Increase Virtio-net vring size #1

The test_path binary is (unlike the other test binaries in tests/tcg) actually intended to be compiled with the same compiler used to build the main QEMU executables. It actually #includes a number of the QEMU source files in an attempt to unit-test the util/path.c functions, and so if it is not compiled with the same compiler used by configure to set CONFIG_ settings then it is liable to fail to build. Fix the makefile to build it with the default C compiler rules, not CC_I386, and fix the test itself not to include a lot of unnecessary trace related source files which cause the build to fail if the trace backend is anything other than 'simple'. Signed-off-by: Peter Maydell <[email protected]> Signed-off-by: Michael Tokarev <[email protected]>

When DPRINTF() has effect, the original author wants to print all ram_load() calling results. So need use 'goto' instead of 'return' within ram_load(), just like other areas have done. Signed-off-by: Chen Gang <[email protected]> Signed-off-by: Michael Tokarev <[email protected]>

Signed-off-by: Stefan Weil <[email protected]> Signed-off-by: Michael Tokarev <[email protected]>

This fixes a warning from the static code analysis (smatch). Signed-off-by: Stefan Weil <[email protected]> Signed-off-by: Michael Tokarev <[email protected]> Acked-by: Aurelien Jarno <[email protected]>

Signed-off-by: Stefan Weil <[email protected]> Signed-off-by: Michael Tokarev <[email protected]>

This fixes a warning from the static code analysis (smatch). Signed-off-by: Stefan Weil <[email protected]> Signed-off-by: Michael Tokarev <[email protected]>

There was already a forward declaration using 'static', but the attribute was missing in the implementation. This fixes a warning from the static code analysis (smatch). Signed-off-by: Stefan Weil <[email protected]> Signed-off-by: Michael Tokarev <[email protected]>

This fixes a warning from the static code analysis (smatch). Signed-off-by: Stefan Weil <[email protected]> Signed-off-by: Michael Tokarev <[email protected]>

This fixes warnings from the static code analysis (smatch). Signed-off-by: Stefan Weil <[email protected]> Signed-off-by: Michael Tokarev <[email protected]>

Reviewed-by: Stefan Weil <[email protected]> Signed-off-by: Jim Meyering <[email protected]> Acked-by: Stefano Stabellini <[email protected]> Signed-off-by: Michael Tokarev <[email protected]>

Signed-off-by: Stefan Weil <[email protected]> Signed-off-by: Michael Tokarev <[email protected]>

glib >= 2.31 always enables thread support and g_thread_supported() is #defined to 1, there's no need to call g_thread_init() anymore, and it definitely does not need to report error which never happens. Keep code for old < 2.31 glibc anyway for now, just #ifdef it differently. Signed-off-by: Michael Tokarev <[email protected]> Reviewed-by: Stefan Hajnoczi <[email protected]> Cc: [email protected]

We have a dedicated header file for wrappers to smooth over glib version differences. Move the g_poll() definition into glib-compat.h for consistency. Signed-off-by: Stefan Hajnoczi <[email protected]> Signed-off-by: Michael Tokarev <[email protected]> Cc: [email protected]

Commit 2e67978 replaced strncpy() with pstrcpy() in one place in libcacard. This is a qemu-specific function, while libcacard is a stand-alone library (or tries to be). But since we know the exact length of the string to copy, and know that it definitely will fit in the destination buffer, use memcpy() instead, and null-terminate the string after that. An alternative is to use g_strlcpy() or strncpy(), but memcpy() is more than adequate in this place. Signed-off-by: Michael Tokarev <[email protected]> Cc: [email protected] Cc: Alon Levy <[email protected]>

Without this, ambiguous path is reported to the user as "not found", which is confusing at least. Signed-off-by: Michael Tokarev <[email protected]>

Signed-off-by: Michael Tokarev <[email protected]> Reviewed-by: Andreas Färber <[email protected]>

Signed-off-by: Hani Benhabiles <[email protected]> Signed-off-by: Michael Tokarev <[email protected]>

.impl.valid should be .impl.unaligned and the description needs some fixes. .old_portio is removed since commit b40acf9 (ioport: Switch dispatching to memory core layer). Signed-off-by: Fam Zheng <[email protected]> Signed-off-by: Michael Tokarev <[email protected]>

Signed-off-by: Alon Levy <[email protected]> Signed-off-by: Michael Tokarev <[email protected]>

Makes it easier to ensure proper naming across the different frontends and backends. Signed-off-by: Lluís Vilanova <[email protected]> Signed-off-by: Stefan Hajnoczi <[email protected]>

Signed-off-by: Lluís Vilanova <[email protected]> Signed-off-by: Stefan Hajnoczi <[email protected]>

The following tracetool cleanup changes the event numbering policy. Signed-off-by: Lluís Vilanova <[email protected]> Signed-off-by: Stefan Hajnoczi <[email protected]>

Backends now only contain the essential backend-specific code, and most of the work is moved to frontend code. Signed-off-by: Lluís Vilanova <[email protected]> Signed-off-by: Stefan Hajnoczi <[email protected]>

It is only used with the simple trace backend. Signed-off-by: Stefan Weil <[email protected]> Signed-off-by: Stefan Hajnoczi <[email protected]>

…staging * remotes/sstabellini/xen-140507-2: xen_disk: add discard support pass an inclusive address range to xc_domain_pin_memory_cacheattr xen: factor out common functions xen: move Xen HVM files under hw/i386/xen xen: move Xen PV machine files to hw/xenpv qemu-xen: free all the pirqs for msi/msix when driver unload exec: Limit translation limiting in address_space_translate to xen Signed-off-by: Peter Maydell <[email protected]>

…5-07' into staging trivial patches for 2014-05-07 # gpg: Signature made Wed 07 May 2014 18:01:15 BST using RSA key ID A4C3D7DB # gpg: Good signature from "Michael Tokarev <[email protected]>" # gpg: aka "Michael Tokarev <[email protected]>" # gpg: aka "Michael Tokarev <[email protected]>" # gpg: WARNING: This key is not certified with a trusted signature! # gpg: There is no indication that the signature belongs to the owner. # Primary key fingerprint: 6EE1 95D1 886E 8FFB 810D 4324 457C E0A0 8044 65C5 # Subkey fingerprint: 6F67 E18E 7C91 C5B1 5514 66A7 BEE5 9D74 A4C3 D7DB * remotes/mjt/tags/trivial-patches-2014-05-07: (21 commits) libcacard: remove unnecessary EOL from debug prints docs/memory.txt: Fix document on MMIO operations readline: Sort completions before printing them. readline: use g_strndup instead of open-coding it qmp: report path ambiguity error libcacard: replace pstrcpy() with memcpy() glib: move g_poll() replacement into glib-compat.h do not call g_thread_init() for glib >= 2.31 hw/9pfs: Add include file for exported symbol xen: remove unused global, xen_xcg hw: Add missing 'static' attributes qemu-timer: Add missing 'static' attribute ui: Add missing 'static' attribute monitor: Add missing 'static' attribute hw/s390x: Add missing 'static' attribute hw/mips: Add missing 'static' and 'const' attributes hw/9pfs: Add missing 'static' attributes arch_init: Be sure of only one exit entry with DPRINTF() for ram_load() tests/tcg: Fix compilation of test_path qga: Fix typo (plural) in comment ... Signed-off-by: Peter Maydell <[email protected]>

In case no listening address was specified, "info spice" reports "0.0.0.0" as address. Which is incorrect in case spice is listening on ipv6. Replace it by a wildcard "*" to indicate it is not limited to a specific address. Note: Being more specific is not possible without extending the spice-server api. The socket is handled by spice-server not qemu, so qemu can't easily figure the actual socket address. Reported-by: David Jaša <[email protected]> Signed-off-by: Gerd Hoffmann <[email protected]>

Only notify spice-server about migration events in case we got target host information beforehand. So we kick the seamless spice client migration only in case a actual live migration happens, not when libvirt uses live-migration-to-file for snapshotting. Signed-off-by: Gerd Hoffmann <[email protected]>

…taging spice: small fixes # gpg: Signature made Thu 08 May 2014 09:50:21 BST using RSA key ID D3E87138 # gpg: Good signature from "Gerd Hoffmann (work) <[email protected]>" # gpg: aka "Gerd Hoffmann <[email protected]>" # gpg: aka "Gerd Hoffmann (private) <[email protected]>" * remotes/spice/tags/pull-spice-7: spice: fix libvirt snapshots spice: fix "info spice" Signed-off-by: Peter Maydell <[email protected]>

…est' into staging Tracing pull request # gpg: Signature made Wed 07 May 2014 18:14:02 BST using RSA key ID 81AB73C8 # gpg: Good signature from "Stefan Hajnoczi <[email protected]>" # gpg: aka "Stefan Hajnoczi <[email protected]>" * remotes/stefanha/tags/tracing-pull-request: configure: Show trace output file conditionally trace: [tracetool] Minimize the amount of per-backend code trace: [simple] Bump up log version number trace: [tracetool] Change format docs to point to the generated file trace: [tracetool] Show list of frontends and backends sorted by name trace: [tracetool] Cosmetic changes trace: [tracetool] Spacing changes trace: [tracetool] Add methods 'Event.copy' and 'Arguments.copy' trace: [tracetool] Add method 'Event.api' to build event names Signed-off-by: Peter Maydell <[email protected]>

The compiling is done in a subdir, so the extraction of per-object libs and cflags are referencing objects with ../ prefixed. So prefix the per-object variables "foo.o-cflags" and "foo.o-libs" to "../foo.o-cflags" and "../foo.o-libs". Signed-off-by: Fam Zheng <[email protected]> Reviewed-by: Michael Tokarev <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]>

On win*, we build QEMU_PROGW (GUI) and create a console app QEMU_PROG from it, while on non-win*, we make only QEMU_PROG using the same rules as used for QEMU_PROGW on win*. Make just one rule for building main executable, and an additional rule for win* to make console app from it. Also consolidate tests for $(QEMU_PROGW). Signed-off-by: Michael Tokarev <[email protected]> [Fix user-mode compilation. - Paolo] Signed-off-by: Paolo Bonzini <[email protected]>

No need to save/restore obj-y, we can just build all-obj-y incrementally. Reviewed-by: Fam Zheng <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]>

Commit 52ba784 replaced $(STRIP_OPT) with $(STRIP) in some places (for example, Makefile.target), but not all of them. There are a few places remain in main Makefile which still uses $(STRIP_OPT). Replace these places with $(STRIP) too. While at it, simplify variable pattern substitution of the surrounding places, change $(patsubst pat,rep,$(var)) into $(var:pat=rep) which is much easier to read (this is probably a good idea to do everywhere). Signed-off-by: Michael Tokarev <[email protected]> Cc: Fam Zheng <[email protected]> Cc: Paolo Bonzini <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]>

$(INSTALL_PROG) is evaluated to libtool if using libtool, while $(INSTALL) is not. Use $(INSTALL_PROG) so that libtool is used with target too when necessary. This allows, for example, to link qemu with shared libcacard. Signed-off-by: Michael Tokarev <[email protected]> Cc: Fam Zheng <[email protected]> Cc: Paolo Bonzini <[email protected]> Cc: Alon Levy <[email protected]> Cc: [email protected] -- This is done on top of previous patch (using $(STRIP)), but it can be used by its own. Signed-off-by: Paolo Bonzini <[email protected]>

We have $(INSTALL_LIB) which is the same as $(INSTALL_PROG) but uses correct permissions. Loadable objects (modules) are like shared libraries, not like programs. Signed-off-by: Michael Tokarev <[email protected]> Cc: Fam Zheng <[email protected]> Cc: Paolo Bonzini <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]>

This is needed in order to use per-object flags variables. Reviewed-by: Michael Tokarev <[email protected]> Reviewed-by: Fam Zheng <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]>

Current Makefile system allows using foo.o-cflags variables to store object-specific CFLAGS. Convert some usages of old syntax (using QEMU_CFLAGS += construct) to the new syntax. Do not touch multifile modules for now, as build system isn't ready for this. Signed-off-by: Michael Tokarev <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]>

Signed-off-by: Lluís Vilanova <[email protected]> Reviewed-by: Eric Blake <[email protected]> Reviewed-by: Markus Armbruster <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

Use an explicit input file on the command-line instead of reading from standard input. It also outputs the proper file name when there's an error. Signed-off-by: Lluís Vilanova <[email protected]> Reviewed-by: Eric Blake <[email protected]> Reviewed-by: Markus Armbruster <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

