number field and an core identity string pointer.
Labels are now defined for all the entry points in the core header
structure so that the linker can relocate the branches to the core.
The core entry points are now branch instructions relative to the
start of the core so the address of the core function does not have to
be calcuated are call time.
Define the two new fields added to the FPE core header in
the arm_fpe_mod_hdr_t structure.
Added prototypes for arm_fpe_getcontext() and arm_fpe_setcontext().
Updated the prototypes for arm_fpe_core_loadcontext() and
arm_fpe_core_savecontext() to pass a fp_context_frame pointer.
CPU_SA110 and CPU_LATE_ABORT.
Updated the CLKF_INTR() macro for changes made to the interrupt system.
Updated some of the CPU ID codes.
Added the CPU ID for the ARM8.
This is for an Acorn A7000 machine with an ARM7500 CPU and no VRAM.
This config should also work for other ARM7500 machines with an
architecture that matches Acorns.
Make fp_reg_t a typedef of fp_extended_precision_t.
Rename the fp_state structure to fpe_sp_state as it describes
the single precision FPE state held in the pcb and is internal to the
kernel.
Define a new fp_state structure that is for user access to the fp
state (e.g. via ptrace()).
the AMD AM53CF94 Enhanced SCSI Controller. The code is based on the
SFAS216 driver as these chips are very similar. There are several
differences but more will follow.
Use definitions from this file for match_podule() rather than hardcoded
values.
Added a routine asc_minphys() in preparation for driver changes when on
card DMA support is added.
Use definitions from this file for match_podule() rather than hardcoded
values.
Reset the interface following a bad packet. This fixes some jams when
the driver failed to recover properly after a bad packet.
podule_data.h instead.
Removed the dead function find_podule().
The 0xf5 entry in the podules chunck directory is now searched for and
the podule description associated with this entry is placed in the
description field of the podule structure.
Generally tidied up all the comments.
- Fix up usage of MBD_ISPID().
- Ensure that we never deal with a NULL proc, and if our proc has
no pcb, punt. Suggested by Gordon Ross.
- Eliminate some redundant NULL pointer checks in the T_MMUFLT cases;
proc0.p_addr is now initialized early, and we make a single test
for sanity at the top of trap() now.
- Initialize proc0.p_addr just after setting up the kernel stack, to avoid
getting NULL pointers in trap(). Change suggested by Gordon Ross.
- Panic if main() returns.
- Tidy up a couple of comments.
(remaps page read/write/cache-inhibit, does write, restores previous
mapping). Kernel text no longer needs to be read/write with DDB/KGDB
is in the kernel.
Based on a similar module written for the Sun3 port by Gordon Ross,
and modified somewhat by me.
- Offset kernel text one page. Stash the PA of this offset page for
use later.
- Add a few comments.
- Free up some registers earlier in the initialization process.
- Use a `prototc' to set the Translation Control register, rather
than relying on a2 pointing just past the MMU trampoline in the
high page. (Suggested by Charles Hannum.)
- Set VBR to the kernel vector table just before turning on the MMU.
- Just before rebooting, set VBR to 0, which is what the BOOTROM expects
it to be.
kernel setroot(). The device type for network boot is set like any
other device now. Also, call the punit entry point for device drivers
to properly set `bootdev'.
The "options GENERIC" entry in kernel config files is not longer necessary
for "swap generic" kernels. Uses new config constructs which work with
some glue in an old config environment. This code will support new config
with minimal changes.
an unrecognized keyboard produced garbage on keypresses in the kernel,
but worked in the boot program (which has this default). The bug is
that the keymap pointers are unitialized.
section. Patch come up with by Bob Baron <rvb+@cs.cmu.edu> and myself.
This entire bit of code (the code which sets daddr/dsize and taddr/tsize)
is very bogus, but it's not clear what the 'right' way to fix it is
and this patch fixes a problem preventing some ELF executables from
being run.
mips_r2000_, mips_R2000_ -> mips1_
mips_r4000_, mips_R4000_ -> mips3_
(which are also, for mnemonic reasons, consistent with gcc flag usage,
rather than using mipsI_ and mipsIII_).
Where CACHED_TO_HYS() is still needed for kernel-virtual-to-physical
or physical-to-uncached mapping (fb drivers), replace with
`#include <mips/cpuregs.h>'.
