| Age | Commit message (Collapse) | Author | |
|---|---|---|---|
| 2001-08-17 | Add RC4 support. | ben | |
| 2001-08-15 | - Attempt to recover from PCI aborts. This is a heavy handed approach which | Jason Wright | |
| involves running down what's in the queue and resetting the chip. - remove a DELAY() from reset and just wait for the proc unit to come out of reset. - never set the LAST bit in a JUMP descriptor - use lengths from each map (not just source map) for total_*_length in command [This does not fix problems with net4501, but may address problems with CA810] | |||
| 2001-08-12 | remove redundant vm includes | Michael Shalayeff | |
| 2001-08-11 | move hifn_stats definition to header | Jason Wright | |
| 2001-08-08 | leave the dma engines polling for a bit after the last descriptor has been | Jason Wright | |
| passed through. After a few seconds (5 currently) of no activity, stop them completely. | |||
| 2001-08-08 | Initial support for RC4 operations (This probably won't work yet as | Jason Wright | |
| the caller of *_process() doesn't seem to initialize crd_klen correctly; it has a zero value... still debugging...). | |||
| 2001-07-21 | - Clear all interrupts during initialization (shouldn't be necessary, but | Jason Wright | |
| doesn't hurt). - Turn the dma queues on and off as needed. This results in fewer PCI aborts on 7751. | |||
| 2001-07-20 | Since the aborts seem recoverable (thus far), don't printf() about them, | Jason Wright | |
| just maintain a counter. | |||
| 2001-07-18 | Enable reception of PCI abort interrupts and attempt to handle them. | Jason Wright | |
| 7751 seems to generate these alot and restarting the DMA unit seems to get things going again. 7951, however, never generates these aborts in the same machine/slot. | |||
| 2001-07-16 | oops, remove some debugging stuff | Jason Wright | |
| 2001-07-16 | - Separate out reset function of proc unit | Jason Wright | |
| - Handle ram detection errors more gracefully - If things go awry after enabling DMA, turn off polling - Follow hifn's specs for readram and writeram commands more closely - Add bus_dmamap_sync()'s during ram probing as well | |||
| 2001-07-08 | Hifn has changed its company name, make the name consistent throughout | Brad Smith | |
| the tree. Hi/fn, Hi/Fn and HiFn -> Hifn. -- Ok'd by deraadt@ | |||
| 2001-06-25 | Have to enable PIER as well as DMAIER in order to get interrupts from the | Jason Wright | |
| bignum engine. | |||
| 2001-06-24 | Remove some redundent copies of the total length/skip values and use the | Jason Wright | |
| descriptors passed in instead. | |||
| 2001-06-24 | reflect reality more closely and remove more invertex stuff | Jason Wright | |
| 2001-06-24 | - oops, masking off wrong bits in destination total length field | Jason Wright | |
| - make sure reserved fields are zero'd | |||
| 2001-06-23 | copydata, not copyback | Angelos D. Keromytis | |
| 2001-06-23 | Correctly handle the IV_PRESENT flag. | Angelos D. Keromytis | |
| 2001-06-23 | be sure to clear (by writing 1) the public done by in the status register | Jason Wright | |
| 2001-06-23 | Conform to new prototype for crypto_register() | Angelos D. Keromytis | |
| 2001-06-23 | add some infrastructure for the public engine | Jason Wright | |
| require reworking interrupt enable and status checking as well as a bit of initialization | |||
| 2001-06-23 | merge crypto/crypto{dev,}.h to crypto/cryptodev.h, to avoid name conflicts ↵ | Theo de Raadt | |
| inside OpenSSL codebase | |||
| 2001-06-23 | - put hw defines in *reg, software defines in *var, etc | Jason Wright | |
| - clean up pub/rng initialization call - attempt to remove more intertex vestiges | |||
| 2001-06-22 | Fix lotsa bugs: | Jason Wright | |
