November 2024 - Linux-kselftest-mirror

[PROBLEM] selftests/bpf/progs/test_tunnel_kern.c: 678: 41-47: ERROR: application of sizeof to pointer

by Mirsad Todorovac

Hi, all! In the linux-next tree, next-20241108, coccinelle found an error. In the line 617, ret = bpf_skb_set_tunnel_opt(skb, gopt, sizeof(local_gopt)); In the line 678, ret = bpf_skb_set_tunnel_opt(skb, gopt, sizeof(gopt)); when 592 struct local_geneve_opt local_gopt; 593 struct geneve_opt *gopt = (struct geneve_opt *) &local_gopt; and 652 struct local_geneve_opt local_gopt; 653 struct geneve_opt *gopt = (struct geneve_opt *) &local_gopt; So, in all other call to bpf_skb_set_tunnel_opt(), the third parameter is the size of the struct, not the size of the pointer: ./tools/testing/selftests/bpf/progs/test_tunnel_kern.c:193: ret = bpf_skb_set_tunnel_opt(skb, &md, sizeof(md)); ./tools/testing/selftests/bpf/progs/test_tunnel_kern.c:273: ret = bpf_skb_set_tunnel_opt(skb, &md, sizeof(md)); ./tools/testing/selftests/bpf/progs/test_tunnel_kern.c:349: ret = bpf_skb_set_tunnel_opt(skb, &md, sizeof(md)); ./tools/testing/selftests/bpf/progs/test_tunnel_kern.c:388: ret = bpf_skb_set_tunnel_opt(skb, &md, sizeof(md)); ./tools/testing/selftests/bpf/progs/test_tunnel_kern.c:617: ret = bpf_skb_set_tunnel_opt(skb, gopt, sizeof(local_gopt)); ./tools/testing/selftests/bpf/progs/test_tunnel_kern.c:678: ret = bpf_skb_set_tunnel_opt(skb, gopt, sizeof(gopt)); =========================================== 587 SEC("tc") 588 int geneve_set_tunnel(struct __sk_buff *skb) 589 { 590 int ret; 591 struct bpf_tunnel_key key; 592 struct local_geneve_opt local_gopt; 593 struct geneve_opt *gopt = (struct geneve_opt *) &local_gopt; 594 595 __builtin_memset(&key, 0x0, sizeof(key)); 596 key.remote_ipv4 = 0xac100164; /* 172.16.1.100 */ 597 key.tunnel_id = 2; 598 key.tunnel_tos = 0; 599 key.tunnel_ttl = 64; 600 601 __builtin_memset(gopt, 0x0, sizeof(local_gopt)); 602 gopt->opt_class = bpf_htons(0x102); /* Open Virtual Networking (OVN) */ 603 gopt->type = 0x08; 604 gopt->r1 = 0; 605 gopt->r2 = 0; 606 gopt->r3 = 0; 607 gopt->length = 2; /* 4-byte multiple */ 608 *(int *) &gopt->opt_data = bpf_htonl(0xdeadbeef); 609 610 ret = bpf_skb_set_tunnel_key(skb, &key, sizeof(key), 611 BPF_F_ZERO_CSUM_TX); 612 if (ret < 0) { 613 log_err(ret); 614 return TC_ACT_SHOT; 615 } 616 617 → ret = bpf_skb_set_tunnel_opt(skb, gopt, sizeof(local_gopt)); 618 if (ret < 0) { 619 log_err(ret); 620 return TC_ACT_SHOT; 621 } 622 623 return TC_ACT_OK; 624 } 648 SEC("tc") 649 int ip6geneve_set_tunnel(struct __sk_buff *skb) 650 { 651 struct bpf_tunnel_key key; 652 struct local_geneve_opt local_gopt; 653 struct geneve_opt *gopt = (struct geneve_opt *) &local_gopt; 654 int ret; 655 656 __builtin_memset(&key, 0x0, sizeof(key)); 657 key.remote_ipv6[3] = bpf_htonl(0x11); /* ::11 */ 658 key.tunnel_id = 22; 659 key.tunnel_tos = 0; 660 key.tunnel_ttl = 64; 661 662 ret = bpf_skb_set_tunnel_key(skb, &key, sizeof(key), 663 BPF_F_TUNINFO_IPV6); 664 if (ret < 0) { 665 log_err(ret); 666 return TC_ACT_SHOT; 667 } 668 669 __builtin_memset(gopt, 0x0, sizeof(local_gopt)); 670 gopt->opt_class = bpf_htons(0x102); /* Open Virtual Networking (OVN) */ 671 gopt->type = 0x08; 672 gopt->r1 = 0; 673 gopt->r2 = 0; 674 gopt->r3 = 0; 675 gopt->length = 2; /* 4-byte multiple */ 676 *(int *) &gopt->opt_data = bpf_htonl(0xfeedbeef); 677 678 → ret = bpf_skb_set_tunnel_opt(skb, gopt, sizeof(gopt)); 679 if (ret < 0) { 680 log_err(ret); 681 return TC_ACT_SHOT; 682 } 683 684 return TC_ACT_OK; 685 } SOLUTION: Fixes: 5ddafcc377f98 ("selftests/bpf: Fix a few tests for GCC related warnings.") ------------------ diff --git a/tools/testing/selftests/bpf/progs/test_tunnel_kern.c b/tools/testing/selftests/bpf/progs/test_tunnel_kern.c index 32127f1cd687..b53d367451b2 100644 --- a/tools/testing/selftests/bpf/progs/test_tunnel_kern.c +++ b/tools/testing/selftests/bpf/progs/test_tunnel_kern.c @@ -675,7 +675,7 @@ int ip6geneve_set_tunnel(struct __sk_buff *skb) gopt->length = 2; /* 4-byte multiple */ *(int *) &gopt->opt_data = bpf_htonl(0xfeedbeef); - ret = bpf_skb_set_tunnel_opt(skb, gopt, sizeof(gopt)); + ret = bpf_skb_set_tunnel_opt(skb, gopt, sizeof(local_gopt)); if (ret < 0) { log_err(ret); return TC_ACT_SHOT; -- Best regards, MT

5 months, 2 weeks

1
0
0 0

[PATCH] kunit: skb: use "gfp" variable instead of hardcoding GFP_KERNEL

by Dan Carpenter

The intent here was clearly to use the gfp variable flags instead of hardcoding GFP_KERNEL. All the callers pass GFP_KERNEL as the gfp flags so this doesn't affect runtime. Fixes: b3231d353a51 ("kunit: add a convenience allocation wrapper for SKBs") Signed-off-by: Dan Carpenter <dan.carpenter(a)linaro.org> --- include/kunit/skbuff.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/include/kunit/skbuff.h b/include/kunit/skbuff.h index 44d12370939a..345e1e8f0312 100644 --- a/include/kunit/skbuff.h +++ b/include/kunit/skbuff.h @@ -29,7 +29,7 @@ static void kunit_action_kfree_skb(void *p) static inline struct sk_buff *kunit_zalloc_skb(struct kunit *test, int len, gfp_t gfp) { - struct sk_buff *res = alloc_skb(len, GFP_KERNEL); + struct sk_buff *res = alloc_skb(len, gfp); if (!res || skb_pad(res, len)) return NULL; -- 2.45.2

