OSDN Git Service

qmiga/qemu.git
9 years agotarget-alpha: Set PC correctly for floating-point exceptions
Richard Henderson [Sat, 28 Jun 2014 20:06:19 +0000 (13:06 -0700)]
target-alpha: Set PC correctly for floating-point exceptions

PC should be one past the faulting insn.  Add better commentary
for the machine-check exception path.

Reported-by: Al Viro <viro@ZenIV.linux.org.uk>
Signed-off-by: Richard Henderson <rth@twiddle.net>
9 years agotarget-alpha: Forget installed round mode after MT_FPCR
Richard Henderson [Sat, 28 Jun 2014 17:25:36 +0000 (10:25 -0700)]
target-alpha: Forget installed round mode after MT_FPCR

When we use QUAL_RM_D, we copy fpcr_dyn_round to float_status.
When we install a new FPCR value, we update fpcr_dyn_round.
Reset the status of the cache so that we re-copy for the next
fp insn that requires dynamic rounding.

Signed-off-by: Richard Henderson <rth@twiddle.net>
9 years agotarget-alpha: Rename floating-point subroutines
Richard Henderson [Fri, 8 Aug 2014 20:54:35 +0000 (10:54 -1000)]
target-alpha: Rename floating-point subroutines

... to match the instructions, which have no leading "f".

Signed-off-by: Richard Henderson <rth@twiddle.net>
9 years agotarget-alpha: Move VAX helpers to a new file
Richard Henderson [Fri, 8 Aug 2014 20:42:45 +0000 (10:42 -1000)]
target-alpha: Move VAX helpers to a new file

Keep the IEEE and VAX floating point emulation separate.

Signed-off-by: Richard Henderson <rth@twiddle.net>
9 years agoMerge remote-tracking branch 'remotes/armbru/tags/pull-qapi-2015-05-15' into staging
Peter Maydell [Fri, 15 May 2015 16:51:20 +0000 (17:51 +0100)]
Merge remote-tracking branch 'remotes/armbru/tags/pull-qapi-2015-05-15' into staging

qapi: Fix qapi mangling of downstream names, and more

# gpg: Signature made Fri May 15 17:41:31 2015 BST using RSA key ID EB918653
# gpg: Good signature from "Markus Armbruster <armbru@redhat.com>"
# gpg:                 aka "Markus Armbruster <armbru@pond.sub.org>"

* remotes/armbru/tags/pull-qapi-2015-05-15: (26 commits)
  qapi: Inline gen_command_decl_prologue(), gen_command_def_prologue()
  qapi: Drop pointless flush() before close()
  qapi: Factor open_output(), close_output() out of generators
  qapi: Turn generators' mandatory option -i into an argument
  qapi: Fix generators to report command line errors decently
  qapi: Factor parse_command_line() out of the generators
  qapi: qapi-commands.py option --type is unused, drop it
  qapi: qapi-event.py option -b does nothing, drop it
  tests: Add missing dependencies on $(qapi-py)
  qapi: Support downstream events and commands
  qapi: Support downstream alternates
  qapi: Support downstream flat unions
  qapi: Support downstream simple unions
  qapi: Support downstream structs
  qapi: Support downstream enums
  qapi: Make c_type() consistently convert qapi names
  qapi: Tidy c_type() logic
  qapi: Move camel_to_upper(), c_enum_const() to closely related code
  qapi: Use c_enum_const() in generate_alternate_qtypes()
  qapi: Simplify c_enum_const()
  ...

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoMerge remote-tracking branch 'remotes/rth/tags/pull-tcg-20150514' into staging
Peter Maydell [Fri, 15 May 2015 15:02:08 +0000 (16:02 +0100)]
Merge remote-tracking branch 'remotes/rth/tags/pull-tcg-20150514' into staging

Per-memop alignment

# gpg: Signature made Thu May 14 20:17:27 2015 BST using RSA key ID 4DD0279B
# gpg: Good signature from "Richard Henderson <rth7680@gmail.com>"
# gpg:                 aka "Richard Henderson <rth@redhat.com>"
# gpg:                 aka "Richard Henderson <rth@twiddle.net>"

* remotes/rth/tags/pull-tcg-20150514:
  tcg: Add MO_ALIGN, MO_UNALN
  tcg: Push merged memop+mmu_idx parameter to softmmu routines
  tcg: Merge memop and mmu_idx parameters to qemu_ld/st

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agotcg: Add MO_ALIGN, MO_UNALN
Richard Henderson [Wed, 13 May 2015 18:25:20 +0000 (11:25 -0700)]
tcg: Add MO_ALIGN, MO_UNALN

These modifiers control, on a per-memory-op basis, whether
unaligned memory accesses are allowed.  The default setting
reflects the target's definition of ALIGNED_ONLY.

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <rth@twiddle.net>
9 years agotcg: Push merged memop+mmu_idx parameter to softmmu routines
Richard Henderson [Wed, 13 May 2015 16:10:33 +0000 (09:10 -0700)]
tcg: Push merged memop+mmu_idx parameter to softmmu routines

The extra information is not yet used but it is now available.
This requires minor changes through all of the tcg backends.

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <rth@twiddle.net>
9 years agotcg: Merge memop and mmu_idx parameters to qemu_ld/st
Richard Henderson [Tue, 12 May 2015 18:51:44 +0000 (11:51 -0700)]
tcg: Merge memop and mmu_idx parameters to qemu_ld/st

At the tcg opcode level, not at the tcg-op.h generator level.
This requires minor changes through all of the tcg backends,
but none of the cpu translators.

Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <rth@twiddle.net>
9 years agoqapi: Inline gen_command_decl_prologue(), gen_command_def_prologue()
Markus Armbruster [Thu, 2 Apr 2015 12:52:55 +0000 (14:52 +0200)]
qapi: Inline gen_command_decl_prologue(), gen_command_def_prologue()

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
9 years agoqapi: Drop pointless flush() before close()
Markus Armbruster [Thu, 2 Apr 2015 12:49:29 +0000 (14:49 +0200)]
qapi: Drop pointless flush() before close()

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
9 years agoqapi: Factor open_output(), close_output() out of generators
Markus Armbruster [Thu, 2 Apr 2015 12:46:39 +0000 (14:46 +0200)]
qapi: Factor open_output(), close_output() out of generators

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
9 years agoqapi: Turn generators' mandatory option -i into an argument
Markus Armbruster [Thu, 2 Apr 2015 11:32:16 +0000 (13:32 +0200)]
qapi: Turn generators' mandatory option -i into an argument

Mandatory option is silly, and the error handling is missing: the
programs crash when -i isn't supplied.  Make it an argument, and check
it properly.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
9 years agoqapi: Fix generators to report command line errors decently
Markus Armbruster [Thu, 2 Apr 2015 11:17:34 +0000 (13:17 +0200)]
qapi: Fix generators to report command line errors decently

Report to stderr, prefix with the program name.  Also reject
extra arguments.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
9 years agoqapi: Factor parse_command_line() out of the generators
Markus Armbruster [Thu, 2 Apr 2015 11:12:21 +0000 (13:12 +0200)]
qapi: Factor parse_command_line() out of the generators

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
9 years agoqapi: qapi-commands.py option --type is unused, drop it
Markus Armbruster [Thu, 2 Apr 2015 09:41:22 +0000 (11:41 +0200)]
qapi: qapi-commands.py option --type is unused, drop it

Anything but --type sync (which is the default) suppresses output
entirely, which makes no sense.

Dates back to the initial commit c17d990.  Commit message says
"Currently only generators for synchronous qapi/qmp functions are
supported", so maybe output other than "synchronous qapi/qmp" was
planned at the time, to be selected with --type.

Should other kinds of output ever materialize, we can put the option
back.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
9 years agoqapi: qapi-event.py option -b does nothing, drop it
Markus Armbruster [Thu, 2 Apr 2015 09:40:21 +0000 (11:40 +0200)]
qapi: qapi-event.py option -b does nothing, drop it

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
9 years agotests: Add missing dependencies on $(qapi-py)
Markus Armbruster [Thu, 2 Apr 2015 11:38:48 +0000 (13:38 +0200)]
tests: Add missing dependencies on $(qapi-py)

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
9 years agoqapi: Support downstream events and commands
Eric Blake [Thu, 14 May 2015 12:51:01 +0000 (06:51 -0600)]
qapi: Support downstream events and commands

Enhance the testsuite to cover downstream events and commands.
Events worked without more tweaks, but commands needed a few final
updates in the generator to mangle names in the appropriate places.
In making those tweaks, it was easier to drop type_visitor() and
inline its actions instead.

