Changeset 220 in xtideuniversalbios


Ignore:
Timestamp:
Jan 25, 2012, 7:13:21 AM (13 years ago)
Author:
gregli@…
google:author:
gregli@hotmail.com
Message:

Serial port code, removed pointer re(normalization) code as the pointer is normalized before we are called and the sector count is kept in the range where we will not need to renormalize.

File:
1 edited

Legend:

Unmodified
Added
Removed
  • trunk/XTIDE_Universal_BIOS/Src/Device/Serial/SerialCommand.asm

    r216 r220  
    211211        pop     es
    212212
     213%if 0
     214;;; no longer needed, since the pointer is normalized before we are called and we do not support
     215;;; more than 128 sectors (and for 128 specifically, the pointer must be segment aligned).
     216;;; See comments below at the point this entry point was called for more details...
    213217.nextSectorNormalize:           
    214 
    215         call    Registers_NormalizeESDI
     218        call    Registers_NormalizeESDI
     219%endif
    216220       
    217221        pop     ax              ; load command byte (done before call to .nextSector on subsequent iterations)
     
    329333        out     dx,al           ; ACK with next sector number
    330334
     335%if 0
     336;;; This code is no longer needed as we do not support more than 128 sectors, and for 128 the pointer
     337;;; must be segment aligned.  If we ever do want to support more sectors, the code can help...
     338       
    331339;
    332340; Normalize buffer pointer for next go round, if needed.
     
    371379        jl      short .nextSector       ; OF<>SF, branch for all cases but 7f, fe, and ff
    372380
    373 %if 0
    374         jpo     short .nextSector       ; if we really wanted to avoid normalizing for ff, this
     381;       jpo     short .nextSector       ; if we really wanted to avoid normalizing for ff, this
    375382                                        ; is one way to do it, but it adds more memory and more
    376383                                        ; cycles for the 7f and fe cases.  IMHO, given that I've
    377384                                        ; never seen a request for more than 7f, this seems unnecessary.
     385
     386        jmp     short .nextSectorNormalize  ; our two renormalization cases (plus one for ff)
     387
     388%else
     389       
     390        jmp     short .nextSector       
     391       
    378392%endif
    379 
    380         jmp     short .nextSectorNormalize  ; our two renormalization cases (plus one for ff)
    381393
    382394;---------------------------------------------------------------------------
Note: See TracChangeset for help on using the changeset viewer.