On Wed, Nov 18, 2015 at 15:13:20 +0100, Andrea Bolognani wrote:
The amount of memory a ppc64 domain might need to lock is different than that of a equally-sized x86 domain, so we need to check the domain's architecture and act accordingly.
Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=1273480 --- src/qemu/qemu_domain.c | 80 +++++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 79 insertions(+), 1 deletion(-)
diff --git a/src/qemu/qemu_domain.c b/src/qemu/qemu_domain.c index 2c0f5af..1e92b9d 100644 --- a/src/qemu/qemu_domain.c +++ b/src/qemu/qemu_domain.c
[...]
+ /* baseLimit := maxMemory / 128 (a) + * + 4 MiB * #PHBs + 8 MiB (b) + * + * (a) is the hash table + * + * (b) is accounting for the 32-bit DMA window - it could be either the + * KVM accelerated TCE tables for emulated devices, or the VFIO + * userspace view. The 4 MiB per-PHB (including the default one) covers + * a 2GiB DMA window: default is 1GiB, but it's possible it'll be + * increased to help performance. The 8 MiB extra should be plenty for + * the TCE table index for any reasonable number of PHBs and several + * spapr-vlan or spapr-vscsi devices (512kB + a tiny bit each) */ + baseLimit = maxMemory / 128 + + 4096 * nPCIHostBridges + + 8192; + + /* passthroughLimit := max( 2 GiB * #PHBs, (c) + * memory (d) + * + memory * 1/512 * #PHBs + 8 MiB ) (e) + * + * (c) is the pre-DDW VFIO DMA window accounting. We're allowing 2 GiB + * rather than 1 GiB + * + * (d) is the with-DDW (and memory pre-registration and related + * features) DMA window accounting - assuming that we only account RAM + * once, even if mapped to multiple PHBs + * + * (e) is the with-DDW userspace view and overhead for the 64-bit DMA + * window. This is based a bit on expected guest behaviour, but there + * really isn't a way to completely avoid that. We assume the guest + * requests a 64-bit DMA window (per PHB) just big enough to map all + * its RAM. 4 kiB page size gives the 1/512; it will be less with 64 + * kiB pages, less still if the guest is mapped with hugepages (unlike + * the default 32-bit DMA window, DDW windows can use large IOMMU + * pages). 8 MiB is for second and further level overheads, like (b) */ + passthroughLimit = MAX(2 * 1024 * 1024 * nPCIHostBridges, + memory + + memory / 512 * nPCIHostBridges + 8192); + + if (usesVFIO) + memKB = baseLimit + passthroughLimit; + else + memKB = baseLimit;
Is there any public documentation where you were taking the info from? Should we link it to the code? Peter