The primitive uses JSON syntax, and include paths are relative to the file using the directive: { 'include': 'path/to/file.json' } Signed-off-by: Lluís Vilanova <[email protected]> Reviewed-by: Eric Blake <[email protected]> Reviewed-by: Markus Armbruster <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

strtosz_suffix() might return negative error, this patch fixes the error handling. This patch also changes to handle error in the if statement rather than handle success specially, this will make this use of strtosz_suffix consistent with all other uses. Signed-off-by: Amos Kong <[email protected]> Reviewed-by: Michael S. Tsirkin <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

qemu_parse_fd() used to handle at least the following strings incorrectly: o "-2": simply let through o "2147483648": returned as LONG_MAX==INT_MAX on ILP32 (with ERANGE ignored); implementation-defined behavior on LP64 Signed-off-by: Laszlo Ersek <[email protected]> Reviewed-by: Eric Blake <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

and rebase monitor_handle_fd_param() to it. (Note that this will slightly change the behavior when the qemu_parse_fd() branch is selected and it fails: we now report (and in case of QMP, set) the error immediately, rather than allowing the caller to set its own error message (if any)). Signed-off-by: Laszlo Ersek <[email protected]> Reviewed-by: Eric Blake <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

Propagate any errors in monitor fd handling up to get_real_device(), and report them there. We'll continue the propagation upwards when get_real_device() becomes a leaf itself (when none of its callees will report errors internally any longer when detecting and returning an error). Signed-off-by: Laszlo Ersek <[email protected]> eviewed-by: Eric Blake <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

This allows us to report the entire error with one error_report() call, easing future error propagation. Signed-off-by: Laszlo Ersek <[email protected]> Reviewed-by: Eric Blake <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

get_real_id() has two thin wrappers (and no other callers), get_real_vendor_id() and get_real_device_id(); it's easiest to convert them in one fell swoop. Signed-off-by: Laszlo Ersek <[email protected]> Reviewed-by: Eric Blake <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

Rename check_irqchip_in_kernel() to verify_irqchip_in_kernel(), so that the name reflects our expectation better. Rather than returning a bool, make it do nothing or set an Error. Signed-off-by: Laszlo Ersek <[email protected]> Reviewed-by: Eric Blake <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

... and rebase pci_add_capability() to it. Signed-off-by: Laszlo Ersek <[email protected]> Reviewed-by: Eric Blake <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

Propagate any errors while adding PCI capabilities to assigned_device_pci_cap_init(). We'll continue the propagation upwards when assigned_device_pci_cap_init() becomes a leaf itself (when none of its callees will report errors internally any longer when detecting and returning them). Signed-off-by: Laszlo Ersek <[email protected]> Reviewed-by: Eric Blake <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

assigned_initfn() get_real_device() read() Signed-off-by: Laszlo Ersek <[email protected]> Reviewed-by: Eric Blake <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

Signed-off-by: Laszlo Ersek <[email protected]> Reviewed-by: Eric Blake <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

The return type is also changed from "int" to "void", because it was used in a success vs. failure sense only (the caller didn't distinguish error codes from each other, and even assigned_dev_register_msix_mmio() masked mmap()'s errno values with a common -EFAULT). Signed-off-by: Laszlo Ersek <[email protected]> Reviewed-by: Eric Blake <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

Signed-off-by: Laszlo Ersek <[email protected]> Reviewed-by: Eric Blake <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

Also, change the return type to "void"; the function is static (with a sole caller) and the negative errno values are not distinguished from each other. Signed-off-by: Laszlo Ersek <[email protected]> Reviewed-by: Eric Blake <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

Among the callers, only assigned_initfn() should set the monitor's stored error. Other callers may run in contexts where the monitor's stored error makes no sense. For example: assigned_dev_pci_write_config() assigned_dev_update_msix() assign_intx() Signed-off-by: Laszlo Ersek <[email protected]> Reviewed-by: Eric Blake <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

Signed-off-by: Laszlo Ersek <[email protected]> Reviewed-by: Eric Blake <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

Signed-off-by: Markus Armbruster <[email protected]> Reviewed-by: Eric Blake <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

If the very first allocation has a length of 0, the free_cluster_index is still 0 after the for loop, which means that subtracting one from it will underflow and signal an invalid range of clusters by returning -EFBIG. However, there is no such range, as its length is 0. Fix this by preventing underflows on free_cluster_index during the check. Signed-off-by: Max Reitz <[email protected]> Reviewed-by: Kevin Wolf <[email protected]> Signed-off-by: Stefan Hajnoczi <[email protected]>

Currently, QEMU's iotests rely on /usr/bin/env to start the correct Python (that is, at least Python 2.4, but not 3). On systems where Python 3 is the default, the user has no clean way of making the iotests use the correct binary. This commit makes the iotests use the Python selected by configure. Signed-off-by: Max Reitz <[email protected]> Signed-off-by: Stefan Hajnoczi <[email protected]>

The help message for qemu-img lists the supported block formats, of which there are 27 as of version 2.0.50. The formats are printed in the order of their driver's position in a linked list, which appears random. This patch prints the formats in sorted order, making it easier to read and to find a specific format in the list. [Added suggestions from Fam Zheng <[email protected]> to declare variables at the top of the scope in help() and to omit explicit cast for void* opaque. --Stefan] Signed-off-by: Mike Day <[email protected]> Signed-off-by: Stefan Hajnoczi <[email protected]>

After the URL has been parsed make sure the server part is valid in order to avoid a segmentation fault when calling nfs_mount(). Signed-off-by: Max Reitz <[email protected]> Signed-off-by: Stefan Hajnoczi <[email protected]>

Signed-off-by: Peter Lieven <[email protected]> Signed-off-by: Stefan Hajnoczi <[email protected]>

If target block driver forces compression, qemu-img convert needs to write by cluster size as well as "-c" option. Particularly, this applies for converting to VMDK streamOptimized format. Signed-off-by: Fam Zheng <[email protected]> Signed-off-by: Stefan Hajnoczi <[email protected]>

Add a wrapper function to support "compressed" path in qemu-img convert. Only support streamOptimized subformat case for now (num_extents == 1 and extent compression is true). Signed-off-by: Fam Zheng <[email protected]> Signed-off-by: Stefan Hajnoczi <[email protected]>

This will return cluster_size and needs_compressed_writes to caller, if all the extents have the same value (or there's only one extent). Otherwise return -ENOTSUP. cluster_size is only reported for sparse formats. Signed-off-by: Fam Zheng <[email protected]> Signed-off-by: Stefan Hajnoczi <[email protected]>

Signed-off-by: Fam Zheng <[email protected]> Signed-off-by: Stefan Hajnoczi <[email protected]>

Signed-off-by: Markus Armbruster <[email protected]> Reviewed-by: Eric Blake <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

Using error_is_set(errp) to check whether a function call failed is fragile: it breaks when errp is null. ga_get_fd_handle() and guest_file_handle_add() don't return a useful value when they fail, but that's just stupid. Fix that, and check them instead. As far as I can tell, errp can't be null there, but this is more robust and more obviously correct. Signed-off-by: Markus Armbruster <[email protected]> Reviewed-by: Eric Blake <[email protected]> Reviewed-by: Michael Roth <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

Null errp argument makes no sense. Assert it's not null, to make this explicit, and guard against misuse. All current callers pass non-null errp. Signed-off-by: Markus Armbruster <[email protected]> Reviewed-by: Eric Blake <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

do_qmp_dispatch()'s test for qmp_dispatch_check_obj() failure examines both the return value and the error object. The latter part is unclean; it works only when do_qmp_dispatch()'s caller passes a non-null errp argument. That's the case, but it's not locally obvious. Unclean. Cleanup would be easy enough, but since the unclean code is also redundant, let's just drop it. Signed-off-by: Markus Armbruster <[email protected]> Reviewed-by: Eric Blake <[email protected]> Reviewed-by: Michael Roth <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

visit_type_TestStruct() does nothing when called with an error set. Callers shouldn't do that, and no caller does. Drop the superfluous test. Signed-off-by: Markus Armbruster <[email protected]> Reviewed-by: Eric Blake <[email protected]> Reviewed-by: Michael Roth <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

Using error_is_set(ERRP) to find out whether a function failed is either wrong, fragile, or unnecessarily opaque. It's wrong when ERRP may be null, because errors go undetected when it is. It's fragile when proving ERRP non-null involves a non-local argument. Else, it's unnecessarily opaque (see commit 84d18f0). The error_is_set(errp) in do_qmp_dispatch() is merely fragile, because the caller never passes a null errp argument. Make the code more robust and more obviously correct: receive the error in a local variable, then propagate it through the parameter. Signed-off-by: Markus Armbruster <[email protected]> Reviewed-by: Eric Blake <[email protected]> Reviewed-by: Michael Roth <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

Using error_is_set(ERRP) to find out whether a function failed is either wrong, fragile, or unnecessarily opaque. It's wrong when ERRP may be null, because errors go undetected when it is. It's fragile when proving ERRP non-null involves a non-local argument. Else, it's unnecessarily opaque (see commit 84d18f0). The error_is_set(errp) in the guest agent command handler functions are merely fragile, because all chall chains (do_qmp_dispatch() via the generated marshalling functions) pass a non-null errp argument. Make the code more robust and more obviously correct: receive the error in a local variable, then propagate it through the parameter. Signed-off-by: Markus Armbruster <[email protected]> Reviewed-by: Eric Blake <[email protected]> Reviewed-by: Michael Roth <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

acquire_privilege(), execute_async() and check_suspend_mode() do nothing when called with an error set. Callers shouldn't do that, and no caller does. Drop the superfluous tests. Signed-off-by: Markus Armbruster <[email protected]> Reviewed-by: Eric Blake <[email protected]> Reviewed-by: Michael Roth <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

Using error_is_set(ERRP) to find out whether to bail out due to previous error is either wrong, fragile, or unnecessarily opaque. It's wrong when ERRP may be null, because errors go undetected when it is. It's fragile when proving ERRP non-null involves a non-local argument. Else, it's unnecessarily opaque (see commit 84d18f0). The error_is_set(state->errp) in qemu_opts_from_qdict_1() is merely fragile, because the callers never pass state argument with null state->errp. Make the code more robust and more obviously correct: test *state->errp directly. Signed-off-by: Markus Armbruster <[email protected]> Reviewed-by: Eric Blake <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

In qmp_dump_guest_memory(), the error must be clear on entry, and we always bail out after setting it, directly or via dump_init(). Therefore, both error_is_set() are always false. Drop them. DumpState member errp is now write-only. Drop it, too. Signed-off-by: Markus Armbruster <[email protected]> Reviewed-by: Eric Blake <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

Using error_is_set(errp) that way can sweep programming errors under the carpet when we get called incorrectly with an error set. encrypted_bdrv_it() does it, because there's no way to make bdrv_iterate() break its loop. Actually safe, because qmp_cont() clears the error before the loop. Clean it up anyway: replace bdrv_iterate() by bdrv_next(), break the loop on error. Replace both occurrences, for consistency. Signed-off-by: Markus Armbruster <[email protected]> Reviewed-by: Eric Blake <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

Our example should use the correct quotes to match what someone could actually pass over the wire. * qmp-commands.hx: Use correct JSON quotes. Signed-off-by: Eric Blake <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

Signed-off-by: Eric Blake <[email protected]> Signed-off-by: Fam Zheng <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

…r()" This reverts commit 25a7017. Turns out the argument *can* be null: QEMU now segfaults if it receives an invalid parameter via a qmp command instead of throwing an error. For example: { "execute": "blockdev-add", "arguments": { "options" : { "driver": "invalid-driver" } } } CC: [email protected] Signed-off-by: Peter Lieven <[email protected]> Reviewed-by: Eric Blake <[email protected]> Reviewed-by: Markus Armbruster <[email protected]> Signed-off-by: Luiz Capitulino <[email protected]>

…aging * remotes/qmp-unstable/queue/qmp: (38 commits) Revert "qapi: Clean up superfluous null check in qapi_dealloc_type_str()" qapi: Document optional arguments' backwards compatibility qmp: use valid JSON in transaction example qmp: Don't use error_is_set() to suppress additional errors dump: Drop pointless error_is_set(), DumpState member errp qemu-option: Clean up fragile use of error_is_set() qga: Drop superfluous error_is_set() qga: Clean up fragile use of error_is_set() qapi: Clean up fragile use of error_is_set() tests/qapi-schema: Drop superfluous error_is_set() qapi: Drop redundant, unclean error_is_set() hmp: Guard against misuse of hmp_handle_error() qga: Use return values instead of error_is_set(errp) error: Consistently name Error ** objects errp, and not err qmp: Consistently name Error ** objects errp, and not err qga: Consistently name Error ** objects errp, and not err qmp hmp: Consistently name Error * objects err, and not errp pci-assign: assigned_initfn(): set monitor error in common error handler pci-assign: propagate errors from assign_intx() pci-assign: propagate errors from assign_device() ... Signed-off-by: Peter Maydell <[email protected]>

The immediately visible effect of this patch is that it fixes committing a temporary snapshot to its backing file. Previously, it would fail with a "permission denied" error because bdrv_inherited_flags() forced the backing file to be read-only, ignoring the r/w reopen of bdrv_commit(). The bigger problem this revealed is that the original open flags must actually only be applied to the temporary snapshot, and the original image file must be treated as a backing file of the temporary snapshot and get the right flags for that. Reported-by: Jan Kiszka <[email protected]> Signed-off-by: Kevin Wolf <[email protected]> Reviewed-by: Max Reitz <[email protected]> Signed-off-by: Stefan Hajnoczi <[email protected]>

