| Age | Commit message (Collapse) | Author | |
|---|---|---|---|
| 2009-06-04 | More fallback from struct disklabel changes, yawn | Miod Vallat | |
| 2009-06-03 | add kern.bufcachepercent sysctl to allow adjusting the buffer cache | Bob Beck | |
| size on a running system. ok art@, oga@ | |||
| 2009-06-03 | Arla client rename from xfs to nnpfs for later upgrades. Tested on various ↵ | Janne Johansson | |
| arches. ok todd@ beck@ | |||
| 2009-05-31 | Remove BUS_DMAMEM_NOSYNC definition. Its name is not consistent with | Miod Vallat | |
| other BUS_DMA_xxx flag names, and nothing uses it. ok many@ | |||
| 2009-05-31 | Fix definition of CL_FIFO_MAX to match what the hardware can do; this in | Miod Vallat | |
| turn fixes a stack smash in cl_rxintr(). | |||
| 2009-04-20 | Add a BUS_DMA_ZERO flag for bus_dmamem_alloc() to return zeroed memory. | Owain Ainsworth | |
| Saves every damned driver calling bzero(), and continues the M_ZERO, PR_ZERO symmetry. | |||
| 2009-04-19 | Rename max_cpus to ncpusfound and compute it regardless of option | Miod Vallat | |
| MULTIPROCESSOR. | |||
| 2009-04-14 | Convert the waitok field of uvm_pglistalloc to "flags", more will be added soon. | Owain Ainsworth | |
| For the possibility of sleeping, the first two flags are UVM_PLA_WAITOK and UVM_PLA_NOWAIT. It is an error not to show intention, so assert that one of the two is provided. Switch over every caller in the tree to using the appropriate flag. ok art@, ariane@ | |||
| 2009-03-29 | make various strings ("can't map mem space" and similar) more consistent | Stuart Henderson | |
| between instances, saving space in the kernel. feedback from many (some incorporated, some left for future work). ok deraadt, kettenis, "why not" miod. | |||
| 2009-03-15 | Generic softinterrupt support for m88k based platforms, adapted from arm | Miod Vallat | |
| with different locking mechanism. 88110 soft ipi are replaced with an ipi callback which is checked upon return from exception (it can not be kept as a softintr, as the generic softinterrupt code doesn't have per-cpu pending softintr queues). | |||
| 2009-03-09 | Switch mvme88k to timecounters. | Miod Vallat | |
| 2009-03-08 | Move more z8536 defines from MVME188 specific code to the MI header and use it. | Miod Vallat | |
| 2009-03-07 | When allocating memory in bus_dmamem_alloc() with uvm_pglistalloc(), do not | Miod Vallat | |
| try to be smart for the address range, uvm_pglistalloc() is smart enough nowadays. | |||
| 2009-03-05 | Change SMP secondary processor startup to have them ``hatch'' earlier, | Miod Vallat | |
| so that they can invoke sched_init_cpu() before the scheduler starts, which allows the horrible kluge in cpu_configure() to die. | |||
| 2009-03-04 | Clean up the ipi code a bit and try to provide useful comments as well. | Miod Vallat | |
| 2009-03-04 | When handling clock interrupts, check the overflow counter to know how many | Miod Vallat | |
| times hardclock() needs to be invoked; fixes clock drift found on 197DP with SMP kernels. | |||
| 2009-03-01 | Make more IPI synchronous. Performance slightly down, stability slightly up. | Miod Vallat | |
| ok deraadt@ | |||
| 2009-02-27 | Rework nmi handling to handle ``complex'' NMI faster, and return as fast as | Miod Vallat | |
| possible from the exception, without doing the AST and softintr dance. This should avoid too much stack usage under load. ok deraadt@ | |||
| 2009-02-27 | Be a bin more paranoid in m88410_inval() in the operation complete check. | Miod Vallat | |
| ok deraadt@ | |||
| 2009-02-21 | Move part of the mp lock logic into per-cpu callbacks; on MVME197DP we need | Miod Vallat | |
| to disable NMI sources in addition to interrupt sources, and we can not use a quick sequence with shadowing frozen as done for atomic ops. This lets GENERIC.MP boot multiuser on MVME197DP boards, and is so far stable enough to be able to recompile a kernel from scratch (with make -j2). | |||
| 2009-02-21 | Get rid of 88110 nmi stacks. This was a good idea, but I outsmarted myself | Miod Vallat | |
| since it was intended to service NMI occuring in user mode, and we could end up invoking preempt() and have another cpu start using this stack, with interesting results. | |||
| 2009-02-20 | When looping to process external interrupts, do not complain (and panic) | Miod Vallat | |
| if another processors sends us an IPI, it will get handled later. | |||
| 2009-02-20 | Let M188 kernel compile. | Miod Vallat | |
| 2009-02-18 | Fix thinko in list traversal in vsbic_cmd_ccb(). | Miod Vallat | |
| 2009-02-17 | Pass a cpu_info * to setsoftipi() so it does not need to curcpu(), which | Miod Vallat | |
| synchronizes the pipeline on 88110. | |||