Signed-off-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
9 years agoqapi: Support downstream alternates
Eric Blake [Thu, 14 May 2015 12:51:00 +0000 (06:51 -0600)]
qapi: Support downstream alternates

Enhance the testsuite to cover downstream alternates, including
whether the branch name or type is downstream.  Update the
generator to mangle alternate names in the appropriate places.

Signed-off-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
9 years agoqapi: Support downstream flat unions
Eric Blake [Thu, 14 May 2015 12:50:59 +0000 (06:50 -0600)]
qapi: Support downstream flat unions

Enhance the testsuite to cover downstream flat unions, including
the base type, discriminator name and type, and branch name and
type.  Update the generator to mangle the union names in the
appropriate places.

Signed-off-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
9 years agoqapi: Support downstream simple unions
Eric Blake [Thu, 14 May 2015 12:50:58 +0000 (06:50 -0600)]
qapi: Support downstream simple unions

Enhance the testsuite to cover downstream simple unions, including
when a union branch is a downstream name.  Update the generator to
mangle the union names in the appropriate places.

Signed-off-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
9 years agoqapi: Support downstream structs
Eric Blake [Thu, 14 May 2015 12:50:57 +0000 (06:50 -0600)]
qapi: Support downstream structs

Enhance the testsuite to cover downstream structs, including struct
members and base structs.  Update the generator to mangle the
struct names in the appropriate places.

Signed-off-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
9 years agoqapi: Support downstream enums
Eric Blake [Thu, 14 May 2015 12:50:56 +0000 (06:50 -0600)]
qapi: Support downstream enums

Enhance the testsuite to cover a downstream enum type and enum
string.  Update the generator to mangle the enum name in the
appropriate places.

Signed-off-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
9 years agoqapi: Make c_type() consistently convert qapi names
Eric Blake [Thu, 14 May 2015 12:50:55 +0000 (06:50 -0600)]
qapi: Make c_type() consistently convert qapi names

Continuing the string of cleanups for supporting downstream names
containing '.', this patch focuses on ensuring c_type() can
handle a downstream name.  This patch alone does not fix the
places where generator output should be calling this function
but was open-coding things instead, but it gets us a step closer.

In particular, the changes to c_list_type() and type_name() mean
that type_name(FOO) now handles the case when FOO contains '.',
'-', or is a ticklish identifier other than a builtin (builtins
are exempted because ['int'] must remain mapped to 'intList' and
not 'q_intList').  Meanwhile, ['unix'] now maps to 'q_unixList'
rather than 'unixList', to match the fact that 'unix' is ticklish;
however, our naming conventions state that complex types should
start with a capital, so no type name following conventions will
ever have the 'q_' prepended.

Likewise, changes to c_type() mean that c_type(FOO) properly
handles an enum or complex type FOO with '.' or '-' in the
name, or is a ticklish identifier (again, a ticklish identifier
as a type name violates conventions).

Signed-off-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
9 years agoqapi: Tidy c_type() logic
Eric Blake [Thu, 14 May 2015 12:50:54 +0000 (06:50 -0600)]
qapi: Tidy c_type() logic

c_type() is designed to be called on both string names and on
array designations, so 'name' is a bit misleading because it
operates on more than strings.  Also, no caller ever passes
an empty string.  Finally, + notation is a bit nicer to read
than '%s' % value for string concatenation.

Signed-off-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
9 years agoqapi: Move camel_to_upper(), c_enum_const() to closely related code
Markus Armbruster [Thu, 14 May 2015 12:50:53 +0000 (06:50 -0600)]
qapi: Move camel_to_upper(), c_enum_const() to closely related code

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Eric Blake <eblake@redhat.com>
9 years agoqapi: Use c_enum_const() in generate_alternate_qtypes()
Markus Armbruster [Thu, 14 May 2015 12:50:52 +0000 (06:50 -0600)]
qapi: Use c_enum_const() in generate_alternate_qtypes()

Missed in commit b0b5819.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Eric Blake <eblake@redhat.com>
9 years agoqapi: Simplify c_enum_const()
Markus Armbruster [Thu, 14 May 2015 12:50:51 +0000 (06:50 -0600)]
qapi: Simplify c_enum_const()

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Eric Blake <eblake@redhat.com>
9 years agoqapi: Rename generate_enum_full_value() to c_enum_const()
Markus Armbruster [Thu, 14 May 2015 12:50:50 +0000 (06:50 -0600)]
qapi: Rename generate_enum_full_value() to c_enum_const()

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Eric Blake <eblake@redhat.com>
9 years agoqapi: Rename _generate_enum_string() to camel_to_upper()
Markus Armbruster [Thu, 14 May 2015 12:50:49 +0000 (06:50 -0600)]
qapi: Rename _generate_enum_string() to camel_to_upper()

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Eric Blake <eblake@redhat.com>
9 years agoqapi: Rename identical c_fun()/c_var() into c_name()
Eric Blake [Thu, 14 May 2015 12:50:48 +0000 (06:50 -0600)]
qapi: Rename identical c_fun()/c_var() into c_name()

Now that the two functions are identical, we only need one of them,
and we might as well give it a more descriptive name.  Basically,
the function serves as the translation from a QAPI name into a
(portion of a) C identifier, without regards to whether it is a
variable or function name.

Signed-off-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
9 years agoqapi: Fix C identifiers generated for names containing '.'
Markus Armbruster [Thu, 14 May 2015 12:50:47 +0000 (06:50 -0600)]
qapi: Fix C identifiers generated for names containing '.'

c_fun() maps '.' to '_', c_var() doesn't.  Nothing prevents '.' in
QAPI names that get passed to c_var().

Which QAPI names get passed to c_fun(), to c_var(), or to both is not
obvious.  Names of command parameters and struct type members get
passed to c_var().

c_var() strips a leading '*', but this cannot happen.  c_fun()
doesn't.

Fix c_var() to work exactly like c_fun().

Perhaps they should be replaced by a single mapping function.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
[add 'import string']
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
9 years agodoc: fix qmp event type
Michael S. Tsirkin [Mon, 11 May 2015 15:17:49 +0000 (17:17 +0200)]
doc: fix qmp event type

Event name for hot unplug errors was wrong.
Make doc match code.

Cc: Zhu Guihua <zhugh.fnst@cn.fujitsu.com>
Reported-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
9 years agoqmp: Add qom_path field to query-cpus command
Eduardo Habkost [Fri, 8 May 2015 19:04:22 +0000 (16:04 -0300)]
qmp: Add qom_path field to query-cpus command

This will allow clients to query additional information directly using
qom-get on the CPU objects.

Reviewed-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Andreas Färber <afaerber@suse.de>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
9 years agoMerge remote-tracking branch 'remotes/agraf/tags/signed-s390-for-upstream' into staging
Peter Maydell [Wed, 13 May 2015 15:06:07 +0000 (16:06 +0100)]
Merge remote-tracking branch 'remotes/agraf/tags/signed-s390-for-upstream' into staging

Patch queue for s390 - 2015-05-13

A few TCG fixes for the s390x target. Nothing special, but with these
applied I can run most of the SLE12 binaries in Linux-user emulation.

# gpg: Signature made Wed May 13 13:49:25 2015 BST using RSA key ID 03FEDC60
# gpg: Good signature from "Alexander Graf <agraf@suse.de>"
# gpg:                 aka "Alexander Graf <alex@csgraf.de>"

* remotes/agraf/tags/signed-s390-for-upstream:
  s390x: Add interlocked access facility 1 instructions
  s390x: Add some documentation in opcode list
  s390x: Fix stoc direction

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoMerge remote-tracking branch 'remotes/jnsnow/tags/ide-cve-pull-request' into staging
Peter Maydell [Wed, 13 May 2015 12:57:44 +0000 (13:57 +0100)]
Merge remote-tracking branch 'remotes/jnsnow/tags/ide-cve-pull-request' into staging

# gpg: Signature made Wed May 13 12:52:19 2015 BST using RSA key ID AAFC390E
# gpg: Good signature from "John Snow (John Huston) <jsnow@redhat.com>"
# gpg: WARNING: This key is not certified with sufficiently trusted signatures!
# gpg:          It is not certain that the signature belongs to the owner.
# Primary key fingerprint: FAEB 9711 A12C F475 812F  18F2 88A9 064D 1835 61EB
#      Subkey fingerprint: F9B7 ABDB BCAC DF95 BE76  CBD0 7DEF 8106 AAFC 390E

* remotes/jnsnow/tags/ide-cve-pull-request:
  fdc: force the fifo access to be in bounds of the allocated buffer

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agos390x: Add interlocked access facility 1 instructions
Alexander Graf [Fri, 8 May 2015 01:07:53 +0000 (03:07 +0200)]
s390x: Add interlocked access facility 1 instructions

We're currently missing all instructions defined by the "interlocked-access
facility 1" which is part of zEC12. This patch implements all of them except
for LPD and LPDG.

