aboutsummaryrefslogtreecommitdiff
path: root/hw/pci
AgeCommit message (Collapse)AuthorFilesLines
2015-09-18Fix bad error handling after memory_region_init_ram()Markus Armbruster1-1/+1
Symptom: $ qemu-system-x86_64 -m 10000000 Unexpected error in ram_block_add() at /work/armbru/qemu/exec.c:1456: upstream-qemu: cannot set up guest memory 'pc.ram': Cannot allocate memory Aborted (core dumped) Root cause: commit ef701d7 screwed up handling of out-of-memory conditions. Before the commit, we report the error and exit(1), in one place, ram_block_add(). The commit lifts the error handling up the call chain some, to three places. Fine. Except it uses &error_abort in these places, changing the behavior from exit(1) to abort(), and thus undoing the work of commit 3922825 "exec: Don't abort when we can't allocate guest memory". The three places are: * memory_region_init_ram() Commit 4994653 (right after commit ef701d7) lifted the error handling further, through memory_region_init_ram(), multiplying the incorrect use of &error_abort. Later on, imitation of existing (bad) code may have created more. * memory_region_init_ram_ptr() The &error_abort is still there. * memory_region_init_rom_device() Doesn't need fixing, because commit 33e0eb5 (soon after commit ef701d7) lifted the error handling further, and in the process changed it from &error_abort to passing it up the call chain. Correct, because the callers are realize() methods. Fix the error handling after memory_region_init_ram() with a Coccinelle semantic patch: @r@ expression mr, owner, name, size, err; position p; @@ memory_region_init_ram(mr, owner, name, size, ( - &error_abort + &error_fatal | err@p ) ); @script:python@ p << r.p; @@ print "%s:%s:%s" % (p[0].file, p[0].line, p[0].column) When the last argument is &error_abort, it gets replaced by &error_fatal. This is the fix. If the last argument is anything else, its position is reported. This lets us check the fix is complete. Four positions get reported: * ram_backend_memory_alloc() Error is passed up the call chain, ultimately through user_creatable_complete(). As far as I can tell, it's callers all handle the error sanely. * fsl_imx25_realize(), fsl_imx31_realize(), dp8393x_realize() DeviceClass.realize() methods, errors handled sanely further up the call chain. We're good. Test case again behaves: $ qemu-system-x86_64 -m 10000000 qemu-system-x86_64: cannot set up guest memory 'pc.ram': Cannot allocate memory [Exit 1 ] The next commits will repair the rest of commit ef701d7's damage. Signed-off-by: Markus Armbruster <armbru@redhat.com> Message-Id: <1441983105-26376-3-git-send-email-armbru@redhat.com> Reviewed-by: Peter Crosthwaite <crosthwaite.peter@gmail.com>
2015-09-16pci: remove Link Training error from AER error listPaolo Bonzini1-4/+0
The spec says: Undefined – The value read from this bit is undefined. In previous versions of this specification, this bit was used to indicate a Link Training Error. System software must ignore the value read from this bit. System software is permitted to write any value to this bit. Do not allow injecting it. Suggested-by: Michael S. Tsirkin <mst@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2015-09-11maint: remove unused include for strings.hDaniel P. Berrange1-1/+0
A number of files were including strings.h but not using any of the functions it provides Signed-off-by: Daniel P. Berrange <berrange@redhat.com> Reviewed-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
2015-09-10hw/pci: fix pci_update_mappings() trace eventsLaszlo Ersek1-2/+2
The current trace prototypes and (matching) trace calls lead to "unorthodox" PCI BDF notation in at least the stderr trace backend. For example, the four BARs of a QXL video card at 00:01.0 (bus 0, slot 1, function 0) are traced like this (PID and timestamps removed): pci_update_mappings_add d=0x7f14a73bf890 00:00.1 0,0x84000000+0x4000000 pci_update_mappings_add d=0x7f14a73bf890 00:00.1 1,0x80000000+0x4000000 pci_update_mappings_add d=0x7f14a73bf890 00:00.1 2,0x88200000+0x2000 pci_update_mappings_add d=0x7f14a73bf890 00:00.1 3,0xd060+0x20 The slot and function values are in reverse order. Stick with the conventional BDF notation. Cc: "Michael S. Tsirkin" <mst@redhat.com> Cc: Don Koch <dkoch@verizon.com> Cc: qemu-trivial@nongnu.org Fixes: 7828d75045 Signed-off-by: Laszlo Ersek <lersek@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
