[haiku-commits] Re: BRANCH pdziepak-github.lock_elision [cd86e1f] headers/private/system src/system/libroot/os/arch/x86_64 src/system/libroot/os/arch/x86 src/add-ons/kernel/drivers src/system/libroot/os/locks

From: Ingo Weinhold <ingo_weinhold@xxxxxx>
To: haiku-commits@xxxxxxxxxxxxx
Date: Fri, 20 Sep 2013 21:00:53 +0200

On 09/20/2013 04:15 AM, pdziepak-github.lock_elision wrote:

cd86e1f: kernel: Add lock elision to kernel mutexes

I would argue the same as for the userland locking primitives. Enablinglock elision generically can hurt performance. So it would be better tointroduce respective locking functions to explicitly request it.


CU, Ingo

Follow-Ups:
- [haiku-commits] Re: BRANCH pdziepak-github.lock_elision [cd86e1f] headers/private/system src/system/libroot/os/arch/x86_64 src/system/libroot/os/arch/x86 src/add-ons/kernel/drivers src/system/libroot/os/locks
  - From: Pawel Dziepak

References:
- [haiku-commits] BRANCH pdziepak-github.lock_elision [cd86e1f] headers/private/system src/system/libroot/os/arch/x86_64 src/system/libroot/os/arch/x86 src/add-ons/kernel/drivers src/system/libroot/os/locks
  - From: pdziepak-github . lock_elision

[haiku-commits] Re: BRANCH pdziepak-github.lock_elision [cd86e1f] headers/private/system src/system/libroot/os/arch/x86_64 src/system/libroot/os/arch/x86 src/add-ons/kernel/drivers src/system/libroot/os/locks

Other related posts: