Commit graph

740 commits

Author SHA1 Message Date
Ben Gras
d69cc76e03 rename struct mem_range to minix_mem_range
. avoid a name clash with gdb
2012-08-15 15:17:25 +02:00
David van Moolenbroek
4aec9db571 dp8390: resolve Coverity warnings (encore) 2012-08-14 09:11:18 +00:00
David van Moolenbroek
bc404a9401 TTY: resolve Coverity warnings 2012-08-14 09:11:17 +00:00
David van Moolenbroek
533f61249b random: resolve Coverity warnings 2012-08-14 09:11:16 +00:00
Arun Thomas
93eadc7040 Make memory driver arch-neutral
-Use PAGE_SIZE instead of I386_PAGE_SIZE
-No need to grant ioport access for ARM
2012-08-13 17:21:47 +02:00
David van Moolenbroek
cff95276d9 pci: resolve Coverity warnings
Or rather, slightly improve the code based on a false positive.
2012-08-06 16:18:40 +00:00
David van Moolenbroek
2289c91010 lance: resolve Coverity warnings 2012-08-06 16:18:39 +00:00
David van Moolenbroek
431cd9f69c fxp: resolve Coverity warnings
This removes some unused and unsupported features, and leaves in support
for another currently unused feature that might be used in the future.
2012-08-06 16:18:38 +00:00
David van Moolenbroek
c984359bb0 floppy: resolve Coverity warnings 2012-08-06 16:18:37 +00:00
David van Moolenbroek
4f877afeb6 e1000: resolve Coverity warnings 2012-08-06 16:18:36 +00:00
David van Moolenbroek
6199f090c8 dpeth: resolve Coverity warnings 2012-08-06 16:18:35 +00:00
David van Moolenbroek
6e3824d78b dp8390: resolve Coverity warnings 2012-08-06 16:18:34 +00:00
David van Moolenbroek
ab6e62a523 dec21140A: resolve Coverity warnings 2012-08-06 16:18:33 +00:00
Arun Thomas
19ffad7692 Remove ACK EM_WSIZE/EM_PSIZE macro usage 2012-08-06 17:49:22 +02:00
Arun Thomas
6723dcfab7 Replace MACHINE/CHIP macros with compiler macros 2012-08-06 17:49:22 +02:00
David van Moolenbroek
90e2b939ab rtl8139: resolve Coverity warnings 2012-07-30 13:59:43 +00:00
David van Moolenbroek
e34f611be3 rtl8169: resolve Coverity warnings 2012-07-30 12:10:15 +00:00
David van Moolenbroek
db7a45926b filter: resolve Coverity warnings 2012-07-30 12:10:11 +00:00
David van Moolenbroek
da915d36b7 atl2: resolve Coverity warnings 2012-07-30 12:10:10 +00:00
David van Moolenbroek
0ae9652177 at_wini: resolve Coverity warnings 2012-07-30 12:10:09 +00:00
David van Moolenbroek
d75ef5a05c ahci: resolve Coverity warnings 2012-07-30 12:10:08 +00:00
Ben Gras
b6ea15115c kernel: facility for user-visible memory
. map all objects named usermapped_*.o with globally visible
	  pages; usermapped_glo_*.o with the VM 'global' bit on, i.e.
	  permanently in tlb (very scarce resource!)
	. added kinfo, machine, kmessages and loadinfo for a start
	. modified log, tty to make use of the shared messages struct
2012-07-28 20:57:38 +00:00
David van Moolenbroek
8caec1b57b libsys: 64-bit numbers support for printf()
Change some drivers accordingly.
2012-07-26 09:45:05 +00:00
Ben Gras
cbcdb838f1 various coverity-inspired fixes
. some strncpy/strcpy to strlcpy conversions
	. new <minix/param.h> to avoid including other minix headers
	  that have colliding definitions with library and commands code,
	  causing parse warnings
	. removed some dead code / assignments
2012-07-16 14:00:56 +02:00
Ben Gras
50e2064049 No more intel/minix segments.
This commit removes all traces of Minix segments (the text/data/stack
memory map abstraction in the kernel) and significance of Intel segments
(hardware segments like CS, DS that add offsets to all addressing before
page table translation). This ultimately simplifies the memory layout
and addressing and makes the same layout possible on non-Intel
architectures.

There are only two types of addresses in the world now: virtual
and physical; even the kernel and processes have the same virtual
address space. Kernel and user processes can be distinguished at a
glance as processes won't use 0xF0000000 and above.

No static pre-allocated memory sizes exist any more.

