xen.git
19 years ago[XEND] Silence Xen API debugging messages.
Alastair Tse [Fri, 3 Nov 2006 10:38:18 +0000 (10:38 +0000)]
[XEND] Silence Xen API debugging messages.

Signed-off-by: Alastair Tse <atse@xensource.com>
19 years ago[XEN] Quieten down tracing when fail (quite expectedly) to map INVALID_MFN.
kfraser@localhost.localdomain [Fri, 3 Nov 2006 10:26:36 +0000 (10:26 +0000)]
[XEN] Quieten down tracing when fail (quite expectedly) to map INVALID_MFN.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years agoI found a small bug in xenmon. Even if I gave xenmon a wrong
Ewan Mellor [Thu, 2 Nov 2006 23:01:32 +0000 (23:01 +0000)]
I found a small bug in xenmon.  Even if I gave xenmon a wrong
parameter, xenmon did not become an error.

This patch adds a checking of wrong parameter into xenmon.
If wrong parameter is given, xenmon shows the following error
messages.

usage: xenmon.py [options]

xenmon.py: error: No parameter required

Signed-off-by: Masaki Kanno <kanno.masaki@jp.fujitsu.com>
19 years ago[XEN] Add 'loglvl' and 'guest_loglvl' boot parameters.
kaf24@localhost.localdomain [Thu, 2 Nov 2006 22:24:20 +0000 (22:24 +0000)]
[XEN] Add 'loglvl' and 'guest_loglvl' boot parameters.

 <lvl> := none|error|warning|info|debug|all

 loglvl=<lvl_print_always>[/<lvl_print_ratelimit>]
  <lvl_print_always>: log level which is always printed
  <lvl_print_rlimit>: log level which is rate-limit printed

 'loglvl' applies to non-guest-related messages.
 'guest_loglvl' applies to guest-related messages.

 Defaults: loglvl=warning ; guest_loglvl=none/warning

Also clean up hvm_print_line().

Signed-off-by: Keir Fraser <keir@xensource.com>
19 years agoMerge
Alastair Tse [Thu, 2 Nov 2006 18:52:04 +0000 (18:52 +0000)]
Merge

19 years ago[XEND] Reverting self.domains UUID indexing
Alastair Tse [Thu, 2 Nov 2006 18:51:23 +0000 (18:51 +0000)]
[XEND] Reverting self.domains UUID indexing

Due to localhost migration, it is possible to have two VMs with the
same UUID on the same host. So instead, we revert back to XendDomain
where there are two lists, a "domains" of active domains indexed by
domid and a "managed_domains" which is a list of all domains whose
lifecycle is managed by Xend.

Signed-off-by: Alastair Tse <atse@xensource.com>
19 years agoDynamically link to libgcc on Solaris. Also clean up duplicate -Wall flags.
kfraser@localhost.localdomain [Thu, 2 Nov 2006 18:19:22 +0000 (18:19 +0000)]
Dynamically link to libgcc on Solaris. Also clean up duplicate -Wall flags.
Signed-off-by: John Levon <john.levon@sun.com>
19 years ago[BLKTAP] Fix Linux and Xen gnttab map operations when mapping into
kfraser@localhost.localdomain [Thu, 2 Nov 2006 18:18:19 +0000 (18:18 +0000)]
[BLKTAP] Fix Linux and Xen gnttab map operations when mapping into
PTEs above 4GB on PAE systems.
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
19 years ago[HVM] Clean up VCPU initialisation in Xen. No longer
kfraser@localhost.localdomain [Thu, 2 Nov 2006 15:55:51 +0000 (15:55 +0000)]
[HVM] Clean up VCPU initialisation in Xen. No longer
parse HVM e820 tables in Xen (add some extra HVM parameters as a
cleaner alternative). Lots of code removal.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[XEND] Fix paused state being overriden by refreshShutdown
Alastair Tse [Thu, 2 Nov 2006 15:48:32 +0000 (15:48 +0000)]
[XEND] Fix paused state being overriden by refreshShutdown

Signed-off-by: Alastair Tse <atse@xensource.com>
19 years agoMerge with tip
Alastair Tse [Thu, 2 Nov 2006 14:27:16 +0000 (14:27 +0000)]
Merge with tip

19 years ago[XEND][XM] Fixes for xm list output
Alastair Tse [Thu, 2 Nov 2006 14:26:01 +0000 (14:26 +0000)]
[XEND][XM] Fixes for xm list output

Do not output image, security if they are empty
Change xm list slightly to be more informative about state.
Add back store_mfn and console_mfn to the SXP so save/restore can work
again.

Signed-off-by: Alastair Tse <atse@xensource.com>
19 years ago[XEN] Better parameter checking for grant table copy operation.
Steven Smith [Thu, 2 Nov 2006 13:20:32 +0000 (13:20 +0000)]
[XEN] Better parameter checking for grant table copy operation.

Signed-off-by: Steven Smith <sos22@cam.ac.uk>
19 years agoPV-on-HVM: Conditionally disable checksum offload for receive on some
Ian Campbell [Thu, 2 Nov 2006 11:56:26 +0000 (11:56 +0000)]
PV-on-HVM: Conditionally disable checksum offload for receive on some
older kernels.

On some older kernels the CHECKSUM_UNNECESSARY flag on received
skbuffs is not honoured by all subsystems, specifically netfilter
under 2.6.9-42.EL

To work around this we need to ensure that the correct checksum is
calculated at some point. We can cause the backend to do this for us
by advertising to the backend that we cannot support checksum offload.

The correlation of this issue with the presence of NETIF_F_TSO is not
100% but it is a good first approximation.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
19 years agoPV-on-HVM: Tweak reboot.c to be compatible with older kernels.
Ian Campbell [Thu, 2 Nov 2006 11:56:26 +0000 (11:56 +0000)]
PV-on-HVM: Tweak reboot.c to be compatible with older kernels.

call_usermodehelper_keys() was not available until 2.6.13 but since we
pass session_keyring as NULL we may as well use call_usermodehelper()
anyway.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
19 years agoPV-on-HVM: Fix PV-on-HVM drivers for IA64.
Ian Campbell [Thu, 2 Nov 2006 11:56:26 +0000 (11:56 +0000)]
PV-on-HVM: Fix PV-on-HVM drivers for IA64.

Thanks to Kasai Takanori and Doi Tsunehisa for testing.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
19 years agoPV-on-HVM: Use ln -nsf in mkbuildtree to avoid creating a symlink loop
Ian Campbell [Thu, 2 Nov 2006 11:56:26 +0000 (11:56 +0000)]
PV-on-HVM: Use ln -nsf in mkbuildtree to avoid creating a symlink loop
if include/xen/interfaces already exists.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
19 years ago[TOOLS] Paravirt guests have their memory allocated in the
kfraser@localhost.localdomain [Thu, 2 Nov 2006 11:11:28 +0000 (11:11 +0000)]
[TOOLS] Paravirt guests have their memory allocated in the
libxenguest builder function.

Again, PPC and IA64 will need to do some fixing up.

Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[XEND] Compare result of blkdev_name_to_number() explicitly against None.
kaf24@firebug.cl.cam.ac.uk [Thu, 2 Nov 2006 07:46:06 +0000 (07:46 +0000)]
[XEND] Compare result of blkdev_name_to_number() explicitly against None.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[XEN] Fix credit scheduler cap mechanism not to over park VCPUs
Emmanuel Ackaouy [Wed, 1 Nov 2006 19:44:34 +0000 (19:44 +0000)]
[XEN] Fix credit scheduler cap mechanism not to over park VCPUs
We used to park a capped VCPU when it had ran its fair share, even
if the fair share was below its cap. With this change, make sure
we only park once a VCPU has overrun its actual cap.
We could also try to make the capping mechanism more accurate by
parking and unparking at a finer granularity (currently done on
30ms boundaries) but that'll be for a different time.
Signed-off-by: Emmanuel Ackaouy <ack@xensource.com>
19 years ago[XEN] Fix x86/64 build.
kfraser@localhost.localdomain [Wed, 1 Nov 2006 18:48:57 +0000 (18:48 +0000)]
[XEN] Fix x86/64 build.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years agoReturn the restored VM at the end of XendDomainInfo.restore.
Ewan Mellor [Wed, 1 Nov 2006 18:39:33 +0000 (18:39 +0000)]
Return the restored VM at the end of XendDomainInfo.restore.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
19 years ago[HVM] Move shadow initialisation into domain-creation hypercall.
kfraser@localhost.localdomain [Wed, 1 Nov 2006 18:37:23 +0000 (18:37 +0000)]
[HVM] Move shadow initialisation into domain-creation hypercall.
Allocate HVM guest memory in the libxc builder function rather
than in xend. Clean up fall out from these changes.

Todo: Fix ia64. Move PV builder to same model (it should allocate the
memory rather than xend doing so -- then it can avoid using
xc_get_pfn_list()).

Signed-off-by: Keir Fraser <keir@xensource.com>
19 years agoPrint errno for each of the "Error when writing to state file" messages.
Ewan Mellor [Wed, 1 Nov 2006 18:32:45 +0000 (18:32 +0000)]
Print errno for each of the "Error when writing to state file" messages.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
19 years agoReturn the domain renaming during migration.
Ewan Mellor [Wed, 1 Nov 2006 18:20:23 +0000 (18:20 +0000)]
Return the domain renaming during migration.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
19 years ago[HVM] HVM is now a flag parameter to domain-creation hypercall.
kfraser@localhost.localdomain [Wed, 1 Nov 2006 16:08:19 +0000 (16:08 +0000)]
[HVM] HVM is now a flag parameter to domain-creation hypercall.
This cleans up HVM start-of-day in Xen and means that the
HVM status of a domain is maintained from cradle to grave.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years agoThis patch adds a checking of wrong parameter into xm dmesg command.
Ewan Mellor [Wed, 1 Nov 2006 15:56:55 +0000 (15:56 +0000)]
This patch adds a checking of wrong parameter into xm dmesg command.
If wrong parameter is given, xm dmesg command shows the following
error messages.