* dcparam() with normal tty t_param interface, which calls
* cold_dcparam() called with explicit dc7085 register address
and flags, which does the work and is also callable when cold,
to set up console (or kgdb) line parameters.
* use mips_round_page,mips_trunc_seg() instead of
pica_round_page(),pica_trunc_page().
* discard (unused) return value from TLBUpdate(), and delete
(unused) temporary variable used to hold it.
* Move mips-I pte (TLBlo) definitions from pmax/include/pte.h
to mips/include/mips1_pte.h
* Move mips-III pte (TLBlo) definitions from pica/include/pte.h
to mips/include/mips3_pte.h
* Add new mips/include/pte.h, which includes exactly one of
mips1_pte.h or mips3_pte.h (which still have namespace collisions),
depending on "options MIPS1" or "options MIPS3". (hack).
Move soft kvtopte(), ptetovk() definitions to mips/include/pte.h
* Add macro PTE_TO_PADDR() to hide the different hardware TLB formats
when mapping from pte to physical address.
* Add macro PTE_READONLY() to hide lack of SW read-only bit in mips-III
tlb. (mips1 pmap uses a sw bit in the PTE, mips3 looks up RO bit in
the kernel pmap.)
* Use macros (not direct TLB frobbing) in mips/trap.c, to make it
mips-1/mips-III indepenndet.
* Change {pmax,pica}/include/pte.h to just do #include <mips/pte.h>.
jump-table) locore entrypoints, so the stack traceback code can use
the end marker to handle entry points specially when doing tracebacks,
even if it doesn't know about them explicitly.
* fold in netBSD style changes:
``if (foo = bar)'' -> ``if ((foo = bar) != 0)''
* add mips1 vector as well as mips3 vector
* change vector names to make mips1 and mips3 locore entrpoints distinct
* add pmax additions (e.g., old fixes to ktrace code, kernel-TLB_miss
instrumentation)
* update stack traceback code to a newer version of the pmax code.
add catch-all case, with distinct mips1 and mips3 ranges for locore
entry points, cases to catch othewise-unknown locore entrypoints and
vector code (which have special entry sequences and require special
support to trace through). The relevant mips1 and mips3 functions are
of course now distinct.
NetBSD-current changes:
* change include paths to be relative to the kernel-source tree
(e.g., `/sys') instead of arch/pmax/conf.
* add explicit options for exec packages (EXEC_ELF32, EXEC_ECOFF
for COMPAT_ULTRIX)
* comment out references to still-unsuppoted MI scsi.
DEC dc503 cursor chip) into Decstation 2100,3100 cfattach front-end
and ``machine-independent'' back-end.
pm_ds.c: pmin/pmax cfattach front-end
pm.c: bt478, 503 back-end
pmvar.h: declarations of back-end normal and console attach
entry points.
- change uiomove() argument from caddr_t to void *; this was the last caddr_t
in systm.h.
- remove parameter names from the b*() routines to make them look like the
mem*() counterparts.
screen geometry.
- Re-arrange et_loadmode() and et_inittextmode() to write into shadow
register set instead of writing to the card-registers directly.
- provide et_hwsave()/et_hwrest() functions to transfer the card-registers
to and from the shadow set.
ELF_ROUND (round to higher alignment boundary), and use them properly.
Also, change a bit of code in elf_load_psection to use the next ELF_ROUND
macro. This fixes a bug found by Robert Baron <rvb+@cs.cmu.edu> where
elf_load_psection, if given a properly aligned address at which to load
the section, would round actually load it at the next highest alignment
boundary.
Scenario: If ip_insertoptions() prepends a new mbuf to the chain, the
bad: label's m_freem(m0) still would free only the original mbuf chain
if the transmission failed for, e.g., no route to host; resulting in
one lost mbuf per failed packet. (The original posting included a
demonstration program).
Original report of this bug was by jinmei@isl.rdc.toshiba.co.jp
(JINMEI Tatuya) on comp.bugs.4bsd.
- const poisoning for the copy routines.
