summaryrefslogtreecommitdiffstats
path: root/arch/x86/kernel
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@linux-foundation.org>2017-07-03 17:27:42 -0700
committerLinus Torvalds <torvalds@linux-foundation.org>2017-07-03 17:27:42 -0700
commit8c073517a992124e385040c0da0df809abfa8f61 (patch)
treeb0b3f2c4c6cf96ca358d58d8376d4b5e8e89eba1 /arch/x86/kernel
parent03ffbcdd7898c0b5299efeb9f18de927487ec1cf (diff)
parentdf65c1bcd9b7b639177a5a15da1b8dc3bee4f5fa (diff)
downloadop-kernel-dev-8c073517a992124e385040c0da0df809abfa8f61.zip
op-kernel-dev-8c073517a992124e385040c0da0df809abfa8f61.tar.gz
Merge branch 'x86-platform-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 PCI updates from Thomas Gleixner: "This update provides the seperation of x86 PCI accessors from the global PCI lock in the generic PCI config space accessors. The reasons for this are: - x86 has it's own PCI config lock for various reasons, so the accessors have to lock two locks nested. - The ECAM (mmconfig) access to the extended configuration space does not require locking. The existing generic locking causes a massive lock contention when accessing the extended config space of the Uncore facility for performance monitoring. The commit which switched the access to the primary config space over to ECAM mode has been removed from the branch, so the primary config space is still accessed with type1 accessors properly serialized by the x86 internal locking. Bjorn agreed on merging this through the x86 tree" * 'x86-platform-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: x86/PCI: Select CONFIG_PCI_LOCKLESS_CONFIG PCI: Provide Kconfig option for lockless config space accessors x86/PCI/ce4100: Properly lock accessor functions x86/PCI: Abort if legacy init fails x86/PCI: Remove duplicate defines
Diffstat (limited to 'arch/x86/kernel')
0 files changed, 0 insertions, 0 deletions
OpenPOWER on IntegriCloud