remoteproc: qcom: pil-info: Fix shift overflow

On platforms with 32-bit phys_addr_t the shift to get the upper word of
the base address of the memory region is invalid. Cast the base to 64
bit to resolv this.

Fixes: 549b67da ("remoteproc: qcom: Introduce helper to store pil info in IMEM")
Tested-by: Nathan Chancellor <natechancellor@gmail.com> # build
Reported-by: Lee Jones <lee.jones@linaro.org>
Reported-by: Nathan Chancellor <natechancellor@gmail.com>
Link: https://lore.kernel.org/r/20200716054817.157608-1-bjorn.andersson@linaro.orgSigned-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>

drivers/remoteproc/qcom_pil_info.c

@@ -108,7 +108,7 @@ int qcom_pil_info_store(const char *image, phys_addr_t base, size_t size)
 found_existing:
 	/* Use two writel() as base is only aligned to 4 bytes on odd entries */
 	writel(base, entry + PIL_RELOC_NAME_LEN);
-	writel(base >> 32, entry + PIL_RELOC_NAME_LEN + 4);
+	writel((u64)base >> 32, entry + PIL_RELOC_NAME_LEN + 4);
 	writel(size, entry + PIL_RELOC_NAME_LEN + sizeof(__le64));
 	mutex_unlock(&pil_reloc_lock);