aboutsummaryrefslogtreecommitdiff
path: root/tools
AgeCommit message (Collapse)AuthorFilesLines
2017-08-31usbip: auto retry for concurrent attachNobuo Iwata1-15/+20
This patch adds recovery from false busy state on concurrent attach operation. The procedure of attach operation is as below. 1) Find an unused port in /sys/devices/platform/vhci_hcd/status. (userspace) 2) Request attach found port to driver through /sys/devices/platform/vhci_hcd/attach. (userspace) 3) Lock table, reserve requested port and unlock table. (vhci driver) Attaching more than one remote devices concurrently, same unused port number will be found in step-1. Then one request will succeed and others will fail even though there are some unused ports. With this patch, driver returns EBUSY when requested port has already been used. In this case, attach command retries from step-1: finding another unused port. If there's no unused port, the attach operation will fail in step-1. Otherwise it retries automatically using another unused port. vhci-hcd's interface (only errno) is changed as following. Current errno New errno Condition EINVAL same as left specified port number is in invalid range EAGAIN same as left platform_get_drvdata() failed EINVAL same as left specified socket fd is not valid EINVAL EBUSY specified port status is not free The errno EBUSY was not used in userspace src/usbip_attach.c:import_device(). It is needed to distinguish the condition to be able to retry from other unrecoverable errors. It is possible to avoid this failure by introducing userspace exclusive control. But it's exaggerated for this special condition. The locking itself has done in driver. As an alternate solution, userspace doesn't specify port number, driver searches unused port and it returns port number to the userspace. With this solution, the interface is much different than this patch. Signed-off-by: Nobuo Iwata <[email protected]> Acked-by: Shuah Khan <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2017-08-31selftests/powerpc: Force ptrace tests to build -fno-pieMichael Neuling1-1/+1
Currently these tests won't build with a `--enable-default-pie` compiler as they require r30 to be clobbered. This gives an error: ptrace-tm-spd-gpr.c:41:2: error: PIC register clobbered by 'r30' in 'asm' This forces these tests to be built no-pie. Signed-off-by: Michael Neuling <[email protected]> Signed-off-by: Michael Ellerman <[email protected]>
2017-08-31powerpc/asm: Convert .llong directives to .8byteTobin C. Harding1-1/+1
.llong is an undocumented PPC specific directive. The generic equivalent is .quad, but even better (because it's self describing) is .8byte. Convert all .llong directives to .8byte. Signed-off-by: Tobin C. Harding <[email protected]> Signed-off-by: Michael Ellerman <[email protected]>
2017-08-30bpf: test_maps: fix typos, "conenct" and "listeen"Colin Ian King1-2/+2
Trivial fix to typos in printf error messages: "conenct" -> "connect" "listeen" -> "listen" thanks to Daniel Borkmann for spotting one of these mistakes Signed-off-by: Colin Ian King <[email protected]> Acked-by: John Fastabend <[email protected]> Acked-by: Daniel Borkmann <[email protected]> Acked-by: Shuah Khan <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2017-08-30perf report: Calculate the average cycles of iterationsJin Yao5-79/+85
