summaryrefslogtreecommitdiffstats
path: root/tools
Commit message (Collapse)AuthorAgeFilesLines
* Merge tag 'net-5.10-rc4' of ↵Linus Torvalds2020-11-129-18/+281
|\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net Pull networking fixes from Jakub Kicinski: "Current release - regressions: - arm64: dts: fsl-ls1028a-kontron-sl28: specify in-band mode for ENETC Current release - bugs in new features: - mptcp: provide rmem[0] limit offset to fix oops Previous release - regressions: - IPv6: Set SIT tunnel hard_header_len to zero to fix path MTU calculations - lan743x: correctly handle chips with internal PHY - bpf: Don't rely on GCC __attribute__((optimize)) to disable GCSE - mlx5e: Fix VXLAN port table synchronization after function reload Previous release - always broken: - bpf: Zero-fill re-used per-cpu map element - fix out-of-order UDP packets when forwarding with UDP GSO fraglists turned on: - fix UDP header access on Fast/frag0 UDP GRO - fix IP header access and skb lookup on Fast/frag0 UDP GRO - ethtool: netlink: add missing netdev_features_change() call - net: Update window_clamp if SOCK_RCVBUF is set - igc: Fix returning wrong statistics - ch_ktls: fix multiple leaks and corner cases in Chelsio TLS offload - tunnels: Fix off-by-one in lower MTU bounds for ICMP/ICMPv6 replies - r8169: disable hw csum for short packets on all chip versions - vrf: Fix fast path output packet handling with async Netfilter rules" * tag 'net-5.10-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (65 commits) lan743x: fix use of uninitialized variable net: udp: fix IP header access and skb lookup on Fast/frag0 UDP GRO net: udp: fix UDP header access on Fast/frag0 UDP GRO devlink: Avoid overwriting port attributes of registered port vrf: Fix fast path output packet handling with async Netfilter rules cosa: Add missing kfree in error path of cosa_write net: switch to the kernel.org patchwork instance ch_ktls: stop the txq if reaches threshold ch_ktls: tcb update fails sometimes ch_ktls/cxgb4: handle partial tag alone SKBs ch_ktls: don't free skb before sending FIN ch_ktls: packet handling prior to start marker ch_ktls: Correction in middle record handling ch_ktls: missing handling of header alone ch_ktls: Correction in trimmed_len calculation cxgb4/ch_ktls: creating skbs causes panic ch_ktls: Update cheksum information ch_ktls: Correction in finding correct length cxgb4/ch_ktls: decrypted bit is not enough net/x25: Fix null-ptr-deref in x25_connect ...
| * selftest: fix flower terse dump testsVlad Buslov2020-11-101-2/+2
| | | | | | | | | | | | | | | | | | | | Iproute2 tc classifier terse dump has been accepted with modified syntax. Update the tests accordingly. Signed-off-by: Vlad Buslov <vlad@buslov.dev> Fixes: e7534fd42a99 ("selftests: implement flower classifier terse dump tests") Link: https://lore.kernel.org/r/20201107111928.453534-1-vlad@buslov.dev Signed-off-by: Jakub Kicinski <kuba@kernel.org>
| * Merge git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpfJakub Kicinski2020-11-068-16/+279
| |\ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Alexei Starovoitov says: ==================== pull-request: bpf 2020-11-06 1) Pre-allocated per-cpu hashmap needs to zero-fill reused element, from David. 2) Tighten bpf_lsm function check, from KP. 3) Fix bpftool attaching to flow dissector, from Lorenz. 4) Use -fno-gcse for the whole kernel/bpf/core.c instead of function attribute, from Ard. * git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf: bpf: Update verification logic for LSM programs bpf: Zero-fill re-used per-cpu map element bpf: BPF_PRELOAD depends on BPF_SYSCALL tools/bpftool: Fix attaching flow dissector libbpf: Fix possible use after free in xsk_socket__delete libbpf: Fix null dereference in xsk_socket__delete libbpf, hashmap: Fix undefined behavior in hash_bits bpf: Don't rely on GCC __attribute__((optimize)) to disable GCSE tools, bpftool: Remove two unused variables. tools, bpftool: Avoid array index warnings. xsk: Fix possible memory leak at socket close bpf: Add struct bpf_redir_neigh forward declaration to BPF helper defs samples/bpf: Set rlimit for memlock to infinity in all samples bpf: Fix -Wshadow warnings selftest/bpf: Fix profiler test using CO-RE relocation for enums ==================== Link: https://lore.kernel.org/r/20201106221759.24143-1-alexei.starovoitov@gmail.com Signed-off-by: Jakub Kicinski <kuba@kernel.org>
| | * bpf: Zero-fill re-used per-cpu map elementDavid Verbeiren2020-11-052-0/+247