Error: No parameter required
Usage: xm dmesg [-c|--clear]

Read and/or clear Xend's message buffer.
  -c, --clear                    Clear dmesg buffer

Signed-off-by: Masaki Kanno <kanno.masaki@jp.fujitsu.com>
19 years agoFix comment.
Ewan Mellor [Wed, 1 Nov 2006 15:56:22 +0000 (15:56 +0000)]
Fix comment.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
19 years agoAdd full flag to domain(domid), fixing xm list --long VM1.
Ewan Mellor [Wed, 1 Nov 2006 15:41:35 +0000 (15:41 +0000)]
Add full flag to domain(domid), fixing xm list --long VM1.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
19 years agoCreate /var/lib/xend/domains on startup if it doesn't exist.
Ewan Mellor [Wed, 1 Nov 2006 13:50:02 +0000 (13:50 +0000)]
Create /var/lib/xend/domains on startup if it doesn't exist.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
19 years ago[XEN] Extend dom0_mem syntax to support min/max clamps.
kfraser@localhost.localdomain [Wed, 1 Nov 2006 11:18:29 +0000 (11:18 +0000)]
[XEN] Extend dom0_mem syntax to support min/max clamps.

dom0_mem=[min:<min_amt>,][max:<max_amt>,][<amt>]

<min_amt>: The minimum amount of memory which should be allocated for dom0.
<max_amt>: The maximum amount of memory which should be allocated for dom0.
<amt>:     The precise amount of memory to allocate for dom0.

Notes:
 1. <amt> is clamped from below by <min_amt> and from above by available
    memory and <max_amt>
 2. <min_amt> is clamped from above by available memory and <max_amt>
 3. <min_amt> is ignored if it is greater than <max_amt>
 4. If <amt> is not specified, it is calculated as follows:
    "All of memory is allocated to domain 0, minus 1/16th which is reserved
     for uses such as DMA buffers (the reservation is clamped to 128MB)."

Each value can be specified as positive or negative:
 If +ve: The specified amount is an absolute value.
 If -ve: The specified amount is subtracted from total available memory.

Signed-off-by: Keir Fraser <keir@xensource.com>
19 years agoMerge.
Ewan Mellor [Wed, 1 Nov 2006 10:41:44 +0000 (10:41 +0000)]
Merge.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
19 years agoMerge xen-api.hg changeset 142:d75413e65ba319d577c764be6c13efa5eae6da1c.
Ewan Mellor [Wed, 1 Nov 2006 10:40:46 +0000 (10:40 +0000)]
Merge xen-api.hg changeset 142:d75413e65ba319d577c764be6c13efa5eae6da1c.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
19 years ago[XEN] Prefetch multiple shadow entries per pagefault
Tim Deegan [Wed, 1 Nov 2006 10:31:11 +0000 (10:31 +0000)]
[XEN] Prefetch multiple shadow entries per pagefault
Also, clean up the shadow *_propagate/fault routines.
This allows us to quickly dispatch some guest-not-present faults
and most MMIO accesses without taking the shadow lock.
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
19 years ago[XEN] Fix building on OpenBSD.
kfraser@localhost.localdomain [Wed, 1 Nov 2006 10:02:00 +0000 (10:02 +0000)]
[XEN] Fix building on OpenBSD.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[XEN] Fix deadlock in printk().
kfraser@localhost.localdomain [Wed, 1 Nov 2006 09:55:43 +0000 (09:55 +0000)]
[XEN] Fix deadlock in printk().
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years agoStop enforcing -g for some of the tools.
kfraser@localhost.localdomain [Tue, 31 Oct 2006 18:24:57 +0000 (18:24 +0000)]
Stop enforcing -g for some of the tools.
Signed-off-by: John Levon <john.levon@sun.com>
19 years ago[XEND] Improve error message for unparsed disk modes.
kfraser@localhost.localdomain [Tue, 31 Oct 2006 18:24:44 +0000 (18:24 +0000)]
[XEND] Improve error message for unparsed disk modes.
Signed-off-by: John Levon <john.levon@sun.com>
19 years ago[HVM] Use correct types for guest physical addresses
Tim Deegan [Tue, 31 Oct 2006 16:42:46 +0000 (16:42 +0000)]
[HVM] Use correct types for guest physical addresses
Guest physical addresses are not guaranteed to fit in either a pointer
or an unsigned long int; use paddr_t for them.
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
19 years ago[HVM] Add support for 'add r8,m8' instruction to memory-mapped I/O.
kfraser@localhost.localdomain [Tue, 31 Oct 2006 16:22:39 +0000 (16:22 +0000)]
[HVM] Add support for 'add r8,m8' instruction to memory-mapped I/O.
Signed-off-by: Kevin Tronkowski <ktronkowski@virtualiron.com>
Signed-off-by: Ben Thomas <bthomas@virtualiron.com>
19 years ago[NET] front: Clean up error handling. This eliminates earlier
kfraser@localhost.localdomain [Tue, 31 Oct 2006 16:18:07 +0000 (16:18 +0000)]
[NET] front: Clean up error handling. This eliminates earlier
workaround patch for an observed crash.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years agoClean up recent changes to reboot code. This fixes PV save/restore.
kfraser@localhost.localdomain [Tue, 31 Oct 2006 15:54:03 +0000 (15:54 +0000)]
Clean up recent changes to reboot code. This fixes PV save/restore.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years agoUpdate .hgignore
kfraser@localhost.localdomain [Tue, 31 Oct 2006 15:04:05 +0000 (15:04 +0000)]
Update .hgignore

19 years ago[HVM] vlapic: Synchronously determine PPR register value whenevr it is needed.
kfraser@localhost.localdomain [Tue, 31 Oct 2006 14:39:50 +0000 (14:39 +0000)]
[HVM] vlapic: Synchronously determine PPR register value whenevr it is needed.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[VMX] Replace vmx_load_cr2() with direct write to %cr2.
kfraser@localhost.localdomain [Tue, 31 Oct 2006 13:26:49 +0000 (13:26 +0000)]
[VMX] Replace vmx_load_cr2() with direct write to %cr2.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[HVM] ACPI: Clean up table definitions, move under hvmloader/
kfraser@localhost.localdomain [Tue, 31 Oct 2006 12:42:50 +0000 (12:42 +0000)]
[HVM] ACPI: Clean up table definitions, move under hvmloader/
directory pending tighter integration of table generation
into hvmloader. This patch also includes the TPM TIS SSDT
generator from Stefan Berger at IBM.

Signed-off-by: Keir Fraser <keir@xensource.com>
19 years agoMerge.
Steven Smith [Tue, 31 Oct 2006 11:44:47 +0000 (11:44 +0000)]
Merge.

19 years ago[XEN] Add a warning to the i8259 if we generate spurious IRQs. Tidy up
Steven Smith [Tue, 31 Oct 2006 11:44:28 +0000 (11:44 +0000)]
[XEN] Add a warning to the i8259 if we generate spurious IRQs.  Tidy up
a little while I'm here.

Signed-off-by: Steven Smith <sos22@cam.ac.uk>
19 years ago[PV-ON-HVM] Don't generate lots of spurious interrupts when using event
Steven Smith [Tue, 31 Oct 2006 11:38:55 +0000 (11:38 +0000)]
[PV-ON-HVM] Don't generate lots of spurious interrupts when using event
channel upcalls.

The issue here was that the Xen platform PCI interrupt is only updated
when you return from the hypervisor into guest context, and so remained
asserted for a short interval after the interrupt handler ran.  If
it happened that the first subsequent trap to the hypervisor was
for unmasking the 8259 interrupt again, the unmasking caused the interrupt
to be reinjected.  This caused an edge on the chaining interrupt from
the slave PIC to the master.  The platform interrupt on the slave
would then be cleared as we returned to the guest, and so you
eventually end up injecting an interrupt on the master chained
interrupt with nothing pending on the slave, which shows up as
a spurious interrupt in the guest.

Signed-off-by: Steven Smith <sos22@cam.ac.uk>
19 years ago[NETFRONT] Make sure we don't crash if the backend goes to state
Steven Smith [Tue, 31 Oct 2006 11:31:34 +0000 (11:31 +0000)]
[NETFRONT] Make sure we don't crash if the backend goes to state
Closing before we have a chance to connect.  This can happen if
there's an error while connecting.

Signed-off-by: Steven Smith <sos22@cam.ac.uk>
19 years agoMerge.
Steven Smith [Tue, 31 Oct 2006 10:12:35 +0000 (10:12 +0000)]
Merge.

19 years ago[HVM][LINUX][TOOLS] Split control/reboot_module into control/feature-reboot
Steven Smith [Tue, 31 Oct 2006 10:11:53 +0000 (10:11 +0000)]
[HVM][LINUX][TOOLS] Split control/reboot_module into control/feature-reboot
and control/feature-sysrq.  This is more consistent with the other
PV device protocols.

Signed-off-by: Steven Smith <sos22@cam.ac.uk>
19 years ago[XEN] Cleanup and optimize layout of credit scheduler stats.
Emmanuel Ackaouy [Tue, 31 Oct 2006 10:10:17 +0000 (10:10 +0000)]
[XEN] Cleanup and optimize layout of credit scheduler stats.
Also carve out per VCPU stats for clarity and add per-VCPU
migration counter.
Signed-off-by: Emmanuel Ackaouy <ack@xensource.com>
19 years ago[HVM][LINUX][TOOLS] Make xm {shutdown|reboot} do something sensible for HVM
Steven Smith [Tue, 31 Oct 2006 09:54:49 +0000 (09:54 +0000)]
[HVM][LINUX][TOOLS] Make xm {shutdown|reboot} do something sensible for HVM
domains with PV drivers loaded.  This patch creates a new PV-on-HVM
module, reboot.ko, which, when loaded, creates control/reboot_module
in the store.  The tools notice this, and disable the watch which would
normally destroy HVM domains which are the target of an xm shutdown
command, allowing the reboot module to shut the domain down cleanly.

Signed-off-by: Tetsu Yamamoto <yamamoto.tetsu@jp.fujitsu.com>
(Checkin comments by Steven Smith <sos22@cam.ac.uk>)

19 years ago[NET] back: Remove warning about non-zero tx queue length.
kfraser@localhost.localdomain [Tue, 31 Oct 2006 09:49:31 +0000 (09:49 +0000)]
[NET] back: Remove warning about non-zero tx queue length.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[XEN] Cleanup and improve accuracy of credit scheduler accounting code
Emmanuel Ackaouy [Mon, 30 Oct 2006 17:35:11 +0000 (17:35 +0000)]
[XEN] Cleanup and improve accuracy of credit scheduler accounting code
Signed-off-by: Emmanuel Ackaouy <ack@xensource.com>
19 years ago[LINUX] Fix build breakage from previous patch.
kfraser@localhost.localdomain [Mon, 30 Oct 2006 14:28:29 +0000 (14:28 +0000)]
[LINUX] Fix build breakage from previous patch.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[LINUX] Use CONFIG_XEN_COMPAT_030002 where possible.
kfraser@localhost.localdomain [Mon, 30 Oct 2006 14:14:46 +0000 (14:14 +0000)]
[LINUX] Use CONFIG_XEN_COMPAT_030002 where possible.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
19 years ago[LINUX] Declare structs as 'static __initdata' where possible.
kfraser@localhost.localdomain [Mon, 30 Oct 2006 14:09:13 +0000 (14:09 +0000)]
[LINUX] Declare structs as 'static __initdata' where possible.

Probably still not catching all possible cases, but using static
(where not already the case) makes the code size smaller, and using __initdata
has the usual advantages.

Signed-off-by: Jan Beulich <jbeulich@novell.com>
19 years ago[LINUX] privcmd: Range-check hypercall index.
kfraser@localhost.localdomain [Mon, 30 Oct 2006 14:04:44 +0000 (14:04 +0000)]
[LINUX] privcmd: Range-check hypercall index.
Otherwise, bugs in e.g. libxc may bring the kernel down.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
19 years ago[XEN] HVM: Clean up and simplify vlapic device-model code.
kfraser@localhost.localdomain [Mon, 30 Oct 2006 13:53:09 +0000 (13:53 +0000)]
[XEN] HVM: Clean up and simplify vlapic device-model code.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[HVM][VMX] Enable VMX TPR shadow feature.
kfraser@localhost.localdomain [Mon, 30 Oct 2006 10:42:27 +0000 (10:42 +0000)]
[HVM][VMX] Enable VMX TPR shadow feature.

x64 Windows uses CR8 to access TPR very frequently. This patch enables
TPR shadow and allows mov-from/to-CR8 to access it directly; tests
indicates it can boost greatly the performance of x64 Windows 2003.

Signed-off-by: Eddie Dong <eddie.dong@intel.com>
Signed-off-by: Yunhong Jiang <yunhong.jiang@intel.com>
Signed-off-by: Dexuan Cui <dexuan.cui@intel.com>
19 years ago[HVM] Fix Qemu-dm serial issues:
kfraser@localhost.localdomain [Mon, 30 Oct 2006 09:45:17 +0000 (09:45 +0000)]
[HVM] Fix Qemu-dm serial issues:
 1. Retry transmit via a polling timer if a byte cannot be written
    immediately to its destination.
 2. Turn off output processing of raw serial lines.

