summaryrefslogtreecommitdiffstats
path: root/pciusb.c
Commit message (Collapse)AuthorAgeFilesLines
* do full-probe by default; remove support for no full-probeThierry Vignaud2003-08-191-6/+1
|
* special case for buggy 0x0 usb entry so that we eventually got aThierry Vignaud2003-01-061-3/+5
| | | | | class, thus normalizing lspcidrake & harddrake output
* enforce gc coding rulesThierry Vignaud2002-10-071-1/+1
|
* - remove gc hack that tried to fix my bugThierry Vignaud2002-10-071-5/+3
| | | | | | | (which i fixed in last commit) - simplify subids match; explay why we can only test for nb==4
* fix "skipping already handled device" (aka kill stupid remaining test)Thierry Vignaud2002-10-071-4/+2
|
* try to detect correctly the module when subv/subd differ with non-subv/subdGuillaume Cottenceau2002-10-041-0/+3
|
* fix ugly case for snd-usb-audio which should have made titi think thatPascal Rigaux2002-09-051-2/+2
| | | | | | something was broken. Really fixing the right way (this fixes automatic detection of unknown usb controllers)
* usb audio devices can use new alsa modules snd-usb-audioThierry Vignaud2002-08-151-1/+3
|
* - struct pciusb_entrie : add already_found flagThierry Vignaud2002-08-081-0/+9
| | | | | | | | | - pciusb::pciusb_initialize() : zero the above flag: - pciusb::pciusb_find_modules() : use the above flag: o if device subids matches the table ones, swtich the already_found flag o skip already handled device (if already_found==1)
* fix "(null) description" bug: don't skip entries where module hasThierry Vignaud2002-07-251-1/+1
| | | | | already be set by pci.c workaround
* teach titi that !(a && !b) is not (!a && !b) but (!a || b)Pascal Rigaux2002-07-161-1/+1
| | | | | | | (the other solution is to teach him to *test*) (oh, remind me to teach him not to re-indent the whole code until he doesn't make stupid bugs)
* s/zcat/gzip -cd/ back for lord pixouThierry Vignaud2002-07-031-2/+3
|
* ifree(e->text) to prevent mem leak with usbThierry Vignaud2002-07-031-0/+1
|
* - fix pipe() error pathThierry Vignaud2002-07-031-22/+22
| | | | | | - make pciusb_find_modules() normal path clearer: rewrite tests cascade into a single path were exceptions're just "go away"
* - make gzip path in fh_open() be clearerThierry Vignaud2002-07-031-36/+39
| | | | | | | | | | | | | | | | | | | | | | - pciusb_find_modules() optimization / cleanups : - only calc module and description once so that if there're multiples identical cards, we just gives the same result - do less copies : - only copy when setting e->{text,module} - it's useless to do copies on the stack, just play with strndup() instead of strdup() - skip comments in {usb,pci}table - remove if (q) test which is uneeded since we've already parsed the buffer with sscanf() and we've skipped comments - remove uneeded ifree (it's impossible to get valid pointer a this point)
* - no need to allocate fname_gz in fast pathThierry Vignaud2002-07-031-28/+31
| | | | | | | - reduce cmd size - use zcat rather than gzip
* get rid of more uneeded copiesThierry Vignaud2002-07-031-14/+14
|
* indent-regionThierry Vignaud2002-07-031-108/+109
|
* don't do uneeded copies (saves 2% of text size)Thierry Vignaud2002-07-031-7/+7
|
* ensure the header file are C++ compliant (do not use "class" for struct ↵Pascal Rigaux2002-06-101-1/+1
| | | | field name)
* fix pb when 2 similar cards are therePascal Rigaux2001-07-031-1/+4
|
* close fdno's of the pipe which are unused or dup2'edPascal Rigaux2001-04-121-0/+2
|
* *** empty log message ***Francois Pons2001-04-111-13/+32
|
* fix some memory leak a few segfaultsPascal Rigaux2001-03-291-2/+3
|
* nasty C, fclose on popen'ed gets a segfault, in /some/ cases :-(Pascal Rigaux2001-03-241-18/+41
|
* handle gzip'ed pcitable/usbtablePascal Rigaux2001-03-231-5/+15
|
* if subid can be useful, use them. use probe_type==-1 to ensure no subid is usedPascal Rigaux2001-03-211-7/+22
|
* added pci(bus,device,function) for DrakX.Francois Pons2001-03-151-0/+3
|
* add some fclose where it shouldPascal Rigaux2001-02-061-0/+1
|
* now detect usbPascal Rigaux2000-12-161-0/+64