Todo Index
$#! · 0-9 · A · B · C · D · E · F · G · H · I · J · K · L · M · N · O · P · Q · R · S · T · U · V · W · X · Y · Z
$#!
 ), Extend to internal testing(buffer etc
 *
0-9
 1,3-17-3-31, UDI_NIC_MEDIATYPE:: 3::1,3:: 2,3:: 3
 12 point 8, math:: h section 17
 3, remainder of UDI physical IO Ch
 4, remainder of UDI physical IO Ch
 5, remainder of UDI physical IO Ch
 8-10, UDI core 2:: 22:: 2::2 22
A
 AC capability detection
 Add mbstate
 Add support for non-x86
 Add testdrivers after flush/ close/ tmpfile is implemented
 Address space switches
 allocation of kernel pages
 AP detection and bootstrap
 assumes udi_trans_size<=2;
 ATAPI/ SATA checking
 atof(),strtof(),strtod(),strtold
 Autodetect IRQ in use by baseaddr
B
 Balance
 base is not power of two, test”odd”overflow,i:: e
 Brendan Bug#1-no kernel found error on P4 with less-than-perfect FDD
 build struct and move it
C
 c, t:: b
 change base10 values for 80 bit float
 check against limits
 Check all settings for sanity
 check CTS if CS option set and start timer
 check destroyed regs
 check errors
 Check for invalid flag combinations
 check if some idiot put an PCI IDE card in a machine with an onboard ISA IDE controller
 Check open file list,flush and close file if open
 Check what happens when fflush(NULL)encounters write errors,in other libs
 Check what happens when ungetc()is called on a stream at offset 0
 check where the new bits go
 Check whether the stored state is consistent
 clean up
 Clean-up on TIMEOUT
 concatenate fail to end if necessary
 copy existing handles
 create something sensible here
 Cyrix detection
D
 Defined to 1 as multibyte characters are not supported yet
 documentation
 Documentation
 Doing this via a static array is not the way to do it
E
 E_intptr, E_long
 Earlier version was missing tolower()here but was not caught by tests
 enable hInit,atexit and fb_RtExit
 equates to TDT
 error
 error and gamma functions
 error handling
 Error handling
 expand detection for sse3,3dnow,syscall
 expand for multiboot stuff
 Extend testing once setvbuf()is finished
 Extended CPUID
F
 fail stub
 FIFO overflow checks
 FIFO setup on UARTs that support it
 FIFO Support on UARTs that support it
 Figure out which Unixy systems have/ don’t have PTHREAD_MUTEX_RECURSIVE[_NP]
 fill me in later
 find a non-pci Mach64 to test manual config of aperture+sparse portio
 fix cpuid bugs for intel class chips
 fix debug stub
 fix issues with bus alignments,
 fix MMX and EMMI detection
 fix unicode support
 Flags,wide chars
 Floats
 fninit bug in 486 and Pentium I series
 fscanf()documentation
G
 get task entry point and create threads for all modules
H
 hand optimize to assembly
 handle scrolling for internal characters/ attributes buffer?
 HyperThreading
I
 implement me properly
 implement stuff to generate this properly
 Improve, Primitive placeholder
 In other than”C”locale,additional patterns may be defined
 insert opcodes for GLX get*functions
 invalid argument
 It is not nice to do this on a stream we just closed
M
 Macro MB_CUR_MAX
 manipulation functions
 MCR_RTS should be set only after waiting for DSR if the OP/ CS protocol option is set
 min max and difference
 move to correct section
 Much room for improvement, Primitive placeholder
 muladd
N
 nearest int functions
 needs lots of fixes
 needs the exact needed frequency calculated
O
 Only if endptr!=NULL-but do we really want*another*parameter?
 only offset differences), fix matching of seperated buffers(i:: e
 optimize for back-to-front copying
P
 Parse memtable
 power and abs functions
 pre-C99 compilers might require modulus corrections
 ProcessorCOMApatch
 ProcessorSYSCpatch
 proper assert handlers
 proper page flushing
R
 Recurse
 release this address
 remainder functions
 Remove tmpfile()files, Flush and close open streams
 replace with a call to pciserv
 replace with real handlers
 resolve virtual pointers
 rest of 2(if there’s anything)?, UDI_NIC_MEDIATYPE
S
 send queued packet
 Set RTS if RS option not set
 set SSE bits in CR4
 Setting mbstate
 Shouldn’t realloc()split the now much-too-large node?
 Special handling for mode changes on std-streams
 Stack usage is log2(nmemb)(minus what T shaves off the worst case)
 standardize the DEV_*structs,or provide a device hook function to get OS handle
 start autoload
 start DCD low timer if CD option set
 start DSR low timer if DS option set
 Support for ASC/ LF options
 Support for slave PIC?
T
 Test SYSENTER/ SYSEXIT-bochs has no clue
 These tests assume two-complement,but conversion should work
 This can run beyond a malformed format string
 This function can change wide orientation of a stream
 This function should interpret format as multibyte characters
 This is proof-of-concept,requires finetuning
 This is ugly,but keeps caller from negating the error value
 This should be dynamic but ATM gives problems, 32 is guaranteed
 thread status checking
 try alternative methods than E820
Todo: UDI Core 20.*
Todo: UDI Core 30.* Todo: UDI Core 31.* Todo: UDI Core 32.* Todo: UDI Core 33.*
Todo: UDI Core 12.* Todo: UDI Core 14.* Todo: UDI Core 18.* UDI core 1.18.2 18.3
these are the amount of occupied/unoccupied pages they should later be set to the actual pages that are used.
CALL InitializeAPIC ; todo CALL Multiprocessor ; needs fixing
showbyte(asc(“!”
Todo: comments
Marcel Sondaar
no-wait instructions can cause an exception in compatibility mode (NE=0) not sure wether we need this as native exceptions seem to be supported in 386+ (it requires mobo circuitry though, might need testing)
It does not matter with the current implementation of clearerr(), but it might start to matter if someone replaced that implementation.
none
for fixing the cyrix 6x86 coma bug
for allowing SYSCALL/SYSRET
OOPS
Worst-case nmemb is platform dependent and should probably be configured through _PDCLIB_config.h.
with my malloc.