diff options
author | Samuel Thibault <samuel.thibault@ens-lyon.org> | 2018-10-30 01:00:28 +0100 |
---|---|---|
committer | Samuel Thibault <samuel.thibault@ens-lyon.org> | 2018-10-30 01:00:28 +0100 |
commit | 96cbadd47a46d07fd242f7d06869e9b1a54ceb79 (patch) | |
tree | f08e50641920c5d0ffc39886c8c042e6515dbac5 /pci-arbiter | |
parent | a830bef66e689f6f1defa9988f257a7ec2bb034d (diff) | |
download | hurd-96cbadd47a46d07fd242f7d06869e9b1a54ceb79.tar.gz hurd-96cbadd47a46d07fd242f7d06869e9b1a54ceb79.tar.bz2 hurd-96cbadd47a46d07fd242f7d06869e9b1a54ceb79.zip |
pci-arbiter: Add TODO file
* pci-arbiter/TODO: New file.
Diffstat (limited to 'pci-arbiter')
-rw-r--r-- | pci-arbiter/TODO | 15 |
1 files changed, 15 insertions, 0 deletions
diff --git a/pci-arbiter/TODO b/pci-arbiter/TODO new file mode 100644 index 00000000..22ae72a8 --- /dev/null +++ b/pci-arbiter/TODO @@ -0,0 +1,15 @@ +- add "--pci 00:1f.3" option to easily specify a PCI device instead of having to + use --bus, --dev, and --func. + +- memset + snprintf looks dubious + +- pci_get_ndevs should be deprecated, applications shouldn't be relying on this + +- we shouldn't duplicate pci_access.[ch] x86_pci.[ch] from libpciaccess, we + should get libpciaccess to expose pci_system_x86_create() to keep the + maintenance of x86 port knocking there. + + At least one difference with libpciaccess is the refresh operation. Perhaps + we'd need to extend libpciaccess to fix that. + + BTW we could also support libpci. |