Fixing the IP accounting issue, if we plan to do so, needs to be better
thought out; the 'fix' introduces a hash lookup and a possible kernel panic.
Reported by: Mark Tinguely
Either they're there early and the ispfw sets have
registered themselves, or they're not.
The module dependency stuff isn't quite what we want
anyway. If the user doesn't want the load placed on
system memory by loading the firmware, they don't
specify it to be loaded (either by being linked in
or via being a module to be loaded and then hooked
in with firmware(9)). It doesn't then make sense to
then override what they want by pulling it in anyway.
This might be able to work if we were able to pull in
just exactly what we needed for the card we have- but
that's an optimization left for the future.
at which the kernel should start allocating physical memory. The primary
purpose of this is to test 64-bit cleanness of the data path by setting
hw.physmemstart=4G so that all physical allocations are above 4GB. AMD64
and i386/PAE could also benefit from having this option.
just the intenral phy on parts supported by the rl and re drivers, the
RTL8201BL for example. He also sent me a nice picture of hundreds of
these chips in a tray to boulder his claim. :-) Therefore remove a
comment that suggested that they were...
is already bounded by hw.physmem to calculate phys_avail[] - previously only
real_phys_avail[] was being bound by hw.physmem so we were allocating memory
that wasn't mapped in the direct map
- shuffle memory range following kernel to the beginning of phys_avail
- have the direct area use 256MB pages where possible
- remove dead code from the end of pmap_bootstrap
- have pmap_alloc_contig_pages check all memory ranges in phys_avail before
giving up
- informal benchmarking indicates a ~5% speedup on buildworld
an "export" flag indicating that we are trying to NFS export the
filesystem, and the MSDOSFS_LARGEFS flag is set on the filesystem,
then deny the mount update and export request. Otherwise,
let the full mount update proceed normally.
MSDOSFS_LARGES and NFS don't mix because of the way inodes are calculated
for MSDOSFS_LARGEFS.
MFC after: 3 days
The symptoms were that outgoing DHCP requests for diskless kernels
had the IP header corrupt. After long investigations, the source of
the problem was found in ether_output() - for SIMPLEX interfaces
and broadcast traffic, a copy of the packet is passed back to the kernel
through if_simloop(). However if_simloop() modifies the mbuf, while
the copy obtained through m_copym() is a readonly one.
The bug has been there forever, but it has been triggered only recently
by a change in sosend_dgram() which passed down mbufs with sufficient
space to prepend the header.
This fix is trivial - use m_dup() instead of m_copy() to create
the copy. As an alternative, we could try and modify if_simloop()
to play safely with readonly mbufs, but i don't think it is worthwhile
because 1) this is a relatively infrequent code path so we do not need
to worry too much about performance, and 2) the cost of doing an
extra m_pullup in if_simloop() is probably the same as doing the
copy of the cluster, anyways.
MFC after: 1 week
field to "unsigned long" so that it actually works.
Thanks to Robert Sciuk for sending me a DVD that
demonstrated ISO9660-formatted media with a file >2G.
I've now fixed this both in libarchive and in the cd9660
filesystem.
MFC after: 14 days
timer in xl_txeof()/xl_txeof_90xB(); xl_poll_locked() unconditionally
invokes xl_txeof()/xl_txeof_90xB(), effectively circumventing that
the watchdog ever fires in the DEVICE_POLLING case as its timer is
constantly reloaded.
- Remove the banal and pedantically outdated comment regarding setting
xl_wdog_timer to 0 in xl_txeof().
Pointed out by: bde
(somewhat) meaningful message and terminate the build. It'd be
nice to print a proper URL from which to fetch the file but that
seems problematic. Leave a suggested starting point in this file
(TBD: add it to the man page).
Submitted by: ru
to workaround the problem with SMP kernels on Turion64 X2 processors
described in kern/104678 and may be useful in other situations too.
MFC after: 3 days
rather than treating them as a fatal exception and halting. At least one
storage BIOS (some newer mpt(4) parts) have a breakpoint instruction in
their disk read routine.
MFC after: 3 days
Make part of John Birrell's KSE patch permanent..
Specifically, remove:
Any reference of the ksegrp structure. This feature was
never fully utilised and made things overly complicated.