The branch history code has a loop detection function. With this, we can get the number of iterations by calculating the removed loops. While it would be nice for knowing the average cycles of iterations. This patch adds up the cycles in branch entries of removed loops and save the result to the next branch entry (e.g. branch entry A). Finally it will display the iteration number and average cycles at the "from" of branch entry A. For example: perf record -g -j any,save_type ./div perf report --branch-history --no-children --stdio --22.63%--main div.c:42 (RET CROSS_2M) compute_flag div.c:28 (cycles:2 iter:173115 avg_cycles:2) | --10.73%--compute_flag div.c:27 (RET CROSS_2M) rand rand.c:28 (cycles:1) rand rand.c:28 (RET CROSS_2M) __random random.c:298 (cycles:1) __random random.c:297 (COND_BWD CROSS_2M) __random random.c:295 (cycles:1) __random random.c:295 (COND_BWD CROSS_2M) __random random.c:295 (cycles:1) __random random.c:295 (RET CROSS_2M) Signed-off-by: Yao Jin <[email protected]> Reviewed-by: Andi Kleen <[email protected]> Cc: Alexander Shishkin <[email protected]> Cc: Jiri Olsa <[email protected]> Cc: Kan Liang <[email protected]> Cc: Peter Zijlstra <[email protected]> Link: http://lkml.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2017-08-30objtool: Handle GCC stack pointer adjustment bugJosh Poimboeuf4-90/+88
Arnd Bergmann reported the following warning with GCC 7.1.1: fs/fs_pin.o: warning: objtool: pin_kill()+0x139: stack state mismatch: cfa1=7+88 cfa2=7+96 And the kbuild robot reported the following warnings with GCC 5.4.1: fs/fs_pin.o: warning: objtool: pin_kill()+0x182: return with modified stack frame fs/quota/dquot.o: warning: objtool: dquot_alloc_inode()+0x140: stack state mismatch: cfa1=7+120 cfa2=7+128 fs/quota/dquot.o: warning: objtool: dquot_free_inode()+0x11a: stack state mismatch: cfa1=7+112 cfa2=7+120 Those warnings are caused by an unusual GCC non-optimization where it uses an intermediate register to adjust the stack pointer. It does: lea 0x8(%rsp), %rcx ... mov %rcx, %rsp Instead of the obvious: add $0x8, %rsp It makes no sense to use an intermediate register, so I opened a GCC bug to track it: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81813 But it's not exactly a high-priority bug and it looks like we'll be stuck with this issue for a while. So for now we have to track register values when they're loaded with stack pointer offsets. This is kind of a big workaround for a tiny problem, but c'est la vie. I hope to eventually create a GCC plugin to implement a big chunk of objtool's functionality. Hopefully at that point we'll be able to remove of a lot of these GCC-isms from the objtool code. Reported-by: Arnd Bergmann <[email protected]> Reported-by: kbuild test robot <[email protected]> Signed-off-by: Josh Poimboeuf <[email protected]> Cc: Linus Torvalds <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Thomas Gleixner <[email protected]> Link: http://lkml.kernel.org/r/6a41a96884c725e7f05413bb7df40cfe824b2444.1504028945.git.jpoimboe@redhat.com Signed-off-by: Ingo Molnar <[email protected]>
2017-08-29tc-testing: add test for testing ife typeAlexander Aring1-0/+50
This patch adds a new testcase for the IFE type setting in tc. In case of user specified the type it will check if the ife is correctly configured to react on it. If it's not specified the default IFE type should be used. Signed-off-by: Alexander Aring <[email protected]> Acked-by: Jamal Hadi Salim <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2017-08-29Merge tag 'perf-core-for-mingo-4.14-20170829' of ↵Ingo Molnar22-43/+185
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: - Fix remote HITM detection for Skylake in 'perf c2c' (Jiri Olsa) - Fixes for the handling of PERF_RECORD_READ records (Jiri Olsa) - Fix kprobes blackist symbol lookup in 'perf probe' (Li Bin) - The PLT header and entry sizes are not the same in !x86, fix it for ARM and AARCH64 (Li Bin) - Beautify pkey_{alloc,free,mprotect} arguments in 'perf trace' (Arnaldo Carvalho de Melo) - Fix CC, AR, LD external definition, allow flex and bison to be externally defined and other related Makefile fixes (David Carrillo-Cisneros) - Sync CPU features kernel ABI headers with tooling headers (Arnaldo Carvalho de Melo) - Fix path to PMU formats in 'perf stat' documentation (Jack Henschel) - Fix static build with newer toolchains (Jiri Olsa) Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2017-08-29tools: PCI: Add a missing option help lineStan Drozd1-0/+1