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Zero-fill element values for all other cpus than current, just as when not using prealloc. This is the only way the bpf program can ensure known initial values for all cpus ('onallcpus' cannot be set when coming from the bpf program). The scenario is: bpf program inserts some elements in a per-cpu map, then deletes some (or userspace does). When later adding new elements using bpf_map_update_elem(), the bpf program can only set the value of the new elements for the current cpu. When prealloc is enabled, previously deleted elements are re-used. Without the fix, values for other cpus remain whatever they were when the re-used entry was previously freed. A selftest is added to validate correct operation in above scenario as well as in case of LRU per-cpu map element re-use. Fixes: 6c9059817432 ("bpf: pre-allocate hash map elements") Signed-off-by: David Verbeiren <david.verbeiren@tessares.net> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Acked-by: Matthieu Baerts <matthieu.baerts@tessares.net> Acked-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20201104112332.15191-1-david.verbeiren@tessares.net
| | * tools/bpftool: Fix attaching flow dissectorLorenz Bauer2020-11-051-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | My earlier patch to reject non-zero arguments to flow dissector attach broke attaching via bpftool. Instead of 0 it uses -1 for target_fd. Fix this by passing a zero argument when attaching the flow dissector. Fixes: 1b514239e859 ("bpf: flow_dissector: Check value of unused flags to BPF_PROG_ATTACH") Reported-by: Jiri Benc <jbenc@redhat.com> Signed-off-by: Lorenz Bauer <lmb@cloudflare.com> Signed-off-by: Alexei Starovoitov <ast@kernel.org> Acked-by: Song Liu <songliubraving@fb.com> Link: https://lore.kernel.org/bpf/20201105115230.296657-1-lmb@cloudflare.com
| | * libbpf: Fix possible use after free in xsk_socket__deleteMagnus Karlsson2020-11-041-2/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix a possible use after free in xsk_socket__delete that will happen if xsk_put_ctx() frees the ctx. To fix, save the umem reference taken from the context and just use that instead. Fixes: 2f6324a3937f ("libbpf: Support shared umems between queues and devices") Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/1604396490-12129-3-git-send-email-magnus.karlsson@gmail.com
| | * libbpf: Fix null dereference in xsk_socket__deleteMagnus Karlsson2020-11-041-1/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix a possible null pointer dereference in xsk_socket__delete that will occur if a null pointer is fed into the function. Fixes: 2f6324a3937f ("libbpf: Support shared umems between queues and devices") Reported-by: Andrii Nakryiko <andrii.nakryiko@gmail.com> Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/1604396490-12129-2-git-send-email-magnus.karlsson@gmail.com
| | * libbpf, hashmap: Fix undefined behavior in hash_bitsIan Rogers2020-11-021-6/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If bits is 0, the case when the map is empty, then the >> is the size of the register which is undefined behavior - on x86 it is the same as a shift by 0. Fix by handling the 0 case explicitly and guarding calls to hash_bits for empty maps in hashmap__for_each_key_entry and hashmap__for_each_entry_safe. Fixes: e3b924224028 ("libbpf: add resizable non-thread safe internal hashmap") Suggested-by: Andrii Nakryiko <andriin@fb.com>, Signed-off-by: Ian Rogers <irogers@google.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Acked-by: Andrii Nakryiko <andrii@kernel.org> Acked-by: Song Liu <songliubraving@fb.com> Link: https://lore.kernel.org/bpf/20201029223707.494059-1-irogers@google.com
| | * tools, bpftool: Remove two unused variables.Ian Rogers2020-10-291-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Avoid an unused variable warning. Signed-off-by: Ian Rogers <irogers@google.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Reviewed-by: Tobias Klauser <tklauser@distanz.ch> Acked-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20201027233646.3434896-2-irogers@google.com
| | * tools, bpftool: Avoid array index warnings.Ian Rogers2020-10-291-1/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The bpf_caps array is shorter without CAP_BPF, avoid out of bounds reads if this isn't defined. Working around this avoids -Wno-array-bounds with clang. Signed-off-by: Ian Rogers <irogers@google.com> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Reviewed-by: Tobias Klauser <tklauser@distanz.ch> Acked-by: Andrii Nakryiko <andrii@kernel.org> Link: https://lore.kernel.org/bpf/20201027233646.3434896-1-irogers@google.com
