29 Jul, 2008
18 commits
-
Number of exits transmitting 10GB Guest->Host before:
network xmit 7858610 recv 118136After:
network xmit 7750233 recv 1Signed-off-by: Rusty Russell
-
To simplify the transition to when we publish indices in the ring
(and make shuffling my patch queue easier), wrap them in a lg_last_avail()
macro.Signed-off-by: Rusty Russell
-
To support my little make-x86-bitops-use-proper-typechecking projectlet.
Cc: Thomas Gleixner
Cc: Andrea Arcangeli
Signed-off-by: Andrew Morton
Acked-by: Ingo Molnar
Signed-off-by: Rusty Russell -
This is a simple patch to add support for the virtio "hardware random
generator" to lguest. It gets about 1.2 MB/sec reading from /dev/hwrng
in the guest.Signed-off-by: Rusty Russell
-
If you've got a nice DHCP configuration which maps MAC
addresses to specific IP addresses, then you're going to
want to start your guest with one of those MAC addresses.Also, in Fedora, we have persistent network interface naming
based on the MAC address, so with randomly assigned
addresses you're soon going to hit eth13. Who knows what
will happen then!Allow assigning a MAC address to the network interface with
e.g.--tunnet=bridge:eth0:00:FF:95:6B:DA:3D
or:
--tunnet=192.168.121.1:00:FF:95:6B:DA:3D
which is pretty unintelligable, but ...
(includes Rusty's minor rework)
Signed-off-by: Mark McLoughlin
Signed-off-by: Rusty Russell -
Signed-off-by: Mark McLoughlin
Signed-off-by: Rusty Russell -
%02x is more appropriate for bytes than %08x.
Signed-off-by: Rusty Russell
-
map_switcher allocates the array, unmap_switcher has to free it
accordingly.Signed-off-by: Johannes Weiner
Signed-off-by: Rusty Russell -
Ron Minnich noticed that guest userspace gets a GPF when it tries to int3:
we need to copy the privilege level from the guest-supplied IDT to the real
IDT. int3 is the only common case where guest userspace expects to invoke
an interrupt, so that's the symptom of failing to do this.Signed-off-by: Rusty Russell
-
6af61a7614a306fe882a0c2b4ddc63b65aa66efc 'x86: clean up max_pfn_mapped
usage - 32-bit' makes the following comment:XEN PV and lguest may need to assign max_pfn_mapped too.
But no CC. Yinghai, wasting fellow developers' time is a VERY bad
habit. If you do it again, I will hunt you down and try to extract
the three hours of my life I just lost :)Signed-off-by: Rusty Russell
Cc: Yinghai Lu -
* 'cpus4096-v2' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
cpu masks: optimize and clean up cpumask_of_cpu()
cpumask: export cpumask_of_cpu_map
cpumask: change cpumask_of_cpu_ptr to use new cpumask_of_cpu
cpumask: put cpumask_of_cpu_map in the initdata section
cpumask: make cpumask_of_cpu_map generic -
There's a brown paper bag compile failure introduced by this patch
commit a01386924874c4d6d67f8a34e66f04452c2abb69
Author: David Sterba
Date: Mon Jul 28 16:53:32 2008 +0200ipwireless: Preallocate received packet buffers with MRU size
Really, it can't ever have been even compile tested. It looks like the
closing bracket is in the wrong place, so this is the fix.Signed-off-by: James Bottomley
Signed-off-by: Linus Torvalds -
Conflicts:
kernel/stop_machine.c
Signed-off-by: Ingo Molnar
-
Clean up and optimize cpumask_of_cpu(), by sharing all the zero words.