Add a missing option help line for performing legacy interrupt test. Signed-off-by: Stan Drozd <[email protected]> Signed-off-by: Kishon Vijay Abraham I <[email protected]> Signed-off-by: Bjorn Helgaas <[email protected]>
2017-08-29selftests: add addrlabel add/delete to rtnetlink.shFlorian Westphal1-0/+41
Signed-off-by: Florian Westphal <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2017-08-29perf symbols: Fix plt entry calculation for ARM and AARCH64Li Bin1-5/+22
On x86, the plt header size is as same as the plt entry size, and can be identified from shdr's sh_entsize of the plt. But we can't assume that the sh_entsize of the plt shdr is always the plt entry size in all architecture, and the plt header size may be not as same as the plt entry size in some architecure. On ARM, the plt header size is 20 bytes and the plt entry size is 12 bytes (don't consider the FOUR_WORD_PLT case) that refer to the binutils implementation. The plt section is as follows: Disassembly of section .plt: 000004a0 <__cxa_finalize@plt-0x14>: 4a0: e52de004 push {lr} ; (str lr, [sp, #-4]!) 4a4: e59fe004 ldr lr, [pc, #4] ; 4b0 <_init+0x1c> 4a8: e08fe00e add lr, pc, lr 4ac: e5bef008 ldr pc, [lr, #8]! 4b0: 00008424 .word 0x00008424 000004b4 <__cxa_finalize@plt>: 4b4: e28fc600 add ip, pc, #0, 12 4b8: e28cca08 add ip, ip, #8, 20 ; 0x8000 4bc: e5bcf424 ldr pc, [ip, #1060]! ; 0x424 000004c0 <printf@plt>: 4c0: e28fc600 add ip, pc, #0, 12 4c4: e28cca08 add ip, ip, #8, 20 ; 0x8000 4c8: e5bcf41c ldr pc, [ip, #1052]! ; 0x41c On AARCH64, the plt header size is 32 bytes and the plt entry size is 16 bytes. The plt section is as follows: Disassembly of section .plt: 0000000000000560 <__cxa_finalize@plt-0x20>: 560: a9bf7bf0 stp x16, x30, [sp,#-16]! 564: 90000090 adrp x16, 10000 <__FRAME_END__+0xf8a8> 568: f944be11 ldr x17, [x16,#2424] 56c: 9125e210 add x16, x16, #0x978 570: d61f0220 br x17 574: d503201f nop 578: d503201f nop 57c: d503201f nop 0000000000000580 <__cxa_finalize@plt>: 580: 90000090 adrp x16, 10000 <__FRAME_END__+0xf8a8> 584: f944c211 ldr x17, [x16,#2432] 588: 91260210 add x16, x16, #0x980 58c: d61f0220 br x17 0000000000000590 <__gmon_start__@plt>: 590: 90000090 adrp x16, 10000 <__FRAME_END__+0xf8a8> 594: f944c611 ldr x17, [x16,#2440] 598: 91262210 add x16, x16, #0x988 59c: d61f0220 br x17 NOTES: In addition to ARM and AARCH64, other architectures, such as s390/alpha/mips/parisc/poperpc/sh/sparc/xtensa also need to consider this issue. Signed-off-by: Li Bin <[email protected]> Acked-by: Namhyung Kim <[email protected]> Cc: Alexander Shishkin <[email protected]> Cc: Alexis Berlemont <[email protected]> Cc: David Tolnay <[email protected]> Cc: Hanjun Guo <[email protected]> Cc: Hemant Kumar <[email protected]> Cc: Masami Hiramatsu <[email protected]> Cc: Milian Wolff <[email protected]> Cc: Namhyung Kim <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Wang Nan <[email protected]> Cc: [email protected] Link: http://lkml.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2017-08-29perf probe: Fix kprobe blacklist checking conditionLi Bin1-1/+1