5 months, 2 weeks

3
2
0 0

[PATCH net-next v8 0/6] Suspend IRQs during application busy periods

by Joe Damato

Greetings: Welcome to v8, see changelog below. This revision is a quick follow-up to v7 and only drops the unneeded exports in patch 2. No other changes. This series introduces a new mechanism, IRQ suspension, which allows network applications using epoll to mask IRQs during periods of high traffic while also reducing tail latency (compared to existing mechanisms, see below) during periods of low traffic. In doing so, this balances CPU consumption with network processing efficiency. Martin Karsten (CC'd) and I have been collaborating on this series for several months and have appreciated the feedback from the community on our RFC [1]. We've updated the cover letter and kernel documentation in an attempt to more clearly explain how this mechanism works, how applications can use it, and how it compares to existing mechanisms in the kernel. I briefly mentioned this idea at netdev conf 2024 (for those who were there) and Martin described this idea in an earlier paper presented at Sigmetrics 2024 [2]. ~ The short explanation (TL;DR) We propose adding a new napi config parameter: irq_suspend_timeout to help balance CPU usage and network processing efficiency when using IRQ deferral and napi busy poll. If this parameter is set to a non-zero value *and* a user application has enabled preferred busy poll on a busy poll context (via the EPIOCSPARAMS ioctl introduced in commit 18e2bf0edf4d ("eventpoll: Add epoll ioctl for epoll_params")), then application calls to epoll_wait for that context will cause device IRQs and softirq processing to be suspended as long as epoll_wait successfully retrieves data from the NAPI. Each time data is retrieved, the irq_suspend_timeout is deferred. If/when network traffic subsides and epoll_wait returns no data, IRQ suspension is immediately reverted back to the existing napi_defer_hard_irqs and gro_flush_timeout mechanism which was introduced in commit 6f8b12d661d0 ("net: napi: add hard irqs deferral feature")). The irq_suspend_timeout serves as a safety mechanism. If userland takes a long time processing data, irq_suspend_timeout will fire and restart normal NAPI processing. For a more in depth explanation, please continue reading. ~ Comparison with existing mechanisms Interrupt mitigation can be accomplished in napi software, by setting napi_defer_hard_irqs and gro_flush_timeout, or via interrupt coalescing in the NIC. This can be quite efficient, but in both cases, a fixed timeout (or packet count) needs to be configured. However, a fixed timeout cannot effectively support both low- and high-load situations: At low load, an application typically processes a few requests and then waits to receive more input data. In this scenario, a large timeout will cause unnecessary latency. At high load, an application typically processes many requests before being ready to receive more input data. In this case, a small timeout will likely fire prematurely and trigger irq/softirq processing, which interferes with the application's execution. This causes overhead, most likely due to cache contention. While NICs attempt to provide adaptive interrupt coalescing schemes, these cannot properly take into account application-level processing. An alternative packet delivery mechanism is busy-polling, which results in perfect alignment of application processing and network polling. It delivers optimal performance (throughput and latency), but results in 100% cpu utilization and is thus inefficient for below-capacity workloads. We propose to add a new packet delivery mode that properly alternates between busy polling and interrupt-based delivery depending on busy and idle periods of the application. During a busy period, the system operates in busy-polling mode, which avoids interference. During an idle period, the system falls back to interrupt deferral, but with a small timeout to avoid excessive latencies. This delivery mode can also be viewed as an extension of basic interrupt deferral, but alternating between a small and a very large timeout. This delivery mode is efficient, because it avoids softirq execution interfering with application processing during busy periods. It can be used with blocking epoll_wait to conserve cpu cycles during idle periods. The effect of alternating between busy and idle periods is that performance (throughput and latency) is very close to full busy polling, while cpu utilization is lower and very close to interrupt mitigation. ~ Usage details IRQ suspension is introduced via a per-NAPI configuration parameter that controls the maximum time that IRQs can be suspended. Here's how it is intended to work: - The user application (or system administrator) uses the netdev-genl netlink interface to set the pre-existing napi_defer_hard_irqs and gro_flush_timeout NAPI config parameters to enable IRQ deferral. - The user application (or system administrator) sets the proposed irq_suspend_timeout parameter via the netdev-genl netlink interface to a larger value than gro_flush_timeout to enable IRQ suspension. - The user application issues the existing epoll ioctl to set the prefer_busy_poll flag on the epoll context. - The user application then calls epoll_wait to busy poll for network events, as it normally would. - If epoll_wait returns events to userland, IRQs are suspended for the duration of irq_suspend_timeout. - If epoll_wait finds no events and the thread is about to go to sleep, IRQ handling using napi_defer_hard_irqs and gro_flush_timeout is resumed. As long as epoll_wait is retrieving events, IRQs (and softirq processing) for the NAPI being polled remain disabled. When network traffic reduces, eventually a busy poll loop in the kernel will retrieve no data. When this occurs, regular IRQ deferral using gro_flush_timeout for the polled NAPI is re-enabled. Unless IRQ suspension is continued by subsequent calls to epoll_wait, it automatically times out after the irq_suspend_timeout timer expires. Regular deferral is also immediately re-enabled when the epoll context is destroyed. ~ Usage scenario The target scenario for IRQ suspension as packet delivery mode is a system that runs a dominant application with substantial network I/O. The target application can be configured to receive input data up to a certain batch size (via epoll_wait maxevents parameter) and this batch size determines the worst-case latency that application requests might experience. Because packet delivery is suspended during the target application's processing, the batch size also determines the worst-case latency of concurrent applications using the same RX queue(s). gro_flush_timeout should be set as small as possible, but large enough to make sure that a single request is likely not being interfered with. irq_suspend_timeout is largely a safety mechanism against misbehaving applications. It should be set large enough to cover the processing of an entire application batch, i.e., the factor between gro_flush_timeout and irq_suspend_timeout should roughly correspond to the maximum batch size that the target application would process in one go. ~ Important call out in the implementation - Enabling per epoll-context preferred busy poll will now effectively lead to a nonblocking iteration through napi_busy_loop, even when busy_poll_usecs is 0. See patch 4. ~ Benchmark configs & descriptions The changes were benchmarked with memcached [3] using the benchmarking tool mutilate [4]. To facilitate benchmarking, a small patch [5] was applied to memcached 1.6.29 to allow setting per-epoll context preferred busy poll and other settings via environment variables. Another small patch [6] was applied to libevent to enable full busy-polling. Multiple scenarios were benchmarked as described below and the scripts used for producing these results can be found on github [7] (note: all scenarios use NAPI-based traffic splitting via SO_INCOMING_ID by passing -N to memcached): - base: - no other options enabled - deferX: - set defer_hard_irqs to 100 - set gro_flush_timeout to X,000 - napibusy: - set defer_hard_irqs to 100 - set gro_flush_timeout to 200,000 - enable busy poll via the existing ioctl (busy_poll_usecs = 64, busy_poll_budget = 64, prefer_busy_poll = true) - fullbusy: - set defer_hard_irqs to 100 - set gro_flush_timeout to 5,000,000 - enable busy poll via the existing ioctl (busy_poll_usecs = 1000, busy_poll_budget = 64, prefer_busy_poll = true) - change memcached's nonblocking epoll_wait invocation (via libevent) to using a 1 ms timeout - suspend0: - set defer_hard_irqs to 0 - set gro_flush_timeout to 0 - set irq_suspend_timeout to 20,000,000 - enable busy poll via the existing ioctl (busy_poll_usecs = 0, busy_poll_budget = 64, prefer_busy_poll = true) - suspendX: - set defer_hard_irqs to 100 - set gro_flush_timeout to X,000 - set irq_suspend_timeout to 20,000,000 - enable busy poll via the existing ioctl (busy_poll_usecs = 0, busy_poll_budget = 64, prefer_busy_poll = true) ~ Benchmark results Tested on: Single socket AMD EPYC 7662 64-Core Processor Hyperthreading disabled 4 NUMA Zones (NPS=4) 16 CPUs per NUMA zone (64 cores total) 2 x Dual port 100gbps Mellanox Technologies ConnectX-5 Ex EN NIC The test machine is configured such that a single interface has 8 RX queues. The queues' IRQs and memcached are pinned to CPUs that are NUMA-local to the interface which is under test. The NIC's interrupt coalescing configuration is left at boot-time defaults. Results: Results are shown below. The mechanism added by this series is represented by the 'suspend' cases. Data presented shows a summary over nearly 10 runs of each test case [8] using the scripts on github [7]. For latency, the median is shown. For throughput and CPU utilization, the average is shown. The results also include cycles-per-query (cpq) and instruction-per-query (ipq) metrics, following the methodology proposed in [2], to augment the CPU utilization numbers, which could be skewed due to frequency scaling. We find that this does not appear to be the case as CPU utilization and low-level metrics show similar trends. These results were captured using the scripts on github [7] to illustrate how this approach compares with other pre-existing mechanisms. This data is not to be interpreted as scientific data captured in a fully isolated lab setting, but instead as best effort, illustrative information comparing and contrasting tradeoffs. The absolute QPS results shift between submissions, but the relative differences are equivalent. As patches are rebased, several factors likely influence overall performance. Compare: - Throughput (MAX) and latencies of base vs suspend. - CPU usage of napibusy and fullbusy during lower load (200K, 400K for example) vs suspend. - Latency of the defer variants vs suspend as timeout and load increases. - suspend0, which sets defer_hard_irqs and gro_flush_timeout to 0, has nearly the same performance as the base case (this is FAQ item #1). The overall takeaway is that the suspend variants provide a superior combination of high throughput, low latency, and low cpu utilization compared to all other variants. Each of the suspend variants works very well, but some fine-tuning between latency and cpu utilization is still possible by tuning the small timeout (gro_flush_timeout). Note: we've reorganized the results to make comparison among testcases with the same load easier. testcase load qps avglat 95%lat 99%lat cpu cpq ipq base 200K 199946 112 239 416 26 12973 11343 defer10 200K 199971 54 124 142 29 19412 17460 defer20 200K 199986 60 130 153 26 15644 14095 defer50 200K 200025 79 144 182 23 12122 11632 defer200 200K 199999 164 254 309 19 8923 9635 fullbusy 200K 199998 46 118 133 100 43658 23133 napibusy 200K 199983 100 237 277 56 24840 24716 suspend0 200K 200020 105 249 432 30 14264 11796 suspend10 200K 199950 53 123 141 32 19518 16903 suspend20 200K 200037 58 126 151 30 16426 14736 suspend50 200K 199961 73 136 177 26 13310 12633 suspend200 200K 199998 149 251 306 21 9566 10203 testcase load qps avglat 95%lat 99%lat cpu cpq ipq base 400K 400014 139 269 707 41 9476 9343 defer10 400K 400016 59 133 166 53 13991 12989 defer20 400K 399952 67 140 172 47 12063 11644 defer50 400K 400007 87 162 198 39 9384 9880 defer200 400K 399979 181 274 330 31 7089 8430 fullbusy 400K 399987 50 123 156 100 21827 16037 napibusy 400K 400014 76 222 272 83 18185 16529 suspend0 400K 400015 127 350 776 47 10699 9603 suspend10 400K 400023 57 129 164 54 13758 13178 suspend20 400K 400043 62 135 169 49 12071 11826 suspend50 400K 400071 76 149 186 42 10011 10301 suspend200 400K 399961 154 269 327 34 7827 8774 testcase load qps avglat 95%lat 99%lat cpu cpq ipq base 600K 599951 149 266 574 