sergiotarxz/nuttx
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

arm/up_allocpage: fix warning: "PG_POOL_MAXL1NDX" is not defined

Browse Source 
Signed-off-by: Xiang Xiao <xiaoxiang@xiaomi.com>
This commit is contained in:
Xiang Xiao 2020-04-29 22:13:18 +08:00 committed by patacongo
parent 5b839061f9
commit 1c483d8ed4
1 changed files with 16 additions and 16 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

32
arch/arm/src/arm/up_allocpage.c
View File

@ -70,12 +70,12 @@ typedef uint16_t pgndx_t;
typedef uint32_t pgndx_t;
#endif
#if PG_POOL_MAXL1NDX < 256
typedef uint8_t L1ndx_t;
#elif PG_POOL_MAXL1NDX < 65536
typedef uint16_t L1ndx_t;
#if PG_POOL_MAXL2NDX < 256
typedef uint8_t l2ndx_t;
#elif PG_POOL_MAXL2NDX < 65536
typedef uint16_t l2ndx_t;
#else
typedef uint32_t L1ndx_t;
typedef uint32_t l2ndx_t;
#endif
/****************************************************************************
@ -93,14 +93,14 @@ typedef uint32_t L1ndx_t;
static pgndx_t g_pgndx;
/* After CONFIG_PAGING_NPAGED have been allocated, the pages will be re-used.
* In order to re-used the page, we will have un-map the page from its previous
* mapping. In order to that, we need to be able to map a physical address to
* to an index into the PTE where it was mapped. The following table supports
* this backward lookup - it is indexed by the page number index, and holds
* another index to the mapped virtual page.
* In order to re-used the page, we will have un-map the page from its
* previous mapping. In order to that, we need to be able to map a physical
* address to to an index into the PTE where it was mapped. The following
* table supports this backward lookup - it is indexed by the page number
* index, and holds another index to the mapped virtual page.
*/
static L1ndx_t g_ptemap[CONFIG_PAGING_NPPAGED];
static l2ndx_t g_ptemap[CONFIG_PAGING_NPPAGED];
/* The contents of g_ptemap[] are not valid until g_pgndx has wrapped at
* least one time.
@ -136,11 +136,11 @@ static bool g_pgwrap;
* NOTE 2: If an in-use page is un-mapped, it may be necessary to flush the
* instruction cache in some architectures.
*
* NOTE 3: Allocating and filling a page is a two step process. up_allocpage()
* NOTE 3: Allocate and fill a page is a two step process. up_allocpage()
* allocates the page, and up_fillpage() fills it with data from some non-
* volatile storage device. This distinction is made because up_allocpage()
* can probably be implemented in board-independent logic whereas up_fillpage()
* probably must be implemented as board-specific logic.
* can probably be implemented in board-independent logic whereas
* up_fillpage() probably must be implemented as board-specific logic.
*
* NOTE 4: The initial mapping of vpage should be read-able and write-
* able (but not cached). No special actions will be required of
@ -206,12 +206,12 @@ int up_allocpage(FAR struct tcb_s *tcb, FAR void **vpage)
pte = up_va2pte(oldvaddr);
*pte = 0;
/* Invalidate the instruction TLB corresponding to the virtual address */
/* Invalidate instruction TLB corresponding to the virtual address */
tlb_inst_invalidate_single(oldvaddr);
/* I do not believe that it is necessary to flush the I-Cache in this
* case: The I-Cache uses a virtual address index and, hence, since the
* case: The I-Cache uses a virtual address index and, hence, since the
* NuttX address space is flat, the cached instruction value should be
* correct even if the page mapping is no longer in place.
*/