vfio-pci: Allow mmap of MSIX BAR
v5 cover vfio-pci: Allow mmap of MSIX BAR
This is to enable MSIX BAR mapping to the guest and by the same time avoid failing on p2p DMA mappings. This attempt makes a try to separate p2p handling from msix.
This is based on sha1
a6506a9 Suraj Jitindar Singh "target/ppc/spapr: Add H-Call H_GET_CPU_CHARACTERISTICS".
Please comment. Thanks.
Alexey Kardashevskiy (2):
vfio/common: Add 'p2p' property to enable DMA mapping of MMIO regions
RFC: vfio-pci: Allow mmap of MSIX BAR
include/hw/vfio/vfio-common.h | 2 ++ linux-headers/linux/vfio.h | 5 +++ hw/ppc/spapr.c | 7 ++++ hw/vfio/common.c | 74 +++++++++++++++++++++++++++++++++++++++---- hw/vfio/pci.c | 11 +++++++ 5 files changed, 92 insertions(+), 7 deletions(-)
Signature · 1 line
2.11.0
PATCH v5 1/2 vfio/common: Add 'p2p' property to enable DMA mapping of MMIO regions
At the moment we map all RAM memory regions for possible DMA including MMIO regions of passed through device as this might be used for P2P PCI. However if DMA map fails for whatever reason, we fail and exit QEMU. Since P2P is not widely used and tested, it makes sense to exclude MMIO regions from DMA mapping by default and add a flag to enable these when needed.
This adds a "p2p" option for "vfio-pci" device and vfio_listener_skipped_section() checks for it so region_add/del() skip these by default. The MMIO region needs initialized mr::owner which is set anyway.
This avoids DMA map when start and/or size of the area is not aligned to the minimal IOMMU page size as it is known to fail; the diagnostic message is printed in this case.
This avoids exiting QEMU if QEMU tried DMA map and failed for some other reason that misalignment; this should allow to experiment when needed.
This adds necessary checks to the vfio_listener_region_del() hook.
--- include/hw/vfio/vfio-common.h | 1 + hw/vfio/common.c | 59 ++++++++++++++++++++++++++++++++++++++----- hw/vfio/pci.c | 1 + 3 files changed, 54 insertions(+), 7 deletions(-)
include/hw/vfio/vfio-common.h +1 -0
@@ -121,6 +121,7 @@ typedef struct VFIODevice {bool reset_works;bool needs_reset;bool no_mmap; +bool p2p;VFIODeviceOps *ops;unsigned int num_irqs;unsigned int num_regions;
hw/vfio/common.c +52 -7
90 lines not on this page · raw message
hw/vfio/pci.c +1 -0
@@ -2999,6 +2999,7 @@ static Property vfio_pci_dev_properties[] = {DEFINE_PROP_UNSIGNED_NODEFAULT ("x-nv-gpudirect-clique" , VFIOPCIDevice, nv_gpudirect_clique, qdev_prop_nv_gpudirect_clique, uint8_t), + DEFINE_PROP_BOOL("p2p", VFIOPCIDevice, vbasedev.p2p, false),/* * TODO - support passed fds... is this necessary? * DEFINE_PROP_STRING("vfiofd", VFIOPCIDevice, vfiofd_name),
Signature · 1 line
2.11.0
Re: [Qemu-devel] [PATCH qemu v5 1/2] vfio/common: Add 'p2p' property to enable DMA mapping of MMIO regions
Quoted 6 lines · On Tue, 30 Jan 2018 14:55:26 +1100 Alexey Kardashevskiy <[email protected]> wrote:
At the moment we map all RAM memory regions for possible DMA including MMIO regions of passed through device as this might be used for P2P PCI. However if DMA map fails for whatever reason, we fail and exit QEMU. Since P2P is not widely used and tested, it makes sense to exclude MMIO regions from DMA mapping by default and add a flag to enable these when needed.
No, it doesn't. You're disabling an intentionally added feature with no evidence that it doesn't work and isn't in use and placing the burden on the user to discover this change and re-enable it. This has really gone off into the weeds. NAK. Thanks,
Alex
Quoted 144 lines
This adds a "p2p" option for "vfio-pci" device and
vfio_listener_skipped_section()checks for it so region_add/del() skip these by default. The MMIO region needs initializedmr::ownerwhich is set anyway.This avoids DMA map when start and/or size of the area is not aligned to the minimal IOMMU page size as it is known to fail; the diagnostic message is printed in this case.