The commit 9aaf5a5f479b ("perf probe: Check kprobes blacklist when adding new events"), 'perf probe' supports checking the blacklist of the fuctions which can not be probed. But the checking condition is wrong, that the end_addr of the symbol which is the start_addr of the next symbol can't be included. Committer notes: IOW make it match its kernel counterpart in kernel/kprobes.c: bool within_kprobe_blacklist(unsigned long addr) Each entry have as its end address not its end address, but the first address _outside_ that symbol, which for related functions, is the first address of the next symbol, like these from kernel/trace/trace_probe.c: 0xffffffffbd198df0-0xffffffffbd198e40 print_type_u8 0xffffffffbd198e40-0xffffffffbd198e90 print_type_u16 0xffffffffbd198e90-0xffffffffbd198ee0 print_type_u32 0xffffffffbd198ee0-0xffffffffbd198f30 print_type_u64 0xffffffffbd198f30-0xffffffffbd198f80 print_type_s8 0xffffffffbd198f80-0xffffffffbd198fd0 print_type_s16 0xffffffffbd198fd0-0xffffffffbd199020 print_type_s32 0xffffffffbd199020-0xffffffffbd199070 print_type_s64 0xffffffffbd199070-0xffffffffbd1990c0 print_type_x8 0xffffffffbd1990c0-0xffffffffbd199110 print_type_x16 0xffffffffbd199110-0xffffffffbd199160 print_type_x32 0xffffffffbd199160-0xffffffffbd1991b0 print_type_x64 But not always: 0xffffffffbd1997b0-0xffffffffbd1997c0 fetch_kernel_stack_address (kernel/trace/trace_probe.c) 0xffffffffbd1c57f0-0xffffffffbd1c58b0 __context_tracking_enter (kernel/context_tracking.c) Signed-off-by: Li Bin <[email protected]> Cc: Masami Hiramatsu <[email protected]> Cc: Namhyung Kim <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: Wang Nan <[email protected]> Cc: [email protected] Fixes: 9aaf5a5f479b ("perf probe: Check kprobes blacklist when adding new events") Link: http://lkml.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2017-08-29Merge branch 'perf/urgent' into perf/core, to pick up fixesIngo Molnar2-1/+29
Signed-off-by: Ingo Molnar <[email protected]>
2017-08-28selftests/bpf: check the instruction dumps are populatedJakub Kicinski1-4/+12
Add a basic test for checking whether kernel is populating the jited and xlated BPF images. It was used to confirm the behaviour change from commit d777b2ddbecf ("bpf: don't zero out the info struct in bpf_obj_get_info_by_fd()"), which made bpf_obj_get_info_by_fd() usable for retrieving the image dumps. Signed-off-by: Jakub Kicinski <[email protected]> Acked-by: Daniel Borkmann <[email protected]> Acked-by: Martin KaFai Lau <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2017-08-28perf trace beauty: Beautify pkey_{alloc,free,mprotect} argumentsArnaldo Carvalho de Melo6-1/+83
Reuse 'mprotect' beautifiers for 'pkey_mprotect'. System wide tracing pkey_alloc, pkey_free and pkey_mprotect calls, with backtraces: # perf trace -e pkey_alloc,pkey_mprotect,pkey_free --max-stack=5 0.000 ( 0.011 ms): pkey/7818 pkey_alloc(init_val: DISABLE_ACCESS|DISABLE_WRITE) = -1 EINVAL Invalid argument syscall (/usr/lib64/libc-2.25.so) pkey_alloc (/home/acme/c/pkey) 0.022 ( 0.003 ms): pkey/7818 pkey_mprotect(start: 0x7f28c3890000, len: 4096, prot: READ|WRITE, pkey: -1) = 0 syscall (/usr/lib64/libc-2.25.so) pkey_mprotect (/home/acme/c/pkey) 0.030 ( 0.002 ms): pkey/7818 pkey_free(pkey: -1 ) = -1 EINVAL Invalid argument syscall (/usr/lib64/libc-2.25.so) pkey_free (/home/acme/c/pkey) The tools/include/uapi/asm-generic/mman-common.h file is used to find the access rights defines for the pkey_alloc syscall second argument. Since we have the detector of changes for the tools/include header files versus its kernel origin (include/uapi/asm-generic/mman-common.h), we'll get whatever new flag appears for that argument automatically. This method should be used in other cases where it is easy to generate those flags tables because the header has properly namespaced defines like PKEY_DISABLE_ACCESS and PKEY_DISABLE_WRITE. Cc: Adrian Hunter <[email protected]> Cc: Arnd Bergmann <[email protected]> Cc: Dave Hansen <[email protected]> Cc: David Ahern <[email protected]> Cc: Jiri Olsa <[email protected]> Cc: Namhyung Kim <[email protected]> Cc: Wang Nan <[email protected]> Link: http://lkml.kernel.org/n/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2017-08-28tools headers: Sync cpu features kernel ABI headers with tooling headersArnaldo Carvalho de Melo2-6/+3
