summaryrefslogtreecommitdiff
path: root/sys/dev/pci
AgeCommit message (Collapse)Author
2006-11-02whitespaces cleanup, no functional changeFelix Kronlage
ok dlg@
2006-11-01Use correct error message as return value. OK dlg@Claudio Jeker
2006-11-01dont walk the device list looking for our scsibus, it is stashed when weDavid Gwynne
attach it.
2006-11-01whitespace fixesDavid Gwynne
2006-11-01fix firmware fatal errors on re-associations.Damien Bergamini
reported and tested by Marc Winiger and dhartmei@
2006-11-01fix the tx path so it can use more than one scatter gather entry whenDavid Gwynne
sending the packet. this makes it less likely that we'll have to repack fragmented packets for transmission.
2006-11-01syncTheo de Raadt
2006-11-01amd 8111 has an ethernet, but we have never seen anyone use it yetTheo de Raadt
2006-11-01let the tx completion path clean up by recording which mbuf we'reDavid Gwynne
sending in the tx forward path. gotta love one line fixes... committed over vic
2006-11-01vmware frames are slightly short, but that is ok. done with clues fromDavid Gwynne
brad@
2006-11-01fix the barriers in setlladdr.David Gwynne
2006-10-31first cut at sensorsJolan Luff
help from & ok dlg@
2006-10-31we're passing physical addresses to the adapter, so say so. add some magicDavid Gwynne
that lets me see packets coming out of the box.
2006-10-31absolutely huge reworking on this driver (sorry reyk). so far attach,David Gwynne
resource mapping, mem allocation, init, and rx is working. the tx path needs work, but that can happen in tree. overall it's closer to working than the old code. if_vicvar.h goes away since nothing else uses it. im getting this in so other people can play with it too. ok brad@
2006-10-30regenBrad Smith
2006-10-30add PCI ids for newer Intel PRO/1000 chipsets.Brad Smith
2006-10-30regenBrad Smith
2006-10-30fix a typo in the comment above the SNI entries, SNC -> SNI.Brad Smith
From Igor Sobrado <igor at string1 dot ciencias dot uniovi dot es>
2006-10-30removed unused variable.David Gwynne
ok brad@
2006-10-30regenBrad Smith
2006-10-30add an ALI PCI-ISA bridge and an NEC PCIe to PCI-X bridge.Brad Smith
2006-10-28M_TRAILINGSPACE(m) returns 0 if M_READONLY(m) is true,Brad Smith
so no need to call both. From tsutsui@NetBSD
2006-10-27regenBrad Smith
2006-10-27add the NVIDIA GeForce 6100 nForce 430 and some more MCP61 devices.Brad Smith
2006-10-26Set Rx early threshold and Tx DMA thesholds.Brad Smith
From Linux via FreeBSD
2006-10-26Fix this by commiting the proper revision of the bge_encap() diff.Brad Smith
2006-10-26bge_encap():Brad Smith
- Move TX ring full sanity check further up and check the number of DMA segments from the DMA map, instead of counting the DMA segments in the for loop and breaking out later. - Unload the DMA map if encountering an error condition. Tested by brad@ sturm@ wilfried@
2006-10-26do the minimal initialization of the firmware so that ASF alwaysBrad Smith
works. From ambrisko@FreeBSD
2006-10-25replace a few more instances of hand rolled code with theBrad Smith
LIST_FOREACH macro.
2006-10-24remove useless chunks of code that do nothing, since endian adjustmentsBrad Smith
only have to be made for little endian architectures.
2006-10-23- remove the TODO comments at the top.Brad Smith
- add the VLAN headers.
2006-10-23remove detach() function. this is not hotplug and this is dead code.Damien Bergamini
2006-10-23make sure wpi_iter_func() is called at splnet.Damien Bergamini
simplify wpi_dma_contig_alloc(). various cosmetic tweaks while i'm here.
2006-10-23use the Sundance TC9021 ids. this was done so that stge(4) will attachBrad Smith
properly with the sample boards and not have dmesg print unknown product.
2006-10-23regenBrad Smith
2006-10-23add the Sundance TC9021 PCI ids.Brad Smith
2006-10-22recognize the BCM5715 A3 chipset.Brad Smith
2006-10-22regenBrad Smith
2006-10-22add the ASPEED Technology vendor id, AST2000 product id and an NVIDIA MCP55Brad Smith
PCI-ISA bridge.
2006-10-22now with the right revision of this diff which compiles. ok pedro, mglocker.Brad Smith
- Ensure that at least 16 TX descriptors are kept unused in the ring. - Use more complete error handling for TX load problems. From scottl@FreeBSD
2006-10-22replace rssadapt(9) with amrr for automatic rate control.Damien Bergamini
as a side-effect, this should fix all the "bogus xmit rate" panics users have been complaining about for some time when operating in HostAP mode.
2006-10-22Check that ni->ni_rates.rs_nrates is not greater than sizeof rs.rates inDamien Bergamini
iwi_auth_and_assoc() before copying the rate set. The firmware command allows a maximum of 12 rates to be defined while the ieee80211_rateset structure can contain up to 15 rates. Notice that this should not happen since the rate set is supposed to be negotiated at that time but Jeremie Le Hen sees some evidence of this happening in FreeBSD. In case it happens, print a diagnostic message and truncate the rate set. Pointed out by Jeremie Le Hen.
2006-10-21does not compileTheo de Raadt
2006-10-21- Ensure that at least 16 TX descriptors are kept unused in the ring.Brad Smith
- Use more complete error handling for TX load problems. From scottl@FreeBSD
2006-10-21syncDamien Bergamini
2006-10-21Intel PRO/Wireless 3965ABGDamien Bergamini
2006-10-20remove some NetBSD code.Brad Smith
2006-10-20ansiBrad Smith
2006-10-19make the exit label naming scheme match the current function names, removesBrad Smith
a FreeBSD-ism from the original driver.
2006-10-19Overhaul the transmit path:Brad Smith
- Eliminate the bnx_dmamap_arg structure. - Refactor the loop that fills the buffer descriptor so that it can be done with a single set of logic in a single loop instead of two sets of logic. - Eliminate the need to cache and pass descriptor indexes between the start loop and the encap function. - Change the start loop to always check the ifnet sendq for more work. From scottl@FreeBSD
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-26 18:08:13 +0000