- fu* su* fixes:
XXX:
1. These are poorly named [better names should have real sizes in them]
2. Not all of them are in use implemented
3. Interface is broken; no way to detect errors in fu*
4. The kernel man 9 pages for copy, fetch and store leave a lot to be
desired.
* New patchable variable maxphysmem to limit physical memory.
That's useful to simulate small machines for testing install
kernels.
* Make debugging crashdumps with gdb -k work.
* microtime is moved to clock.c.
* Reserve memory for symbol table when ddb is configured.
* Move checks for RB_KDB into consinit.
* Fix a typo in delay().
* Disable generation of profiling prologue for proc_trampoline,
sigcode, idle, handle_trap, flush_icache, svc and interrupt.
* A few tweaks to make kgdb and ddb work nicer.
* Support for bpf added.
* Don't duplicate code, use ether_ifattach instead.
* Recode plipreceive and pliptransmit in assembler.
This improves performance from 40 kbyte/s to 50 kbyte/s
between a 30mhz pc532 and a 25mhz i386.
dev/microcode/aic7xxx_seq.h,
dev/ic/aic7xxxreg.h:
Remove intrinsic knowledge about SDTR and WDTR messages and replace it
with a generic message system that allows the kernel driver to handle
SDTR, WDTR and any other type of extended message it chooses too. This
makes the sequencer code much simpler, makes extended message handling
debuggable since the bulk of the work is in the kernel driver, and saves
lots of instruction space.
Regen microcode header file.
dev/ic/aic7xxx.c, dev/ic/aic7xxxvar.h:
Add code to handle WDTR and SDTR negotiation in light of the changes in
the message interface to the sequencer. Don't reject targets that
negotiate async by sending an SDTR with a 0 offset. Use an sdtr message
with 0,0 to negotiate async when a target suggests a period that is too
long for us to handle. Some tape and cdrom drives don't like us doing
the message reject that we did in the past.
Fix a problem with handing the QUEUE FULL condition.
Fix a race condition (most likely the cause of the SCB paging problems) that
might allow the sequencer to get unpaused before the condition that caused
it to be paused (a SEQINT) was handled.
Race condition pointed out by Doug Ledford <dledford@dialnet.net> and
by "Dan Willis" <dan@plutotech.com>.
dev/pci/ahc_pci.c:
Add support for the 2940AU, an aic7860 based controller.
dev/pci/pcidevs.h, dev/pci/pcidevs_data.h:
Add product IDs for the 2940AU, aic7860 and aic7855.
Regen data file.
scsi/scsi_message.h:
Add MSG_EXT_SDTR_LEN and MSG_EXT_WDTR_LEN - the length of bytes in these
extended messages.
Thanks to Chuck Cranor <chuck@maria.wustl.edu> for testing these changes
out for me.
for NOEXEC and NOSUID, and make sure the interpreter file is executable.
The mount point checks are done because, even though the interpreter
is not the program being 'executed', code from the interpreter is being
executed, and so the mount point's flags should be respected.
exceptions, which puts the address of the instruction we faulted
on in a different location. Copy it and handle as we normally would,
restoring the saved PC before returning.
The FPE should probably be reworked to take advantage of the 68LC040's
precalculated effective address, at some point.
to signal that the build is happening on a machine with an ELF
toolchain. This is temporary, until a better toolchain-recognition
scheme is worked out.
- Only pass user trace traps and breakpoints on to trap().
Gets rid of some hair in the trace/breakpoint trap cases.
- Before entering the debugger, switch to a temporary
stack so that the debugger can alter the stack pointer.
- Add glue for KGDB (still not complete).
Some other minor cleanup:
- Protect against some bad pointer derefs.
- Be more a little more verbose when a fatal trap
occurs to aid debugging.
- Only pass user trace traps and breakpoints on to trap().
- Before entering the debugger, switch to a temporary
stack so that the debugger can alter the stack pointer.
- Add glue for KGDB (still not complete).
Clearly mark the MMU enable trampoline code.
the keyboard to work. Fixes a bug where booting with `-d' worked
only on systems using a serial console.
While I'm here, eliminate some redundancy in the ite console intialization
code.
This fixes a critical bug where a clock interrupt would happen sometime
between the call to hp300_calibrate_delay() and when proc0 is initialized.
