platforms which do not use loaders or kernels that want to hardcode
options or for FDT passed in by loader.
Also fix a build issue by putting the kmdp variable accessed back under
the #ifdef FDT; we may wish to revisit decision in which case more
code needs changing.
Submitted by: brooks
by adding an argument to the VM_SUSPEND ioctl that specifies how the virtual
machine should be suspended, viz. VM_SUSPEND_RESET or VM_SUSPEND_POWEROFF.
The disposition of VM_SUSPEND is also made available to the exit handler
via the 'u.suspended' member of 'struct vm_exit'.
This capability is exposed via the '--force-reset' and '--force-poweroff'
arguments to /usr/sbin/bhyvectl.
Discussed with: grehan@
old devd's and thus hosts that get IP addresses from DHCP was too much
of a POLA violation.
The sysctl may be removed again after r263758 has been merged to at
least stable/9 and stable/10, and releases have been cut from those
branches.
Discussed with: mjg
Reported by: theraven, rwatson
It exposes I/O resources to user space, so that programs can peek
and poke at the hardware. It does not itself have knowledge about
the hardware device it attaches to.
Sponsored by: Juniper Networks, Inc.
Instead opening/closing provider by each of metadata classes, do it only
once in core code. Since for SCSI disks open/close means sending some
SCSI commands to the device, this change reduces taste time.
MFC after: 2 weeks
Sponsored by: iXsystems, Inc.
FD_SETSIZE is often used as an argument to select or compared with an
integer file descriptor. Rather than force 3rd party software to add
explicit casts, just make it a plain (int) constant as on other
operating systems.
Previous discussion:
http://lists.freebsd.org/pipermail/freebsd-standards/2012-July/002410.html
returns ZIO_PIPELINE_CONTINUE from vdev_op_io_start to zio_execute resulting
in the wrong ZIO continuing its pipeline.
This is a serious issue which could cause data loss / corruption but appears
to be limited to error handling such as when vdev_readable(vd) returns false.
MFC after: 2 days
the MYBEACON RX filter (only receive beacons which match the BSSID)
or all beacons on the current channel.
* Add the relevant RX filter entry for MYBEACON.
Tested:
* AR5416, STA
* AR9285, STA
TODO:
* once the code is in -HEAD, just make sure that the code which uses it
correctly sets BEACON for pre-AR5416 chips.
Obtained from: QCA, Linux ath9k
the QCA HAL.
This fires off an interrupt if the TSF from the AP / IBSS peer is
wildly out of range. I'll add some code to the ath(4) driver soon
which makes use of this.
TODO:
* verify this didn't break TDMA!
to the hardware.
The QCA HAL has a comment noting that if this isn't done, modifications
to AR_IMR_S2 before AR_IMR is flushed may produce spurious interrupts.
Obtained from: QCA
Flushing the caches is required before doing a panic dump, but ARM
doesn't provide a flavor of flush that gets broadcast to other cores.
However, all cores except one are stopped before doing a dump, so this
works around the lack of a global flush/invalidate by doing it locally
on each CPU as part of stopping.
Discussed with: cognet@
This was added ca. 2004 for the purpose of ensuring the caches were in the
right state after the debugger set a breakpoint. kdb_cpu_sync_icache()
was added in 2007 to handle that situation, and now the wbinv_all is
actually harmful because the operation isn't broadcast to other cores.
* memory is now allocated as early as possible, without holding locks.
* sysctl users are now guaranteed to get a response (M_WAITOK buffer prealloc).
* socket users are more likely to use on-stack buffer for replies.
* standard kernel malloc/free functions are now used instead of radix wrappers.
rt_msg2() has been renamed to rtsock_msg_buffer().
MFC after: 1 month
#gpio-cells property.
Add a new ofw_bus method (OFW_BUS_MAP_GPIOS()) that allows the GPIO
controller to implement its own mapping to deal with gpio-specifiers,
allowing the decoding of gpio-specifiers to be controller specific.
The default ofw_bus_map_gpios() decodes the linux standard (#gpio-cells =
<2>) and the FreeBSD standard (#gpio-cells = <3>).
It pass the gpio-specifier flag field to the children as an ivar variable so
they can act upon.
checks for special pcid values in invlpg_pcid_handler(). Forst check
for special values, and only then do PCID-specific page invalidation.
Minor fix to the style compliance, declare local variable at the
function start.