Instead of stupidly generating all possible i=0...NR_CPUS 2^i patterns
creating a huge array of constant bitmasks, realize that the zero words
can be shared.In other words, on a 64-bit architecture, we only ever need 64 of these
arrays - with a different bit set in one single world (with enough zero
words around it so that we can create any bitmask by just offsetting in
that big array). And then we just put enough zeroes around it that we
can point every single cpumask to be one of those things.So when we have 4k CPU's, instead of having 4k arrays (of 4k bits each,
with one bit set in each array - 2MB memory total), we have exactly 64
arrays instead, each 8k bits in size (64kB total).And then we just point cpumask(n) to the right position (which we can
calculate dynamically). Once we have the right arrays, getting
"cpumask(n)" ends up being:static inline const cpumask_t *get_cpu_mask(unsigned int cpu)
{
const unsigned long *p = cpu_bit_bitmap[1 + cpu % BITS_PER_LONG];
p -= cpu / BITS_PER_LONG;
return (const cpumask_t *)p;
}This brings other advantages and simplifications as well:
- we are not wasting memory that is just filled with a single bit in
various different places- we don't need all those games to re-create the arrays in some dense
format, because they're already going to be dense enough.if we compile a kernel for up to 4k CPU's, "wasting" that 64kB of memory
is a non-issue (especially since by doing this "overlapping" trick we
probably get better cache behaviour anyway).[ mingo@elte.hu:
Converted Linus's mails into a commit. See:
http://lkml.org/lkml/2008/7/27/156
http://lkml.org/lkml/2008/7/28/320Also applied a family filter - which also has the side-effect of leaving
out the bits where Linus calls me an idio... Oh, never mind ;-)
]Signed-off-by: Ingo Molnar
Cc: Rusty Russell
Cc: Andrew Morton
Cc: Al Viro
Cc: Mike Travis
Signed-off-by: Ingo Molnar -
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dtor/input:
Input: add driver for Atmel integrated touchscreen controller
Input: ads7846 - optimize order of calculating Rt in ads7846_rx()
Input: ads7846 - fix sparse endian warnings
Input: uinput - remove duplicate include
Input: serio - offload resume to kseriod
Input: serio - mark serio_register_driver() __must_check -
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/teigland/dlm:
dlm: fix uninitialized variable for search_rsb_list callers
dlm: release socket on error
dlm: fix basts for granted CW waiting PR/CW
dlm: check for null in device_write -
* 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc: (25 commits)
powerpc: Disable 64K hugetlb support when doing 64K SPU mappings
powerpc/powermac: Fixup default serial port device for pmac_zilog
powerpc/powermac: Use sane default baudrate for SCC debugging
powerpc/mm: Implement _PAGE_SPECIAL & pte_special() for 64-bit
powerpc: Show processor cache information in sysfs
powerpc: Make core id information available to userspace
powerpc: Make core sibling information available to userspace
powerpc/vio: More fallout from dma_mapping_error API change
ibmveth: Fix multiple errors with dma_mapping_error conversion
powerpc/pseries: Fix CMO sysdev attribute API change fallout
powerpc: Enable tracehook for the architecture
powerpc: Add TIF_NOTIFY_RESUME support for tracehook
powerpc: Add asm/syscall.h with the tracehook entry points
powerpc: Make syscall tracing use tracehook.h helpers
powerpc: Call tracehook_signal_handler() when setting up signal frames
powerpc: Update cpu_sibling_maps dynamically
powerpc: register_cpu_online should be __cpuinit
powerpc: kill useless SMT code in prom_hold_cpus
powerpc: Fix 8xx build failure
powerpc: Fix vio build warnings
...
28 Jul, 2008
22 commits
-
* git://git.kernel.org/pub/scm/linux/kernel/git/kkeil/ISDN-2.6:
Remove deprecated virt_to_bus() -
* git://git.kernel.org/pub/scm/linux/kernel/git/lethal/sh-2.6: (72 commits)
sh: SuperH Mobile CEU and camera platform data for AP325RXA
sh: Update smc911x platform data for AP325RXA
sh: SuperH Mobile LCDC platform data for AP325RXA
sh: Add SuperH Mobile CEU platform data for Migo-R
sh: Add SuperH Mobile LCDC platform data for Migo-R
sh: Move asid_cache() out of ifdef to fix SH-3/4 nommu build.
sh: Workaround for __put_user_asm() bug with gcc 4.x on big-endian.
sh: Wire up new syscalls.
