summaryrefslogtreecommitdiff
path: root/src/linux_sysfs.c
AgeCommit message (Collapse)AuthorFilesLines
2009-08-06pciaccess: make linux rom reading fallback optionalDave Airlie1-0/+5
On kernels which have ROM files, the boot VGA rom file will contain the 0xc0000 file anyways. If another card is missing a rom file there is no need for this. This is fallback is disabled by default, on the premise that distro shipping this new a pciaccess will be running on kernels with working rom support which is been upstream for a long time
2009-07-16add support for finding if something has a kernel driverDave Airlie1-0/+21
2009-05-01Don't try to write boot_vga since that fails with EACCESAaron Plattner1-1/+1
2009-05-01pciaccess: provide a method to detect if a device is boot VGADave Airlie1-0/+31
When the linux kernel exposes this information, we can use this interface in the X server to detect whether the kernel believes the device we are looking at is the boot VGA device. Signed-off-by: Dave Airlie <airlied@redhat.com>
2008-10-15Fix various typos (mostly in comments)Alan Coopersmith1-2/+2
2008-06-23Support write combine resource files in Linux sysfsJesse Barnes1-0/+39
Starting with version 2.6.26, Linux will support resourceN_wc files which export write combining mappings of PCI resource ranges, so support them if present in libpciaccess.
2008-06-23Fixed typo in output ("performace" --> "performance").Stefan Dirsch1-1/+1
2008-06-05Catch and recover from yet another linux kernel bug in mprotect.Eric Anholt1-5/+20
2008-05-21Linux: Fail gracefully on machines without PCI.Adam Jackson1-4/+3
2008-05-21linux: add pci_device_enable entrypoint and sysfs support for itDave Airlie1-1/+24
2008-04-16Kludge around linux bug and turn off write-through and cache-disable bitsKeith Packard1-0/+3
When mmaping the PCI device, the kernel turns on the write-through and cache-disable bits in the allocated PTEs. This disables write-combining mode and dramatically reduces write bandwidth to the frame buffer. While that should be fixed in the kernel, we'll kludge around it here by using mprotect to rewrite the PTEs and get those bits turned off.
2008-03-14Bug #13988: Fix reads from "0"-sized ROMs.Stuart Bennett1-2/+6
2008-02-27Don't export private symbols.Julien Cristau1-1/+1
2007-08-31Don't add MTRR for uncached regions. Remove MTRR on unmap.Keith Packard1-8/+67
MTRR regions aren't needed for uncached mappings, so don't add them. Also, when unmapping memory, remove the MTRR entry.
2007-08-30New interfaces to map memory with MTRR (or similar) support.Ian Romanick1-62/+67
Added new functions pci_device_map_range and pci_device_unmap_range to handle mapping of PCI device BARs. These new interfaces allow the possiblity of MTRRs on platforms that support them. These additional APIs necessitated changing some internal interfaces. The code for FreeBSD and Solaris has been updated but has not been compiled or tested. Old interfaces are marked deprecated and will eventually be removed.
2007-06-18Initial support for reading expansion ROM via VGA BIOS address.Ian Romanick1-1/+6
2007-03-27Populate device and vendor ID fields at device list creation.Ian Romanick1-22/+52
2007-01-11Fix bad cast that caused bit truncation.Ian Romanick1-1/+1
2006-07-06Files missed (for reasons I still don't understand) on the previousIan Romanick1-0/+4
commit. REALLY add support for querying bridge information. Bump to version 0.5.0.
2006-07-06Fix byte ordering of the PCI class.Ian Romanick1-2/+2
2006-05-22Fix foolish endianess bug. Bump to version 0.4.1.Ian Romanick1-5/+6
2006-04-12Bump to version 0.4.0.Ian Romanick1-5/+25
Add multiple-inclusion protection. Add new function to write masked bits to PCI config space. This mirrors functionality currently available in X.org that is slated to be removed. Gut old regex based search mechanism with a new mechanism that is modeled after the Linux kernel. In addition to searching for devices by device / vendor ID, it is possible to search for devices by their domain / bus / slot / function. Fix serious bus in the reading of ROMs and in the unmapping of regions. The main point is that the map routine depens on the pci_mem_region::memory pointer being non-NULL only when the region is mapped. Therefore, the unmap routine should set it to NULL after unmapping. Update to use new search API.
2006-03-18Initial import of libpciaccess.Ian Romanick1-0/+493