All code in the scheduler that tried to make threaded programs
fair to unthreaded programs. Libpthread processes will already
do this to some extent and libthr processes already disable it.
Also:
Since this makes such a big change to the scheduler(s), take the opportunity
to rename some structures and elements that had to be moved anyhow.
This makes the code a lot more readable.
The ULE scheduler compiles again but I have no idea if it works.
The 4bsd scheduler still reqires a little cleaning and some functions that now do
ALMOST nothing will go away, but I thought I'd do that as a separate commit.
Tested by David Xu, and Dan Eischen using libthr and libpthread.
driving xl_watchdog() in order to avoid races accessing if_timer.
While at it relax the watchdog a bit by reloading it in xl_txeof()/
xl_txeof_90xB() if there are still packets enqueued.
- Use bus_get_dma_tag() so xl(4) works on platforms requiring it.
- Don't bother to set if_mtu to ETHERMTU, ether_ifattach() does that.
hme_watchdog() in order to avoid races accessing if_timer.
- Use bus_get_dma_tag() so hme(4) works on platforms requiring it.
- Don't bother to set if_mtu to ETHERMTU, ether_ifattach() does that.
gem_watchdog() in order to avoid races accessing if_timer.
While at it relax the watchdog a bit by reloading it in gem_tint()
if there are still packets enqueued.
- Don't bother to set if_mtu to ETHERMTU, ether_ifattach() does that.
- Fix inconsistencies in prototypes.
depending on the NIC and isn't used at all with HomePNA links)
instead of if_slowtimo() for driving dc_watchdog() in order to
avoid races accessing if_timer.
- Use bus_get_dma_tag() so dc(4) works on platforms requiring it.
- Don't bother to set if_mtu to ETHERMTU, ether_ifattach() does that.
- Remove an alpha remnant in dc_softc.
of the nvenet lib upgrade (the constant went from 63 (2^n - 1) to
32 (2^n)). For reasons that are not obvious to me this fixes the driver
on at least some NICs.
MFC after: 3 days
with- not hope for the best. Change some things which were gated
off of 24XX to be gated off of 2K login support. Convert some
isp_prt calls to xpt_print calls.
Add a note that suggests a cleanup.
Note: This patch was derived based on looking at the pvrxxx/pvr250
ports' Makefiles only, and may be incomplete. It is not derived from
anything I saw from Hauppage.
was written into a user's address space. The fix is to modify uiomove_fromphys
to sync the icache when an executable user-space page is written into.
Alan Cox suggested that there should probably be a higher-level interface
to this in the ptrace code, but agreed that this is an OK short-term solution.
Files changed:
pmap.h - declaration of pmap_page_executable()
pmap_dispatch.c - pass through the page_executable call to the mmu object
mmu_oea.c - implement the page_executable method by examining the PTE_EXEC
field in the vm_page_t
uio_machdep.c - in uiomove_fromphys(), if the op was a UIO_WRITE to user-space,
and if the page is executable, sync the icache since this is at the least
a breakpoint-write from gdb.
Reported by: marcel
Tested by: marcel, grehan on g3+g4
Discussed with: alc
MFC after: 2 weeks
Fixes for 'blocking in fifoor state' problem of LTP tests.
linux_*stat*() functions were opening files with O_RDONLY to get
major/minor pair for char/block special files. Unfortunately,
when these functions are used against fifo, it is blocked forever
because there is no writer. Instead, we only open char/block special
files for major/minor conversion. We have to get rid of kern_open()
entirely from translate_path_major_minor() but today is not the day.
While I am here, add checks for errors before calling
translate_path_major_minor().
if waittime was zero (the lock was uncontested) l->lpo_waittime
in the hash table would not get initialized.
Inspection prompted by questions from: Attilio Rao
- create real_phys_avail which includes all memory ranges to be added to the direct map
- merge in nucleus memory to real_phys_avail
- distinguish between tag VA and index VA in tsb_set_tte_real for cases where page_size != index_page_size
- clean up direct map loop
of the bridge port and path cost have been administratively set or
calculated automatically by RSTP.
Make sure to transition from non-edge to edge when the port goes down
and the edge flag was manually set before.
This is needed to comply with the condition
((!portEnabled && AdminEdge) || ....)
in the Bridge Detection State Machine (IEE802.1D-2004, p. 171).