| | * selftest/bpf: Fix profiler test using CO-RE relocation for enumsAndrii Nakryiko2020-10-261-3/+8
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Instead of hard-coding invalid pids_cgrp_id, use Kconfig to detect the presence of that enum value and CO-RE to capture its actual value in the hosts's kernel. Fixes: 03d4d13fab3f ("selftests/bpf: Add profiler test") Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net> Tested-by: Song Liu <songliubraving@fb.com> Link: https://lore.kernel.org/bpf/20201022202739.3667367-1-andrii@kernel.org
* | | Merge branch 'turbostat' of ↵Linus Torvalds2020-11-104-138/+507
|\ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux Pull turbostat updates from Len Brown: "Update update to version 20.09.30, one kernel side fix" * 'turbostat' of git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux: tools/power turbostat: update version number powercap: restrict energy meter to root access tools/power turbostat: harden against cpu hotplug tools/power turbostat: adjust for temperature offset tools/power turbostat: Build with _FILE_OFFSET_BITS=64 tools/power turbostat: Support AMD Family 19h tools/power turbostat: Remove empty columns for Jacobsville tools/power turbostat: Add a new GFXAMHz column that exposes gt_act_freq_mhz. tools/power x86_energy_perf_policy: Input/output error in a VM tools/power turbostat: Skip pc8, pc9, pc10 columns, if they are disabled tools/power turbostat: Support additional CPU model numbers tools/power turbostat: Fix output formatting for ACPI CST enumeration tools/power turbostat: Replace HTTP links with HTTPS ones: TURBOSTAT UTILITY tools/power turbostat: Use sched_getcpu() instead of hardcoded cpu 0 tools/power turbostat: Enable accumulate RAPL display tools/power turbostat: Introduce functions to accumulate RAPL consumption tools/power turbostat: Make the energy variable to be 64 bit tools/power turbostat: Always print idle in the system configuration header tools/power turbostat: Print /dev/cpu_dma_latency
| * | | tools/power turbostat: update version numberLen Brown2020-11-101-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | goodbye summer... Signed-off-by: Len Brown <len.brown@intel.com>
| * | | tools/power turbostat: harden against cpu hotplugLen Brown2020-10-231-9/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | turbostat tends to get confused when CPUs are added and removed while it is running. There are races, such as checking the current cpu, and then reading a sysfs file that depends on that cpu number. Close the two issues that seem to come up the most. First, there is an infinite reset loop detector -- change that to allow more resets before giving up. Secondly, one of those file reads didn't really need to exit the program on failure... Signed-off-by: Len Brown <len.brown@intel.com>
| * | | tools/power turbostat: adjust for temperature offsetLen Brown2020-10-231-33/+29
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | cpu1: MSR_IA32_TEMPERATURE_TARGET: 0x05640000 (95 C) (100 default - 5 offset) Account for the new "offset" field in MSR_TEMPERATURE_TARGET. While this field is usually zero, ignoring it results in over-stating the current temperature, both per-core and per-package. Signed-off-by: Len Brown <len.brown@intel.com>
| * | | tools/power turbostat: Build with _FILE_OFFSET_BITS=64Alexander Monakov2020-09-031-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | For compatibility reasons, Glibc off_t is a 32-bit type on 32-bit x86 unless _FILE_OFFSET_BITS=64 is defined. Add this define, as otherwise reading MSRs with index 0x80000000 and above attempts a pread with a negative offset, which fails. Signed-off-by: Alexander Monakov <amonakov@ispras.ru> Tested-by: Liwei Song <liwei.song@windriver.com> Signed-off-by: Len Brown <len.brown@intel.com>
| * | | tools/power turbostat: Support AMD Family 19hKim Phillips2020-09-031-23/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Family 19h processors have the same RAPL (Running average power limit) hardware register interface as Family 17h processors. Change the family checks to succeed for Family 17h and above to enable core and package energy measurement on Family 19h machines. Also update the TDP to the largest found at the bottom of the page at amd.com->processors->servers->epyc->2nd-gen-epyc, i.e., the EPYC 7H12. Signed-off-by: Kim Phillips <kim.phillips@amd.com> Cc: Len Brown <len.brown@intel.com> Cc: Len Brown <lenb@kernel.org> Cc: linux-pm@vger.kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by: Len Brown <len.brown@intel.com>
