layer, but with a twist.
The twist has to do with the fact that Microsoft supports structured
exception handling in kernel mode. On the i386 arch, exception handling
is implemented by hanging an exception registration list off the
Thread Environment Block (TEB), and the TEB is accessed via the %fs
register. The problem is, we use %fs as a pointer to the pcpu stucture,
which means any driver that tries to write through %fs:0 will overwrite
the curthread pointer and make a serious mess of things.
To get around this, Project Evil now creates a special entry in
the GDT on each processor. When we call into Windows code, a context
switch routine will fix up %fs so it points to our new descriptor,
which in turn points to a fake TEB. When the Windows code returns,
or calls out to an external routine, we swap %fs back again. Currently,
Project Evil makes use of GDT slot 7, which is all 0s by default.
I fully expect someone to jump up and say I can't do that, but I
couldn't find any code that makes use of this entry anywhere. Sadly,
this was the only method I could come up with that worked on both
UP and SMP. (Modifying the LDT works on UP, but becomes incredibly
complicated on SMP.) If necessary, the context switching stuff can
be yanked out while preserving the convention calling wrappers.
(Fortunately, it looks like Microsoft uses some special epilog/prolog
code on amd64 to implement exception handling, so the same nastiness
won't be necessary on that arch.)
The advantages are:
- Any driver that uses %fs as though it were a TEB pointer won't
clobber pcpu.
- All the __stdcall/__fastcall/__regparm stuff that's specific to
gcc goes away.
Also, while I'm here, switch NdisGetSystemUpTime() back to using
nanouptime() again. It turns out nanouptime() is way more accurate
than just using ticks(). On slower machines, the Atheros drivers
I tested seem to take a long time to associate due to the loss
in accuracy.
ndis_timercall() in NdisMInitializeTimer(), we can't use the raw
function pointer. This is because ntoskrnl_run_dpc() expects to
invoke a function with Microsoft calling conventions. On i386,
this works because ndis_timercall() is declared with the __stdcall
attribute, but this is a no-op on amd64. To do it correctly, we
have to generate a wrapper for ndis_timercall() and us the wrapper
instead of of the raw function pointer.
Fix this by adding ndis_timercall() to the funcptr table in subr_ndis.c,
and create ndis_findwrap() to extract the wrapped function from the
table in NdisMInitializeTimer() instead of just passing ndis_timercall()
to KeInitializeDpc() directly.
ExAllocatePoolWithTag(), not malloc(), so it should be released
with ExFreePool(), not free(). Fix a couple if instances of
free(fh, ...) that got overlooked.
- On amd64, InterlockedPushEntrySList() and InterlockedPopEntrySList()
are mapped to ExpInterlockedPushEntrySList and
ExpInterlockedPopEntrySList() via macros (which do the same thing).
Add IMPORT_FUNC_MAP()s for these.
- Implement ExQueryDepthSList().
alloc and free routine pointers in the lookaside list with pointers
to ExAllocatePoolWithTag() and ExFreePool() (in the case where the
driver does not provide its own alloc and free routines). For amd64,
this is wrong: we have to use pointers to the wrapped versions of these
functions, not the originals.
nll_obsoletelock field in the lookaside list structure is only defined
for the i386 arch. For amd64, the field is gone, and different list
update routines are used which do their locking internally. Apparently
the Inprocomm amd64 driver uses lookaside lists. I'm not positive this
will make it work yet since I don't have an Inprocomm NIC to test, but
this needs to be fixed anyway.
work on SMP" saga. After several weeks and much gnashing of teeth,
I have finally tracked down all the problems, despite their best
efforts to confound and annoy me.
Problem nunmber one: the Atheros windows driver is _NOT_ a de-serialized
miniport! It used to be that NDIS drivers relied on the NDIS library
itself for all their locking and serialization needs. Transmit packet
queues were all handled internally by NDIS, and all calls to
MiniportXXX() routines were guaranteed to be appropriately serialized.
This proved to be a performance problem however, and Microsoft
introduced de-serialized miniports with the NDIS 5.x spec. Microsoft
still supports serialized miniports, but recommends that all new drivers
written for Windows XP and later be deserialized. Apparently Atheros
wasn't listening when they said this.
This means (among other things) that we have to serialize calls to
MiniportSendPackets(). We also have to serialize calls to MiniportTimer()
that are triggered via the NdisMInitializeTimer() routine. It finally
dawned on me why NdisMInitializeTimer() takes a special
NDIS_MINIPORT_TIMER structure and a pointer to the miniport block:
the timer callback must be serialized, and it's only by saving the
miniport block handle that we can get access to the serialization
lock during the timer callback.
Problem number two: haunted hardware. The thing that was _really_
driving me absolutely bonkers for the longest time is that, for some
reason I couldn't understand, my test machine would occasionally freeze
or more frustratingly, reset completely. That's reset and in *pow!*
back to the BIOS startup. No panic, no crashdump, just a reset. This
appeared to happen most often when MiniportReset() was called. (As
to why MiniportReset() was being called, see problem three below.)
I thought maybe I had created some sort of horrible deadlock
condition in the process of adding the serialization, but after three
weeks, at least 6 different locking implementations and heroic efforts
to debug the spinlock code, the machine still kept resetting. Finally,
I started single stepping through the MiniportReset() routine in
the driver using the kernel debugger, and this ultimately led me to
the source of the problem.
One of the last things the Atheros MiniportReset() routine does is
call NdisReadPciSlotInformation() several times to inspect a portion
of the device's PCI config space. It reads the same chunk of config
space repeatedly, in rapid succession. Presumeably, it's polling
the hardware for some sort of event. The reset occurs partway through
this process. I discovered that when I single-stepped through this
portion of the routine, the reset didn't occur. So I inserted a 1
microsecond delay into the read loop in NdisReadPciSlotInformation().
Suddenly, the reset was gone!!
I'm still very puzzled by the whole thing. What I suspect is happening
is that reading the PCI config space so quickly is causing a severe
PCI bus error. My test system is a Sun w2100z dual Opteron system,
and the NIC is a miniPCI card mounted in a miniPCI-to-PCI carrier card,
plugged into a 100Mhz PCI slot. It's possible that this combination of
hardware causes a bus protocol violation in this scenario which leads
to a fatal machine check. This is pure speculation though. Really all I
know for sure is that inserting the delay makes the problem go away.
(To quote Homer Simpson: "I don't know how it works, but fire makes
it good!")
Problem number three: NdisAllocatePacket() needs to make sure to
initialize the npp_validcounts field in the 'private' section of
the NDIS_PACKET structure. The reason if_ndis was calling the
MiniportReset() routine in the first place is that packet transmits
were sometimes hanging. When sending a packet, an NDIS driver will
call NdisQueryPacket() to learn how many physical buffers the packet
resides in. NdisQueryPacket() is actually a macro, which traverses
the NDIS_BUFFER list attached to the NDIS_PACKET and stashes some
of the results in the 'private' section of the NDIS_PACKET. It also
sets the npp_validcounts field to TRUE To indicate that the results are
now valid. The problem is, now that if_ndis creates a pool of transmit
packets via NdisAllocatePacketPool(), it's important that each time
a new packet is allocated via NdisAllocatePacket() that validcounts
be initialized to FALSE. If it isn't, and a previously transmitted
NDIS_PACKET is pulled out of the pool, it may contain stale data
from a previous transmission which won't get updated by NdisQueryPacket().
This would cause the driver to miscompute the number of fragments
for a given packet, and botch the transmission.
Fixing these three problems seems to make the Atheros driver happy
on SMP, which hopefully means other serialized miniports will be
happy too.
And there was much rejoicing.
Other stuff fixed along the way:
- Modified ndis_thsuspend() to take a mutex as an argument. This
allows KeWaitForSingleObject() and KeWaitForMultipleObjects() to
avoid any possible race conditions with other routines that
use the dispatcher lock.
- Fixed KeCancelTimer() so that it returns the correct value for
'pending' according to the Microsoft documentation
- Modfied NdisGetSystemUpTime() to use ticks and hz rather than
calling nanouptime(). Also added comment that this routine wraps
after 49.7 days.
- Added macros for KeAcquireSpinLock()/KeReleaseSpinLock() to hide
all the MSCALL() goop.
- For x86, KeAcquireSpinLockRaiseToDpc() needs to be a separate
function. This is because it's supposed to be _stdcall on the x86
arch, whereas KeAcquireSpinLock() is supposed to be _fastcall.
On amd64, all routines use the same calling convention so we can
just map KeAcquireSpinLockRaiseToDpc() directly to KfAcquireSpinLock()
and it will work. (The _fastcall attribute is a no-op on amd64.)
- Implement and use IoInitializeDpcRequest() and IoRequestDpc() (they're
just macros) and use them for interrupt handling. This allows us to
move the ndis_intrtask() routine from if_ndis.c to kern_ndis.c.
- Fix the MmInitializeMdl() macro so that is uses sizeof(vm_offset_t)
when computing mdl_size instead of uint32_t, so that it matches the
MmSizeOfMdl() routine.
- Change a could of M_WAITOKs to M_NOWAITs in the unicode routines in
subr_ndis.c.
- Use the dispatcher lock a little more consistently in subr_ntoskrnl.c.