Changes to booting:
        . The pre_init.c leaves the kernel and modules exactly as
          they were left by the bootloader in physical memory
        . The kernel starts running using physical addressing,
          loaded at a fixed location given in its linker script by the
          bootloader.  All code and data in this phase are linked to
          this fixed low location.
        . It makes a bootstrap pagetable to map itself to a
          fixed high location (also in linker script) and jumps to
          the high address. All code and data then use this high addressing.
        . All code/data symbols linked at the low addresses is prefixed by
          an objcopy step with __k_unpaged_*, so that that code cannot
          reference highly-linked symbols (which aren't valid yet) or vice
          versa (symbols that aren't valid any more).
        . The two addressing modes are separated in the linker script by
          collecting the unpaged_*.o objects and linking them with low
          addresses, and linking the rest high. Some objects are linked
          twice, once low and once high.
        . The bootstrap phase passes a lot of information (e.g. free memory
          list, physical location of the modules, etc.) using the kinfo
          struct.
        . After this bootstrap the low-linked part is freed.
        . The kernel maps in VM into the bootstrap page table so that VM can
          begin executing. Its first job is to make page tables for all other
          boot processes. So VM runs before RS, and RS gets a fully dynamic,
          VM-managed address space. VM gets its privilege info from RS as usual
          but that happens after RS starts running.
        . Both the kernel loading VM and VM organizing boot processes happen
	  using the libexec logic. This removes the last reason for VM to
	  still know much about exec() and vm/exec.c is gone.

Further Implementation:
        . All segments are based at 0 and have a 4 GB limit.
        . The kernel is mapped in at the top of the virtual address
          space so as not to constrain the user processes.
        . Processes do not use segments from the LDT at all; there are
          no segments in the LDT any more, so no LLDT is needed.
        . The Minix segments T/D/S are gone and so none of the
          user-space or in-kernel copy functions use them. The copy
          functions use a process endpoint of NONE to realize it's
          a physical address, virtual otherwise.
        . The umap call only makes sense to translate a virtual address
          to a physical address now.
        . Segments-related calls like newmap and alloc_segments are gone.
        . All segments-related translation in VM is gone (vir2map etc).
        . Initialization in VM is simpler as no moving around is necessary.
        . VM and all other boot processes can be linked wherever they wish
          and will be mapped in at the right location by the kernel and VM
          respectively.

Other changes:
        . The multiboot code is less special: it does not use mb_print
          for its diagnostics any more but uses printf() as normal, saving
          the output into the diagnostics buffer, only printing to the
          screen using the direct print functions if a panic() occurs.
        . The multiboot code uses the flexible 'free memory map list'
          style to receive the list of free memory if available.
        . The kernel determines the memory layout of the processes to
          a degree: it tells VM where the kernel starts and ends and
          where the kernel wants the top of the process to be. VM then
          uses this entire range, i.e. the stack is right at the top,
          and mmap()ped bits of memory are placed below that downwards,
          and the break grows upwards.

Other Consequences:
        . Every process gets its own page table as address spaces
          can't be separated any more by segments.
        . As all segments are 0-based, there is no distinction between
          virtual and linear addresses, nor between userspace and
          kernel addresses.
        . Less work is done when context switching, leading to a net
          performance increase. (8% faster on my machine for 'make servers'.)
	. The layout and configuration of the GDT makes sysenter and syscall
	  possible.
2012-07-15 22:30:15 +02:00
Thomas Veerman
39dfb2bc55 Don't build ext2 ramdisk for cross compilation
. as mkfs.mfs is now cross compilable, we can remove the ext2 ramdisk
  special case for cross compilation.
2012-06-18 10:54:56 +00:00
Thomas Veerman
aab6eb7615 Clean up generated files for ash,ramdisk,libutil,libc 2012-06-18 10:54:56 +00:00
Thomas Veerman
791aa4e16f Cross compile MFS ramdisk 2012-06-18 10:54:55 +00:00
Thomas Veerman
03ea1c1676 Dynamically determine which ramdisk to use
.Also, rename ext2_ramdisk to ramdisk_ext2.
2012-06-18 10:53:36 +00:00
Thomas Veerman
f09c2e014f Use MACHINE_ARCH instead of ARCH 2012-06-18 10:53:35 +00:00
Thomas Veerman
f93afa00e9 Remove MINIXSRCDIR and use NETBSDSRCDIR
NETBSDSRCDIR is used all over the place anyway, and this reduces
our diff with NetBSD a little.
2012-06-18 10:53:35 +00:00
Thomas Veerman
d1301b26b3 Use buildsystem commands instead of {cp,mkdir} 2012-06-18 10:53:33 +00:00
Thomas Veerman
acac1f2fda Cross-compilation fixes
. Some Makefile fixes to automatically differentiate between a normal
  compilation and cross-compilation. Also, build compressed images.
. Harmonize ramdisk rc scripts for normal use case and ext2 ramdisk.
. ext2_ramdisk filesystem prototype fixes.
2012-06-18 10:53:32 +00:00
Thomas Veerman
0bcc4c15c9 newroot is gone too 2012-06-18 10:53:31 +00:00
Thomas Veerman
d0c7c2a5d2 bios_wini is no longer amongst us 2012-06-18 10:53:31 +00:00
Dirk Vogt
67ff0d71ed ext2 ramdisk: copy rc to correct destination 2012-06-18 10:53:28 +00:00
Dirk Vogt
56f6a77ead ramdisk: don't do anything at dependall 2012-06-18 10:53:27 +00:00
Dirk Vogt
ed4df03113 Include ext2 ramdisk in build and remove sync
As the ramdisk and ext2 ramdisk don't build anything for the install
and all targets, memory does not have to wait for them anymore.
2012-06-18 10:53:26 +00:00
Dirk Vogt
bfacef76c5 Add ext2 ramdisk, intended for cross compilation 2012-06-18 10:53:25 +00:00
Dirk Vogt
a8f6f3ebc9 Make ramdisk location configurable
Also, the ramdisk is now built when the memory driver is compiled.
2012-06-18 10:53:25 +00:00
Dirk Vogt
c2c39af6bb Make TTY cross-compilable 2012-06-18 10:53:23 +00:00
Ben Gras
2bfeeed885 drop segment from safecopy invocations
. all invocations were S or D, so can safely be dropped
	  to prepare for the segmentless world
	. still assign D to the SCP_SEG field in the message
	  to make previous kernels usable
2012-06-16 16:22:51 +00:00
Ben Gras
73379f0be5 set major version of shlibs to 0
WARNING: this will break existing dynamically linked binaries if they
exist. If you have any:
	. re-build world statically first if necessary
	. remove libraries from /lib and /usr/lib
	. then build world