| * | | tools/power turbostat: Remove empty columns for JacobsvilleAntti Laakso2020-09-031-3/+30
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Jacobsville doesn't have Package C2 and C6. Also Core and DRAM RAPL are not available. Adjust output accordingly. Signed-off-by: Antti Laakso <antti.laakso@linux.intel.com> Signed-off-by: Len Brown <len.brown@intel.com>
| * | | tools/power turbostat: Add a new GFXAMHz column that exposes gt_act_freq_mhz.Rafael Antognolli2020-09-031-0/+50
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The column already present called GFXMHz reads from gt_cur_freq_mhz, which represents the GT frequency that was requested, but power management might not be able to do that. So the new column will display what the actual frequency GT is running at. Signed-off-by: Rafael Antognolli <rafael.antognolli@intel.com> Signed-off-by: Len Brown <len.brown@intel.com>
| * | | tools/power x86_energy_perf_policy: Input/output error in a VMOndřej Lysoněk2020-09-031-13/+54
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | I've encountered an issue with x86_energy_perf_policy. If I run it on a machine that I'm told is a qemu-kvm virtual machine running inside a privileged container, I get the following error: x86_energy_perf_policy: /dev/cpu/0/msr offset 0x1ad read failed: Input/output error I get the same error in a Digital Ocean droplet, so that might be a similar environment. I created the following patch which is intended to give a more user-friendly message. It's based on a patch for turbostat from Prarit Bhargava that was posted some time ago. The patch is "[v2] turbostat: Running on virtual machine is not supported" [1]. Given my limited knowledge of the topic, I can't say with confidence that this is the right solution, though (that's why this is not an official patch submission). Also, I'm not sure what the convention with exit codes is in this tool. Also, instead of the error message, perhaps the tool should just not print anything in this case, which is how it behaves in a "regular" VM? [1] https://patchwork.kernel.org/patch/9868587/ Signed-off-by: Ondřej Lysoněk <olysonek@redhat.com> Signed-off-by: Len Brown <len.brown@intel.com>
| * | | tools/power turbostat: Skip pc8, pc9, pc10 columns, if they are disabledLen Brown2020-09-031-3/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Like we skip PC3 and PC6 columns when the package C-state limit disables them, skip PC8/PC9/CP10 under analogous conditions. Reported-by: Zhang Rui <rui.zhang@intel.com> Signed-off-by: Len Brown <len.brown@intel.com>
| * | | tools/power turbostat: Support additional CPU model numbersLen Brown2020-09-031-0/+4
| | | | | | | | | | | | | | | | | | | | | | | | Initial support for models recently added to intel-family.h. Signed-off-by: Len Brown <len.brown@intel.com>
| * | | tools/power turbostat: Fix output formatting for ACPI CST enumerationDavid Arcari2020-09-031-0/+20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | turbostat formatting is broken with ACPI CST for enumeration. The problem is that the CX_ACPI% is eight characters long which does not work with tab formatting. One simple solution is to remove the underbar from the state name such that C1_ACPI will be displayed as C1ACPI. Signed-off-by: David Arcari <darcari@redhat.com> Cc: Len Brown <lenb@kernel.org> Cc: linux-kernel@vger.kernel.org Signed-off-by: Len Brown <len.brown@intel.com>
| * | | tools/power turbostat: Replace HTTP links with HTTPS ones: TURBOSTAT UTILITYAlexander A. Klimov2020-09-031-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Rationale: Reduces attack surface on kernel devs opening the links for MITM as HTTPS traffic is much harder to manipulate. Deterministic algorithm: For each file: If not .svg: For each line: If doesn't contain `\bxmlns\b`: For each link, `\bhttp://[^# \t\r\n]*(?:\w|/)`: If neither `\bgnu\.org/license`, nor `\bmozilla\.org/MPL\b`: If both the HTTP and HTTPS versions return 200 OK and serve the same content: Replace HTTP with HTTPS. Signed-off-by: Alexander A. Klimov <grandmaster@al2klimov.de> Signed-off-by: Len Brown <len.brown@intel.com>
| * | | tools/power turbostat: Use sched_getcpu() instead of hardcoded cpu 0Prarit Bhargava2020-09-031-3/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Disabling cpu 0 results in an error turbostat: /sys/devices/system/cpu/cpu0/topology/thread_siblings: open failed: No such file or directory Use sched_getcpu() instead of a hardcoded cpu 0 to get the max cpu number. Signed-off-by: Prarit Bhargava <prarit@redhat.com> Signed-off-by: Len Brown <len.brown@intel.com>