Signed-off-by: Xiaowei Yang <xiaowei.yang@intel.com>
Signed-off-by: Yunhong Jiang <yunhong.jiang@intel.com>
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years agoFix handling of SXP for devices through device_add -- this solves the
Ewan Mellor [Sun, 29 Oct 2006 19:28:26 +0000 (19:28 +0000)]
Fix handling of SXP for devices through device_add -- this solves the
Invalid device (uname) problem that breaks just about everything once you've
done xm block-attach.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
19 years ago[XEN] Add implicit "\n" back into MEM_LOG(). Removed by mistake.
kaf24@firebug.cl.cam.ac.uk [Sat, 28 Oct 2006 11:30:38 +0000 (12:30 +0100)]
[XEN] Add implicit "\n" back into MEM_LOG(). Removed by mistake.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[XEN] Some cleanups to the log-level stuff. Largely this
kaf24@localhost.localdomain [Fri, 27 Oct 2006 22:47:30 +0000 (23:47 +0100)]
[XEN] Some cleanups to the log-level stuff. Largely this
is a rename and slight interface change to DPRINTK (now
replaced by dprintk() and gdprintk()). Also shuffle some
log-level definitions around and tweak the semantics of
the upper threshold.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[QEMU] Add a TIS device model compliant to the 1.2 TPM specification.
kaf24@localhost.localdomain [Fri, 27 Oct 2006 18:15:48 +0000 (19:15 +0100)]
[QEMU] Add a TIS device model compliant to the 1.2 TPM specification.
It implements all registers necessary to make the Linux TIS driver
work (tpm_tis.c). All of the basic registers supported by this type of
device are implemented. Also the locality selection has been
implemented, but has not been tested. The legacy registers as
described in the specification are not supported.

Current caveat: The device has so far not yet been integrated with the
virtual TPM available in the repository. It will require changes to
the virtual TPM spawned by the vTPM manager to offer an additional message
interface. The TIS interface itself then needs to have an additional
transport implemented. (see vTPMTransmit array).

The relevant specification for the device model can be found here:
https://www.trustedcomputinggroup.org/groups/pc_client/TCG_PCClientTPMSpecification_1-20_1-00_FINAL.pdf

Signed-off-by: Stefan Berger <stefanb@us.ibm.com>
19 years ago[QEMU] Helper functions to interface with the xenstore and read device information...
kaf24@localhost.localdomain [Fri, 27 Oct 2006 18:13:51 +0000 (19:13 +0100)]
[QEMU] Helper functions to interface with the xenstore and read device information from it.

 - detect what types of devices a domain has or whether a domain has a
   device of a certain type
 - read the content of a variable related to a device, i.e.,
   hotplug-status
 - subscribe to changes of the hotplug status of a device for not
   having to poll the status

Signed-off-by: Stefan Berger <stefanb@us.ibm.com>
19 years ago[XEN] Logging parameters for ia64 code.
kaf24@localhost.localdomain [Fri, 27 Oct 2006 17:57:02 +0000 (18:57 +0100)]
[XEN] Logging parameters for ia64 code.
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
19 years ago[XEN] Logging parameters for powerpc code.
kaf24@localhost.localdomain [Fri, 27 Oct 2006 17:56:44 +0000 (18:56 +0100)]
[XEN] Logging parameters for powerpc code.
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
19 years ago[XEN] Logging parameters for x86 code.
kaf24@localhost.localdomain [Fri, 27 Oct 2006 17:56:24 +0000 (18:56 +0100)]
[XEN] Logging parameters for x86 code.
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
19 years ago[XEN] Logging parameters for common code.
kaf24@localhost.localdomain [Fri, 27 Oct 2006 17:55:16 +0000 (18:55 +0100)]
[XEN] Logging parameters for common code.
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
19 years ago[XEN] Implement log levels. It adds the thresholds and code to printk
kaf24@localhost.localdomain [Fri, 27 Oct 2006 17:52:55 +0000 (18:52 +0100)]
[XEN] Implement log levels. It adds the thresholds and code to printk
to implement the thresholds.
Signed-off-by: Steven Rostedt <srostedt@redhat.com>
19 years ago[XEN] Implement rate-limited logging.
kaf24@localhost.localdomain [Fri, 27 Oct 2006 17:46:20 +0000 (18:46 +0100)]
[XEN] Implement rate-limited logging.

Taken from Andi Kleen's rate limit in the Linux kernel.

This keeps large amounts of prints in the HV down.

Signed-off-by: Steven Rostedt <srostedt@redhat.com>
19 years ago[XEND] Do not delete VNC password from configuration.
kaf24@localhost.localdomain [Fri, 27 Oct 2006 17:15:14 +0000 (18:15 +0100)]
[XEND] Do not delete VNC password from configuration.
Fixes reboot of HVM guest when VNC authentication is in use.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[XEND] Do not delete VNC password from configuration.
kaf24@localhost.localdomain [Fri, 27 Oct 2006 17:15:14 +0000 (18:15 +0100)]
[XEND] Do not delete VNC password from configuration.
Fixes reboot of HVM guest when VNC authentication is in use.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years ago[XEN] perf counters: Fix NULL-pointer check. Should happen later.
kaf24@localhost.localdomain [Fri, 27 Oct 2006 17:10:37 +0000 (18:10 +0100)]
[XEN] perf counters: Fix NULL-pointer check. Should happen later.
Signed-of-by: Kouya Shimura <kouya@jp.fujitsu.com>
19 years ago[HVM] Ensure that, if AP is blocked, it is woken by interrupt delivered by IOAPIC.
kaf24@localhost.localdomain [Fri, 27 Oct 2006 17:02:40 +0000 (18:02 +0100)]
[HVM] Ensure that, if AP is blocked, it is woken by interrupt delivered by IOAPIC.

