summaryrefslogtreecommitdiff
path: root/sys/arch/hppa
AgeCommit message (Collapse)Author
2010-06-29Add support for mapping ACPI to PCI devicesJordan Hargrave
ok kettenis, deraadt
2010-06-29Remove COMPAT_HPUX. No one wanted to support it and its fewmets werePhilip Guenthe
blocking other cleanups ok miod@
2010-06-29Ansify function declarations.Joel Sing
ok kettenis@
2010-06-29Factor out code used to save and flushe process FPU context.Joel Sing
ok kettenis@
2010-06-29Store pointer to process FPU state in struct cpu_info.Joel Sing
ok kettenis@
2010-06-28Allow tty drivers to request larger buffers at attach time using aTheo de Raadt
max-baud-rate hint. Adjust TTYHOG (the nearly full logic) to this new situation. The larger buffers are required by the very high speed KDDI devices in Japan (CF com, or USB ucom) so those are the only two drivers which currently ask for a larger buffer size. ok yasuoka miod
2010-06-27uvm constraints. Add two mandatory MD symbols, uvm_md_constraintsThordur I. Bjornsson
which contains the constraints for DMA/memory allocation for each architecture, and dma_constraints which contains the range of addresses that are dma accessable by the system. This is based on ariane@'s physcontig diff, with lots of bugfixes and additions the following additions by my self: Introduce a new function pool_set_constraints() which sets the address range for which we allocate pages for the pool from, this is now used for the mbuf/mbuf cluster pools to keep them dma accessible. The !direct archs no longer stuff pages into the kernel object in uvm_km_getpage_pla but rather do a pmap_extract() in uvm_km_putpages. Tested heavily by my self on i386, amd64 and sparc64. Some tests on alpha and SGI. "commit it" beck, art, oga, deraadt "i like the diff" deraadt
2010-06-27Use a mutex to make atomic operations atomic on multiprocessor kernels.Joel Sing
ok kettenis@
2010-06-26Allow processes to be scheduled onto secondary CPUs.Joel Sing
ok kettenis@
2010-06-26Add GENERIC.MP.Joel Sing
ok kettenis@
2010-06-26Do not store curcpu() in syscall and trap handlers since we can go toJoel Sing
sleep and wake up on another CPU. ok kettenis@
2010-06-26Avoid a potential race when unlocking a mutex.Joel Sing
ok kettenis@
2010-06-19Remove unused md BUS_SPACE_MAP_NOEXTENT flag; ok jsing@Miod Vallat
2010-06-10Declare safepri at the MD level on each platform, so that the kern_synch.cTheo de Raadt
does not have to deal with it as a common. Some platforms may be missed by this commit... if you spot one, fix it the same way. ok miod
2010-06-09Move declaration of cn_tab from dev/cons.c to dev/cninit.c, or appropriateMiod Vallat
md files for ports which do not use dev/cninit.c. This gets rid of a common at kernel link time. feedback drahn@ ok deraadt@
2010-06-09Wire up device nodes for disk mapper.Joel Sing
ok deraadt@
2010-06-08Add missing function prototype.Joel Sing
2010-06-03Rename ci_spinup_stack to ci_stack and use it as the stack for both CPUJoel Sing
spin up and FPU emulation. Since all CPUs need a stack for FPU emulation, move the allocation code to cpuattach(). ok kettenis@
2010-06-03Add missing function prototypes for MP kernels.Joel Sing
2010-06-01use --warn-common for linking (some arch's will show new warnings), andTheo de Raadt
prefer binutils-compatible options in STRIPFLAGS (now that our non-binutils strip(1) can handle them too) ok drahn; miod kettenis (for parts)
2010-05-31Completely rewrite the tags file creation process. use config(8) againstTheo de Raadt
a fat kernel (GERERIC or GENERIC.MP) into a temporary dir, then extract some variables using make -V, and after some more singing and dancing use this full and (more) correct list of files with ctags. Don't read this code. ok guenther
2010-05-28${PARAM} is now included in ${CPPFLAGS}, so stop passing it separatelyPhilip Guenthe
to genassym.sh ok deraadt
2010-05-27We no longer need -traditional-cpp for hppa.Joel Sing
2010-05-27Generate exported names using token pasting rather than a macro. OtherwiseJoel Sing
without -traditional-cpp the trap name gets preprocessed to the defined value rather than remaining as text. ok kettenis@
2010-05-25The _arch and _mach variables now come from config(8), decreasing even furtherTheo de Raadt
the differences between these files. You will need a newer config(8) binary to be able to build kernels. ok kettenis miod
2010-05-25Use token paste operator to avoid nested __CONCAT() macros. This allowsJoel Sing
the code to compile without -traditional-cpp. ok kettenis@
2010-05-25Hand-list the two-step assym.h -> .[Ss] -> .o dependencies. TheTheo de Raadt
result: kernels built without 'make depend'-provided information (ie. the .depend file) are more likely to have their *.[Ss] file compilations track changes to *.h files. The "*.o: assym.h" dependencies listed are gotten from reading the .depend output --- from the biggest kernel possible (ie. GENERIC.MP). When an architecture changes in a substantial way (new .[sS] files), the list should be updated in the prettiest way possible. This is not encouraging people to skip 'make depend'; other issues are not resolved and may be solved later with a change guenther is working on. You can still screwed really easily, so continue running make depend as config tells you. Idea from a discussion with drahn ok drahn, kettenis likes the idea too
2010-05-25${SRCS} is only used once, and fits onto the line where it is usedTheo de Raadt
idea that came out of discussion with drahn
2010-05-24give up on -Wvariable-declTheo de Raadt
delete the archaic links: target which is easily misused handle special .[sS] files in a portable way
2010-05-24Unify the dependency time check for whether an architecture has SFILESTheo de Raadt
2010-05-24Add missing prototypesTheo de Raadt
ok jsing kettenis
2010-05-24Reduce differences between the Makefile.* files for all architectures.Theo de Raadt
ok various people, tested by fewer people, tested by me on 15.