| * | | tools/power turbostat: Enable accumulate RAPL displayChen Yu2020-09-031-22/+16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Enable the accumulated RAPL display by default. Signed-off-by: Chen Yu <yu.c.chen@intel.com> Signed-off-by: Len Brown <len.brown@intel.com>
| * | | tools/power turbostat: Introduce functions to accumulate RAPL consumptionChen Yu2020-09-032-7/+219
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Since the RAPL Joule Counter is 32 bit, turbostat would only print a *star* instead of printing the actual energy consumed to indicate the overflow due to long duration. This does not meet the requirement from servers as the sampling time of turbostat is usually very long on servers. So maintain a set of MSR buffer, and update them periodically before the 32bit MSR register is wrapped round, so as to avoid the overflow. The idea is similar to the implementation of ktime_get(): Periodical MSR timer: total_rapl_sum += (current_rapl_msr - last_rapl_msr); Using get_msr_sum() to get the accumulated RAPL: return (current_rapl_msr - last_rapl_msr) + total_rapl_sum; The accumulated RAPL mechanism will be turned on in next patch. Originally-by: Aaron Lu <aaron.lwe@gmail.com> Reviewed-by: Doug Smythies <dsmythies@telus.net> Tested-by: Doug Smythies <dsmythies@telus.net> Signed-off-by: Chen Yu <yu.c.chen@intel.com> Signed-off-by: Len Brown <len.brown@intel.com>
| * | | tools/power turbostat: Make the energy variable to be 64 bitChen Yu2020-09-031-17/+13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Change the energy variable from 32bit to 64bit, so that it can record long time duration. After this conversion, adjust the DELTA_WRAP32() accordingly. Signed-off-by: Chen Yu <yu.c.chen@intel.com> Signed-off-by: Len Brown <len.brown@intel.com>
| * | | tools/power turbostat: Always print idle in the system configuration headerDoug Smythies2020-09-031-3/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | If the --quiet option is not used, turbostat prints a useful system configuration header during startup. But inclusion of idle system configuration information in this header is currently a function of inclusion in the columns chosen to be displayed. Always list this idle system configuration. Signed-off-by: Doug Smythies <dsmythies@telus.net> Signed-off-by: Len Brown <len.brown@intel.com>
| * | | tools/power turbostat: Print /dev/cpu_dma_latencyLen Brown2020-09-031-0/+28
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Users are puzzled when they use tuned performance and all their C-states vanish. Dump /dev/cpu_dma_latency and state whether the value is default, or constraining, to explain this situation. Signed-off-by: Len Brown <len.brown@intel.com>
* | | | Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvmLinus Torvalds2020-11-0923-277/+2252
|\ \ \ \ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Pull kvm fixes from Paolo Bonzini: "ARM: - fix compilation error when PMD and PUD are folded - fix regression in reads-as-zero behaviour of ID_AA64ZFR0_EL1 - add aarch64 get-reg-list test x86: - fix semantic conflict between two series merged for 5.10 - fix (and test) enforcement of paravirtual cpuid features selftests: - various cleanups to memory management selftests - new selftests testcase for performance of dirty logging" * tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm: (30 commits) KVM: selftests: allow two iterations of dirty_log_perf_test KVM: selftests: Introduce the dirty log perf test KVM: selftests: Make the number of vcpus global KVM: selftests: Make the per vcpu memory size global KVM: selftests: Drop pointless vm_create wrapper KVM: selftests: Add wrfract to common guest code KVM: selftests: Simplify demand_paging_test with timespec_diff_now KVM: selftests: Remove address rounding in guest code KVM: selftests: Factor code out of demand_paging_test KVM: selftests: Use a single binary for dirty/clear log test KVM: selftests: Always clear dirty bitmap after iteration KVM: selftests: Add blessed SVE registers to get-reg-list KVM: selftests: Add aarch64 get-reg-list test selftests: kvm: test enforcement of paravirtual cpuid features selftests: kvm: Add exception handling to selftests selftests: kvm: Clear uc so UCALL_NONE is being properly reported selftests: kvm: Fix the segment descriptor layout to match the actual layout KVM: x86: handle MSR_IA32_DEBUGCTLMSR with report_ignored_msrs kvm: x86: request masterclock update any time guest uses different msr kvm: x86: ensure pv_cpuid.features is initialized when enabling cap ...