- Get rid of the "wait for link event" hack in ndis_init(). Now that
I fixed NdisReadPciSlotInformation(), it seems I don't need it anymore.
This should fix the witness panic a couple of people have reported.
- Use MSCALL1() when calling the MiniportHangCheck() function in
ndis_ticktask(). I accidentally missed this one when adding the
wrapping for amd64.
at some point result in a status event being triggered (it should
be a link down event: the Microsoft driver design guide says you
should generate one when the NIC is initialized). Some drivers
generate the event during MiniportInitialize(), such that by the
time MiniportInitialize() completes, the NIC is ready to go. But
some drivers, in particular the ones for Atheros wireless NICs,
don't generate the event until after a device interrupt occurs
at some point after MiniportInitialize() has completed.
The gotcha is that you have to wait until the link status event
occurs one way or the other before you try to fiddle with any
settings (ssid, channel, etc...). For the drivers that set the
event sycnhronously this isn't a problem, but for the others
we have to pause after calling ndis_init_nic() and wait for the event
to arrive before continuing. Failing to wait can cause big trouble:
on my SMP system, calling ndis_setstate_80211() after ndis_init_nic()
completes, but _before_ the link event arrives, will lock up or
reset the system.
What we do now is check to see if a link event arrived while
ndis_init_nic() was running, and if it didn't we msleep() until
it does.
Along the way, I discovered a few other problems:
- Defered procedure calls run at PASSIVE_LEVEL, not DISPATCH_LEVEL.
ntoskrnl_run_dpc() has been fixed accordingly. (I read the documentation
wrong.)
- Similarly, the NDIS interrupt handler, which is essentially a
DPC, also doesn't need to run at DISPATCH_LEVEL. ndis_intrtask()
has been fixed accordingly.
- MiniportQueryInformation() and MiniportSetInformation() run at
DISPATCH_LEVEL, and each request must complete before another
can be submitted. ndis_get_info() and ndis_set_info() have been
fixed accordingly.
- Turned the sleep lock that guards the NDIS thread job list into
a spin lock. We never do anything with this lock held except manage
the job list (no other locks are held), so it's safe to do this,
and it's possible that ndis_sched() and ndis_unsched() can be
called from DISPATCH_LEVEL, so using a sleep lock here is
semantically incorrect. Also updated subr_witness.c to add the
lock to the order list.
both consuming 1K of stack space. This is unfriendly. Allocate the buffers
off the heap instead. It's a little slower, but these aren't performance
critical routines.
Also, add a spinlock to NdisAllocatePacketPool(), NdisAllocatePacket(),
NdisFreePacketPool() and NdisFreePacket(). The pool is maintained as a
linked list. I don't know for a fact that it can be corrupted, but why
take chances.
malloc(sizeof(device_object), ...) by mistake. Correct this, and
rename "dobj" to "drv" to make it a bit clearer what this variable
is supposed to be.
Spotted by: Mikore Li at Sun dot comnospamplzkthx
that describe a buffer of variable size). The problem is, allocating
MDLs off the heap is slow, and it can happen that drivers will allocate
lots and lots of lots of MDLs as they run.
As a compromise, we now do the following: we pre-allocate a zone for
MDLs big enough to describe any buffer with 16 or less pages. If
IoAllocateMdl() needs a MDL for a buffer with 16 or less pages, we'll
allocate it from the zone. Otherwise, we allocate it from the heap.
MDLs allocate from the zone have a flag set in their mdl_flags field.
When the MDL is released, IoMdlFree() will uma_zfree() the MDL if
it has the MDL_ZONE_ALLOCED flag set, otherwise it will release it
to the heap.
The assumption is that 16 pages is a "big number" and we will rarely
need MDLs larger than that.
- Moved the ndis_buffer zone to subr_ntoskrnl.c from kern_ndis.c
and named it mdl_zone.
- Modified IoAllocateMdl() and IoFreeMdl() to use uma_zalloc() and
uma_zfree() if necessary.
- Made ndis_mtop() use IoAllocateMdl() instead of calling uma_zalloc()
directly.
Inspired by: discussion with Giridhar Pemmasani
when we create a PDO, the driver_object associated with it is that
of the parent driver, not the driver we're trying to attach. For
example, if we attach a PCI device, the PDO we pass to the NdisAddDevice()
function should contain a pointer to fake_pci_driver, not to the NDIS
driver itself. For PCI or PCMCIA devices this doesn't matter because
the child never needs to talk to the parent bus driver, but for USB,
the child needs to be able to send IRPs to the parent USB bus driver, and
for that to work the parent USB bus driver has to be hung off the PDO.
This involves modifying windrv_lookup() so that we can search for
bus drivers by name, if necessary. Our fake bus drivers attach themselves
as "PCI Bus," "PCCARD Bus" and "USB Bus," so we can search for them
using those names.
The individual attachment stubs now create and attach PDOs to the
parent bus drivers instead of hanging them off the NDIS driver's
object, and in if_ndis.c, we now search for the correct driver
object depending on the bus type, and use that to find the correct PDO.
With this fix, I can get my sample USB ethernet driver to deliver
an IRP to my fake parent USB bus driver's dispatch routines.
- Add stub modules for USB support: subr_usbd.c, usbd_var.h and
if_ndis_usb.c. The subr_usbd.c module is hooked up the build
but currently doesn't do very much. It provides the stub USB
parent driver object and a dispatch routine for
IRM_MJ_INTERNAL_DEVICE_CONTROL. The only exported function at
the moment is USBD_GetUSBDIVersion(). The if_ndis_usb.c stub
compiles, but is not hooked up to the build yet. I'm putting
these here so I can keep them under source code control as I
flesh them out.
- In kern_ndis.c:ndis_unload_driver(), test that ndis_block->nmb_rlist
is not NULL before trying to free() it.
- In subr_pe.c:pe_get_import_descriptor(), do a case-insensitive
match on the import module name. Most drivers I have encountered
link against "ntoskrnl.exe" but the ASIX USB ethernet driver I'm
testing with wants "NTOSKRNL.EXE."
- In subr_ntoskrnl.c:IoAllocateIrp(), return a pointer to the IRP
instead of NULL. (Stub code leftover.)
- Also in subr_ntoskrnl.c, add ExAllocatePoolWithTag() and ExFreePool()
to the function table list so they'll get exported to drivers properly.
and a machine-independent though inefficient InterlockedExchange().
In Windows, InterlockedExchange() appears to be implemented in header
files via inline assembly. I would prefer using an atomic.h macro for
this, but there doesn't seem to be one that just does a plain old
atomic exchange (as opposed to compare and exchange). Also implement
IoSetCancelRoutine(), which is just a macro that uses InterlockedExchange().
Fill in IoBuildSynchronousFsdRequest(), IoBuildAsynchronousFsdRequest()
and IoBuildDeviceIoControlRequest() so that they do something useful,
and add a bunch of #defines to ntoskrnl_var.h to help make these work.
These may require some tweaks later.
copy op to shift arguments on the stack instead of transfering each
argument one by one through a register. Probably doesn't affect overall
operation, but makes the code a little less grotty and easier to update
later if I choose to make the wrapper handle more args. Also add
comments.
for now) exactly the same as KfAcquireSpinLock() and KfReleaseSpinLock().
I implemented the former as small routines in subr_ntoskrnl.c that just
turned around and invoked the latter. But I don't really need the wrapper
routines: I can just create an entries in the ntoskrnl func table that
map KeAcquireSpinLockRaiseToDpc() and KeReleaseSpinLock() to
KfAcquireSpinLock() and KfReleaseSpinLock() directly. This means
the stubs can go away.
Ville-Pertti Keinonen (will at exomi dot comohmygodnospampleasekthx)
deserves a big thanks for submitting initial patches to make it
work. I have mangled his contributions appropriately.
The main gotcha with Windows/x86-64 is that Microsoft uses a different
calling convention than everyone else. The standard ABI requires using
6 registers for argument passing, with other arguments on the stack.
Microsoft uses only 4 registers, and requires the caller to leave room
on the stack for the register arguments incase the callee needs to
spill them. Unlike x86, where Microsoft uses a mix of _cdecl, _stdcall
and _fastcall, all routines on Windows/x86-64 uses the same convention.
This unfortunately means that all the functions we export to the
driver require an intermediate translation wrapper. Similarly, we have
to wrap all calls back into the driver binary itself.
The original patches provided macros to wrap every single routine at
compile time, providing a secondary jump table with a customized
wrapper for each exported routine. I decided to use a different approach:
the call wrapper for each function is created from a template at
runtime, and the routine to jump to is patched into the wrapper as
it is created. The subr_pe module has been modified to patch in the
wrapped function instead of the original. (On x86, the wrapping
routine is a no-op.)
There are some minor API differences that had to be accounted for:
- KeAcquireSpinLock() is a real function on amd64, not a macro wrapper
around KfAcquireSpinLock()
- NdisFreeBuffer() is actually IoFreeMdl(). I had to change the whole
NDIS_BUFFER API a bit to accomodate this.
Bugs fixed along the way:
- IoAllocateMdl() always returned NULL
- kern_windrv.c:windrv_unload() wasn't releasing private driver object
extensions correctly (found thanks to memguard)
This has only been tested with the driver for the Broadcom 802.11g
chipset, which was the only Windows/x86-64 driver I could find.
