search for: squeue

1. 566 squeue_enter(squeue_t *sqp, mblk_t *mp, mblk_t *tail, uint32_t cnt, Why cnt is declared as uint32_t (fixed size?) Why not just int or uint_t? 554 * squeue_enter() - enter squeue sqp with mblk mp (which can be 555 * a chain), while tail points to the end and cnt in number of 556 * mblks in the chai...

Folks, I just posted an updated Crossbow virtualization architecture document. The new revision is available at: http://opensolaris.org/os/project/crossbow/Docs/crossbow-virt.pdf The main changes are the addition of support for multiple MAC addresses per client, and an explicit separation between consolidation private and project private MAC API entry points. See in particular the

...39;t assume that the sending CPU will be reasonably static for a given flow. (I gotta think this wreaks havoc on the caches involved... but that''s a different problem.) _If_ transmitted packets are sent to the stack and always land in a delivery queue, then perhaps the outbound queue (squeue?) can have a worker thread that doesn''t migrate around. But in order for that to happen, we have to stop having sending threads deliver right to the driver driver when intervening queues are empty. I _think_ this will work better for throughput. It may hurt latency slightly though....

...a short description on assigning priorities to a link (via dladm). Flows will have a similar mechanism (via flowadm). thanks, -venu ---------------------------------------------------------------------------- The components associated with a link or flow, such as Soft Ring Sets, Soft Rings and squeues, have kernel threads processing inbound/outbound packets. The scheduler processes threads based on a thread''s priority level. The priority level of a thread determine how it is scheduled w.r.t other kernel threads. Thus by assigning a priority to a link/flow, we can control how the thre...

...r/src/uts/common/inet/ip/ip_cksum.c usr/src/uts/common/inet/ip/ip_ftable.c usr/src/uts/common/inet/ip/ip_if.c usr/src/uts/common/inet/ip/ip_ire.c usr/src/uts/common/inet/ip/ip_mroute.c usr/src/uts/common/inet/ip/ip_multi.c usr/src/uts/common/inet/ip/ip_netinfo.c usr/src/uts/common/inet/ip/ip_squeue.c usr/src/uts/common/inet/ip/spd.c usr/src/uts/common/inet/ip/tun.c usr/src/uts/common/inet/ip_ftable.h usr/src/uts/common/inet/ip_if.h usr/src/uts/common/inet/ip_impl.h usr/src/uts/common/inet/ip_ire.h usr/src/uts/common/inet/ip_stack.h usr/src/uts/common/inet/ipclassifier.h usr/src/uts/c...

...f loads and stores. > > Each load (or store) consumes an entry in the load (or store) queue. The number > of slots in the load/store queues is unknown by the tool, since there is no > mention of it in the scheduling model. In practice, users can specify flag > `-lqueue=N` (vic. `-squeue=N`) to limit the number of entries in the queue to be > equal to exactly N (an unsigned value). If N is zero, then the tool assumes an > unbounded queue (this is the default). > > LSUnit implements a relaxed consistency model for memory loads and stores. The > rules are: > 1) A y...

...or speculative execution of loads and stores. Each load (or store) consumes an entry in the load (or store) queue. The number of slots in the load/store queues is unknown by the tool, since there is no mention of it in the scheduling model. In practice, users can specify flag `-lqueue=N` (vic. `-squeue=N`) to limit the number of entries in the queue to be equal to exactly N (an unsigned value). If N is zero, then the tool assumes an unbounded queue (this is the default). LSUnit implements a relaxed consistency model for memory loads and stores. The rules are: 1) A younger load is allowed to pass...

...es. > > Each load (or store) consumes an entry in the load (or store) queue.  > The number > of slots in the load/store queues is unknown by the tool, since there > is no > mention of it in the scheduling model.  In practice, users can specify > flag > `-lqueue=N` (vic. `-squeue=N`) to limit the number of entries in the > queue to be > equal to exactly N (an unsigned value). If N is zero, then the tool > assumes an > unbounded queue (this is the default). > > LSUnit implements a relaxed consistency model for memory loads and > stores. The > rules...

...oads and stores. > > Each load (or store) consumes an entry in the load (or store) queue. The > number > of slots in the load/store queues is unknown by the tool, since there is no > mention of it in the scheduling model. In practice, users can specify flag > `-lqueue=N` (vic. `-squeue=N`) to limit the number of entries in the queue > to be > equal to exactly N (an unsigned value). If N is zero, then the tool > assumes an > unbounded queue (this is the default). > > LSUnit implements a relaxed consistency model for memory loads and stores. > The > rules ar...

...oads and stores. > > Each load (or store) consumes an entry in the load (or store) queue. The > number > of slots in the load/store queues is unknown by the tool, since there is no > mention of it in the scheduling model. In practice, users can specify flag > `-lqueue=N` (vic. `-squeue=N`) to limit the number of entries in the queue > to be > equal to exactly N (an unsigned value). If N is zero, then the tool > assumes an > unbounded queue (this is the default). > > LSUnit implements a relaxed consistency model for memory loads and stores. > The > rules ar...