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
entry and performs the upgrade of the entry permissions from read-only
to read-write, we must allow to search for the source pages in the
backing object, like we do in the case of forking the read-only wired
entry. For the fork case, the behaviour is allowed by src_readonly
boolean, which in fact is only used to assert that read-write case
provides all source pages in the top-level object.
Eliminate the src_readonly variable. Allow for the copy loop to look
into the backing objects, add explicit asserts to ensure that only
read-only and upgrade case actually does.
Expand comments. Change the panic call into assert.
Reported by: markj
Tested by: markj, pho (previous version)
Reviewed by: alc
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
kernel will now boot on both BIOS and EFI systems without modification.
Equivalent functionality in GENERIC requires making vt(9) the default console
driver, which is probably appropriate at this point.
using armv7_idcache_wbinv_all, because wbinv_all doesn't broadcast the
operation to other cores. In elf_cpu_load_file() use icache_sync_all()
and explain why it's needed (and why other sync operations aren't).
As part of doing this, all callers of cpu_icache_sync_all() were
inspected to ensure they weren't relying on the old side effect of
doing a wbinv_all along with the icache work.
RTM_CHANGE is now handled inside route.c:rtrequest1_fib() as it should be.
Note change change handler is a separate function rtrequest1_fib_change().
MFC after: 1 month
the cpufreq code. Replace its use with smp_started. There's at least
one userland tool that still looks at the kern.smp.active sysctl, so
preserve it but point it to smp_started as well.
Discussed with: peter, jhb
MFC after: 3 days
Obtained from: Netflix
define a few imx_ccm_foo() functions that are implemented by the imx51 or
imx6 ccm code. Of course, the imx6 ccm code is still more a wish than
reality, so for now its implementations just return hard-coded numbers.
copy of the code from boot1.chrp again.
The resulting image is installed to /boot/boot1.efifat. If dd'ed to an 800K
"efi" partition, it should result in a bootable system.
replace the vt_vga driver with vt_efifb.
This is intended to help with snapshot builds only.
There is no intention to MFC this commit.
Sponsored by: The FreeBSD Foundation
a jtag debugging product, which was used on early Beaglebone boards (later
boards used a standard FTDI 2232C product ID). Change the name accordingly,
and also add an entry for XDS100V3, the latest version of that product
which has its own new product ID number.
Device type and revision is now determined from the bcdDevice field and
doesn't need to be in the table at all. The feature that skips creation
of /dev/ttyU* entries for jtag and gpio interfaces is enhanced:
- The feature is now optional, but enabled by default. A tunable and
sysctl are available to control it: hw.usb.uftdi.skip_jtag_interfaces.
- We no longer assume interface #0 is the only jtag interface. Up to
eight interfaces per chip can be flagged as jtag. (Current ftdi chips
support a max of 4 interfaces; this leaves room for growth.)
- Some manufacturers don't change the product ID or use the same ID for
different devices intended for both serial-comms and jtag/gpio use.
Often while the product ID is the same, the product name string is
different, so it's now possible to search for the product name in a
table of strings and get the set of non-tty interfaces from that table.
- DIOCADDADDR adds addresses and puts them into V_pf_pabuf
- DIOCADDRULE takes all addresses from V_pf_pabuf and links
them into rule.
The ugly part is that if address is a table, then it is initialized
in DIOCADDRULE, because we need ruleset, and DIOCADDADDR doesn't
supply ruleset. But if address is a dynaddr, we need address family,
and address family could be different for different addresses in one
rule, so dynaddr is initialized in DIOCADDADDR.
This leads to the entangled state of addresses on V_pf_pabuf. Some are
initialized, and some not. That's why running pf_empty_pool(&V_pf_pabuf)
can lead to a panic on a NULL table address.
Since proper fix requires API/ABI change, for now simply plug the panic
in pf_empty_pool().
Reported by: danger
request during SLEEP results in a hang.
Whilst I'm here, add in some disabled code that will transition to RUN
if there's multicast traffic. It's not needed for Atheros hardware but
it may be for other hardware.
Tested:
* AR5416, STA mode (powersave)
* AR5212, STA mode (powersave)
These two bugs are closely related. The root cause is that ifa_ifwithnet
does not consider FIBs when searching for an interface address.
sys/net/if_var.h
sys/net/if.c
Add a fib argument to ifa_ifwithnet and ifa_ifwithdstadddr. Those
functions will only return an address whose interface fib equals the
argument.
sys/net/route.c
Update calls to ifa_ifwithnet and ifa_ifwithdstaddr with fib
arguments.
sys/netinet/in.c
Update in_addprefix to consider the interface fib when adding
prefixes. This will prevent it from not adding a subnet route when
one already exists on a different fib.
sys/net/rtsock.c
sys/netinet/in_pcb.c
sys/netinet/ip_output.c
sys/netinet/ip_options.c
sys/netinet6/nd6.c
Add RT_DEFAULT_FIB arguments to ifa_ifwithdstaddr and ifa_ifwithnet.