Windows DRIVER_OBJECT and DEVICE_OBJECT mechanism so that we can
simulate driver stacking.
In Windows, each loaded driver image is attached to a DRIVER_OBJECT
structure. Windows uses the registry to match up a given vendor/device
ID combination with a corresponding DRIVER_OBJECT. When a driver image
is first loaded, its DriverEntry() routine is invoked, which sets up
the AddDevice() function pointer in the DRIVER_OBJECT and creates
a dispatch table (based on IRP major codes). When a Windows bus driver
detects a new device, it creates a Physical Device Object (PDO) for
it. This is a DEVICE_OBJECT structure, with semantics analagous to
that of a device_t in FreeBSD. The Windows PNP manager will invoke
the driver's AddDevice() function and pass it pointers to the DRIVER_OBJECT
and the PDO.
The AddDevice() function then creates a new DRIVER_OBJECT structure of
its own. This is known as the Functional Device Object (FDO) and
corresponds roughly to a private softc instance. The driver uses
IoAttachDeviceToDeviceStack() to add this device object to the
driver stack for this PDO. Subsequent drivers (called filter drivers
in Windows-speak) can be loaded which add themselves to the stack.
When someone issues an IRP to a device, it travel along the stack
passing through several possible filter drivers until it reaches
the functional driver (which actually knows how to talk to the hardware)
at which point it will be completed. This is how Windows achieves
driver layering.
Project Evil now simulates most of this. if_ndis now has a modevent
handler which will use MOD_LOAD and MOD_UNLOAD events to drive the
creation and destruction of DRIVER_OBJECTs. (The load event also
does the relocation/dynalinking of the image.) We don't have a registry,
so the DRIVER_OBJECTS are stored in a linked list for now. Eventually,
the list entry will contain the vendor/device ID list extracted from
the .INF file. When ndis_probe() is called and detectes a supported
device, it will create a PDO for the device instance and attach it
to the DRIVER_OBJECT just as in Windows. ndis_attach() will then call
our NdisAddDevice() handler to create the FDO. The NDIS miniport block
is now a device extension hung off the FDO, just as it is in Windows.
The miniport characteristics table is now an extension hung off the
DRIVER_OBJECT as well (the characteristics are the same for all devices
handled by a given driver, so they don't need to be per-instance.)
We also do an IoAttachDeviceToDeviceStack() to put the FDO on the
stack for the PDO. There are a couple of fake bus drivers created
for the PCI and pccard buses. Eventually, there will be one for USB,
which will actually accept USB IRP.s
Things should still work just as before, only now we do things in
the proper order and maintain the correct framework to support passing
IRPs between drivers.
Various changes:
- corrected the comments about IRQL handling in subr_hal.c to more
accurately reflect reality
- update ndiscvt to make the drv_data symbol in ndis_driver_data.h a
global so that if_ndis_pci.o and/or if_ndis_pccard.o can see it.
- Obtain the softc pointer from the miniport block by referencing
the PDO rather than a private pointer of our own (nmb_ifp is no
longer used)
- implement IoAttachDeviceToDeviceStack(), IoDetachDevice(),
IoGetAttachedDevice(), IoAllocateDriverObjectExtension(),
IoGetDriverObjectExtension(), IoCreateDevice(), IoDeleteDevice(),
IoAllocateIrp(), IoReuseIrp(), IoMakeAssociatedIrp(), IoFreeIrp(),
IoInitializeIrp()
- fix a few mistakes in the driver_object and device_object definitions
- add a new module, kern_windrv.c, to handle the driver registration
and relocation/dynalinkign duties (which don't really belong in
kern_ndis.c).
- made ndis_block and ndis_chars in the ndis_softc stucture pointers
and modified all references to it
- fixed NdisMRegisterMiniport() and NdisInitializeWrapper() so they
work correctly with the new driver_object mechanism
- changed ndis_attach() to call NdisAddDevice() instead of ndis_load_driver()
(which is now deprecated)
- used ExAllocatePoolWithTag()/ExFreePool() in lookaside list routines
instead of kludged up alloc/free routines
- added kern_windrv.c to sys/modules/ndis/Makefile and files.i386.
attributes in casts (i.e. foo = (__stdcall sometype)bar). This only
happens in two places where we need to set up function pointers, so
work around the problem with some void pointer magic.
USB device support):
- Convert all of my locally chosen function names to their actual
Windows equivalents, where applicable. This is a big no-op change
since it doesn't affect functionality, but it helps avoid a bit
of confusion (it's now a lot easier to see which functions are
emulated Windows API routines and which are just locally defined).
- Turn ndis_buffer into an mdl, like it should have been. The structure
is the same, but now it belongs to the subr_ntoskrnl module.
- Implement a bunch of MDL handling macros from Windows and use them where
applicable.
- Correct the implementation of IoFreeMdl().
- Properly implement IoAllocateMdl() and MmBuildMdlForNonPagedPool().
- Add the definitions for struct irp and struct driver_object.
- Add IMPORT_FUNC() and IMPORT_FUNC_MAP() macros to make formatting
the module function tables a little cleaner. (Should also help
with AMD64 support later on.)
- Fix if_ndis.c to use KeRaiseIrql() and KeLowerIrql() instead of
the previous calls to hal_raise_irql() and hal_lower_irql() which
have been renamed.
The function renaming generated a lot of churn here, but there should
be very little operational effect.
calls MiniportQueryInformation(), it will return NDIS_STATUS_PENDING.
When this happens, ndis_get_info() will sleep waiting for a completion
event. If two threads call ndis_get_info() and both end up having to
sleep, they will both end up waiting on the same wait channel, which
can cause a panic in sleepq_add() if INVARIANTS are turned on.
Fix this by having ndis_get_info() use a common mutex rather than
using the process mutex with PROC_LOCK(). Also do the same for
ndis_set_info(). Note that Pierre's original patch also made ndis_thsuspend()
use the new mutex, but ndis_thsuspend() shouldn't need this since
it will make each thread that calls it sleep on a unique wait channel.
Also, it occured to me that we probably don't want to enter
MiniportQueryInformation() or MiniportSetInformation() from more
than one thread at any given time, so now we acquire a Windows
spinlock before calling either of them. The Microsoft documentation
says that MiniportQueryInformation() and MiniportSetInformation()
are called at DISPATCH_LEVEL, and previously we would call
KeRaiseIrql() to set the IRQL to DISPATCH_LEVEL before entering
either routine, but this only guarantees mutual exclusion on
uniprocessor machines. To make it SMP safe, we need to use a real
spinlock. For now, I'm abusing the spinlock embedded in the
NDIS_MINIPORT_BLOCK structure for this purpose. (This may need to be
applied to some of the other routines in kern_ndis.c at a later date.)
Export ntoskrnl_init_lock() (KeInitializeSpinlock()) from subr_ntoskrnl.c
since we need to use in in kern_ndis.c, and since it's technically part
of the Windows kernel DDK API along with the other spinlock routines. Use
it in subr_ndis.c too rather than frobbing the spinlock directly.
These normally only manifest if the ndis compat module is statically
compiled into a kernel image by way of 'options NDISAPI'.
Submitted by: Dmitri Nikulin
Approved by: wpaul
PR: kern/71449
MFC after: 1 week
This was tested with a Netgear WG311v2 802.11b/g PCI card. Things
that were fixed:
- This chip has two memory mapped regions, one at PCIR_BAR(0) and the
other at PCIR_BAR(1). This is a little different from the other
chips I've seen with two PCI shared memory regions, since they tend
to have the second BAR ad PCIR_BAR(2). if_ndis_pci.c tests explicitly
for PCIR_BAR(2). This has been changed to simply fill in ndis_res_mem
first and ndis_res_altmem second, if a second shared memory range
exists. Given that NDIS drivers seem to scan for BARs in ascending
order, I think this should be ok.
- Fixed the code that tries to process firmware images that have been
loaded as .ko files. To save a step, I was setting up the address
mapping in ndis_open_file(), but ndis_map_file() flags pre-existing
mappings as an error (to avoid duplicate mappings). Changed this so
that the mapping is now donw in ndis_map_file() as expected.
- Made the typedef for 'driver_entry' explicitly include __stdcall
to silence gcc warning in ndis_load_driver().
NOTE: the Texas Instruments ACX111 driver needs firmware. With my
card, there were 3 .bin files shipped with the driver. You must
either put these files in /compat/ndis or convert them with
ndiscvt -f and kldload them so the driver can use them. Without
the firmware image, the NIC won't work.
- In ntoskrnl_var.h, I had defined compat macros for
ntoskrnl_acquire_spinlock() and ntoskrnl_release_spinlock() but
never used them. This is fortunate since they were stale. Fix them
to work properly. (In Windows/x86 KeAcquireSpinLock() is a macro that
calls KefAcquireSpinLock(), which lives in HAL.dll. To imitate this,
ntoskrnl_acquire_spinlock() is just a macro that calls hal_lock(),
which lives in subr_hal.o.)
- Add macros for ntoskrnl_raise_irql() and ntoskrnl_lower_irql() that
call hal_raise_irql() and hal_lower_irql().
