eliminates some performance degradation seen in some workloads when
the change was made from logical APIC addresses to physical APIC addresses.
Approved by: re (scottl)
associated changes. More details below:
Remove public declarations of variables that were forgotten when they were
made static.
Revision Changes Path
1.31 +0 -1 src/sys/sys/interrupt.h
Make sure the interrupt is masked before processing it, or bad things
can happen.
Revision Changes Path
1.10 +3 -3 src/sys/arm/arm/intr.c
Reorganize the interrupt handling code a bit to make a few things cleaner
and increase flexibility to allow various different approaches to be tried
in the future.
- Split struct ithd up into two pieces. struct intr_event holds the list
of interrupt handlers associated with interrupt sources.
struct intr_thread contains the data relative to an interrupt thread.
Currently we still provide a 1:1 relationship of events to threads
with the exception that events only have an associated thread if there
is at least one threaded interrupt handler attached to the event. This
means that on x86 we no longer have 4 bazillion interrupt threads with
no handlers. It also means that interrupt events with only INTR_FAST
handlers no longer have an associated thread either.
- Renamed struct intrhand to struct intr_handler to follow the struct
intr_foo naming convention. This did require renaming the powerpc
MD struct intr_handler to struct ppc_intr_handler.
- INTR_FAST no longer implies INTR_EXCL on all architectures except for
powerpc. This means that multiple INTR_FAST handlers can attach to the
same interrupt and that INTR_FAST and non-INTR_FAST handlers can attach
to the same interrupt. Sharing INTR_FAST handlers may not always be
desirable, but having sio(4) and uhci(4) fight over an IRQ isn't fun
either. Drivers can always still use INTR_EXCL to ask for an interrupt
exclusively. The way this sharing works is that when an interrupt
comes in, all the INTR_FAST handlers are executed first, and if any
threaded handlers exist, the interrupt thread is scheduled afterwards.
This type of layout also makes it possible to investigate using interrupt
filters ala OS X where the filter determines whether or not its companion
threaded handler should run.
- Aside from the INTR_FAST changes above, the impact on MD interrupt code
is mostly just 's/ithread/intr_event/'.
- A new MI ddb command 'show intrs' walks the list of interrupt events
dumping their state. It also has a '/v' verbose switch which dumps
info about all of the handlers attached to each event.
- We currently don't destroy an interrupt thread when the last threaded
handler is removed because it would suck for things like ppbus(8)'s
braindead behavior. The code is present, though, it is just under
#if 0 for now.
- Move the code to actually execute the threaded handlers for an interrrupt
event into a separate function so that ithread_loop() becomes more
readable. Previously this code was all in the middle of ithread_loop()
and indented halfway across the screen.
- Made struct intr_thread private to kern_intr.c and replaced td_ithd
with a thread private flag TDP_ITHREAD.
- In statclock, check curthread against idlethread directly rather than
curthread's proc against idlethread's proc. (Not really related to intr
changes)
Tested on: alpha, amd64, i386, sparc64
Tested on: arm, ia64 (older version of patch by cognet and marcel)
Revision Changes Path
1.88 +43 -29 src/sys/alpha/alpha/interrupt.c
1.38 +5 -5 src/sys/alpha/isa/isa.c
1.16 +58 -52 src/sys/amd64/amd64/intr_machdep.c
1.6 +1 -1 src/sys/amd64/include/intr_machdep.h
1.16 +2 -2 src/sys/amd64/isa/atpic.c
1.11 +28 -22 src/sys/arm/arm/intr.c
1.462 +2 -2 src/sys/dev/sio/sio.c
1.6 +1 -1 src/sys/dev/uart/uart_kbd_sun.c
1.24 +2 -2 src/sys/dev/uart/uart_tty.c
1.15 +58 -52 src/sys/i386/i386/intr_machdep.c
1.8 +1 -1 src/sys/i386/include/intr_machdep.h
1.21 +2 -2 src/sys/i386/isa/atpic.c
1.52 +32 -25 src/sys/ia64/ia64/interrupt.c
1.180 +3 -2 src/sys/kern/kern_clock.c
1.127 +437 -270 src/sys/kern/kern_intr.c
1.206 +0 -1 src/sys/kern/subr_witness.c
1.6 +3 -3 src/sys/powerpc/include/intr_machdep.h
1.7 +35 -32 src/sys/powerpc/powerpc/intr_machdep.c
1.14 +1 -1 src/sys/sparc64/include/intr_machdep.h
1.24 +43 -36 src/sys/sparc64/sparc64/intr_machdep.c
1.32 +36 -36 src/sys/sys/interrupt.h
1.440 +1 -3 src/sys/sys/proc.h
Catch up with interrupt-thread changes.
Revision Changes Path
1.32 +1 -1 src/sys/dev/zs/zs.c
Catch up with new interrupt handling code.
Revision Changes Path
1.16 +3 -3 src/sys/netgraph/bluetooth/drivers/bt3c/ng_bt3c_pccard.c
Catch up with new interrupt handling code.
Revision Changes Path
1.162 +2 -2 src/sys/dev/cy/cy.c
1.101 +2 -2 src/sys/dev/rc/rc.c
Catch up with new interrupt handling code.
Revision Changes Path
1.50 +2 -2 src/sys/dev/cx/if_cx.c
1.41 +1 -1 src/sys/dev/sab/sab.c
1.238 +2 -2 src/sys/pc98/cbus/sio.c
Add a swi_remove() function to teardown software interrupt handlers. For
now it just calls intr_event_remove_handler(), but at some point it might
also be responsible for tearing down interrupt events created via swi_add.
Revision Changes Path
1.128 +17 -0 src/sys/kern/kern_intr.c
1.33 +1 -0 src/sys/sys/interrupt.h
- Use swi_remove() to teardown swi handlers rather than
intr_event_remove_handler().
- Remove tty: prefix from a couple of swi handler names.
Revision Changes Path
1.51 +1 -1 src/sys/dev/cx/if_cx.c
1.102 +2 -2 src/sys/dev/rc/rc.c
1.42 +1 -1 src/sys/dev/sab/sab.c
1.25 +1 -1 src/sys/dev/uart/uart_tty.c
1.33 +1 -1 src/sys/dev/zs/zs.c
1.17 +2 -2 src/sys/netgraph/bluetooth/drivers/bt3c/ng_bt3c_pccard.c
Remove a stray return statement in the interrupt dispatch function
that caused a premature exit after calling a fast interrupt handler
and bypassing a much needed critical_exit() and the scheduling of
the interrupt thread for non-fast handlers. In short: unbreak :-)
Revision Changes Path
1.53 +0 -1 src/sys/ia64/ia64/interrupt.c
If we get a stray interrupt, return after logging it. In the extremely
rare case of a stray interrupt to an unregistered source (such as a stray
interrupt from the 8259As when using APIC), this could result in a page
fault when it tried to walk the list of interrupt handlers to execute
INTR_FAST handlers. This bug was introduced with the intr_event changes,
so it's not present in 5.x or 6.x.
Submitted by: Mark Tinguely tinguely at casselton dot net
Revision Changes Path
1.17 +1 -0 src/sys/amd64/amd64/intr_machdep.c
1.16 +1 -0 src/sys/i386/i386/intr_machdep.c
Approved by: re (scottl)
osigpending, osigvec, osigblock, osigsetmask, osigsuspend, osigstack,
clock_gettime, clock_settime, and clock_getres.
Also correct the prototype for freebsd32_nanosleep in syscalls.master.
- i915 (i830-i915) driver ported.
- S3 Savage driver ported.
- Added support for ATI_fragment_shader registers for r200.
- Improved r300 support, needed for latest r300 DRI driver.
- (possibly) r300 PCIE support, needs X.Org server from CVS.
- Added support for PCI Matrox cards.
- Software fallbacks fixed for Rage 128, which used to render badly or hang.
- Some issues reported by WITNESS are fixed.
Begin and end the initialization of pvzone in pmap_init().
Previously, pvzone's initialization was split between pmap_init() and
pmap_init2(). This split initialization was the underlying cause of
some UMA panics during initialization. Specifically, if the UMA boot
pages was exhausted before the pvzone was fully initialized, then UMA,
through no fault of its own, would use an inappropriate back-end
allocator leading to a panic. (Previously, as a workaround, we have
increased the UMA boot pages.) Fortunately, there is no longer any
reason that pvzone's initialization cannot be completed in
pmap_init().
Eliminate a check for whether pv_entry_high_water has been initialized
or not from get_pv_entry(). Since pvzone's initialization is
completed in pmap_init(), this check is no longer needed.
Use cnt.v_page_count, the actual count of available physical pages,
instead of vm_page_array_size to compute the maximum number of pv
entries.
Introduce the vm.pmap.pv_entries tunable on alpha and ia64.
Eliminate some unnecessary white space.
When support for 2MB/4MB pages was added in revision 1.148 an error was
made in pmap_protect(): The pmap's resident count should not be reduced
unless mappings are removed.
The errant change to the pmap's resident count could result in a later
pmap_remove() failing to remove any mappings if the errant change has set
the pmap's resident count to zero.
MFC revision 1.518
Decouple the unrefing of a page table page from the removal of a pv entry.
In other words, change pmap_remove_entry() such that it no longer unrefs
the page table page. Now, it only removes the pv entry.
MFC revision 1.519
Eliminate unneeded diagnostic code.
MFC revision 1.520
Eliminate unneeded diagnostic code.