Reviewed by: thompsa
Approved by: bz (mentor)
pthread_cancel()ed, it is expected that the thread will not
consume a pthread_cond_signal(), therefor, we use thr_wake()
to mark a flag, the flag tells a thread calling do_cv_wait()
in umtx code to not block on a condition variable.
Thread library is expected that once a thread detected itself
is in pthread_cond_wait, it will call the thr_wake() for itself
in its SIGCANCEL handler.
- In hme_eint() print MIF register contents on MIF interrupts.
- In hme_mifinit() don't bother to preserve the previous MIF config.
This was mainly done in order to preserve the PHY select bit (external
or internal PHY) but which only needs to be set as appropriate when
reading from or writing to the desired PHY in hme_mii_{read,write}reg().
Similarly don't bother to set the PHY select bit in hme_mii_statchg().
- In hme_mii_{read,write}reg() ignore requests to PHYs other than the
external and internal PHY one.
- Move enabling/disabling the MII drivers of the external transceiver
from hme_init_locked() and based on the sheer presence of an external
to hme_mifinit() and based on the currently selected media, defaulting
to the internal transceiver when the media hasn't been set, yet.
Invoke hme_mifinit() from the newly added hme_mediachange_locked() so
the setting of the MII drivers is updated when changing media.
These changes keep the MII bus from wedging (which manifests in the HME
and the PHYs no longer being able to communicate with each other) when
the PHY device drivers isolate the unused PHY in two-PHY configurations
as present in f.e. Netra t1 100 while changing media, either from
hme_init_locked() (see also below) or via ifconfig(8). They also allow
for using both transceivers/PHYs.
- In the newly added hme_mediachange_locked() also reset the PHYs in two-
PHY configurations before invoking mii_mediachg(). This is required
for successfully unisolating the previously unused PHY when switching
between PHYs.
- Now that changing media should no longer cause problems back out rev.
1.27 and re-enable setting the current media in hme_init_locked() (see
the commit message of rev. 1.23 for more info).
These changes are roughly a merge of NetBSD gem.c rev. 1.32 - 1.35 (1.30
was already fixed differently in our 1.36; 1.31 and 1.32 were wrong) with
some parts reworked and things that don't make sense like setting the MII
drivers and restoring the previous MIF and XIF settings in hme_mii_{read,
write}reg() omitted.
MFC after: 2 weeks
Use TAILQ_FOREACH_SAFE instead of the unsafe one where an item is removed
from the queue.
This prevents a panic on kldunload.
Submitted by: rdivacky
Tested by: bsam
set birthtime to FAT CTime (creation time) and in the other cases
set birthtime to -1.
o Set ctime to mtime instead of FAT CTime which has completely
different meaning.
PR: kern/106018
Submitted by: Oliver Fromme
MFC after: 1 month
read wasn't flagging the SYNC mode was enabled. The temp
values for offset and sync period were uint8_t, but were
being assigned and shifted from a uint32_t value.
This didn't show up in testing because a random number
of 1030 cards set a bit that says "honor BIOS negotiation",
which means this whole code path was skipped.
This should clear up at least some of the negotation
issues that have been seen.
author can't remember why it was there.
The CTS_SCSI_FLAGS_TAG_ENB remains in place, and makes sense, and is
checked all over the place.
The CTS_SPI_FLAGS_TAG_ENB was probably an attempt to distinguish
protocol and transport tag capabilities. At the very least this can
be confusing and prone to many bugs, so let's just assume that the
transport tag case just flows from the protocol (and vice versa)
for now.
and by only delaying when an RTC register is written to. The delay
after writing to the data register is now not just a workaround.
This reduces the number of ISA accesses in the usual case from 4 to
1. The usual case is 2 rtcin()'s for each RTC interrupt. The index
register is almost always RTC_INTR for this. The 3 extra ISA accesses
were 1 for writing the index and 2 for delays. Some delays are needed
in theory, but in practice they now just slow down slow accesses some
more since almost eveyone including us does them wrong so modern systems
enforce sufficient delays in hardware. I used to have the delays ifdefed
out, but with the index register optimization the delays are rarely
executed so the old magic ones can be kept or even implemented non-
magically without significant cost.