61 9265 8876 defer10 600K 600006 71 147 203 76 11866 10936 defer20 600K 600123 76 152 203 66 10430 10342 defer50 600K 600162 95 172 217 54 8526 9142 defer200 600K 599942 200 301 357 46 6977 8212 fullbusy 600K 599990 55 127 177 100 14551 13983 napibusy 600K 600035 63 160 250 96 13937 14140 suspend0 600K 599903 127 320 732 68 10166 8963 suspend10 600K 599908 63 137 192 69 10902 11100 suspend20 600K 599961 66 141 194 65 9976 10370 suspend50 600K 599973 80 159 204 57 8678 9381 suspend200 600K 600010 157 277 346 48 7133 8381 testcase load qps avglat 95%lat 99%lat cpu cpq ipq base 800K 800039 181 300 536 87 9585 8304 defer10 800K 800038 181 530 939 96 10564 8970 defer20 800K 800029 112 225 329 90 10056 8935 defer50 800K 799999 120 208 296 82 9234 8562 defer200 800K 800066 227 338 401 63 7117 8129 fullbusy 800K 800040 61 134 190 100 10913 12608 napibusy 800K 799944 64 141 214 99 10828 12588 suspend0 800K 799911 126 248 509 85 9346 8498 suspend10 800K 800006 69 143 200 83 9410 9845 suspend20 800K 800120 74 150 207 78 8786 9454 suspend50 800K 799989 87 168 224 71 7946 8833 suspend200 800K 799987 160 292 357 62 6923 8229 testcase load qps avglat 95%lat 99%lat cpu cpq ipq base 1000K 906879 4079 5751 6216 98 9496 7904 defer10 1000K 860849 3643 6274 6730 99 10040 8676 defer20 1000K 896063 3298 5840 6349 98 9620 8237 defer50 1000K 919782 2962 5513 5807 97 9284 7951 defer200 1000K 970941 3059 5348 5984 95 8593 7959 fullbusy 1000K 999950 70 150 207 100 8732 10777 napibusy 1000K 999996 78 154 223 100 8722 10656 suspend0 1000K 949706 2666 5770 6660 99 9071 8046 suspend10 1000K 1000024 80 160 220 92 8137 9035 suspend20 1000K 1000059 83 165 226 89 7850 8804 suspend50 1000K 999955 95 180 240 84 7411 8459 suspend200 1000K 999914 163 299 366 77 6833 8078 testcase load qps avglat 95%lat 99%lat cpu cpq ipq base MAX 1037654 4184 5453 5810 100 8411 7938 defer10 MAX 905607 4840 6151 6380 100 9639 8431 defer20 MAX 986463 4455 5594 5796 100 8848 8110 defer50 MAX 1077030 4000 5073 5299 100 8104 7920 defer200 MAX 1040728 4152 5385 5765 100 8379 7849 fullbusy MAX 1247536 3518 3935 3984 100 6998 7930 napibusy MAX 1136310 3799 7756 9964 100 7670 7877 suspend0 MAX 1057509 4132 5724 6185 100 8253 7918 suspend10 MAX 1215147 3580 3957 4041 100 7185 7944 suspend20 MAX 1216469 3576 3953 3988 100 7175 7950 suspend50 MAX 1215871 3577 3961 4075 100 7181 7949 suspend200 MAX 1216882 3556 3951 3988 100 7175 7955 ~ FAQ - Why is a new parameter needed? Does irq_suspend_timeout override gro_flush_timeout? Using the suspend mechanism causes the system to alternate between polling mode and irq-driven packet delivery. During busy periods, irq_suspend_timeout overrides gro_flush_timeout and keeps the system busy polling, but when epoll finds no events, the setting of gro_flush_timeout and napi_defer_hard_irqs determine the next step. There are essentially three possible loops for network processing and packet delivery: 1) hardirq -> softirq -> napi poll; basic interrupt delivery 2) timer -> softirq -> napi poll; deferred irq processing 3) epoll -> busy-poll -> napi poll; busy looping Loop 2 can take control from Loop 1, if gro_flush_timeout and napi_defer_hard_irqs are set. If gro_flush_timeout and napi_defer_hard_irqs are set, Loops 2 and 3 "wrestle" with each other for control. During busy periods, irq_suspend_timeout is used as timer in Loop 2, which essentially tilts this in favour of Loop 3. If gro_flush_timeout and napi_defer_hard_irqs are not set, Loop 3 cannot take control from Loop 1. Therefore, setting gro_flush_timeout and napi_defer_hard_irqs is the recommended usage, because otherwise setting irq_suspend_timeout might not have any discernible effect. This is shown in the results above: compare suspend0 with the base case. Note that the lack of napi_defer_hard_irqs and gro_flush_timeout produce similar results for both, which encourages the use of napi_defer_hard_irqs and gro_flush_timeout in addition to irq_suspend_timeout. - Can the new timeout value be threaded through the new epoll ioctl ? It is possible, but presents challenges for userspace. User applications must ensure that the file descriptors added to epoll contexts have the same NAPI ID to support busy polling. An epoll context is not permanently tied to any particular NAPI ID. So, a user application could decide to clear the file descriptors from the context and add a new set of file descriptors with a different NAPI ID to the context. Busy polling would work as expected, but the meaning of the suspend timeout becomes ambiguous because IRQs are not inherently associated with epoll contexts, but rather with the NAPI. The user program would need to reissue the ioctl to set the irq_suspend_timeout, but the napi_defer_hard_irqs and gro_flush_timeout settings would come from the NAPI's napi_config (which are set either by sysfs or by netlink). Such an interface seems awkard to use from a user perspective. Further, IRQs are related to NAPIs, which is why they are stored in the napi_config space. Putting the irq_suspend_timeout in the epoll context while other IRQ deferral mechanisms remain in the NAPI's napi_config space seems like an odd design choice. We've opted to keep all of the IRQ deferral parameters together and place the irq_suspend_timeout in napi_config. This has nice benefits for userspace: if a user app were to remove all file descriptors from an epoll context and add new file descriptors with a new NAPI ID, the correct suspend timeout for that NAPI ID would be used automatically without the user application needing to do anything (like re-issuing an ioctl, for example). All IRQ deferral related parameters are in one place and can all be set the same way: with netlink. - Can irq suspend be built by combining NIC coalescing and gro_flush_timeout ? No. The problem is that the long timeout must engage if and only if prefer-busy is active. When using NIC coalescing for the short timeout (without napi_defer_hard_irqs/gro_flush_timeout), an interrupt after an idle period will trigger softirq, which will run napi polling. At this point, prefer-busy is not active, so NIC interrupts would be re-enabled. Then it is not possible for the longer timeout to interject to switch control back to polling. In other words, only by using the software timer for the short timeout, it is possible to extend the timeout without having to reprogram the NIC timer or reach down directly and disable interrupts. Using gro_flush_timeout for the long timeout also has problems, for the same underlying reason. In the current napi implementation, gro_flush_timeout is not tied to prefer-busy. We'd either have to change that and in the process modify the existing deferral mechanism, or introduce a state variable to determine whether gro_flush_timeout is used as long timeout for irq suspend or whether it is used for its default purpose. In an earlier version, we did try something similar to the latter and made it work, but it ends up being a lot more convoluted than our current proposal. - Isn't it already possible to combine busy looping with irq deferral? Yes, in fact enabling irq deferral via napi_defer_hard_irqs and gro_flush_timeout is a precondition for prefer_busy_poll to have an effect. If the application also uses a tight busy loop with essentially nonblocking epoll_wait (accomplished with a very short timeout parameter), this is the fullbusy case shown in the results. An application using blocking epoll_wait is shown as the napibusy case in the results. It's a hybrid approach that provides limited latency benefits compared to the base case and plain irq deferral, but not as good as fullbusy or suspend. ~ Special thanks Several people were involved in earlier stages of the development of this mechanism whom we'd like to thank: - Peter Cai (CC'd), for the initial kernel patch and his contributions to the paper. - Mohammadamin Shafie (CC'd), for testing various versions of the kernel patch and providing helpful feedback. Thanks, Martin and Joe [1]: https://lore.kernel.org/netdev/20240812125717.413108-1-jdamato@fastly.com/ [2]: https://doi.org/10.1145/3626780 [3]: https://github.com/memcached/memcached/blob/master/doc/napi_ids.txt [4]: https://github.com/leverich/mutilate [5]: https://raw.githubusercontent.com/martinkarsten/irqsuspend/main/patches/mem… [6]: https://raw.githubusercontent.com/martinkarsten/irqsuspend/main/patches/lib… [7]: https://github.com/martinkarsten/irqsuspend [8]: https://github.com/martinkarsten/irqsuspend/tree/main/results v8: - Update patch 2 to drop the exports, as requested by Jakub. v7: https://lore.kernel.org/netdev/20241108023912.98416-1-jdamato@fastly.com/ - Jakub noted that patch 2 adds unnecessary complexity by checking the suspend timeout in the NAPI loop. This makes the code more complicated and difficult to reason about. He's right; we've dropped patch 2 which simplifies this series. - Updated the cover letter with a full re-run of all test cases. - Updated FAQ #2. v6: https://lore.kernel.org/netdev/20241104215542.215919-1-jdamato@fastly.com/ - Updated the cover letter with a full re-run of all test cases, including a new case suspend0, as requested by Sridhar previously. - Updated the kernel documentation in patch 7 as suggested by Bagas Sanjaya, which improved the htmldoc output. v5: https://lore.kernel.org/netdev/20241103052421.518856-1-jdamato@fastly.com/ - Adjusted patch 5 to only suspend IRQs when ep_send_events returns a positive return value. This issue was pointed out by Hillf Danton. - Updated the commit message of patch 6 which still mentioned netcat, despite the code being updated in v4 to replace it with socat and fixed misspelling of netdevsim. - Fixed a minor typo in patch 7 and removed an unnecessary paragraph. - Added Sridhar Samudrala's Reviewed-by to patch 1-5 and 7. v4: https://lore.kernel.org/netdev/20241102005214.32443-1-jdamato@fastly.com/ - Added a new FAQ item to cover letter. - Updated patch 6 to use socat instead of nc in busy_poll_test.sh and updated busy_poller.c to use netlink directly to configure napi params. - Updated the kernel documentation in patch 7 to include more details. - Dropped Stanislav's Acked-by and Bagas' Reviewed-by from patch 7 since the documentation was updated. v3: https://lore.kernel.org/netdev/20241101004846.32532-1-jdamato@fastly.com/ - Added Stanislav Fomichev's Acked-by to every patch except the newly added selftest. - Added Bagas Sanjaya's Reviewed-by to the documentation patch. - Fixed the commit message of patch 2 to remove a reference to the now non-existent sysfs setting. - Added a self test which tests both "regular" busy poll and busy poll with suspend enabled. This was added as patch 6 as requested by Paolo. netdevsim was chosen instead of veth due to netdevsim's pre-existing support for netdev-genl. See the commit message of patch 6 for more details. v2: https://lore.kernel.org/bpf/20241021015311.95468-1-jdamato@fastly.com/ - Cover letter updated, including a re-run of test data. - Patch 1 rewritten to use netdev-genl instead of sysfs. - Patch 3 updated with a comment added to napi_resume_irqs. - Patch 4 rebased to apply now that commit b9ca079dd6b0 ("eventpoll: Annotate data-race of busy_poll_usecs") has been picked up from VFS. - Patch 6 updated the kernel documentation. rfc -> v1: - Cover letter updated to include more details. - Patch 1 updated to remove the documentation added. This was moved to patch 6 with the rest of the docs (see below). - Patch 5 updated to fix an error uncovered by the kernel build robot. See patch 5's changelog for more details. - Patch 6 added which updates kernel documentation. Joe Damato (2): selftests: net: Add busy_poll_test docs: networking: Describe irq suspension Martin Karsten (4): net: Add napi_struct parameter irq_suspend_timeout net: Add control functions for irq suspension eventpoll: Trigger napi_busy_loop, if prefer_busy_poll is set eventpoll: Control irq suspension for prefer_busy_poll Documentation/netlink/specs/netdev.yaml | 7 + Documentation/networking/napi.rst | 170 ++++++++- fs/eventpoll.c | 36 +- include/linux/netdevice.h | 2 + include/net/busy_poll.h | 3 + include/uapi/linux/netdev.h | 1 + net/core/dev.c | 39 +++ net/core/dev.h | 25 ++ net/core/netdev-genl-gen.c | 5 +- net/core/netdev-genl.c | 12 + tools/include/uapi/linux/netdev.h | 1 + tools/testing/selftests/net/.gitignore | 1 + tools/testing/selftests/net/Makefile | 3 +- tools/testing/selftests/net/busy_poll_test.sh | 164 +++++++++ tools/testing/selftests/net/busy_poller.c | 328 ++++++++++++++++++ 15 files changed, 790 insertions(+), 7 deletions(-) create mode 100755 tools/testing/selftests/net/busy_poll_test.sh create mode 100644 tools/testing/selftests/net/busy_poller.c base-commit: dc7c381bb8649e3701ed64f6c3e55316675904d7 -- 2.25.1