These changes made the tools/arch/x86/include/ headers to drift from its kernel origins: 910448bbed06 ("perf/x86/amd/uncore: Rename cpufeatures macro for cache counters") 5442c2699552 ("x86/cpufeature, kvm/svm: Rename (shorten) the new "virtualized VMSAVE/VMLOAD" CPUID flag") cba4671af755 ("x86/mm: Disable PCID on 32-bit kernels") Which was detected while building perf: make: Entering directory '/home/acme/git/linux/tools/perf' BUILD: Doing 'make -j4' parallel build Warning: Kernel ABI header at 'tools/arch/x86/include/asm/disabled-features.h' differs from latest version at 'arch/x86/include/asm/disabled-features.h' Warning: Kernel ABI header at 'tools/arch/x86/include/asm/cpufeatures.h' differs from latest version at 'arch/x86/include/asm/cpufeatures.h' This sync causes just these perf object files to be rebuilt: CC /tmp/build/perf/bench/mem-memcpy-x86-64-asm.o CC /tmp/build/perf/bench/mem-memset-x86-64-asm.o And the changes in the above changesets don't entail any need for change in the above 'perf bench' files. Cc: Adrian Hunter <[email protected]> Cc: Andy Lutomirski <[email protected]> Cc: Borislav Petkov <[email protected]> Cc: David Ahern <[email protected]> Cc: Janakarajan Natarajan <[email protected]> Cc: Jiri Olsa <[email protected]> Cc: Namhyung Kim <[email protected]> Cc: Wang Nan <[email protected]> Link: http://lkml.kernel.org/n/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2017-08-28perf tools: Pass full path of FEATURES_DUMPDavid Carrillo-Cisneros1-1/+1
When building with an external FEATURES_DUMP, bpf complains that features dump file is not found. Fix it by passing full file path. Signed-off-by: David Carrillo-Cisneros <[email protected]> Acked-by: Jiri Olsa <[email protected]> Cc: Alexander Shishkin <[email protected]> Cc: Paul Turner <[email protected]> Cc: Stephane Eranian <[email protected]> Link: http://lkml.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2017-08-28perf tools: Robustify detection of clang binaryDavid Carrillo-Cisneros5-6/+8
Prior to this patch, make scripts tested for CLANG with ifeq ($(CC), clang), failing to detect CLANG binaries with different names. Fix it by testing for the existence of __clang__ macro in the list of compiler defined macros. Signed-off-by: David Carrillo-Cisneros <[email protected]> Acked-by: Jiri Olsa <[email protected]> Cc: Alexander Shishkin <[email protected]> Cc: Paul Turner <[email protected]> Cc: Stephane Eranian <[email protected]> Link: http://lkml.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2017-08-28tools lib: Allow external definition of CC, AR and LDDavid Carrillo-Cisneros1-3/+3
Use already defined values for CC, AR and LD when available. Signed-off-by: David Carrillo-Cisneros <[email protected]> Acked-by: Jiri Olsa <[email protected]> Cc: Alexander Shishkin <[email protected]> Cc: Paul Turner <[email protected]> Cc: Stephane Eranian <[email protected]> Link: http://lkml.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2017-08-28perf tools: Allow external definition of flex and bison binary namesDavid Carrillo-Cisneros1-2/+2
Allow user to define flex and bison binary names by passing FLEX and BISON variables. Signed-off-by: David Carrillo-Cisneros <[email protected]> Acked-by: Jiri Olsa <[email protected]> Cc: Alexander Shishkin <[email protected]> Cc: Paul Turner <[email protected]> Cc: Stephane Eranian <[email protected]> Link: http://lkml.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2017-08-28tools build tests: Don't hardcode gcc nameDavid Carrillo-Cisneros1-1/+1