Signed-off-by: Alexander Graf <agraf@suse.de>
Reviewed-by: Richard Henderson <rth@twiddle.net>
9 years agos390x: Add some documentation in opcode list
Alexander Graf [Fri, 8 May 2015 01:06:41 +0000 (03:06 +0200)]
s390x: Add some documentation in opcode list

I find it really hard to grasp what each field in the opcode list means.
Slowly walking through its semantics myself, I figured I'd write a small
summary at the top of the file to make life easier for me and whoever
looks at the file next.

Signed-off-by: Alexander Graf <agraf@suse.de>
Reviewed-by: Richard Henderson <rth@twiddle.net>
9 years agos390x: Fix stoc direction
Alexander Graf [Wed, 15 Apr 2015 01:45:41 +0000 (03:45 +0200)]
s390x: Fix stoc direction

The store conditional instruction wants to store when the condition
is fulfilled, so we should branch out when it's not true.

The code today branches out when the condition is true, clearly
reversing the logic. Fix it up by negating the condition.

Signed-off-by: Alexander Graf <agraf@suse.de>
Reviewed-by: Richard Henderson <rth@twiddle.net>
9 years agofdc: force the fifo access to be in bounds of the allocated buffer
Petr Matousek [Wed, 6 May 2015 07:48:59 +0000 (09:48 +0200)]
fdc: force the fifo access to be in bounds of the allocated buffer

During processing of certain commands such as FD_CMD_READ_ID and
FD_CMD_DRIVE_SPECIFICATION_COMMAND the fifo memory access could
get out of bounds leading to memory corruption with values coming
from the guest.

Fix this by making sure that the index is always bounded by the
allocated memory.

This is CVE-2015-3456.

Signed-off-by: Petr Matousek <pmatouse@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Signed-off-by: John Snow <jsnow@redhat.com>
9 years agoMerge remote-tracking branch 'remotes/pmaydell/tags/pull-target-arm-20150512' into...
Peter Maydell [Tue, 12 May 2015 11:11:32 +0000 (12:11 +0100)]
Merge remote-tracking branch 'remotes/pmaydell/tags/pull-target-arm-20150512' into staging

target-arm queue:
 * Support TZ and grouping in the GIC
 * hw/sd: sd_reset cleanup
 * armv7m_nvic: fix bug in systick device

# gpg: Signature made Tue May 12 12:02:26 2015 BST using RSA key ID 14360CDE
# gpg: Good signature from "Peter Maydell <peter.maydell@linaro.org>"

* remotes/pmaydell/tags/pull-target-arm-20150512:
  hw/arm/highbank.c: Wire FIQ between CPU <> GIC
  hw/arm/vexpress.c: Wire FIQ between CPU <> GIC
  hw/arm/virt.c: Wire FIQ between CPU <> GIC
  hw/intc/arm_gic: Add grouping support to gic_update()
  hw/intc/arm_gic: Change behavior of IAR writes
  hw/intc/arm_gic: Change behavior of EOIR writes
  hw/intc/arm_gic: Handle grouping for GICC_HPPIR
  hw/intc/arm_gic: Restrict priority view
  hw/intc/arm_gic: Implement Non-secure view of RPR
  hw/intc/arm_gic: Make ICCICR/GICC_CTLR banked
  hw/intc/arm_gic: Make ICCBPR/GICC_BPR banked
  hw/intc/arm_gic: Make ICDDCR/GICD_CTLR banked
  hw/intc/arm_gic_kvm.c: Save and restore GICD_IGROUPRn state
  hw/intc/arm_gic: Add Interrupt Group Registers
  hw/intc/arm_gic: Switch to read/write callbacks with tx attributes
  hw/intc/arm_gic: Add Security Extensions property
  hw/intc/arm_gic: Create outbound FIQ lines
  hw/sd: Don't pass BlockBackend to sd_reset()
  armv7m_nvic: systick: Reload the RELOAD value and count down only if ENABLE bit is set

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agohw/arm/highbank.c: Wire FIQ between CPU <> GIC
Peter Maydell [Tue, 12 May 2015 10:57:19 +0000 (11:57 +0100)]
hw/arm/highbank.c: Wire FIQ between CPU <> GIC

Connect FIQ output of the GIC CPU interfaces to the CPUs.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1430502643-25909-18-git-send-email-peter.maydell@linaro.org

9 years agohw/arm/vexpress.c: Wire FIQ between CPU <> GIC
Fabian Aggeler [Tue, 12 May 2015 10:57:18 +0000 (11:57 +0100)]
hw/arm/vexpress.c: Wire FIQ between CPU <> GIC

Connect FIQ output of the GIC CPU interfaces to the CPUs.

Signed-off-by: Fabian Aggeler <aggelerf@ethz.ch>
Signed-off-by: Greg Bellows <greg.bellows@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1430502643-25909-17-git-send-email-peter.maydell@linaro.org
Message-id: 1429113742-8371-3-git-send-email-greg.bellows@linaro.org
[PMM: minor format tweak]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agohw/arm/virt.c: Wire FIQ between CPU <> GIC
Greg Bellows [Tue, 12 May 2015 10:57:18 +0000 (11:57 +0100)]
hw/arm/virt.c: Wire FIQ between CPU <> GIC

Connect FIQ output of the GIC CPU interfaces to the CPUs.

Signed-off-by: Greg Bellows <greg.bellows@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1430502643-25909-16-git-send-email-peter.maydell@linaro.org
Message-id: 1429113742-8371-4-git-send-email-greg.bellows@linaro.org
[PMM: minor format tweak]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agohw/intc/arm_gic: Add grouping support to gic_update()
Peter Maydell [Tue, 12 May 2015 10:57:18 +0000 (11:57 +0100)]
hw/intc/arm_gic: Add grouping support to gic_update()

Add support to gic_update() for determining the current IRQ
and FIQ status when interrupt grouping is supported. This
simply requires that instead of always raising IRQ we
check the group of the highest priority pending interrupt
and the GICC_CTLR.FIQEn bit to see whether we should raise
IRQ or FIQ.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Message-id: 1430502643-25909-15-git-send-email-peter.maydell@linaro.org

9 years agohw/intc/arm_gic: Change behavior of IAR writes
Fabian Aggeler [Tue, 12 May 2015 10:57:18 +0000 (11:57 +0100)]
hw/intc/arm_gic: Change behavior of IAR writes

Grouping (GICv2) and Security Extensions change the behavior of IAR
reads. Acknowledging Group0 interrupts is only allowed from Secure
state and acknowledging Group1 interrupts from Secure state is only
allowed if AckCtl bit is set.

Signed-off-by: Fabian Aggeler <aggelerf@ethz.ch>
Signed-off-by: Greg Bellows <greg.bellows@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1430502643-25909-14-git-send-email-peter.maydell@linaro.org
Message-id: 1429113742-8371-14-git-send-email-greg.bellows@linaro.org
[PMM: simplify significantly by reusing the existing
 gic_get_current_pending_irq() rather than reimplementing the
 same logic here]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agohw/intc/arm_gic: Change behavior of EOIR writes
Fabian Aggeler [Tue, 12 May 2015 10:57:18 +0000 (11:57 +0100)]
hw/intc/arm_gic: Change behavior of EOIR writes

Grouping (GICv2) and Security Extensions change the behavior of EOIR
writes. Completing Group0 interrupts is only allowed from Secure state.

Signed-off-by: Fabian Aggeler <aggelerf@ethz.ch>
Signed-off-by: Greg Bellows <greg.bellows@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1430502643-25909-13-git-send-email-peter.maydell@linaro.org
Message-id: 1429113742-8371-13-git-send-email-greg.bellows@linaro.org
[PMM: Rather than go to great lengths to ignore the UNPREDICTABLE case
 of a Secure EOI of a Group1 (NS) irq with AckCtl == 0, we just let
 it fall through; add a comment about it.]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agohw/intc/arm_gic: Handle grouping for GICC_HPPIR
Fabian Aggeler [Tue, 12 May 2015 10:57:18 +0000 (11:57 +0100)]
hw/intc/arm_gic: Handle grouping for GICC_HPPIR

Grouping (GICv2) and Security Extensions change the behaviour of reads
of the highest priority pending interrupt register (ICCHPIR/GICC_HPPIR).