Clocks are initialized in qemu_init_main_loop. They are not needed before it. Initializing them twice is not only unnecessary but is harmful: it results in memory leak and potentially can lead to a situation where different parts of QEMU use different sets of timers. To avoid it remove init_clocks call from main and add an assertion to qemu_clock_init that corresponding clock has not been initialized yet. Signed-off-by: Kirill Batuzov <[email protected]> Signed-off-by: Stefan Hajnoczi <[email protected]>

The docs for glfs_init suggest that the function sets errno on every failure. In fact it doesn't. As other functions such as qemu_gluster_open() in the gluster block code report their errors based on this fact we need to make sure that errno is set on each failure. This fixes a crash of qemu-img/qemu when a gluster brick isn't accessible from given host while the server serving the volume description is. Thread 1 (Thread 0x7ffff7fba740 (LWP 203880)): #0 0x00007ffff77673f8 in glfs_lseek () from /usr/lib64/libgfapi.so.0 snabbco#1 0x0000555555574a68 in qemu_gluster_getlength () #2 0x0000555555565742 in refresh_total_sectors () #3 0x000055555556914f in bdrv_open_common () #4 0x000055555556e8e8 in bdrv_open () #5 0x000055555556f02f in bdrv_open_image () #6 0x000055555556e5f6 in bdrv_open () #7 0x00005555555c5775 in bdrv_new_open () #8 0x00005555555c5b91 in img_info () #9 0x00007ffff62c9c05 in __libc_start_main () from /lib64/libc.so.6 #10 0x00005555555648ad in _start () Signed-off-by: Stefan Hajnoczi <[email protected]>

The current version of raw-posix always uses ioctl(FS_IOC_FIEMAP) if FIEMAP is available; lseek with SEEK_HOLE/SEEK_DATA are not even compiled in in this case. However, there may be implementations which support the latter but not the former (e.g., NFSv4.2) as well as vice versa. To cover both cases, try FIEMAP first (as this will return -ENOTSUP if not supported instead of returning a failsafe value (everything allocated as a single extent)) and if that does not work, fall back to SEEK_HOLE/SEEK_DATA. Signed-off-by: Max Reitz <[email protected]> Signed-off-by: Stefan Hajnoczi <[email protected]>

This creates some common functions for bash language qemu-iotests to control, and communicate with, a running QEMU process. 4 functions are introduced: 1. _launch_qemu() This launches the QEMU process(es), and sets up the file descriptors and fifos for communication. You can choose to launch each QEMU process listening for either QMP or HMP monitor. You can call this function multiple times, and save the handle returned from each. The returned handle is in $QEMU_HANDLE. You must copy this value. Commands 2 and 3 use the handle received from _launch_qemu(), to talk to the appropriate process. 2. _send_qemu_cmd() Sends a command string, specified by $2, to QEMU. If $3 is non-NULL, _send_qemu_cmd() will wait to receive $3 as a required result string from QEMU. Failure to receive $3 will cause the test to fail. The command can optionally be retried $qemu_cmd_repeat number of times. Set $qemu_error_no_exit to not force the test the fail on exit; in this case, $QEMU_STATUS[$1] will be set to -1 on failure. 3. _timed_wait_for() Waits for a response, for up to a default of 10 seconds. If $2 is not seen in that time (anywhere in the response), then the test fails. Primarily used by _send_qemu_cmd, but could be useful standalone, as well. To prevent automatic exit (and therefore test failure), set $qemu_error_no_exit to a non-NULL value. If $silent is a non-NULL value, then output to stdout will be suppressed. 4. _cleanup_qemu() Kills the running QEMU processes, and removes the fifos. Signed-off-by: Jeff Cody <[email protected]> Signed-off-by: Stefan Hajnoczi <[email protected]>

The new functionality of common.qemu implements the QEMU control and communication functionality that was originally in test 085. This removes that now-duplicate functionality, and uses the common.qemu functions. The QEMU commandline changes slightly due to this; in addition to monitor and qmp i/o options, the new QEMU commandline from inside common.qemu now introduces -machine accel=qtest. Reviewed-by: Benoit Canet <[email protected]> Reviewed-by: Fam Zheng <[email protected]> Signed-off-by: Jeff Cody <[email protected]> Signed-off-by: Stefan Hajnoczi <[email protected]>

This is an initial, simple live migration test from one running VM to another, using monitor commands. This is also an example of using the new common.qemu functions for controlling multiple running qemu instances, for tests that need a live qemu vm. Reviewed-by: Fam Zheng <[email protected]> Signed-off-by: Jeff Cody <[email protected]> Signed-off-by: Stefan Hajnoczi <[email protected]>

g_poll has a problem on Windows when using timeouts < 10ms, in glib/gpoll.c: /* If not, and we have a significant timeout, poll again with * timeout then. Note that this will return indication for only * one event, or only for messages. We ignore timeouts less than * ten milliseconds as they are mostly pointless on Windows, the * MsgWaitForMultipleObjectsEx() call will timeout right away * anyway. */ if (retval == 0 && (timeout == INFINITE || timeout >= 10)) retval = poll_rest (poll_msgs, handles, nhandles, fds, nfds, timeout); so whenever g_poll is called with timeout < 10ms it does a quick poll instead of wait, this causes significant performance degradation of QEMU, thus we should use WaitForMultipleObjectsEx directly Signed-off-by: Stanislav Vorobiov <[email protected]> Signed-off-by: Stefan Hajnoczi <[email protected]>

fix-obj-vars has the undesired side effect of breaking -cflags -objs and -libs variables in the toplevel Makefile.objs. The variables in the toplevel Makefile.objs do not need any fix, so fix-obj-vars need not do anything. Since we are touching it, remove the now unnecessary $(if) in the callers. Signed-off-by: Paolo Bonzini <[email protected]>

Currently all what's needed for single file libcacard/vcard_emul_nss.c (libnss cflags) and hw/usb/ccid-card-emulated.c (libcacard includes) together with the libs is added to global QEMU_CFLAGS and libs_softmmu. Use the cflags only where really used (for two mentioned files), and libs only where needed. While at it, rename variables to better reflect reality: libcacard_* is really nss_*. This needs a bit more tweaking: $(NSS_LIBS) should not contain $glib_libs (ditto for _cflags). But in order to fix it, some more preparations should be made first. So add a FIXME comment. Signed-off-by: Michael Tokarev <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]>

It's a loop from i < num_sg and the array is VIRTQUEUE_MAX_SIZE - so it's OK if the value read is VIRTQUEUE_MAX_SIZE. Not a big problem in practice as people don't use such big queues, but it's inelegant. Reported-by: "Dr. David Alan Gilbert" <[email protected]> Cc: [email protected] Signed-off-by: Michael S. Tsirkin <[email protected]>

The code which patches x86 jump instructions assumes it can do an unaligned write of a uint32_t. This is actually safe on x86, but it's still undefined behaviour. We have infrastructure for doing efficient unaligned accesses which doesn't engage in undefined behaviour, so use it. This is technically fractionally less efficient, at least with gcc 4.6; instead of one instruction: 7b2: 89 3e mov %edi,(%rsi) we get an extra spurious store to the stack slot: 7b2: 89 7c 24 64 mov %edi,0x64(%rsp) 7b6: 89 3e mov %edi,(%rsi) Reviewed-by: Alex Bennée <[email protected]> Signed-off-by: Peter Maydell <[email protected]> Signed-off-by: Richard Henderson <[email protected]>

Avoid stores to unaligned addresses in TCG code generation, by using the usual memcpy() approach. (Using bswap.h would drag a lot of QEMU baggage into TCG, so it's simpler just to do direct memcpy() here.) Reviewed-by: Alex Bennée <[email protected]> Signed-off-by: Peter Maydell <[email protected]> Signed-off-by: Richard Henderson <[email protected]>

To avoid C undefined behaviour when patching generated code, provide wrappers tcg_patch8/16/32/64 which use the usual memcpy trick, and use them in the i386 backend. Reviewed-by: Alex Bennée <[email protected]> Signed-off-by: Peter Maydell <[email protected]> Signed-off-by: Richard Henderson <[email protected]>

Reviewed-by: Peter Maydell <[email protected]> Signed-off-by: Richard Henderson <[email protected]>

To be defined by the tcg backend based on the elemental unit of the ISA. During the transition, allow TCG_TARGET_INSN_UNIT_SIZE to be undefined, which allows us to default tcg_insn_unit to the current uint8_t. Reviewed-by: Peter Maydell <[email protected]> Signed-off-by: Richard Henderson <[email protected]>

And use tcg pointer differencing functions as appropriate. Reviewed-by: Alex Bennée <[email protected]> Reviewed-by: Peter Maydell <[email protected]> Signed-off-by: Richard Henderson <[email protected]>

And use tcg pointer differencing functions as appropriate. Reviewed-by: Alex Bennée <[email protected]> Signed-off-by: Richard Henderson <[email protected]>

And use tcg pointer differencing functions as appropriate. Reviewed-by: Alex Bennée <[email protected]> Reviewed-by: Peter Maydell <[email protected]> Signed-off-by: Richard Henderson <[email protected]>

And use tcg pointer differencing functions as appropriate. Reviewed-by: Peter Maydell <[email protected]> Signed-off-by: Richard Henderson <[email protected]>

And use tcg pointer differencing functions as appropriate. Acked-by: Claudio Fontana <[email protected]> Reviewed-by: Peter Maydell <[email protected]> Signed-off-by: Richard Henderson <[email protected]>

And use tcg pointer differencing functions as appropriate. Signed-off-by: Richard Henderson <[email protected]>

Using a 16-byte aligned structure achieves best results, both for code cleanliness and compiled code size. However, this means that we can't use the trick of encoding the slot number into the low 2 bits. Thankfully, we only ever use slot2, so make that explicit in the names of the relocation functions, and drop the code for other slots. Signed-off-by: Richard Henderson <[email protected]>

And use tcg pointer differencing functions as appropriate. Reviewed-by: Peter Maydell <[email protected]> Signed-off-by: Richard Henderson <[email protected]>

Now that all backends do define TCG_TARGET_INSN_UNIT_SIZE, remove the fallback definition. Reviewed-by: Alex Bennée <[email protected]> Signed-off-by: Richard Henderson <[email protected]>

Signed-off-by: Richard Henderson <[email protected]>

Merge the existing tcg_out_call into tcg_out_op. Signed-off-by: Richard Henderson <[email protected]>

Rename the existing tcg_out_calli to tcg_out_call_nodelay. Signed-off-by: Richard Henderson <[email protected]>

Signed-off-by: Richard Henderson <[email protected]>

Avoid allocating a tcg temporary to hold the constant address, and instead place it directly into the op_call arguments. At the same time, convert to the newly introduced tcg_out_call backend function, rather than invoking tcg_out_op for the call. Signed-off-by: Richard Henderson <[email protected]>

The move opcodes are special in that their constraints must cover all available registers. So instead of checking the constraints, just use the available registers. Signed-off-by: Richard Henderson <[email protected]>

The INDEX_op_call case has just been obsoleted; the mov and movi cases have not been reachable for years. Attempt to document this both in each tcg_out_op switch, and via TCG_OPF_NOT_PRESENT. Because of the TCG_OPF_NOT_PRESENT change, this must be done for all targets in a single commit. Signed-off-by: Richard Henderson <[email protected]>

Older Intel manuals (pre-2010) and current AMD manuals describe Z as undefined, but newer Intel manuals describe Z as unchanged. Cc: [email protected] Reviewed-by: Paolo Bonzini <[email protected]> Signed-off-by: Richard Henderson <[email protected]>

Signed-off-by: Edgar E. Iglesias <[email protected]>

The MER register only has two valid bits. This is now used by the linux kernel to auto-detect endianness, and causes Linux 3.15-rc1 and later to hang when run under qemu-microblaze. Mask valid bits before writing the register to solve the problem. Signed-off-by: Guenter Roeck <[email protected]> Reviewed-by: Edgar E. Iglesias <[email protected]> [Edgar: Untabified] Signed-off-by: Edgar E. Iglesias <[email protected]>

The TCSR register has only 11 valid bits. This is now used by the linux kernel to auto-detect endianness, and causes Linux 3.15-rc1 and later to hang when run under qemu-microblaze. Mask valid bits before writing the register to solve the problem. Signed-off-by: Guenter Roeck <[email protected]> Reviewed-by: Edgar E. Iglesias <[email protected]> Signed-off-by: Edgar E. Iglesias <[email protected]>

There was some modulo logic to ensure that Microblaze always booted into physical RAM regardless of the elf entry. Removed it, as QEMU should fail gracefully when given a bad elf, rather than attempt to run it. Signed-off-by: Peter Crosthwaite <[email protected]> Signed-off-by: Edgar E. Iglesias <[email protected]>