5 months, 2 weeks

2
7
0 0

[PATCH] selftests/mm: Define PKEY_UNRESTRICTED for pkey_sighandler_tests

by Kevin Brodsky

Commit 6e182dc9f268 ("selftests/mm: Use generic pkey register manipulation") makes use of PKEY_UNRESTRICTED in pkey_sighandler_tests. The macro has been proposed for addition to uapi headers [1], but the patch hasn't landed yet. Define PKEY_UNRESTRICTED in pkey-helpers.h for the time being to fix the build. [1] https://lore.kernel.org/all/20241028090715.509527-2-yury.khrustalev@arm.com/ Fixes: 6e182dc9f268 ("selftests/mm: Use generic pkey register manipulation") Reported-by: Aishwarya TCV <aishwarya.tcv(a)arm.com> Signed-off-by: Kevin Brodsky <kevin.brodsky(a)arm.com> --- Based on arm64 for-next/pkey-signal (49f59573e9e0). --- tools/testing/selftests/mm/pkey-helpers.h | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/tools/testing/selftests/mm/pkey-helpers.h b/tools/testing/selftests/mm/pkey-helpers.h index 9ab6a3ee153b..319f5b6b7132 100644 --- a/tools/testing/selftests/mm/pkey-helpers.h +++ b/tools/testing/selftests/mm/pkey-helpers.h @@ -112,6 +112,10 @@ void record_pkey_malloc(void *ptr, long size, int prot); #define PKEY_MASK (PKEY_DISABLE_ACCESS | PKEY_DISABLE_WRITE) #endif +#ifndef PKEY_UNRESTRICTED +#define PKEY_UNRESTRICTED 0x0 +#endif + #ifndef set_pkey_bits static inline u64 set_pkey_bits(u64 reg, int pkey, u64 flags) { -- 2.43.0