Signed-off-by: Fabian Aggeler <aggelerf@ethz.ch>
Signed-off-by: Greg Bellows <greg.bellows@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1430502643-25909-12-git-send-email-peter.maydell@linaro.org
Message-id: 1429113742-8371-12-git-send-email-greg.bellows@linaro.org
[PMM: make utility fn static; coding style fixes; AckCtl has an effect
 for GICv2 without security extensions as well; removed checks on enable
 bits because these are done when we set current_pending[cpu]]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agohw/intc/arm_gic: Restrict priority view
Fabian Aggeler [Tue, 12 May 2015 10:57:17 +0000 (11:57 +0100)]
hw/intc/arm_gic: Restrict priority view

GICs with Security Extensions restrict the non-secure view of the
interrupt priority and priority mask registers.

Signed-off-by: Fabian Aggeler <aggelerf@ethz.ch>
Signed-off-by: Greg Bellows <greg.bellows@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1430502643-25909-11-git-send-email-peter.maydell@linaro.org
Message-id: 1429113742-8371-15-git-send-email-greg.bellows@linaro.org
[PMM: minor code tweaks; fixed missing masking in gic_set_priority_mask
and gic_set_priority]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agohw/intc/arm_gic: Implement Non-secure view of RPR
Fabian Aggeler [Tue, 12 May 2015 10:57:17 +0000 (11:57 +0100)]
hw/intc/arm_gic: Implement Non-secure view of RPR

For GICs with Security Extensions Non-secure reads have a restricted
view on the current running priority.

Signed-off-by: Fabian Aggeler <aggelerf@ethz.ch>
Signed-off-by: Greg Bellows <greg.bellows@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1430502643-25909-10-git-send-email-peter.maydell@linaro.org
Message-id: 1429113742-8371-11-git-send-email-greg.bellows@linaro.org
[PMM: make function static, minor comment tweak]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agohw/intc/arm_gic: Make ICCICR/GICC_CTLR banked
Fabian Aggeler [Tue, 12 May 2015 10:57:17 +0000 (11:57 +0100)]
hw/intc/arm_gic: Make ICCICR/GICC_CTLR banked

ICCICR/GICC_CTLR is banked in GICv1 implementations with Security
Extensions or in GICv2 in independent from Security Extensions.
This makes it possible to enable forwarding of interrupts from
the CPU interfaces to the connected processors for Group0 and Group1.

We also allow to set additional bits like AckCtl and FIQEn by changing
the type from bool to uint32. Since the field does not only store the
enable bit anymore and since we are touching the vmstate, we use the
opportunity to rename the field to cpu_ctlr.

Signed-off-by: Fabian Aggeler <aggelerf@ethz.ch>
Signed-off-by: Greg Bellows <greg.bellows@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1430502643-25909-9-git-send-email-peter.maydell@linaro.org
Message-id: 1429113742-8371-9-git-send-email-greg.bellows@linaro.org
[PMM: rewrote to store state in a single uint32_t rather than
 keeping the NS and S banked variants separate; this considerably
 simplifies the get/set functions]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agohw/intc/arm_gic: Make ICCBPR/GICC_BPR banked
Fabian Aggeler [Tue, 12 May 2015 10:57:17 +0000 (11:57 +0100)]
hw/intc/arm_gic: Make ICCBPR/GICC_BPR banked

This register is banked in GICs with Security Extensions. Storing the
non-secure copy of BPR in the abpr, which is an alias to the non-secure
copy for secure access. ABPR itself is only accessible from secure state
if the GIC implements Security Extensions.

Signed-off-by: Fabian Aggeler <aggelerf@ethz.ch>
Signed-off-by: Greg Bellows <greg.bellows@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1430502643-25909-8-git-send-email-peter.maydell@linaro.org
Message-id: 1429113742-8371-10-git-send-email-greg.bellows@linaro.org
[PMM: rewrote to fix style issues and correct handling of GICv2
 without security extensions]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agohw/intc/arm_gic: Make ICDDCR/GICD_CTLR banked
Fabian Aggeler [Tue, 12 May 2015 10:57:17 +0000 (11:57 +0100)]
hw/intc/arm_gic: Make ICDDCR/GICD_CTLR banked

ICDDCR/GICD_CTLR is banked if the GIC has the security extensions,
and the S (or only) copy has separate enable bits for Group0 and
Group1 enable if the GIC implements interrupt groups.

EnableGroup0 (Bit [1]) in GICv1 is architecturally IMPDEF. Since this
bit (Enable Non-secure) is present in the integrated GIC of the Cortex-A9
MPCore, we support this bit in our GICv1 implementation too.

Signed-off-by: Fabian Aggeler <aggelerf@ethz.ch>
Signed-off-by: Greg Bellows <greg.bellows@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1430502643-25909-7-git-send-email-peter.maydell@linaro.org
Message-id: 1429113742-8371-8-git-send-email-greg.bellows@linaro.org
[PMM: rewritten to store the state in a single s->ctlr uint32,
 with the NS register handled as an alias of bit 1 in that value;
 added vmstate version bump]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agohw/intc/arm_gic_kvm.c: Save and restore GICD_IGROUPRn state
Peter Maydell [Tue, 12 May 2015 10:57:17 +0000 (11:57 +0100)]
hw/intc/arm_gic_kvm.c: Save and restore GICD_IGROUPRn state

Now that the GIC base class has state fields for the GICD_IGROUPRn
registers, make kvm_arm_gic_get() and kvm_arm_gic_put() write and
read them. This allows us to remove the check that made us
fail migration if the guest had set any of the group register bits.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1430502643-25909-6-git-send-email-peter.maydell@linaro.org

9 years agohw/intc/arm_gic: Add Interrupt Group Registers
Fabian Aggeler [Tue, 12 May 2015 10:57:17 +0000 (11:57 +0100)]
hw/intc/arm_gic: Add Interrupt Group Registers

The Interrupt Group Registers allow the guest to configure interrupts
into one of two groups, where Group0 are higher priority and may
be routed to IRQ or FIQ, and Group1 are lower priority and always
routed to IRQ. (In a GIC with the security extensions Group0 is
Secure interrupts and Group 1 is NonSecure.)
The GICv2 always supports interrupt grouping; the GICv1 does only
if it implements the security extensions.

This patch implements the ability to read and write the registers;
the actual functionality the bits control will be added in a
subsequent patch.

Signed-off-by: Fabian Aggeler <aggelerf@ethz.ch>
Signed-off-by: Greg Bellows <greg.bellows@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1430502643-25909-5-git-send-email-peter.maydell@linaro.org
Message-id: 1429113742-8371-7-git-send-email-greg.bellows@linaro.org
[PMM: bring GIC_*_GROUP macros into line with the others, ie a
 simple SET/CLEAR/TEST rather than GROUP0/GROUP1;
 utility gic_has_groups() function;
 minor style fixes;
 bump vmstate version]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agohw/intc/arm_gic: Switch to read/write callbacks with tx attributes
Peter Maydell [Tue, 12 May 2015 10:57:16 +0000 (11:57 +0100)]
hw/intc/arm_gic: Switch to read/write callbacks with tx attributes

Switch the GIC's MMIO callback functions to the read_with_attrs
and write_with_attrs functions which provide MemTxAttrs. This will
allow the GIC to correctly handle secure and nonsecure register
accesses.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Message-id: 1430502643-25909-4-git-send-email-peter.maydell@linaro.org

9 years agohw/intc/arm_gic: Add Security Extensions property
Fabian Aggeler [Tue, 12 May 2015 10:57:16 +0000 (11:57 +0100)]
hw/intc/arm_gic: Add Security Extensions property

Add a QOM property which allows the GIC Security Extensions to be
enabled. These are an optional part of the GICv1 and GICv2 architecture.
This commit just adds the property and some sanity checks that it
is only enabled on GIC revisions that support it.

Signed-off-by: Fabian Aggeler <aggelerf@ethz.ch>
Signed-off-by: Greg Bellows <greg.bellows@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1430502643-25909-3-git-send-email-peter.maydell@linaro.org
Message-id: 1429113742-8371-5-git-send-email-greg.bellows@linaro.org
[PMM: changed property name, added checks that it isn't set for
 older GIC revisions or if using the KVM VGIC; reworded commit message]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agohw/intc/arm_gic: Create outbound FIQ lines
Fabian Aggeler [Tue, 12 May 2015 10:57:16 +0000 (11:57 +0100)]
hw/intc/arm_gic: Create outbound FIQ lines

Create the outbound FIQ lines from the GIC to the CPUs; these are
used if the GIC has security extensions or grouping support.