- Use these macros in kern_ndis.c, subr_ndis.c and subr_ntoskrnl.c.
- Along the way, I realised subr_ndis.c:ndis_lock() was not calling
hal_lock() correctly (it was using the FASTCALL2() wrapper when
in reality this routine is FASTCALL1()). Using the
ntoskrnl_acquire_spinlock() fixes this. Not sure if this actually
caused any bugs since hal_lock() would have just ignored what
was in %edx, but it was still bogus.
This hides many of the uses of the FASTCALLx() macros which makes the
code a little cleaner. Should not have any effect on generated object
code, other than the one fix in ndis_lock().
- Give ndiscvt(8) the ability to process a .SYS file directly into
a .o file so that we don't have to emit big messy char arrays into
the ndis_driver_data.h file. This behavior is currently optional, but
may become the default some day.
- Give ndiscvt(8) the ability to turn arbitrary files into .ko files
so that they can be pre-loaded or kldloaded. (Both this and the
previous change involve using objcopy(1)).
- Give NdisOpenFile() the ability to 'read' files out of kernel memory
that have been kldloaded or pre-loaded, and disallow the use of
the normal vn_open() file opening method during bootstrap (when no
filesystems have been mounted yet). Some people have reported that
kldloading if_ndis.ko works fine when the system is running multiuser
but causes a panic when the modile is pre-loaded by /boot/loader. This
happens with drivers that need to use NdisOpenFile() to access
external files (i.e. firmware images). NdisOpenFile() won't work
during kernel bootstrapping because no filesystems have been mounted.
To get around this, you can now do the following:
o Say you have a firmware file called firmware.img
o Do: ndiscvt -f firmware.img -- this creates firmware.img.ko
o Put the firmware.img.ko in /boot/kernel
o add firmware.img_load="YES" in /boot/loader.conf
o add if_ndis_load="YES" and ndis_load="YES" as well
Now the loader will suck the additional file into memory as a .ko. The
phony .ko has two symbols in it: filename_start and filename_end, which
are generated by objcopy(1). ndis_open_file() will traverse each module
in the module list looking for these symbols and, if it finds them, it'll
use them to generate the file mapping address and length values that
the caller of NdisOpenFile() wants.
As a bonus, this will even work if the file has been statically linked
into the kernel itself, since the "kernel" module is searched too.
(ndiscvt(8) will generate both filename.o and filename.ko for you).
- Modify the mechanism used to provide make-pretend FASTCALL support.
Rather than using inline assembly to yank the first two arguments
out of %ecx and %edx, we now use the __regparm__(3) attribute (and
the __stdcall__ attribute) and use some macro magic to re-order
the arguments and provide dummy arguments as needed so that the
arguments passed in registers end up in the right place. Change
taken from DragonflyBSD version of the NDISulator.
actually work.
Make the PCI and PCCARD attachments provide a bus_get_resource_list()
method so that resource listing for PCCARD works. PCCARD does not
have a bus_get_resource_list() method (yet), so I faked up the
resource list management in if_ndis_pccard.c, and added
bus_get_resource_list() methods to both if_ndis_pccard.c and if_ndis_pci.c.
The one in the PCI attechment just hands off to the PCI bus code.
The difference is transparent to the NDIS resource handler code.
Fixed ndis_open_file() so that opening files which live on NFS
filesystems work: pass an actual ucred structure to VOP_GETATTR()
(NFS explodes if the ucred structure is NOCRED).
Make NdisMMapIoSpace() handle mapping of PCMCIA attribute memory
resources correctly.
Turn subr_ndis.c:my_strcasecmp() into ndis_strcasecmp() and export
it so that if_ndis_pccard.c can use it, and junk the other copy
of my_strcasecmp() from if_ndis_pccard.c.
- In subr_ndis.c:ndis_allocate_sharemem(), create the busdma tags
used for shared memory allocations with a lowaddr of 0x3E7FFFFF.
This forces the buffers to be mapped to physical/bus addresses within
the first 1GB of physical memory. It seems that at least one card
(Linksys Instant Wireless PCI V2.7) depends on this behavior. I
don't know if this is a hardware restriction, or if the NDIS
driver for this card is truncating the addresses itself, but using
physical/bus addresses beyong the 1GB limit causes initialization
failures.
- Create am NDIS_INITIALIZED() macro in if_ndisvar.h and use it in
if_ndis.c to test whether the device has been initialized rather
than checking for the presence of the IFF_UP flag in if_flags.
While debugging the previous problem, I noticed that bringing
up the device would always produce failures from ndis_setmulti().
It turns out that the following steps now occur during device
initialization:
- IFF_UP flag is set in if_flags
- ifp->if_ioctl() called with SIOCSIFADDR (which we don't handle)
- ifp->if_ioctl() called with SIOCADDMULTI
- ifp->if_ioctl() called with SIOCADDMULTI (again)
- ifp->if_ioctl() called with SIOCADDMULTI (yet again)
- ifp->if_ioctl() called with SIOCSIFFLAGS
Setting the receive filter and multicast filters can only be done
when the underlying NDIS driver has been initialized, which is done
by ifp->if_init(). However, we don't call ifp->if_init() until
ifp->if_ioctl() is called with SIOCSIFFLAGS and IFF_UP has been
set. It appears that now, the network stack tries to add multicast
addresses to interface's filter before those steps occur. Normally,
ndis_setmulti() would trap this condition by checking for the IFF_UP
flag, but the network code has in fact set this flag already, so
ndis_setmulti() is fooled into thinking the interface has been
initialized when it really hasn't.
It turns out this is usually harmless because the ifp->if_init()
routine (in this case ndis_init()) will set up the multicast
filter when it initializes the hardware anyway, and the underlying
routines (ndis_get_info()/ndis_set_info()) know that the driver/NIC
haven't been initialized yet, but you end up spurious error messages
on the console all the time.
Something tells me this new behavior isn't really correct. I think
the intention was to fix it so that ifp->if_init() is only called
once when we ifconfig an interface up, but the end result seems a
little bogus: the change of the IFF_UP flag should be propagated
down to the driver before calling any other ioctl() that might actually
require the hardware to be up and running.
make the key name matching case-insensitive. There are some drivers
and .inf files that have mismatched cases, e.g. the driver will look
for "AdhocBand" whereas the .inf file specifies a registry key to be
created called "AdHocBand." The mismatch is probably a typo that went
undetected (so much for QA), but since Windows seems to be case-insensitive,
we should be too.
In if_ndis.c, initialize rates and channels correctly so that specify
frequences correctly when trying to set channels in the 5Ghz band, and
so that 802.11b rates show up for some a/b/g cards (which otherwise
appear to have no 802.11b modes).
Also, when setting OID_802_11_CONFIGURATION in ndis_80211_setstate(),
provide default values for the beacon interval, ATIM window and dwelltime.
The Atheros "Aries" driver will crash if you try to select ad-hoc mode
and leave the beacon interval set to 0: it blindly uses this value and
does a division by 0 in the interrupt handler, causing an integer
divide trap.
- Use the dh_inserted member of the dispatch header in the Windows
timer structure to indicate that the timer has been "inserted into
the timer queue" (i.e. armed via timeout()). Use this as the value
to return to the caller in KeCancelTimer(). Previously, I was using
callout_pending(), but you can't use that with timeout()/untimeout()
without creating a potential race condition.
- Make ntoskrnl_init_timer() just a wrapper around ntoskrnl_init_timer_ex()
(reduces some code duplication).
- Drop Giant when entering if_ndis.c:ndis_tick() and
subr_ntorkrnl.c:ntoskrnl_timercall(). At the moment, I'm forced to
use system callwheel via timeout()/untimeout() to handle timers rather
than the callout API (struct callout is too big to fit inside the
Windows struct KTIMER, so I'm kind of hosed). Unfortunately, all
the callouts in the callwhere are not marked as MPSAFE, so when
one of them fires, it implicitly acquires Giant before invoking the
callback routine (and releases it when it returns). I don't need to
hold Giant, but there's no way to stop the callout code from acquiring
it as long as I'm using timeout()/untimeout(), so for now we cheat
by just dropping Giant right away (and re-acquiring it right before
the routine returns so keep the callout code happy). At some point,
I will need to solve this better, but for now this should be a suitable
workaround.
(I hope.)
My original instinct to make ndis_return_packet() asynchronous was correct.
Making ndis_rxeof() submit packets to the stack asynchronously fixes
one recursive spinlock acquisition, but it's also possible for it to
happen via the ndis_txeof() path too. So:
- In if_ndis.c, revert ndis_rxeof() to its old behavior (and don't bother
putting ndis_rxeof_serial() back since we don't need it anymore).
- In kern_ndis.c, make ndis_return_packet() submit the call to the
MiniportReturnPacket() function to the "ndis swi" thread so that
it always happens in another context no matter who calls it.
ntoskrnl_unlocl_dpc().
- hal_raise_irql(), hal_lower_irql() and hal_irql() didn't work right
on SMP (priority inheritance makes things... interesting). For now,
use only two states: DISPATCH_LEVEL (PI_REALTIME) and PASSIVE_LEVEL
(everything else). Tested on a dual PIII box.