2015-09-10pci: Fix pci_device_iommu_address_space() bus propagationBenjamin Herrenschmidt1-8/+5
he current code walks up the bus tree for an iommu, however it passes to the iommu_fn() callback the bus/devfn of the immediate child of the level where the callback was found, rather than the original bus/devfn where the search started from. This prevents iommu's like POWER8 (and in fact also Q35) to properly provide an address space for a subset of devices that aren't immediate children of the iommu. PCIe carries the originator bdfn acccross to the iommu on all DMA transactions, so we must be able to properly identify devices at all levels. This changes the function pci_device_iommu_address_space() to pass the original pointers to the iommu_fn() callback instead. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
2015-09-07hw/pci: Use pow2ceil() rather than hand-calculationPeter Maydell2-6/+2
A couple of places in hw/pci use an inline calculation to round a size up to the next largest power of 2. We have a utility routine for this, so use it. (The behaviour of the old code is different if the size value is 0 -- it would leave it as 0 rather than rounding up to 1, but in both cases we know the size can't be 0. In the case where the size value had bit 31 set, the old code would invoke undefined behaviour; the new code will give a result of 0. Presumably that could never happen either.) Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Message-id: 1437741192-20955-2-git-send-email-peter.maydell@linaro.org
2015-08-13pci: allow 0 address for PCI IO/MEM regionsLaurent Vivier1-3/+9
Some kernels program a 0 address for io regions. PCI 3.0 spec section 6.2.5.1 doesn't seem to disallow this. based on patch by Michael Roth <mdroth@linux.vnet.ibm.com> Add pci_allow_0_addr in MachineClass to conditionally allow addr 0 for pseries, as this can break other architectures. This patch allows to hotplug PCI card in pseries machine, as the first added card BAR0 is always set to 0 address. This as a temporary hack, waiting to fix PCI memory priorities for more machine types... Signed-off-by: Laurent Vivier <lvivier@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
2015-07-20pci_add_capability: remove duplicate commentsChen Hanxiao1-4/+2
Signed-off-by: Chen Hanxiao <chenhanxiao@cn.fujitsu.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
2015-07-08pcie: Set the "link active" in the link status registerBenjamin Herrenschmidt1-1/+1
Some firmwares can test that and assume the device hasn't come up if that bit isn't set Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
2015-06-23Print error when failing to load PCI config dataDr. David Alan Gilbert1-0/+4
When loading migration fails due to a disagreement about PCI config data we don't currently get any errors explaining that was the cause of the problem or which byte in the config data was at fault. Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
2015-06-22Include qapi/qmp/qerror.h exactly where neededMarkus Armbruster3-2/+1
In particular, don't include it into headers. Signed-off-by: Markus Armbruster <armbru@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Luiz Capitulino <lcapitulino@redhat.com>
2015-06-22qerror: Move #include out of qerror.hMarkus Armbruster1-0/+1
Signed-off-by: Markus Armbruster <armbru@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Luiz Capitulino <lcapitulino@redhat.com>
2015-06-22qerror: Clean up QERR_ macros to expand into a single stringMarkus Armbruster1-1/+1
These macros expand into error class enumeration constant, comma, string. Unclean. Has been that way since commit 13f59ae. The error class is always ERROR_CLASS_GENERIC_ERROR since the previous commit. Clean up as follows: * Prepend every use of a QERR_ macro by ERROR_CLASS_GENERIC_ERROR, and delete it from the QERR_ macro. No change after preprocessing. * Rewrite error_set(ERROR_CLASS_GENERIC_ERROR, ...) into error_setg(...). Again, no change after preprocessing. Signed-off-by: Markus Armbruster <armbru@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Luiz Capitulino <lcapitulino@redhat.com>
2015-06-22qobject: Use 'bool' inside qdictEric Blake1-2/+2