Signed-off-by: Fabian Aggeler <aggelerf@ethz.ch>
Signed-off-by: Greg Bellows <greg.bellows@linaro.org>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1430502643-25909-2-git-send-email-peter.maydell@linaro.org
Message-id: 1429113742-8371-2-git-send-email-greg.bellows@linaro.org
[PMM: added FIQ lines to kvm-arm-gic so its interface is the same;
tweaked commit message]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agohw/sd: Don't pass BlockBackend to sd_reset()
Peter Maydell [Tue, 12 May 2015 10:57:16 +0000 (11:57 +0100)]
hw/sd: Don't pass BlockBackend to sd_reset()

The only valid BlockBackend to pass to sd_reset() is the one for
the SD card, which is sd->blk. Drop the second argument from this
function in favour of having it just use sd->blk.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-id: 1430683444-9797-1-git-send-email-peter.maydell@linaro.org

9 years agoarmv7m_nvic: systick: Reload the RELOAD value and count down only if ENABLE bit is set
Adrian Huang [Tue, 12 May 2015 10:57:16 +0000 (11:57 +0100)]
armv7m_nvic: systick: Reload the RELOAD value and count down only if ENABLE bit is set

Consider the following pseudo code to configure SYSTICK (The
recommended programming sequence from "the definitive guide to the
arm cortex-m3"):
    SYSTICK Reload Value Register = 0xffff
    SYSTICK Current Value Register = 0
    SYSTICK Control and Status Register = 0x7

The pseudo code "SYSTICK Current Value Register = 0" leads to invoking
systick_reload(). As a consequence, the systick.tick member is updated
and the systick timer starts to count down when the ENABLE bit of
SYSTICK Control and Status Register is cleared.

The worst case is that: during the system initialization, the reset
value of the SYSTICK Control and Status Register is 0x00000000.
When the code "SYSTICK Current Value Register = 0" is executed, the
systick.tick member is accumulated with "(s->systick.reload + 1) *
systick_scale(s)". The systick_scale() gets the external_ref_clock
scale because the CLKSOURCE bit of the SYSTICK Control and Status
Register is cleared. This is the incorrect behavior because of the
code "SYSTICK Control and Status Register = 0x7". Actually, we want
the processor clock instead of the external reference clock.

This incorrect behavior defers the generation of the first interrupt.

The patch fixes the above-mentioned issue by setting the systick.tick
member and modifying the systick timer only if the ENABLE bit of
the SYSTICK Control and Status Register is set.

In addition, the Cortex-M3 Devices Generic User Guide mentioned that
"When ENABLE is set to 1, the counter loads the RELOAD value from the
SYST RVR register and then counts down". This patch adheres to the
statement of the user guide.

Signed-off-by: Adrian Huang <adrianhuang0701@gmail.com>
Reviewed-by: Jim Huang <jserv.tw@gmail.com>
[PMM: minor tweak to comment text]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoMerge remote-tracking branch 'remotes/stefanha/tags/net-pull-request' into staging
Peter Maydell [Tue, 12 May 2015 09:40:31 +0000 (10:40 +0100)]
Merge remote-tracking branch 'remotes/stefanha/tags/net-pull-request' into staging

# gpg: Signature made Mon May 11 16:25:58 2015 BST using RSA key ID 81AB73C8
# gpg: Good signature from "Stefan Hajnoczi <stefanha@redhat.com>"
# gpg:                 aka "Stefan Hajnoczi <stefanha@gmail.com>"

* remotes/stefanha/tags/net-pull-request:
  rocker: timestamp on the debug logs helps correlate with events in the VM
  MAINTAINERS: add rocker
  rocker: add tests
  rocker: add new rocker switch device
  pci: add network device class 'other' for network switches
  pci: add rocker device ID
  rocker: add register programming guide
  virtio-net: use qemu_mac_strdup_printf
  net: add MAC address string printer

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoMerge remote-tracking branch 'remotes/qmp-unstable/tags/for-upstream' into staging
Peter Maydell [Tue, 12 May 2015 08:01:51 +0000 (09:01 +0100)]
Merge remote-tracking branch 'remotes/qmp-unstable/tags/for-upstream' into staging

QMP pull request

# gpg: Signature made Mon May 11 14:15:19 2015 BST using RSA key ID E24ED5A7
# gpg: Good signature from "Luiz Capitulino <lcapitulino@gmail.com>"

* remotes/qmp-unstable/tags/for-upstream:
  scripts: qmp-shell: Add verbose flag
  scripts: qmp-shell: add transaction subshell
  scripts: qmp-shell: Expand support for QMP expressions
  scripts: qmp-shell: refactor helpers
  MAINTAINERS: New maintainer for QMP and QAPI
  json-parser: Accept 'null' in QMP
  qobject: Add a special null QObject
  qobject: Clean up around qtype_code
  QJSON: Use OBJECT_CHECK

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoMerge remote-tracking branch 'remotes/mst/tags/for_upstream' into staging
Peter Maydell [Mon, 11 May 2015 15:21:50 +0000 (16:21 +0100)]
Merge remote-tracking branch 'remotes/mst/tags/for_upstream' into staging

pc, virtio enhancements

Memory hot-unplug support for pc, MSI-X
mapping update speedup for virtio-pci,
misc refactorings and bugfixes.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
# gpg: Signature made Mon May 11 08:23:43 2015 BST using RSA key ID D28D5469
# gpg: Good signature from "Michael S. Tsirkin <mst@kernel.org>"
# gpg:                 aka "Michael S. Tsirkin <mst@redhat.com>"

* remotes/mst/tags/for_upstream: (28 commits)
  acpi: update expected files for memory unplug
  virtio-scsi: Move DEFINE_VIRTIO_SCSI_FEATURES to virtio-scsi
  virtio-net: Move DEFINE_VIRTIO_NET_FEATURES to virtio-net
  pci: Merge pci_nic_init() into pci_nic_init_nofail()
  acpi: add a missing backslash to the \_SB scope.
  qmp-event: add event notification for memory hot unplug error
  acpi: add hardware implementation for memory hot unplug
  acpi: fix "Memory device control fields" register
  acpi: extend aml_field() to support UpdateRule
  acpi, mem-hotplug: add unplug cb for memory device
  acpi, mem-hotplug: add unplug request cb for memory device
  acpi, mem-hotplug: add acpi_memory_slot_status() to get MemStatus
  docs: update documentation for memory hot unplug
  virtio: coding style tweak
  pci: remove hard-coded bar size in msix_init_exclusive_bar()
  virtio-pci: speedup MSI-X masking and unmasking
  virtio: introduce vector to virtqueues mapping
  virtio-ccw: using VIRTIO_NO_VECTOR instead of 0 for invalid virtqueue
  monitor: check return value of qemu_find_net_clients_except()
  monitor: replace the magic number 255 with MAX_QUEUE_NUM
  ...

Conflicts:
hw/s390x/s390-virtio-bus.c

[PMM: fixed conflict in s390_virtio_scsi_properties and
s390_virtio_net_properties arrays; since the result of the
two conflicting patches is to empty the property arrays
completely, the conflict resolution is to remove them entirely.]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoMerge remote-tracking branch 'remotes/bkoppelmann/tags/pull-tricore-20150511' into...
Peter Maydell [Mon, 11 May 2015 14:07:12 +0000 (15:07 +0100)]
Merge remote-tracking branch 'remotes/bkoppelmann/tags/pull-tricore-20150511' into staging

TriCore bugfixes

# gpg: Signature made Mon May 11 13:26:40 2015 BST using RSA key ID 6B69CA14
# gpg: Good signature from "Bastian Koppelmann <kbastian@mail.uni-paderborn.de>"

* remotes/bkoppelmann/tags/pull-tricore-20150511:
  target-tricore: fix rfe not restoring the PC
  target-tricore: fix rslcx restoring the upper context instead of the lower
  target-tricore: fix BO_OFF10_SEXT calculating the wrong offset
  target-tricore: fix SLR_LD_W and SLR_LD_W_POSTINC insn being a 2 byte memory access insted of 4
  target-tricore: Fix LOOP using wrong register for compare

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agorocker: timestamp on the debug logs helps correlate with events in the VM
David Ahern [Sat, 14 Mar 2015 04:09:33 +0000 (21:09 -0700)]
rocker: timestamp on the debug logs helps correlate with events in the VM

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: Scott Feldman <sfeldma@gmail.com>
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Message-id: 1426306173-24884-10-git-send-email-sfeldma@gmail.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoMAINTAINERS: add rocker
Scott Feldman [Sat, 14 Mar 2015 04:09:32 +0000 (21:09 -0700)]
MAINTAINERS: add rocker

Signed-off-by: Scott Feldman <sfeldma@gmail.com>
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Message-id: 1426306173-24884-9-git-send-email-sfeldma@gmail.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agorocker: add tests
Scott Feldman [Sat, 14 Mar 2015 04:09:31 +0000 (21:09 -0700)]
rocker: add tests

Add some basic test for rocker to test L2/L3/L4 functionality.  Requires an
external test environment, simp, located here:

https://github.com/scottfeldman/simp

To run tests, simp environment must be installed and a suitable VM image built
and installed with a Linux 3.18 (or greater) kernel with rocker driver support
enabled.