Use $(CC) instead of harcoded gcc binary name. Signed-off-by: David Carrillo-Cisneros <[email protected]> Acked-by: Jiri Olsa <[email protected]> Cc: Alexander Shishkin <[email protected]> Cc: Paul Turner <[email protected]> Cc: Stephane Eranian <[email protected]> Link: http://lkml.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2017-08-28perf report: Group stat values on global event idJiri Olsa2-2/+2
There's no big value on displaying counts for every event ID, which is one per every CPU. Rather than that, displaying the whole sum for the event. $ perf record -c 100000 -e cycles:u -s test $ perf report -T Before: # PID TID cycles:u cycles:u cycles:u cycles:u ... [20 more columns of 'cycles:u'] 3339 3339 0 0 0 0 3340 3340 0 0 0 0 3341 3341 0 0 0 0 3342 3342 0 0 0 0 Now: # PID TID cycles:u 3339 3339 19678 3340 3340 18744 3341 3341 17335 3342 3342 26414 Signed-off-by: Jiri Olsa <[email protected]> Cc: Alexander Shishkin <[email protected]> Cc: Andi Kleen <[email protected]> Cc: David Ahern <[email protected]> Cc: Mark Rutland <[email protected]> Cc: Namhyung Kim <[email protected]> Cc: Peter Zijlstra <[email protected]> Link: http://lkml.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2017-08-28perf values: Zero value buffersJiri Olsa1-2/+7
We need to make sure the array of value pointers are zero initialized, because we use them in realloc later on and uninitialized non zero value will cause allocation error and aborted execution. Signed-off-by: Jiri Olsa <[email protected]> Cc: Alexander Shishkin <[email protected]> Cc: Andi Kleen <[email protected]> Cc: David Ahern <[email protected]> Cc: Mark Rutland <[email protected]> Cc: Namhyung Kim <[email protected]> Cc: Peter Zijlstra <[email protected]> Link: http://lkml.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2017-08-28perf values: Fix allocation checkJiri Olsa1-1/+1
Bailing out in case the allocation failed, not the other way round. Signed-off-by: Jiri Olsa <[email protected]> Cc: Alexander Shishkin <[email protected]> Cc: Andi Kleen <[email protected]> Cc: David Ahern <[email protected]> Cc: Mark Rutland <[email protected]> Cc: Namhyung Kim <[email protected]> Cc: Peter Zijlstra <[email protected]> Link: http://lkml.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2017-08-28perf values: Fix thread index bugJiri Olsa1-2/+2
We are taking wrong index (+1) for first thread, which leaves thread with index 0 unused and uninitialized. Signed-off-by: Jiri Olsa <[email protected]> Cc: Alexander Shishkin <[email protected]> Cc: Andi Kleen <[email protected]> Cc: David Ahern <[email protected]> Cc: Mark Rutland <[email protected]> Cc: Namhyung Kim <[email protected]> Cc: Peter Zijlstra <[email protected]> Link: http://lkml.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2017-08-28perf report: Add dump_read functionJiri Olsa2-4/+25
Adding dump_read function to gather all the dump output of read function. Adding output of enabled and running times and id if enabled (3 new lines with '...' prefix below). $ perf record -s ... $ perf report -D 958358311769 0x91f8 [0x40]: PERF_RECORD_READ: 3339 3339 cycles:u 0 ... time enabled : 958358313731 ... time running : 958358313731 ... id : 80 Committer note: Do not use 'read' as a variable name as it breaks the build on older systems, such as RHEL6: CC /tmp/build/perf/util/session.o cc1: warnings being treated as errors util/session.c: In function 'dump_read': util/session.c:1132: error: declaration of 'read' shadows a global declaration /usr/include/bits/unistd.h:35: error: shadowed declaration is here mv: cannot stat `/tmp/build/perf/util/.session.o.tmp': No such file or directory Signed-off-by: Jiri Olsa <[email protected]> Cc: Alexander Shishkin <[email protected]> Cc: Andi Kleen <[email protected]> Cc: David Ahern <[email protected]> Cc: Mark Rutland <[email protected]> Cc: Namhyung Kim <[email protected]> Cc: Peter Zijlstra <[email protected]> Link: http://lkml.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2017-08-28bpf: test_maps add sockmap stress testJohn Fastabend1-1/+28
