- 03 Feb, 2016 29 commits
-
-
Jiri Olsa authored
With the hist object having the perf_hpp_list we can now iterate sort format entries based in the hists object. Adding hists__for_each_sort_list macro to do that. Signed-off-by:
Jiri Olsa <jolsa@kernel.org> Cc: David Ahern <dsahern@gmail.com> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/1453109064-1026-27-git-send-email-jolsa@kernel.orgSigned-off-by:
Arnaldo Carvalho de Melo <acme@redhat.com>
-
Jiri Olsa authored
With the hist object having the perf_hpp_list we can now iterate output format entries based in the hists object. Adding hists__for_each_format macro to do that. Signed-off-by:
-
Jiri Olsa authored
Adding hpp_list into struct hists object. Initializing struct hists_evsel hists object to carry global perf_hpp_list list. Signed-off-by:
-
Jiri Olsa authored
Adding struct perf_hpp_list argument to following helper functions: void perf_hpp__setup_output_field(struct perf_hpp_list *list); void perf_hpp__reset_output_field(struct perf_hpp_list *list); void perf_hpp__append_sort_keys(struct perf_hpp_list *list); so they could be used on hists's hpp_list. Signed-off-by:
-
Jiri Olsa authored
Introducing perf_hpp_list__for_each_sort_list_safe macro to iterate perf_hpp_list object's sort entries safely. Signed-off-by:
-
Jiri Olsa authored
Introducing perf_hpp_list__for_each_sort_list macro to iterate perf_hpp_list object's sort entries. Signed-off-by:
-
Jiri Olsa authored
Introducing perf_hpp_list__for_each_format_safe macro to iterate perf_hpp_list object's output entries safely. Signed-off-by:
-
Jiri Olsa authored
Introducing perf_hpp_list__for_each_format macro to iterate perf_hpp_list object's output entries. Signed-off-by:
-
Jiri Olsa authored
Passing perf_hpp_list all the way through setup_output_list so the output entry could be added on the arbitrary list. Signed-off-by:
-
Jiri Olsa authored
Adding 2 perf_hpp_list register helpers: perf_hpp_list__column_register() perf_hpp_list__register_sort_field() to be called within existing helpers: perf_hpp__column_register() perf_hpp__register_sort_field() to register format entries within global perf_hpp_list object. Signed-off-by:
-
Jiri Olsa authored
Introducing perf_hpp_list__init function to have an easy way to initialize perf_hpp_list struct. Signed-off-by:
-
Jiri Olsa authored
Gather output and sort lists under struct perf_hpp_list, so we could have multiple instancies of sort/output format entries. Replacing current perf_hpp__list and perf_hpp__sort_list lists with single perf_hpp_list instance. Signed-off-by:
-
Jiri Olsa authored
Separating output fields parsing into setup_output_list function, so it's separated from field_order string setup and could be reused later in following patches. Signed-off-by:
-
Jiri Olsa authored
Separating sort fields parsing into setup_sort_list function, so it's separated from sort_order string setup and could be reused later in following patches. Signed-off-by:
-
Jiri Olsa authored
With multiple list holding format entries, we need the support properly releasing format output/sort fields. Signed-off-by:
-
Jiri Olsa authored
Those functions are no longer needed. They operate over perf_hpp__format array which is now used only as template for dynamic entries. Signed-off-by:
-
Jiri Olsa authored
Currently we use static output fields, because we have single global list of all sort/output fields. We will add hists specific sort and output lists in following patches, so we need all format entries to be dynamically allocated. Adding support to allocate output sort field. Signed-off-by:
-
Arnaldo Carvalho de Melo authored
The ui initialization changes hpp format callbacks, based on the used browser. Thus we need this init being processed before setup_sorting. Replica of a patch by Jiri for 'perf report'. Cc: David Ahern <dsahern@gmail.com> Cc: Jiri Olsa <jolsa@kernel.org> Cc: Namhyung Kim <namhyung@kernel.org> Cc: Peter Zijlstra <a.p.zijlstra@chello.nl> Link: http://lkml.kernel.org/r/1453109064-1026-9-git-send-email-jolsa@kernel.orgSigned-off-by:
-
Jiri Olsa authored
The ui initialization changes hpp format callbacks, based on the used browser. Thus we need this init being processed before setup_sorting. Signed-off-by:
-
Jiri Olsa authored
Now that we have the 'equal' method implemented for hpp format entries we can ease up the logic in the following functions and make them generic wrt comparing format entries: perf_hpp__setup_output_field perf_hpp__append_sort_keys Signed-off-by:
-
Jiri Olsa authored
Adding 'hpp__equal' callback function to compare hpp output format entries. Signed-off-by:
-
Jiri Olsa authored
To easily compare format entries and make it available for all kinds of format entries. Signed-off-by:
-
Jiri Olsa authored
We are going to add dynamic hpp format fields, so we need to make the 'len' change for the format itself, not in the perf_hpp__format template. Signed-off-by:
-
Jiri Olsa authored
Currently there's no way of comparing hpp format entries, which is needed in following patches. Adding _idx fields into struct perf_hpp_fmt to recognize and be able to compare hpp format entries. Signed-off-by:
-
Jiri Olsa authored
Adding evsel specific function to sort hists_evsel based hists. The hists__output_resort can be now used to sort common hists object. Signed-off-by:
-
Jiri Olsa authored
Currently hists__output_resort() depends on hists based on hists_evsel struct, but we need to be able to sort common hists as well. Cutting out the sorting base sorting code into output_resort function, so it can be reused in following patch. Signed-off-by:
-
Ingo Molnar authored
Merge tag 'perf-core-for-mingo-3' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/core Pull perf/core callchain fixes and improvements from Arnaldo Carvalho de Melo <acme@redhat.com: User visible changes: - Make --percent-limit apply to callchains also and fix some bugs related to --percent-limit (Namhyung Kim) Signed-off-by:
Ingo Molnar <mingo@kernel.org>
-
Ingo Molnar authored
Merge tag 'perf-core-for-mingo-2' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/core Pull perf tooling changes from Arnaldo Carvalho de Melo: New features: - Port 'perf kvm stat' to PowerPC (Hemant Kumar) Infrastructure changes: - Use the 'feature-dump' target to do the feature checks just once and then add code to reuse that in the tests/make makefile, speeding up the 'make -C tools/perf build-test' target (Wang Nan) - Reduce the number of tests the 'build-test' target do to those that don't pollute the source tree (Arnaldo Carvalho de Melo) - Improve the output of the build tests a bit by aligning the name of the tests, more can be done to filter out uninteresting info in the output (Arnaldo Carvalho de Melo) - Add perf_evlist pointer to *info_priv_size(), more prep work for supporting the coresight architecture (Mathieu Poirier) - Improve the 'perf test bp_signal' test (Wang Nan) - Check environment before starting the BPF 'perf test', so that we can just 'Skip' older kernels instead of 'FAIL'ing them (Wang Nan) - Fix cpumode of synthesized buildid event (Wang Nan) Signed-off-by:
-
Ingo Molnar authored
Merge tag 'perf-core-for-mingo' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux into perf/core Pull perf/core improvements and fixes from Arnaldo Carvalho de Melo: User visible changes: - Rename the "colors.code" ~/.perfconfig variable to "colors.jump_arrows", as it controls just the that UI element in the annotate browser (Taeung Song) - Avoid trying to read ELF symtabs from device files, noticed while doing memory profiling work (Jiri Olsa) - Improve context detection when offering options in the hists browser, i.e. some options don't make sense when the browser is not working with a perf.data file ('perf top' mode), only in 'perf report' mode, like scripting (Namhyung Kim) Infrastructure changes: - Elliminate duplication in the hists browser filter functions, getting the common part into a function that receives callbacks for filtering by DSO, thread, etc. (Namhyung Kim) - Fix misleadingly indented assignment, found using gcc6 -Wmisleading-indentation (Markus Trippelsdorf) - Handle LLVM relocation oddities in libbpf, introducing a 'perf test' that detects such problems and then fixing the problem, so that the test now passes (Wang Nan) - More improvements to the build infrastructure to allow reusing the feature detection facilities (Wang Nan) - Auto initialize the globals needed by cpu__max_{cpu,node}() routines (Arnaldo Carvalho de Melo) Documentation changes: - Document the perf sysctls in Documentation/sysctl/kernel.txt (Ben Hutchings) - Document a bunch more ~/.perfconfig knobs (Taeung Song) Signed-off-by:
-
- 01 Feb, 2016 11 commits
-
-
git://git.kernel.org/pub/scm/linux/kernel/git/davem/netLinus Torvalds authored