sh: fix uImage Entry Point
sh_keysc: remove request_mem_region() and release_mem_region()
sh: Don't miss pending signals returning to user mode after signal processing
sh: Use clk_always_enable() on sh7366
sh: Use clk_always_enable() on sh7343 / SE77343
sh: Use clk_always_enable() on sh7722 / Migo-R / SE7722
sh: Use clk_always_enable() on sh7723 / ap325rxa
sh: Introduce clk_always_enable() function
sh: Show all clocks and their state in /proc/clocks
sh: Merge sh7343 and sh7722 clock code
sh: Add SuperH Mobile MSTPCR bits to clock framework
sh: Use arch_flags to simplify sh7722 siu clock code
... -
* git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-for-linus:
stop_machine: fix up ftrace.c
stop_machine: Wean existing callers off stop_machine_run()
stop_machine(): stop_machine_run() changed to use cpu mask
Hotplug CPU: don't check cpu_online after take_cpu_down
Simplify stop_machine
stop_machine: add ALL_CPUS option
module: fix build warning with !CONFIG_KALLSYMS -
* git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/v4l-dvb: (98 commits)
V4L/DVB (8549): mxl5007: Fix an error at include file
V4L/DVB (8548): pwc: Fix compilation
V4L/DVB (8546): add tuner-3036 and dpc7146 drivers to feature-removal-schedule.txt
V4L/DVB (8546): saa7146: fix read from uninitialized memory
V4L/DVB (8544): gspca: probe/open race.
V4L/DVB (8543): em28xx: Rename #define for Compro VideoMate ForYou/Stereo
V4L/DVB (8542): em28xx: AMD ATI TV Wonder HD 600 entry at cards struct is duplicated
V4L/DVB (8541): em28xx: HVR-950 entry is duplicated.
V4L/DVB (8540): em28xx-cards: Add Compro VideoMate ForYou/Stereo model
V4L/DVB (8539): em28xx-cards: New supported IDs for analog models
V4L/DVB (8538): em28xx-cards: Add GrabBeeX+ USB2800 model
V4L/DVB (8534): remove select's of FW_LOADER
V4L/DVB (8522): videodev2: Fix merge conflict
V4L/DVB (8532): mxl5007t: remove excessive locks
V4L/DVB (8531): mxl5007t: move i2c gate handling outside of mutex protected code blocks
V4L/DVB (8530): au0828: add support for new revision of HVR950Q
V4L/DVB (8529): mxl5007t: enable _init and _sleep power management functionality
V4L/DVB (8528): add support for MaxLinear MxL5007T silicon tuner
V4L/DVB (8526): saa7146: fix VIDIOC_ENUM_FMT
V4L/DVB (8525): fix a few assorted spelling mistakes.
... -
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc-2.6:
sparc: Set CONFIG_HAVE_ARCH_TRACEHOOK
sparc: Add task_pt_regs().
sparc: Add call to tracehook_signal_handler().
sparc: Create and use TIF_NOTIFY_RESUME.
sparc: Use tracehook routines in syscall_trace().
sparc64: tracehook: CONFIG_HAVE_ARCH_TRACEHOOK
sparc: Add user_stack_pointer().
sparc64: tracehook_signal_handler
sparc64: tracehook: TIF_NOTIFY_RESUME
sparc: Add asm/syscall.h
sparc64: tracehook syscall
sparc: enable headers_export again
sparc, sparc64: use arch/sparc/include -
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6:
net: missing bits of net-namespace / sysctl
ipcomp: Fix warnings after ipcomp consolidation.
dccp: Add check for truncated ICMPv6 DCCP error packets
dccp: Fix incorrect length check for ICMPv4 packets
dccp: Add check for sequence number in ICMPv6 message
dccp: Fix sequence number check for ICMPv4 packets
dccp: Bug-Fix - AWL was never updated
dccp: Allow to distinguish original and retransmitted packets -
Add git pull command info and diffstat summary info so that we don't
have to search email archives for it repeatedly.Signed-off-by: Randy Dunlap
Signed-off-by: Linus Torvalds -
Noticed because of this warning:
drivers/base/memory.c:279: warning: initialization from incompatible pointer type
Signed-off-by: Stephen Rothwell
Signed-off-by: Linus Torvalds -
pm_idle_save resp. pm_idle_old can be NULL when the restore code in
acpi_processor_cst_has_changed() resp. cpuidle_uninstall_idle_handler()
is called. This can set pm_idle unconditinally to NULL, which causes the
kernel to panic when calling pm_idle in the x86 idle code. This was
covered by an extra check for !pm_idle in the x86 idle code, which was
removed during the x86 idle code refactoring.Instead of restoring the pm_idle check in the x86 code prevent the
acpi/cpuidle code to set pm_idle to NULL.Reported by: Dhaval Giani http://lkml.org/lkml/2008/7/2/309
Based on a debug patch from Ingo MolnarSigned-off-by: Thomas Gleixner
Signed-off-by: Linus Torvalds -
* 'i2c-for-linus' of git://aeryn.fluff.org.uk/bjdooks/linux:
i2c: Blackfin I2C Driver: Functional power management support
i2c: Documentation: upgrading clients HOWTO
i2c: S3C24XX I2C frequency scaling support.
