From: Matthias Weisser <weisserm@arcor.de>
To: u-boot@lists.denx.de
Subject: [U-Boot] [[PATCH V2]] sandbox: Add improved RAM simulation
Date: Thu, 03 Nov 2011 18:55:12 +0100 [thread overview]
Message-ID: <4EB2D580.4010405@arcor.de> (raw)
In-Reply-To: <CAPnjgZ3MfWE=xk45132vgLUXDB_4om5UKaB21LTkmK21LSNMVw@mail.gmail.com>
Am 02.11.2011 21:56, schrieb Simon Glass:
> On Wed, Nov 2, 2011 at 1:12 PM, Matthias Weisser <weisserm@arcor.de> wrote:
>> Using mmap to allocate memory from the OS for RAM simulation we can use
>> u-boots own malloc implementation.
>>
>> Signed-off-by: Matthias Weisser <weisserm@arcor.de>
>> ---
>> Changes in V2:
>> Removed the address hint for mmap
>> Removed the special handling of dlmalloc in common
>> Set gd->bd->bi_dram[0].start to 0 again
>>
>> arch/sandbox/cpu/os.c | 7 +++++++
>> arch/sandbox/lib/board.c | 17 ++++++++++-------
>> common/Makefile | 2 --
>> include/os.h | 8 ++++++++
>> 4 files changed, 25 insertions(+), 9 deletions(-)
>>
>> diff --git a/arch/sandbox/cpu/os.c b/arch/sandbox/cpu/os.c
>> index 6c175d4..05f91f2 100644
>> --- a/arch/sandbox/cpu/os.c
>> +++ b/arch/sandbox/cpu/os.c
>> @@ -24,6 +24,7 @@
>> #include <unistd.h>
>> #include <sys/types.h>
>> #include <sys/stat.h>
>> +#include <sys/mman.h>
>>
>> #include <os.h>
>>
>> @@ -53,3 +54,9 @@ void os_exit(int exit_code)
>> {
>> exit(exit_code);
>> }
>> +
>> +void *os_malloc(size_t length)
>> +{
>> + return mmap(NULL, length, PROT_READ | PROT_WRITE,
>> + MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
>> +}
>> diff --git a/arch/sandbox/lib/board.c b/arch/sandbox/lib/board.c
>> index ae5a517..d71cd6f 100644
>> --- a/arch/sandbox/lib/board.c
>> +++ b/arch/sandbox/lib/board.c
>> @@ -45,8 +45,12 @@
>> #include <version.h>
>> #include <serial.h>
>>
>> +#include <os.h>
>> +
>> DECLARE_GLOBAL_DATA_PTR;
>>
>> +static gd_t gd_mem;
>> +
>> /************************************************************************
>> * Init Utilities *
>> ************************************************************************
>> @@ -147,7 +151,7 @@ void board_init_f(ulong bootflag)
>> uchar *mem;
>> unsigned long addr_sp, addr, size;
>>
>> - gd = malloc(sizeof(gd_t));
>> + gd = &gd_mem;
>> assert(gd);
>>
>> memset((void *)gd, 0, sizeof(gd_t));
>> @@ -158,7 +162,8 @@ void board_init_f(ulong bootflag)
>> }
>>
>> size = CONFIG_SYS_SDRAM_SIZE;
>> - mem = malloc(size);
>> + mem = os_malloc(CONFIG_SYS_SDRAM_SIZE);
>> +
>
> Just a suggestion if you like: perhaps also assert that
> CONFIG_SYS_SDRAM_SIZE is greater than TOTAL_MALLOC_LEN. Or assert
> lower down at the end of board_init_f() that addr_sp is still greatrer
> than mem?
I will leave out this point for a separate patch.
>> assert(mem);
>> gd->ram_buf = mem;
>> addr = (ulong)(mem + size);
>> @@ -214,11 +219,9 @@ void board_init_r(gd_t *id, ulong dest_addr)
>> post_output_backlog();
>> #endif
>>
>> -#if 0 /* Sandbox uses system malloc for now */
>> - /* The Malloc area is immediately below the monitor copy in DRAM */
>> - malloc_start = dest_addr - TOTAL_MALLOC_LEN;
>> - mem_malloc_init(malloc_start, TOTAL_MALLOC_LEN);
>> -#endif
>> + /* The Malloc area is at the top of simulated DRAM */
>> + mem_malloc_init(gd->ram_buf + gd->ram_size - TOTAL_MALLOC_LEN,
>> + TOTAL_MALLOC_LEN);
>
> I get:
>
> board.c:224:4: warning: passing argument 1 of ?mem_malloc_init? makes
> integer from pointer without a cast [enabled by default]
Fixed. Will send an updated version.
>>
>> /* initialize environment */
>> env_relocate();
>> diff --git a/common/Makefile b/common/Makefile
>> index 1b672ad..919be42 100644
>> --- a/common/Makefile
>> +++ b/common/Makefile
>> @@ -29,9 +29,7 @@ LIB = $(obj)libcommon.o
>> ifndef CONFIG_SPL_BUILD
>> COBJS-y += main.o
>> COBJS-y += command.o
>> -ifndef CONFIG_SANDBOX
>> COBJS-y += dlmalloc.o
>> -endif
>
> I think this can be removed. It was added unconditionally lower down
> in the Makefile by this commit:
>
> 2d01dd9 omap: spl: fix build break due to changes in FAT
>
> (incidentally this is what has broken sandbox).
Will do.
Matthias
next prev parent reply other threads:[~2011-11-03 17:55 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-11-01 14:07 [U-Boot] [PATCH] sandbox: Add improved RAM simulation Matthias Weisser
2011-11-01 15:28 ` Mike Frysinger
2011-11-01 16:05 ` Matthias Weisser
2011-11-01 16:45 ` Simon Glass
2011-11-01 18:37 ` Matthias Weisser
2011-11-01 18:52 ` Mike Frysinger
2011-11-01 19:01 ` Matthias Weisser
2011-11-01 20:10 ` Mike Frysinger
2011-11-02 18:35 ` Matthias Weisser
2011-11-02 19:20 ` Mike Frysinger
2011-11-01 17:12 ` Mike Frysinger
2011-11-01 16:42 ` Simon Glass
2011-11-02 20:12 ` [U-Boot] [[PATCH V2]] " Matthias Weisser
2011-11-02 20:39 ` Mike Frysinger
2011-11-02 20:56 ` Simon Glass
2011-11-03 17:55 ` Matthias Weisser [this message]
2011-11-03 18:02 ` [U-Boot] [PATCH V3] " Matthias Weisser
2011-11-03 18:05 ` Mike Frysinger
2011-11-03 22:55 ` Simon Glass
2011-11-05 10:40 ` [U-Boot] [PATCH V4] " Matthias Weisser
2011-11-05 15:05 ` Simon Glass
2011-11-05 16:47 ` Mike Frysinger
2011-12-02 17:08 ` Mike Frysinger
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4EB2D580.4010405@arcor.de \
--to=weisserm@arcor.de \
--cc=u-boot@lists.denx.de \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox