John Baldwin [Wed, 6 Feb 2013 17:06:51 +0000 (17:06 +0000)]
Rework the handling of stop signals in the NFS client. The changes in
195702, 195703, and 195821 prevented a thread from suspending while holding
locks inside of NFS by forcing the thread to fail sleeps with EINTR or
ERESTART but defer the thread suspension to the user boundary. However,
this had the effect that stopping a process during an NFS request could
abort the request and trigger EINTR errors that were visible to userland
processes (previously the thread would have suspended and completed the
request once it was resumed).
This change instead effectively masks stop signals while in the NFS client.
It uses the existing TDF_SBDRY flag to effect this since SIGSTOP cannot
be masked directly. Also, instead of setting PBDRY on individual sleeps,
the NFS client now sets the TDF_SBDRY flag around each NFS request and
stop signals are masked for all sleeps during that region (the previous
change missed sleeps in lockmgr locks). The end result is that stop
signals sent to threads performing an NFS request are completely
ignored until after the NFS request has finished processing and the
thread prepares to return to userland. This restores the behavior of
stop signals being transparent to userland processes while still
preventing threads from suspending while holding NFS locks.
Make sure that all mouse buttons are released when clients
using /dev/consolectl close. This fixes a problem where if
a USB mouse is detached while a button is pressed, that
button is never released.
Gleb Smirnoff [Wed, 6 Feb 2013 07:27:25 +0000 (07:27 +0000)]
Fixes to QUEUE_MACRO_DEBUG support:
- Add const quilifiers to fields that store value of __FILE__.
- Use long type for fields that store value of __LINE__.
- Sort and style(9) debugging fields.
- Add initializer for debugging fields into TAILQ_INITIALIZER macro.
Neel Natu [Wed, 6 Feb 2013 04:53:00 +0000 (04:53 +0000)]
Compute the number of initial kernel page table pages (NKPT) dynamically.
This eliminates the need to recompile the kernel when the default value
of NKPT is not big enough - for e.g. when loading large kernel modules
or memory disk images from the loader.
If NKPT is defined in the kernel configuration file then it overrides the
dynamic calculation.
Andrew Turner [Tue, 5 Feb 2013 20:03:58 +0000 (20:03 +0000)]
* Add the integer div & mod functions and ARM EABI support functions to
libstand.
* Stop linking the ARM U-Boot loader against libgcc now libstand has the
required symbols.
Pyun YongHyeon [Tue, 5 Feb 2013 00:37:45 +0000 (00:37 +0000)]
Rework jumbo frame handling. QAC confirmed that the controller
requires 8 bytes alignment on RX buffer. Given that non-jumbo
frame works on any alignments I guess this DMA limitation for RX
buffer could be jumbo frame specific one. Also I'm not sure
whether this DMA limitation is related with 64bit DMA. Previously
age(4) disabled 64bit DMA addressing due to silent data corruption.
So we may need more testing on re-enabling 64bit DMA in future.
While I'm here, change mbuf chaining algorithm to use fixed sized
buffer and force software checksum if controller reports length
error. According to QAC, RFD is not updated at all for jumbo frame
so it works just like alc(4) controllers. This change also added
alignment fixup for strict alignment architectures. Because I'm
not aware of any non-x86 machines that use age(4) controllers it's
just for completeness at this moment.
Wit this change, jumbo frame should work with age(4).
Tested by: Christian Gusenbauer < c47g <> gmx dot at >
MFC after: 1 week
Andrew Turner [Mon, 4 Feb 2013 10:05:55 +0000 (10:05 +0000)]
Cast *tabchar, a wchar_t, to a wint_t as it is the type the %lc printf
format string expects. This is only an issue on ARM EABI where wint_t is
different to wchar_t.
Andrew Turner [Mon, 4 Feb 2013 09:28:36 +0000 (09:28 +0000)]
Allow the unwind functions int libgcc_s to interact correctly with libthr.
