Changeset 383

Show
Ignore:
Timestamp:
04/11/99 23:33:26 (10 years ago)
Author:
frodo
Message:

Updated and subject-sorted TODO list

Files:

Legend:

Unmodified
Added
Removed
Modified
Copied
Moved

  • lm-sensors/trunk/TODO

    r370 r383  
    11Many, many things. Most notably: 
    22 
    3 * Sensors program needs to print a + before lm75 temperatures 
    4 * Timer chips - block read problems? 
    5 * Library: reload does not work (Bison/Flex problem?) 
    6 * Create wait-queues for bus locking instead of pure semaphores, to make 
    7   it more interactive 
    8 * Check for block read/writes in smbus-on-i2c emulation code 
    9 * Better bounds checking for writes to /proc; fan_min values now checked 
    10   but other features may need too. 
     3KERNEL MODULES 
     4============== 
     5 
     6* Module documentation!!! 
     7* maxilife.c: Introduce new insmod variables 
     8* maxilife.c: Round REG_TO_VID correctly 
     9* icspll.c: Rewrite. The current implementation simply can't work at all. 
     10  It needs i2c-level access (too bad for SMBus-only adapters). 
     11* Create wait-queues for bus locking instead of pure semaphores in the i2c 
     12  base modules, to make it more interactive (right now one process can  
     13  monopolize the entire adapter) 
     14* Find some soulation for block reads in smbus-on-i2c emulation code 
     15  We need a hack in the i2c level code for this :-( 
    1116* At least the bit-lp and bit-velle modules do no detection on loading; 
    1217  ask Simon whether this is possible to add. 
    13 * kmalloc with GFP_KERNEL can cause a context switch! Check whether this is 
    14   safe everywhere it is used. 
    1518* Support 10-bit addresses. At this moment, they are supported nowhere, except 
    1619  in Simon Vogl's i2c modules. 
    …… 
    2225  done for the i2c modules, too. 
    2326* Better general locking, or at least a once-over to check no race-conditions 
    24   remain. 
    25 * Make lm78.c detect 'double hits', (same chip connected to both SMBus and 
    26   ISA). 
    27 * Better chip detection; insmod-time paramters to set addresses  
    28   (difficult, because of all i2c busses!). 
    29 * At least one case report that unused port addresses do not return 0xff 
     27  remain. This is part of the SMP-safeness, and can better be done at once. 
    3028* Register which i2c addresses are used (a la ISA) for each adapter. 
    31 * Rename i2c-proc.c to i2c-user.c, and extend it with /dev entries; 
    32   or, create a better i2c-dev.c, that understands SMBus commands. 
    33 * Write all other drivers 
    34 * How to make drivers detect thing *not*; for example, if a Winbond is  
    35   present, the LM75 driver should not try to access its simulated 
    36   LM75 chips. 
    37 * Some library routines are pretty inefficient right now 
    3829* Test SIS5595(/91/98) sensor support. Write generic i2c support.  
    3930  Thomas Dickel <Thomas.Dickel@mni.fh-giessen.de> owns one SIS5595. 
    4031* Make GL518SM rev 00 voltage iteration run as a kernel thread. 
    4132  Add libsensors support for the new readable registers. 
    42 * Sensors, sensors.conf.eg: add ADM1021 and ADM9240 support. 
     33* adm9240: check whether the current voltage computations are correct. 
     34  Probably not, as they are different from the datasheet specifications. 
     35  Also check for the supported dallas chip. 
     36 
     37LIBRARY 
     38======= 
     39 
     40* reload does not work (Bison/Flex problem?). 
     41* Some library routines are pretty inefficient right now. 
     42* Library should be split in more separate files, for better linking 
     43  behaviour. 
     44* Many chips are not yet supported by the library. 
     45 
     46PROGRAMS 
     47======== 
     48 
     49* Sensors program needs to print a + before lm75 temperatures 
     50* Sensors program needs to be extended for many new chips