Optimizing RTC interrupt handling is more interesting than it used to
be because RTC interrupts are currently needed to fix the more efficient
apic timer interrupts on some systems. apic_timer_hz is normally 2000
so the RTC interrupt rate needs to be 2048 to keep the apic timer
firing on such systems. Without these changes, each RTC interrupt
normally took 10 ISA accesses (2 PIC accesses and 2 sets of 4 RTC
accesses). Each ISA access takes 1-1.5uS so 10 of then at 2048 Hz
takes 2-3% of a CPU. Now 4 of them take 0.8-1.2% of a CPU.
priority mutex implemented, it is the time to introduce this stuff,
now we can use umutex and ucond together to implement pthread's
condition wait/signal.
Fix things to use the LSI-Logic Fusion Library mask and shift names for
offset and sync, no matter how awkward they are, in preference to just
plain numbers.
- Don't set MIIF_NOISOLATE so amphy(4) can be used in configurations
with multiple PHYs. There doesn't seem to be a problem with isolating
AM79c873 and workalikes per se nor in combination with the NICs they're
used with and amphy(4) was already adding IFM_NONE anyway.
- Use mii_phy_add_media() instead of mii_add_media() so the latter can
be eventually retired.
- Take advantage of mii_phy_setmedia().
- Fix a whitespace nit.
Obtained from: NetBSD dmphy(4) (except for the last item)
MFC after: 2 weeks
the currently selected media is of type IFM_AUTO as auto-negotiation
doesn't need to be kicked anyway.
- Fix a whitespace nit.
- Probe another Altima PHY, which is a AC101 workalike and integrated
in at least ADMtek ADM8511 but apparently is not mentioned in any
publically available data sheet so the actual identifier is unknown.
- Don't set MIIF_NOISOLATE so acphy(4) can be used in configurations
with multiple PHYs. There doesn't seem to be a problem with isolating
AC101 and workalikes per se nor in combination with the NICs they're
used with.
- Use mii_phy_add_media() instead of mii_add_media() so the latter can
be eventually retired.
- Take advantage of mii_phy_setmedia().
Obtained from: NetBSD (except for the first and second item)
MFC after: 2 weeks
in at least ADMtek ADM8511 but apparently is not mentioned in any
publically available data sheet so the actual identifier is unknown.
- Add Davicom DM9102 PHY.
- Add DM9101 to the description of AMD 79C873 as at least some Davicom
DM9101F identify identical to AMD 79C873.
Obtained from: NetBSD
MFC after: 2 weeks
with multiple PHYs. There doesn't seem to be a problem with isolating
78Q2120 per se nor in combination with the NICs they're used with and
tdkphy(4) was already adding IFM_NONE anyway.
- Set MIIF_NOLOOP as loopback doesn't work with this PHY. The MIIF_NOLOOP
flag currently triggers nothing but hopefully will be respected by
mii_phy_setmedia() later on.
- Use mii_phy_add_media() instead of mii_add_media() so the latter can
be eventually retired.
- Take advantage of mii_phy_setmedia().
Thanks to Hans-Joerg Sirtl for lending me test hardware.
Obtained from: NetBSD tqphy(4)
MFC after: 2 weeks
with multiple PHYs and un-comment case IFM_NONE in case MII_MEDIACHG
rgephy_service(). There doesn't seem to be a problem with isolating
RTL8169S and their internal PHY.
- Take advantage of mii_phy_add_media(). [1]
Obtained from: NetBSD [1]
Tested by: yongari
MFC after: 2 weeks
- Fix some whitespace nits.
- Fix some spelling in comments.
- Use MII_ANEGTICKS instead of 5.
- Don't define variables in nested scope.
- Remove superfluous returns at the end of void functions.
- Remove unused static global rgephy_mii_model.
- Remove dupe $Id$ in tdkphy(4).
- Sort brgphys table.
MFC after: 2 weeks
and Daichi GOTO <daichi@FreeBSD.org> for submitting this
major rewrite of unionfs. This rewrite was done to
try to solve many of the longstanding crashing and locking
issues in the existing unionfs implementation. This
implementation also adds a 'MASQUERADE mode', which allows
the user to set different user, group, and file permission
modes in the upper layer.
Submitted by: daichi, Masanori OZAWA
Reviewed by: rodrigc (modified for minor style issues)
- Playback and headphone/speaker automute works.