_Unwind_ForcedUnwind in libgcc_s takes as one of it's parameters a stop
function to tell it when to stop unwinding. One of the stop function's
parameters is a _Unwind_Exception_Class. On most architectures this is an
int64_t, however on ARM EABI the gcc developers have made this a char array
with 8 items. While both of these take the same space they are passed into
the stop function differently, an int64_t is passed in in registers r2 and
r3, while the char[8] is passed in as a pointer to the first item in
register r2.
Because libthr expects the value to be an int64_t we would get incorrect
results when it passes a function that take an int64_t but libgcc passes in
a pointer to a char array including crashing.
The fix is to update libgcc_s to make it pass an int64_t to the stop
function and to libstdc++ as it expects _Unwind_Exception_Class to be an
array.
Andrew Turner [Mon, 4 Feb 2013 06:59:33 +0000 (06:59 +0000)]
Extend GDB to check the value in the .note.tag section along with the
.note.ABI-tag section.
This helps on ARM EABI where the OS/ABI field is zero. It would be better
to use the NOTES program header however this would require a more invasive
change.
Dimitry Andric [Sun, 3 Feb 2013 20:40:41 +0000 (20:40 +0000)]
Add several missing symbols to libcxxrt's symbol version map, and remove
a few duplicates. This should fix building world with -stdlib=libc++
after r246028.
crunchide(1): support non-custom elf object layout
The crunchide utility presumes the last 3 chunks of an ELF object
layout are section headers, symbol table, and then string table.
However, this is not specified in the ELF standards, and linkers
may generate different layouts when doing partial linking (-r).
This change is required to build FreeBSD with mclinker or the
gold linker.
Andriy Gapon [Sun, 3 Feb 2013 18:42:20 +0000 (18:42 +0000)]
zfs: fix, improve and re-organize page_lookup and page_unlock
Now they are split into two pairs: page_hold/page_unhold for mappedread
and page_busy/page_unbusy for update_pages.
For mappedread we simply hold a page that is to be used as a source if it
is resident and valid (and not busy). This is sufficient since we are
only doing page -> user buffer copying. There is no page <-> backing
storage I/O involved.
update_pages is now better split to properly handle the putpages case
(page -> arc) and the regular write case (arc -> page).
For the latter we use complete protocol of marking an object with
paging-in-progress and marking a page with io_start (busy count).
Also, in this case we remove the write bit from all page mappings and
clear dirty bits of the pages, the former is needed to ensure that the
latter does the right thing.
Additionally we update a page if it is cached instead of just freeing it
as was done before. This needs to be verified.
A minor detail: ZFS-backed pages should always be either fully valid
or fully invalid. Assert this and use simpler API that does not deal
with sub-page blocks.
Kirk McKusick [Sun, 3 Feb 2013 17:16:32 +0000 (17:16 +0000)]
For UFS2 i_blocks is unsigned. The current "sanity" check that it
has gone below zero after the blocks in its inode are freed is a
no-op which the compiler fails to warn about because of the use of
the DIP macro. Change the sanity check to compare the number of
blocks being freed against the value i_blocks. If the number of
blocks being freed exceeds i_blocks, just set i_blocks to zero.
Reported by: Pedro Giffuni (pfg@)
MFC after: 2 weeks
Jilles Tjoelker [Sun, 3 Feb 2013 15:54:57 +0000 (15:54 +0000)]
sh: Expand here documents in the current process.
Expand here documents at the same point other redirections are expanded but
use a non-fork subshell environment (like simple command substitutions) for
compatibility. Substitition errors result in an empty here document like
before.
As a result, a fork is avoided for short (<4K) expanded here documents.
Unexpanded here documents (with quoted end marker after <<) are not affected
by this change. They already only forked when >4K.
Side effects:
* Order of expansion is slightly different.
* Slow expansions are not executed in parallel with the redirected command.
* A non-fork subshell environment is subtly different from a forked process.
Hiroki Sato [Sun, 3 Feb 2013 10:26:24 +0000 (10:26 +0000)]
- Add CHECKSUM.* support in Makefile[1].