This ends up dereferencing a bad pointer in itimerdecr(), which scribbles
over the first page of kernel text, specifically vectors 46 and 47 (decimal).
To complicate matters, the way the bug manifested itself was different
depending on whether or not DDB was configured into the kernel. When
DDB is in the kernel, kernel text is mapped read/write. When DDB is not
in the kernel, kernel text is mapped read-only. Note that the kernel
scribble happens early, typically before the console is initialized.
In the non-DDB case, the kernel will hang as soon as it's loaded because
the access causes a fault (before the console is initialized, so you
don't see the trap).
In the DDB case, the access does _not_ cause a fault. However, the
mechanism used to enter the kernel debugger is to issue a "trap #15".
Conveniently, this is one of the corrupted vectors (47), thus rendering
DDB useless (it actually caused a recursive panic/trap loop).
This _WILL_ be in the first 1.2 official patch.
include files containing model-specific I/O device configuration.
Add more options and devices (ccd, LKM, etc), as appropriate for
documentation and as examples in a "template" GENERIC config file,
even though not all of these work completely yet.
and shell script support to be optional (conditioned on EXEC_SCRIPT).
Remove the implicit inclusion of EXEC_ECOFF when COMPAT_OSF1 and/or
COMPAT_ULTRIX is included, and of EXEC_ELF32 when COMPAT_LINUX and/or
COMPAT_SVR4 is included.
queue.h list/queue head initializer macros. mountlist was converted so
that panics (or other reboots) early on in kernel startup don't cause
sys_sync() to croak. vnode_free_list was converted because it was nearby.
- This driver supports the on-board mbus-based cgfourteen (sometimes referred
to as "SX") video hardware present on SS20-class machines.
- It does *not* support any of the SX acceleration features.
- It does support the 8-bit mode of the hardware, and looks to X like
a cgthree.
- It does support the cg6-style hardware cursor, even when running X in
cgthree emulation.
- It does support DPMS power-down of compatible displays on later-revision
cg14's.
- There is code to support the true color (32-bit) mode of the cg14 as
cg8 emulation, but it is disabled by default because it is most likely
broken. #define CG14_CG8 to turn it on.
The driver is not yet installed in the conf files, but I will do so
shortly...
Check only the IR bit of the CIAA Interrupt Control Register when testing
for a level 6 interrupt. An interrupt only occurs if IR is set, and IR is
only set if the individual mask bits are set. The individual interrupt
status bits can be set without causing an interrupt if the corresponding
enable bits are not set.
from arch/mips/mips/locore.S to arch/pmax/pmax/locore_machdep.S.
* Move ARC-specific locore code (vmstat -i counters) to
arch/pica/pica/locore_machdep.S.
* When the mips3 locore support is known to work, both ports can now use
arch/mips/mips/locore.S.
wrappers for:
gets.c getenv.c strcat.c strcpy.c
so they compile again (if _KERNEL is defined).
For bootblock space reasons it would be prereable to call the PROM entrypoints
directly, via the macro wappers in de_prom.h, rather than C function wrappers.
* Create arch/mips/Makefile.inc with source list of generic MIPS-cpu
files for tags
* Use mips/Makefile.inc and updated tag list in pmax/Makefile
* Try building bootblocks in arch/pmax/stand.
clock_attach() time (for now).
This removes our dependance on the DraCo ROM access timing and frees
the second CIA on Amigas.
b) support for DraCo rev. >= 4 native timer chips.
* When we are transferring in DATA (in asc_dma_in) and the target
is an async device, there is sometimes an extra byte in the FIFO.
If so, we need to drain that byte out of the fifo, but if and only
if the target is async. See also the comments in asc_dma_in()
in the related Mach mk84 asc driver (scsi_53C94_hdw.c), which
has an identical fix but applied in more restrictive conditions
than we need, with async *disk* targets, as well as async tapes.
* Add a watchdog and timeout active SCSI requests, to eliminate any
potential for deadlock due to applying the fix above on newer
silicon versions of the 53c94 which may not have the above problem.
Should use the MI scsi per-target timeout instead, when available.
for a level 6 interrupt. An interrupt only occurs if IR is set, and IR is
only set if the individual mask bits are set. The individual interrupt
status bits can be set without causing an interrupt if the corresponding
enable bits are not set.
