summaryrefslogtreecommitdiffstats
Commit message (Expand)AuthorAgeFilesLines
* net: prevent if_clone_destroy from racing with rest of stackHEADmasterJason A. Donenfeld3 days1-1/+35
* Eliminate pointless 'child' variable and just cast the config_found() result tokrw3 days1-4/+3
* add kstat support for reading hardware counters.dlg4 days1-217/+352
* it's not an error if the API doesnt support reading sfp/qsfp stuff.dlg4 days1-4/+10
* actually tell people that they need a recent fw/API version.dlg4 days1-2/+15
* Minor code shuffle to close ranks of sc_link initializations.krw4 days1-3/+5
* False alarm, back it goes. Managed to mangle my kernel build directorykrw4 days2-5/+3
* Revert previous. Broke something somehow.krw4 days2-3/+5
* timekeep_sz now already includes the round_page() adjustment; ok kettenis@naddy4 days1-2/+2
* Cache the adapter's port_wwn and node_wwn in mpi_softc, not thekrw4 days2-4/+9
* Cache a scsibus_softc * and use it rather than cludging indirectkrw4 days2-4/+6
* Garbage collect unused vioscsi_softc member "sc_scsibus".krw4 days1-3/+2
* Garbage collect unused nvme_softc member "sc_scsibus".krw4 days2-5/+3
* Cache a pointer to the scsibus_softc rather than hoping the copy inkrw4 days2-4/+6
* Set the default ABI for OpenBSD to ELFv2 in the clang frontend as well.kettenis4 days1-0/+4
* Add usertc.c.kettenis4 days1-0/+1
* Add missing usertc.c file.kettenis4 days1-0/+21
* Oops. One too many lines deleted, leaving 'buswidth' uninitialized butkrw4 days1-1/+2
* Shuffle sc_link.openings initialization down to its friends and justkrw4 days1-6/+4
* Some whitespace fixes for the inline assembly.kettenis4 days1-6/+6
* Synchronize each core's CP0 cycle counter using the IO clock counter.visa4 days6-11/+103
* Optimize rasops_vcons_copyrows() so write-only framebuffer consolesfcambus4 days1-11/+26
* Implement linear and power-of-two histograms: hist() and lhist() keywords.mpi4 days7-37/+359
* Build 'flags' in intermediate variable and shuffle sc_linkkrw4 days1-16/+17
* Expunge some Captain Obvious comments to make upcoming diff(s) smallerkrw4 days1-5/+1
* Expunge some Captain Obvious comments, tweak whitespace a bit, move a debugkrw4 days31-112/+47
* Don't reload the FPU registers when they're already loaded.kettenis4 days1-3/+5
* Implement a machdep.altivec sysctl that always returns 1. This makeskettenis4 days2-3/+17
* Change locks which don't use spin_lock_irq() or spin_lock_irqsave()jsg5 days2-3/+3
* test for gem_size > 0 on pool_destroy() matching pool_init() calljsg5 days1-1/+2
* don't complain if the API is too low to support "query phy types".dlg5 days1-6/+8
* Add hw.cpuspeed and hw.setperfgkoehler5 days3-6/+117
* table fix;jmc5 days1-3/+3
* Document which crypto transforms are enabled by default.tobhe5 days1-62/+62
* Shuffle a bit of code to stop using sc->sc_link to hold values forkrw5 days2-9/+11
* Oops, I didn't intend to commit these bits.kettenis5 days2-18/+5
* Implement process_read_regs() such that core dumps become usable.kettenis5 days3-28/+28
* syncderaadt5 days1-0/+1
* Parse a bootduid= argument passed on the command line and use it to set thekettenis5 days1-1/+33
* Pass bootduid (from disklabel) on the bootline to the kernelderaadt5 days1-3/+4
* adjust %n description to vaguely say "pointer", becuase the followingderaadt5 days1-4/+6
* Clear FPU state if we try to restore the state when the process didn't usekettenis5 days1-2/+6
* Add check for userland VSID overflow.kettenis5 days1-1/+2
* base tree had only two uses of the printf %n format string, in this file.deraadt5 days1-2/+2
* As suggested by deraadt@, rewrite most of the printf(3) manual pageschwarze5 days1-453/+586
* Use OPAL_REINIT_CPUS calls during early boot to set endianness and tokettenis5 days1-1/+11
* Change users of IFQ_SET_MAXLEN() and IFQ_IS_EMPTY() to use the "new" API.patrick5 days117-315/+308
* Change users of IFQ_PURGE() to use the "new" API.patrick5 days8-24/+19
* Change users of IFQ_DEQUEUE(), IFQ_ENQUEUE() and IFQ_LEN() to use thepatrick5 days72-158/+147
* Add OPAL_REINIT_CPUS.kettenis5 days2-2/+12