- Use ln -fs to create a symlink.
- Remove pkgadd for docports.
- Use WITHOUT_JADETEX=yes instead of WITH_JADETEX=no.
- Add {WORLD,KERNEL}_FLAGS to [BTWK]MAKE.
- Use makefs(8) and gpart(8) for sparc64 ISO image[2].
- Add publisher option to makefs(8)[2].
Based on work by: gjb[1]
Discussed with: marius, nwhitehorn[2]
Andriy Gapon [Sun, 3 Feb 2013 09:57:39 +0000 (09:57 +0000)]
allow for large KTR_ENTRIES values by allocating ktr_buf using malloc(9)
Only during very early boot, before malloc(9) is functional (SI_SUB_KMEM),
the static ktr_buf_init is used. Size of the static buffer is determined
by a new kernel option KTR_BOOT_ENTRIES. Its default value is 1024.
Tim Kientzle [Sun, 3 Feb 2013 01:08:01 +0000 (01:08 +0000)]
Another overhaul of the CPSW driver for BeagleBone
Major changes:
* Finally tracked down the flow control setting that
seems to have been causing TX stalls and watchdog timeouts
* RX and TX paths now share a lot more code
* TX interrupt is no longer used; we instead GC finished
tx queue entries at the bottom of the start routine.
* TX start now queues fragmented packets directly; it only
invokes defrag() for occasional very fragmented packets.
* "sysctl dev.cpsw" dumps controller statistics and queue counts
* Host Error Interrupt will give extensive debugging information
if the controller chokes on the queued data.
Justin Hibbits [Sun, 3 Feb 2013 00:19:34 +0000 (00:19 +0000)]
Fix the PowerPC DTrace copy functions. The kernel doesn't hold the same view to
the user map, so use the md copy in/out functions provided by the kernel.
Dimitry Andric [Sat, 2 Feb 2013 22:28:29 +0000 (22:28 +0000)]
Pull in r170135 from upstream clang trunk:
Dont use/link ARCMT, StaticAnalyzer and Rewriter to clang when the user
specifies not to. Dont build ASTMatchers with Rewriter disabled and
StaticAnalyzer when it's disabled.
Without all those three, the clang binary shrinks (x86_64) from ~36MB
to ~32MB (unstripped).
To disable these clang components, and get a smaller clang binary built
and installed, set WITHOUT_CLANG_FULL in src.conf(5). During the
initial stages of buildworld, those extra components are already
disabled automatically, to save some build time.
- Remove unused extern declarations in fs.h
- Correct comments in ext2_dir.h
- Several panic() messages showed wrong function names.
- Remove commented out stray line in ext2_alloc.c.
- Remove the unused macro EXT2_BLOCK_SIZE_BITS() and the then
write-only member e2fs_blocksize_bits from struct m_ext2fs.
- Remove the unused macro EXT2_FIRST_INO() and the then write-only
member e2fs_first_inode from struct m_ext2fs.
- Remove EXT2_DESC_PER_BLOCK() and the member e2fs_descpb from
struct m_ext2fs.
- Remove the unused members e2fs_bmask, e2fs_dbpg and
e2fs_mount_opt from struct m_ext2fs
- Correct harmless off-by-one error for fspath in ext2_vfsops.c.
- Remove the unused and broken macros EXT2_ADDR_PER_BLOCK_BITS()
and EXT2_DESC_PER_BLOCK_BITS().
- Remove the !_KERNEL versions of the EXT2_* macros.
Marius Strobl [Sat, 2 Feb 2013 21:57:06 +0000 (21:57 +0000)]
Improve r238673 to additionally allow for odd-aligned buffers as
passed in by smartd of smartmontools.
While at it, hint the compiler that 32-bit PIO is the most likely
case (idea from Linux) and use bus_{read,write}_stream_2(9) instead
of bus_{read,write}_multi_stream_2(9) for single count reads/writes.
