jhb [Tue, 17 Jun 2014 14:47:49 +0000 (14:47 +0000)]
Fix teardown of static DMA allocations in various NIC drivers:
- Add missing calls to bus_dmamap_unload() in et(4).
- Check the bus address against 0 to decide when to call
bus_dmamap_unload() instead of comparing the bus_dma map against NULL.
- Check the virtual address against NULL to decide when to call
bus_dmamem_free() instead of comparing the bus_dma map against NULL.
- Don't clear bus_dma map pointers to NULL for static allocations.
Instead, treat the value as completely opaque.
- Pass the correct virtual address to bus_dmamem_free() in wpi(4) instead
of trying to free a pointer to the virtual address.
emaste [Tue, 17 Jun 2014 13:12:30 +0000 (13:12 +0000)]
Add glyphs from converted syscons iso* fonts
This consists of the unique glyphs from the following font files in
/usr/share/syscons/fonts:
iso*.fnt ISO-8859-1 West European
iso02*.fnt ISO-8859-2 Central European
iso04*.fnt ISO-8859-4 Baltic
iso05*.fnt ISO-8859-5 Cyrillic
iso07*.fnt ISO-8859-7 Greek
iso08*.fnt ISO-8859-8 Hebrew
iso09*.fnt ISO-8859-9 Turkish
iso15*.fnt ISO-8859-15 West European
kib [Tue, 17 Jun 2014 09:33:22 +0000 (09:33 +0000)]
Do not reference native_lapic_ipi_*() functions in the UP build.
The functions' definitions are protected by #ifdef SMP.
Keeping apic_ops.ipi_*() methods NULL would allow to catch the use
on UP machines.
Reviewed by: royger
Sponsored by: The FreeBSD Foundation
kib [Tue, 17 Jun 2014 07:11:00 +0000 (07:11 +0000)]
In msdosfs_setattr(), add a check for result of the utimes(2)
permissions test, forgotten in r164033.
Refactor the permission checks for utimes(2) into vnode helper
function vn_utimes_perm(9), and simplify its code comparing with the
UFS origin, by writing the call to VOP_ACCESSX only once. Use the
helper for UFS(5), tmpfs(5), devfs(5) and msdosfs(5).
Reported by: bde
Reviewed by: bde, trasz
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
attilio [Mon, 16 Jun 2014 18:15:27 +0000 (18:15 +0000)]
- Modify vm_page_unwire() and vm_page_enqueue() to directly accept
the queue where to enqueue pages that are going to be unwired.
- Add stronger checks to the enqueue/dequeue for the pagequeues when
adding and removing pages to them.
Of course, for unmanaged pages the queue parameter of vm_page_unwire() will
be ignored, just as the active parameter today.
This makes adding new pagequeues quicker.
This change effectively modifies the KPI. __FreeBSD_version will be,
however, bumped just when the full cache of free pages will be
evicted.
pfg [Mon, 16 Jun 2014 14:55:09 +0000 (14:55 +0000)]
Update license to strptime(3) implementation.
Our strptime(3) implementation was the base for the illumos
implementation and after contacting the author, Kevin Rudy
stated the code is under a 2-Clause BSD License [1]
After reviewing our local changes to the file in question,
the FreeBSD Foundation has agreed that their contributions
to this file are not required to carry clause 3 or 4 so
the file can be relicensed as in Illumos [2].
mav [Mon, 16 Jun 2014 11:00:14 +0000 (11:00 +0000)]
Add support for VERIFY(10/12/16) and COMPARE AND WRITE SCSI commands.
Make data_submit backends method support not only read and write requests,
but also two new ones: verify and compare. Verify just checks readability
of the data in specified location without transferring them outside.
Compare reads the specified data and compares them to received data,
returning error if they are different.
VERIFY(10/12/16) commands request either verify or compare from backend,
depending on BYTCHK CDB field. COMPARE AND WRITE command executed in two
stages: first it requests compare, and then, if succeesed, requests write.
Atomicity of operation is guarantied by CTL request ordering code.
royger [Mon, 16 Jun 2014 08:48:42 +0000 (08:48 +0000)]
xen: unify gnttab initialization for PVHVM and PVH
Switch the initialization of gnttab to use an unused physical memory
range for both PVHVM and PVH.
In the past PVHVM was using the xenpci BAR, but there's no reason to
do that, and in fact FreeBSD was probably doing it because it was the
way it was done in Windows, were drivers cannot probably request for
unused physical memory ranges, but it was never enforced in the
hypervisor.
Sponsored by: Citrix Systems R&D
Approved by: gibbs
xen/gnttab.c:
- Allocate contiguous physical memory for grant table frames for both
PVHVM and PVH.
