on V3D 4.2 and 7.1

So doing an over-allocation of 64-bytes at v3dv_AllocateMemory is
the simplest method to make these MMU errors itp disapear.

Running ./deqp-vk for an hour, we can see that ~%40 of allocations
would need an extra page (4096 bytes) to accomodate this 64 bytes
padding.

Fixes: ca330f7f04 ("v3dv: implement VK_EXT_memory_budget")
Reviewed-by: Iago Toral Quiroga <itoral@igalia.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/34475>

src/broadcom/vulkan/v3dv_device.c

@@ -2208,9 +2208,13 @@ v3dv_AllocateMemory(VkDevice _device,
    assert(pAllocateInfo->sType == VK_STRUCTURE_TYPE_MEMORY_ALLOCATE_INFO);
 
    /* We always allocate device memory in multiples of a page, so round up
-    * requested size to that.
+    * requested size to that. We need to add a V3D_TFU_READHAEAD padding to
+    * avoid invalid reads done by the TFU unit after the end of the last page
+    * allocated.
     */
-   const VkDeviceSize alloc_size = align64(pAllocateInfo->allocationSize, 4096);
+
+   const VkDeviceSize alloc_size = align64(pAllocateInfo->allocationSize +
+                                           V3D_TFU_READAHEAD_SIZE, 4096);
 
    if (unlikely(alloc_size > MAX_MEMORY_ALLOCATION_SIZE))
       return vk_error(device, VK_ERROR_OUT_OF_DEVICE_MEMORY);

src/broadcom/vulkan/v3dv_private.h

@@ -591,6 +591,11 @@ struct v3dv_device {
    struct util_dynarray device_address_bo_list; /* Array of struct v3dv_bo * */
 };
 
+/* TFU has readhead of 64 bytes. So to avoid the unit reading unmaped memory
+ * it is needed to overallocate buffers that could be read by the TFU
+ */
+#define V3D_TFU_READAHEAD_SIZE 64
+
 struct v3dv_device_memory {
    struct vk_device_memory vk;