Now that qbool is fixed, let's fix getting and setting a bool value to a qdict member to also use C99 bool rather than int. I audited all callers to ensure that the changed return type will not cause any changed semantics. Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: Alberto Garcia <berto@igalia.com> Acked-by: Luiz Capitulino <lcapitulino@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
2015-06-12virtio-vga: add '-vga virtio' supportGerd Hoffmann1-0/+2
Some convinience fluff: Add support for '-vga virtio', also add virtio-vga to the list of vga cards so '-device virtio-vga' will turn off the default vga. Written by Dave Airlie and Gerd Hoffmann. Signed-off-by: Dave Airlie <airlied@redhat.com> Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2015-06-04Merge remote-tracking branch 'remotes/mst/tags/for_upstream' into stagingPeter Maydell1-12/+66
pc, acpi, virtio, tpm This includes pxb support by Marcel, as well as multiple enhancements all over the place. Signed-off-by: Michael S. Tsirkin <mst@redhat.com> # gpg: Signature made Thu Jun 4 11:51:02 2015 BST using RSA key ID D28D5469 # gpg: Good signature from "Michael S. Tsirkin <mst@kernel.org>" # gpg: aka "Michael S. Tsirkin <mst@redhat.com>" * remotes/mst/tags/for_upstream: (28 commits) vhost: logs sharing hw/acpi: piix4_pm_init(): take fw_cfg object no more hw/acpi: move "etc/system-states" fw_cfg file from PIIX4 to core hw/acpi: acpi_pm1_cnt_init(): take "disable_s3" and "disable_s4" pc-dimm: don't assert if pc-dimm alignment != hotpluggable mem range size docs: Add PXB documentation apci: fix PXB behaviour if used with unsupported BIOS hw/pxb: add numa_node parameter hw/pci: add support for NUMA nodes hw/pxb: add map_irq func hw/pci: inform bios if the system has extra pci root buses hw/pci: introduce PCI Expander Bridge (PXB) hw/pci: removed 'rootbus nr is 0' assumption from qmp_pci_query hw/acpi: remove from root bus 0 the crs resources used by other buses. hw/acpi: add _CRS method for extra root busses hw/apci: add _PRT method for extra PCI root busses hw/acpi: add support for i440fx 'snooping' root busses hw/pci: extend PCI config access to support devices behind PXB hw/i386: query only for q35/pc when looking for pci host bridge hw/pci: made pci_bus_num a PCIBusClass method ... Conflicts: hw/i386/pc_piix.c Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2015-06-04Merge remote-tracking branch 'remotes/agraf/tags/signed-ppc-for-upstream' ↵Peter Maydell1-1/+1
into staging Patch queue for ppc - 2015-06-03 Highlights this time around: - sPAPR: endian fixes, speedups, bug fixes, hotplug basics - add default ram size capability for machines (sPAPR defaults to 512MB now) # gpg: Signature made Wed Jun 3 22:59:09 2015 BST using RSA key ID 03FEDC60 # gpg: Good signature from "Alexander Graf <agraf@suse.de>" # gpg: aka "Alexander Graf <alex@csgraf.de>" * remotes/agraf/tags/signed-ppc-for-upstream: (40 commits) softmmu: support up to 12 MMU modes tcg: add TCG_TARGET_TLB_DISPLACEMENT_BITS tci: do not use CPUArchState in tcg-target.h Add David Gibson for sPAPR in MAINTAINERS file pseries: Enable in-kernel H_LOGICAL_CI_{LOAD, STORE} implementations spapr: override default ram size to 512MB machine: add default_ram_size to machine class spapr_pci: emit hotplug add/remove events during hotplug spapr_pci: enable basic hotplug operations pci: make pci_bar useable outside pci.c spapr_pci: create DRConnectors for each PCI slot during PHB realize spapr_pci: add dynamic-reconfiguration option for spapr-pci-host-bridge spapr_drc: add spapr_drc_populate_dt() spapr_events: event-scan RTAS interface spapr_events: re-use EPOW event infrastructure for hotplug events spapr_rtas: add ibm, configure-connector RTAS interface spapr: add rtas_st_buffer_direct() helper spapr_rtas: add get-sensor-state RTAS interface spapr_rtas: add set-indicator RTAS interface spapr_rtas: add get/set-power-level RTAS interfaces ... Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2015-06-03pci: make pci_bar useable outside pci.cMichael Roth1-1/+1
We need to work with PCI BARs to generate OF properties during PCI hotplug for sPAPR guests. Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com> Reviewed-by: David Gibson <david@gibson.dropbear.id.au> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: David Gibson <david@gibson.dropbear.id.au> Signed-off-by: Alexander Graf <agraf@suse.de>
