Syslinux - Apr 2011 - [syslinux:lwip] PXE Cleanups, allow for 128 open files

On Mon, Apr 25, 2011 at 11:51:44AM -0700, syslinux-bot for H. Peter Anvin
wrote:
> Commit-ID:  9ca79e92e60dd243c60e257a2dc60ba7b3677f37
> Gitweb:    
http://syslinux.zytor.com/commit/9ca79e92e60dd243c60e257a2dc60ba7b3677f37
> 
> PXE Cleanups, allow for 128 open files
> 
    <snip/>
> --- a/core/include/fs.h
> +++ b/core/include/fs.h
> @@ -11,12 +11,9 @@
>  #include "disk.h"
>  
>  /*
> - * Maximum number of open files.  This is *currently* constrained by the
> - * fact that PXE needs to be able to fit all its packet buffers into a
> - * 64K segment; this should be fixed by moving the packet buffers to high
> - * memory.
> + * Maximum number of open files.
>   */
> -#define MAX_OPEN_LG2	5
> +#define MAX_OPEN_LG2	7
>  #define MAX_OPEN	(1 << MAX_OPEN_LG2)
>  
>  #define FILENAME_MAX_LG2 8
> 
> --- a/core/fs/pxe/pxe.h
> +++ b/core/fs/pxe/pxe.h
> @@ -21,27 +21,15 @@
>  #define PXE_H
>  
>  #include <syslinux/pxe_api.h>
> -#include "fs.h"			/* For MAX_OPEN, should go away */
> +#include "fs.h"			/* Mostly for FILENAME_MAX */
FWIW, I would expect
  #include "fs.h"			/* Mostly for MAX_OPEN */


Groeten
Geert Stappers
-- 
> And is there a policy on top-posting vs. bottom-posting?
Yes.

On 04/25/2011 09:05 PM, Geert Stappers wrote:
>> --- a/core/fs/pxe/pxe.h
>> +++ b/core/fs/pxe/pxe.h
>> @@ -21,27 +21,15 @@
>>  #define PXE_H
>>  
>>  #include <syslinux/pxe_api.h>
>> -#include "fs.h"			/* For MAX_OPEN, should go away */
>> +#include "fs.h"			/* Mostly for FILENAME_MAX */
> 
> FWIW, I would expect
>   #include "fs.h"			/* Mostly for MAX_OPEN */
> 
There are no references to MAX_OPEN anymore, but FILENAME_MAX is used
for some buffer sizes.

	-hpa


-- 
H. Peter Anvin, Intel Open Source Technology Center
I work for Intel.  I don't speak on their behalf.