5 months, 2 weeks

3
5
0 0

[PATCH] selftests: netfilter: Fix the bug of missing return values

by guanjing

Fixed the bug of some functions were missing return values. Fixes: eff3c558bb7e ("netfilter: ctnetlink: support filtering by zone") Signed-off-by: Guan Jing <guanjing(a)cmss.chinamobile.com> --- .../testing/selftests/net/netfilter/conntrack_dump_flush.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/tools/testing/selftests/net/netfilter/conntrack_dump_flush.c b/tools/testing/selftests/net/netfilter/conntrack_dump_flush.c index 254ff03297f0..5f827e10717d 100644 --- a/tools/testing/selftests/net/netfilter/conntrack_dump_flush.c +++ b/tools/testing/selftests/net/netfilter/conntrack_dump_flush.c @@ -43,6 +43,8 @@ static int build_cta_tuple_v4(struct nlmsghdr *nlh, int type, mnl_attr_nest_end(nlh, nest_proto); mnl_attr_nest_end(nlh, nest); + + return 0; } static int build_cta_tuple_v6(struct nlmsghdr *nlh, int type, @@ -71,6 +73,8 @@ static int build_cta_tuple_v6(struct nlmsghdr *nlh, int type, mnl_attr_nest_end(nlh, nest_proto); mnl_attr_nest_end(nlh, nest); + + return 0; } static int build_cta_proto(struct nlmsghdr *nlh) @@ -90,6 +94,8 @@ static int build_cta_proto(struct nlmsghdr *nlh) mnl_attr_nest_end(nlh, nest_proto); mnl_attr_nest_end(nlh, nest); + + return 0; } static int conntrack_data_insert(struct mnl_socket *sock, struct nlmsghdr *nlh, -- 2.33.0

5 months, 2 weeks

1
0
0 0

Some thoughts on Linux benchmarks results & processing

by Konstantin Belov

Hello colleagues, Following up on Tim Bird's presentation "Adding benchmarks results support to KTAP/kselftest", I would like to share some thoughts on kernel benchmarking and kernel performance evaluation. Tim suggested sharing these comments with the wider kselftest community for discussion. The topic of performance evaluation is obviously extremely complex, so I’ve organised my comments into several paragraphs, each of which focuses on a specific aspect. This should make it easier to follow and understand the key points, such as metrics, reference values, results data lake, interpretation of contradictory results, system profiles, analysis and methodology. # Metrics A few remarks on benchmark metrics which were called “values” in the original presentation: - Metrics must be accompanied by standardised units. This standardisation ensures consistency across different tests and environments, simplifying accurate comparisons and analysis. - Each metric should be clearly labelled with its nature or kind (throughput, speed, latency, etc). This classification is essential for proper interpretation of the results and prevents misunderstandings that could lead to incorrect conclusions. - Presentation contains "May also include allowable variance", but variance must be included into the analysis as we deal with statistical calculations and multiple randomised values. - I would like to note that other statistical parameters are also worth including into comparison, like confidence levels, sample size and so on. # Reference Values The concept of "reference values" introduced in the slides could be significantly enhanced by implementing a collaborative, transparent system for data collection and validation. This system could operate as follows: - Data Collection: Any user could submit benchmark results to a centralised and public repository. This would allow for a diverse range of hardware configurations and use cases to be represented. - Vendor Validation: Hardware vendors would have the opportunity to review submitted results pertaining to their products. They could then mark certain results as "Vendor Approved," indicating that the results align with their own testing and expectations. - Community Review: The broader community of users and experts could also review and vote on submitted results. Results that receive substantial positive feedback could be marked as "Community Approved," providing an additional layer of validation. - Automated Validation: Reference values must be checked, validated and supported by multiple sources. This can be done only in an automatic way as those processes are time consuming and require extreme attention to details. - Transparency: All submitted results would need to be accompanied by detailed information about the testing environment, hardware specifications, and methodology used. This would ensure reproducibility and allow others to understand the context of each result. - Trust Building: The combination of vendor and community approval would help establish trust in the reference values. It would mitigate concerns about marketing bias and provide a more reliable basis for performance comparisons. - Accessibility: The system would be publicly accessible, allowing anyone to reference and utilise this data in their own testing and analysis. Implementation of such a system would require careful consideration of governance and funding. A community-driven, non-profit organisation sponsored by multiple stakeholders could be an appropriate model. This structure would help maintain neutrality and avoid potential conflicts of interest. While the specifics of building and managing such a system would need further exploration, this approach could significantly improve the reliability and usefulness of reference values in benchmark testing. It would foster a more collaborative and transparent environment for performance evaluation in the Linux ecosystem as well as attract interested vendors to submit and review results. I’m not very informed about the current state of the community in this field, but I’m sure you know better how exactly this can be done. # Results Data Lake Along with reference values it’s important to collect results on a regular basis as the kernel evolves so results must follow this evolution as well. To do this cloud-based data lake is needed (a self-hosted system will be too expensive from my point of view). This data lake should be able to collect and process incoming data as well as to serve reference values for users. Data processing flow should be quite standard: Collection -> Parsing + Enhancement -> Storage -> Analysis -> Serving. Tim proposed to use file names for reference files, I would like to note that such approach could fail pretty fast if system will collect more and more data and there will rise a need to have more granular and detailed features to identify reference results and this can lead to very long filenames, which will be hard to use. I propose to use UUID4-based identification, which provides very low chances for collision. Those IDs will be keys in the database with all information required for clear identification of relevant results and corresponding details. Moreover this approach can be easily extended on the database side if more data is needed. Yes, UUID4 is not human-readable, but do we need such an option if we have tools, which can provide a better interface? For example, this could be something like: --- request: results-cli search -b "Test Suite D" -v "v1.2.3" -o "Ubuntu 22.04" -t "baseline" -m "response_time>100" response: [ { "id": "550e8400-e29b-41d4-a716-446655440005", "benchmark": "Test Suite A", "version": "v1.2.3", "target_os": "Ubuntu 22.04", "metrics": { "cpu_usage": 70.5, "memory_usage": 2048, "response_time": 120 }, "tags": ["baseline", "v1.0"], "created_at": "2024-10-25T10:00:00Z" }, ... ] --- or request: results-cli search "<Domain-Specific-Language-Query>" response: [ {}, {}, {}...] --- or request: results-cli get 550e8400-e29b-41d4-a716-446655440005 response: { "id": "550e8400-e29b-41d4-a716-446655440005", "benchmark": "Test Suite A", "version": "v1.2.3", "target_os": "Ubuntu 22.04", "metrics": { "cpu_usage": 70.5, "memory_usage": 2048, "response_time": 120 }, "tags": ["baseline", "v1.0"], "created_at": "2024-10-25T10:00:00Z" } --- or request: curl -X POST http://api.example.com/references/search \ -d '{ "query": "benchmark = \"Test Suite A\" AND (version >= \"v1.2\" OR tag IN [\"baseline\", \"regression\"]) AND cpu_usage > 60" }' ... --- Another point of use DB-based approach is the following: in case when a user works with particular hardware and/or would like to use a reference he/she does not need a full database with all collected reference values, but only a small slice of it. This slice can be downloaded from public repo or accessed via API. # Large results dataset If we collect a large benchmarks dataset in one place accompanied with detailed information about target systems from which this dataset was collected, then it will allow us to calculate precise baselines across different compositions of parameters, making performance deviations easier to detect. Long-term trend analysis can identify small changes and correlate them with updates, revealing performance drift. Another use of such a database - predictive modelling, which can provide forecasts of expected results and setting dynamic performance thresholds, enabling early issue detection. Anomaly detection becomes more effective with context, distinguishing unusual deviations from normal behaviour. # Interpretation of contradictory results It’s not clear how to deal with contradictory results to make a decision on regression presence. For example, we have a set of 10 tests, which test more or less the same, for example disk performance. It’s unclear what to do when one subset of tests show degradation and another subset shows neutral status or improvements. Is there a regression? I suppose that the availability of historical data can help to deal with such situations as historical data can show behaviour of particular tests and allow to assign weights in decision-making algorithms, but it’s just my guess. # System Profiles Tim's idea to reduce results to - “pass / fail” and my experience with various people trying to interpret benchmarking results led me to think of “profiles” - a set of parameters and metrics collected from a reference system while execution of a particular configuration of a particular benchmark. Profiles can be used for A/B comparison with pass/fail outcomes or match/not match, and this approach does not hide/miss the details and allows to capture multiple characteristics of the experiment, like presence of outliers/errors or skewed distribution form. Interested persons (like kernel developers or performance engineers, for example) can dig deeper to find a reason for such mismatch and those who are interested just in high-level results - pass/fail should be enough. Here is how I imaging a structure of a profile: --- profile_a: system packages: - pkg_1 - pkg_2 # Additional packages... settings: - cmdline # Additional settings... indicators: cpu: null ram: null loadavg: null # Additional indicators... benchmark: settings: param_1: null param_2: null param_x: null metrics: metric_1: null metric_2: null metric_x: null --- - System Packages, System Settings: Usually we do not pay much attention to this, but I think it’s worth highlighting that base OS is an important factor, as there are distribution-specific modifications present in the filesystem. Most commonly developers and researchers use Ubuntu (as the most popular distro) or Debian (as a cleaner and lightweight version of Ubuntu), but distributions apply their own patches to the kernel and system libraries, which may impact performance. Another kind of base OS - cloud OS images which can be modified by cloud providers to add internal packages & services which could potentially affect performance as well. While comparing we must take into account this aspect to compare apples-to-apples. - System Indicators: These are periodic statistics like CPU utilisation, RAM consumption, and other params collected before benchmarking, while benchmarking and after benchmarking. - Benchmark Settings: Benchmarking systems have multiple parameters, so it’s important to capture them and use them in analysis. - Benchmark Metrics: That’s obviously - benchmark results. It’s not a rare case when a benchmark test provides more than a single number. # Analysis Proposed rules-based analysis will work only for highly determined environments and systems, where rules can describe all the aspects. Rule-based systems are easier to understand and implement than other types, but for a small set of rules. However, we deal with the live system and it constantly evolves, so rules will deprecate extremely fast. It's the same story as with rule-based recommended systems in early years of machine learning. If you want to follow a rules-based approach, it's probably worth taking a look at https://www.clipsrules.net as this will allow to decouple results from analysis and avoid reinventing the analysis engine. Declaration of those rules will be error-prone due to the nature of their origin - they must be declared and maintained by humans. IMHO a human-less approach and use modern ML methods instead would be more beneficial in the long run. # Methodology Used methodology - another aspect which is not directly related to Tim's slides, but it's an important topic for results processing and interpretation, probably an idea of automated results interpretation can force use of one or another methodology. # Next steps I would be glad to participate in further discussions and share experience to improve kernel performance testing automation, analysis and interpretation of results. If there is interest, I'm open to collaborating on implementing some of these ideas. -- Best regards, Konstantin Belov