2015-06-03hw/pci: add support for NUMA nodesMarcel Apfelbaum1-0/+11
PCI root buses can be attached to a specific NUMA node. PCI buses are not attached by default to a NUMA node. Signed-off-by: Marcel Apfelbaum <marcel@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Acked-by: Laszlo Ersek <lersek@redhat.com>
2015-06-03hw/pci: removed 'rootbus nr is 0' assumption from qmp_pci_queryMarcel Apfelbaum1-1/+2
Use the newer pci_bus_num to correctly get the root bus number. Signed-off-by: Marcel Apfelbaum <marcel@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Acked-by: Laszlo Ersek <lersek@redhat.com>
2015-06-03hw/pci: extend PCI config access to support devices behind PXBMarcel Apfelbaum1-5/+29
PXB buses are assumed to be children of bus 0. Look for them while scanning the buses. Signed-off-by: Marcel Apfelbaum <marcel@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Acked-by: Laszlo Ersek <lersek@redhat.com>
2015-06-03hw/pci: made pci_bus_num a PCIBusClass methodMarcel Apfelbaum1-3/+10
Refactoring it as a method of PCIBusClass will allow different implementations for subclasses. Signed-off-by: Marcel Apfelbaum <marcel@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Acked-by: Laszlo Ersek <lersek@redhat.com>
2015-06-03hw/pci: made pci_bus_is_root a PCIBusClass methodMarcel Apfelbaum1-3/+14
Refactoring it as a method of PCIBusClass will allow different implementations for subclasses. Removed the assumption that the root bus does not have a parent device because is specific only to the default class implementation. Signed-off-by: Marcel Apfelbaum <marcel@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Acked-by: Laszlo Ersek <lersek@redhat.com>
2015-06-02Merge remote-tracking branch 'remotes/armbru/tags/pull-monitor-2015-06-02' ↵Peter Maydell2-30/+23
into staging Monitor patches # gpg: Signature made Tue Jun 2 09:16:07 2015 BST using RSA key ID EB918653 # gpg: Good signature from "Markus Armbruster <armbru@redhat.com>" # gpg: aka "Markus Armbruster <armbru@pond.sub.org>" * remotes/armbru/tags/pull-monitor-2015-06-02: (21 commits) monitor: Change return type of monitor_cur_is_qmp() to bool monitor: Rename monitor_ctrl_mode() to monitor_is_qmp() monitor: Turn int command_mode into bool in_command_mode monitor: Drop do_qmp_capabilities()'s superfluous QMP check monitor: Unbox Monitor member mc and rename to qmp monitor: Rename monitor_control_read(), monitor_control_event() monitor: Rename handle_user_command() to handle_hmp_command() monitor: Limit QError use to command handlers monitor: Inline monitor_has_error() into its only caller monitor: Wean monitor_protocol_emitter() off mon->error monitor: Propagate errors through invalid_qmp_mode() monitor: Propagate errors through qmp_check_input_obj() monitor: Propagate errors through qmp_check_client_args() monitor: Drop unused "new" HMP command interface monitor: Use trad. command interface for HMP pcie_aer_inject_error monitor: Use traditional command interface for HMP device_add monitor: Use traditional command interface for HMP drive_del monitor: Convert client_migrate_info to QAPI monitor: Improve and document client_migrate_info protocol error monitor: Clean up after previous commit ... Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2015-06-02xen: don't allow guest to control MSI mask registerJan Beulich1-4/+0
It's being used by the hypervisor. For now simply mimic a device not capable of masking, and fully emulate any accesses a guest may issue nevertheless as simple reads/writes without side effects. This is XSA-129. Signed-off-by: Jan Beulich <jbeulich@suse.com> Reviewed-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
2015-06-02monitor: Use trad. command interface for HMP pcie_aer_inject_errorMarkus Armbruster2-30/+23