Signed-off-by: Scott Feldman <sfeldma@gmail.com>
Message-id: 1426306173-24884-8-git-send-email-sfeldma@gmail.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agorocker: add new rocker switch device
Scott Feldman [Sat, 14 Mar 2015 04:09:30 +0000 (21:09 -0700)]
rocker: add new rocker switch device

Rocker is a simulated ethernet switch device.  The device supports up to 62
front-panel ports and supports L2 switching and L3 routing functions, as well
as L2/L3/L4 ACLs.  The device presents a single PCI device for each switch,
with a memory-mapped register space for device driver access.

Rocker device is invoked with -device, for example a 4-port switch:

  -device rocker,name=sw1,len-ports=4,ports[0]=dev0,ports[1]=dev1, \
         ports[2]=dev2,ports[3]=dev3

Each port is a netdev and can be paired with using -netdev id=<port name>.

Signed-off-by: Scott Feldman <sfeldma@gmail.com>
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Acked-by: Scott Feldman <sfeldma@gmail.com>
Acked-by: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: David Ahern <dsahern@gmail.com>
Message-id: 1426306173-24884-7-git-send-email-sfeldma@gmail.com

rocker: fix clang compiler errors

Consolidate all forward typedef declarations to rocker.h.

Signed-off-by: David Ahern <dsahern@gmail.com>
Acked-by: Scott Feldman <sfeldma@gmail.com>
Acked-by: Jiri Pirko <jiri@resnulli.us>
rocker: add support for flow modification

We had support for flow add/del.  This adds support for flow mod.  I needed
this for L3 support where an existing route is modified using NLM_F_REPLACE.
For example:

  ip route add 12.0.0.0/30 nexthop via 11.0.0.1 dev swp1
  ip route change 12.0.0.0/30 nexthop via 11.0.0.9 dev swp2

The first cmd adds the route.  The second cmd changes the existing route by
changing its nexthop info.

In the device, a mod operation results in the matching flow enty being modified
with the new settings.  This is atomic to the device.

Signed-off-by: Scott Feldman <sfeldma@gmail.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agopci: add network device class 'other' for network switches
Scott Feldman [Sat, 14 Mar 2015 04:09:29 +0000 (21:09 -0700)]
pci: add network device class 'other' for network switches

Rocker is an ethernet switch device, so add 'other' network device class as
defined by PCI to cover these types of devices.

Signed-off-by: Scott Feldman <sfeldma@gmail.com>
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Message-id: 1426306173-24884-6-git-send-email-sfeldma@gmail.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agopci: add rocker device ID
Scott Feldman [Sat, 14 Mar 2015 04:09:28 +0000 (21:09 -0700)]
pci: add rocker device ID

Signed-off-by: Scott Feldman <sfeldma@gmail.com>
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Message-id: 1426306173-24884-5-git-send-email-sfeldma@gmail.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agorocker: add register programming guide
Scott Feldman [Sat, 14 Mar 2015 04:09:27 +0000 (21:09 -0700)]
rocker: add register programming guide

This is the register programming guide for the Rocker device.  It's intended
for driver writers and device writers.  It covers the device's PCI space,
the register set, DMA interface, and interrupts.

Signed-off-by: Scott Feldman <sfeldma@gmail.com>
Signed-off-by: Jiri Pirko <jiri@resnulli.us>
Message-id: 1426306173-24884-4-git-send-email-sfeldma@gmail.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agovirtio-net: use qemu_mac_strdup_printf
Scott Feldman [Sat, 14 Mar 2015 04:09:26 +0000 (21:09 -0700)]
virtio-net: use qemu_mac_strdup_printf

Signed-off-by: Scott Feldman <sfeldma@gmail.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Message-id: 1426306173-24884-3-git-send-email-sfeldma@gmail.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agonet: add MAC address string printer
Scott Feldman [Sat, 14 Mar 2015 04:09:25 +0000 (21:09 -0700)]
net: add MAC address string printer

We can use this in virtio-net code as well as new Rocker driver code, so
up-level this.

Signed-off-by: Scott Feldman <sfeldma@gmail.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Message-id: 1426306173-24884-2-git-send-email-sfeldma@gmail.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
9 years agoscripts: qmp-shell: Add verbose flag
John Snow [Wed, 29 Apr 2015 19:14:04 +0000 (15:14 -0400)]
scripts: qmp-shell: Add verbose flag

Add a verbose flag that shows the QMP command that was
constructed, to allow for later copy/pasting, reference,
debugging, etc.

The QMP is converted from a Python literal to JSON first,
to ensure that it is viable input to the actual QMP parser.

As a side-effect, this JSON output will helpfully show all
the necessary conversions that were performed on the input,
illustrating that "True" was transformed back into "true",
literal values are now escaped with "" instead of '', and so on.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Tested-by: Kashyap Chamarthy <kchamart@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
9 years agoscripts: qmp-shell: add transaction subshell
John Snow [Wed, 29 Apr 2015 19:14:03 +0000 (15:14 -0400)]
scripts: qmp-shell: add transaction subshell

Add a special processing mode to craft transactions.

By entering "transaction(" the shell will enter a special
mode where each subsequent command will be saved as a transaction
instead of executed as an individual command.

The transaction can be submitted by entering ")" on a line by itself.

Examples:

Separate lines:

(QEMU) transaction(
TRANS> block-dirty-bitmap-add node=drive0 name=bitmap1
TRANS> block-dirty-bitmap-clear node=drive0 name=bitmap0
TRANS> )

With a transaction action included on the first line:

(QEMU) transaction( block-dirty-bitmap-add node=drive0 name=bitmap2
TRANS> block-dirty-bitmap-add node=drive0 name=bitmap3
TRANS> )

As a one-liner, with just one transaction action:

(QEMU) transaction( block-dirty-bitmap-add node=drive0 name=bitmap0 )

As a side-effect of this patch, blank lines are now parsed as no-ops,
regardless of which shell mode you are in.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Tested-by: Kashyap Chamarthy <kchamart@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
9 years agoscripts: qmp-shell: Expand support for QMP expressions
John Snow [Wed, 29 Apr 2015 19:14:02 +0000 (15:14 -0400)]
scripts: qmp-shell: Expand support for QMP expressions

This includes support for [] expressions, single-quotes in
QMP expressions (which is not strictly a part of JSON), and
the ability to use "True", "False" and "None" literals instead
of JSON's equivalent true, false, and null literals.

qmp-shell currently allows you to describe values as
JSON expressions:
key={"key":{"key2":"val"}}

But it does not currently support arrays, which are needed
for serializing and deserializing transactions:
key=[{"type":"drive-backup","data":{...}}]

qmp-shell also only currently accepts doubly quoted strings
as-per JSON spec, but QMP allows single quotes.

Lastly, python allows you to utilize "True" or "False" as
boolean literals, but JSON expects "true" or "false". Expand
qmp-shell to allow the user to type either, converting to the
correct type.

As a consequence of the above, the key=val parsing is also improved
to give better error messages if a key=val token is not provided.

CAVEAT: The parser is still extremely rudimentary and does not
expect to find spaces in {} nor [] expressions. This patch does
not improve this functionality.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Tested-by: Kashyap Chamarthy <kchamart@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
9 years agoscripts: qmp-shell: refactor helpers
John Snow [Wed, 29 Apr 2015 19:14:01 +0000 (15:14 -0400)]
scripts: qmp-shell: refactor helpers

Refactor the qmp-shell command line processing function
into two components. This will be used to allow sub-expressions,
which will assist us in adding transactional support to qmp-shell.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Tested-by: Kashyap Chamarthy <kchamart@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
9 years agoMAINTAINERS: New maintainer for QMP and QAPI
Luiz Capitulino [Tue, 5 May 2015 14:39:15 +0000 (10:39 -0400)]
MAINTAINERS: New maintainer for QMP and QAPI

Markus is taking over maintership of QMP and the QAPI from
me. Markus has always been a great reviewer and contributor
to those subsystems. In the last few months he's also doing
pull requests that are a lot more relevant than the ones I
was able to do. So, this is a natural move.