This avoids exiting QEMU if QEMU tried DMA map and failed for some other reason that misalignment; this should allow to experiment when needed.
This adds necessary checks to the
vfio_listener_region_del()hook.Signed-off-by Alexey Kardashevskiy <[email protected]>--- include/hw/vfio/vfio-common.h | 1 + hw/vfio/common.c | 59 ++++++++++++++++++++++++++++++++++++++----- hw/vfio/pci.c | 1 + 3 files changed, 54 insertions(+), 7 deletions(-)include/hw/vfio/vfio-common.h +1
-0@@ -121,6 +121,7 @@ typedef struct VFIODevice {bool reset_works;bool needs_reset;bool no_mmap; +bool p2p;VFIODeviceOps *ops;unsigned int num_irqs;unsigned int num_regions;hw/vfio/common.c +52
-7@@ -305,7 +305,13 @@ static bool vfio_listener_skipped_section(MemoryRegionSection *section)* are never accessed by the CPU and beyond the address width of * some IOMMU hardware. TODO: VFIO should tell us the IOMMU width. */ - section->offset_within_address_space & (1ULL <<63 ); + section->offset_within_address_space & (1ULL <<63 ) || +/* +* Allow mapping of MMIO only if the device has p2p=true. +*/ + (memory_region_is_ram_device (section->mr) && + (!section->mr->owner || + !object_property_get_bool (section->mr->owner,"p2p" ,NULL ))); }/* Called with rcu_read_lock held. */ @@ -508,6 +514,18 @@ static void vfio_listener_region_add(MemoryListener *listener, }/* Here we assume that memory_region_is_ram(section->mr)==true */ +if (memory_region_is_ram_device (section->mr)) { +hwaddr pgmask = (1ULL <<ctz64 (hostwin->iova_pgsizes)) -1 ; + +if ((section->offset_within_region & pgmask) || + (int128_getlo (section->size) & pgmask)) { +error_report ("Region %" HWADDR_PRIx"..%" HWADDR_PRIx" is not aligned to %" HWADDR_PRIx" and cannot be mapped for DMA" , + section->offset_within_region, +int128_getlo (section->size), + pgmask +1 ); +return ; + } + } vaddr =memory_region_get_ram_ptr (section->mr) + section->offset_within_region +@@ -523,6 +541,9 @@ static void vfio_listener_region_add(MemoryListener *listener,error_report ("vfio_dma_map(%p, 0x%" HWADDR_PRIx", " "0x%" HWADDR_PRIx", %p) = %d (%m)" , container, iova,int128_get64 (llsize), vaddr, ret); +if (memory_region_is_ram_device (section->mr)) { +return ; + } goto fail; }@@ -550,6 +571,7 @@ static void vfio_listener_region_del(MemoryListener *listener,hwaddr iova, end;Int128 llend, llsize;int ret; +bool try_unmap = true;if (vfio_listener_skipped_section (section)) { trace_vfio_listener_region_del_skip(@@ -602,13 +624,36 @@ static void vfio_listener_region_del(MemoryListener *listener,trace_vfio_listener_region_del (iova, end); - ret =vfio_dma_unmap (container, iova,int128_get64 (llsize)); +if (memory_region_is_ram_device (section->mr)) { +hwaddr pgmask; +VFIOHostDMAWindow *hostwin; +bool hostwin_found; + + hostwin_found = false; + QLIST_FOREACH(hostwin, &container->hostwin_list, hostwin_next) { +if (hostwin->min_iova <= iova && end <= hostwin->max_iova) { + hostwin_found = true; +break ; + } + } +assert (hostwin_found); + + pgmask = (1ULL <<ctz64 (hostwin->iova_pgsizes)) -1 ; + + try_unmap = !(section->offset_within_region & pgmask) && + !(int128_getlo (section->size) & pgmask); + } + +if (try_unmap) { + ret =vfio_dma_unmap (container, iova,int128_get64 (llsize)); +if (ret) { +error_report ("vfio_dma_unmap(%p, 0x%" HWADDR_PRIx", " +"0x%" HWADDR_PRIx") = %d (%m)" , + container, iova,int128_get64 (llsize), ret); + } + } +memory_region_unref (section->mr); -if (ret) { -error_report ("vfio_dma_unmap(%p, 0x%" HWADDR_PRIx", " -"0x%" HWADDR_PRIx") = %d (%m)" , - container, iova,int128_get64 (llsize), ret); - }if (container->iommu_type == VFIO_SPAPR_TCE_v2_IOMMU) { vfio_spapr_remove_window(container,hw/vfio/pci.c +1