All QMP commands use the "new" handler interface (mhandler.cmd_new). Most HMP commands still use the traditional interface (mhandler.cmd), but a few use the "new" one. Complicates handle_user_command() for no gain, so I'm converting these to the traditional interface. pcie_aer_inject_error's implementation is split into the hmp_pcie_aer_inject_error() and pcie_aer_inject_error_print(). The former is a peculiar crossbreed between HMP and QMP handler. On success, it works like a QMP handler: store QDict through ret_data parameter, return 0. Printing the QDict is left to pcie_aer_inject_error_print(). On failure, it works more like an HMP handler: print error to monitor, return negative number. To convert to the traditional interface, turn pcie_aer_inject_error_print() into a command handler wrapping around hmp_pcie_aer_inject_error(). By convention, this command handler should be called hmp_pcie_aer_inject_error(), so rename the existing hmp_pcie_aer_inject_error() to do_pcie_aer_inject_error(). Signed-off-by: Markus Armbruster <armbru@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Luiz Capitulino <lcapitulino@redhat.com>
2015-05-31Add stream ID to MSI writePavel Fedin2-3/+10
GICv3 ITS distinguishes between devices by using hardwired device IDs passed on the bus. This patch implements passing these IDs in qemu. SMMU is also known to use stream IDs, therefore this addition can also be useful for implementing platforms with SMMU. Signed-off-by: Pavel Fedin <p.fedin@samsung.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Changes from v1: - Added bus number to the stream ID - Added stream ID not only to MSI-X, but also to plain MSI. Some common code was made into msi_send_message() function. Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
2015-05-11Merge remote-tracking branch 'remotes/mst/tags/for_upstream' into stagingPeter Maydell2-40/+33
pc, virtio enhancements Memory hot-unplug support for pc, MSI-X mapping update speedup for virtio-pci, misc refactorings and bugfixes. Signed-off-by: Michael S. Tsirkin <mst@redhat.com> # gpg: Signature made Mon May 11 08:23:43 2015 BST using RSA key ID D28D5469 # gpg: Good signature from "Michael S. Tsirkin <mst@kernel.org>" # gpg: aka "Michael S. Tsirkin <mst@redhat.com>" * remotes/mst/tags/for_upstream: (28 commits) acpi: update expected files for memory unplug virtio-scsi: Move DEFINE_VIRTIO_SCSI_FEATURES to virtio-scsi virtio-net: Move DEFINE_VIRTIO_NET_FEATURES to virtio-net pci: Merge pci_nic_init() into pci_nic_init_nofail() acpi: add a missing backslash to the \_SB scope. qmp-event: add event notification for memory hot unplug error acpi: add hardware implementation for memory hot unplug acpi: fix "Memory device control fields" register acpi: extend aml_field() to support UpdateRule acpi, mem-hotplug: add unplug cb for memory device acpi, mem-hotplug: add unplug request cb for memory device acpi, mem-hotplug: add acpi_memory_slot_status() to get MemStatus docs: update documentation for memory hot unplug virtio: coding style tweak pci: remove hard-coded bar size in msix_init_exclusive_bar() virtio-pci: speedup MSI-X masking and unmasking virtio: introduce vector to virtqueues mapping virtio-ccw: using VIRTIO_NO_VECTOR instead of 0 for invalid virtqueue monitor: check return value of qemu_find_net_clients_except() monitor: replace the magic number 255 with MAX_QUEUE_NUM ... Conflicts: hw/s390x/s390-virtio-bus.c [PMM: fixed conflict in s390_virtio_scsi_properties and s390_virtio_net_properties arrays; since the result of the two conflicting patches is to empty the property arrays completely, the conflict resolution is to remove them entirely.] Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2015-05-05qapi: Drop inline nested structs in query-pciEric Blake1-19/+23
A future patch will be using a 'name':{dictionary} entry in the QAPI schema to specify a default value for an optional argument (see previous commit message for more details why); but existing use of inline nested structs conflicts with that goal. This patch fixes one of only two commands relying on nested types, by breaking the nesting into an explicit type; it means that the type is now boxed instead of unboxed in C code, but the QMP wire format is unaffected by this change. Prefer the safer g_new0() while making the conversion, and reduce some long lines. Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: Markus Armbruster <armbru@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
2015-04-28pci: Merge pci_nic_init() into pci_nic_init_nofail()Thomas Huth1-29/+14