| * | | | KVM: selftests: allow two iterations of dirty_log_perf_testPaolo Bonzini2020-11-091-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Even though one iteration is not enough for the dirty log performance test (due to the cost of building page tables, zeroing memory etc.) two is okay and it is the default. Without this patch, "./dirty_log_perf_test" without any further arguments fails. Cc: Ben Gardon <bgardon@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
| * | | | KVM: selftests: Introduce the dirty log perf testBen Gardon2020-11-086-8/+396
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The dirty log perf test will time verious dirty logging operations (enabling dirty logging, dirtying memory, getting the dirty log, clearing the dirty log, and disabling dirty logging) in order to quantify dirty logging performance. This test can be used to inform future performance improvements to KVM's dirty logging infrastructure. This series was tested by running the following invocations on an Intel Skylake machine: dirty_log_perf_test -b 20m -i 100 -v 64 dirty_log_perf_test -b 20g -i 5 -v 4 dirty_log_perf_test -b 4g -i 5 -v 32 demand_paging_test -b 20m -v 64 demand_paging_test -b 20g -v 4 demand_paging_test -b 4g -v 32 All behaved as expected. Signed-off-by: Ben Gardon <bgardon@google.com> Message-Id: <20201027233733.1484855-6-bgardon@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
| * | | | KVM: selftests: Make the number of vcpus globalAndrew Jones2020-11-082-20/+20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We also check the input number of vcpus against the maximum supported. Signed-off-by: Andrew Jones <drjones@redhat.com> Message-Id: <20201104212357.171559-8-drjones@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
| * | | | KVM: selftests: Make the per vcpu memory size globalAndrew Jones2020-11-082-12/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Rename vcpu_memory_bytes to something with "percpu" in it in order to be less ambiguous. Also make it global to simplify things. Signed-off-by: Andrew Jones <drjones@redhat.com> Message-Id: <20201104212357.171559-7-drjones@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
| * | | | KVM: selftests: Drop pointless vm_create wrapperAndrew Jones2020-11-084-9/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Andrew Jones <drjones@redhat.com> Message-Id: <20201104212357.171559-3-drjones@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
| * | | | KVM: selftests: Add wrfract to common guest codeBen Gardon2020-11-082-1/+7
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Wrfract will be used by the dirty logging perf test introduced later in this series to dirty memory sparsely. This series was tested by running the following invocations on an Intel Skylake machine: dirty_log_perf_test -b 20m -i 100 -v 64 dirty_log_perf_test -b 20g -i 5 -v 4 dirty_log_perf_test -b 4g -i 5 -v 32 demand_paging_test -b 20m -v 64 demand_paging_test -b 20g -v 4 demand_paging_test -b 4g -v 32 All behaved as expected. Signed-off-by: Ben Gardon <bgardon@google.com> Message-Id: <20201027233733.1484855-5-bgardon@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
| * | | | KVM: selftests: Simplify demand_paging_test with timespec_diff_nowBen Gardon2020-11-083-15/+27
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add a helper function to get the current time and return the time since a given start time. Use that function to simplify the timekeeping in the demand paging test. This series was tested by running the following invocations on an Intel Skylake machine: dirty_log_perf_test -b 20m -i 100 -v 64 dirty_log_perf_test -b 20g -i 5 -v 4 dirty_log_perf_test -b 4g -i 5 -v 32 demand_paging_test -b 20m -v 64 demand_paging_test -b 20g -v 4 demand_paging_test -b 4g -v 32 All behaved as expected. Signed-off-by: Ben Gardon <bgardon@google.com> Message-Id: <20201027233733.1484855-4-bgardon@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