-0@@ -2999,6 +2999,7 @@ static Property vfio_pci_dev_properties[] = {DEFINE_PROP_UNSIGNED_NODEFAULT ("x-nv-gpudirect-clique" , VFIOPCIDevice, nv_gpudirect_clique, qdev_prop_nv_gpudirect_clique, uint8_t), + DEFINE_PROP_BOOL("p2p", VFIOPCIDevice, vbasedev.p2p, false),/* * TODO - support passed fds... is this necessary? * DEFINE_PROP_STRING("vfiofd", VFIOPCIDevice, vfiofd_name),
PATCH v5 2/2 RFC: vfio-pci: Allow mmap of MSIX BAR
In reply to Alexey KardashevskiyThis makes use of a new VFIO_REGION_INFO_CAP_MSIX_MAPPABLE capability which tells that a region with MSIX data can be mapped entirely, i.e. the VFIO PCI driver won't prevent MSIX vectors area from being mapped.
With this change, all BARs are mapped in a single chunk and MSIX vectors are emulated on top unless the machine requests not to by defining and enabling a new "vfio-no-msix-emulation" property. At the moment only sPAPR machine does so - it prohibits MSIX emulation and does not allow enabling it as it does not define the "set" callback for the new property; the new property also does not appear in "-machine pseries,help".
This requires the kernel change - "vfio-pci: Allow mapping MSIX BAR" - for the new capability: https://www.spinics.net/lists/kvm/msg160282.html
---
Changes:
v5:
* rebased on top of 'p2p' proposed patch
v4:
* silenced dma map errors if unaligned mapping is attempted - they are going to fail anyway
v3:
* vfio_listener_region_add() won't make qemu exit if failed on MMIO MR
--- include/hw/vfio/vfio-common.h | 1 + linux-headers/linux/vfio.h | 5 +++++ hw/ppc/spapr.c | 7 +++++++ hw/vfio/common.c | 15 +++++++++++++++ hw/vfio/pci.c | 10 ++++++++++ 5 files changed, 38 insertions(+)
include/hw/vfio/vfio-common.h +1 -0
@@ -172,6 +172,7 @@ int vfio_get_region_info(VFIODevice *vbasedev, int index, struct vfio_region_info **info);int vfio_get_dev_region_info (VFIODevice *vbasedev,uint32_t type,uint32_t subtype,struct vfio_region_info **info); +bool vfio_is_cap_present (VFIODevice *vbasedev,uint16_t cap_type,int region);#endif extern const MemoryListener vfio_prereg_listener;
linux-headers/linux/vfio.h +5 -0
@@ -301,6 +301,11 @@ struct vfio_region_info_cap_type {#define VFIO_REGION_SUBTYPE_INTEL_IGD_HOST_CFG (2)#define VFIO_REGION_SUBTYPE_INTEL_IGD_LPC_CFG (3) +/* +* The MSIX mappable capability informs that MSIX data of a BAR can be mmapped. +*/ +#define VFIO_REGION_INFO_CAP_MSIX_MAPPABLE 3 +/** * VFIO_DEVICE_GET_IRQ_INFO - _IOWR(VFIO_TYPE, VFIO_BASE + 9, * struct vfio_irq_info)
hw/ppc/spapr.c +7 -0
@@ -2830,6 +2830,11 @@ static void spapr_set_modern_hotplug_events(Object *obj, bool value, spapr->use_hotplug_event_source = value; } +static bool spapr_get_msix_emulation (Object *obj,Error **errp) +{ +return true; +} +static char *spapr_get_resize_hpt (Object *obj,Error **errp) {sPAPRMachineState *spapr =SPAPR_MACHINE (obj);