In some cases it there wasn't a clear specific fib number to use.
In others, I was unable to test those functions so I chose
RT_DEFAULT_FIB to minimize divergence from current behavior. I will
fix some of the latter changes along with PR kern/187553.
tests/sys/netinet/fibs_test.sh
tests/sys/netinet/udp_dontroute.c
tests/sys/netinet/Makefile
Revert r263738. The udp_dontroute test was right all along.
However, bugs kern/187550 and kern/187553 cancelled each other out
when it came to this test. Because of kern/187553, ifa_ifwithnet
searched the default fib instead of the requested one, but because
of kern/187550, there was an applicable subnet route on the default
fib. The new test added in r263738 doesn't work right, however. I
can verify with dtrace that ifa_ifwithnet returned the wrong address
before I applied this commit, but route(8) miraculously found the
correct interface to use anyway. I don't know how.
Clear expected failure messages for kern/187550 and kern/187552.
PR: kern/187550
PR: kern/187552
Reviewed by: melifaro
MFC after: 3 weeks
Sponsored by: Spectra Logic
SLEEP rather than RUN.
Without this things like 'ifconfig wlan0 list sta' don't work when the
NIC is power save.
Tested:
* AR5212, STA mode (with powersave)
* AR5416, STA mode (with powersave)
commit 003649d9622ce252a2794ae5891ee7e7c209caca
Author: Robert N. M. Watson <robert.watson@cl.cam.ac.uk>
Date: Wed Feb 5 18:32:09 2014 +0000
Teach the FreeBSD/beri boot to "auto-detect" whether argument 4 (a3) is a
memory size of pointer to a struct bootinfo * by looking at its value and
seeing whether it is pointer-like. If a pointer, assume it's a bootinfo
and extract memsize from it instead; otherwise, use it as memsize directly.
This allows kernels to support bootinfo being passed by loader (and boot2)
while still supporting older Miniboot setups.
commit f7045af9a1e92b6bd92541fe5d25abf66d824e8f
Author: Robert N. M. Watson <robert.watson@cl.cam.ac.uk>
Date: Thu Feb 6 13:45:34 2014 +0000
When the module metadata pointer is available from loader, use it in the
kernel.
commit 52e0e1ff2cba9dfcfab9e1d0a31fb7fdf7317450
Author: Robert N. M. Watson <robert.watson@cl.cam.ac.uk>
Date: Thu Feb 6 19:57:48 2014 +0000
In the BERI kernel boot code, extract 'boothowto' (which includes boot flags
such as '-s') and 'envp' from passed module data. Booting to single-user
mode using boot flags now works.
Sponsored by: DARPA, AFRL
for the VOP_READ() call. This patch fixes both the old and new
server for this case.
PR: 185232
Submitted by: PR had patch for old server
Reviewed by: kib
MFC after: 2 weeks
sys/net/route.c
In rtinit1, use the interface fib instead of the process fib. The
latter wasn't very useful because ifconfig(8) is usually invoked
with the default process fib. Changing ifconfig(8) to use setfib(2)
would be redundant, because it already sets the interface fib.
tests/sys/netinet/fibs_test.sh
Clear the expected ATF failure
sys/net/if.c
Pass the interface fib in calls to rtrequest1_fib and rtalloc1_fib
sys/netinet/in.c
sys/net/if_var.h
Add a fibnum argument to ifa_switch_loopback_route, a subroutine of
in_scrubprefix. Pass it the interface fib.
PR: kern/187549
Reviewed by: melifaro
MFC after: 3 weeks
Sponsored by: Spectra Logic Corporation
Instead of allocating up to 16MB or RAM at once to handle whole I/O,
allocate up to 1MB at a time, but do multiple ctl_datamove() and storage
I/Os if needed.
Some of the code in xen-locore.S was picked from Cherry G. Mathew
amd64 Xen PV branch, but I've failed to set the proper copyright, so
do it now.
Approved by: gibbs
This transitions the VAP in and out of SLEEP state based on:
* whether there's been an active transmission in the last (hardcoded) 500ms;
* whether the TIM from the AP indicates there is data available.