- Since gnttab is not a device, use the xenpv device in order to
request for this allocation.
dev/xen/xenpci/xenpcivar.h:
dev/xen/xenpci/xenpci.c:
- Remove the now unused xenpci_alloc_space and xenpci_alloc_space_int
functions.
xen/gnttab.h:
- Change the prototype of gnttab_init and gnttab_resume, that now
takes a device_t parameter.
dev/xen/control/control.c:
x86/xen/xenpv.c:
- Changes to accomodate the new prototype of gnttab_init and
gnttab_resume.
royger [Mon, 16 Jun 2014 08:44:33 +0000 (08:44 +0000)]
xen: introduce xenpv bus
Create a dummy bus so top level Xen devices can attach to it (instead
of attaching directly to the nexus). This allows to have all the Xen
related devices grouped under a single bus.
Sponsored by: Citrix Systems R&D
Approved by: gibbs
x86/xen/xenpv.c:
- Attach the xenpv bus when running as a Xen guest.
- Attach the ISA bus if needed, in order to attach syscons.
conf/files.amd6:
conf/files.i386:
- Include the xenpv.c file in the build of i386/amd64 kernels using
XENHVM.
dev/xen/console/console.c:
dev/xen/timer/timer.c:
xen/xenstore/xenstore.c:
- Attach to the xenpv bus instead of the Nexus.
dev/xen/xenpci/xenpci.c:
- Xen specific devices on PVHVM guests are no longer attached to the
xenpci device, they are instead attached to the xenpv bus, remove
the now unused methods.
royger [Mon, 16 Jun 2014 08:43:45 +0000 (08:43 +0000)]
xen: add hooks for Xen PV APIC
Create the necessary hooks in order to provide a Xen PV APIC
implementation that can be used on PVH. Most of the lapic ops
shouldn't be called on Xen, since we trap those operations at a higher
layer.
Sponsored by: Citrix Systems R&D
Approved by: gibbs
x86/xen/hvm.c:
x86/xen/xen_apic.c:
- Move IPI related code to xen_apic.c
x86/xen/xen_apic.c:
- Introduce Xen PV APIC implementation, most of the functions of the
lapic interface should never be called when running as PV(H) guest,
so make sure FreeBSD panics when trying to use one of those.
- Define the Xen APIC implementation in xen_apic_ops.
xen/xen_pv.h:
- Extern declaration of the xen_apic struct.
x86/xen/pv.c:
- Use xen_apic_ops as apic_ops when running as PVH guest.
conf/files.amd64:
conf/files.i386:
- Include the xen_apic.c file in the build of i386/amd64 kernels
using XENHVM.
amd64/include/cpu.h:
amd64/amd64/mp_machdep.c:
i386/include/cpu.h:
i386/i386/mp_machdep.c:
- Remove lapic_ipi_vectored hook from cpu_ops, since it's now
implemented in the lapic hooks.
amd64/amd64/mp_machdep.c:
i386/i386/mp_machdep.c:
- Use lapic_ipi_vectored directly, since it's now an inline function
that will call the appropiate hook.
x86/x86/local_apic.c:
- Prefix bare metal public lapic functions with native_ and mark them
as static.
- Define default implementation of apic_ops.
x86/include/apicvar.h:
- Declare the apic_ops structure and create inline functions to
access the hooks, so the change is transparent to existing users of
the lapic_ functions.
bryanv [Mon, 16 Jun 2014 04:32:27 +0000 (04:32 +0000)]
Force two byte alignment for all control message headers
The header structure consists of two 1-byte elements, but it must always
be describable by a single SG entry. Note for consistency, specify the
alignment everywhere, even if the structure has the appropriate natural
alignment since it contains a uint16_t.
bryanv [Mon, 16 Jun 2014 04:25:04 +0000 (04:25 +0000)]
Move the VIRTIO_RING_F_* defines out of virtqueue.h into virtio_config.h
These defines are applicable to userland too, but virtqueue.h contains
the kernel virtqueue interface, and is therefore not usable in userland.
Note that Linux places these defines in virtio_ring.h, but I don't want
the drivers including this header file to keep the VirtIO ring opaque to
everything but the virtqueue.
bryanv [Mon, 16 Jun 2014 04:16:31 +0000 (04:16 +0000)]
Remove kernel specific macro out of the VirtIO PCI header file
The eventual goal is to share this file with userland, so
remove the macro that is only specific for virtio_pci(4).
Instead, add the VIRTIO_PCI_CONFIG_OFF macro from Linux to
get the config size whether MSIX is enabled or not.
ian [Sun, 15 Jun 2014 13:45:37 +0000 (13:45 +0000)]
Create a mechanism for providing fine-grained build order dependencies
during SUBDIR_PARALLEL builds. This augments the coarse .WAIT mechanism,
which is still useful if you've got a situation such as "almost everything
depends on A and B".
Because the parallel subdir mechanism uses non-obvious mangling of
target names, which should probably remain a private detail of the
implementation, it's not easy to do things like "libfoo: libbar", so
instead the new mechanism lets you set a variable that lists dependencies:
SUBDIR_DEPEND_libfoo= libgroodah libpouet
Note that while I'm using libraries as an example here, it really has
nothing to do with the generated library files. This is really saying
"build in directory libfoo after building in the libgroodah and libpouet
directories."