5 months, 2 weeks

2
1
0 0

[PATCH net-next v7 0/6] Suspend IRQs during application busy periods

by Joe Damato

Greetings: Welcome to v7, see changelog below. This revision drops patch 2 from the previous version, which was unnecessary as Jakub pointed out. The tests were fully re-run after dropping the patch to confirm the performance numbers hold; they do - see below. FAQ #2 has been updated to be more verbose, as well. This series introduces a new mechanism, IRQ suspension, which allows network applications using epoll to mask IRQs during periods of high traffic while also reducing tail latency (compared to existing mechanisms, see below) during periods of low traffic. In doing so, this balances CPU consumption with network processing efficiency. Martin Karsten (CC'd) and I have been collaborating on this series for several months and have appreciated the feedback from the community on our RFC [1]. We've updated the cover letter and kernel documentation in an attempt to more clearly explain how this mechanism works, how applications can use it, and how it compares to existing mechanisms in the kernel. I briefly mentioned this idea at netdev conf 2024 (for those who were there) and Martin described this idea in an earlier paper presented at Sigmetrics 2024 [2]. ~ The short explanation (TL;DR) We propose adding a new napi config parameter: irq_suspend_timeout to help balance CPU usage and network processing efficiency when using IRQ deferral and napi busy poll. If this parameter is set to a non-zero value *and* a user application has enabled preferred busy poll on a busy poll context (via the EPIOCSPARAMS ioctl introduced in commit 18e2bf0edf4d ("eventpoll: Add epoll ioctl for epoll_params")), then application calls to epoll_wait for that context will cause device IRQs and softirq processing to be suspended as long as epoll_wait successfully retrieves data from the NAPI. Each time data is retrieved, the irq_suspend_timeout is deferred. If/when network traffic subsides and epoll_wait returns no data, IRQ suspension is immediately reverted back to the existing napi_defer_hard_irqs and gro_flush_timeout mechanism which was introduced in commit 6f8b12d661d0 ("net: napi: add hard irqs deferral feature")). The irq_suspend_timeout serves as a safety mechanism. If userland takes a long time processing data, irq_suspend_timeout will fire and restart normal NAPI processing. For a more in depth explanation, please continue reading. ~ Comparison with existing mechanisms Interrupt mitigation can be accomplished in napi software, by setting napi_defer_hard_irqs and gro_flush_timeout, or via interrupt coalescing in the NIC. This can be quite efficient, but in both cases, a fixed timeout (or packet count) needs to be configured. However, a fixed timeout cannot effectively support both low- and high-load situations: At low load, an application typically processes a few requests and then waits to receive more input data. In this scenario, a large timeout will cause unnecessary latency. At high load, an application typically processes many requests before being ready to receive more input data. In this case, a small timeout will likely fire prematurely and trigger irq/softirq processing, which interferes with the application's execution. This causes overhead, most likely due to cache contention. While NICs attempt to provide adaptive interrupt coalescing schemes, these cannot properly take into account application-level processing. An alternative packet delivery mechanism is busy-polling, which results in perfect alignment of application processing and network polling. It delivers optimal performance (throughput and latency), but results in 100% cpu utilization and is thus inefficient for below-capacity workloads. We propose to add a new packet delivery mode that properly alternates between busy polling and interrupt-based delivery depending on busy and idle periods of the application. During a busy period, the system operates in busy-polling mode, which avoids interference. During an idle period, the system falls back to interrupt deferral, but with a small timeout to avoid excessive latencies. This delivery mode can also be viewed as an extension of basic interrupt deferral, but alternating between a small and a very large timeout. This delivery mode is efficient, because it avoids softirq execution interfering with application processing during busy periods. It can be used with blocking epoll_wait to conserve cpu cycles during idle periods. The effect of alternating between busy and idle periods is that performance (throughput and latency) is very close to full busy polling, while cpu utilization is lower and very close to interrupt mitigation. ~ Usage details IRQ suspension is introduced via a per-NAPI configuration parameter that controls the maximum time that IRQs can be suspended. Here's how it is intended to work: - The user application (or system administrator) uses the netdev-genl netlink interface to set the pre-existing napi_defer_hard_irqs and gro_flush_timeout NAPI config parameters to enable IRQ deferral. - The user application (or system administrator) sets the proposed irq_suspend_timeout parameter via the netdev-genl netlink interface to a larger value than gro_flush_timeout to enable IRQ suspension. - The user application issues the existing epoll ioctl to set the prefer_busy_poll flag on the epoll context. - The user application then calls epoll_wait to busy poll for network events, as it normally would. - If epoll_wait returns events to userland, IRQs are suspended for the duration of irq_suspend_timeout. - If epoll_wait finds no events and the thread is about to go to sleep, IRQ handling using napi_defer_hard_irqs and gro_flush_timeout is resumed. As long as epoll_wait is retrieving events, IRQs (and softirq processing) for the NAPI being polled remain disabled. When network traffic reduces, eventually a busy poll loop in the kernel will retrieve no data. When this occurs, regular IRQ deferral using gro_flush_timeout for the polled NAPI is re-enabled. Unless IRQ suspension is continued by subsequent calls to epoll_wait, it automatically times out after the irq_suspend_timeout timer expires. Regular deferral is also immediately re-enabled when the epoll context is destroyed. ~ Usage scenario The target scenario for IRQ suspension as packet delivery mode is a system that runs a dominant application with substantial network I/O. The target application can be configured to receive input data up to a certain batch size (via epoll_wait maxevents parameter) and this batch size determines the worst-case latency that application requests might experience. Because packet delivery is suspended during the target application's processing, the batch size also determines the worst-case latency of concurrent applications using the same RX queue(s). gro_flush_timeout should be set as small as possible, but large enough to make sure that a single request is likely not being interfered with. irq_suspend_timeout is largely a safety mechanism against misbehaving applications. It should be set large enough to cover the processing of an entire application batch, i.e., the factor between gro_flush_timeout and irq_suspend_timeout should roughly correspond to the maximum batch size that the target application would process in one go. ~ Important call out in the implementation - Enabling per epoll-context preferred busy poll will now effectively lead to a nonblocking iteration through napi_busy_loop, even when busy_poll_usecs is 0. See patch 4. ~ Benchmark configs & descriptions The changes were benchmarked with memcached [3] using the benchmarking tool mutilate [4]. To facilitate benchmarking, a small patch [5] was applied to memcached 1.6.29 to allow setting per-epoll context preferred busy poll and other settings via environment variables. Another small patch [6] was applied to libevent to enable full busy-polling. Multiple scenarios were benchmarked as described below and the scripts used for producing these results can be found on github [7] (note: all scenarios use NAPI-based traffic splitting via SO_INCOMING_ID by passing -N to memcached): - base: - no other options enabled - deferX: - set defer_hard_irqs to 100 - set gro_flush_timeout to X,000 - napibusy: - set defer_hard_irqs to 100 - set gro_flush_timeout to 200,000 - enable busy poll via the existing ioctl (busy_poll_usecs = 64, busy_poll_budget = 64, prefer_busy_poll = true) - fullbusy: - set defer_hard_irqs to 100 - set gro_flush_timeout to 5,000,000 - enable busy poll via the existing ioctl (busy_poll_usecs = 1000, busy_poll_budget = 64, prefer_busy_poll = true) - change memcached's nonblocking epoll_wait invocation (via libevent) to using a 1 ms timeout - suspend0: - set defer_hard_irqs to 0 - set gro_flush_timeout to 0 - set irq_suspend_timeout to 20,000,000 - enable busy poll via the existing ioctl (busy_poll_usecs = 0, busy_poll_budget = 64, prefer_busy_poll = true) - suspendX: - set defer_hard_irqs to 100 - set gro_flush_timeout to X,000 - set irq_suspend_timeout to 20,000,000 - enable busy poll via the existing ioctl (busy_poll_usecs = 0, busy_poll_budget = 64, prefer_busy_poll = true) ~ Benchmark results Tested on: Single socket AMD EPYC 7662 64-Core Processor Hyperthreading disabled 4 NUMA Zones (NPS=4) 16 CPUs per NUMA zone (64 cores total) 2 x Dual port 100gbps Mellanox Technologies ConnectX-5 Ex EN NIC The test machine is configured such that a single interface has 8 RX queues. The queues' IRQs and memcached are pinned to CPUs that are NUMA-local to the interface which is under test. The NIC's interrupt coalescing configuration is left at boot-time defaults. Results: Results are shown below. The mechanism added by this series is represented by the 'suspend' cases. Data presented shows a summary over nearly 10 runs of each test case [8] using the scripts on github [7]. For latency, the median is shown. For throughput and CPU utilization, the average is shown. The results also include cycles-per-query (cpq) and instruction-per-query (ipq) metrics, following the methodology proposed in [2], to augment the CPU utilization numbers, which could be skewed due to frequency scaling. We find that this does not appear to be the case as CPU utilization and low-level metrics show similar trends. These results were captured using the scripts on github [7] to illustrate how this approach compares with other pre-existing mechanisms. This data is not to be interpreted as scientific data captured in a fully isolated lab setting, but instead as best effort, illustrative information comparing and contrasting tradeoffs. The absolute QPS results shift between submissions, but the relative differences are equivalent. As patches are rebased, several factors likely influence overall performance. Compare: - Throughput (MAX) and latencies of base vs suspend. - CPU usage of napibusy and fullbusy during lower load (200K, 400K for example) vs suspend. - Latency of the defer variants vs suspend as timeout and load increases. - suspend0, which sets defer_hard_irqs and gro_flush_timeout to 0, has nearly the same performance as the base case (this is FAQ item #1). The overall takeaway is that the suspend variants provide a superior combination of high throughput, low latency, and low cpu utilization compared to all other variants. Each of the suspend variants works very well, but some fine-tuning between latency and cpu utilization is still possible by tuning the small timeout (gro_flush_timeout). Note: we've reorganized the results to make comparison among testcases with the same load easier. testcase load qps avglat 95%lat 99%lat cpu cpq ipq base 200K 199946 112 239 416 26 12973 11343 defer10 200K 199971 54 124 142 29 19412 17460 defer20 200K 199986 60 130 153 26 15644 14095 defer50 200K 200025 79 144 182 23 12122 11632 defer200 200K 199999 164 254 309 19 8923 9635 fullbusy 200K 199998 46 118 133 100 43658 23133 napibusy 200K 199983 100 237 277 56 24840 24716 suspend0 200K 200020 105 249 432 30 14264 11796 suspend10 200K 199950 53 123 141 32 19518 16903 suspend20 200K 200037 58 126 151 30 16426 14736 suspend50 200K 199961 73 136 177 26 13310 12633 suspend200 200K 199998 149 251 306 21 9566 10203 testcase load qps avglat 95%lat 99%lat cpu cpq ipq base 400K 400014 139 269 707 41 9476 9343 defer10 400K 400016 59 133 166 53 13991 12989 defer20 400K 399952 67 140 172 47 12063 11644 defer50 400K 400007 87 162 198 39 9384 9880 defer200 400K 399979 181 274 330 31 7089 8430 fullbusy 400K 399987 50 123 156 100 21827 16037 napibusy 400K 400014 76 222 272 83 18185 16529 suspend0 400K 400015 127 350 776 47 10699 9603 suspend10 400K 400023 57 129 164 54 13758 13178 suspend20 400K 400043 62 135 169 49 12071 11826 suspend50 400K 400071 76 149 186 42 10011 10301 suspend200 400K 399961 154 269 327 34 7827 8774 testcase load qps avglat 95%lat 99%lat cpu cpq ipq base 600K 599951 149 266 574 61 9265 8876 defer10 600K 600006 71 147 203 76 11866 10936 defer20 600K 600123 76 152 203 66 10430 10342 defer50 600K 600162 95 172 217 54 8526 9142 defer200 600K 599942 200 301 357 46 6977 8212 fullbusy 600K 599990 55 127 177 100 14551 13983 napibusy 600K 600035 63 160 250 96 13937 14140 suspend0 600K 599903 127 320 732 68 10166 8963 suspend10 600K 599908 63 137 192 69 10902 11100 suspend20 600K 599961 66 141 194 65 9976 10370 suspend50 600K 599973 80 159 204 57 8678 9381 suspend200 600K 600010 157 277 346 48 7133 8381 testcase load qps avglat 95%lat 99%lat cpu cpq ipq base 800K 800039 181 300 536 87 9585 8304 defer10 800K 800038 181 530 939 96 10564 8970 defer20 800K 800029 112 225 329 90 10056 8935 defer50 800K 799999 120 208 296 82 9234 8562 defer200 800K 800066 227 338 401 63 7117 8129 fullbusy 800K 800040 61 134 190 100 10913 12608 napibusy 800K 799944 64 141 214 99 10828 12588 suspend0 800K 799911 126 248 509 85 9346 8498 suspend10 800K 800006 69 143 200 83 9410 9845 suspend20 800K 800120 74 150 207 78 8786 9454 suspend50 800K 799989 87 168 224 71 7946 8833 suspend200 800K 799987 160 292 357 62 6923 8229 testcase load qps avglat 95%lat 99%lat cpu cpq ipq base 1000K 906879 4079 5751 6216 98 9496 7904 defer10 1000K 860849 3643 6274 6730 99 10040 8676 defer20 1000K 896063 3298 5840 6349 98 9620 8237 defer50 1000K 919782 2962 5513 5807 97 9284 7951 defer200 1000K 970941 3059 5348 5984 95 8593 7959 fullbusy 1000K 999950 70 150 207 100 8732 10777 napibusy 1000K 999996 78 154 223 100 8722 10656 suspend0 1000K 949706 2666 5770 6660 99 9071 8046 suspend10 1000K 1000024 80 160 220 92 8137 9035 suspend20 1000K 1000059 83 165 226 89 7850 8804 suspend50 1000K 999955 95 180 240 84 7411 8459 suspend200 1000K 999914 163 299 366 77 6833 8078 testcase load qps avglat 95%lat 99%lat cpu cpq ipq base MAX 1037654 4184 5453 5810 100 8411 7938 defer10 MAX 905607 4840 6151 6380 100 9639 8431 defer20 MAX 986463 4455 5594 5796 100 8848 8110 defer50 MAX 1077030 4000 5073 5299 100 8104 7920 defer200 MAX 1040728 4152 5385 5765 100 8379 7849 fullbusy MAX 1247536 3518 3935 3984 100 6998 7930 napibusy MAX 1136310 3799 7756 9964 100 7670 7877 suspend0 MAX 1057509 4132 5724 6185 100 8253 7918 suspend10 MAX 1215147 3580 3957 4041 100 7185 7944 suspend20 MAX 1216469 3576 3953 3988 100 7175 7950 suspend50 MAX 1215871 3577 3961 4075 100 7181 7949 suspend200 MAX 1216882 3556 3951 3988 100 7175 7955 ~ FAQ - Why is a new parameter needed? Does irq_suspend_timeout override gro_flush_timeout? Using the suspend mechanism causes the system to alternate between polling mode and irq-driven packet delivery. During busy periods, irq_suspend_timeout overrides gro_flush_timeout and keeps the system busy polling, but when epoll finds no events, the setting of gro_flush_timeout and napi_defer_hard_irqs determine the next step. There are essentially three possible loops for network processing and packet delivery: 1) hardirq -> softirq -> napi poll; basic interrupt delivery 2) timer -> softirq -> napi poll; deferred irq processing 3) epoll -> busy-poll -> napi poll; busy looping Loop 2 can take control from Loop 1, if gro_flush_timeout and napi_defer_hard_irqs are set. If gro_flush_timeout and napi_defer_hard_irqs are set, Loops 2 and 3 "wrestle" with each other for control. During busy periods, irq_suspend_timeout is used as timer in Loop 2, which essentially tilts this in favour of Loop 3. If gro_flush_timeout and napi_defer_hard_irqs are not set, Loop 3 cannot take control from Loop 1. Therefore, setting gro_flush_timeout and napi_defer_hard_irqs is the recommended usage, because otherwise setting irq_suspend_timeout might not have any discernible effect. This is shown in the results above: compare suspend0 with the base case. Note that the lack of napi_defer_hard_irqs and gro_flush_timeout produce similar results for both, which encourages the use of napi_defer_hard_irqs and gro_flush_timeout in addition to irq_suspend_timeout. - Can the new timeout value be threaded through the new epoll ioctl ? It is possible, but presents challenges for userspace. User applications must ensure that the file descriptors added to epoll contexts have the same NAPI ID to support busy polling. An epoll context is not permanently tied to any particular NAPI ID. So, a user application could decide to clear the file descriptors from the context and add a new set of file descriptors with a different NAPI ID to the context. Busy polling would work as expected, but the meaning of the suspend timeout becomes ambiguous because IRQs are not inherently associated with epoll contexts, but rather with the NAPI. The user program would need to reissue the ioctl to set the irq_suspend_timeout, but the napi_defer_hard_irqs and gro_flush_timeout settings would come from the NAPI's napi_config (which are set either by sysfs or by netlink). Such an interface seems awkard to use from a user perspective. Further, IRQs are related to NAPIs, which is why they are stored in the napi_config space. Putting the irq_suspend_timeout in the epoll context while other IRQ deferral mechanisms remain in the NAPI's napi_config space seems like an odd design choice. We've opted to keep all of the IRQ deferral parameters together and place the irq_suspend_timeout in napi_config. This has nice benefits for userspace: if a user app were to remove all file descriptors from an epoll context and add new file descriptors with a new NAPI ID, the correct suspend timeout for that NAPI ID would be used automatically without the user application needing to do anything (like re-issuing an ioctl, for example). All IRQ deferral related parameters are in one place and can all be set the same way: with netlink. - Can irq suspend be built by combining NIC coalescing and gro_flush_timeout ? No. The problem is that the long timeout must engage if and only if prefer-busy is active. When using NIC coalescing for the short timeout (without napi_defer_hard_irqs/gro_flush_timeout), an interrupt after an idle period will trigger softirq, which will run napi polling. At this point, prefer-busy is not active, so NIC interrupts would be re-enabled. Then it is not possible for the longer timeout to interject to switch control back to polling. In other words, only by using the software timer for the short timeout, it is possible to extend the timeout without having to reprogram the NIC timer or reach down directly and disable interrupts. Using gro_flush_timeout for the long timeout also has problems, for the same underlying reason. In the current napi implementation, gro_flush_timeout is not tied to prefer-busy. We'd either have to change that and in the process modify the existing deferral mechanism, or introduce a state variable to determine whether gro_flush_timeout is used as long timeout for irq suspend or whether it is used for its default purpose. In an earlier version, we did try something similar to the latter and made it work, but it ends up being a lot more convoluted than our current proposal. - Isn't it already possible to combine busy looping with irq deferral? Yes, in fact enabling irq deferral via napi_defer_hard_irqs and gro_flush_timeout is a precondition for prefer_busy_poll to have an effect. If the application also uses a tight busy loop with essentially nonblocking epoll_wait (accomplished with a very short timeout parameter), this is the fullbusy case shown in the results. An application using blocking epoll_wait is shown as the napibusy case in the results. It's a hybrid approach that provides limited latency benefits compared to the base case and plain irq deferral, but not as good as fullbusy or suspend. ~ Special thanks Several people were involved in earlier stages of the development of this mechanism whom we'd like to thank: - Peter Cai (CC'd), for the initial kernel patch and his contributions to the paper. - Mohammadamin Shafie (CC'd), for testing various versions of the kernel patch and providing helpful feedback. Thanks, Martin and Joe [1]: https://lore.kernel.org/netdev/20240812125717.413108-1-jdamato@fastly.com/ [2]: https://doi.org/10.1145/3626780 [3]: https://github.com/memcached/memcached/blob/master/doc/napi_ids.txt [4]: https://github.com/leverich/mutilate [5]: https://raw.githubusercontent.com/martinkarsten/irqsuspend/main/patches/mem… [6]: https://raw.githubusercontent.com/martinkarsten/irqsuspend/main/patches/lib… [7]: https://github.com/martinkarsten/irqsuspend [8]: https://github.com/martinkarsten/irqsuspend/tree/main/results v7: - Jakub noted that patch 2 adds unnecessary complexity by checking the suspend timeout in the NAPI loop. This makes the code more complicated and difficult to reason about. He's right; we've dropped patch 2 which simplifies this series. - Updated the cover letter with a full re-run of all test cases. - Updated FAQ #2. v6: https://lore.kernel.org/netdev/20241104215542.215919-1-jdamato@fastly.com/ - Updated the cover letter with a full re-run of all test cases, including a new case suspend0, as requested by Sridhar previously. - Updated the kernel documentation in patch 7 as suggested by Bagas Sanjaya, which improved the htmldoc output. v5: https://lore.kernel.org/netdev/20241103052421.518856-1-jdamato@fastly.com/ - Adjusted patch 5 to only suspend IRQs when ep_send_events returns a positive return value. This issue was pointed out by Hillf Danton. - Updated the commit message of patch 6 which still mentioned netcat, despite the code being updated in v4 to replace it with socat and fixed misspelling of netdevsim. - Fixed a minor typo in patch 7 and removed an unnecessary paragraph. - Added Sridhar Samudrala's Reviewed-by to patch 1-5 and 7. v4: https://lore.kernel.org/netdev/20241102005214.32443-1-jdamato@fastly.com/ - Added a new FAQ item to cover letter. - Updated patch 6 to use socat instead of nc in busy_poll_test.sh and updated busy_poller.c to use netlink directly to configure napi params. - Updated the kernel documentation in patch 7 to include more details. - Dropped Stanislav's Acked-by and Bagas' Reviewed-by from patch 7 since the documentation was updated. v3: https://lore.kernel.org/netdev/20241101004846.32532-1-jdamato@fastly.com/ - Added Stanislav Fomichev's Acked-by to every patch except the newly added selftest. - Added Bagas Sanjaya's Reviewed-by to the documentation patch. - Fixed the commit message of patch 2 to remove a reference to the now non-existent sysfs setting. - Added a self test which tests both "regular" busy poll and busy poll with suspend enabled. This was added as patch 6 as requested by Paolo. netdevsim was chosen instead of veth due to netdevsim's pre-existing support for netdev-genl. See the commit message of patch 6 for more details. v2: https://lore.kernel.org/bpf/20241021015311.95468-1-jdamato@fastly.com/ - Cover letter updated, including a re-run of test data. - Patch 1 rewritten to use netdev-genl instead of sysfs. - Patch 3 updated with a comment added to napi_resume_irqs. - Patch 4 rebased to apply now that commit b9ca079dd6b0 ("eventpoll: Annotate data-race of busy_poll_usecs") has been picked up from VFS. - Patch 6 updated the kernel documentation. rfc -> v1: - Cover letter updated to include more details. - Patch 1 updated to remove the documentation added. This was moved to patch 6 with the rest of the docs (see below). - Patch 5 updated to fix an error uncovered by the kernel build robot. See patch 5's changelog for more details. - Patch 6 added which updates kernel documentation. Joe Damato (2): selftests: net: Add busy_poll_test docs: networking: Describe irq suspension Martin Karsten (4): net: Add napi_struct parameter irq_suspend_timeout net: Add control functions for irq suspension eventpoll: Trigger napi_busy_loop, if prefer_busy_poll is set eventpoll: Control irq suspension for prefer_busy_poll Documentation/netlink/specs/netdev.yaml | 7 + Documentation/networking/napi.rst | 170 ++++++++- fs/eventpoll.c | 36 +- include/linux/netdevice.h | 2 + include/net/busy_poll.h | 3 + include/uapi/linux/netdev.h | 1 + net/core/dev.c | 41 +++ net/core/dev.h | 25 ++ net/core/netdev-genl-gen.c | 5 +- net/core/netdev-genl.c | 12 + tools/include/uapi/linux/netdev.h | 1 + tools/testing/selftests/net/.gitignore | 1 + tools/testing/selftests/net/Makefile | 3 +- tools/testing/selftests/net/busy_poll_test.sh | 164 +++++++++ tools/testing/selftests/net/busy_poller.c | 328 ++++++++++++++++++ 15 files changed, 792 insertions(+), 7 deletions(-) create mode 100755 tools/testing/selftests/net/busy_poll_test.sh create mode 100644 tools/testing/selftests/net/busy_poller.c base-commit: dc7c381bb8649e3701ed64f6c3e55316675904d7 -- 2.25.1

5 months, 2 weeks

1
1
0 0

[PATCH net-next v1 0/7] devmem TCP fixes

by Mina Almasry

A few unrelated devmem TCP fixes bundled in a series for some convenience (if that's ok). Patch 1-2: fix naming and provide page_pool_alloc_netmem for fragged netmem. Patch 3-4: fix issues with dma-buf dma addresses being potentially passed to dma_sync_for_* helpers. Patch 5-6: fix syzbot SO_DEVMEM_DONTNEED issue and add test for this case. Mina Almasry (6): net: page_pool: rename page_pool_alloc_netmem to *_netmems net: page_pool: create page_pool_alloc_netmem page_pool: disable sync for cpu for dmabuf memory provider netmem: add netmem_prefetch net: fix SO_DEVMEM_DONTNEED looping too long ncdevmem: add test for too many token_count Samiullah Khawaja (1): page_pool: Set `dma_sync` to false for devmem memory provider include/net/netmem.h | 7 ++++ include/net/page_pool/helpers.h | 50 ++++++++++++++++++-------- include/net/page_pool/types.h | 2 +- net/core/devmem.c | 9 +++-- net/core/page_pool.c | 11 +++--- net/core/sock.c | 46 ++++++++++++++---------- tools/testing/selftests/net/ncdevmem.c | 11 ++++++ 7 files changed, 93 insertions(+), 43 deletions(-) -- 2.47.0.163.g1226f6d8fa-goog

5 months, 2 weeks

3
9
0 0

[PATCH v6 00/10] iommufd: Add vIOMMU infrastructure (Part-2: vDEVICE)

by Nicolin Chen

Following the previous vIOMMU series, this adds another vDEVICE structure, representing the association from an iommufd_device to an iommufd_viommu. This gives the whole architecture a new "v" layer: _______________________________________________________________________ | iommufd (with vIOMMU/vDEVICE) | | _____________ _____________ | | | | | | | | |----------------| vIOMMU |<---| vDEVICE |<------| | | | | | |_____________| | | | | ______ | | _____________ ___|____ | | | | | | | | | | | | | | | IOAS |<---|(HWPT_PAGING)|<---| HWPT_NESTED |<--| DEVICE | | | | |______| |_____________| |_____________| |________| | |______|________|______________|__________________|_______________|_____| | | | | | ______v_____ | ______v_____ ______v_____ ___v__ | struct | | PFN | (paging) | | (nested) | |struct| |iommu_device| |------>|iommu_domain|<----|iommu_domain|<----|device| |____________| storage|____________| |____________| |______| This vDEVICE object is used to collect and store all vIOMMU-related device information/attributes in a VM. As an initial series for vDEVICE, add only the virt_id to the vDEVICE, which is a vIOMMU specific device ID in a VM: e.g. vSID of ARM SMMUv3, vDeviceID of AMD IOMMU, and vRID of Intel VT-d to a Context Table. This virt_id helps IOMMU drivers to link the vID to a pID of the device against the physical IOMMU instance. This is essential for a vIOMMU-based invalidation, where the request contains a device's vID for a device cache flush, e.g. ATC invalidation. Therefore, with this vDEVICE object, support a vIOMMU-based invalidation, by reusing IOMMUFD_CMD_HWPT_INVALIDATE for a vIOMMU object to flush cache with a given driver data. As for the implementation of the series, add driver support in ARM SMMUv3 for a real world use case. This series is on Github: https://github.com/nicolinc/iommufd/commits/iommufd_viommu_p2-v6 (QEMU branch for testing will be provided in Jason's nesting series) Changelog v6 * Fixed kdoc in the uAPI header * Fixed indentations in iommufd.rst * Replaced vdev->idev with vdev->dev * Added "Reviewed-by" from Kevin and Jason * Updated kdoc of struct iommu_vdevice_alloc * Fixed lockdep function call in iommufd_viommu_find_dev * Added missing iommu_dev validation between viommu and idev * Skipped SMMUv3 driver changes (to post in a separate series) * Replaced !cache_invalidate_user in WARN_ON of the allocation path with cache_invalidate_user validation in iommufd_hwpt_invalidate v5 https://lore.kernel.org/all/cover.1729897278.git.nicolinc@nvidia.com/ * Dropped driver-allocated vDEVICE support * Changed vdev_to_dev helper to iommufd_viommu_find_dev v4 https://lore.kernel.org/all/cover.1729555967.git.nicolinc@nvidia.com/ * Added missing brackets in switch-case * Fixed the unreleased idev refcount issue * Reworked the iommufd_vdevice_alloc allocator * Dropped support for IOMMU_VIOMMU_TYPE_DEFAULT * Added missing TEST_LENGTH and fail_nth coverages * Added a verification to the driver-allocated vDEVICE object * Added an iommufd_vdevice_abort for a missing mutex protection * Added a u64 structure arm_vsmmu_invalidation_cmd for user command conversion v3 https://lore.kernel.org/all/cover.1728491532.git.nicolinc@nvidia.com/ * Added Jason's Reviewed-by * Split this invalidation part out of the part-1 series * Repurposed VDEV_ID ioctl to a wider vDEVICE structure and ioctl * Reduced viommu_api functions by allowing drivers to access viommu and vdevice structure directly * Dropped vdevs_rwsem by using xa_lock instead * Dropped arm_smmu_cache_invalidate_user v2 https://lore.kernel.org/all/cover.1724776335.git.nicolinc@nvidia.com/ * Limited vdev_id to one per idev * Added a rw_sem to protect the vdev_id list * Reworked driver-level APIs with proper lockings * Added a new viommu_api file for IOMMUFD_DRIVER config * Dropped useless iommu_dev point from the viommu structure * Added missing index numnbers to new types in the uAPI header * Dropped IOMMU_VIOMMU_INVALIDATE uAPI; Instead, reuse the HWPT one * Reworked mock_viommu_cache_invalidate() using the new iommu helper * Reordered details of set/unset_vdev_id handlers for proper lockings v1 https://lore.kernel.org/all/cover.1723061377.git.nicolinc@nvidia.com/ Thanks! Nicolin Jason Gunthorpe (1): iommu: Add iommu_copy_struct_from_full_user_array helper Nicolin Chen (9): iommufd/viommu: Add IOMMUFD_OBJ_VDEVICE and IOMMU_VDEVICE_ALLOC ioctl iommufd/selftest: Add IOMMU_VDEVICE_ALLOC test coverage iommu/viommu: Add cache_invalidate to iommufd_viommu_ops iommufd: Allow hwpt_id to carry viommu_id for IOMMU_HWPT_INVALIDATE iommufd/viommu: Add iommufd_viommu_find_dev helper iommufd/selftest: Add mock_viommu_cache_invalidate iommufd/selftest: Add IOMMU_TEST_OP_DEV_CHECK_CACHE test command iommufd/selftest: Add vIOMMU coverage for IOMMU_HWPT_INVALIDATE ioctl Documentation: userspace-api: iommufd: Update vDEVICE drivers/iommu/iommufd/iommufd_private.h | 18 ++ drivers/iommu/iommufd/iommufd_test.h | 30 +++ include/linux/iommu.h | 48 ++++- include/linux/iommufd.h | 22 ++ include/uapi/linux/iommufd.h | 31 ++- tools/testing/selftests/iommu/iommufd_utils.h | 83 +++++++ drivers/iommu/iommufd/driver.c | 13 ++ drivers/iommu/iommufd/hw_pagetable.c | 40 +++- drivers/iommu/iommufd/main.c | 6 + drivers/iommu/iommufd/selftest.c | 98 ++++++++- drivers/iommu/iommufd/viommu.c | 76 +++++++ tools/testing/selftests/iommu/iommufd.c | 204 +++++++++++++++++- .../selftests/iommu/iommufd_fail_nth.c | 4 + Documentation/userspace-api/iommufd.rst | 41 +++- 14 files changed, 688 insertions(+), 26 deletions(-) -- 2.43.0

5 months, 2 weeks

4
25
0 0

[PATCH v6 00/13] iommufd: Add vIOMMU infrastructure (Part-1)

by Nicolin Chen

This series introduces a new vIOMMU infrastructure and related ioctls. IOMMUFD has been using the HWPT infrastructure for all cases, including a nested IO page table support. Yet, there're limitations for an HWPT-based structure to support some advanced HW-accelerated features, such as CMDQV on NVIDIA Grace, and HW-accelerated vIOMMU on AMD. Even for a multi-IOMMU environment, it is not straightforward for nested HWPTs to share the same parent HWPT (stage-2 IO pagetable), with the HWPT infrastructure alone: a parent HWPT typically hold one stage-2 IO pagetable and tag it with only one ID in the cache entries. When sharing one large stage-2 IO pagetable across physical IOMMU instances, that one ID may not always be available across all the IOMMU instances. In other word, it's ideal for SW to have a different container for the stage-2 IO pagetable so it can hold another ID that's available. And this container will be able to hold some advanced feature too. For this "different container", add vIOMMU, an additional layer to hold extra virtualization information: _______________________________________________________________________ | iommufd (with vIOMMU) | | _____________ | | | | | | |----------------| vIOMMU | | | | ______ | | _____________ ________ | | | | | | | | | | | | | | | IOAS |<---|(HWPT_PAGING)|<---| HWPT_NESTED |<--| DEVICE | | | | |______| |_____________| |_____________| |________| | | | | | | | | |______|________|______________|__________________|_______________|_____| | | | | | ______v_____ | ______v_____ ______v_____ ___v__ | struct | | PFN | (paging) | | (nested) | |struct| |iommu_device| |------>|iommu_domain|<----|iommu_domain|<----|device| |____________| storage|____________| |____________| |______| The vIOMMU object should be seen as a slice of a physical IOMMU instance that is passed to or shared with a VM. That can be some HW/SW resources: - Security namespace for guest owned ID, e.g. guest-controlled cache tags - Non-device-affiliated event reporting, e.g. invalidation queue errors - Access to a sharable nesting parent pagetable across physical IOMMUs - Virtualization of various platforms IDs, e.g. RIDs and others - Delivery of paravirtualized invalidation - Direct assigned invalidation queues - Direct assigned interrupts On a multi-IOMMU system, the vIOMMU object must be instanced to the number of the physical IOMMUs that have a slice passed to (via device) a guest VM, while being able to hold the shareable parent HWPT. Each vIOMMU then just needs to allocate its own individual ID to tag its own cache: ---------------------------- ---------------- | | paging_hwpt0 | | hwpt_nested0 |--->| viommu0 ------------------ ---------------- | | IDx | ---------------------------- ---------------------------- ---------------- | | paging_hwpt0 | | hwpt_nested1 |--->| viommu1 ------------------ ---------------- | | IDy | ---------------------------- As an initial part-1, add IOMMUFD_CMD_VIOMMU_ALLOC ioctl for an allocation only. More vIOMMU-based structs and ioctls will be introduced in the follow-up series to support vDEVICE, vIRQ (vEVENT) and vQUEUE objects. Although we repurposed the vIOMMU object from an earlier RFC, just for a referece: https://lore.kernel.org/all/cover.1712978212.git.nicolinc@nvidia.com/ This series is on Github: https://github.com/nicolinc/iommufd/commits/iommufd_viommu_p1-v6 (QEMU branch for testing will be provided in Jason's nesting series) Changelog v6 * Improved comment lines * Added a TEST_F for IO page fault * Fixed indentations in iommufd.rst * Revised kdoc of the viommu_alloc op * Added "Reviewed-by" from Kevin and Jason * Passed in "flags" to ->alloc_domain_nested * Renamed "free" op to "destroy" in viommu_ops * Skipped SMMUv3 driver changes (to post in a separate series) * Fixed "flags" validation in iommufd_viommu_alloc_hwpt_nested * Added CONFIG_IOMMUFD_DRIVER_CORE for sharing between iommufd core and IOMMU dirvers * Replaced iommufd_verify_unfinalized_object with xa_cmpxchg in iommufd_object_finalize/abort functions v5 https://lore.kernel.org/all/cover.1729897352.git.nicolinc@nvidia.com/ * Added "Reviewed-by" from Kevin * Reworked iommufd_viommu_alloc helper * Revised the uAPI kdoc for vIOMMU object * Revised comments for pluggable iommu_dev * Added a couple of cleanup patches for selftest * Renamed domain_alloc_nested op to alloc_domain_nested * Updated a few commit messages to reflect the latest series * Renamed iommufd_hwpt_nested_alloc_for_viommu to iommufd_viommu_alloc_hwpt_nested, and added flag validation v4 https://lore.kernel.org/all/cover.1729553811.git.nicolinc@nvidia.com/ * Added "Reviewed-by" from Jason * Dropped IOMMU_VIOMMU_TYPE_DEFAULT support * Dropped iommufd_object_alloc_elm renamings * Renamed iommufd's viommu_api.c to driver.c * Reworked iommufd_viommu_alloc helper * Added a separate iommufd_hwpt_nested_alloc_for_viommu function for hwpt_nested allocations on a vIOMMU, and added comparison between viommu->iommu_dev->ops and dev_iommu_ops(idev->dev) * Replaced s2_parent with vsmmu in arm_smmu_nested_domain * Replaced domain_alloc_user in iommu_ops with domain_alloc_nested in viommu_ops * Replaced wait_queue_head_t with a completion, to delay the unplug of mock_iommu_dev * Corrected documentation graph that was missing struct iommu_device * Added an iommufd_verify_unfinalized_object helper to verify driver- allocated vIOMMU/vDEVICE objects * Added missing test cases for TEST_LENGTH and fail_nth v3 https://lore.kernel.org/all/cover.1728491453.git.nicolinc@nvidia.com/ * Rebased on top of Jason's nesting v3 series https://lore.kernel.org/all/0-v3-e2e16cd7467f+2a6a1-smmuv3_nesting_jgg@nvid… * Split the series into smaller parts * Added Jason's Reviewed-by * Added back viommu->iommu_dev * Added support for driver-allocated vIOMMU v.s. core-allocated * Dropped arm_smmu_cache_invalidate_user * Added an iommufd_test_wait_for_users() in selftest * Reworked test code to make viommu an individual FIXTURE * Added missing TEST_LENGTH case for the new ioctl command v2 https://lore.kernel.org/all/cover.1724776335.git.nicolinc@nvidia.com/ * Limited vdev_id to one per idev * Added a rw_sem to protect the vdev_id list * Reworked driver-level APIs with proper lockings * Added a new viommu_api file for IOMMUFD_DRIVER config * Dropped useless iommu_dev point from the viommu structure * Added missing index numnbers to new types in the uAPI header * Dropped IOMMU_VIOMMU_INVALIDATE uAPI; Instead, reuse the HWPT one * Reworked mock_viommu_cache_invalidate() using the new iommu helper * Reordered details of set/unset_vdev_id handlers for proper lockings v1 https://lore.kernel.org/all/cover.1723061377.git.nicolinc@nvidia.com/ Thanks! Nicolin Nicolin Chen (13): iommufd: Move struct iommufd_object to public iommufd header iommufd: Move _iommufd_object_alloc helper to a sharable file iommufd: Introduce IOMMUFD_OBJ_VIOMMU and its related struct iommufd: Verify object in iommufd_object_finalize/abort() iommufd/viommu: Add IOMMU_VIOMMU_ALLOC ioctl iommufd: Add alloc_domain_nested op to iommufd_viommu_ops iommufd: Allow pt_id to carry viommu_id for IOMMU_HWPT_ALLOC iommufd/selftest: Add container_of helpers iommufd/selftest: Prepare for mock_viommu_alloc_domain_nested() iommufd/selftest: Add refcount to mock_iommu_device iommufd/selftest: Add IOMMU_VIOMMU_TYPE_SELFTEST iommufd/selftest: Add IOMMU_VIOMMU_ALLOC test coverage Documentation: userspace-api: iommufd: Update vIOMMU drivers/iommu/iommufd/Kconfig | 5 + drivers/iommu/iommufd/Makefile | 8 +- drivers/iommu/iommufd/iommufd_private.h | 33 +-- drivers/iommu/iommufd/iommufd_test.h | 2 + include/linux/iommu.h | 14 + include/linux/iommufd.h | 83 ++++++ include/uapi/linux/iommufd.h | 54 +++- tools/testing/selftests/iommu/iommufd_utils.h | 28 ++ drivers/iommu/iommufd/driver.c | 40 +++ drivers/iommu/iommufd/hw_pagetable.c | 72 ++++- drivers/iommu/iommufd/main.c | 54 ++-- drivers/iommu/iommufd/selftest.c | 266 +++++++++++++----- drivers/iommu/iommufd/viommu.c | 81 ++++++ tools/testing/selftests/iommu/iommufd.c | 128 +++++++++ .../selftests/iommu/iommufd_fail_nth.c | 11 + Documentation/userspace-api/iommufd.rst | 69 ++++- 16 files changed, 795 insertions(+), 153 deletions(-) create mode 100644 drivers/iommu/iommufd/driver.c create mode 100644 drivers/iommu/iommufd/viommu.c -- 2.43.0

5 months, 2 weeks

3
23
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-kselftest-mirror November 2024