Pull networking fixes from David Miller: "This looks like a lot but it's a mixture of regression fixes as well as fixes for longer standing issues. 1) Fix on-channel cancellation in mac80211, from Johannes Berg. 2) Handle CHECKSUM_COMPLETE properly in xt_TCPMSS netfilter xtables module, from Eric Dumazet. 3) Avoid infinite loop in UDP SO_REUSEPORT logic, also from Eric Dumazet. 4) Avoid a NULL deref if we try to set SO_REUSEPORT after a socket is bound, from Craig Gallek. 5) GRO key comparisons don't take lightweight tunnels into account, from Jesse Gross. 6) Fix struct pid leak via SCM credentials in AF_UNIX, from Eric Dumazet. 7) We need to set the rtnl_link_ops of ipv6 SIT tunnels before we register them, otherwise the NEWLINK netlink message is missing the proper attributes. From Thadeu Lima de Souza Cascardo. 8) Several Spectrum chip bug fixes for mlxsw switch driver, from Ido Schimmel 9) Handle fragments properly in ipv4 easly socket demux, from Eric Dumazet. 10) Don't ignore the ifindex key specifier on ipv6 output route lookups, from Paolo Abeni" * git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (128 commits) tcp: avoid cwnd undo after receiving ECN irda: fix a potential use-after-free in ircomm_param_request net: tg3: avoid uninitialized variable warning net: nb8800: avoid uninitialized variable warning net: vxge: avoid unused function warnings net: bgmac: clarify CONFIG_BCMA dependency net: hp100: remove unnecessary #ifdefs net: davinci_cpdma: use dma_addr_t for DMA address ipv6/udp: use sticky pktinfo egress ifindex on connect() ipv6: enforce flowi6_oif usage in ip6_dst_lookup_tail() netlink: not trim skb for mmaped socket when dump vxlan: fix a out of bounds access in __vxlan_find_mac net: dsa: mv88e6xxx: fix port VLAN maps fib_trie: Fix shift by 32 in fib_table_lookup net: moxart: use correct accessors for DMA memory ipv4: ipconfig: avoid unused ic_proto_used symbol bnxt_en: Fix crash in bnxt_free_tx_skbs() during tx timeout. bnxt_en: Exclude rx_drop_pkts hw counter from the stack's rx_dropped counter. bnxt_en: Ring free response from close path should use completion ring net_sched: drr: check for NULL pointer in drr_dequeue ... -
git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6Linus Torvalds authored
Pull crypto fixes from Herbert Xu: "This fixes the following issues: API: - algif_hash needs to wait for init operations to complete. - The has_key setting for shash was always true. Algorithms: - Add missing selections of CRYPTO_HASH. - Fix pkcs7 authentication. Drivers: - Fix stack alignment bug in chacha20-ssse3. - Fix performance regression in caam due to incorrect setting. - Fix potential compile-only build failure of stm32" * 'linus' of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6: crypto: atmel-aes - remove calls of clk_prepare() from atomic contexts crypto: algif_hash - wait for crypto_ahash_init() to complete crypto: shash - Fix has_key setting hwrng: stm32 - Fix dependencies for !HAS_IOMEM archs crypto: ghash,poly1305 - select CRYPTO_HASH where needed crypto: chacha20-ssse3 - Align stack pointer to 64 bytes PKCS#7: Don't require SpcSpOpusInfo in Authenticode pkcs7 signatures crypto: caam - make write transactions bufferable on PPC platforms
-
git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimmLinus Torvalds authored
Pull libnvdimm fixes from Dan Williams: "1/ Fixes to the libnvdimm 'pfn' device that establishes a reserved area for storing a struct page array. 2/ Fixes for dax operations on a raw block device to prevent pagecache collisions with dax mappings. 3/ A fix for pfn_t usage in vm_insert_mixed that lead to a null pointer de-reference. These have received build success notification from the kbuild robot across 153 configs and pass the latest ndctl tests" * 'libnvdimm-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/nvdimm/nvdimm: phys_to_pfn_t: use phys_addr_t mm: fix pfn_t to page conversion in vm_insert_mixed block: use DAX for partition table reads block: revert runtime dax control of the raw block device fs, block: force direct-I/O for dax-enabled block devices devm_memremap_pages: fix vmem_altmap lifetime + alignment handling libnvdimm, pfn: fix restoring memmap location libnvdimm: fix mode determination for e820 devices -
Namhyung Kim authored
When all callchains of a hist entry is percent-limited, do not add a blank line at the end. It makes the entry look like it doesn't have callchains. Reported-and-Tested-by:
Namhyung Kim <namhyung@kernel.org> Cc: Andi Kleen <andi@firstfloor.org> Cc: David Ahern <dsahern@gmail.com> Cc: Frederic Weisbecker <fweisbec@gmail.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Wang Nan <wangnan0@huawei.com> Link: http://lkml.kernel.org/r/20160128122454.GA27446@danjae.kornetSigned-off-by:
-
Namhyung Kim authored
When there's only a single callchain, perf doesn't print its percentage in front of the symbols. This is because it assumes that the percentage is same as parents. But if a percent limit is applied, it's possible that there are actually a couple of child nodes but only one of them is shown. In this case it should display the percent to prevent misunderstanding of its percentage is same as the parent's. For example, let's see the following callchain. $ perf report --no-children --percent-limit 0.01 --tui ... - 0.06% sleep [kernel.vmlinux] [k] kmem_cache_alloc_trace kmem_cache_alloc_trace - perf_event_mmap - 0.04% mmap_region do_mmap_pgoff - vm_mmap_pgoff + 0.02% sys_mmap_pgoff + 0.02% vm_mmap + 0.02% mprotect_fixup Current code omits the percent if 'mmap_region' becomes the only node when percent limit is set to 0.03%, its percent is not 0.06% but users will assume it incorrectly. Before: $ perf report --no-children --percent-limit 0.03 --tui ... 0.06% sleep [kernel.vmlinux] [k] kmem_cache_alloc_trace kmem_cache_alloc_trace - perf_event_mmap - mmap_region do_mmap_pgoff vm_mmap_pgoff After: $ perf report --no-children --percent-limit 0.03 --tui ... 0.06% sleep [kernel.vmlinux] [k] kmem_cache_alloc_trace kmem_cache_alloc_trace - perf_event_mmap - 0.04% mmap_region do_mmap_pgoff vm_mmap_pgoff Signed-off-by: -
Namhyung Kim authored
Pass parent node's total period to callchain print functions. This info is needed by later patch to determine whether it can omit percent or not correctly. No functional change intended. Signed-off-by:
-
Namhyung Kim authored
The commit 8c430a34 ("perf hists browser: Support folded callchains") missed to update hist_browser__dump() so it always shows graph-style callchains regardless of current setting. To fix that, factor out callchain printing code and rename the existing function which prints graph-style callchain. Signed-off-by:
-
Namhyung Kim authored
When there's only a single callchain, perf doesn't print its percentage in front of the symbols. This is because it assumes that the percentage is same as parents. But if a percent limit is applied, it's possible that there are actually a couple of child nodes but only one of them is shown. In this case it should display the percent to prevent misunderstanding of its percentage is same as the parent's. For example, let's see the following callchain. $ perf report -s comm --percent-limit 0.01 --stdio ... 9.95% swapper | |--7.57%--intel_idle | cpuidle_enter_state | cpuidle_enter | call_cpuidle | cpu_startup_entry | | | |--4.89%--start_secondary | | | --2.68%--rest_init | start_kernel | x86_64_start_reservations | x86_64_start_kernel | |--0.15%--__schedule | | | |--0.13%--schedule | | schedule_preempt_disable | | cpu_startup_entry | | | | | |--0.09%--start_secondary | | | | | --0.04%--rest_init | | start_kernel | | x86_64_start_reservations | | x86_64_start_kernel | | | --0.01%--schedule_preempt_disabled | cpu_startup_entry ... Current code omits the percent if 'intel_idle' becomes the only node when percent limit is set to 0.5%, its percent is not 9.95% but users will assume it incorrectly. Before: $ perf report --percent-limit 0.5 --stdio ... 9.95% swapper | ---intel_idle cpuidle_enter_state cpuidle_enter call_cpuidle cpu_startup_entry | |--4.89%--start_secondary | --2.68%--rest_init start_kernel x86_64_start_reservations x86_64_start_kernel After: $ perf report --percent-limit 0.5 --stdio ... 9.95% swapper | --7.57%--intel_idle cpuidle_enter_state cpuidle_enter call_cpuidle cpu_startup_entry | |--4.89%--start_secondary | --2.68%--rest_init start_kernel x86_64_start_reservations x86_64_start_kernel Signed-off-by: -
Namhyung Kim authored
Pass hist entry's period to graph callchain print function. This info is needed by later patch to determine whether it can omit percentage of top-level node or not. No functional change intended. Signed-off-by:
-
Namhyung Kim authored
It's just a wrapper function to align the start position ofcallchains to 'comm' of each thread if it's a first sort key. But it doesn't not work with tracepoint events and also with upcoming hierarchy view. Signed-off-by:
-
Namhyung Kim authored
Currently --percent-limit option only works for hist entries. However it'd be better to have same effect to callchains as well Requested-by:
Andi Kleen <andi@firstfloor.org> Signed-off-by:
-