…t' into staging Block pull request # gpg: Signature made Fri 09 May 2014 19:57:53 BST using RSA key ID 81AB73C8 # gpg: Good signature from "Stefan Hajnoczi <[email protected]>" # gpg: aka "Stefan Hajnoczi <[email protected]>" * remotes/stefanha/tags/block-pull-request: glib: fix g_poll early timeout on windows block: qemu-iotests - test for live migration block: qemu-iotests - update 085 to use common.qemu block: qemu-iotests - add common.qemu, for bash-controlled qemu tests block/raw-posix: Try both FIEMAP and SEEK_HOLE gluster: Correctly propagate errors when volume isn't accessible vl.c: remove init_clocks call from main block: Fix open flags with BDRV_O_SNAPSHOT qemu-iotests: Test converting to streamOptimized from small cluster size vmdk: Implement .bdrv_get_info() vmdk: Implement .bdrv_write_compressed qemu-img: Convert by cluster size if target is compressed block/iscsi: bump year in copyright notice block/nfs: Check for NULL server part qemu-img: sort block formats in help message iotests: Use configured python qcow2: Fix alloc_clusters_noref() overflow detection Signed-off-by: Peter Maydell <[email protected]>

* remotes/bonzini/configure: libcacard: remove libcacard-specific CFLAGS and LIBS from global vars build: simplify and fix fix-obj-vars build: convert some obj-specific CFLAGS to use new foo.o-cflags syntax build: add support for per-object -cflags and -libs to all rules Makefile: use $(INSTALL_LIB) for modules not $(INSTALL_PROG) Makefile.target: use $(INSTALL_PROG) for installing, not $(INSTALL) Makefile: strip tools and modules too build: simplify Makefile.target around unnest-vars invocations build: simplify Makefile.target a bit, use just one rule for softmmu build: Fix per-object variables for Makefile.target Signed-off-by: Peter Maydell <[email protected]>

Signed-off-by: Stefan Weil <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]>

s390x introduced helper functions for getting/setting one_regs with commit 860643b. However, nothing about these is s390-specific. Alexey Kardashevskiy had already posted a general version, so let's merge the two patches and massage the code a bit. CC: Alexey Kardashevskiy <[email protected]> Signed-off-by: Cornelia Huck <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]>

When mmapping memory for the MSI-X table failsthe dev->msix_table is not set to NULL and assigned_dev_unregister_msix_mmio() will cause a segfault when trying to munmap it. Signed-off-by: Gonglei Arei <[email protected]> Reviewed-by: Michael S. Tsirkin <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]> Signed-off-by: Gonglei <[email protected]>

KVM only supports MSIX table size up to 256 vectors, but some assigned devices support more vectors, at the moment attempts to assign them fail with EINVAL. Tweak the MSIX capability exposed to guest to limit table size to a supported value. Signed-off-by: Michael S. Tsirkin <[email protected]> Tested-by: Gonglei <[email protected]> Cc: [email protected] Acked-by: Alex Williamson <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]>

…helper.c The svm_load_seg_cache() function calls cpu_x86_load_seg_cache() which inspects env->eflags. So, make sure all changes to eflags are done prior to loading the segment cache. Signed-off-by: Kevin O'Connor <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]>

…he() in smm_helper.c The cpu_x86_load_seg_cache() function inspects cr0 and eflags, so make sure all changes to eflags and cr0 are done prior to loading the segment caches. Signed-off-by: Kevin O'Connor <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]>

…seg_helper.c The cpu_x86_load_seg_cache() function inspects eflags, so make sure all changes to eflags are done prior to loading the segment caches. Signed-off-by: Kevin O'Connor <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]>

…s accordingly. Instead of manually calling cpu_x86_set_cpl() when the CPL changes, check for CPL changes on calls to cpu_x86_load_seg_cache(R_CS). Every location that called cpu_x86_set_cpl() also called cpu_x86_load_seg_cache(R_CS), so cpu_x86_set_cpl() is no longer required. This fixes the SMM handler code as it was not setting/restoring the CPL level manually. Signed-off-by: Kevin O'Connor <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]>

Now that we have a CPU object with a reset method, it is better to keep the KVM reset close to the CPU reset. Using qemu_register_reset as we do now keeps them far apart. With this patch, PPC no longer calls the kvm_arch_ function, so it can get removed there. Other arches call it from their CPU reset handler, and the function gets an ARMCPU/X86CPU/S390CPU. Note that ARM- and s390-specific functions are called kvm_arm_* and kvm_s390_*, while x86-specific functions are called kvm_arch_*. That follows the convention used by the different architectures. Changing that is the topic of a separate patch. Reviewed-by: Gleb Natapov <[email protected]> Reviewed-by: Michael S. Tsirkin <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]>

Reviewed-by: Gleb Natapov <[email protected]> Reviewed-by: Michael S. Tsirkin <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]>

BND0-3, BNDCFGU, BNDCFGS, BNDSTATUS were not zeroed on reset, but they should be (Intel Instruction Set Extensions Programming Reference 319433-015, pages 9-4 and 9-6). Same for YMM. XCR0 should be reset to 1. TSC and TSC_RESET were zeroed already by the memset, remove the explicit assignments. Cc: Andreas Faerber <[email protected]> Reviewed-by: Michael S. Tsirkin <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]>

Most MSRs, plus the FPU, MMX, MXCSR, XMM and YMM registers should not be zeroed on INIT (Table 9-1 in the Intel SDM). Copy them out of CPUX86State and back in, instead of special casing env->pat. The relevant fields are already consecutive except PAT and SMBASE. However: - KVM and Hyper-V MSRs should be reset because they include memory locations written by the hypervisor. These MSRs are moved together at the end of the preserved area. - SVM state can be moved out of the way since it is written by VMRUN. Cc: Andreas Faerber <[email protected]> Reviewed-by: Michael S. Tsirkin <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]>

SIPI interrupts are ignored on the bootstrap. Never accept one. Cc: Andreas Faerber <[email protected]> Reviewed-by: Michael S. Tsirkin <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]>

On the x86, some devices need access to the CPU reset pin (INIT#). Provide a generic service to do this, using one of the internal cpu_interrupt targets. Generalize the PPC-specific code for CPU_INTERRUPT_RESET to other targets. Since PPC does not support migration across QEMU versions (its machine types are not versioned yet), I picked the value that is used on x86, CPU_INTERRUPT_TGT_INT_1. Consequently, TGT_INT_2 and TGT_INT_3 are shifted down by one while keeping their value. Reviewed-by: Anthony Liguori <[email protected]> Reviewed-by: Michael S. Tsirkin <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]>

The PIIX datasheet says that "before another INIT pulse can be generated via [port 92h], [bit 0] must be written back to a zero. This bug is masked right now because a full reset will clear the value of port 92h. But once we implement soft reset correctly, the next attempt to enable the A20 line by setting bit 1 (and leaving the others untouched) will cause another reset. Reviewed-by: Anthony Liguori <[email protected]> Reviewed-by: Michael S. Tsirkin <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]>

Commit e586822 broke the bsd-user build when it removed the CONFIG_UNAME_RELEASE define but forgot to remove the use of it in bsd-user. Fix this in the simplest possible way (bsd-user doesn't make any use at all of the qemu_uname_release variable except to allow it to be pointlessly set by the user, so this is all we need to do.) Signed-off-by: Peter Maydell <[email protected]> Message-id: [email protected]

…o staging tcg updates for 20140512 # gpg: Signature made Tue 13 May 2014 00:19:56 BST using RSA key ID 4DD0279B # gpg: Can't check signature: public key not found * remotes/rth/tags/pull-tcg-20140512: (26 commits) tcg: Remove unreachable code in tcg_out_op and op_defs tcg: Use tcg_target_available_regs in tcg_reg_alloc_mov tcg: Make call address a constant parameter tci: Create tcg_out_call tcg-mips: Split out tcg_out_call tcg-sparc: Create tcg_out_call tcg-ppc64: Rename tcg_out_calli to tcg_out_call tcg-ppc: Split out tcg_out_call tcg-s390: Rename tgen_calli to tcg_out_call tcg-i386: Rename tcg_out_calli to tcg_out_call tcg: Require TCG_TARGET_INSN_UNIT_SIZE tci: Define TCG_TARGET_INSN_UNIT_SIZE tcg-mips: Define TCG_TARGET_INSN_UNIT_SIZE tcg-ia64: Define TCG_TARGET_INSN_UNIT_SIZE tcg-s390: Define TCG_TARGET_INSN_UNIT_SIZE tcg-aarch64: Define TCG_TARGET_INSN_UNIT_SIZE tcg-arm: Define TCG_TARGET_INSN_UNIT_SIZE tcg-sparc: Define TCG_TARGET_INSN_UNIT_SIZE tcg-ppc: Define TCG_TARGET_INSN_UNIT_SIZE tcg-ppc64: Define TCG_TARGET_INSN_UNIT_SIZE ... Signed-off-by: Peter Maydell <[email protected]>

Update our copy of libvixl to upstream's 1.4 release. Note that we no longer need any local fixes for compilation on 32 bit hosts -- they have all been integrated upstream. Signed-off-by: Peter Maydell <[email protected]> Message-id: [email protected] Acked-by: Richard Henderson <[email protected]>

After commit 767adce, they are redundant. This way we don't assign them except when needed. Once there, there were lots of cases where the ".fields" indentation was wrong: .fields = (VMStateField []) { and .fields = (VMStateField []) { Change all the combinations to: .fields = (VMStateField[]){ The biggest problem (apart from aesthetics) was that checkpatch complained when we copy&pasted the code from one place to another. Signed-off-by: Juan Quintela <[email protected]> [PMM: fixed minor conflict, corrected commit message typos] Signed-off-by: Peter Maydell <[email protected]>

The current tx_fifo code has a corner case where the guest can overrun the fifo buffer: if automatic CRCs are disabled we allow the guest to write the CRC word even if there isn't actually space for it in the FIFO. The datasheet is unclear about exactly how the hardware deals with this situation; the most plausible answer seems to be that the CRC word is just lost. Implement this fix by separating the "can we stuff another word in the FIFO" logic from the "should we transmit the packet now" check. This also moves us closer to the real hardware, which has a number of ways it can be configured to trigger sending the packet, some of which we don't implement. Signed-off-by: Peter Maydell <[email protected]> Reviewed-by: Dr. David Alan Gilbert <[email protected]> Cc: [email protected]

The PADEN bit in the transmit control register enables padding of short data packets out to the required minimum length. However a typo here meant we were adjusting tx_fifo_len rather than tx_frame_len, so the padding didn't actually happen. Fix this bug. Signed-off-by: Peter Maydell <[email protected]> Reviewed-by: Dr. David Alan Gilbert <[email protected]> Cc: [email protected]

The datasheet is clear that the frame length written to the DATA register is actually stored in the TX FIFO; this means we don't need to keep both tx_frame_len and tx_fifo_len state separately. Signed-off-by: Peter Maydell <[email protected]> Reviewed-by: Dr. David Alan Gilbert <[email protected]>

Packet transmission for the stellaris ethernet controller can be triggered in one of two ways: * by setting a threshold value in the THR register; when the FIFO fill level reaches the threshold, the h/w starts transmitting. Software has to finish filling the FIFO before the transmit process completes to avoid a (silent) underrun * by software writing to the TR register to explicitly trigger transmission Since QEMU transmits packets instantaneously (from the guest's point of view), implement "transmit based on threshold" with our existing mechanism of "transmit as soon as we have the whole packet", with the additional wrinkle that we don't transmit if the packet size is below the specified threshold, and implement "transmit by specific request" properly. Signed-off-by: Peter Maydell <[email protected]> Reviewed-by: Dr. David Alan Gilbert <[email protected]>

Fix various debug format strings which were incorrect for the data type, so that building with debug enabled is possible. Signed-off-by: Peter Maydell <[email protected]> Reviewed-by: Dr. David Alan Gilbert <[email protected]>

The rx_fifo pointer is awkward to migrate, and is actually redundant since it is always possible to determine it from the current rx[].len/.data and rx_fifo_len. Remove both rx_fifo and rx_fifo_len from the state, replacing them with a simple rx_fifo_offset which points at the current location in the RX fifo. Signed-off-by: Peter Maydell <[email protected]> Reviewed-by: Dr. David Alan Gilbert <[email protected]>

Convert this device to use vmstate for its save/load, including providing a post_load function that sanitizes inbound data to avoid possible buffer overflows if it is malicious. The sanitizing fixes CVE-2013-4532 (though nobody should be relying on the security properties of most of the unmaintained ARM board models anyway, and migration doesn't actually work on this board due to issues in other device models). Signed-off-by: Peter Maydell <[email protected]> Reviewed-by: Dr. David Alan Gilbert <[email protected]> Reviewed-by: Michael S. Tsirkin <[email protected]>

Linux makes a habit of writing the same value to the SCTLR that it already holds. In a sample boot of the kernel to a shell prompt it wrote the SCTLR with the value it already held 325465 times, and wrote different values just 3 times. Skip flushing the TLB if the SCTLR value isn't actually being changed; this speeds up my sample boot by 3-5%. Reported-by: Laurent Desnogues <[email protected]> Reviewed-by: Laurent Desnogues <[email protected]> Signed-off-by: Peter Maydell <[email protected]> Message-id: [email protected]

Add missing 'break' after handling of AW_A10_PIC_BASE_ADDR write. Signed-off-by: Peter Maydell <[email protected]> Reviewed-by: Peter Crosthwaite <[email protected]>

Commit 191946c moved the code to handle padding to minimum length from after the handling of the CRC to before it. This means that the CRC code doesn't need to cope with the possibility that the size is less than 60; remove this dead code. Signed-off-by: Peter Maydell <[email protected]> Reviewed-by: Peter Crosthwaite <[email protected]>

When writing to the YEARS_REG register, if the year value is 99 then the multiplication by 31536000 will overflow into the sign bit of a 32 bit value and then be erroneously sign-extended if time_t is 64 bits. Add a cast to avoid this. Signed-off-by: Peter Maydell <[email protected]> Reviewed-by: Peter Crosthwaite <[email protected]>

Add missing (uint32_t) casts in cases where we're trying to put a uint16_t value into the top half of a 32-bit field. These were already present in some but not all places. Signed-off-by: Peter Maydell <[email protected]> Reviewed-by: Peter Crosthwaite <[email protected]>

Add casts to avoid potentially overflowing the multiplications of 32 bit quantities in exynos4210_ltick_recalc_count(). Signed-off-by: Peter Maydell <[email protected]> Reviewed-by: Peter Crosthwaite <[email protected]>

We don't implement very much of the GPTM TAR register, and what we do is wrong. The "are we in RT mode?" field is in s->config, not s->control. Correct this, use LOG_UNIMP rather than hw_error() for the cases we don't support, and avoid an unlabelled fallthrough that makes Coverity complain. Signed-off-by: Peter Maydell <[email protected]> Reviewed-by: Peter Crosthwaite <[email protected]>

In fill_prefetch_fifo(), if the device we are reading from is 16 bit, then we must not try to transfer an odd number of bytes into the FIFO. This could otherwise have resulted in our overrunning the prefetch.fifo array by one byte. Signed-off-by: Peter Maydell <[email protected]> Reviewed-by: Peter Crosthwaite <[email protected]>

We got the wrong version of stl_p, the one that bswaps as appropriate for the target. Since x86 is always little-endian, the "_le_" routine will resolve to what we want. Signed-off-by: Richard Henderson <[email protected]>

* remotes/rth/fix-i386: exec-all.h: Use stl_le_p instead of stl_p patching x86 Signed-off-by: Peter Maydell <[email protected]>

For xbzrle_decode_buffer(), when decoding contents will exceed writing buffer, it will return -1, so need not check the return value whether large than writing buffer. And when failure occurs within load_xbzrle(), it always return -1 without any resources which need release. So can remove the related checking statements, and also can remove 'rc' and 'ret' local variables, Signed-off-by: Chen Gang <[email protected]> Signed-off-by: Juan Quintela <[email protected]>

ram_save_block is getting a bit too complicated, and does two separate things: 1) Finds a page to send 2) Sends the page (dealing with compression etc) Split into 'ram_save_page' to send the page and deal with compression (2) Rename remaining function to 'ram_find_and_save_block' Signed-off-by: Dr. David Alan Gilbert <[email protected]> Signed-off-by: Juan Quintela <[email protected]>

After previous Peter patch, they are redundant. This way we don't assign them except when needed. Once there, there were lots of case where the ".fields" indentation was wrong: .fields = (VMStateField []) { and .fields = (VMStateField []) { Change all the combinations to: .fields = (VMStateField[]){ The biggest problem (appart from aesthetics) was that checkpatch complained when we copy&pasted the code from one place to another. Signed-off-by: Juan Quintela <[email protected]> Acked-by: Gerd Hoffmann <[email protected]>

After previous Peter patch, they are redundant. This way we don't assign them except when needed. Once there, there were lots of case where the ".fields" indentation was wrong: .fields = (VMStateField []) { and .fields = (VMStateField []) { Change all the combinations to: .fields = (VMStateField[]){ The biggest problem (appart from aesthetics) was that checkpatch complained when we copy&pasted the code from one place to another. Signed-off-by: Juan Quintela <[email protected]> Reviewed-by: Peter Maydell <[email protected]>

currently the value of the throughput field contains whatever was the last calculated throughput shortly before the migration finished. This patch updates the post migration contents of the field to the average throughput. Signed-off-by: Peter Lieven <[email protected]> Reviewed-by: Paolo Bonzini <[email protected]> Reviewed-by: Juan Quintela <[email protected]> Signed-by: Juan Quintela <[email protected]> Signed-off-by: Juan Quintela <[email protected]>

Correct post load checks: 1. dev->setup_len == sizeof(dev->data_buf) seems fine, no need to fail migration 2. When state is DATA, passing index > len will cause memcpy with negative length, resulting in heap overflow First of the issues was reported by dgilbert. Reported-by: "Dr. David Alan Gilbert" <[email protected]> Signed-off-by: Michael S. Tsirkin <[email protected]> Signed-off-by: Juan Quintela <[email protected]>

Commit af3cbfb hoisted some "common" loads of the temporary type, forgetting that the types could differ during truncating moves. This affects the correctness of the memory offset on big-endian hosts. Tested-by: Tom Musta <[email protected]> Signed-off-by: Richard Henderson <[email protected]>

Signed-off-by: Richard Henderson <[email protected]>

That is, the old LDST_OPTIMIZATION. Signed-off-by: Richard Henderson <[email protected]>

Elides two insns from the sequence. The resulting tlb compare sequence is satisfyingly minimal: risbg %r2,%r8,51,186,56 risbg %r3,%r8,61,178,0 cg %r3,904(%r10,%r2) lg %r2,920(%r10,%r2) jlh tlb_miss Signed-off-by: Richard Henderson <[email protected]>

Signed-off-by: Richard Henderson <[email protected]>

There are a variety of common cases for which we can use carry tricks to avoid a conditional branch. On very new hardware, use LOAD ON CONDITION instead of a conditional branch. Signed-off-by: Richard Henderson <[email protected]>

Code grew a bit, had to disable CONFIG_USB_UAS to make the bios for the old machine types continue top fit into 128k. git shortlog since 1.7.4: Eric Northup (2): init_virtio_blk, init_virtio_scsi: reset HBA on error vp_init_simple: enable PCI bus-mastering before relying on DMA. Gabriel L. Somlo (1): SMBIOS: Check for aggregate tables & entry point in fw_cfg Gerd Hoffmann (4): smbios: catch zero-length strings pci: improve io address space allocation xhci: allocate scratch pad buffers vga: rework bar detection, add virtio-vga Johannes Krampf (5): build: Remove unneccesary semicolons from acpi_extract scripts. build: Make print statements in scripts python3 compatible. build: Be explicit that we want integers when dividing for python3 compat. build: Avoid sort() on unordered classes for python3 compatibility. build: Be careful with unicode and byte strings for python3 compatibility. Kevin O'Connor (79): build: Don't trash the .config file when src/Kconfig changes. Remove the pmm handle argument from _malloc(). Clear the total block transfer count on error. Remove unnecesary updates of the disk op->count field. Always perform thread cleanup on MainThread stack. pvscsi: Don't store reference to struct pci_device. pvscsi: Always run entirely in 32bit mode. pvscsi: Remove use of LOWFLAT and GLOBALFLAT macros. usb: Replace EHCI to UHCI/OHCI synchronization with new scheme. Minor - pciinit.c - order header files and use static on local functions. Minor - fix misspelling in romfile.c comment. ehci: memset the qTD structures in ehci_alloc_intr_pipe build: Update kconfig to version in Linux 3.13. scripts: Allow encodeint.py to take integers in hex notation. coreboot: Support alternative locations for CBFS. debug: Add runtime option to not report the debug serial port to the OS. debug: Only call serial_debug_preinit() at startup. debug: Unify the SeaBIOS debug version banner. xhci: Use 64bit writes to ERDP register. xhci: Fix incorrect direction setting on status transmissions. xhci: Set the interval parameter on interrupt pipes. xhci: Verify PAGESIZE register before initializing driver. xhci: Allocate and free the xhci inctx structure on each use. xhci: Move set_address code from xhci_control to xhci_alloc_pipe. xhci: Eliminate 'struct xhci_device'. xhci: Support xhci controllers with 64 byte contexts. xhci: Allow the XHCI USB controller to be enabled for coreboot. xhci: Don't use a dummy endpoint count in configure command. xhci: Report the "protocol defined" and PSIC bits of extended capabilities. xhci: Support USB hubs on xhci controllers. Add call32_params() helper function. Add space between DTYPE_* definitions. xhci: Run the XHCI driver entirely in 32bit mode. Remove pci_writel() and pci_readl() functions. xhci: Check if scratch pad allocation fails. Increase maximum size of permanent high memory area. usb: USB 3 devices encode bMaxPacketSize0 field. usb: Move default pipe max packet size code from xhci to main code. xhci: Use the same endpoint initialization code in xhci_alloc_pipe() xhci: Add copyright notice to usb-xhci.c vgabios: Make sure exported structs use PACKED. vgabios: Simplify the bios save state area. vgabios: Simplify save/restore mechanism. vgabios: Attempt to detect old x86emu and force a fault. vgabios: Don't interpret TAB character. vgabios: Refactor write_teletype and write_chars. vgabios: Make sure stdvga_list_modes() doesn't overrun the buffer. coreboot: Add support for a "links" file to have aliases in CBFS. Enhance nullTrailingSpace() so that it can also skip leading spaces. coreboot: Move links file processing to its own function. vgabios: Fix PMM allocation request size. smbios: Default all values to zero. usb: Only disable HID devices if both USB_MOUSE and USB_KEYBOARD are off. Minor - fix some typos in comments. Replace CONFIG_THREAD_OPTIONROMS with a runtime config setting. acpi: Move acpi parsing logic from acpi.c to biostables.c / paravirt.c. smbios: Move smbios parsing logic from smbios.c to biostables.c. Move PirAddr definition from pirtable.c to biostables.c. Use biostables.c for copying bios tables even when generating them. Document no new changes to pirtable.c, mptable.c, acpi.c, and smbios.c. apm: Remove old Bochs mechanism for shutdown/suspend/standby. Minor - remove unused includes from pci.c. vgabios: Add option to control use of standard VGA IO ports. vgabios: Split vgafb_scroll() into separate move and clear functions. vgabios: Rewrite vgafb.c graphics operations to set of 4 standard operators. vgabios: Introduce text_address(). vgabios: Add support for manipulating framebuffers in high memory. Set the color attribute when calling vgabios print character. vgabios: PMM scan was incorrectly depending on a zero %ds segment. coreboot: Make coreboot table searching available to other code. vgabios: Initial support for coreboot native vga vgabios. Minor - replace some tab characters that slipped into the code. Factor out smbios table walking from display_uuid to smbios_next(). vgabios: Define PlatformRunningOn to make compile happy. If an int 1587 call is made from an option rom, stay in bigreal mode. xhci: Use msleep() instead of mdelay() for port status delay. build: Make sure romlayout.o is recompiled on Kconfig change. Fix int 1589 calls when CONFIG_ENTRY_EXTRASTACK is enabled. acpi: Fix acpi dsdt compile warning. Laszlo Ersek (1): init_virtio_scsi(): reset the HBA before freeing its virtio ring Marcel Apfelbaum (3): resume: restore piix pm config registers after resume hw/pci: reserve IO and mem for pci-2-pci bridges with no devices attached hw/pci: check if pci2pci bridges implement optional limit registers Roger Pau Monne (1): ld: don't hardcode "-melf_i386" Sander Eikelenboom (1): Prevent build failure by not splitting line in the middle of arguments to checkrom.py script Signed-off-by: Gerd Hoffmann <[email protected]>

