| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | Introduce pci_probe_device_hook(pci_chipset_tag_t, struct pci_attach_args *). |   miod | 2011-10-10 | 1 | -1/+8 |
| * | Some devices, like parts of the NVIDIA C51 chipset, have a broken Capabilities |   kettenis | 2011-06-12 | 1 | -4/+7 |
| * | Enable MSI for HyperTransport devices that have MSI remapping enabled. | kettenis | 2011-05-30 | 1 | -3/+57 |
| * | Save and restore the MSI capability registers on suspend/resume. | kettenis | 2011-05-19 | 1 | -4/+41 |
| * | Pass flags down the PCI bus hierarchy. Soon to be used to selectively enable | kettenis | 2011-05-14 | 1 | -2/+4 |
| * | We need to delay 10msec after changing to/from D3 state. note: There are |   deraadt | 2011-04-12 | 1 | -2/+7 |
| * | Only disable I/O and memory space and bus mastering if we're actually going | kettenis | 2010-12-30 | 1 | -14/+16 |
| * | Check the register offset against the device's configuration space size in | miod | 2010-12-04 | 1 | -3/+11 |
| * | remove the powerhook code. All architectures now use the ca_activate tree | deraadt | 2010-09-07 | 1 | -17/+1 |
| * | All PWR_{SUSPEND,RESUME} can now be replaced by DVACT_{SUSPEND,RESUME} | deraadt | 2010-09-06 | 1 | -3/+3 |
| * | Add an ioctl to figure out the non-sticky bits in PCI BARs used for determining | kettenis | 2010-09-05 | 1 | -4/+67 |
| * | Add DVACT_QUIECE support. This is called before splhigh() and before | deraadt | 2010-08-31 | 1 | -1/+4 |
| * | Make PCI Power Management optional, and only enable it when acpi(4) attaches. | kettenis | 2010-08-27 | 1 | -32/+72 |
| * | Power management for PCI devices. For now just put everything in the D3 | kettenis | 2010-08-17 | 1 | -25/+55 |
| * | Proper fix for uninitialized variable warning. | kettenis | 2010-07-02 | 1 | -5/+5 |
| * | the variables work better when initialized. ok deraadt |   tedu | 2010-07-02 | 1 | -5/+5 |
| * | Preserve and restore BARs on suspend/resume for all devices, not just |   mlarkin | 2010-06-30 | 1 | -4/+1 |
| * | Add support for mapping ACPI to PCI devices |   jordan | 2010-06-29 | 1 | -1/+3 |
| * | let this compile on RAMDISKS, which have no ioctl | deraadt | 2010-04-21 | 1 | -3/+2 |
| * | Directly comparing pcitag_t's doesn't work on all platforms so decompose the | kettenis | 2010-04-21 | 1 | -2/+7 |
| * | First stab at a VGA arbiter. The VGA arbiter makes sure that only a single | kettenis | 2010-04-21 | 1 | -2/+176 |
| * | backout last, it causes mem conflicts and device trouble; |   otto | 2010-01-26 | 1 | -18/+7 |
| * | Don't whack conflicting BARs if they are enabled. | kettenis | 2010-01-13 | 1 | -7/+18 |
| * | pci should return the result of it's children's suspend/resume calls back | mlarkin | 2009-11-23 | 1 | -3/+3 |
| * | Walk our children in the activate function. | deraadt | 2009-11-23 | 1 | -4/+6 |
| * | PCI suspend/resume plumbing. This code is currently not called. | mlarkin | 2009-11-23 | 1 | -2/+19 |
| * | Add a new PCIOCGETROMLEN ioctl, to give users of PCIOGETROM a chance to | kettenis | 2009-11-05 | 1 | -2/+11 |
| * | Add PCIOCGETROM, and ioctl(2) to read PCI ROMs. | kettenis | 2009-07-26 | 1 | -13/+80 |
| * | Better parameter validation in pciioctl(): check for ioctl number and | miod | 2009-07-25 | 1 | -13/+17 |
| * | Don't forget to read the high bits I/O range register when registering | miod | 2009-07-18 | 1 | -1/+4 |
| * | Kill structs and ioctls that were never implemented. | kettenis | 2009-07-14 | 1 | -5/+1 |
| * | Whack bridges that have conflicting windows. This prevents us from choosing | kettenis | 2009-06-29 | 1 | -5/+15 |
| * | Whack conflicting BARs, such that they get assigned a non-conflicting value | kettenis | 2009-06-03 | 1 | -3/+9 |
| * | Add extents for prefetchable memory mapped I/O. Should remove conflict | kettenis | 2009-05-05 | 1 | -3/+17 |
| * | Set a first step on the road towards proper accounting of PCI resources by | kettenis | 2009-04-06 | 1 | -1/+114 |
| * | Remove confusion between "raw" and "swizzled" interrupt pins. Remove some | kettenis | 2008-12-28 | 1 | -12/+1 |
| * | properly indent something horridly misleading | deraadt | 2008-06-13 | 1 | -17/+17 |
| * | Initialize the return value in pci_probe_device() function, so that we do | deraadt | 2008-06-13 | 1 | -2/+2 |
| * | Make sure we match the correct pci domain; clean up the code a bit and remove | kettenis | 2008-02-27 | 1 | -13/+9 |
| * | On some machines we get hotplug interrupts when they're resumed, so don't | kettenis | 2007-12-31 | 1 | -2/+22 |
| * | Make pci(4) detachable. | kettenis | 2007-11-30 | 1 | -2/+9 |
| * | Add PCI VPD read/write functions from NetBSD required for some things |   jsg | 2007-11-26 | 1 | -1/+78 |
| * | Allow pci_get_capability to decode capabilities for PCI bridges. |   brad | 2007-11-17 | 1 | -2/+3 |
| * | Add pci_set_powerstate() to change the pm state for a given device. |   fgsch | 2007-10-22 | 1 | -1/+18 |
| * | Implement deep interrupt swizzling by mapping all four PCI interrupt pins | kettenis | 2007-05-21 | 1 | -1/+3 |
| * | Make pci subsystem aware of domains. Each host bridge gets assigned a unique | kettenis | 2006-12-14 | 1 | -2/+11 |
| * | On i386, swizzle interrupts for devices for which we don't have explicit | kettenis | 2006-12-11 | 1 | -1/+2 |
| * | Do not assume that the pci(4) unit number matches the actual PCI bus number. | kettenis | 2006-10-04 | 1 | -9/+9 |
| * | Do not blindly enable io and mem space for all matched PCI devices. This is | kettenis | 2006-07-04 | 1 | -20/+2 |
| * | Add pci_find_device() a helper function for enumerating the | jsg | 2006-03-22 | 1 | -1/+19 |