i2c: i2c_gpio: keep probe resident for hotplugged devices.
i2c: S3C2410: Pass the I2C bus number via drivers platform data -
ipwireless: Preallocate received packet buffers with MRU size
Packets are assembled from link size (~300 bytes) up to PPP MRU
(1500 by default). Try to preallocate full size rather than
repeatedly advance buffer size by 256 bytes.Signed-off-by: David Sterba
Signed-off-by: Jiri Kosina
Signed-off-by: Linus Torvalds -
ipwireless: Put packets to pool start
Put packets to pool start, try to reuse cached memory.
Signed-off-by: David Sterba
Signed-off-by: Jiri Kosina
Signed-off-by: Linus Torvalds -
ipwireless: Increase PPP outgoing queue size
Increase default size of PPP outgoing queue. Currently set to 1, which
means that a packet quickly following another pushed by PPP must wait
until hardware actually sends the previous and PPP has to be waken up
by ppp_wakeup(). This slows down upstream.Now PPP can push more packets at once which get buffered inside driver
and pushed immediatelly to hardware when previous packet is out.Experiments show that size = 10 is quite good for all connection types
(GPRS/EDGE/UMTS) and gains 4 KB/sec of upload for UMTS for batch uploads.
Need for higher queue size than 10 occures in only < 0.1 % of cases.Signed-off-by: David Sterba
Signed-off-by: Jiri Kosina
Signed-off-by: Linus Torvalds -
ipwireless: Explicitly request io and mem regions
Documentation/pcmcia/driver-changes.txt says, that driver should call request_region
for used memory/io regions since PCMCIA does not do this (since 2.6.8).Signed-off-by: David Sterba
Signed-off-by: Jiri Kosina
Signed-off-by: Linus Torvalds -
ipwireless: Constify buffer variables
Signed-off-by: David Sterba
Signed-off-by: Jiri Kosina
Signed-off-by: Linus Torvalds -
ipwireless: Do not return value from sending funcs
Do not return value from do_send_fragment and do_send_packet, it's not used.
The packet size checks are not useful too:* zero length packet will never be sent, caller always passes packet_header
size which is either 1 or 3
* MTU check is done in caller, no need to repeatSigned-off-by: David Sterba
Signed-off-by: Jiri Kosina
Signed-off-by: Linus Torvalds -
ipwireless: Remove endian-dependent bitfields
Remove endian-dependent bitfields and use bitmasks to transform
packet header bitfields from/to machine order.Signed-off-by: David Sterba
Signed-off-by: Jiri Kosina
Signed-off-by: Linus Torvalds -
ipwireless: Glue splitted printk strings back
Signed-off-by: David Sterba
Signed-off-by: Jiri Kosina
Signed-off-by: Linus Torvalds -
ipwireless: Remove pt_regs from interrupt handler
Signed-off-by: David Sterba
Signed-off-by: Jiri Kosina
Signed-off-by: Linus Torvalds -
ipwireless: Rename spinlock variables to lock
Signed-off-by: David Sterba
Signed-off-by: Jiri Kosina
Signed-off-by: Linus Torvalds -
ipwireless: Remove unused defines
Remove unused defines, defines hiding variables, defines hiding 0.
Signed-off-by: David Sterba
Signed-off-by: Jiri Kosina
Signed-off-by: Linus Torvalds -
ipwireless: Misc cleanups
- remove likely() and some extra () in ifs
- use unsigned in for loops
- remove useless typecasts
- remove obvious comments
- add () around ?:Signed-off-by: David Sterba
Signed-off-by: Jiri Kosina
Signed-off-by: Linus Torvalds