Now that ipxe has separate settings for load / boot banner timeouts re-enable the boot banner while keeping the load banner turned off, so we don't add a delay to non-pxe boots. git shortlog: Adrian Jamróz (2): [rhine] Rewrite VIA Rhine driver [velocity] Rewrite VIA Velocity driver Alex Williamson (4): [romprefix] Allow ROM banner timeout to be configured independently [autoboot] Enable infrastructure to specify an autoboot device location [prefix] Allow prefix to specify a PCI autoboot device location [romprefix] Store boot bus:dev.fn address as autoboot device location Alexander Chernyakhovsky (1): [ocsp] Handle OCSP responses that don't provide certificates Bo Yang (1): [build] Avoid strict-aliasing warning for gcc 4.3 Brandon Penglase (1): [build] Add vmware build target Christian Hesse (2): [build] Update build system for Syslinux 6.x [build] Fix LABEL name for .liso images Daniel P. Berrange (1): [zbin] Fix size used for memset in alloc_output_file Frediano Ziglio (1): [romprefix] Fix incorrect pointer offset in undiloader.S Kevin Tran (1): [tg3] Fix various tg3 issues Marin Hannache (8): [linux] Add missing #include <stddef.h> [legal] Add FILE_LICENCE for valgrind headers [legal] Add FILE_LICENCE for core/errno.c [legal] Add FILE_LICENCE for ath9k driver headers [legal] Add missing FILE_LICENCE declarations [cmdline] Add "poweroff" command [nfs] Add support for NFS protocol [nfs] Fix an issue with the selection of a local port Michael Brown (363): [netdevice] Separate VLAN support from presence of VLAN-supporting drivers [undi] Work around specific devices with known broken interrupt behaviour [menu] Prevent character code zero from acting as a shortcut key [menu] Prevent separators with shortcut keys from being selected [tftp] Allow TFTP block size to be controlled via the PXE TFTP API [pcbios] Add extra debugging messages when unhiding iPXE from memory [nbiprefix] Set up real-mode stack before jumping to .text16 [prefix] Use %cs as implicit parameter to uninstall() [pcbios] Inhibit all calls to INT 15,e820 and INT 15,e801 during POST [romprefix] Display only one "Ctrl-B" prompt per PCI device during POST [efi] Update to latest EDK2 headers [autoboot] Split main control flow out of main() into a new function ipxe() [efi] Add last_opened_snpdev() [efi] Expose downloaded images via EFI_SIMPLE_FILE_SYSTEM_PROTOCOL [efi] Add our own EFI_LOAD_FILE_PROTOCOL implementation [efi] Fix building with newer binutils [lkrnprefix] Allow relocation when no initrd is present [efi] Fix minor typos in efi_image.c [efi] Add EFI-specific debugging macros [int13] Split out ISO9660 and El Torito definitions to separate header files [smbios] Provide SMBIOS version number via smbios_version() [smbios] Mangle UUIDs for SMBIOS version 2.6 and newer [dhcp] Use PXE byte ordering for UUID in DHCP option 97 [uuid] Abstract UUID mangling code out to a separate uuid_mangle() function [efi] Provide efi_guid_ntoa() for printing EFI GUIDs [efi] Add "reboot" command for EFI [cmdline] Add ability to perform a warm reboot [efi] Enable "cpuid" command by default for EFI [cs89x0] Remove conflicting macro definitions [igbvf] Remove conflicting macro definitions [prism2] Use standard type names [efi] Fetch device path for loaded image during initialisation [settings] Expose build architecture and platform via settings [bios] Fix screen clearing on buggy BIOSes [intel] Remove hardcoded offsets for descriptor ring registers [intel] Expose functionality to be shared with intelx driver [intel] Add intelx driver for Intel 10 Gigabit Ethernet NICs [efi] Remove obsolete EFI I/O implementation using EFI_CPU_IO_PROTOCOL [libc] Use __einfo() tuple as first argument to EUNIQ() [libc] Redefine low 8 bits of error code as "platform error code" [efi] Perform meaningful error code conversions [efi] Add sample platform-generated error disambiguations [netdevice] Use link-layer address as part of RNG seed [build] Allow sparse to find compiler.h [build] Define __WINT_TYPE__ if necessary [build] Allow sparse to be invoked via "make C=1" [build] Avoid sparse undeclared symbol warning for PROVIDE_SYMBOL() [process] Mark process descriptor as static in PERMANENT_PROCESS [realtek] Print bad MAC address in debug message when inferring no EEPROM [build] Use -Wno-decl when running sparse [build] Fix uses of literal 0 as a NULL pointer [build] Fix dubious uses of bitwise operators [build] Default to short wchar_t in stddef.h [realtek] Use ID word to detect EEPROM presence [errdb] Strip platform error code for non-platform-generated errors [pxe] Convert external PXE API errors into iPXE platform-generated errors [realtek] Allow reaction time between writing RTL_CAPR and reading RTL_CR [romprefix] Report failure cause when unable to open payload [realtek] Ensure EEPROM writes reach chip before starting udelay() [dhcp] Remove obsolete bootp.h header [netdevice] Add netdev_tx_defer() to allow drivers to defer transmissions [realtek] Defer packets when no transmit descriptors are available [settings] Eliminate settings "tag magic" [smbios] Allow access to unreferenced SMBIOS strings [smbios] Allow access to multiple instances of SMBIOS structures [crypto] Report meaningful error when certificate chain validation fails [build] Use $(eval) if available [build] Provide "allXXXs" targets for all media on all platforms [build] Add efidrv.cab target for UEFI Secure Boot signing [netdevice] Add "chip" setting [netdevice] Reset MAC address when asked to clear the "mac" setting [realtek] Fix reopening of legacy-mode 8139 NIC [crypto] Accept OCSP responses containing multiple certificates [bzimage] Align initrd images to page boundary [bzimage] Fix spurious uninitialised-variable warning on some gcc versions [build] Work around bug in gcc >= 4.8 [autoboot] Use next-server from filename's settings block [tcp] Do not send RST for unrecognised connections [tcp] Fix comment to match code behaviour [test] Add self-tests for base16 [base16] Generalise base16_decode() to hex_decode() [settings] Use hex_decode() to parse hex settings [settings] Add "hexraw" setting type [netdevice] Add "bustype" and "busloc" settings [settings] Add "busdevfn" setting type [linux] Add support for accessing PCI configuration space via /proc/bus/pci [settings] Add config/settings.h [settings] Expose PCI configuration space via settings mechanism [settings] Make "netX" settings block function as a symbolic link [cmdline] Accept "netX" in iPXE commands [lotest] Include sequence number within loopback test packets [intel] Incorporate ring producer and consumer counters in diagnostics [base16] Ensure base16_encode() always terminates its result string [realtek] Report RX error detail in debug messages [script] Allow for backslash continuation of script lines [settings] Add support for navigation keys in "config" user interface [realtek] Allow extra space in RX buffers [ifmgmt] Avoid relying on global variable within ifcommon_exec() [script] Avoid trying to read final character of a zero-length string [legal] Add missing FILE_LICENCE declarations [build] Fix %.licence build target [settings] Clarify usage of the term "named setting" [settings] Avoid potentially large stack allocations [settings] Remove temporary name buffer parameter from parse_setting_name() [settings] Remove default_type parameter from parse_setting_name() [settings] Expose parse_setting_name() [settings] Change "not-found" semantics of fetch_setting_copy() [settings] Add fetchf_setting_copy() [settings] Eliminate call to fetchf_named_setting() in expand_settings() [settings] Eliminate call to store_named_setting() in nslookup.c [settings] Remove now-unused store_named_setting() [parseopt] Allow parsed option to be modified [parseopt] Move parse_settings() to parseopt.c [parseopt] Add parse_setting() [settings] Eliminate call to storef_named_setting() in choose_exec() [settings] Eliminate calls to {fetch,store}f_named_setting() in NVO commands [settings] Remove now-unused fetchf_named_setting() and storef_named_setting() [script] Allow initial whitespace on lines containing labels [script] Allow commands following a script label [settings] Introduce the generalised concept of a numeric setting [cmdline] Add "inc" command [pci] Add pci_find_next() to iterate over existent PCI devices [cmdline] Add "pciscan" command to allow iteration over PCI devices [udp] Move high-frequency debug messages to DBGLVL_EXTRA [ipv6] Rename sin_{family,port} to sin6_{family,port} in struct sockaddr_in6 [tcpip] Allow binding to unspecified privileged ports (below 1024) [settings] Expose CPUID instruction via settings mechanism [test] Add self-tests for snprintf() [settings] Allow numeric_setting_value() to handle long setting values [settings] Expose memory map via settings mechanism [params] Add support for the general concept of a form parameter list [cmdline] Add "params" and "param" commands to manage form parameter lists [uri] Allow URIs to incorporate a parameter list [http] Add support for HTTP POST [settings] Make built-in settings a linker table [linux] Give tap devices a name and bus type [ipv4] Generalise fragment reassembly mechanism [netdevice] Increase maximum network-layer address length [netdevice] Add concept of a network device index [netdevice] Add method for generating EUI-64 address from link-layer address [ethernet] Add support for generating multicast hash for IPv6 addresses [linux] Apply MAC address prior to registering network device [tcpip] Pass through network device to transport layer protocols [neighbour] Generalise concept of neighbour discovery [ipv4] Abstract out protocol-specific portions of "route" command [ipv6] Replace IPv6 stack [ipv6] Fix uninitialised-variable warning [tcp] Reduce path MTU to 1280 bytes [ipv6] Add inet6_aton() [build] Work around bug in gcc >= 4.8 [neighbour] Add nstat() function to print out neighbour table [cmdline] Add "nstat" command [tcp] Add AF_INET6 socket opener [udp] Add AF_INET6 socket opener [netdevice] Add find_netdev_by_index() [socket] Add concept of a generalised socket address converter [ipv4] Add IPv4 socket address converter [ipv6] Add IPv6 socket address converter [resolv] Use sock_aton() to allow parsing of arbitrary numeric addresses [icmp] Add support for sending ICMP echo requests [ping] Add concept of a ping socket [ping] Add generic ping mechanism [ping] Add top-level ping() function to ping a host [cmdline] Add "ping" command [ipv6] Handle IPv6 option length correctly [ipv6] Extract link layer addresses from router advertisements [ipv6] Support stateless address autoconfiguration (SLAAC) [ipv6] Treat sin6_scope_id consistently [ipv6] Automatically choose source for link-local and multicast destinations [pxe] Always retrieve cached DHCPACK and apply to relevant network device [ipv6] Add ndp_tx_router_solicitation() to send router solicitations [parseopt] Add parse_timeout() [netdevice] Make all net_driver methods optional [interface] Default to calling intf_restart() in response to intf_close() [job] Allow job_progress() to return an ongoing job status code, if known [monojob] Add timeout parameter to monojob_wait() [monojob] Report ongoing job status as overall return status on timeout [cmdline] Rewrite "sync" command to use monojob_wait() [autoboot] Fix shell banner timeout [ifmgmt] Rewrite iflinkwait() to use monojob_wait() [cmdline] Allow "if<xxx>" commands to take options [netdevice] Add generic concept of a network device configurator [dhcp] Add DHCP network device configurator [parseopt] Add parse_netdev_configurator() [ifmgmt] Add ifconf() to carry out network device configuration [autoboot] Use ifconf() to configure network device [cmdline] Add "ifconf" command [cmdline] Make "dhcp" command a synonym for "ifconf" [dhcp] Remove obsolete dhcp() function [ipv6] Add IPv6 network device configurator [cmdline] Generate command option help text automatically [pxe] Ensure cached DHCPACK is retrieved prior to network device creation [base16] Fix comparison of signed and unsigned integers [settings] Move user-class setting from dhcp.c to settings.c [ipv6] Use given source address only if it is not the unspecified address [ipv6] Add "ipv6" setting type [dhcpv6] Add basic support for stateful and stateless DHCPv6 [ipv6] Separate the concepts of prefix and address creation [dhcpv6] Allow stateful DHCPv6 to apply obtained IPv6 addresses [test] Include failing code within failed test result output [main] Defer "initialising devices" message until initialising devices [console] Pass escape sequence context to ANSI escape sequence handlers [pixbuf] Add generic concept of a pixel buffer [image] Allow for non-executable image formats [image] Add image_pixbuf() to create pixel buffer from image [pnm] Add support for PNM images [mucurses] Use "<ESC>[2J" ANSI escape sequence to clear screen [console] Allow console input and output to be disabled independently [console] Add concept of generic console configuration [cmdline] Add "console" command to configure console [fbcon] Add generic concept of a frame buffer console [vesafb] Add VESA frame buffer console [vesafb] Include raw status value within VBE error messages [vesafb] Skip modes for which we cannot get mode information [vesafb] Return meaningful error when no suitable mode is found [vesafb] Select an optimal mode, rather than the first acceptable mode [build] Fix building on OpenBSD 5.4 [console] Allow '?' as an intermediate byte in ANSI escape sequences [mucurses] Implement curs_set() to control cursor visibility [settings] Hide cursor when not actively editing a setting [menu] Hide cursor when displaying menu [fbcon] Add support for displaying a cursor [settings] Explicitly separate the concept of a completed fetched setting [settings] Allow for IPv6 setting types in non-IPv6 builds [settings] Add fetch_ipv6_setting() [dns] Add support for resolving IPv6 addresses via AAAA records [socket] Ensure socket address structure sizes are fixed [syslog] Add support for IPv6 syslog server [ipv6] Add support for resolving IPv6 addresses via the "nslookup" command [linux] Provide access to SMBIOS via /dev/mem [dhcpv6] Use DUID-UUID form of client DUID [settings] Force settings into alphabetical order within sections [settings] Allow for multiple definitions of each predefined setting [settings] Merge SETTING_IPv4 and SETTING_IPv6 [dhcpv6] Add DHCPv6 "filename" setting [ipv6] Expose NDP-provided settings (including the DNS server) [vesafb] Work around data corruption bug in bochs/qemu VBE implementation [settings] Adapt user interface to fit display size [login] Adapt user interface to fit display size [menu] Adapt user interface to fit display size [console] Allow consoles to update the recorded console size [fbcon] Update the console width and height after changing mode [lkrnprefix] Include iPXE version string in image header [fbcon] Always draw cursor using current foreground and background colours [console] Add centralised concept of colours and colour pairs [mucurses] Use centralised concept of colour pairs [cmdline] Add the "colour" and "cpair" commands [console] Add concept of a "magic" colour [vesafb] Set "magic" colour to transparent when a background picture is used [build] Remove obsolete check for GNU as version 2.9.1 [ipv6] Fix compilation under FreeBSD [build] Build ROMs used by qemu and VMware as part of "make all" [libc] Add isprint() [test] Add okx() macro taking an explicit file name and line number [deflate] Add support for DEFLATE decompression [deflate] Fix literal data length calculation [libc] Add abs() [uaccess] Add memcmp_user() [test] Rewrite pnm_ok() using okx() [test] Generalise pnm_ok() to pixbuf_ok() [png] Add support for PNG images [vesafb] Handle failures from fbcon_init() [fbcon] Centre background picture on console [fbcon] Allow for an arbitrary margin around the text area [vesafb] Allow for an arbitrary margin around the text area [cmdline] Rename "console" command's --bpp option to --depth [cmdline] Add margin options to the "console" command [fbcon] Allow ANSI CUP with missing arguments [cmdline] Always clear screen after reconfiguring console [intel] Add some missing PCI IDs [dns] Support DNS search lists [dns] Update end-of-name pointer after processing CNAME record [dhcp] Copy exactly the required length when resizing DHCP options [settings] Remove "uristring" setting type [params] Use reference counters for form parameter lists [uri] Refactor URI parsing and formatting [image] Ensure every image has a fully resolved URI [tcpip] Add IP statistics collection as per RFC 4293 [cmdline] Add the "ipstat" command [prefix] Ignore PCI autoboot device location if set to 00:00.0 [tcpip] Provide tcpip_netdev() to determine the transmitting network device [tcpip] Provide tcpip_mtu() to determine the maximum transmission unit [tcp] Calculate correct MSS from peer address [bzimage] Report exact initrd length via bzImage header [realtek] Include link status register details in debug messages [romprefix] Do not clobber stack segment when returning to BIOS [netdevice] Mark devices as open before calling open() method [tcp] Update window even if ACK does not acknowledge new data [linux] Avoid starting currticks() from zero every time [http] Use a retry timer to trigger retried requests [http] Automatically retry request on a 503 Service Unavailable [mii] Add mii_dump() to dump all MII registers [realtek] Dump all MII register contents when link status changes [monojob] Reset timeout when progress is made [image] Add "--timeout" parameter to image downloading commands [efi] Allow for 64-bit EFI_STATUS codes [efi] Allow driver to be unloaded [efi] Connect driver to devices as part of installation [console] Fix display of characters with top bit set [lotest] Allow loopback testing on shared networks [netdevice] Notify upper-layer drivers when RX processing is (un)frozen [efi] Avoid accidentally calling main() twice [efi] Unload our own image before exiting UEFI application [efi] Disable SNP devices when running iPXE as the application [realtek] Add ability to dump all internal registers [realtek] Clear bit 24 of RCR [bios] Fix screen clearing on even more buggy BIOSes [http] Accept Content-Length header with trailing whitespace [crypto] Remove dynamically-allocated storage for certificate name [crypto] Remove dynamically-allocated storage for certificate OCSP URI [build] Remove long-obsolete mechanism for wrapping embedded images [build] Disable ccache for all relevant build targets [build] Add dependency of generated files upon Makefile [crypto] Add pubkey_match() to check for matching public/private key pairs [crypto] Generalise X.509 cache to a full certificate store [crypto] Use fingerprint when no common name is available for debug messages [crypto] Allow signed timestamp error margin to be configured at build time [test] Rewrite X.509 tests using okx() [test] Rewrite CMS tests using okx() [test] Add tests for x509_check_name() [test] Add subject alternative names to X.509 server test certificate [crypto] Add support for subjectAltName and wildcard certificates [crypto] Allow wildcard matches on commonName as well as subjectAltName [intel] Avoid completely filling the TX descriptor ring [test] Rewrite TCP/IP tests using okx() [test] Add self-tests for flsl() [libc] Add inline assembly implementation of flsl() using BSR instruction [test] Add tests for 64-bit division [build] Fix __libgcc attribute for recent gcc versions [test] Check for correct -mrtd assumption on libgcc arithmetic functions [libc] Add isqrt() function to find integer square roots [libc] Add flsll() [profile] Add generic profiling infrastructure [cmdline] Add "profstat" command to display profiling statistics [netdevice] Profile common operations [intel] Profile common virtual machine operations [intel] Push new RX descriptors in batches [librm] Use genuine real mode to accelerate operation in virtual machines [ipv4] Profile transmit and receive datapaths [tcp] Profile transmit and receive datapaths [http] Profile receive datapath [downloader] Profile receive datapath [build] Allow for a debug level of zero [librm] Allow interrupts in protected mode [comboot] Use built-in interrupt reflector [pcbios] Do not switch to real mode to check for timer interrupt [pcbios] Do not switch to real mode to sleep the CPU [test] Print out profiling statistics after a successful test run [librm] Add a profiling self-test for measuring mode transition times [librm] Add meaningful labels at section changes [librm] Speed up protected-to-real mode transition under KVM [librm] Speed up real-to-protected mode transition under KVM [librm] Speed up protected-mode calls under KVM [profile] Provide methods for profiling individual stages of operations [librm] Add profiling self-tests for complete real_call and prot_call cycles [pxe] Profile all PXE API calls [pxe] Profile UNDI transmit datapath [pxe] Work around missing PXENV_UNDI_OPEN only when necessary [undi] Profile all PXE API calls [undi] Profile transmit and receive datapaths [undi] Report any PXENV_UNDI_ISR errors via netdev_rx_err() [undi] Do not switch to real mode to check for NIC interrupt [undi] Place an upper limit on the number of PXENV_UNDI_ISR calls per poll Peter Pickford (1): [serial] Enable UART FIFOs Robin Smidsrød (2): [settings] Add "version" builtin setting [build] Include ipxe.pxe in default build Stefan Hajnoczi (2): [netdevice] Add vlan_tag() to get the VLAN tag of a network device [iscsi] Include 802.1Q VLAN identifier in iBFT Thomas Miletich (3): [3c90x] Stall upload engine before setting RX ring address [3c90x] Don't round up transmit packet length [3c90x] Fix High-MTU packet reception lolipop (1): [intel] Add Intel I217 Gigabit Ethernet PCI ID Signed-off-by: Gerd Hoffmann <[email protected]>

