PV ticket locks without expanding spinlock
Get series with b4
b4 am [email protected]Versions newest first
- v1 Latest14 patches 53 repliesIn review
Patches in v1 No applies recorded yet
| # | Subject and files | Size | Replies | Applied |
|---|---|---|---|---|
| 1/14 | x86/ticketlock: clean up types and accessorsarch/ arch/ | +28 | 2 replies | |
| 1/14 | x86/ticketlock: clean up types and accessorsarch/ arch/ | +28 | 0 replies | |
| 2/14 | x86/ticketlock: convert spin loop to Carch/ | +30 | 0 replies | |
| 2/14 | x86/ticketlock: convert spin loop to Carch/ | +30 | 0 replies | |
| 3/14 | x86/ticketlock: Use C for __ticket_spin_unlockarch/ | +17 | 0 replies | |
| 3/14 | x86/ticketlock: Use C for __ticket_spin_unlockarch/ | +17 | 0 replies | |
| 4/14 | x86/ticketlock: make large and small ticket versions of spin_lock the samearch/ | +8 | 0 replies | |
| 4/14 | x86/ticketlock: make large and small ticket versions of spin_lock the samearch/ | +8 | 0 replies | |
| 5/14 | x86/ticketlock: make __ticket_spin_lock commonarch/ | +20 | 0 replies | |
| 5/14 | x86/ticketlock: make __ticket_spin_lock commonarch/ | +20 | 0 replies | |
| 6/14 | x86/ticketlock: make __ticket_spin_trylock commonarch/ arch/ | +14 | 0 replies | |
| 6/14 | x86/ticketlock: make __ticket_spin_trylock commonarch/ arch/ | +14 | 0 replies | |
| 7/14 | x86/spinlocks: replace pv spinlocks with pv ticketlocksarch/ arch/ arch/ +2 more | +50 | 0 replies | |
| 7/14 | x86/spinlocks: replace pv spinlocks with pv ticketlocksarch/ arch/ arch/ +2 more | +50 | 0 replies | |
| 8/14 | x86/ticketlock: collapse a layer of functionsarch/ | +5 | 0 replies | |
| 8/14 | x86/ticketlock: collapse a layer of functionsarch/ | +5 | 0 replies | |
| 9/14 | xen/pvticketlock: Xen implementation for PV ticket locksarch/ | +36 | 10 replies | |
| 9/14 | xen/pvticketlock: Xen implementation for PV ticket locksarch/ | +36 | 0 replies | |
| 10/14 | x86/pvticketlock: use callee-save for lock_spinningarch/ arch/ arch/ +1 more | +5 | 0 replies | |
| 10/14 | x86/pvticketlock: use callee-save for lock_spinningarch/ arch/ arch/ +1 more | +5 | 0 replies | |
| 11/14 | x86/ticketlock: don't inline _spin_unlock when using paravirt spinlocksarch/ kernel/ | +4 | 0 replies | |
| 11/14 | x86/ticketlock: don't inline _spin_unlock when using paravirt spinlocksarch/ kernel/ | +4 | 0 replies | |
| 12/14 | x86/ticketlocks: when paravirtualizing ticket locks, increment by 2arch/ arch/ | +18 | 0 replies | |
| 12/14 | x86/ticketlocks: when paravirtualizing ticket locks, increment by 2arch/ arch/ | +18 | 0 replies | |
| 13/14 | x86/ticketlock: add slowpath logicarch/ arch/ arch/ +1 more | +88 | 37 replies | |
| 13/14 | x86/ticketlock: add slowpath logicarch/ arch/ arch/ +1 more | +88 | 0 replies | |
| 14/14 | x86/ticketlocks: tidy up __ticket_unlock_kick()arch/ arch/ | +1 | 0 replies | |
| 14/14 | x86/ticketlocks: tidy up __ticket_unlock_kick()arch/ arch/ | +1 | 0 replies |
Reviews trailers per patch
No Reviewed-by, Acked-by or Tested-by trailers for v1 yet.
Cover letter [PATCH 00/14] PV ticket locks without expanding spinlock
From: Jeremy Fitzhardinge <[email protected]>
Hi all,
This is a revised version of the pvticket lock series.
The early part of the series is mostly unchanged: it converts the bulk of the ticket lock code into C and makes the "small" and "large" ticket code common. The only changes are the incorporation of various review comments.
The latter part of the series converts from pv spinlocks to pv ticket locks (ie, using the ticket lock fastpath as-is, but adding pv ops for the ticketlock slowpaths).
The significant difference here is that rather than adding a new ticket_t-sized element to arch_spinlock_t - effectively doubling the size - I steal the LSB of the tickets themselves to store a bit. This allows the structure to remain the same size, but at the cost of halving the max number of CPUs (127 for a 8-bit ticket, and a hard max of 32767 overall).
The extra bit (well, two, but one is unused) in indicates whether the lock has gone into "slowpath state", which means one of its lockers has entered its slowpath and has blocked in the hypervisor. This means the current lock-holder needs to make sure it gets kicked out of the hypervisor on unlock.
The spinlock remains in slowpath state until the last unlock happens (ie there are no more queued lockers).
This code survives for a while with moderate testing, (make -j 100 on 8 VCPUs on a 4 PCPU system), but locks up after about 20 iterations, so there's still some race/deadlock in there (probably something misordered), but I think the basic approach is sound.
Thanks,
J
Jeremy Fitzhardinge (14):
x86/ticketlock: clean up types and accessors x86/ticketlock: convert spin loop to C
x86/ticketlock: Use C for __ticket_spin_unlock x86/ticketlock: make large and small ticket versions of spin_lock the
same
x86/ticketlock: make __ticket_spin_lock common x86/ticketlock: make __ticket_spin_trylock common x86/spinlocks: replace pv spinlocks with pv ticketlocks x86/ticketlock: collapse a layer of functions xen/pvticketlock: Xen implementation for PV ticket locks x86/pvticketlock: use callee-save for lock_spinning x86/ticketlock: don't inline _spin_unlock when using paravirt
spinlocks
x86/ticketlocks: when paravirtualizing ticket locks, increment by 2 x86/ticketlock: add slowpath logic
x86/ticketlocks: tidy up __ticket_unlock_kick()
arch/x86/Kconfig | 3 + arch/x86/include/asm/paravirt.h | 30 +--- arch/x86/include/asm/paravirt_types.h | 8 +- arch/x86/include/asm/spinlock.h | 236 +++++++++++++--------------- arch/x86/include/asm/spinlock_types.h | 32 ++++- arch/x86/kernel/paravirt-spinlocks.c | 52 +++++-- arch/x86/xen/spinlock.c | 281 +++++---------------------------- kernel/Kconfig.locks | 2 +- 8 files changed, 231 insertions(+), 413 deletions(-)
Signature · 1 line
1.7.2.3