| - Hi/Fn length fields in command structures are 18 bits (feature!), but descriptor lengths are 16 bits. - define dmamap maximum lengths correctly (2^18 for total length, 2^16 for segments). - Make the defines more consistent, and add other modes - split source_count in command descriptors into a 16 bit length, and 16bit reserved part upshot: blocks as large as 2^18 - 8 work now for userland crypto | |||
| 2001-06-22 | Add support for RNG on 7951; many thanks to Soren Kristensen | Jason Wright | |
| <soren@soekris.com> for donating the cards. | |||
| 2001-06-14 | hackish auto-IV mode for IOV operations | Theo de Raadt | |
| 2001-06-14 | dst ring corruption fix; jason | Theo de Raadt | |
| 2001-06-14 | KNF | Theo de Raadt | |
| 2001-06-12 | Make pci_mapreg_map take an extra argument where we can | Niklas Hallqvist | |
| put a size limitation of the PCI region to map. That makes the PERC 3/Di raid controller usable, as it publishes too much PCI memory for us to map in the kernel virtual memory. As we only access the first 256 byte it is of no use to map 128MB of kvm. | |||
| 2001-05-14 | use bus_dmamap_sync() as appropriate | Jason Wright | |
| 2001-05-14 | - simplify hifn_dmamap_aligned() | Jason Wright | |
| - simplify hifn_crypto() and remove duplicate work of determining alignment - add ability to share a map for src and dst instead of creating a new one if all of the alignment constraints are met. | |||
| 2001-05-14 | use real uio | Theo de Raadt | |
| 2001-05-13 | First try at bus_dma(9)-ifying hifn7751 driver... advantages: | Jason Wright | |
| - vtophys is dead - now builds on alpha | |||
| 2001-05-13 | missing else | Jason Wright | |
| 2001-05-13 | initial cut at /dev/crypto support. takes original mbuf "try, and discard | Theo de Raadt | |
| if we fail" semantics and extends to two varients of data movement: mbuf, or an iovec style block. | |||
| 2001-05-11 | I think this will match and work on a hifn 7951 | Theo de Raadt | |
| 2001-05-08 | simplify prints | Theo de Raadt | |
| 2001-04-06 | typo in error message | Jason Wright | |
| 2001-03-28 | Allow tdbi's to appear in mbufs throughout the stack; this allows | Angelos D. Keromytis | |
| security properties of the packets to be pushed up to the application (not done yet). Eventually, this will be turned into a packet attributes framework. Make sure tdbi's are free'd/cleared properly whenever drivers (or NFS) does weird things with mbufs. | |||
| 2001-02-18 | shorten message | Theo de Raadt | |
| 2001-01-31 | before copying the packet header, make sure we actually got the mbuf | Jason Wright | |
| 2000-11-22 | even bigger delay in ram probe; deraadt@ ok | Michael Shalayeff | |
| 2000-11-17 | *HMAC96->*HMAC | Angelos D. Keromytis | |
| Bear in mind, you will need to recompile both isakmpd/ipsecadm and your kernel --- otherwise things won't work together. Naturally, all these changes will not be folded into -STABLE, since they would break binary compatibility. | |||
| 2000-10-26 | shadow the dmaier to avoid an additional pci read in the interrupt path | Jason Wright | |
| if mastering isn't enabled, scream document netsec rev A workaround | |||
| 2000-10-24 | Duh, C_WAIT will always be set if the card is inactive, but that interrupt | Jason Wright | |
| isn't always enabled. (CSR & IER) == 0, means not to claim the intr. deraadt ok | |||
| 2000-10-23 | unreset takes time too, and you notice this on a p3/933 | Theo de Raadt | |
| 2000-10-13 | delays that permit more machines to work; itojun | Theo de Raadt | |
| 2000-10-11 | printf format string typo in HIFN_DEBUG section. | Jun-ichiro itojun Hagino | |
| size_t on printf (cast to u_long and use %lu) | |||
| 2000-08-15 | - deallocate all resources grabbed during hifn_attach() when an error occurs. | Jason Wright | |
| - temporary workaround for netsec hifn7751 rev A: half the amount of memory because two of the address lines were left floating (better memory check coming soon). | |||
| 2000-08-13 | indent | Theo de Raadt | |
 |
index : src | |
| OpenBSD base system |
| summaryrefslogtreecommitdiff |