This reverts commit f915db0. This commit is broken because it does not account for the build tree and the source tree being different, and can cause build failures for out-of-tree builds. Revert it until we can identify a better solution to the problem. Signed-off-by: Peter Maydell <[email protected]> Message-id: [email protected] Acked-by: Kevin Wolf <[email protected]>

…' into staging Fix BT zero flag for new Intel manuals # gpg: Signature made Mon 12 May 2014 22:26:01 BST using RSA key ID 4DD0279B # gpg: Can't check signature: public key not found * remotes/rth/tags/pull-tgt-i386-20140512: target-i386: Preserve the Z bit for bt/bts/btr/btc Signed-off-by: Peter Maydell <[email protected]>

* remotes/kvm/uq/master: pc: port 92 reset requires a low->high transition cpu: make CPU_INTERRUPT_RESET available on all targets apic: do not accept SIPI on the bootstrap processor target-i386: preserve FPU and MSR state on INIT target-i386: fix set of registers zeroed on reset kvm: forward INIT signals coming from the chipset kvm: reset state from the CPU's reset method target-i386: the x86 CPL is stored in CS.selector - auto update hflags accordingly. target-i386: set eflags prior to calling cpu_x86_load_seg_cache() in seg_helper.c target-i386: set eflags and cr0 prior to calling cpu_x86_load_seg_cache() in smm_helper.c target-i386: set eflags prior to calling svm_load_seg_cache() in svm_helper.c pci-assign: limit # of msix vectors pci-assign: Fix a bug when map MSI-X table memory failed kvm: make one_reg helpers available for everyone target-i386: Remove unused data from local array Signed-off-by: Peter Maydell <[email protected]>

…0140513' into staging target-arm queue: * update libvixl to 1.4 * remove version_minimum_id_old from ARM devices * stellaris_enet tx/rx/migration overhaul * various minor fixes for coverity issues # gpg: Signature made Tue 13 May 2014 16:25:12 BST using RSA key ID 14360CDE # gpg: Good signature from "Peter Maydell <[email protected]>" * remotes/pmaydell/tags/pull-target-arm-20140513: hw/arm/omap_gpmc: Avoid buffer overrun filling prefetch FIFO hw/arm/stellaris: Correct handling of GPTM TAR register hw/timer/exynos4210_mct: Avoid overflow in exynos4210_ltick_recalc_count hw/dma/omap_dma: Add (uint32_t) casts when shifting uint16_t by 16 hw/arm/omap1: Avoid unintended sign extension writing omap_rtc YEARS_REG hw/net/cadence_gem: Remove dead code hw/intc/allwinner-a10-pic: Add missing 'break' target-arm/helper.c: Don't flush the TLB if SCTLR is rewritten unchanged hw/net/stellaris_enet: Convert to vmstate hw/net/stellaris_enet: Get rid of rx_fifo pointer hw/net/stellaris_enet: Fix debug format strings hw/net/stellaris_enet: Correctly implement the TR and THR registers hw/net/stellaris_enet: Rewrite tx fifo handling code hw/net/stellaris_enet: Correct handling of packet padding hw/net/stellaris_enet: Restructure tx_fifo code to avoid buffer overrun savevm: Remove all the unneeded version_minimum_id_old (arm) disas/libvixl: Update to libvixl 1.4 Signed-off-by: Peter Maydell <[email protected]>

* remotes/rth/fix-mov: tcg: Fix tcg_reg_alloc_mov vs no-op truncation Signed-off-by: Peter Maydell <[email protected]>

While we still require the LONG DISPLACEMENT facility, defaults have moved on since then. Don't override the system compiler, whose default may be set to z9-109 or later. Acked-by: Christian Borntraeger <[email protected]> Signed-off-by: Richard Henderson <[email protected]>

Signed-off-by: Richard Henderson <[email protected]>

…40515' into staging migration/next for 20140515 # gpg: Signature made Thu 15 May 2014 02:32:25 BST using RSA key ID 5872D723 # gpg: Can't check signature: public key not found * remotes/juanquintela/tags/migration/20140515: usb: fix up post load checks migration: show average throughput when migration finishes savevm: Remove all the unneeded version_minimum_id_old (rest) savevm: Remove all the unneeded version_minimum_id_old (usb) Split ram_save_block arch_init: Simplify code for load_xbzrle() Signed-off-by: Peter Maydell <[email protected]>

…ging virtio update looks like a quiet week minor bugfix in virtio by myself Signed-off-by: Michael S. Tsirkin <[email protected]> # gpg: Signature made Tue 13 May 2014 14:37:23 BST using RSA key ID D28D5469 # gpg: Good signature from "Michael S. Tsirkin <[email protected]>" # gpg: aka "Michael S. Tsirkin <[email protected]>" * remotes/mst/tags/for_upstream: virtio: allow mapping up to max queue size Signed-off-by: Peter Maydell <[email protected]>

…taging seabios: update to 1.7.5-rc1 ipxe: update to current git # gpg: Signature made Thu 15 May 2014 13:29:37 BST using RSA key ID D3E87138 # gpg: Good signature from "Gerd Hoffmann (work) <[email protected]>" # gpg: aka "Gerd Hoffmann <[email protected]>" # gpg: aka "Gerd Hoffmann (private) <[email protected]>" * remotes/kraxel/tags/pull-roms-2: ipxe: update to current git seabios: update to 1.7.5-rc1 Signed-off-by: Peter Maydell <[email protected]>

…' into staging tcg/s390 updates # gpg: Signature made Thu 15 May 2014 17:24:40 BST using RSA key ID 4DD0279B # gpg: Can't check signature: public key not found * remotes/rth/tags/pull-tcg-s390-20140515: tcg-s390: Implement direct chaining of TBs tcg-s390: Don't force -march=z990 tcg-s390: Improve setcond tcg-s390: Allow immediate operands to add2 and sub2 tcg-s390: Implement tcg_register_jit tcg-s390: Use more risbg in the tlb sequence tcg-s390: Move ldst helpers out of line tcg-s390: Convert to new ldst opcodes tcg-s390: Integrate endianness into TCGMemOp tcg-s390: Convert to TCGMemOp tcg-s390: Fix off-by-one in wraparound andi Signed-off-by: Peter Maydell <[email protected]>

Signed-off-by: Wanlong Gao <[email protected]> Reviewed-by: Eduardo Habkost <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]> Signed-off-by: Hu Tao <[email protected]>

If the total number of the assigned numa nodes memory is not equal to the assigned ram size, it will write the wrong data to ACPI table, then the guest will ignore the wrong ACPI table and recognize all memory to one node. It's buggy, we should check it to ensure that we write the right data to ACPI table. Signed-off-by: Wanlong Gao <[email protected]> Reviewed-by: Eduardo Habkost <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]> Signed-off-by: Hu Tao <[email protected]>