- Recording untested due to me being deaf doing back-and-forth
remote debugging.
Free Macbook donation is highly appreciated :)
Tested by: Dennis Pielken <mips128@gmx.net>
mii_phy_match() API and takes care of the PHY device probe based on
the struct mii_phydesc array and the match return value provided.
Convert PHY drivers to take advantage of mii_phy_dev_probe(),
converting drivers to provide a mii_phydesc table in the first
place where necessary.
Reviewed by: yongari
MFC after: 2 weeks
- Currently LINUX_MAX_COMM_LEN is smaller than MAXCOMLEN, but in case
this will change we have a buffer overflow. Apply some defensive
programming to DTRT when this should happen.
- Use copyinstr() instead of copyin where appropriate.
* Fallback to copyin() in case of ENAMETOOLONG. [1]
* Use the right source and destination (it was wrong before).
- Use strlcpy instead of strcpy.
- Properly lock the read case (PR_GET_NAME) like the write case.
Reviewed by: rwatson (except [1])
Suggested by: rwatson [1]
on the arm. Add an assert to ensure that the size is 8 to prefent others
from falling into this trap (we should have more of these).
Why the construct:
struct foo {
union bar {
struct {
...
} __packed fred;
...
} __packed wilma;
} __packed;
has a different packing than:
struct foo {
union bar {
struct {
...
} fred __packed;
...
} wilma __packed;
} __packed;
is beyond my ability to ferret out of the gcc documentation. Most
likely some subtle binding issue (eg before it says the struct itself
is packed, while after it means that the whole struct is packed into
the thing it is in). Pointers to relevant documentation would be
appreciated.
sizeof ether_header is 2 * ETHER_ADDR_LEN + 2 (14) bytes long
sizeof ether_addr is ETHER_ADDR_LEN bytes long
On arm, this shows that struct ether_addr needs to be __packed.
The first condition muts be true for the bridging code to not dump core.
The second one appears to be implicitly relied upon by wi (but many
of the rids it sends down likely need __packed too to be safe) and
maybe others. It appears to not hurt anything.
discarded RX packets to input error for BCM5705 or newer chipset as the others.
Unfortunately we cannot do the same for output errors because ifOutDiscards
equivalent register does not exist. While I am here, replace misleading and
wrong BGE_RX_STATS/BGE_TX_STATS with BGE_MAC_STATS. They were reversed but
worked accidently.
machines and both TX and RX were broken on big-endian machines.
The chip design is crazy -- on RX, it puts the 16-bit VLAN tag
in network byte order (big-endian) in the 32-bit little-endian
register!
Thanks to John Baldwin for helping me document this change! ;-)
Tested by: sat (amd64), test program (sparc64)
PR: kern/105054
MFC after: 3 days
if_watchdog/if_timer interface doesn't fit modern SMP network
stack design.
Device drivers that need watchdog to monitor their hardware should
implement it theirselves.
Eventually the if_watchdog/if_timer API will be removed. For now,
warn that driver uses it.
Reviewed by: scottl
__stop_<section> symbols generated by the static linker for elf
sections. This is done only for the final link, and not for ld -r.
Augment elf_obj in-kernel linker by recognizing such special symbols,
and resolving them to the start and end of the section automatically.
As result, linker sets on amd64 could be used in the same way as on
other architectures, without explicit calls to linker_file_lookup_set().
Requested by: rdivacky
No objections from: peter, jhb
by default for sun4v where it is absolutely required.
This change moves the buffer from struct pcpu to the stack to avoid
using the critical section which created a LOR in a couple of cases
due to interaction with the tty code and kqueue. The LOR can't be
fixed with the critical section and the pcpu buffer can't be used
without the critical section.
Putting the buffer on the stack was my initial solution, but it was
pointed out that the stress on the stack might cause problems
depending on the call path. We don't have a way of creating tests
for those possible cases, so it's best to leave this as an option
for the time being. In time we may get enough data to enable this
option more generally.
aue_tick calls several synchronous usb functions from a timeout(9),
this is very broken since a timeout(9) is run as an interrupt
and the usb functions tsleep.
A stopgap fix is to schedule a taskqueue task from the timeout
and defer work to that taskqueue task.
