On Tue, Nov 27, 2018 at 05:50:50PM +0000, Jean-Philippe Brucker wrote: > On 23/11/2018 22:02, Michael S. Tsirkin wrote: > >> +/* > >> + * __viommu_sync_req - Complete all in-flight requests > >> + * > >> + * Wait for all added requests to complete. When this function returns, all > >> + * requests that were in-flight at the time of the call have completed. > >> + */ > >> +static int __viommu_sync_req(struct viommu_dev *viommu) > >> +{ > >> + int ret = 0; > >> + unsigned int len; > >> + size_t write_len; > >> + struct viommu_request *req; > >> + struct virtqueue *vq = viommu->vqs[VIOMMU_REQUEST_VQ]; > >> + > >> + assert_spin_locked(&viommu->request_lock); > >> + > >> + virtqueue_kick(vq); > >> + > >> + while (!list_empty(&viommu->requests)) { > >> + len = 0; > >> + req = virtqueue_get_buf(vq, &len); > >> + if (!req) > >> + continue; > >> + > >> + if (!len) > >> + viommu_set_req_status(req->buf, req->len, > >> + VIRTIO_IOMMU_S_IOERR); > >> + > >> + write_len = req->len - req->write_offset; > >> + if (req->writeback && len == write_len) > >> + memcpy(req->writeback, req->buf + req->write_offset, > >> + write_len); > >> + > >> + list_del(&req->list); > >> + kfree(req); > >> + } > > > > I didn't notice this in the past but it seems this will spin > > with interrupts disabled until host handles the request. > > Please do not do this - host execution can be another > > task that needs the same host CPU. This will then disable > > interrupts for a very very long time. > > In the guest yes, but that doesn't prevent the host from running another > task right? Doesn't prevent it but it will delay it significantly until scheduler decides to kick the VCPU task out. > My tests run fine when QEMU is bound to a single CPU, even > though vcpu and viommu run in different threads > > > What to do then? Queue in software and wake up task. > > Unfortunately I can't do anything here, because IOMMU drivers can't > sleep in the iommu_map() or iommu_unmap() path. > > The problem is the same > for all IOMMU drivers. That's because the DMA API allows drivers to call > some functions with interrupts disabled. For example > Documentation/DMA-API-HOWTO.txt allows dma_alloc_coherent() and > dma_unmap_single() to be called in interrupt context. In fact I don't really understand how it's supposed to work at all: you only sync when ring is full. So host may not have seen your map request if ring is not full. Why is it safe to use the address with a device then? > > As kick is vm exit, kick under interrupts disabled is discouraged too: > > better to prepare for kick enable interrupts then kick. > > That was on my list of things to look at, because it could relax > things for device drivers that don't call us with interrupts disabled. I > just tried it and I can see some performance improvement (7% and 4% on > tcp_stream and tcp_maerts respectively, +/-2.5%). > > Since it's an optimization I'll leave it for later (ACPI and module > support is higher on my list). The resulting change is complicated > because we now need to deal with threads adding new requests while > sync() is running. With my current prototype one thread could end up > staying in sync() while other threads add new async requests, so I need > to find a way to bound it. > > Thanks, > Jean