It uses the beacon reception to trigger the active traffic check.
This way there's no further timer running to wake up the CPU
from its own sleep states.
Right now the VAP isn't woken up for multicast traffic - mostly because
the only NIC I plan on doing this for right will auto wakeup and stay
awake for multicast traffic indicated in the TIM. So I don't have
to manually keep the hardware awake.
This doesn't do anything if the NIC doesn't advertise it implements
the new SWSLEEP capability AND if the VAP doesn't have powersave
enabled.
It also doesn't do much with ath(4) as it doesn't currently implement
the SLEEP state.
Tested:
* AR5416, STA mode (with local ath(4) changes)
This simplifies the code and should avoid the clang sparc
port from generating an abort() call.
Requested by: rdivacky
Submitted by: jhb
MFC after: 2 weeks
Frames transmitted during SLEEP state should be queued in the
power save queue before waking the unit up. Otherwise DHCP
requests and such will be dropped if the NIC is asleep - the
NIC will wake up but not transmit the frame.
- Add a comment about FTDI and ZLPs.
- Correctly check odditiy of baud rate divisor.
- Correct IOCTL handling for "error" and "event" char.
MFC after: 1 weeks
concurrent updates from any completing transmits in other threads.
This was exposed when doing power save work - net80211 is constantly
doing reassociations and it's causing the rate control state to get
blanked out. This could cause the rate control code to assert.
This should be MFCed to stable/10 as it's a stability fix.
Tested:
* AR5416, STA
MFC after: 7 days
The MAC filter set may be called without softc_lock held in the case of
SIOCADDMULTI and SIOCDELMULTI ioctls. The ioctl handler checks IFF_DRV_RUNNING
flag which implies port started, but it is not guaranteed to remain.
softc_lock shared lock can't be held in the case of these ioctls processing,
since it results in failure where kernel complains that non-sleepable
lock is held in sleeping thread.
Both problems are repeatable on LAG with LACP proto bring up.
Submitted by: Andrew Rybchenko <Andrew.Rybchenko at oktetlabs.ru>
Sponsored by: Solarflare Communications, Inc.
MFC after: 2 weeks
Make sure not to start I/O bigger than MAXPHYS bytes.
Quoting r264504:
When we detect the condition, we'll reduce the block count and perform
a "short" read. In g_uncompress_done() we need to consider the original
I/O length and stop early if we're about to deflate a block that we didn't
read. By using bio_completed in the cloned BIO and not bio_length to
check for this, we automatically and gracefully handle short reads that
our providers may be doing on top of the short reads we may initiate
ourselves.
Reviewed by: marcel
The existing cleanup code was based on the Atheros reference driver
from way back and stuff that was in Linux ath9k. It turned out to be ..
rather silly.
Specifically:
* The whole method of determining whether there's hardware-queued frames
was fragile and the BAW would never quite work right afterwards.
* The cleanup path wouldn't correctly pull apart aggregate frames in the
queue, so frames would not be freed and the BAW wouldn't be correctly
updated.
So to implement this:
* Pull the aggregate frames apart correctly and handle each separately;
* Make the atid->incomp counter just track the number of hardware queued
frames rather than try to figure it out from the BAW;
* Modify the aggregate completion path to handle it as a single frame
(atid->incomp tracks the one frame now, not the subframes) and
remove the frames from the BAW before completing them as normal frames;
* Make sure bf->bf_next is NULled out correctly;
* Make both aggregate session and non-aggregate path frames now be
handled via the incompletion path.
TODO:
* kill atid->incomp; the driver tracks the hardware queued frames
for each TID and so we can just use that.
This is a stability fix that should be merged back to stable/10.
Tested:
* AR5416, STA
MFC after: 7 days
MAC
* Now that the paused < 0 bugs have been identified, make the DPRINTF()
a device_printf() again. Anything else that shows up here needs to be
fixed immediately.
Tested:
* AR5416, STA mode
MFC after: 7 days
During power save testing I noticed that the cleanup code is being
called during a RUN->RUN state transition. It's because the net80211
stack is treating that (for reasons I don't quitey know yet) as a
reassociation and this calls the node cleanup code. The reason it's
seeing a RUN->RUN transition is because during active power save
stuff it's possible that the RUN->SLEEP and SLEEP->RUN transitions
happen so quickly that the deferred net80211 vap state code
"loses" a transition, namely the intermediary SLEEP transition.