The error reporting in pci_nic_init() is quite erratic: Some errors are printed directly with error_report(), and some are passed back to the caller pci_nic_init_nofail() via an Error pointer. Since pci_nic_init() is only used by pci_nic_init_nofail(), the functions can be simply merged to clean up this inconsistency. Signed-off-by: Thomas Huth <thuth@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Reviewed-by: Markus Armbruster <armbru@redhat.com>
2015-04-28Convert (ffs(val) - 1) to ctz32(val)Stefan Hajnoczi4-13/+13
This commit was generated mechanically by coccinelle from the following semantic patch: @@ expression val; @@ - (ffs(val) - 1) + ctz32(val) The call sites have been audited to ensure the ffs(0) - 1 == -1 case never occurs (due to input validation, asserts, etc). Therefore we don't need to worry about the fact that ctz32(0) == 32. Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com> Message-id: 1427124571-28598-5-git-send-email-stefanha@redhat.com Signed-off-by: Kevin Wolf <kwolf@redhat.com>
2015-04-27pci: remove hard-coded bar size in msix_init_exclusive_bar()Jason Wang1-11/+19
This patch lets msix_init_exclusive_bar() can calculate the bar and pba size based on the number of MSI-X vectors other than using a hard-coded limit 4096. This is needed to allow device to have more than 128 MSI_X vectors. To keep migration compatibility, keep using 4096 as bar size and 2048 for pba offset. Notes: We don't care about the case that using vectors > 128 for legacy machine type. Since we limit the queue max to 64, so vectors >= 65 is meaningless. Virtio device will be the first user for this. Cc: Keith Busch <keith.busch@intel.com> Cc: Kevin Wolf <kwolf@redhat.com> Cc: Stefan Hajnoczi <stefanha@redhat.com> Cc: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Jason Wang <jasowang@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
2015-04-26Switch non-CPU callers from ld/st*_phys to address_space_ld/st*Peter Maydell2-2/+4
Switch all the uses of ld/st*_phys to address_space_ld/st*, except for those cases where the address space is the CPU's (ie cs->as). This was done with the following script which generates a Coccinelle patch. A few over-80-columns lines in the result were rewrapped by hand where Coccinelle failed to do the wrapping automatically, as well as one location where it didn't put a line-continuation '\' when wrapping lines on a change made to a match inside a macro definition. ===begin=== #!/bin/sh -e # Usage: # ./ldst-phys.spatch.sh > ldst-phys.spatch # spatch -sp_file ldst-phys.spatch -dir . | sed -e '/^+/s/\t/ /g' > out.patch # patch -p1 < out.patch for FN in ub uw_le uw_be l_le l_be q_le q_be uw l q; do cat <<EOF @ cpu_matches_ld_${FN} @ expression E1,E2; identifier as; @@ ld${FN}_phys(E1->as,E2) @ other_matches_ld_${FN} depends on !cpu_matches_ld_${FN} @ expression E1,E2; @@ -ld${FN}_phys(E1,E2) +address_space_ld${FN}(E1,E2, MEMTXATTRS_UNSPECIFIED, NULL) EOF done for FN in b w_le w_be l_le l_be q_le q_be w l q; do cat <<EOF @ cpu_matches_st_${FN} @ expression E1,E2,E3; identifier as; @@ st${FN}_phys(E1->as,E2,E3) @ other_matches_st_${FN} depends on !cpu_matches_st_${FN} @ expression E1,E2,E3; @@ -st${FN}_phys(E1,E2,E3) +address_space_st${FN}(E1,E2,E3, MEMTXATTRS_UNSPECIFIED, NULL) EOF done ===endit=== Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
2015-04-13pci: Fix crash with illegal "-net nic, model=xxx" optionThomas Huth1-1/+3
Current QEMU crashes when specifying an illegal model with the "-net nic,model=xxx" option, e.g.: $ qemu-system-x86_64 -net nic,model=n/a qemu-system-x86_64: Unsupported NIC model: n/a Program received signal SIGSEGV, Segmentation fault. The gdb backtrace looks like this: 0x0000555555965fe0 in error_get_pretty (err=0x0) at util/error.c:152 152 return err->msg; (gdb) bt 0 0x0000555555965fe0 in error_get_pretty (err=0x0) at util/error.c:152 1 0x0000555555965ffd in error_report_err (err=0x0) at util/error.c:157 2 0x0000555555809c90 in pci_nic_init_nofail (nd=0x555555e49860 <nd_table>, rootbus=0x5555564409b0, default_model=0x55555598c37b "e1000", default_devaddr=0x0) at hw/pci/pci.c:1663 3 0x0000555555691e42 in pc_nic_init (isa_bus=0x555556f71900, pci_bus=0x5555564409b0) at hw/i386/pc.c:1506 4 0x000055555569396b in pc_init1 (machine=0x5555562abbf0, pci_enabled=1, kvmclock_enabled=1) at hw/i386/pc_piix.c:248 5 0x0000555555693d27 