Andriy Gapon [Sat, 2 Feb 2013 11:58:35 +0000 (11:58 +0000)]
print compiler version in the kernel banner
And provide kernel compiler version as a sysctl as well.
This is useful while we have gcc and clang cohabitation.
This could be even more useful when we have support
for external toolchains.
Andriy Gapon [Sat, 2 Feb 2013 11:38:26 +0000 (11:38 +0000)]
uart: add resume method and enable it for attachments on the most common
x86 buses
Otherwise the uart hardware could be in such a state after the resume
where IER is cleared and thus no interrupts are generated.
This behavior is observed and tested with QEMU, so I am comitting this
change to help with my debugging.
There has been no feedback from users of serial ports on real hardware.
The MSDOSFSMNT_WAITONFAT flag is bogus and broken. It does less than
track the MNT_SYNCHRONOUS flag. It is set to the latter at mount time
but not updated by MNT_UPDATE.
Use MNT_SYNCHRONOUS to decide to write the FAT updates syncrhonously.
Backup FATs were sometimes marked dirty by copying their first block
from the primary FAT, and then they were not marked clean on unmount.
Force marking them clean when appropriate.
The directory entry for dotdot was corrupted in the FAT32 case when moving
a directory to a subdir of the root directory from somewhere else.
For all directory moves that change the parent directory, the dotdot
entry must be fixed up. For msdosfs, the root directory is magic for
non-FAT32. It is less magic for FAT32, but needs the same magic for
the dotdot fixup. It didn't have it.
Both chkdsk and fsck_msdosfs fix the corrupt directory entries with no
problems.
The fix is to use the same magic for dotdot in msdosfs_rename() as in
msdosfs_mkdir().
For msdosfs_mkdir(), document the magic. When writing the dotdot entry
in mkdir, use explicitly set pcl variable instead on relying on the
start cluster of the root directory typically has a value < 65536.
The mountmsdosfs() function had an insane sanity test, remove it.
Trying FAT32 on a small partition failed to mount because
pmp->pm_Sectors was nonzero. Normally, FAT32 file systems are so
large that the 16-bit pm_Sectors can't hold the size. This is
indicated by setting it to 0 and using only pm_HugeSectors. But at
least old versions of newfs_msdos use the 16-bit field if possible,
and msdosfs supports this except for breaking its own support in the
sanity check. This is quite different from the handling of pm_FATsecs
-- now the 16-bit value is always ignored for FAT32 except for
checking that it is 0, and newfs_msdos doesn't use the 16-bit value
for FAT32.
Neel Natu [Fri, 1 Feb 2013 16:58:59 +0000 (16:58 +0000)]
Add support for MSI-X interrupts in the virtio block device and make that
the default.
The current behavior of advertising a single MSI vector can be requested by
setting the environment variable "BHYVE_USE_MSI" to "yes". The use of MSI
is not compliant with the virtio specification and will be eventually phased
out.
Assert that the mbuf in the chain has sane length. Proper place for
this check is somewhere in the network code, but this assertion
already proven to be useful in catching what seems to be driver bugs
causing NFS scrambling random memory.
The change to reduce default smp_tsc_shift caused tsc shift to become
zero on slower machines, which make the fenced get_timecount methods
not used despite needed. Remove the (shift > 0) condition when
selecting the get_timecount() implementation.
Rename smp_tsc_shift to tsc_shift, and apply it for the UP case too.
Allow shift to reach value of 31 instead of 30, as it was previously
(should be nop).
Reorganize the tc quality calculation to remove the conditionally
compiled block. Rename test_smp_tsc() to test_tsc() and provide
separate versions for SMP and UP builds. The check for virtialized
hardware is more natural to perform in the smp version of the
test_tsc(), since it is only done for smp case.
Noted and reviewed by: bde (previous version)
MFC after: 12 days
Pietro Cerutti [Fri, 1 Feb 2013 13:04:06 +0000 (13:04 +0000)]
- Fix more style(9)-related issues (copyright header, spaces after function
names, unnecessary casts)
- Change type of boolean variable from char to bool