Add the numa_info structure to contain the numa nodes memory, VCPUs information and the future added numa nodes host memory policies. Reviewed-by: Eduardo Habkost <[email protected]> Signed-off-by: Andre Przywara <[email protected]> Signed-off-by: Wanlong Gao <[email protected]> [Fix hw/ppc/spapr.c - Paolo] Signed-off-by: Paolo Bonzini <[email protected]> Signed-off-by: Hu Tao <[email protected]>

Signed-off-by: Wanlong Gao <[email protected]> Signed-off-by: Igor Mammedov <[email protected]> Tested-by: Eduardo Habkost <[email protected]> Reviewed-by: Eduardo Habkost <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]> Signed-off-by: Hu Tao <[email protected]>

libnuma choosed 128 for MAX_NODES, so we follow libnuma here. Signed-off-by: Wanlong Gao <[email protected]> Reviewed-by: Eduardo Habkost <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]> Signed-off-by: Hu Tao <[email protected]>

The -numa option documentation in qemu's manpage lacks the command-line options and some information regarding how it relates to options -m and -smp. This commit fills in the missing text. Signed-off-by: Luiz Capitulino <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]> Signed-off-by: Hu Tao <[email protected]>

Follow the lines of the HMP implementation, using OptsVisitor to parse the options. This gives access to OptsVisitor's rich parsing of integer lists. Signed-off-by: Paolo Bonzini <[email protected]> Signed-off-by: Hu Tao <[email protected]>

Add object to /objects before calling user_creatable_complete() handler, so that object might be able to call object_get_canonical_path() in its completion handler. Signed-off-by: Igor Mammedov <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]> Signed-off-by: Hu Tao <[email protected]>

Use QERR_INVALID_PARAMETER_VALUE for consistency. Signed-off-by: Paolo Bonzini <[email protected]> Signed-off-by: Hu Tao <[email protected]>

Reviewed-by: Michael S. Tsirkin <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]> Signed-off-by: Hu Tao <[email protected]>

Signed-off-by: Paolo Bonzini <[email protected]> Signed-off-by: Hu Tao <[email protected]>

Provides framework for splitting host RAM allocation/ policies into a separate backend that could be used by devices. Initially only legacy RAM backend is provided, which uses memory_region_init_ram() allocator and compatible with every CLI option that affects memory_region_init_ram(). Signed-off-by: Igor Mammedov <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]> Signed-off-by: Hu Tao <[email protected]>

This option provides the infrastructure for binding guest NUMA nodes to host NUMA nodes. For example: -object memory-ram,size=1024M,policy=bind,host-nodes=0,id=ram-node0 \ -numa node,nodeid=0,cpus=0,memdev=ram-node0 \ -object memory-ram,size=1024M,policy=interleave,host-nodes=1-3,id=ram-node1 \ -numa node,nodeid=1,cpus=1,memdev=ram-node1 The option replaces "-numa node,mem=". Signed-off-by: Paolo Bonzini <[email protected]> Signed-off-by: Hu Tao <[email protected]>

Split the internal interface in exec.c to a separate function, and push the check on mem_path up to memory_region_init_ram. Signed-off-by: Paolo Bonzini <[email protected]> Signed-off-by: Hu Tao <[email protected]>

Like the previous patch did in exec.c, split memory_region_init_ram and memory_region_init_ram_from_file, and push mem_path one step further up. Other RAM regions than system memory will now be backed by regular RAM. Also, boards that do not use memory_region_allocate_system_memory will not support -mem-path anymore. This can be changed before the patches are merged by migrating boards to use the function. Signed-off-by: Paolo Bonzini <[email protected]> Signed-off-by: Hu Tao <[email protected]>

Right now, -mem-path will fall back to RAM-based allocation in some cases. This should never happen with "-object memory-file", prepare the code by adding correct error propagation. Signed-off-by: Paolo Bonzini <[email protected]> Signed-off-by: Hu Tao <[email protected]>

So that backends can use it. Signed-off-by: Paolo Bonzini <[email protected]> Signed-off-by: Hu Tao <[email protected]>

Prepare for adding more flags. The "_MASK" suffix is unique, kill it. Signed-off-by: Paolo Bonzini <[email protected]> Signed-off-by: Hu Tao <[email protected]>

Signed-off-by: Paolo Bonzini <[email protected]> Signed-off-by: Hu Tao <[email protected]>

This allows the superclass to set various policies on the memory region that the subclass creates. Signed-off-by: Paolo Bonzini <[email protected]> Signed-off-by: Hu Tao <[email protected]>

Signed-off-by: Paolo Bonzini <[email protected]> Signed-off-by: Hu Tao <[email protected]>

And allow preallocation of file-based memory even without -mem-prealloc. Some care is necessary because -mem-prealloc does not allow disabling preallocation for hostmem-file. Signed-off-by: Paolo Bonzini <[email protected]> Signed-off-by: Hu Tao <[email protected]>

A new "share" property can be used with the "memory-file" backend to map memory with MAP_SHARED instead of MAP_PRIVATE. Signed-off-by: Paolo Bonzini <[email protected]> Signed-off-by: Hu Tao <[email protected]>

Add detection of libnuma (mostly contained in the numactl package) to the configure script. Can be enabled or disabled on the command line, default is use if available. Signed-off-by: Andre Przywara <[email protected]> Signed-off-by: Wanlong Gao <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]> Signed-off-by: Hu Tao <[email protected]>

Signed-off-by: Hu Tao <[email protected]> [Raise errors on setting properties if !CONFIG_NUMA. Add BUILD_BUG_ON checks. - Paolo] Signed-off-by: Paolo Bonzini <[email protected]> Signed-off-by: Marcelo Tosatti <[email protected]> Signed-off-by: Hu Tao <[email protected]> Signed-off-by: Hu Tao <[email protected]>

Signed-off-by: Hu Tao <[email protected]>

…6List Signed-off-by: Hu Tao <[email protected]>

Add qmp command query-memdev to query for information of memory devices Signed-off-by: Hu Tao <[email protected]>

This is the hmp counterpart of qmp query-memdev. Signed-off-by: Hu Tao <[email protected]>

Add a function to check if the eventfd capability is present in KVM in the host kernel. Signed-off-by: Antonios Motakis <[email protected]> Signed-off-by: Nikolay Nikolaev <[email protected]>

This function will attempt to read data from the chardev trying to fill the buffer up to the given length. Add tcp_chr_disconnect to reuse disconnection code where needed. Signed-off-by: Antonios Motakis <[email protected]> Signed-off-by: Nikolay Nikolaev <[email protected]>

This will set an array of file descriptors to the internal structures. The next time a message is send the array will be send as ancillary data. This feature works on the UNIX domain socket backend only. Signed-off-by: Antonios Motakis <[email protected]> Signed-off-by: Nikolay Nikolaev <[email protected]>

This extends the existing qemu_chr_fe_get_msgfd by allowing to read a set of fds. The function for receiving the fds - unix_process_msgfd is extended to allocate the needed array size. Signed-off-by: Antonios Motakis <[email protected]> Signed-off-by: Nikolay Nikolaev <[email protected]>

This is used to detect that the remote end has disconnected. Just call tcp_char_disconnect on receiving this event. Signed-off-by: Antonios Motakis <[email protected]> Signed-off-by: Nikolay Nikolaev <[email protected]>

Generalize the features get/ack to be used for both vhost-net and vhost-scsi. In vhost-net add vhost_net_get_feature_bits to select the feature bit set depending on the NetClient kind. Signed-off-by: Nikolay Nikolaev <[email protected]>

The poll callback needs to be called when bringing up or down the vhost_net instance. As it is not mandatory for an NetClient to implement it, invoke it only when it is set. Signed-off-by: Antonios Motakis <[email protected]> Signed-off-by: Nikolay Nikolaev <[email protected]>

This decouples virtio-net from the TAP netdev backend and allows support for other backends to be implemented. Signed-off-by: Antonios Motakis <[email protected]> Signed-off-by: Nikolay Nikolaev <[email protected]>

vhost_dev_init will replace devfd and devpath with a single opaque argument. This is initialised with a file descriptor. When TAP is used (through vhost_net), open /dev/vhost-net and pass the fd as an opaque parameter in VhostNetOptions. The same applies to vhost-scsi - open /dev/vhost-scsi and pass the fd. Signed-off-by: Antonios Motakis <[email protected]> Signed-off-by: Nikolay Nikolaev <[email protected]>

Decouple vhost from the Linux kernel by introducing vhost_ops. The intention is to provide different backends - a 'kernel' backend based on the ioctl interface, and an 'user' backend based on a UNIX domain socket and shared memory interface. Signed-off-by: Antonios Motakis <[email protected]> Signed-off-by: Nikolay Nikolaev <[email protected]>

Use vhost_set_backend_type to initialise a proper vhost_ops structure. In vhost_net_init and vhost_net_start_one call conditionally TAP related initialisation depending on the vhost backend type. Signed-off-by: Antonios Motakis <[email protected]> Signed-off-by: Nikolay Nikolaev <[email protected]>

The initialization takes a chardev backed by a unix domain socket. It should implement qemu_fe_set_msgfds in order to be able to pass file descriptors to the remote process. Each ioctl request of vhost-kernel has a vhost-user message equivalent, which is sent over the control socket. The general approach is to copy the data from the supplied argument pointer to a designated field in the message. If a file descriptor is to be passed it will be placed in the fds array for inclusion in the sendmsg control header. VHOST_SET_MEM_TABLE ignores the supplied vhost_memory structure and scans the global ram_list for ram blocks with a valid fd field set. This would be set when the '-object memory-file' option with share=on property is used. Signed-off-by: Antonios Motakis <[email protected]> Signed-off-by: Nikolay Nikolaev <[email protected]>

Handle the feature bits negotiation when using vhost-user. Allow the underlying implementation to have a finer control over all the bits except the VIRTIO_NET_F_MAC. Signed-off-by: Nikolay Nikolaev <[email protected]>

Add a new QEMU netdev backend that is intended to invoke vhost_net with the vhost-user backend. It uses an Unix socket chardev to establish a communication with the 'slave' (client and server mode supported). At runtime the netdev will handle OPEN/CLOSE events from the chardev. Upon disconnection it will set link_down accordingly and notify virtio-net; the virtio-net interface will go down. Signed-off-by: Antonios Motakis <[email protected]> Signed-off-by: Nikolay Nikolaev <[email protected]>

The supplied chardev id will be inspected for supported options. Only a socket backend, with a set path (i.e. a Unix socket) and optionally the server parameter set, will be allowed. Other options (nowait, telnet) will make the chardev unusable and the netdev will not be initialised. Additional checks for validity: - requires `-numa node,memdev=..` - requires `-device virtio-net-*` The `vhostforce` option is used to force vhost-net when we deal with non-MSIX guests. Signed-off-by: Antonios Motakis <[email protected]> Signed-off-by: Nikolay Nikolaev <[email protected]>

This document describes the basic message format used by vhost-user for communication over a unix domain socket. The protocol is based on the existing ioctl interface used for the kernel version of vhost. Signed-off-by: Antonios Motakis <[email protected]> Signed-off-by: Nikolay Nikolaev <[email protected]>

chardev depends on lots of external symbols that are not necessarily needed to be able to use, for example, 'socket chardev'. So add stubs for these functions: - bdrv_commit_all - qemu_chr_open_msmouse - is_daemonized - qemu_add_machine_init_done_notifier - monitor_init - qemu_notify_event - vc_init and this array: - serial_hds Signed-off-by: Antonios Motakis <[email protected]> Signed-off-by: Nikolay Nikolaev <[email protected]>

This test creates a 'server' chardev to listen for vhost-user messages. Once VHOST_USER_SET_MEM_TABLE is received it mmaps each received region, and read 1k bytes from it. The read data is compared to data from readl. The test requires hugetlbfs to be already mounted and writable. The mount point defaults to '/hugetlbfs' and can be specified via the environment variable QTEST_HUGETLBFS_PATH. The rom pc-bios/pxe-virtio.rom is used to instantiate a virtio pcicontroller. Signed-off-by: Antonios Motakis <[email protected]> Signed-off-by: Nikolay Nikolaev <[email protected]>

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Increase Virtio-net vring size #1

Increase Virtio-net vring size #1

Commits on May 7, 2014

Commits on May 8, 2014

Commits on May 9, 2014

Commits on May 12, 2014

Commits on May 13, 2014

Commits on May 14, 2014

Commits on May 15, 2014

Commits on May 21, 2014

Commits on May 27, 2014

Commits on Jun 8, 2014

Increase Virtio-net vring size #1

Are you sure you want to change the base?

Increase Virtio-net vring size #1

Commits on May 7, 2014

Commits on May 8, 2014

Commits on May 9, 2014

Commits on May 12, 2014

Commits on May 13, 2014

Commits on May 14, 2014

Commits on May 15, 2014

Commits on May 21, 2014

Commits on May 27, 2014

Commits on Jun 8, 2014