Use correct memory barrier on x86
[akaros.git] / kern / arch /
2011-11-03 Andrew WatermanUse correct memory barrier on x86
2011-11-03 Andrew WatermanMade stuff work on SPARC HW
2011-11-03 Andrew Watermanvcore_id() is now implemented using TLS
2011-11-03 Barret RhodenLoad balancing for EXPER_TRADPROC
2011-11-03 Barret RhodenFP save/restore on x86
2011-11-03 Kevin KluesFix to e1000 driver
2011-11-03 Andrew WatermanFixed backtracing w.r.t. SPARC kernel entry
2011-11-03 Andrew WatermanFixed register clobbering in SPARC
2011-11-03 Kevin KluesAdded fillmeup support for the e1000 driver
2011-11-03 Barret RhodenFixed DISABLE_SMT to report the right max_vcores
2011-11-03 David Zhuchange fillmeup syscall to use an int32 field to indica...
2011-11-03 Kevin KluesAdd fillmeup support for the realtek driver
2011-11-03 Barret RhodenRevert "Make floating point work (no ancillary ...
2011-11-03 Kevin KluesAdded a fillmeup syscall
2011-11-03 Kevin KluesMake floating point work (no ancillary state yet though)
2011-11-03 Kevin KluesMake the realtek driver route interrupts like the other...
2011-11-03 Andrew WatermanMassive cleanup of SPARC kernel entry code
2011-11-03 Barret RhodenBasic FP support on x86
2011-11-03 Kevin KluesZero out gs and fs whenever we enter the kernel
2011-11-03 Kevin KluesEnabled interrupts in the int driven kernel monitor
2011-11-03 Kevin KluesForce interrupts to core 0 on network drivers
2011-11-03 Barret RhodenTrigger the monitor with an interrupt
2011-11-03 Barret Rhodensys_halt_core() and an idle process
2011-11-03 Barret RhodenExp: per-core runqueues and timer ticks
2011-11-03 Kevin KluesAdded appserver support to the e1000 driver
2011-11-03 Kevin KluesChanges to support the 64 core boxboro machine
2011-11-03 Kevin KluesMoved the E1000 MMIO addr to a makelocal variable.
2011-11-03 Kevin KluesFirst run at an E1000 driver, with various LWIP fixes.
2011-11-03 Andrew WatermanExposed per-core timer interrupts on SPARC
2011-11-03 Andrew WatermanMade pthreads work on SPARC
2011-11-03 Barret Rhodenpthread_yield()
2011-11-03 Barret RhodenPthread create, join, and exit
2011-11-03 Andrew WatermanMade ros_pop_tf work on SPARC
2011-11-03 Barret RhodenThe kernel migrates silly state when becoming _M
2011-11-03 Andrew WatermanFixed cpu_halt on SPARC
2011-11-03 Kevin KluesMassive renaming/cleanup of harts->vcores
2011-11-03 Kevin KluesCheck to make sure an ldt exists before loading it
2011-11-03 Kevin KluesAppserver up and running on hardware for x86...
2011-11-03 Barret RhodenPreemption of user cores
2011-11-03 Barret Rhodenpop_ros_tf can handle missed IPIs
2011-11-03 Barret RhodenUser stackpointers moved to the vcpd structs
2011-11-03 Barret RhodenEnsures __notify took the user trapframe as an arg
2011-11-03 Barret Rhodenset_current_tf only applies to user trapframes
2011-11-03 Barret RhodenFixes trap-based syscalls on x86
2011-11-03 Kevin KluesFixed smp_percpu_init() on sparc; added asserts
2011-11-03 Andrew WatermanFixed up register window fill code
2011-11-03 Barret RhodenFixes bug in x86 atomics
2011-11-03 Andrew WatermanFixed kernel messages on SPARC
2011-11-03 Barret RhodenUserspace no longer includes the kernel's arch/*
2011-11-03 Barret RhodenCompare and Swap
2011-11-03 Barret Rhodenproc messages are routine instead of immediate
2011-11-03 Barret RhodenCall arbitrary functions from the monitor
2011-11-03 Barret RhodenHandles routine kmsgs before returning to userspace
2011-11-03 Barret RhodenRoutine kmsg processing
2011-11-03 Barret RhodenSparc kernel messages
2011-11-03 Barret RhodenKernel messages infrastructure
2011-11-03 Barret Rhodenx86: waits after sending IPIs
2011-11-03 Barret RhodenNew vcoremap in procinfo
2011-11-03 Barret RhodenInitial notification and preemption headers
2011-11-03 Barret Rhodenx86: Fixes TLS bug causing kernel page faults
2011-11-03 Barret Rhodenx86: allows panic earlier in the boot process
2011-11-03 Andrew WatermanMoved brk/sbrk into user-space
2011-11-03 Kevin KluesFinally got KFS stuff sorted out on the new setup
2011-11-03 Kevin KluesCleaned up the way ifdefs are handled
2011-11-03 Andrew WatermanFixed TLS on i386
2011-11-03 Kevin KluesDedicated core 1 to NIC functionality
2011-11-03 Kevin KluesWe never actually removed <ros/error.h>, now we do
2011-11-03 Andrew WatermanFixed bug in SPARC trap entry
2011-11-03 Andrew WatermanAdded icache flushing
2011-11-03 Andrew WatermanAdded demand paging support
2011-11-03 Andrew WatermanMoved pte_t definition to arch-specific file
2011-11-03 Andrew Watermanfrontend_syscall now accepts NULL for &errno
2011-11-03 Andrew WatermanFixed bugs in SPARC trap handlers
2011-11-03 Andrew WatermanAdded arch framework for handling page faults
2011-11-03 Kevin KluesAdded support for returning errors from syscalls
2011-11-03 Kevin KluesRemoved intptr_t
2011-11-03 Kevin KluesSet the default syscall return value to 0
2011-11-03 Kevin KluesFixed bug in sysenter/sysexit
2011-11-03 Kevin Kluesproc_init_trapframe now zeros all fields initially
2011-11-03 Kevin KluesAdded a bad implementation of get_free_va_range
2011-11-03 Kevin KluesCleaned up NIC code for appserver
2011-11-03 Kevin KluesRemoved references to intptr_t
2011-11-03 Kevin KluesRemoved proc_init_arch; created frontend_proc_init
2011-11-03 Kevin KluesFixed kernel page fault on x86 port
2011-11-03 Andrew Watermanadded front-end link code
2011-11-03 Kevin KluesMore work towards getting the appservers combined
2011-11-03 Kevin KluesMajor reorganization in prep for appserver merge
2011-11-03 Andrew WatermanUser programs now tell kernel the stack pointers for...
2011-11-03 Andrew Watermancleand up and sped up SPAARC kernel entry
2011-11-03 Andrew Watermangcc-glibc compiles on both SPARC and i386
2011-11-03 Andrew WatermanExposed SPARC iobus to userspace
2011-11-03 Andrew WatermanAdded back harts support
2011-11-03 Andrew Watermanglibc might compile for x86 now
2010-03-24 Kevin KluesCheckin of Andrews stuff needed to compile glibc
2010-03-24 Paul PearceBrief NE2K driver fix
2010-03-24 Paul PearceFixed NE2K support. WIP Commit
2010-03-24 Paul PearceFunctonal webserver with supporting code changes. WIP...
2010-03-24 Paul PearceFinished glue logic to link LWIP to the OS. This is...
2010-03-13 Barret RhodenFixes race in x86 smp boot