summaryrefslogtreecommitdiff
path: root/sys/dev/pci/hifn7751.c
AgeCommit message (Expand)Author
2001-04-06typo in error messageJason Wright
2001-03-28Allow tdbi's to appear in mbufs throughout the stack; this allowsAngelos D. Keromytis
2001-02-18shorten messageTheo de Raadt
2001-01-31before copying the packet header, make sure we actually got the mbufJason Wright
2000-11-22even bigger delay in ram probe; deraadt@ okMichael Shalayeff
2000-11-17*HMAC96->*HMACAngelos D. Keromytis
2000-10-26shadow the dmaier to avoid an additional pci read in the interrupt pathJason Wright
2000-10-24Duh, C_WAIT will always be set if the card is inactive, but that interruptJason Wright
2000-10-23unreset takes time too, and you notice this on a p3/933Theo de Raadt
2000-10-13delays that permit more machines to work; itojunTheo de Raadt
2000-10-11printf format string typo in HIFN_DEBUG section.Jun-ichiro itojun Hagino
2000-08-15- deallocate all resources grabbed during hifn_attach() when an error occurs.Jason Wright
2000-08-13indentTheo de Raadt
2000-07-28Use the correct bits for determining which session to delete. Some style fixes.Jason Wright
2000-06-20call crypto_done()Jason Wright
2000-06-17knfJason Wright
2000-06-13use the source length to calculate the output length instead of relying onJason Wright
2000-06-13Be more careful when setting up the mbufs for destination buffersJason Wright
2000-06-02callback is always staticTheo de Raadt
2000-05-04fixed context memory sizingJason Wright
2000-04-25if we have to allocate a new mbuf, don't forget to copy the header info fromJason Wright
2000-04-25fixing the frontend means fixing the backend too... add a countdown whenJason Wright
2000-04-25use the size of the allocation, not the minimum of what was requested andJason Wright
2000-04-23freesession takes u_int64_tAngelos D. Keromytis
2000-04-19split out mbuf scatter gather functionTheo de Raadt
2000-04-13- let hifn_process() compute the mask bits for the command descriptorsJason Wright
2000-04-13- complete rewrite of hifn_write_command() to avoid copies by removing theJason Wright
2000-04-13match netsec 7751 card, and oh golly gee, will ya look at that, it'sTheo de Raadt
2000-04-11- add support for buffers that are not aligned (or end on a non 4byte boundary)Jason Wright
2000-04-11- Don't bother with PCI_COMMAND_IO_ENABLE, the card doesn't support it.Jason Wright
2000-04-10- Add back code from the original aeon driver to handle missed interrupts byJason Wright
2000-04-05run at splnet() instead of splimp()Jason Wright
2000-04-05maintain state about keys, upload them to context ram once (and avoid theJason Wright
2000-04-04add mac result pointer to the callback routineJason Wright
2000-03-31Remove the timeout and print_stats functions (debugging code)Jason Wright
2000-03-30document hook functionsJason Wright
2000-03-30instrument bad crp's and bad crp callbacks, tooJason Wright
2000-03-30add some instrumentation for tracking input (commands/bytes) and outputJason Wright
2000-03-29- update copyrightsJason Wright
2000-03-29remove round-robin use of the boards. We're depending on the upper layerJason Wright
2000-03-29Long over due checkpoint:Jason Wright
2000-03-22fix arguments to failure printf's (correct numbers and types)Jason Wright
2000-03-22Oops... catch up with variable renaming in header fileJason Wright
2000-03-18cap the maximum sessions at 2048 since that's all that is available inJason Wright
2000-03-17Be sure to mask out the chip stepping when examining the chip levelJason Wright
2000-03-17Fix some space->\tJason Wright
2000-03-16document that other vendors use the hifn pci id tooTheo de Raadt
2000-03-16move aeon to hifn7751Theo de Raadt
2000-03-15Bring in bit definitions from the datasheet and allow shared interrupts.Jason Wright
2000-03-10repair ramcfg whacking code; mickeyTheo de Raadt
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-26 11:24:54 +0000