- Use ndis_thsuspend() in ndis_sleep() instead of tsleep(). (I added
ndis_thsuspend() and ndis_thresume() to replace kthread_suspend()
and kthread_resume(); the former will preserve a thread's priority
when it wakes up, the latter will not.)
- Change use of tsleep() in ndis_stop_thread() to prevent priority
change on wakeup.
attempting to duplicate Windows spinlocks. Windows spinlocks differ
from FreeBSD spinlocks in the way they block preemption. FreeBSD
spinlocks use critical_enter(), which masks off _all_ interrupts.
This prevents any other threads from being scheduled, but it also
prevents ISRs from running. In Windows, preemption is achieved by
raising the processor IRQL to DISPATCH_LEVEL, which prevents other
threads from preempting you, but does _not_ prevent device ISRs
from running. (This is essentially what Solaris calls dispatcher
locks.) The Windows spinlock itself (kspin_lock) is just an integer
value which is atomically set when you acquire the lock and atomically
cleared when you release it.
FreeBSD doesn't have IRQ levels, so we have to cheat a little by
using thread priorities: normal thread priority is PASSIVE_LEVEL,
lowest interrupt thread priority is DISPATCH_LEVEL, highest thread
priority is DEVICE_LEVEL (PI_REALTIME) and critical_enter() is
HIGH_LEVEL. In practice, only PASSIVE_LEVEL and DISPATCH_LEVEL
matter to us. The immediate benefit of all this is that I no
longer have to rely on a mutex pool.
Now, I'm sure many people will be seized by the urge to criticize
me for doing an end run around our own spinlock implementation, but
it makes more sense to do it this way. Well, it does to me anyway.
Overview of the changes:
- Properly implement hal_lock(), hal_unlock(), hal_irql(),
hal_raise_irql() and hal_lower_irql() so that they more closely
resemble their Windows counterparts. The IRQL is determined by
thread priority.
- Make ntoskrnl_lock_dpc() and ntoskrnl_unlock_dpc() do what they do
in Windows, which is to atomically set/clear the lock value. These
routines are designed to be called from DISPATCH_LEVEL, and are
actually half of the work involved in acquiring/releasing spinlocks.
- Add FASTCALL1(), FASTCALL2() and FASTCALL3() macros/wrappers
that allow us to call a _fastcall function in spite of the fact
that our version of gcc doesn't support __attribute__((__fastcall__))
yet. The macros take 1, 2 or 3 arguments, respectively. We need
to call hal_lock(), hal_unlock() etc... ourselves, but can't really
invoke the function directly. I could have just made the underlying
functions native routines and put _fastcall wrappers around them for
the benefit of Windows binaries, but that would create needless bloat.
- Remove ndis_mtxpool and all references to it. We don't need it
anymore.
- Re-implement the NdisSpinLock routines so that they use hal_lock()
and friends like they do in Windows.
- Use the new spinlock methods for handling lookaside lists and
linked list updates in place of the mutex locks that were there
before.
- Remove mutex locking from ndis_isr() and ndis_intrhand() since they're
already called with ndis_intrmtx held in if_ndis.c.
- Put ndis_destroy_lock() code under explicit #ifdef notdef/#endif.
It turns out there are some drivers which stupidly free the memory
in which their spinlocks reside before calling ndis_destroy_lock()
on them (touch-after-free bug). The ADMtek wireless driver
is guilty of this faux pas. (Why this doesn't clobber Windows I
have no idea.)
- Make NdisDprAcquireSpinLock() and NdisDprReleaseSpinLock() into
real functions instead of aliasing them to NdisAcaquireSpinLock()
and NdisReleaseSpinLock(). The Dpr routines use
KeAcquireSpinLockAtDpcLevel() level and KeReleaseSpinLockFromDpcLevel(),
which acquires the lock without twiddling the IRQL.
- In ndis_linksts_done(), do _not_ call ndis_80211_getstate(). Some
drivers may call the status/status done callbacks as the result of
setting an OID: ndis_80211_getstate() gets OIDs, which means we
might cause the driver to recursively access some of its internal
structures unexpectedly. The ndis_ticktask() routine will call
ndis_80211_getstate() for us eventually anyway.
- Fix the channel setting code a little in ndis_80211_setstate(),
and initialize the channel to IEEE80211_CHAN_ANYC. (The Microsoft
spec says you're not supposed to twiddle the channel in BSS mode;
I may need to enforce this later.) This fixes the problems I was
having with the ADMtek adm8211 driver: we were setting the channel
to a non-standard default, which would cause it to fail to associate
in BSS mode.
- Use hal_raise_irql() to raise our IRQL to DISPATCH_LEVEL when
calling certain miniport routines, per the Microsoft documentation.
I think that's everything. Hopefully, other than fixing the ADMtek
driver, there should be no apparent change in behavior.
before calling BUS_GET_RESOURCE_LIST(). Previously, the list head would
only be initialized if BUS_GET_RESOURCE_LIST() succeeded; it needs to
be initialized unconditionally so that the list cleanup code won't
trip over potential stack garbage.
case we should wait for the resetdone handler to be called before
returning.
- When providing resources via ndis_query_resources(), uses the
computed rsclen when using bcopy() to copy out the resource data
rather than the caller-supplied buffer length.
- Avoid using ndis_reset_nic() in if_ndis.c unless we really need
to reset the NIC because of a problem.
- Allow interrupts to be fielded during ndis_attach(), at least
as far as allowing ndis_isr() and ndis_intrhand() to run.
- Use ndis_80211_rates_ex when probing for supported rates. Technically,
this isn't supposed to work since, although Microsoft added the extended
rate structure with the NDIS 5.1 update, the spec still says that
the OID_802_11_SUPPORTED_RATES OID uses ndis_80211_rates. In spite of
this, it appears some drivers use it anyway.
- When adding in our guessed rates, check to see if they already exist
so that we avoid any duplicates.
- Add a printf() to ndis_open_file() that alerts the user when a
driver attempts to open a file under /compat/ndis.
With these changes, I can get the driver for the SMC 2802W 54g PCI
card to load and run. This board uses a Prism54G chip. Note that in
order for this driver to work, you must place the supplied smc2802w.arm
firmware image under /compat/ndis. (The firmware is not resident on
the device.)
Note that this should also allow the 3Com 3CRWE154G72 card to work
as well; as far as I can tell, these cards also use a Prism54G chip.
resource_var.h.
In kern_ndis.c:ndis_convert_res(), fill in the cprd_flags and
cprd_sharedisp fields as best we can.
In if_ndis.c:ndis_setmulti(), don't bother updating the multicast
filter if our multicast address list is empty.
Add some missing updates to ndis_var.h and ntoskrnl_var.h that I
forgot to check in when I added the KeDpc stuff.
objects rather than synchronization objects. When a sync object is
signaled, only the first thread waiting on it is woken up, and then
it's automatically reset to the not-signaled state. When a
notification object is signaled, all threads waiting on it will
be woken up, and it remains in the signaled state until someone
resets it manually. We want the latter behavior for NDIS events.
- In kern_ndis.c:ndis_convert_res(), we have to create a temporary
copy of the list returned by BUS_GET_RESOURCE_LIST(). When the PCI
bus code probes resources for a given device, it enters them into
a singly linked list, head first. The result is that traversing
this list gives you the resources in reverse order. This means when
we create the Windows resource list, it will be in reverse order too.
Unfortunately, this can hose drivers for devices with multiple I/O
ranges of the same type, like, say, two memory mapped I/O regions (one
for registers, one to map the NVRAM/bootrom/whatever). Some drivers
test the range size to figure out which region is which, but others
just assume that the resources will be listed in ascending order from
lowest numbered BAR to highest. Reversing the order means such drivers
will choose the wrong resource as their I/O register range.
Since we can't traverse the resource SLIST backwards, we have to
make a temporary copy of the list in the right order and then build
the Windows resource list from that. I suppose we could just fix
the PCI bus code to use a TAILQ instead, but then I'd have to track
down all the consumers of the BUS_GET_RESOURCE_LIST() and fix them
too.
which pulls a job off a thread work queue (assuming it hasn't run yet).
This is needed for KeRemoveQueueDpc().
- In subr_ntoskrnl.c, implement KeInsertQueueDpc() and KeRemoveQueueDpc(),
to go with KeInitializeDpc() to round out the API. Also change the
KeTimer implementation to use this API instead of the private
timer callout scheduler. Functionality of the timer API remains
unchanged, but we get a couple new Windows kernel API routines and
more closely imitate the way thing works in Windows. (As of yet
I haven't encountered any drivers that use KeInsertQueueDpc() or
KeRemoveQueueDpc(), but it doesn't hurt to have them.)
when it associates with a net. Because FreeBSD's kstack size is only
2 pages by default, this blows the stack and causes a double fault.
To deal with this, we now create all our kthreads with 8 stack pages.
Also, we now run all timer callouts in the ndis swi thread (since
they would otherwise run in the clock ithread, whose stack is too
small). It happens that the alloca() in this case was occuring within
the interrupt handler, which was already running in the ndis swi
thread, but I want to deal with the callouts too just to be extra
safe.
NOTE: this will only work if you update vm_machdep.c with the change
I just committed. If you don't include this fix, setting the number
of stack pages with kthread_create() has essentially no effect.
are actually layered on top of the KeTimer API in subr_ntoskrnl.c, just
as it is in Windows. This reduces code duplication and more closely
imitates the way things are done in Windows.