NOTE: THESE FILES ARE NOW IDENTICAL TO THEIR ALPHA COUNTERPARTS.
The C preprocessor does the Right Thing when these files are built
on a SPARC. This makes it significantly easier to diff the two
versions (until a real MI 53c9x driver is done, hint hint).
* Handle message retransmissions and partially sent messages correctly.
* Make sure we clear ATN after the last message is sent.
* Do the right thing if the target initiates negotiation for async mode
after having negotiated sync mode.
* Fix some cases where we set ATN with no message queued, or schedule a
message without setting ATN.
* Issue a REQUEST SENSE after an unexpected disconnect, per SCSI spec.
* Fix abort handling in a number of cases.
* Recognize selection timeouts better (to speed up probing).
ELF-outputting version of the assembler. (It was dying when it saw
some CPP line number markers.) This is temporary. (Workaround suggested
by Matt Thomas.)
would generate two interrupts, one real and one spurious. The solution
is to force a drain of the SBus->MBus write buffers after writing to the
lance to clear the interrupt. Thanks to Chris Torek for pointing out a much
easier way to do this than I had planned...
these are hacks (s/long/int/, etc.), but this code really needs a heavy
cleaning (including fixed-size typing) and I don't have time to give it one
now.
consistency with the way machdep headers for other things are done.
(the creation of the ecoff_machdep.h files was done on the CVS server, to
keep the RCS logs intact.)
macros to use to remove #ifdefs from the machine ID case check.
Eventually, these headers will contain other information, e.g.
machine-dependent relocation information, etc.
use the pre-autoconfig cold serial-console inititialization entry
points from the bus-specific cfattach front-end code.
* Delete post-autoconfig code to switch from PROM output to kernel
driver for machines with dc serial chips.
cfattach code for TC SCSI option cards and ioasic 53c94 baseboard SCSI.
ascvar.h: shared softc declarations
asc_ioasic.c: ioasic front-end code.
asc_tc.c: Turbochannel option (and 5000/200 basebard) front-end code.
* ioasic_attach meeds more work to eliminate pmax_type dependency
and to verify the clocks speed passed to 53c94.
* Add prototypes for asc script entry points; should compile with
-Wstrict-prototypes -Wmissing-prototypes.
* Use tcvar.h interface. The usage of tc_syncbus() and tc_mb() may
not be quite stylistically on an Alhpa, but it apparently makes no
difference on the eerly-generation Alpha CPUs in TC Alphas.
cfattach front-end code:
dc_ds for the decstation 2100( pmin), 3100 (pmax), and 5100.
dc_ioasic for the decstation 5000/200 (3max) which does not have a
DEC TC-style IOCTL asic, but is configured as if it did.
* Add pre-autoconfig code initialization of kerenl dc driver console I/O
for remote serial consoles. The hack to use PROM serial I/O until the
dc device is autoconfigured is no longer necessary.
* Add MIPS make variable pointing at $S/arch/mips
* Build locore from $MIPS/mips locore source
* Build locore.o and fp.o (fp emulation) from arch/mips locore source
* Add target and rules to build pmax-specific locore code locore_machdep.o
from $PMAX/pmax/lcore_machdep.S
>* Use `-S' rather than `-x' to remove debugging symbols.
because the pmax toolchain and gcc disagree about what is a "local"
symbol, and ld requires a "-x" to let elf2aout build bootable a.out kernels.
to call mcount(). This is needed because the ``link a6,#0'' insn used
trips up gcc's ANSI preprocessor (A # in a function-type macro must be
followed by a macro argument). _PROF_PROLOG is also used in the i386
asm.h.
Solaris' asm_linkage.h has a MCOUNT macro similar to _PROF_PROLOG
except it expands to different code sequences based on whether a
function is being compiled with "prof" or "gprof" instrumentation.
I also discovered that the m68k ALTENTRY is very different than the
implementation used by other NetBSD ports. Usually ALTENTRY simply
provides an alternate function entry point. The m68k version takes a
second argument and jumps inside the second function when profiling is
enabled. The m68k behavior is similar to the ENTRY2 macro found in
solaris.
Providing ENTRY2 and changing all the code that uses ALTENTRY to use
it would be a desirable change.
mark