I'm still the maintainer of HMP and QObjects, but I'm
looking for someone to take over those too.

PS: This commit also fixes the file listing for the QMP
    entry.

Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
Reviewed-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
9 years agojson-parser: Accept 'null' in QMP
Eric Blake [Wed, 29 Apr 2015 21:35:06 +0000 (15:35 -0600)]
json-parser: Accept 'null' in QMP

We document that in QMP, the client may send any json-value
for the optional "id" key, and then return that same value
on reply (both success and failures, insofar as the failure
happened after parsing the id).  [Note that the output may
not be identical to the input, as whitespace may change and
since we may reorder keys within a json-object, but that this
still constitutes the same json-value].  However, we were not
handling the JSON literal null, which counts as a json-value
per RFC 7159.

Also, down the road, given the QAPI schema of {'*foo':'str'} or
{'*foo':'ComplexType'}, we could decide to allow the QMP client
to pass { "foo":null } instead of the current representation of
{ } where omitting the key is the only way to get at the default
NULL value.  Such a change might be useful for argument
introspection (if a type in older qemu lacks 'foo' altogether,
then an explicit "foo":null probe will force an easily
distinguished error message for whether the optional "foo" key
is even understood in newer qemu).  And if we add default values
to optional arguments, allowing an explicit null would be
required for getting a NULL value associated with an optional
string that has a non-null default.  But all that can come at a
later day.

The 'check-unit' testsuite is enhanced to test that parsing
produces the same object as explicitly requesting a reference
to the special qnull object.  In addition, I tested with:

$ ./x86_64-softmmu/qemu-system-x86_64 -qmp stdio -nodefaults
{"QMP": {"version": {"qemu": {"micro": 91, "minor": 2, "major": 2}, "package": ""}, "capabilities": []}}
{"execute":"qmp_capabilities","id":null}
{"return": {}, "id": null}
{"id":{"a":null,"b":[1,null]},"execute":"quit"}
{"return": {}, "id": {"a": null, "b": [1, null]}}
{"timestamp": {"seconds": 1427742379, "microseconds": 423128}, "event": "SHUTDOWN"}

Signed-off-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
9 years agoqobject: Add a special null QObject
Markus Armbruster [Wed, 29 Apr 2015 21:35:05 +0000 (15:35 -0600)]
qobject: Add a special null QObject

I'm going to fix the JSON parser to recognize null.  The obvious
representation of JSON null as (QObject *)NULL doesn't work, because
the parser already uses it as an error value.  Perhaps we should
change it to free NULL for null, but that's more than I can do right
now.  Create a special null QObject instead.

The existing QDict, QList, and QString all represent something that
is a pointer in C and could therefore be associated with NULL.  But
right now, all three of these sub-types are always non-null once
created, so the new null sentinel object is intentionally unrelated
to them.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
9 years agoqobject: Clean up around qtype_code
Markus Armbruster [Wed, 29 Apr 2015 21:35:04 +0000 (15:35 -0600)]
qobject: Clean up around qtype_code

QTYPE_NONE is a sentinel value.  No QObject has this type code.
Document it properly.

Fix dump_qobject() to abort() on QTYPE_NONE, just like for any other
invalid type code.

Fix to_json() to abort() on all invalid type codes, not just
QTYPE_MAX.

Clean up Property member qtype's type: it's a qtype_code.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
9 years agoQJSON: Use OBJECT_CHECK
Eduardo Habkost [Sat, 25 Apr 2015 15:28:06 +0000 (12:28 -0300)]
QJSON: Use OBJECT_CHECK

The QJSON code used casts to (QJSON*) directly, instead of OBJECT_CHECK.
There were even some functions using object_dynamic_cast() calls
followed by assert(), which is exactly what OBJECT_CHECK does (by
calling object_dynamic_cast_assert()).

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
9 years agoMerge remote-tracking branch 'remotes/mjt/tags/pull-trivial-patches-2015-05-09' into...
Peter Maydell [Mon, 11 May 2015 12:54:00 +0000 (13:54 +0100)]
Merge remote-tracking branch 'remotes/mjt/tags/pull-trivial-patches-2015-05-09' into staging

trivial patches for 2015-05-09

# gpg: Signature made Fri May  8 22:58:42 2015 BST using RSA key ID A4C3D7DB
# gpg: Good signature from "Michael Tokarev <mjt@tls.msk.ru>"
# gpg:                 aka "Michael Tokarev <mjt@corpit.ru>"
# gpg:                 aka "Michael Tokarev <mjt@debian.org>"

* remotes/mjt/tags/pull-trivial-patches-2015-05-09:
  docs: update BLOCK_IMAGE_CORRUPTED documentation
  glib-compat.h: change assert to g_assert
  Remove various unused functions
  sheepdog: fix resource leak with sd_snapshot_create
  xhci: remove unused code

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agotarget-tricore: fix rfe not restoring the PC
Bastian Koppelmann [Tue, 5 May 2015 17:41:10 +0000 (19:41 +0200)]
target-tricore: fix rfe not restoring the PC

Signed-off-by: Bastian Koppelmann <kbastian@mail.uni-paderborn.de>
9 years agotarget-tricore: fix rslcx restoring the upper context instead of the lower
Bastian Koppelmann [Tue, 5 May 2015 17:39:18 +0000 (19:39 +0200)]
target-tricore: fix rslcx restoring the upper context instead of the lower

Signed-off-by: Bastian Koppelmann <kbastian@mail.uni-paderborn.de>
9 years agotarget-tricore: fix BO_OFF10_SEXT calculating the wrong offset
Bastian Koppelmann [Tue, 5 May 2015 17:36:55 +0000 (19:36 +0200)]
target-tricore: fix BO_OFF10_SEXT calculating the wrong offset

The lower part of the combined offset was sign extended and could lead to
wrong results.

Signed-off-by: Bastian Koppelmann <kbastian@mail.uni-paderborn.de>
9 years agotarget-tricore: fix SLR_LD_W and SLR_LD_W_POSTINC insn being a 2 byte memory access...
Bastian Koppelmann [Fri, 3 Apr 2015 12:29:22 +0000 (14:29 +0200)]
target-tricore: fix SLR_LD_W and SLR_LD_W_POSTINC insn being a 2 byte memory access insted of 4

Signed-off-by: Bastian Koppelmann <kbastian@mail.uni-paderborn.de>
9 years agotarget-tricore: Fix LOOP using wrong register for compare
Bastian Koppelmann [Tue, 9 Dec 2014 16:04:46 +0000 (16:04 +0000)]
target-tricore: Fix LOOP using wrong register for compare

Signed-off-by: Bastian Koppelmann <kbastian@mail.uni-paderborn.de>
9 years agoMerge remote-tracking branch 'remotes/bonzini/tags/for-upstream' into staging
Peter Maydell [Mon, 11 May 2015 11:01:09 +0000 (12:01 +0100)]
Merge remote-tracking branch 'remotes/bonzini/tags/for-upstream' into staging

- build bugfix from Fam and new configure check from Emilio
- two improvements to "info mtere" from Gerd
- KVM support for memory transaction attributes
- one more small step towards unlocked MMIO dispatch
- one piece of the qemu-nbd errno fixes
- trivial-ish patches from Denis and Thomas

# gpg: Signature made Fri May  8 13:47:29 2015 BST using RSA key ID 78C7AE83
# gpg: Good signature from "Paolo Bonzini <bonzini@gnu.org>"
# gpg:                 aka "Paolo Bonzini <pbonzini@redhat.com>"
# gpg: WARNING: This key is not certified with sufficiently trusted signatures!
# gpg:          It is not certain that the signature belongs to the owner.
# Primary key fingerprint: 46F5 9FBD 57D6 12E7 BFD4  E2F7 7E15 100C CD36 69B1
#      Subkey fingerprint: F133 3857 4B66 2389 866C  7682 BFFB D25F 78C7 AE83

* remotes/bonzini/tags/for-upstream:
  qemu-nbd: only send a limited number of errno codes on the wire
  rules.mak: Force CFLAGS for all objects in DSO
  configure: require __thread support
  exec: move rcu_read_lock/unlock to address_space_translate callers
  kvm: add support for memory transaction attributes
  mtree: also print disabled regions
  mtree: tag & indent a bit better
  apic_common: improve readability of apic_reset_common
  kvm: Silence warning from valgrind

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoMerge remote-tracking branch 'remotes/kraxel/tags/pull-gtk-20150508-1' into staging
Peter Maydell [Mon, 11 May 2015 09:43:08 +0000 (10:43 +0100)]
Merge remote-tracking branch 'remotes/kraxel/tags/pull-gtk-20150508-1' into staging

gtk: add ui_info support, cleanups + fixes.