This change:
	. avoids possible future dismay when interfacing other
	  systems' binaries; done until they are abi-compatible

Thanks to Antoine Leca for pointing this out.
2012-05-10 16:38:24 +02:00
Ben Gras
cfb2d7bca5 retire BIOS_SEG and umap_bios
. readbios call is now a physical copy with range check in
	  the kernel call instead of BIOS_SEG+umap_bios
	. requires all access to physical memory in bios range to go
	  through sys_readbios
	. drivers/dpeth: wasn't using it
	. adjusted printer
2012-05-09 19:03:59 +02:00
Ben Gras
b611e4c226 brazilian keymap
contributed by David Augusto.
2012-05-04 13:56:04 +02:00
Ben Gras
b332803b6f release fixes
. make ramdisk buildable without ../etc having pwd.db
	. add cat to release bootstrap cmds
	. support running dynamically linked executables for
	  release bootstrap cmds
	. import netbsd chroot to help
2012-04-17 16:58:58 +02:00
Ben Gras
53f94f8ed4 dynamic executables on ramdisk support
See UPDATING about upgrading clang for dynamic linking.

	. allow executables on ramdisk to be dynamically linked; this means
	  putting a few required shared libraries and ld.elf_so on the ramdisk.
	. this makes the ramdisk (usage) smaller when they are dynamic, but
	  bigger when they're not.
	. also we can safely ditch newroot and call mount directly as that is
	  all newroot does.
	. create proto.common to share a bunch of entries between
	  small/nonsmall cases
2012-04-16 14:06:09 +02:00
Ben Gras
4b999f1962 build shared versions of libraries
building defaults to off until clang is updated.

current clang does not handle -shared, necessary to change the ld
invocation to build shared libraries properly. a new clang should be
installed and MKPIC defaults to no unless the newer clang is detected.

changes:

	. mainly small imports of a Makefile or two and small fixes
	  (turning things back on that were turned off in Makefiles)
	. e.g.: dynamic librefuse now depends on dynamic
	  libpuffs, so libpuffs has to be built dynamically too
	  and a make dependency barrier is needed in lib/Makefile
	. all library objects now have a PIC (for .so) and non-PIC
	  version, so everything is built twice.
	. generate PIC versions of the compat (un-RENAMEd) jump files,
	  include function type annotation in generated assembly
	. build progs with -static by default for now
	. also build ld.elf_so
	. also import NetBSD ldd
2012-04-16 05:21:20 +02:00
Thomas Veerman
ca7a466f48 TTY: don't allow multiple readers on tty minor
TTY has no way of keeping track of multiple readers for a tty minor
device. Instead, it stores a read request for the last reader only.
Consequently, the first ("overwritten") reader gets stuck on a read
request that's never going to be finished. Also, the overwriting
causes a grant mismatch in VFS when TTY returns a reply for the
second reader.

This patch is a work around for the actual problem (i.e., keeping track
of multiple readers). It checks whether there is a read operation in
progress and returns an error if it is --preventing that reader from
getting overwritten and stuck. It fixes a bug triggered by executing
'top | more' and pressing the space bar for a while (easily reproducable
in a VM, not on hardware).
2012-04-13 13:22:13 +00:00
Ben Gras
511c977f93 retire bios_wini 2012-04-12 03:02:21 +02:00