- Modify ndis_encode_parm() to deal with the case where we have
a registry key expressed as a hex value ("0x1") which is being
read via NdisReadConfiguration() as an int. Previously, we tried
to decode things like "0x1" with strtol() using a base of 10, which
would always yield 0. This is what was causing problems with the
Intel 2200BG Centrino 802.11g driver: the .inf file that comes
with it has a key called RadioEnable with a value of 0x1. We
incorrectly decoded this value to '0' when it was queried, hence
the driver thought we wanted the radio turned off.
- In if_ndis.c, most drivers don't accept NDIS_80211_AUTHMODE_AUTO,
but NDIS_80211_AUTHMODE_SHARED may not be right in some cases,
so for now always use NDIS_80211_AUTHMODE_OPEN.
NOTE: There is still one problem with the Intel 2200BG driver: it
happens that the kernel stack in Windows is larger than the kernel
stack in FreeBSD. The 2200BG driver sometimes eats up more than 2
pages of stack space, which can lead to a double fault panic.
For the moment, I got things to work by adding the following to
my kernel config file:
options KSTACK_PAGES=8
I'm pretty sure 8 is too big; I just picked this value out of a hat
as a test, and it happened to work, so I left it. 4 pages might be
enough. Unfortunately, I don't think you can dynamically give a
thread a larger stack, so I'm not sure how to handle this short of
putting a note in the man page about it and dealing with the flood
of mail from people who never read man pages.
snprintf() and vsnprintf() in FreeBSD kernel land).
This is needed by the Intel Centrino 2200BG driver. Unfortunately, this
driver still doesn't work right with Project Evil even with this tweak,
but I'm unable to diagnose the problem since I don't have access to a
sample card.
for Windows are deserialized miniports. Such drivers maintain their own
queues and do their own locking. This particular driver is not deserialized
though, and we need special support to handle it correctly.
Typically, in the ndis_rxeof() handler, we pass all incoming packets
directly to (*ifp->if_input)(). This in turn may cause another thread
to run and preempt us, and the packet may actually be processed and
then released before we even exit the ndis_rxeof() routine. The
problem with this is that releasing a packet calls the ndis_return_packet()
function, which hands the packet and its buffers back to the driver.
Calling ndis_return_packet() before ndis_rxeof() returns will screw
up the driver's internal queues since, not being deserialized,
it does no locking.
To avoid this problem, if we detect a serialized driver (by checking
the attribute flags passed to NdisSetAttributesEx(), we use an alternate
ndis_rxeof() handler, ndis_rxeof_serial(), which puts the call to
(*ifp->if_input)() on the NDIS SWI work queue. This guarantees the
packet won't be processed until after ndis_rxeof_serial() returns.
Note that another approach is to always copy the packet data into
another mbuf and just let the driver retain ownership of the ndis_packet
structure (ndis_return_packet() never needs to be called in this
case). I'm not sure which method is faster.
that I added recently:
- When a periodic timer fires, it's automatically re-armed. We must
make sure to re-arm the timer _before_ invoking any caller-supplied
defered procedure call: the DPC may choose to call KeCancelTimer(),
and re-arming the timer after the DPC un-does the effect of the
cancel.
- Fix similar issue with periodic timers in subr_ndis.c.
- When calling KeSetTimer() or KeSetTimerEx(), if the timer is
already pending, untimeout() it first before timeout()ing
it again.
- The old Atheros driver for the 5211 seems to use KeSetTimerEx()
incorrectly, or at the very least in a very strange way that
doesn't quite follow the Microsoft documentation. In one case,
it calls KeSetTimerEx() with a duetime of 0 and a period of 5000.
The Microsoft documentation says that negative duetime values
are relative to the current time and positive values are absolute.
But it doesn't say what's supposed to happen with positive values
that less than the current time, i.e. absolute values that are
in the past.
Lacking any further information, I have decided that timers with
positive duetimes that are in the past should fire right away (or
in our case, after only 1 tick). This also takes care of the other
strange usage in the Atheros driver, where the duetime is
specified as 500000 and the period is 50. I think someone may
have meant to use -500000 and misinterpreted the documentation.
- Also modified KeWaitForSingleObject() and KeWaitForMultipleObjects()
to make the same duetime adjustment, since they have the same rules
regarding timeout values.
- Cosmetic: change name of 'timeout' variable in KeWaitForSingleObject()
and KeWaitForMultipleObjects() to 'duetime' to avoid senseless
(though harmless) overlap with timeout() function name.
With these fixes, I can get the 5211 card to associate properly with
my adhoc net using driver AR5211.SYS version 2.4.1.6.
if_ndis.c has been split into if_ndis_pci.c and if_ndis_pccard.c.
The ndiscvt(8) utility should be able to parse device info for PCMCIA
devices now. The ndis_alloc_amem() has moved from kern_ndis.c to
if_ndis_pccard.c so that kern_ndis.c no longer depends on pccard.
NOTE: this stuff is not guaranteed to work 100% correctly yet. So
far I have been able to load/init my PCMCIA Cisco Aironet 340 card,
but it crashes in the interrupt handler. The existing support for
PCI/cardbus devices should still work as before.
along with KeInitializeTimerEx(), KeSetTimer(), KeSetTimerEx(),
KeCancelTimer(), KeReadStateTimer() and KeInitializeDpc(). I don't
know for certain that these will make the Atheros driver happy since
I don't have the card/driver combo needed to test it, but these are
fairly independent so they shouldn't break anything else.
- Debugger() is present even in kernels without options DDB, so no
conditional compilation is necessary (pointed out by bde).
- Remove the extra km_acquirecnt member that I added to struct kmutant
and embed it within an unused portion of the structure instead, so that
we don't make the structure larger than it's defined to be in Windows.
I don't know what crack I was smoking when I decided it was ok to do
this, but it's worn off now.
routines to guard against problems caused by (possibly) buggy drivers.
The RealTek 8180 wireless driver calls NdisFreeBuffer() to release
some of its buffers _after_ it's already called NdisFreeBufferPool()
to destroy the pool to which the buffers belong. In our implementation,
this error causes NdisFreeBuffer() to touch stale heap memory.
If you are running a release kernel, and hence have INVARIANTS et al
turned off, it turns out nothing happens. But if you're using a
development kernel config with INVARIANTS on, the malloc()/free()
sanity checks will scribble over the pool memory with 0xdeadc0de
once it's released so that any attempts to touch it will cause a
trap, and indeed this is what happens. It happens that I run 5.2-RELEASE
on my laptop, so when I tested the rtl8180.sys driver, it worked fine
for me, but people trying to run it with development systems checked
out or cvsupped from -current would get a page fault on driver load.
I can't find any reason why the NDISulator would cause the RealTek
driver to do the NdisFreeBufferPool() prematurely, and the same driver
obviously works with Windows -- or at least, it doesn't cause a crash:
the Microsoft documentation for NdisFreeBufferPool() says that failing
to return all buffers to the pool before calling NdisFreeBufferPool()
causes a memory leak.
I've written to my contacts at RealTek asking them to check if this
is indeed a bug in their driver. In the meantime, these new sanity checks
will catch this problem and issue a warning rather than causing a trap.
The trick is to keep a count of outstanding buffers for each buffer pool,
and if the driver tries to call NdisFreeBufferPool() while there are still
buffers outstanding, we mark the pool for deletion and then defer
destroying it until after the last buffer has been reclaimed.
is set, since some drivers with debug info can be very chatty.
Also implement DbgBreakPoint(), which is the Windows equivalent of
Debugger(). Unfortunately, this forces subr_ntoskrnl.c to include
opt_ddb.h.
- When adding new waiting threads to the waitlist for an object,
use INSERT_LIST_TAIL() instead of INSERT_LIST_HEAD() so that new
waiters go at the end of the list instead of the beginning. When we
wake up a synchronization object, only the first waiter is awakened,
and this needs to be the first thread that actually waited on the object.
- Correct missing semicolon in INSERT_LIST_TAIL() macro.
- Implement lookaside lists correctly. Note that the Am1771 driver
uses lookaside lists to manage shared memory (i.e. DMAable) buffers
by specifying its own alloc and free routines. The Microsoft documentation
says you should avoid doing this, but apparently this did not deter
the developers at AMD from doing it anyway.
With these changes (which are the result of two straight days of almost
non-stop debugging), I think I finally have the object/thread handling
semantics implemented correctly. The Am1771 driver no longer crashes
unexpectedly during association or bringing the interface up.
The Am1771 driver will sometimes do the following:
- Some thread-> NdisScheduleWorkItem(some work)
- Worker thread -> do some work, KeWaitForSingleObject(some event)
- Some other thread -> NdisScheduleWorkItem(some other work)
When the second call to NdisScheduleWorkItem() occurs, the NDIS worker
thread (in our case ndis taskqueue) is suspended in KeWaitForSingleObject()
and waiting for an event to be signaled. This is different from when
the worker thread is idle and waiting on NdisScheduleWorkItem() to
send it more jobs. However, the ndis_sched() function in kern_ndis.c
always calls kthread_resume() when queueing a new job. Normally this
would be ok, but here this causes KeWaitForSingleObject() to return
prematurely, which is not what we want.