So, this was causing the node reassociation code to sometimes be called
twice in quick succession and this would result in ath_tx_tid_cleanup()
to be called again. The code calling it would always call pause, and
then only call resume if the TID didn't have "cleanup_inprogress" set.
Unfortunately it didn't check if it was already set on entry, so it
would pause but not call resume. Thus, paused would be called more
than once (once before each entry into ath-tx_tid_cleanup()) but resume
would only be called once when the cleanup state was finished.
This doesn't entirely fix all of the issues seen in the cleanup path
but it's a necessary first step.
Since this is a stability fix, it should be merged to stable/10 at some
point.
Tested:
* AR5416, STA mode
MFC after: 7 days
post-create/mkdir directory attributes. This allows the RPC to
name cache the newly created directory and reduces the lookup RPC
count for applications creating a lot of directories.
MFC after: 2 weeks
- These were needed on armv4/5 (VIVT cache), not needed on armv6.
- The wbinv_all call can't be used on SMP systems; cache operations by
set/way are not broadcast to other cores.
- The TLB maintenance operations needed for pmap_growkernel() happen in
pmap_grow_l2_bucket(), so there's no need to flush all TLB entries at
the end.
- There may not be any need for the TLB flush at the beginning of
pmap_release(), but it's left in for now pending more investigation.
Pointed out by: Svatopluk Kraus <onwahe@gmail.com>
Discussed with: cognet@
De-virtualize UMA zone pf_mtag_z and move to global initialization part.
The m_tag struct does not know about vnet context and the pf_mtag_free()
callback is called unaware of current vnet. This causes a panic.
MFC after: 1 week
Since <sys/capsicum.h> does not comply to any standards, standards-compliant
applications do not include it and it is not useful to restrict what is
exposed depending on the requested standard.
In addition, the use of types like u_int makes the header fail in strict
standards compliance modes anyway.
post-open/create directory attributes. This allows the RPC to
name cache the newly created file and reduces the lookup RPC
count by about 10% for software builds.
MFC after: 2 weeks
attributes. This allows the client to cache directory names
when they are looked up, reducing the Lookup RPC count by
about 40% for software builds.
MFC after: 2 weeks
4754 io issued to near-full luns even after setting noalloc threshold
4755 mg_alloc_failures is no longer needed
illumos/illumos@b6240e830b
MFC after: 2 weeks
worse when filling up a device and then trying to erase files to make
space. Without enough space, you can't do that. Also, ensure that the
metadata writes don't generate ENOSPC. They will be retried later
since the buffers are still dirty...
Submitted by: mjg@
Status and Control register at port 0x61.
Be more conservative about "catching up" callouts that were supposed
to fire in the past by skipping an interrupt if it was
scheduled too far in the past.
Restore the PIT ACPI DSDT entries and add an entry for NMISC too.
Approved by: neel (co-mentor)
network interfaces limited to 32 transmit segments, there
are two known issues.
The more serious one is that for an I/O of slightly less than 64K,
the net device driver prepends an ethernet header, resulting in a
TSO segment slightly larger than 64K. Since m_defrag() copies this
into 33 mbuf clusters, the transmit fails with EFBIG.
A tester indicated observing a similar failure using iSCSI.
The second less critical problem is that the network
device driver must copy the mbuf chain via m_defrag()
(m_collapse() is not sufficient), resulting in measurable overhead.
This patch reduces the default size of if_hw_tsomax
slightly, so that the first issue is avoided.
Fixing the second issue will require a way for the
network device driver to inform tcp_output() that it
is limited to 32 transmit segments.
Reported and tested by: csforgeron@gmail.com, markus.gebert@hostpoint.ch
MFC after: 2 weeks
Also, remove #if __BSD_VISIBLE where it is redundant. When __BSD_VISIBLE is
defined to 1, __POSIX_VISIBLE, __XSI_VISIBLE and __ISO_C_VISIBLE are also
defined to the newest supported version.
PR: 188173
Reviewed by: pluknet
boot/mips/beri/loader/metadata.c allow FDT configuration to set
command line options.
This leads to an interesting quesiton of future interactions with loader.
However for configurations without loader this allows bootverbose or boot
single user to be set by compiling a new kernel, which is good enough for
testing and debugging.
Reviewed by: rwatson
MFC after: 1 week
NetFPGA-10G Embedded CPU Ethernet Core.
The current version operates on a simple PIO based interface connected
to a NetFPGA-10G port.