This fixes 64-bit SMP Windows 2k3 boot.

Signed-off-by: Xiaowei Yang <xiaowei.yang@intel.com>
Signed-off-by: Xin Li <xin.b.li@intel.com>
19 years ago[HVM] Fix SMP timer issues:
kaf24@localhost.localdomain [Fri, 27 Oct 2006 17:00:03 +0000 (18:00 +0100)]
[HVM] Fix SMP timer issues:
 * Sync AP TSCs with BP at startup
 * Only halt BP TSC when descheduled
 * Correctly handle IPIs on timer vector
Signed-off-by: Xiaowei Yang <xiaowei.yang@intel.com>
19 years ago[QEMU] ne2000: Stop memory access beyond buffer
kaf24@localhost.localdomain [Fri, 27 Oct 2006 16:50:40 +0000 (17:50 +0100)]
[QEMU] ne2000: Stop memory access beyond buffer

As a program that runs in dom0 which serves users from guests,
the qemu drivers need to be vigilant to the input that comes
from the guests since they may be malicious.

As it is there are multiple ways to get ne2000 to read/write
memory beyond the 48K buffer that it has allocated for each
adapter.

This patch checks the addresses and prevents this from occuring.

The boundary is checked each time since it's changed for every
packet received while the other parameters are only changed
(by the guest) during setup.

Signed-off: Herbert Xu <herbert@gondor.apana.org.au>

19 years agoImprove I/O performance when competing with CPU intensive workloads.
Emmanuel Ackaouy [Fri, 27 Oct 2006 14:44:27 +0000 (15:44 +0100)]
Improve I/O performance when competing with CPU intensive workloads.
Allow non CPU consuming wake-to-run latency sensitive VCPUs to
preempt CPU consuming ones.

Signed-off-by: Emmanuel Ackaouy <ack@xensource.com>
19 years ago[TOOLS] Uncompress and allocate memory for gzipped kernel and initrd images on
Ian Campbell [Thu, 26 Oct 2006 15:56:16 +0000 (16:56 +0100)]
[TOOLS] Uncompress and allocate memory for gzipped kernel and initrd images on
the fly. We cannot rely on the length contained in the gzip trailer to determine
the length of the decompressed data because images have been observed which have
trailing junk.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
19 years ago[HVM] Fix qemu's test for whether physcal addresses are RAM.
Tim Deegan [Thu, 26 Oct 2006 14:08:20 +0000 (15:08 +0100)]
[HVM] Fix qemu's test for whether physcal addresses are RAM.
HVM guests have a memory hole below 4GB, so can't just check whether
addresses are < the amount of RAM allocated.
Signed-off-by: Tim Deegan <Tim.Deegan@xensource.com>
19 years ago[NET] back: Advertise that we do not support rx-flip path any more.
kfraser@localhost.localdomain [Thu, 26 Oct 2006 10:50:17 +0000 (11:50 +0100)]
[NET] back: Advertise that we do not support rx-flip path any more.
This will now be used only by older guests who do not understand
the feature-rx-{copy,flip} feature flags.
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years agoMerge with xen-isa64-unstable.hg
kfraser@localhost.localdomain [Thu, 26 Oct 2006 10:26:05 +0000 (11:26 +0100)]
Merge with xen-isa64-unstable.hg

19 years agoFix ramdisk Makefile so that downloading from xm-test.xensource.com actually
Ewan Mellor [Thu, 26 Oct 2006 09:20:22 +0000 (10:20 +0100)]
Fix ramdisk Makefile so that downloading from xm-test.xensource.com actually
works.