To fix this, the NDIS threads created by kern_ndis.c maintain a state
variable to indicate whether they are running (scanning the job list
and executing jobs) or sleeping (blocked on kthread_suspend() in
ndis_runq()), and ndis_sched() will only call kthread_resume() if
the thread is in the sleeping state.
Note that we can't just check to see if the thread is on the run queue:
in both cases, the thread is sleeping, but it's sleeping for different
reasons.
This stops the Am1771 driver from emitting various "NDIS ERROR" messages
and fixes some cases where it crashes.
routines to do anything except return error if the miniport adapter context
is not set (meaning we either having init'ed the driver yet, or the
initialization failed).
Also, be sure to NULL out the adapter context along with the
miniport characteristics pointers if calling the MiniportInitialize()
method fails.
802.11b chipset work. This chip is present on the SMC2602W version 3
NIC, which is what was used for testing. This driver creates kernel
threads (12 of them!) for various purposes, and required the following
routines:
PsCreateSystemThread()
PsTerminateSystemThread()
KeInitializeEvent()
KeSetEvent()
KeResetEvent()
KeInitializeMutex()
KeReleaseMutex()
KeWaitForSingleObject()
KeWaitForMultipleObjects()
IoGetDeviceProperty()
and several more. Also, this driver abuses the fact that NDIS events
and timers are actually Windows events and timers, and uses NDIS events
with KeWaitForSingleObject(). The NDIS event routines have been rewritten
to interface with the ntoskrnl module. Many routines with incorrect
prototypes have been cleaned up.
Also, this driver puts jobs on the NDIS taskqueue (via NdisScheduleWorkItem())
which block on events, and this interferes with the operation of
NdisMAllocateSharedMemoryAsync(), which was also being put on the
NDIS taskqueue. To avoid the deadlock, NdisMAllocateSharedMemoryAsync()
is now performed in the NDIS SWI thread instead.
There's still room for some cleanups here, and I really should implement
KeInitializeTimer() and friends.
Since we have a worker thread now, we can actually do the allocation
asynchronously in that thread's context. Also, we need to return a
status value: if we're unable to queue up the async allocation, we
return NDIS_STATUS_FAILURE, otherwise we return NDIS_STATUS_PENDING
to indicate the allocation has been queued and will occur later.
This replaces the kludge where we just invoked the callback routine
right away in the current context.
that Asus provides on its CDs has both a MiniportSend() routine
and a MiniportSendPackets() function. The Microsoft NDIS docs say
that if a driver has both, only the MiniportSendPackets() routine
will be used. Although I think I implemented the support correctly,
calling the MiniportSend() routine seems to result in no packets going
out on the air, even though no error status is returned. The
MiniportSendPackets() function does work though, so at least in
this case it doesn't matter.
In if_ndis.c:ndis_getstate_80211(), if ndis_get_assoc() returns
an error, don't bother trying to obtain any other state since the
calls may fail, or worse cause the underlying driver to crash.
(The above two changes make the Asus-supplied Centrino work.)
Also, when calling the OID_802_11_CONFIGURATION OID, remember
to initialize the structure lengths correctly.
In subr_ndis.c:ndis_open_file(), set the current working directory
to rootvnode if we're in a thread that doesn't have a current
working directory set.
and NdisCancelTimer(). NdisInitializeTimer() doesn't accept an NDIS
miniport context argument, so we have to derive it from the timer
function context (which is supposed to be the adapter private context).
NdisCancelTimer is now an alias for NdisMCancelTimer().
Also add stubs for NdisMRegisterDevice() and NdisMDeregisterDevice().
These are no-ops for now, but will likely get fleshed in once I start
working on the Am1771/Am1772 wireless driver.
attached when shutting down, kill our kthreads, but don't destroy
the mutex pool and uma zone resources since the driver shutdown
routine may need them later.
problem with using taskqueue_swi is that some of the things we defer
into threads might block for up to several seconds. This is an unfriendly
thing to do to taskqueue_swi, since it is assumed the taskqueue threads
will execute fairly quickly once a task is submitted. Reorganized the
locking in if_ndis.c in the process.
Cleaned up ndis_write_cfg() and ndis_decode_parm() a little.
According to the Windows DDK header files, KSPIN_LOCK is defined like this:
typedef ULONG_PTR KSPIN_LOCK;
From basetsd.h (SDK, Feb. 2003):
typedef [public] unsigned __int3264 ULONG_PTR, *PULONG_PTR;
typedef unsigned __int64 ULONG_PTR, *PULONG_PTR;
typedef _W64 unsigned long ULONG_PTR, *PULONG_PTR;
The keyword __int3264 specifies an integral type that has the following
properties:
+ It is 32-bit on 32-bit platforms
+ It is 64-bit on 64-bit platforms
+ It is 32-bit on the wire for backward compatibility.
It gets truncated on the sending side and extended appropriately
(signed or unsigned) on the receiving side.
Thus register_t seems the proper mapping onto FreeBSD for spin locks.
the definitions for NDIS_BUS_SPACE_IO and NDIS_BUS_SPACE_MEM logically
belong in hal_var.h. At least, that's my story, and I'm sticking to it.
Also, remove definition of __stdcall from if_ndis.c now that it's pulled
in from pe_var.h.
map ranges that are smaller than what our resource manager code knows
is available, rather than requiring that they match exactly. This
fixes a problem with the Intel PRO/1000 gigE driver: it wants to map
a range of 32 I/O ports, even though some chips appear set up to
decode a range of 64. With this fix, it loads and runs correctly.
unexpected interrupts. If an interrupt is triggered and we're not
finished initializing yet, bail. If we have finished initializing,
but IFF_UP isn't set yet, drain the interrupt with ndis_intr() or
ndis_disable_intr() as appropriate, then return _without_ scheduling
ndis_intrtask().
In kern_ndis.c:ndis_load_driver() only relocate/dynalink a given driver
image once. Trying to relocate an image that's already been relocated
will trash the image. We poison a part of the image header that we
don't otherwise need with a magic value to indicate it's already been
fixed up. This fixes the case where there are multiple units of the
same kind of device.
these add support for listing BSSIDs via wicontrol -l. I added code
to call OID_802_11_BSSID_LIST_SCAN to allow scanning for any nearby
wirelsss nets.
Convert from using individual mutexes to a mutex pool, created in
subr_ndis.c. This deals with the problem of drivers creating locks
in their DriverEntry() routines which might get trashed later.
Put some messages under IFF_DEBUG.
which has two important flags in it: the 'allocated by NDIS' flag
and the 'media specific info present' flag. There are two Windows macros
for getting/setting media specific info fields within the ndis_packet
structure which can behave improperly if these flags are not initialized
correctly when a packet is allocated. It seems the correct thing
to do is always set the NDIS_PACKET_ALLOCATED_BY_NDIS flag on
all newly allocated packets.
This fixes the crashes with the Intel Centrino wireless driver.
My sample card now seems to work correctly.
Also, fix a potential LOR involving ndis_txeof() in if_ndis.c.
By default, we search for files in /compat/ndis. This can be changed with
a systcl. These routines are used by some drivers which need to download
firmware or microcode into their respective devices during initialization.
Also, remove extraneous newlines from the 'built-in' sysctl/registry
variables.
the ni_dpccountlock member is an ndis_kspin_lock, not an
ndis_spin_lock (the latter is too big).
Run if_ndis.c:ndis_tick() via taskqueue_schedule(). Also run
ndis_start() via taskqueue in certain circumstances.
Using these tweaks, I can now get the Broadcom BCM5701 NDIS
driver to load and run. Unfortunately, the version I have seems
to suffer from the same bug as the SMC 83820 driver, which is
that it creates a spinlock during its DriverEntry() routine.
I'm still debating the right way to deal with this.
held. However, if we need to translate a unicode message table message,
ndis_unicode_to_ascii() might malloc() some memory, which causes
a warning from witness. Avoid this by using some stack space to hold
the translated message. (Also bounds check to make sure we don't
overrun the stack buffer.)
in subr_ndis and subr_ntoskrnl. This is faster and avoids potential
LOR whinage from witness (an LOR couldn't happen with the old code
since the interlocked inc/dec routines could not sleep with a lock
held, but this will keep witness happy and it's more efficient
anyway. I think.)
so we increment the right thing. (All work and not enough parens
make Bill something something...) This makes the RealTek 8139C+
driver work correctly.
Also fix some mtx_lock_spin()s and mtx_unlock_spin()s that should
have been just plain mtx_lock()s and mtx_unlock()s.
In kern_ndis.c: remove duplicate code from ndis_send_packets() and
just call the senddone handler (ndis_txeof()).
flag so that it can see if the message string is unicode or not and
do the conversion itself rather than doing it in subr_pe.c. This
prevents subr_pe.c from being dependent on subr_ndis.c.
the RT_MESSAGETABLE resources that some driver binaries have.
This allows us to print error messages in ndis_syslog().
- Correct the implementation of InterlockedIncrement() and
InterlockedDecrement() -- they return uint32_t, not void.