To avoid confusion: this driver operates on a CPU running on the FPGA,
e.g. BERI/mips, and is not suited for the PCI host interface.
MFC after: 1 week
Relnotes: yes
Sponsored by: DARPA/AFRL
This code was removed from the opensolaris and darwin's
netsmb implementations, in DfBSD it also has been disabled.
PR: 36566, 87859, 139407, 161579, 175557, 178412, 186652
MFC after: 2 weeks
Sponsored by: Yandex LLC
Due to the way those timers are implemented, we can't handle very short
intervals. In addition to that mentioned patch caused math overflows
for short intervals. To avoid that round those intervals to 1 tick.
PR: kern/187668
MFC after: 1 week
and normal mode; this makes it possible to compile with the former
by default, but use it only when neccessary. That's especially
important for the userland part.
Sponsored by: The FreeBSD Foundation
was stacked on top of a network interface that set if_hw_tsomax,
tcp_output() would see the default value instead of the value
set by the network interface. This patch modifies vlan so that
it sets if_hw_tsomax to the value of the parent interface.
Reviewed by: glebius
MFC after: 2 weeks
problems in our providers, such as a KASSERT in md(4). We can initiate
I/O for more than MAXPHYS bytes if we've been given a BIO for MAXPHYS
bytes, the blocks from which we're reading couldn't be compressed and
we had compression in preceeding blocks resulting in misalignment of
the blocks we're trying to read relative to the sector. We're forced to
round up the I/O length to make it an multiple of the sector size.
When we detect the condition, we'll reduce the block count and perform
a "short" read. In g_uzip_done() we need to consider the original I/O
length and stop early if we're about to deflate a block that we didn't
read. By using bio_completed in the cloned BIO and not bio_length to
check for this, we automatically and gracefully handle short reads that
our providers may be doing on top of the short reads we may initiate
ourselves.
Obtained from: Juniper Networks, Inc.
When server doesn't support this request, try to use SMB_INFO_ALLOCATION.
And use SMB_COM_QUERY_INFORMATION_DISK request as fallback.
MFC after: 2 weeks
was stacked on top of network interfaces that set if_hw_tsomax,
tcp_output() would see the default value instead of the value
set by the network interface(s). This patch modifies lagg so that
it sets if_hw_tsomax to the minimum of the value(s) for the
underlying network interfaces.
Reviewed by: glebius
MFC after: 2 weeks
this set of patches fixes support for systems with > 32 cores.
Details include
sfxge: RXQ index (not label) comes from FW in flush done/failed events
Change the second argument name of the efx_rxq_flush_done_ev_t and
efx_rxq_flush_failed_ev_t prototypes to highlight that RXQ index (not label)
comes from FW in flush done and failed events.
sfxge: TXQ index (not label) comes from FW in flush done event
Change the second argument name of the efx_txq_flush_done_ev_t prototype to
highlight that TXQ index (not label) comes from FW in flush done event.
sfxge: use TXQ type as label to support more than 32 TXQs
There are 3 TXQs in event queue 0 and 1 TXQ (with TCP/UDP checksum offload)
in all other event queues.
Submitted by: Andrew Rybchenko <Andrew.Rybchenko at oktetlabs.ru>
Sponsored by: Solarflare Communications, Inc.
sites and installing a hook at the kernel's trap handler. The fasttrap code
will emulate the overwritten instruction in some common cases, but otherwise
copies it out into some scratch space in the traced process' address space
and ensures that it's executed after returning from the trap.
In Solaris and illumos, this (per-thread) scratch space comes from some
reserved space in TLS, accessible via the fs segment register. This
approach is somewhat unappealing on FreeBSD since it would require some
modifications to rtld and jemalloc (for static TLS) to ensure that TLS is
executable, and would thus introduce dependencies on their implementation
details. I think it would also be impossible to safely trace static binaries
compiled without these modifications.
This change implements the functionality in a different way, by having
fasttrap map pages into the target process' address space on demand. Each
page is divided into 64-byte chunks for use by individual threads, and
fasttrap's process descriptor struct has been extended to keep track of
any scratch space allocated for the corresponding process.
With this change it's possible to trace all libc functions in a program,
e.g. with
pid$target:libc.so.*::entry {@[probefunc] = count();}
Previously this would generally cause the victim process to crash, as
tracing memcpy on amd64 requires the functionality described above.
Tested by: Prashanth Kumar <pra_udupi@yahoo.co.in> (earlier version)
MFC after: 6 weeks