Eliminate an unused #include. (Kernel stack allocation and deallocation
long ago migrated to the machine-independent code.)
MFC revision 1.521
Simplify the page table page reference counting by pmap_enter()'s change of
mapping case.
Eliminate a stale comment from pmap_enter().
MFC revision 1.522
Correct a performance bug in revision 1.462. The effect of the bug is to
execute the outer loop in procedures such as pmap_protect() many more times
than necessary.
MFC revision 1.523
Introduce pmap_pml4e_to_pdpe() and pmap_pdpe_to_pde() and use them to avoid
recomputation of the pml4e and pdpe in pmap_copy(), pmap_protect(), and
pmap_remove().
MFC revision 1.524
Change pmap_extract() and pmap_extract_and_hold() to use PG_FRAME rather
than ~PDRMASK to extract the physical address of a superpage from a PDE.
The use of ~PDRMASK is problematic if the PDE has PG_NX set. Specifically,
the PG_NX bit will be included in the physical address if ~PDRMASK is used.
MFC revision 1.525
Pass the PDE from pmap_remove() to pmap_remove_page() so that the latter
procedure doesn't have to recompute it.
MFC revision 1.526
Remedy the following three problems:
1. The amd64 pmap, unlike the i386 pmap, maintains a reference count
for each page directory (PD) page. However, in the transformation
of the i386 pmap into the amd64 pmap, operations, such as
pmap_copy() and pmap_object_init_pt(), that create 2MB "superpage"
mappings by setting the PG_PS bit in a PD entry were not modified
to adjust the underlying PD page's reference count. Consequently,
superpage mappings could disappear prematurely.
2. pmap_object_init_pt() could crash or corrupt memory if either the
virtual address range being mapped crosses a 1GB boundary in the
virtual address space or nothing is mapped in the 1GB area.
3. When pmap_allocpte() destroys a 2MB "superpage" mapping it does not
reduce the pmap's resident count accordingly. It should. (This
bug is inherited from i386.)
MFC revision 1.528
Eliminate unnecessary TLB invalidations by pmap_enter(). Specifically,
eliminate TLB invalidations when permissions are relaxed, such as when a
read-only mapping is changed to a read/write mapping. Additionally,
eliminate TLB invalidations when bits that are ignored by the hardware,
such as PG_W ("wired mapping"), are changed.
than 8 entries in the 0xe821 map. This includes things like ACPI
exclusion zones, memory segments and recoverable memory segments after
the kernel has booted.
Specifically panic() in the case where pmap_insert_entry() fails to
get a new pv under high system load where the available pv entries
have been exhausted before the pagedaemon has a chance to wake up
to reclaim some.
Approved by: re (scottl)
The major change here is the support for amd64, as well as possible
support for PAE. Many thanks to Highpoint for continuing to support
FreeBSD.
Obtained from: Steve Chang @ Highpoint
Merged revisions:
| 1.3 dev/hptmv/access601.h
| 1.1 dev/hptmv/amd64-elf.raid.o.uu
| 1.3 dev/hptmv/array.h
| 1.3 dev/hptmv/atapi.h
| 1.3 dev/hptmv/command.h
| 1.9+1.10 dev/hptmv/entry.c
| 1.4 dev/hptmv/global.h
| 1.5 dev/hptmv/gui_lib.c
| 1.3 dev/hptmv/hptintf.h
| 1.6 dev/hptmv/hptproc.c
| 1.5 dev/hptmv/i386-elf.raid.o.uu
| 1.5 dev/hptmv/ioctl.c
| 1.3 dev/hptmv/mv.c
| 1.4 dev/hptmv/mvOs.h
| 1.3 dev/hptmv/mvSata.h
| 1.4 dev/hptmv/mvStorageDev.h
| 1.5 dev/hptmv/osbsd.h
| 1.3 dev/hptmv/raid5n.h
| 1.3 dev/hptmv/readme.txt
| 1.3 dev/hptmv/vdevice.h
| 1.4 modules/hptmv/Makefile
Approved by: re (scottl)
- Special-case NMI handling on the AMD64.
On entry or exit from the kernel the 'alltraps' and 'doreti' the code
used by normal traps disables interrupts to protect the critical sections
where it is setting up %gs.
This protection is insufficient in the presence of NMIs since NMIs
can be taken even when the processor has disabled normal interrupts.
Thus the NMI handler needs to actually read MSR_GBASE on entry to
the kernel to determine whether a swap of %gs using 'swapgs' is
needed. However, reads of MSRs are expensive and integrating this
check into the 'alltraps'/'doreti' path would penalize normal
interrupts.
- Teach DDB about the 'nmi_calltrap' symbol.
Reviewed by: bde, peter (an older version) and ups
Approved by: re (scottl)
