summaryrefslogtreecommitdiffstats
path: root/arch/x86/pci
diff options
context:
space:
mode:
authorStefano Stabellini <stefano.stabellini@eu.citrix.com>2011-02-02 18:32:59 +0000
committerKonrad Rzeszutek Wilk <konrad.wilk@oracle.com>2011-03-14 11:17:14 -0400
commit706cc9d2a4cb9b03217e15b0bb3d117f4d5109ee (patch)
tree60075b399bdbb0a6f4b9d27bcfc17813e5c72734 /arch/x86/pci
parent146c4e511717e581065800938537b276173d8548 (diff)
downloadop-kernel-dev-706cc9d2a4cb9b03217e15b0bb3d117f4d5109ee.zip
op-kernel-dev-706cc9d2a4cb9b03217e15b0bb3d117f4d5109ee.tar.gz
xen/m2p: Check whether the MFN has IDENTITY_FRAME bit set..
If there is no proper PFN value in the M2P for the MFN (so we get 0xFFFFF.. or 0x55555, or 0x0), we should consult the M2P override to see if there is an entry for this. [Note: we also consult the M2P override if the MFN is past our machine_to_phys size]. We consult the P2M with the PFN. In case the returned MFN is one of the special values: 0xFFF.., 0x5555 (which signify that the MFN can be either "missing" or it belongs to DOMID_IO) or the p2m(m2p(mfn)) != mfn, we check the M2P override. If we fail the M2P override check, we reset the PFN value to INVALID_P2M_ENTRY. Next we try to find the MFN in the P2M using the MFN value (not the PFN value) and if found, we know that this MFN is an identity value and return it as so. Otherwise we have exhausted all the posibilities and we return the PFN, which at this stage can either be a real PFN value found in the machine_to_phys.. array, or INVALID_P2M_ENTRY value. [v1: Added Review-by tag] Reviewed-by: Ian Campbell <ian.campbell@citrix.com> Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Diffstat (limited to 'arch/x86/pci')
0 files changed, 0 insertions, 0 deletions
OpenPOWER on IntegriCloud