# gpg: Signature made Fri May  8 12:47:04 2015 BST using RSA key ID D3E87138
# gpg: Good signature from "Gerd Hoffmann (work) <kraxel@redhat.com>"
# gpg:                 aka "Gerd Hoffmann <gerd@kraxel.org>"
# gpg:                 aka "Gerd Hoffmann (private) <kraxel@gmail.com>"

* remotes/kraxel/tags/pull-gtk-20150508-1:
  gtk: update mouse position in mouse_set()
  gtk: create gtk.h
  gtk: add ui_info support
  console: add dpy_ui_info_supported
  console: delayed ui_info guest notification

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoMerge remote-tracking branch 'remotes/kraxel/tags/pull-usb-20150508-1' into staging
Peter Maydell [Mon, 11 May 2015 08:42:20 +0000 (09:42 +0100)]
Merge remote-tracking branch 'remotes/kraxel/tags/pull-usb-20150508-1' into staging

usb: qomify, bugfixes for xhci & uhci.

# gpg: Signature made Fri May  8 12:39:28 2015 BST using RSA key ID D3E87138
# gpg: Good signature from "Gerd Hoffmann (work) <kraxel@redhat.com>"
# gpg:                 aka "Gerd Hoffmann <gerd@kraxel.org>"
# gpg:                 aka "Gerd Hoffmann (private) <kraxel@gmail.com>"

* remotes/kraxel/tags/pull-usb-20150508-1:
  uhci: controller is halted after reset
  usb: usb-serial QOMify
  usb: usb-redir QOMify
  usb: usb-wacom-tablet QOMify
  usb: usb-uas QOMify
  usb: usb-storage QOMify
  usb: usb-ccid QOMify
  usb: usb-net QOMify
  usb-mtp: fix segmentation fault
  usb: usb-mtp QOMify
  usb: usb-hub QOMify
  usb: usb-hid QOMify
  usb: usb-bt QOMify
  usb: usb-audio QOMify
  uhci: QOMify
  xhci: fix events for setup trb.
  Revert "xhci: generate a Transfer Event for each Transfer TRB with the IOC bit set"
  xhci: set timer to retry xfers
  usb: fix usb-net segfault

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoacpi: update expected files for memory unplug
Michael S. Tsirkin [Mon, 11 May 2015 07:21:37 +0000 (09:21 +0200)]
acpi: update expected files for memory unplug

commit c06b2ffb02bfcc642c67300d2c4dffd5aa54932b
    acpi: add hardware implementation for memory hot unplug

Changed both the DSDT and the SSDT. Update the expected files
accordingly.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
9 years agoMerge remote-tracking branch 'remotes/cohuck/tags/s390x-20150508' into staging
Peter Maydell [Sun, 10 May 2015 20:40:54 +0000 (21:40 +0100)]
Merge remote-tracking branch 'remotes/cohuck/tags/s390x-20150508' into staging

Assorted s390x patches:
- updates for virtio-ccw and s390-virtio, making them more similar
  to virtio-pci
- improvements regarding per-vcpu interrupts and migration

# gpg: Signature made Fri May  8 09:45:09 2015 BST using RSA key ID C6F02FAF
# gpg: Good signature from "Cornelia Huck <huckc@linux.vnet.ibm.com>"
# gpg:                 aka "Cornelia Huck <cornelia.huck@de.ibm.com>"

* remotes/cohuck/tags/s390x-20150508:
  s390x/kvm: migrate vcpu interrupt state
  s390x: move fpu regs into a subsection of the vmstate
  s390x/kvm: use ioctl KVM_S390_IRQ for vcpu interrupts
  virtio-ccw: implement ->device_plugged
  virtio-ccw: change realization sequence
  s390-virtio: clear {used,avail}_event_idx on reset as well
  s390-virtio: use common features
  s390-virtio: Accommodate guests using virtqueues too early

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
9 years agoqemu-nbd: only send a limited number of errno codes on the wire
Paolo Bonzini [Thu, 7 May 2015 15:25:10 +0000 (17:25 +0200)]
qemu-nbd: only send a limited number of errno codes on the wire

Right now, NBD includes potentially platform-specific error values in
the wire protocol.

Luckily, most common error values are more or less universal: in
particular, of all errno values <= 34 (up to ERANGE), they are all the
same on supported platforms except for 11 (which is EAGAIN on Windows and
Linux, but EDEADLK on Darwin and the *BSDs).  So, in order to guarantee
some portability, only keep a handful of possible error codes and squash
everything else to EINVAL.

This patch defines a limited set of errno values that are valid for the
NBD protocol, and specifies recommendations for what error to return
in specific corner cases.  The set of errno values is roughly based on
the errors listed in the read(2) and write(2) man pages, with some
exceptions:

- ENOMEM is added for servers that implement copy-on-write or other
  formats that require dynamic allocation.

- EDQUOT is not part of the universal set of errors; it can be changed
  to ENOSPC on the wire format.

- EFBIG is part of the universal set of errors, but it is also changed
  to ENOSPC because it is pretty similar to ENOSPC or EDQUOT.

Incoming values will in general match system errno values, but not
on the Hurd which has different errno values (they have a "subsystem
code" equal to 0x10 in bits 24-31).  The Hurd is probably not something
to which QEMU has been ported, but still do the right thing and
reverse-map the NBD errno values to the system errno values.

The corresponding patch to the NBD protocol description can be found at
http://article.gmane.org/gmane.linux.drivers.nbd.general/3154.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agorules.mak: Force CFLAGS for all objects in DSO
Fam Zheng [Thu, 7 May 2015 06:55:15 +0000 (14:55 +0800)]
rules.mak: Force CFLAGS for all objects in DSO

Because of the trick of process-archive-undefs, all .mo objects, even
with --enable-modules, are dependencies of executables.

This breaks CFLAGS propogation because the compiling of module object
will happen too early before building for DSO.

With GCC 5, the linking would fail because .o doesn't have -fPIC. Also,
BUILD_DSO will be missed. (module-common.o will have it, so the stamp
symbol was still liked in .so).

Fix the problem by forcing the CFLAGS on individual .o-cflags during
unnest-vars.

Reported-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Fam Zheng <famz@redhat.com>
Cc: qemu-stable@nongnu.org # 2.3
Message-Id: <1430981715-31465-1-git-send-email-famz@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
9 years agodocs: update BLOCK_IMAGE_CORRUPTED documentation
Alberto Garcia [Thu, 7 May 2015 14:58:26 +0000 (17:58 +0300)]
docs: update BLOCK_IMAGE_CORRUPTED documentation

Label the "size" and "offset" fields in BLOCK_IMAGE_CORRUPTED as
optional, and clarify that the latter refers to the host's offset into
the image.

Signed-off-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
9 years agoglib-compat.h: change assert to g_assert
Michael Tokarev [Thu, 7 May 2015 10:38:02 +0000 (13:38 +0300)]
glib-compat.h: change assert to g_assert

include/glib-compat.h defines a bunch of functions based on glib primitives,
and uses assert() without including assert.h.  Replace assert() with
g_assert() to make the file more self-contained, and to fix compilation
breakage after 28507a415a9b1e.

Reported-by: Laurent Desnogues <laurent.desnogues@gmail.com>
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
Tested-by: Laurent Desnogues <laurent.desnogues@gmail.com>
9 years agoRemove various unused functions
Thomas Huth [Sun, 3 May 2015 08:47:22 +0000 (10:47 +0200)]
Remove various unused functions

The functions tpm_backend_thread_tpm_reset() and iothread_find()
are completely unused, let's remove them.

Signed-off-by: Thomas Huth <huth@tuxfamily.org>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
9 years agosheepdog: fix resource leak with sd_snapshot_create
zhanghailiang [Tue, 5 May 2015 01:48:03 +0000 (09:48 +0800)]
sheepdog: fix resource leak with sd_snapshot_create

Signed-off-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>