|
From: | Avi Kivity |
Subject: | Re: [Qemu-devel] [RFC v1] Add declarations for hierarchical memory region API |
Date: | Fri, 20 May 2011 12:26:20 +0300 |
User-agent: | Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.17) Gecko/20110428 Fedora/3.1.10-1.fc14 Thunderbird/3.1.10 |
On 05/20/2011 12:04 AM, Stefan Weil wrote:
Am 19.05.2011 16:12, schrieb Avi Kivity:The memory API separates the attributes of a memory region (its size, how reads or writes are handled, dirty logging, and coalescing) from where itis mapped and whether it is enabled. This allows a device to configurea memory region once, then hand it off to its parent bus to map it accordingto the bus configuration.Hierarchical registration also allows a device to compose a region out of a number of sub-regions with different properties; for example some may beRAM while others may be MMIO.
--- /dev/null +++ b/memory.h @@ -0,0 +1,142 @@ +#ifndef MEMORY_H +#define MEMORY_H + +#include <stdint.h> +#include <stdbool.h>stdbool.h is already included in qemu-common.h, stdint.h (indirectly) too. Therefore both include statements can be removed.
We shouldn't rely on indirect includes, it makes updating headers very hard. Each header should #include what it directly needs and no more.
+typedef struct CoalescedMemoryRange CoalescedMemoryRange; + +struct CoalescedMemoryRange { + target_phys_addr_t start; + target_phys_addr_t size; + QTAILQ_ENTRY(coalesced_ranges) link; +}; + +struct MemoryRegion { + /* All fields are private - violators will be prosecuted */Is it possible to move this private declaration into the implementation file (or a private header file if the declaration is needed by more than one file)?
No, the structure size is needed by clients. -- I have a truly marvellous patch that fixes the bug which this signature is too narrow to contain.
[Prev in Thread] | Current Thread | [Next in Thread] |