| * | | | KVM: selftests: Remove address rounding in guest codeBen Gardon2020-11-081-1/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Rounding the address the guest writes to a host page boundary will only have an effect if the host page size is larger than the guest page size, but in that case the guest write would still go to the same host page. There's no reason to round the address down, so remove the rounding to simplify the demand paging test. This series was tested by running the following invocations on an Intel Skylake machine: dirty_log_perf_test -b 20m -i 100 -v 64 dirty_log_perf_test -b 20g -i 5 -v 4 dirty_log_perf_test -b 4g -i 5 -v 32 demand_paging_test -b 20m -v 64 demand_paging_test -b 20g -v 4 demand_paging_test -b 4g -v 32 All behaved as expected. Signed-off-by: Ben Gardon <bgardon@google.com> Message-Id: <20201027233733.1484855-3-bgardon@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
| * | | | KVM: selftests: Factor code out of demand_paging_testBen Gardon2020-11-082-181/+210
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Much of the code in demand_paging_test can be reused by other, similar multi-vCPU-memory-touching-perfromance-tests. Factor that common code out for reuse. No functional change expected. This series was tested by running the following invocations on an Intel Skylake machine: dirty_log_perf_test -b 20m -i 100 -v 64 dirty_log_perf_test -b 20g -i 5 -v 4 dirty_log_perf_test -b 4g -i 5 -v 32 demand_paging_test -b 20m -v 64 demand_paging_test -b 20g -v 4 demand_paging_test -b 4g -v 32 All behaved as expected. Signed-off-by: Ben Gardon <bgardon@google.com> Message-Id: <20201027233733.1484855-2-bgardon@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
| * | | | KVM: selftests: Use a single binary for dirty/clear log testPeter Xu2020-11-083-39/+156
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Remove the clear_dirty_log test, instead merge it into the existing dirty_log_test. It should be cleaner to use this single binary to do both tests, also it's a preparation for the upcoming dirty ring test. The default behavior will run all the modes in sequence. Reviewed-by: Andrew Jones <drjones@redhat.com> Signed-off-by: Peter Xu <peterx@redhat.com> Message-Id: <20201001012233.6013-1-peterx@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
| * | | | KVM: selftests: Always clear dirty bitmap after iterationPeter Xu2020-11-081-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We used not to clear the dirty bitmap before because KVM_GET_DIRTY_LOG would overwrite it the next time it copies the dirty log onto it. In the upcoming dirty ring tests we'll start to fetch dirty pages from a ring buffer, so no one is going to clear the dirty bitmap for us. Reviewed-by: Andrew Jones <drjones@redhat.com> Signed-off-by: Peter Xu <peterx@redhat.com> Message-Id: <20201001012228.5916-1-peterx@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
| * | | | KVM: selftests: Add blessed SVE registers to get-reg-listAndrew Jones2020-11-084-42/+217
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add support for the SVE registers to get-reg-list and create a new test, get-reg-list-sve, which tests them when running on a machine with SVE support. Signed-off-by: Andrew Jones <drjones@redhat.com> Message-Id: <20201029201703.102716-5-drjones@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
| * | | | KVM: selftests: Add aarch64 get-reg-list testAndrew Jones2020-11-085-0/+703
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Check for KVM_GET_REG_LIST regressions. The blessed list was created by running on v4.15 with the --core-reg-fixup option. The following script was also used in order to annotate system registers with their names when possible. When new system registers are added the names can just be added manually using the same grep. while read reg; do if [[ ! $reg =~ ARM64_SYS_REG ]]; then printf "\t$reg\n" continue fi encoding=$(echo "$reg" | sed "s/ARM64_SYS_REG(//;s/),//") if ! name=$(grep "$encoding" ../../../../arch/arm64/include/asm/sysreg.h); then printf "\t$reg\n" continue fi name=$(echo "$name" | sed "s/.*SYS_//;s/[\t ]*sys_reg($encoding)$//") printf "\t$reg\t/* $name */\n" done < <(aarch64/get-reg-list --core-reg-fixup --list) Signed-off-by: Andrew Jones <drjones@redhat.com> Message-Id: <20201029201703.102716-3-drjones@redhat.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