- Correct the declarations of the 64-bit arithmetic shift
routines in subr_ntoskrnl.c (_allshr, allshl, etc...). These
do not follow the _stdcall convention: instead, they appear
to be __attribute__((regparm(3)).
- Change the implementation of KeInitializeSpinLock(). There is
no complementary KeFreeSpinLock() function, so creating a new
mutex on each call to KeInitializeSpinLock() leaks resources
when a driver is unloaded. For now, KeInitializeSpinLock()
returns a handle to the ntoskrnl interlock mutex.
- Use a driver's MiniportDisableInterrupt() and MiniportEnableInterrupt()
routines if they exist. I'm not sure if I'm doing this right
yet, but at the very least this shouldn't break any currently
working drivers, and it makes the Intel PRO/1000 driver work.
- In ndis_register_intr(), save some state that might be needed
later, and save a pointer to the driver's interrupt structure
in the ndis_miniport_block.
- Save a pointer to the driver image for use by ndis_syslog()
when it calls pe_get_message().
and MiniportHandleInterrupt() is fired off later via a task queue in
ndis_intrtask(). This more accurately follows the NDIS interrupt handling
model, where the ISR does a minimal amount of work in interrupt context
and the handler is defered and run at a lower priority.
Create a separate ndis_intrmtx mutex just for the guarding the ISR.
Modify NdisSynchronizeWithInterrupt() to aquire the ndis_intrmtx
mutex before invoking the synchronized procedure. (The purpose of
this function is to provide mutual exclusion for code that shares
variables with the ISR.)
Modify NdisMRegisterInterrupt() to save a pointer to the miniport
block in the ndis_miniport_interrupt structure so that
NdisSynchronizeWithInterrupt() can grab it later and derive
ndis_intrmtx from it.
driver was compiled with.
Remove debug printf from ndis_assicn_pcirsc(). It doesn't serve
much purpose.
Implement NdisMIndicateStatus() and NdisMIndicateStatusComplete()
as functions in subr_ndis.c. In NDIS 4.0, they were functions. In
NDIS 5.0 and later, they're just macros.
Allocate a few extra packets/buffers beyond what the driver asks
for since sometimes it seems they can lie about how many they really
need, and some extra stupid ones don't check to see if NdisAllocatePacket()
and/or NdisAllocateBuffer() actually succeed.
calling the haltfunc. If an interrupt is triggered by the init
or halt func, the IFF_UP flag must be set in order for us to be able
to service it.
In kern_ndis.c: implement a handler for NdisMSendResourcesAvailable()
(currently does nothing since we don't really need it).
In subr_ndis.c:
- Correct ndis_init_string() and ndis_unicode_to_ansi(),
which were both horribly broken.
- Implement NdisImmediateReadPciSlotInformation() and
NdisImmediateWritePciSlotInformation().
- Implement NdisBufferLength().
- Work around my first confirmed NDIS driver bug.
The SMC 9462 gigE driver (natsemi 83820-based copper)
incorrectly creates a spinlock in its DriverEntry()
routine and then destroys it in its MiniportHalt()
handler. This is wrong: spinlocks should be created
in MiniportInit(). In a Windows environment, this is
often not a problem because DriverEntry()/MiniportInit()
are called once when the system boots and MiniportHalt()
or the shutdown handler is called when the system halts.
With this stuff in place, this driver now seems to work:
ndis0: <SMC EZ Card 1000> port 0xe000-0xe0ff mem 0xda000000-0xda000fff irq 10 at device 9.0 on pci0
ndis0: assign PCI resources...
ndis_open_file("FLASH9.hex", 18446744073709551615)
ndis0: Ethernet address: 00:04:e2:0e:d3:f0
subr_ndis.c: implement NdisDprAllocatePacket() and NdisDprFreePacket()
(which are aliased to NdisAllocatePacket() and NdisFreePacket()), and
bump the value we return in ndis_mapreg_cnt() to something ridiculously
large, since some drivers apparently expect to be able to allocate
way more than just 64.
These changes allow the Level 1 1000baseSX driver to work for
the following card:
ndis0: <SMC TigerCard 1000 Adapter> port 0xe000-0xe0ff mem 0xda004000-0xda0043ff irq 10 at device 9.0 on pci0
ndis0: Ethernet address: 00:e0:29:6f:cc:04
This is already supported by the lge(4) driver, but I decided
to take a try at making the Windows driver that came with it work too,
since I still had the floppy diskette for it lying around.
the NTx86 section decoration).
subr_ndis.c: correct the behavior of ndis_query_resources(): if the
caller doesn't provide enough space to return the resources, tell it
how much it needs to provide and return an error.
subr_hal.c & subr_ntoskrnl.c: implement/stub a bunch of new routines;
ntoskrnl:
KefAcquireSpinLockAtDpcLevel
KefReleaseSpinLockFromDpcLevel
MmMapLockedPages
InterlockedDecrement
InterlockedIncrement
IoFreeMdl
KeInitializeSpinLock
HAL:
KfReleaseSpinLock
KeGetCurrentIrql
KfAcquireSpinLock
Lastly, correct spelling of "_aullshr" in the ntoskrnl functable.
copyrights to the inf parser files.
Add a -n flag to ndiscvt to allow the user to override the default
device name of NDIS devices. Instead of "ndis0, ndis1, etc..."
you can have "foo0, foo1, etc..." This allows you to have more than
one kind of NDIS device in the kernel at the same time.
Convert from printf() to device_printf() in if_ndis.c, kern_ndis.c
and subr_ndis.c.
Create UMA zones for ndis_packet and ndis_buffer structs allocated
on transmit. The zones are created and destroyed in the modevent
handler in kern_ndis.c.
printf() and UMA changes submitted by green@freebsd.org
peter and jhb: use __volatile__ to prevent gcc from possibly reordering
code, use a null inline instruction instead of a no-op movl (I would
have done this myself if I knew it was allowed) and combine two register
assignments into a single asm statement.
- if_ndis.c: set the NDIS_STATUS_PENDING flag on all outgoing packets
in ndis_start(), make the resource allocation code a little smarter
about how it selects the altmem range, correct a lock order reversal
in ndis_tick().
ndis_var.h
- In kern_ndis.c:ndis_send_packets(), avoid dereferencing NULL pointers
created when the driver's send routine immediately calls the txeof
handler (which releases the packets for us anyway).
- In if_ndis.c:ndis_80211_setstate(), implement WEP support.
method with something a little more intelligent: use BUS_GET_RESOURCE_LIST()
to run through all resources allocated to us and map them as needed. This
way we know exactly what resources need to be mapped and what their RIDs
are without having to guess. This simplifies both ndis_attach() and
ndis_convert_res(), and eliminates the unfriendly "ndisX: couldn't map
<foo>" messages that are sometimes emitted during driver load.
nb_size field in an ndis_buffer is meant to represent, but it does not
represent the original allocation size, so the sanity check doesn't
make any sense now that we're using the Windows-mandated initialization
method.
Among other things, this makes the following card work with the
NDISulator:
ndis0: <NETGEAR PA301 Phoneline10X PCI Adapter> mem 0xda004000-0xda004fff irq 10 at device 9.0 on pci0
This is that notoriously undocumented 10Mbps HomePNA Broadcom chipset
that people wanted support for many moons ago. Sadly, the only other
HomePNA NIC I have handy is a 1Mbps device, so I can't actually do
any 10Mbps performance tests, but it talks to my 1Mbps ADMtek card
just fine.
For received packets, an status of NDIS_STATUS_RESOURCES means we need
to copy the packet data and return the ndis_packet to the driver immediatel.
NDIS_STATUS_SUCCESS means we get to hold onto the packet, but we have
to set the status to NDIS_STATUS_PENDING so the driver knows we're
going to hang onto it for a while.
For transmit packets, NDIS_STATUS_PENDING means the driver will
asynchronously return the packet to us via the ndis_txeof() routine,
and NDIS_STATUS_SUCCESS means the driver sent the frame, and NDIS
(i.e. the OS) retains ownership of the packet and can free it
right away.
evaluate them. Whatever they're meant to do, they're doing it wrong.
Also:
- Clean up last bits of NULL fallout in subr_pe
- Don't let ndis_ifmedia_sts() do anything if the IFF_UP flag isn't set
- Implement NdisSystemProcessorCount() and NdisQueryMapRegisterCount().
packet being freed has NDIS_STATUS_PENDING in the status field of
the OOB data. Finish implementing the "alternative" packet-releasing
function so it doesn't crash.
For those that are curious about ndis0: <ORiNOCO 802.11abg ComboCard Gold>:
1123 packets transmitted, 1120 packets received, 0% packet loss
round-trip min/avg/max/stddev = 3.837/6.146/13.919/1.925 ms
Not bad!
mbuf<->packet housekeeping. Instead, add a couple of extra fields
to the end of ndis_packet. These should be invisible to the Windows
driver module.
This also lets me get rid of a little bit of evil from ndis_ptom()
(frobbing of the ext_buf field instead of relying on the MEXTADD()
macro).
- Fix ndis_time().
- Implement NdisGetSystemUpTime().
- Implement RtlCopyUnicodeString() and RtlUnicodeStringToAnsiString().
- In ndis_getstate_80211(), use sc->ndis_link to determine connect
status.
Submitted by: Brian Feldman <green@freebsd.org>
