Konrad Rzeszutek Wilk
2013-Sep-07 13:46 UTC
[PATCH] Bug-fixes to enable PV ticketlock to work under Xen PVHVM with Linux v3.12. (v1)
After a bit of false starts, lots of debugging, and tons of help from Stefano and David on how event mechanism is suppose to work I am happy to present a set of bug-fixes that make PV ticketlocks work under Xen PVHVM with Linux v3.12. v3.12 has now thanks to commit 816434ec4a674fcdb3c2221a6dffdc8f34020550 (Merge branch ''x86-spinlocks-for-linus'' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip) PV ticketlocks. That means: - Xen PV bytelock has been replaced by Xen PV ticketlock. - Xen PVHVM is using ticketlock (not PV variant) - this patch makes it PV. - baremetal is still using ticketlocks. In other words everything in the kernel is ticketlock based with the virtualizations having the ''PV'' part to aid. Please take a look at the patches. If you are interested in testing them out I will post a git tree on Monday based on v3.11 and v3.12-rc0. Note that I had not run any performance tests and I am not sure when I will be able to (got other bugs to squash now). I''ve asked our internal performance engineer to do some benchmarking - but those results won''t be available for some time as it takes time to do good benchmarking. arch/x86/xen/enlighten.c | 1 - arch/x86/xen/smp.c | 17 +++++++++++++++++ arch/x86/xen/spinlock.c | 45 ++++++++------------------------------------- 3 files changed, 25 insertions(+), 38 deletions(-) (oh neat, more deletions!) Konrad Rzeszutek Wilk (5): xen/spinlock: Fix locking path engaging too soon under PVHVM. xen/spinlock: We don''t need the old structure anymore xen/smp: Update pv_lock_ops functions before alternative code starts under PVHVM xen/spinlock: Don''t setup xen spinlock IPI kicker if disabled. Revert "xen/spinlock: Disable IRQ spinlock (PV) allocation on PVHVM"
David Vrabel
2013-Sep-09 10:34 UTC
Re: [PATCH] Bug-fixes to enable PV ticketlock to work under Xen PVHVM with Linux v3.12. (v1)
On 07/09/13 14:46, Konrad Rzeszutek Wilk wrote:> After a bit of false starts, lots of debugging, and tons of help from Stefano and > David on how event mechanism is suppose to work I am happy to present a set > of bug-fixes that make PV ticketlocks work under Xen PVHVM with Linux v3.12. > > v3.12 has now thanks to commit 816434ec4a674fcdb3c2221a6dffdc8f34020550 > (Merge branch ''x86-spinlocks-for-linus'' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip) > PV ticketlocks. That means: > - Xen PV bytelock has been replaced by Xen PV ticketlock. > - Xen PVHVM is using ticketlock (not PV variant) - this patch makes it PV. > - baremetal is still using ticketlocks. > > In other words everything in the kernel is ticketlock based with the virtualizations > having the ''PV'' part to aid. > > Please take a look at the patches. If you are interested in testing them > out I will post a git tree on Monday based on v3.11 and v3.12-rc0. > > Note that I had not run any performance tests and I am not sure when I will > be able to (got other bugs to squash now). I''ve asked our internal performance > engineer to do some benchmarking - but those results won''t be available for > some time as it takes time to do good benchmarking.If there was favourable performance data I would be happy for these to go in 3.12, without I think they''ll have to wait for 3.13. David
Konrad Rzeszutek Wilk
2013-Sep-09 13:12 UTC
Re: [PATCH] Bug-fixes to enable PV ticketlock to work under Xen PVHVM with Linux v3.12. (v1)
On Mon, Sep 09, 2013 at 11:34:42AM +0100, David Vrabel wrote:> On 07/09/13 14:46, Konrad Rzeszutek Wilk wrote: > > After a bit of false starts, lots of debugging, and tons of help from Stefano and > > David on how event mechanism is suppose to work I am happy to present a set > > of bug-fixes that make PV ticketlocks work under Xen PVHVM with Linux v3.12. > > > > v3.12 has now thanks to commit 816434ec4a674fcdb3c2221a6dffdc8f34020550 > > (Merge branch ''x86-spinlocks-for-linus'' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip) > > PV ticketlocks. That means: > > - Xen PV bytelock has been replaced by Xen PV ticketlock. > > - Xen PVHVM is using ticketlock (not PV variant) - this patch makes it PV. > > - baremetal is still using ticketlocks. > > > > In other words everything in the kernel is ticketlock based with the virtualizations > > having the ''PV'' part to aid. > > > > Please take a look at the patches. If you are interested in testing them > > out I will post a git tree on Monday based on v3.11 and v3.12-rc0. > > > > Note that I had not run any performance tests and I am not sure when I will > > be able to (got other bugs to squash now). I''ve asked our internal performance > > engineer to do some benchmarking - but those results won''t be available for > > some time as it takes time to do good benchmarking. > > If there was favourable performance data I would be happy for these to > go in 3.12, without I think they''ll have to wait for 3.13.Then they will have to wait. I am not going to be able to provide those within this month.> > David > > _______________________________________________ > Xen-devel mailing list > Xen-devel@lists.xen.org > http://lists.xen.org/xen-devel