2010-05-23no need for -Dhppa or -Dhppa64 on these architecturesTheo de Raadt
2010-05-22Remove instruction TLB purges for direct mappings that will never have theMark Kettenis
X bit set.
2010-05-21Add support for IPIs on hppa.Joel Sing
ok kettenis@
2010-05-19Spin up secondary CPUs on hppa multiprocessor kernels. At this stage weJoel Sing
enable clock interrupts, however do not allow processes to be scheduled onto the secondary CPUs - hopefully we can change this shortly... ok kettenis@
2010-05-17Fix locking logic, remembering that hppa locks are inverted. Whilst hereJoel Sing
also add mpl_cpu and curcpu() to the debugging messages. Issue spotted by and ok kettenis@
2010-05-16Use a per CPU trap save area, ensuring that we maintain 64-byte alignmentJoel Sing
so that it stays within a single cache line. Feedback and ok kettenis@
2010-05-09Prevent "taken branch" exceptions from kernel space by clearing the PSL_TMark Kettenis
bit before we enter a signal handler. "if you say so" miod@, "seems to make sense" jsing@
2010-05-09Make single stepping a system call work. Instead of single stepping throughMark Kettenis
the syscall gateway page, which doesn't work since that page is shared between processes, this makes us step over that bit by setting a breakpoint on the instruction where the system call returns. ok miod@, jsing@
2010-05-09Handle .rodata.* sections emitted by gcc4 the same way as .rodata and .rodata1.Mark Kettenis
2010-05-06Ansify functions and use "clock" rather than the CPU name for clockJoel Sing
interrupts. ok kettenis@
2010-05-06Count the number of CPUs in the system.Joel Sing
ok kettenis@
2010-05-05Since we never set the X bit for directly mappings, nothing can be moved intoMark Kettenis
the instruction cache through them. So we don't need to flush the instruction cache and purge instruction TLBs for these mappings. ok jsing@, miod@
2010-05-02Properly initialize the members of `struct mdproc' that deal withMark Kettenis
single-stepping when we fork and reset them upon exec. ok miod@
2010-04-30Right now, if anything internal changes with a uvm object, diverseOwain Ainsworth
places in the tree need to be touched to update the object initialisation with respect to that. So, make a function (uvm_initobj) that takes the refcount, object and pager ops and does this initialisation for us. This should save on maintainance in the future. looked good to fgs@. Tedu complained about the British spelling but OKed it anyway.
2010-04-29Currently pdc_scanbus() scans in reverse, which results in physical CPU0Joel Sing
attaching as cpu1 and physical CPU1 attaching as cpu0. Split the device scan into two parts - the first pass scans forward and attaches memory and CPUs, whilst the second pass scans in reverse attaching the remaining devices, thus preserving current semantics. Found the hard way whilst trying to spin up the secondary CPU on my j6750. ok miod@ kettenis@
2010-04-29Use per CPU hardware interrupt mask.Joel Sing
ok kettenis@
2010-04-29Store hardware timer information per CPU.Joel Sing
ok kettenis@
2010-04-28Do not modify hardware interrupt mask when establishing a soft interruptJoel Sing
handler. Also, do not allow a hardware interrupt handler to be established for a soft interrupt. ok kettenis@
generated by cgit v1.2.3 (git 2.25.1) at 2025-01-25 13:05:06 +0000