@@ -2911,6 +2916,8 @@ static void spapr_instance_init(Object *obj)object_property_set_description (obj,"vsmt" ,"Virtual SMT: KVM behaves as if this were" " the host's SMT mode" , &error_abort); +object_property_add_bool (obj,"vfio-no-msix-emulation" , + spapr_get_msix_emulation,NULL ,NULL ); }static void spapr_machine_finalizefn (Object *obj)
hw/vfio/common.c +15 -0
@@ -1431,6 +1431,21 @@ int vfio_get_dev_region_info(VFIODevice *vbasedev, uint32_t type,return -ENODEV ; } +bool vfio_is_cap_present (VFIODevice *vbasedev,uint16_t cap_type,int region) +{ +struct vfio_region_info *info =NULL ; +bool ret = false; + +if (!vfio_get_region_info (vbasedev, region, &info)) { +if (vfio_get_region_info_cap (info, cap_type)) { + ret = true; + } +g_free (info); + } + +return ret; +} +/* * Interfaces for IBM EEH (Enhanced Error Handling) */
hw/vfio/pci.c +10 -0
@@ -1289,6 +1289,11 @@ static void vfio_pci_fixup_msix_region(VFIOPCIDevice *vdev)off_t start, end;VFIORegion *region = &vdev->bars[vdev->msix->table_bar].region; +if (vfio_is_cap_present (&vdev->vbasedev,VFIO_REGION_INFO_CAP_MSIX_MAPPABLE , + vdev->msix->table_bar)) { +return ; + } +/* * We expect to find a single mmap covering the whole BAR, anything else * means it's either unsupported or already setup.
@@ -1473,6 +1478,11 @@ static int vfio_msix_setup(VFIOPCIDevice *vdev, int pos, Error **errp)*/ memory_region_set_enabled (&vdev->pdev.msix_pba_mmio, false); +if (object_property_get_bool (OBJECT (qdev_get_machine ()), +"vfio-no-msix-emulation" ,NULL )) { +memory_region_set_enabled (&vdev->pdev.msix_table_mmio, false); + } +return 0 ; }
Signature · 1 line
2.11.0
This series seems to have some coding style problems. See output below for more information:
Hi,
This series seems to have some coding style problems. See output below for more information:
Type: series
Message-id: [email protected]
Subject: [Qemu-devel] [PATCH qemu v5 0/2] vfio-pci: Allow mmap of MSIX BAR
=== TEST SCRIPT BEGIN ===
#!/bin/bash
BASE=base
n=1
total=$(git log --oneline $BASE.. | wc -l)
failed=0
git config --local diff.renamelimit 0
git config --local diff.renames True
commits="$(git log --format=%H --reverse $BASE..)" for c in $commits; do
echo "Checking PATCH $n/$total: $(git log -n 1 --format=%s $c)..."
if ! git show $c --format=email | ./scripts/checkpatch.pl --mailback -; then
failed=1
echo
fi
n=$((n+1))done
exit $failed
=== TEST SCRIPT END ===
Updating 3c8cf5a9c21ff8782164d1def7f44bd888713384
Switched to a new branch 'test'261b4a1327 RFC: vfio-pci: Allow mmap of MSIX BARefd447736c vfio/common: Add 'p2p' property to enable DMA mapping of MMIO regions
=== OUTPUT BEGIN ===
Checking PATCH 1/2: vfio/common: Add 'p2p' property to enable DMA mapping of MMIO regions...
ERROR: line over 90 characters
#59: FILE: hw/vfio/common.c:523:
+ error_report("Region %"HWADDR_PRIx"..%"HWADDR_PRIx" is not aligned to %"HWADDR_PRIx" and cannot be mapped for DMA",
total: 1 errors, 0 warnings, 105 lines checked
Your patch has style problems, please review. If any of these errors are false positives report them to the maintainer, see CHECKPATCH in MAINTAINERS.
Checking PATCH 2/2: RFC: vfio-pci: Allow mmap of MSIX BAR... === OUTPUT END ===
Test command exited with code: 1
---
Email generated automatically by Patchew [http://patchew.org/].
Please send your feedback to [email protected]