Signed-off-by: Tony Breeds <tony@bakeyournoodle.com>
19 years ago[LIBXC][IA64] fix build warning
awilliam@xenbuild.aw [Wed, 25 Oct 2006 21:29:00 +0000 (15:29 -0600)]
[LIBXC][IA64] fix build warning

With warnings becoming errors, this fixes the libxc build on ia64

Signed-off-by: Alex Williamson <alex.williamson@hp.com>
19 years ago[IA64] NUMA support
awilliam@xenbuild.aw [Wed, 25 Oct 2006 20:48:55 +0000 (14:48 -0600)]
[IA64] NUMA support

Based on patch from Tristan Gingold

Signed-off-by: Alex Williamson <alex.williamson@hp.com>
19 years agomerge with xen-unstable.hg
awilliam@xenbuild.aw [Wed, 25 Oct 2006 18:12:01 +0000 (12:12 -0600)]
merge with xen-unstable.hg

19 years ago[IA64][HVM][QEMU] Add buffer IO mechanism for IA64/VTi domain.
kfraser@localhost.localdomain [Wed, 25 Oct 2006 14:29:36 +0000 (15:29 +0100)]
[IA64][HVM][QEMU] Add buffer IO mechanism for IA64/VTi domain.

Signed-off-by:Zhang xiantao <xiantao.zhang@intel.com>

19 years agoImport Xend part of xen-unstable changeset
Ewan Mellor [Wed, 25 Oct 2006 14:29:08 +0000 (15:29 +0100)]
Import Xend part of xen-unstable changeset
11922:1e6b0a8a0822a22d313d9058fa0dc10c7a319821.

[IA64][HVM] Add buffer IO mechanism for Xen/VTi domain. Current
implementation can accelerate Windows guest's dense IO operations
at boot time.

Signed-off-by: Zhang xiantao <xiantao.zhang@intel.com>
19 years ago[IA64][HVM] Add buffer IO mechanism for Xen/VTi domain. Current
kfraser@localhost.localdomain [Wed, 25 Oct 2006 14:29:08 +0000 (15:29 +0100)]
[IA64][HVM] Add buffer IO mechanism for Xen/VTi domain. Current
implementation can accelerate Windows guest's dense IO operations
at boot time.

Signed-off-by: Zhang xiantao <xiantao.zhang@intel.com>
19 years ago[LIBXC] Add -Wmissing-prototypes to CFLAGS, fix warnings resulting from that.
kfraser@localhost.localdomain [Wed, 25 Oct 2006 14:25:13 +0000 (15:25 +0100)]
[LIBXC] Add -Wmissing-prototypes to CFLAGS, fix warnings resulting from that.

Signed-off-by: Gerd Hoffmann <kraxel@suse.de>
19 years ago[XEN] Clean up NUMA stuff and disable by default ('numa=on' enables it).
kfraser@localhost.localdomain [Wed, 25 Oct 2006 14:17:40 +0000 (15:17 +0100)]
[XEN] Clean up NUMA stuff and disable by default ('numa=on' enables it).
Signed-off-by: Keir Fraser <keir@xensource.com>
19 years agoPrint the time to xend-debug.log, when starting or restarting Xend.
Ewan Mellor [Wed, 25 Oct 2006 14:01:51 +0000 (15:01 +0100)]
Print the time to xend-debug.log, when starting or restarting Xend.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
19 years agoPrint the time to xend-debug.log, when starting or restarting Xend.
Ewan Mellor [Wed, 25 Oct 2006 14:01:51 +0000 (15:01 +0100)]
Print the time to xend-debug.log, when starting or restarting Xend.

Signed-off-by: Ewan Mellor <ewan@xensource.com>
19 years agoPV-to-HVM: Add a compatibility define for the end_that_request_last
Ian Campbell [Wed, 25 Oct 2006 12:58:30 +0000 (13:58 +0100)]
PV-to-HVM: Add a compatibility define for the end_that_request_last
which changed prototype in 2.6.16.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Signed-off-by: K. Y. Srinivasan <ksrinivasan@novell.com>
Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>
19 years agoPV-on-HVM: Include compatibility kzalloc implementation for kernels
Ian Campbell [Wed, 25 Oct 2006 12:58:30 +0000 (13:58 +0100)]
PV-on-HVM: Include compatibility kzalloc implementation for kernels
before 2.6.14.

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Signed-off-by: K. Y. Srinivasan <ksrinivasan@novell.com>
Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>
19 years agoPV-to-HVM: Impletement compatibility version of
Ian Campbell [Wed, 25 Oct 2006 12:58:30 +0000 (13:58 +0100)]
PV-to-HVM: Impletement compatibility version of
schedule_timeout_interruptible for kernels before 2.6.14

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Signed-off-by: K. Y. Srinivasan <ksrinivasan@novell.com>
Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>
19 years agoPV-on-HVM: Implement and export a compatibility version of do_exit().
Ian Campbell [Wed, 25 Oct 2006 12:58:30 +0000 (13:58 +0100)]
PV-on-HVM: Implement and export a compatibility version of do_exit().
Kernels prior to 2.6.12 did not export do_exit().

Signed-off-by: Ian Campbell <ian.campbell@xensource.com>
Signed-off-by: K. Y. Srinivasan <ksrinivasan@novell.com>
Signed-off-by: Tsunehisa Doi <Doi.Tsunehisa@jp.fujitsu.com>