| * | | | selftests: kvm: test enforcement of paravirtual cpuid featuresOliver Upton2020-11-087-0/+308
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add a set of tests that ensure the guest cannot access paravirtual msrs and hypercalls that have been disabled in the KVM_CPUID_FEATURES leaf. Expect a #GP in the case of msr accesses and -KVM_ENOSYS from hypercalls. Cc: Jim Mattson <jmattson@google.com> Signed-off-by: Oliver Upton <oupton@google.com> Reviewed-by: Peter Shier <pshier@google.com> Reviewed-by: Aaron Lewis <aaronlewis@google.com> Message-Id: <20201027231044.655110-7-oupton@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
| * | | | selftests: kvm: Add exception handling to selftestsAaron Lewis2020-11-089-9/+244
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Add the infrastructure needed to enable exception handling in selftests. This allows any of the exception and interrupt vectors to be overridden in the guest. Signed-off-by: Aaron Lewis <aaronlewis@google.com> Reviewed-by: Alexander Graf <graf@amazon.com> Message-Id: <20201012194716.3950330-4-aaronlewis@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
| * | | | selftests: kvm: Clear uc so UCALL_NONE is being properly reportedAaron Lewis2020-11-083-0/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Ensure the out value 'uc' in get_ucall() is properly reporting UCALL_NONE if the call fails. The return value will be correctly reported, however, the out parameter 'uc' will not be. Clear the struct to ensure the correct value is being reported in the out parameter. Signed-off-by: Aaron Lewis <aaronlewis@google.com> Reviewed-by: Andrew Jones <drjones@redhat.com> Reviewed-by: Alexander Graf <graf@amazon.com> Message-Id: <20201012194716.3950330-3-aaronlewis@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
| * | | | selftests: kvm: Fix the segment descriptor layout to match the actual layoutAaron Lewis2020-11-082-2/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix the layout of 'struct desc64' to match the layout described in the SDM Vol 3, Chapter 3 "Protected-Mode Memory Management", section 3.4.5 "Segment Descriptors", Figure 3-8 "Segment Descriptor". The test added later in this series relies on this and crashes if this layout is not correct. Signed-off-by: Aaron Lewis <aaronlewis@google.com> Reviewed-by: Alexander Graf <graf@amazon.com> Message-Id: <20201012194716.3950330-2-aaronlewis@google.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
* | | | | Merge tag 'linux-kselftest-fixes-5.10-rc3' of ↵Linus Torvalds2020-11-0634-56/+65
|\ \ \ \ \ | |_|_|/ / |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest Pull Kselftest fixes from Shuah Khan: "Fixes to the ftrace test and several fixes from Tommi Rantala for various other tests" * tag 'linux-kselftest-fixes-5.10-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest: selftests: binderfs: use SKIP instead of XFAIL selftests: clone3: use SKIP instead of XFAIL selftests: core: use SKIP instead of XFAIL in close_range_test.c selftests: proc: fix warning: _GNU_SOURCE redefined selftests: pidfd: drop needless linux/kcmp.h inclusion in pidfd_setns_test.c selftests: pidfd: add CONFIG_CHECKPOINT_RESTORE=y to config selftests: pidfd: skip test on kcmp() ENOSYS selftests: pidfd: use ksft_test_result_skip() when skipping test selftests/harness: prettify SKIP message whitespace again selftests: pidfd: fix compilation errors due to wait.h selftests: filter kselftest headers from command in lib.mk selftests/ftrace: check for do_sys_openat2 in user-memory test selftests/ftrace: Use $FUNCTION_FORK to reference kernel fork function
| * | | | selftests: binderfs: use SKIP instead of XFAILTommi Rantala2020-11-051-4/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | XFAIL is gone since commit 9847d24af95c ("selftests/harness: Refactor XFAIL into SKIP"), use SKIP instead. Fixes: 9847d24af95c ("selftests/harness: Refactor XFAIL into SKIP") Signed-off-by: Tommi Rantala <tommi.t.rantala@nokia.com> Reviewed-by: Kees Cook <keescook@chromium.org> Acked-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
generated by cgit v1.2.3 (git 2.25.1) at 2024-10-22 02:50:44 +0000