in pc_init_pci (machine=0x5555562abbf0) at hw/i386/pc_piix.c:310 6 0x000055555572ddf5 in main (argc=3, argv=0x7fffffffe018, envp=0x7fffffffe038) at vl.c:4226 The problem is that pci_nic_init_nofail() does not check whether the err parameter from pci_nic_init has been set up and thus passes a NULL pointer to error_report_err(). Fix it by correctly checking the err parameter. Signed-off-by: Thomas Huth <thuth@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Reviewed-by: Jason Wang <jasowang@redhat.com> Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2015-03-18pcie_aer: fix comment to match pcie specMichael S. Tsirkin1-2/+2
Code comment says "table 6-2" but in fact it's is not a table, it is "Figure 6-2" on page 479. Cc: Chen Fan <chen.fan.fnst@cn.fujitsu.com> Reported-by: Michael Tokarev <mjt@tls.msk.ru> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
2015-03-18aer: fix a wrong init PCI_ERR_COR_STATUS w1cmask type registerChen Fan1-1/+1
Error Status Register, so this patch fix a wrong definition for PCI_ERR_COR_STATUS register with w1cmask type. Signed-off-by: Chen Fan <chen.fan.fnst@cn.fujitsu.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
2015-03-18pcie_aer: fix typos in pcie_aer_inject_error commentChen Fan1-3/+3
Refer to "PCI Express Base Spec3.0", this comments can't fit the description in spec, so we should fix them. Signed-off-by: Chen Fan <chen.fan.fnst@cn.fujitsu.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
2015-03-18aer: fix wrong check on expose aer tlp prefix logChen Fan1-1/+1
when specify TLP Prefix log as using pcie_aer_inject_error, the TLP prefix log is always discarded. because the check is incorrect, the End-End TLP Prefix Supported bit (PCI_EXP_DEVCAP2_EETLPP) should be in Device Capabilities 2 Register. Signed-off-by: Chen Fan <chen.fan.fnst@cn.fujitsu.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
2015-03-18pcie: correct mistaken register bit for End-End TLP Prefix BlockingChen Fan1-1/+1
from pcie spec 7.8.17, the End-End TLP Prefix Blocking bit local is 15(e.g. 0x8000) in device control 2 register. Signed-off-by: Chen Fan <chen.fan.fnst@cn.fujitsu.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
2015-03-11pci: Convert pci_nic_init() to Error to avoid qdev_init()Markus Armbruster1-4/+14
qdev_init() is deprecated, and will be removed when its callers have been weaned off it. Signed-off-by: Markus Armbruster <armbru@redhat.com> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
2015-03-11pci/shpc: fix signed integer overflowMichael S. Tsirkin1-1/+1
clang undefined behaviour sanitizer reports: > hw/pci/shpc.c:162:27: runtime error: left shift of 1 by 31 places > cannot be represented in type 'int' Caused by the usual lack of a 'U' qualifier on a constant 1 being shifted left. Fix it up. Reported-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
2015-03-01pci-hotplug-old: Has been dead for five major releases, buryMarkus Armbruster2-344/+0
Commit 79ca616 (v1.6.0) accidentally disabled legacy x86-only HMP commands pci_add, pci_del: it defined CONFIG_PCI_HOTPLUG only as make variable, not as preprocessor macro, killing the code conditional on defined(CONFIG_PCI_HOTPLUG_OLD). In all this time, nobody reported the loss. I only noticed it when I tried to test some error reporting change that forced me to touch this old crap again. Fun: git-log hw/pci/pci-hotplug-old.c shows our faith in the backward compatibility god has been strong enough to sacrifice at its altar about a dozen times, but not strong enough to even once verify the legacy feature's still there, let alone works. Remove the commands along with the code backing them. Signed-off-by: Markus Armbruster <armbru@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
2015-03-01pci: Give a few helpers internal linkageMarkus Armbruster1-7/+7
None of them should be used in new code. Signed-off-by: Markus Armbruster <armbru@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
2015-02-26pci: Permit incremental conversion of device models to realizeMarkus Armbruster1-5/+19
Call the new PCIDeviceClass method realize(). Default it to pci_default_realize(), which calls old method init(). To convert a device model, make it implement realize() rather than init(). Signed-off-by: Markus Armbruster <armbru@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Reviewed-by: Gonglei <arei.gonglei@huawei.com>