| 2009-02-17 | Be sure to pass a valid confargs * to child devices of vme0 when attaching, | Miod Vallat | |
| for they will attempt to dereference it and having NULL readable while in the kernel is just cheating. | |||
| 2009-02-16 | More 88110 SMP work. Contains, horribly entangled: | Miod Vallat | |
| - dma_cachectl() split into a ``local cpu only'' and ``all cpus'', and an ipi to broadcast ``local dma_cachectl'' is added. - cpu_info fields are rearranged, to have the 88100-specific information and the 88110-specific information overlap, and has many more 88110 ugly things. - more ipi handling in the 197-specific area. Since it is not possible to have the second processor receive any hardware interrupt (selection is done on a level basis via ISEL, and we definitely do not want the main cpu to lose interrupts), the best we can do is to inflict ourselves a soft interrupt for late ipi processing. It gets used for softclock and hardclock on the secondary processor, but since the soft interrupt dispatcher doesn't have an exception frame, we have to remember parts of it to build a fake clockframe from the soft ipi handler (ugly but works). This now lets GENERIC.MP run a few userland binaries before bugs trigger. | |||
| 2009-02-16 | Since NMI are now handled separately, remove the ``interrupt type'' argument | Miod Vallat | |
| from interrupt() and related function pointers. | |||
| 2009-02-16 | Extend the scsi_adapter minphys() callback to take a struct scsi_link * | Miod Vallat | |
| as additional argument. This will allow intermediate layers between scsi devices such as sd and scsi host adapters to take appropriate action if necessary. | |||
| 2009-02-15 | Properly send a request sense command if the command status asks for it. | Miod Vallat | |
| 2009-02-14 | Sync device lists with GENERIC. | Miod Vallat | |
| 2009-02-14 | Add vsbic(4), a drive for the MVME327A SCSI and floppy controller, only for | Miod Vallat | |
| the SCSI part so far. | |||
| 2009-02-14 | Fix a few defines. | Miod Vallat | |
| 2009-02-13 | Use a different dispatcher for the NMI traps on 88110, these are too | Miod Vallat | |
| different from regular hardware interrupts to be worth handling the same way. Disable IPI reception while we are handling pending IPIs. And do not reenable them by mistake if we need to send an IPI in return. This lets GENERIC.MP boot single user on a MVME197DP. There are still many bugs to fix. | |||
| 2009-02-13 | Release the secondary processor initialization interlock a bit later, | Miod Vallat | |
| after the currently running processor has advertized itself as ready. | |||
| 2009-02-13 | Allow IPIs on the secondary processors later in the bootstrap process, when | Miod Vallat | |
| we are sure the NMI stack exists. | |||
| 2009-02-13 | Provide a specific delay() routine using separate timers for the two cpus | Miod Vallat | |
| on MVME197DP boards running the MP kernel. | |||
| 2009-02-13 | Make delay() a per-board function pointer. | Miod Vallat | |
| 2009-02-12 | BPP routines for ``intelligent'' Motorola MVME boards. Currently not used | Miod Vallat | |
| by anything, will eventually be used by the vx(4) MVME332 driver (once I take the time to fix it) and the vsbic(4) MVME327 driver being worked on. | |||
| 2009-02-08 | Oops, initialize the nmi stack correctly for the boot processor. | Miod Vallat | |
| 2009-02-08 | On 88110 processors, use a separate stack to handle NMI; these can occur | Miod Vallat | |
| while we are switching pcbs and all sort of bad things could happen. | |||
| 2009-02-08 | Use the correct flags when bus_dmamap_sync()ing the s/g lists. | Miod Vallat | |
| 2009-02-01 | Nuke whitespace at EOL. | Miod Vallat | |
| 2009-02-01 | Remove dma_cachectl() and rename dma_cachectl_pa() to dma_cachectl() now that | Miod Vallat | |
| the old vs(4) code is gone. | |||
| 2009-02-01 | Get the cpu speed from the pcctwo configuration on MVME187. | Miod Vallat | |
| 2009-02-01 | Finally switch this driver to bus_dma. | Miod Vallat | |
| 2009-01-29 | Add a few cache operations after I/O completes. | Miod Vallat | |
| 2009-01-29 | The ``flush before inval'' 88410 workaround needs to be done in | Miod Vallat | |
| m88410_initialize_cpu() as well. Since we can't flush unknown data, just make sure all 88410 lines are filled with valid data before flushing it. This should fix the spurious ``197DP hangs or loops on osiop errors after autoconf'' issue I've been seeing. | |||
| 2009-01-25 | Remove /dev/drum and related code. | Miod Vallat | |
| 2009-01-18 | Define HEAP_START so that we can use space below it for ethernet buffers; | Miod Vallat | |
| this prevents if_ie buffers from possibly overwriting the stack. Crank bootloader versions. | |||
 |
index : src | |
| OpenBSD base system |
| summaryrefslogtreecommitdiff |