This updates lib/Makefile with dependency information based on the old
almost-accurate comment block and by combing through lib/* makefiles
looking for LDADD dependencies to other libraries within lib/*.
alc [Sun, 15 Jun 2014 07:52:59 +0000 (07:52 +0000)]
One of the intentions behind r267254 was that the global variable "sgrowsiz"
would be read once and cached in a local variable so that the resource limit
check and map entry insertion would be guaranteed to use the same value.
However, the value being passed to vm_map_insert() is still from "sgrowsiz"
and not the local variable. Correct this oversight.
kib [Sun, 15 Jun 2014 05:01:35 +0000 (05:01 +0000)]
Fix some cosmetic issues with the use of kmem_malloc() in the i386 LDT
sysarch(2) code.
Use M_ZERO instead of explicit bzero(9). Do not check for failed
allocation when M_WAITOK is specified (which is specified always).
Use malloc(9) when allocating memory for the intermediate copy of the
user-supplied buffer.
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
kib [Sun, 15 Jun 2014 04:51:53 +0000 (04:51 +0000)]
Use vn_io_fault for the writes from core dumping code. Recursing into
VM due to copyin(9) faulting while VFS locks are held is
deadlock-prone there in the same way as for the write(2) syscall.
Reported and tested by: pho
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
pfg [Sun, 15 Jun 2014 03:54:23 +0000 (03:54 +0000)]
patch: unsign the line length to avoid overflows.
Patch(1) uses a short int for the line length, which is usually
sufficient for regular diffs, but makes no effort to signal
when there is an overflow.
Change the line length to an unsigned short int to better use
the fact that a length is never negative. The change is loosely
inspired on a related change in DragonFly, but we avoid spending
more memory than necessary.
While here adjust the messages to be clearer on what is happening.
gjb [Sat, 14 Jun 2014 01:03:24 +0000 (01:03 +0000)]
Switch release/Makefile knobs from WITHOUT_FOO=1 to MK_FOO=no.
This allows greater granular control over the release
build, and avoids WITH_FOO=1 and WITHOUT_FOO=1 collision.
This change was verified to produce correct results
compared to an earlier build by diffing find(1) output
of the resulting bootonly/, dvd/, and release/ directories.
Tested on: head@r267449
MFC after: 1 month
X-MFC-Note: Requires commits not yet MFC'd, so likely
longer than 1 month.
Sponsored by: The FreeBSD Foundation
jhb [Fri, 13 Jun 2014 19:40:02 +0000 (19:40 +0000)]
- Don't check the dma map address for a static DMA mapping against NULL
to determine if the mapping is valid.
- Don't explicitly destroy the dmap map for static DMA mappings.
tijl [Fri, 13 Jun 2014 08:44:03 +0000 (08:44 +0000)]
- Make invalids variable local to do_conv such that it prints the number
of invalid characters of the current file instead of an accumulated
value.
- Make do_conv return an error when invalid characters have been found.
Return EXIT_FAILURE from main if any file contained invalid characters.
This matches the behaviour of GNU iconv.
- Mark usage with __dead2 attribute.
- Make the long_options array const.
pfg [Thu, 12 Jun 2014 19:01:57 +0000 (19:01 +0000)]
Avoid zeroing during allocation.
This change reverts a change from OpenBSD which made use of
calloc, and therefore wasted time initializing arrays that
will later be realloc'ed. Consistently use FreeBSD's
reallocf().
While here also merge the changes from OpenBSD's manpage
patch.1 Rev 1.27:
"patch was moved from user portability (UP) to base in issue 7
and is no longer optional"
jmmv [Thu, 12 Jun 2014 18:54:52 +0000 (18:54 +0000)]
Enable MK_TESTS by default.
This was never intended to be off by default but was done this way
because the initial patches broke the build. Things seem stable now
(have been so for a while too) and "make tinderbox" is clean so let's
try again.
Announced in freebsd-current; all reported shortcomings have been
addressed.
emaste [Thu, 12 Jun 2014 18:31:32 +0000 (18:31 +0000)]
Build vt(4) fonts during buildworld
vtfontcvt(8) is now built during buildworld, so can be used as a
bootstrap tool to create vt(4) fonts from source .hex or .bdf font
files, rather than having uuencoded binary fonts in the tree.
jmg [Thu, 12 Jun 2014 16:31:15 +0000 (16:31 +0000)]
clear the write bit... This allows my AVILA board to survive a
portsnap extract, where previously it would panic.. clearly someone
who knows pmap should optimize this code per alc's comment...
mav [Thu, 12 Jun 2014 12:43:48 +0000 (12:43 +0000)]
Implement simple direct-mapped cache for popular filesystem identifiers to
avoid congestion on global mountlist_mtx mutex in vfs_busyfs(), while
traversing through the list of mount points.
This change significantly improves NFS server scalability, since it had
to do this translation for every request, and the global lock becomes quite
congested.
This code is more optimized for relatively small number of mount points.
On systems with hundreds of active mount points this simple cache may have
many collisions. But the original traversal code in that case should also
behave much worse, so we are not loosing much.