2015-02-26pci: Convert core to realizeMarkus Armbruster1-44/+49
Implement DeviceClass methods realize() and unrealize() instead of init() and exit(). The core's initialization errors now get propagated properly, and QMP sends them instead of an unspecific "Device initialization failed" error. Unrealize can't fail, so no change there. PCIDeviceClass is unchanged: it still provides init() and exit(). Therefore, device models' errors are still not propagated. Signed-off-by: Markus Armbruster <armbru@redhat.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Reviewed-by: Gonglei <arei.gonglei@huawei.com>
2015-02-26Merge remote-tracking branch 'remotes/armbru/tags/pull-error-2015-02-18' ↵Peter Maydell2-4/+2
into staging Clean up around error_get_pretty(), qerror_report_err() # gpg: Signature made Wed Feb 18 10:10:07 2015 GMT using RSA key ID EB918653 # gpg: Good signature from "Markus Armbruster <armbru@redhat.com>" # gpg: aka "Markus Armbruster <armbru@pond.sub.org>" * remotes/armbru/tags/pull-error-2015-02-18: qemu-char: Avoid qerror_report_err() outside QMP command handlers qemu-img: Avoid qerror_report_err() outside QMP command handlers vl: Avoid qerror_report_err() outside QMP command handlers tpm: Avoid qerror_report_err() outside QMP command handlers numa: Avoid qerror_report_err() outside QMP command handlers net: Avoid qerror_report_err() outside QMP command handlers monitor: Avoid qerror_report_err() outside QMP command handlers monitor: Clean up around monitor_handle_fd_param() error: Use error_report_err() where appropriate error: New convenience function error_report_err() vhost-scsi: Improve error reporting for invalid vhostfd Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2015-02-25Merge remote-tracking branch 'remotes/armbru/tags/pull-monitor-2015-02-18' ↵Peter Maydell3-4/+4
into staging hmp: Normalize HMP command handler names # gpg: Signature made Wed Feb 18 10:59:44 2015 GMT using RSA key ID EB918653 # gpg: Good signature from "Markus Armbruster <armbru@redhat.com>" # gpg: aka "Markus Armbruster <armbru@pond.sub.org>" * remotes/armbru/tags/pull-monitor-2015-02-18: hmp: Name HMP info handler functions hmp_info_SUBCOMMAND() hmp: Name HMP command handler functions hmp_COMMAND() hmp: Clean up declarations for long-gone info handlers Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2015-02-18hmp: Name HMP command handler functions hmp_COMMAND()Markus Armbruster3-4/+4
Some are called do_COMMAND() (old ones, usually), some hmp_COMMAND(), and sometimes COMMAND pointlessly differs in spelling. Normalize to hmp_COMMAND(), where COMMAND is exactly the command name with '-' replaced by '_'. Exceptions: * do_device_add() and client_migrate_info() *not* renamed to hmp_device_add(), hmp_client_migrate_info(), because they're also QMP handlers. They still need to be converted to QAPI. * do_memory_dump(), do_physical_memory_dump(), do_ioport_read(), do_ioport_write() renamed do hmp_* instead of hmp_x(), hmp_xp(), hmp_i(), hmp_o(), because those names are too cryptic for my taste. * do_info_help() renamed to hmp_info_help() instead of hmp_info(), because it only covers help. Signed-off-by: Markus Armbruster <armbru@redhat.com>
2015-02-18error: Use error_report_err() where appropriateMarkus Armbruster2-4/+2
Coccinelle semantic patch: @@ expression E; @@ - error_report("%s", error_get_pretty(E)); - error_free(E); + error_report_err(E); @@ expression E, S; @@ - error_report("%s", error_get_pretty(E)); + error_report_err(E); ( exit(S); | abort(); ) Trivial manual touch-ups in block/sheepdog.c. Signed-off-by: Markus Armbruster <armbru@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com>
2015-02-16pci: split shpc_cleanup and shpc_freePaolo Bonzini1-1/+10
object_unparent should not be called until the parent device is going to be destroyed. Only remove the capability and do memory_region_del_subregion at unrealize time. Freeing the data structures is left in shpc_free, to be called from the instance_finalize callback. Acked-by: Michael S. Tsirkin <mst@redhat.com> Reviewed-by: Matthew Rosato <mjrosato@linux.vnet.ibm.com> Reviewed-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>