Sockmap is a bit different than normal stress tests that can run in parallel as is. We need to reuse the same socket pool and map pool to get good stress test cases. Signed-off-by: John Fastabend <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2017-08-28bpf: harden sockmap program attach to ensure correct map typeJohn Fastabend2-4/+30
When attaching a program to sockmap we need to check map type is correct. Fixes: 174a79ff9515 ("bpf: sockmap with sk redirect support") Signed-off-by: John Fastabend <[email protected]> Acked-by: Alexei Starovoitov <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2017-08-28bpf: more SK_SKB selftestsJohn Fastabend1-0/+98
Tests packet read/writes and additional skb fields. Signed-off-by: John Fastabend <[email protected]> Acked-by: Alexei Starovoitov <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2017-08-28bpf: additional sockmap self testsJohn Fastabend3-46/+96
Add some more sockmap tests to cover, - forwarding to NULL entries - more than two maps to test list ops - forwarding to different map Signed-off-by: John Fastabend <[email protected]> Acked-by: Alexei Starovoitov <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2017-08-28bpf: convert sockmap field attach_bpf_fd2 to typeJohn Fastabend7-99/+68
In the initial sockmap API we provided strparser and verdict programs using a single attach command by extending the attach API with a the attach_bpf_fd2 field. However, if we add other programs in the future we will be adding a field for every new possible type, attach_bpf_fd(3,4,..). This seems a bit clumsy for an API. So lets push the programs using two new type fields. BPF_SK_SKB_STREAM_PARSER BPF_SK_SKB_STREAM_VERDICT This has the advantage of having a readable name and can easily be extended in the future. Updates to samples and sockmap included here also generalize tests slightly to support upcoming patch for multiple map support. Signed-off-by: John Fastabend <[email protected]> Fixes: 174a79ff9515 ("bpf: sockmap with sk redirect support") Suggested-by: Alexei Starovoitov <[email protected]> Acked-by: Alexei Starovoitov <[email protected]> Signed-off-by: David S. Miller <[email protected]>
2017-08-28perf: cs-etm: Fix ETMv4 CONFIGR entry in perf.data fileMike Leach2-1/+32
The value passed into the perf.data file for the CONFIGR register in ETMv4 was incorrectly being set to the command line options/ETMv3 value. Adds bit definitions and function to remap this value to the correct ETMv4 CONFIGR bit values for all selected options. Signed-off-by: Mike Leach <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2017-08-28coresight: pmu: Adds return stack option to perf coresight pmuMike Leach1-0/+1
Return stack is a programmable option on some ETM and PTM hardware. Adds the option flags to enable this from the perf event command line. Signed-off-by: Mike Leach <[email protected]> Signed-off-by: Mathieu Poirier <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]>
2017-08-28perf record: Set read_format for inherit_statJiri Olsa1-1/+6
Set read_format for what we expect to get from read event generated by perf_event_attr::inherit_stat. Signed-off-by: Jiri Olsa <[email protected]> Cc: Alexander Shishkin <[email protected]> Cc: Andi Kleen <[email protected]> Cc: David Ahern <[email protected]> Cc: Mark Rutland <[email protected]> Cc: Namhyung Kim <[email protected]> Cc: Peter Zijlstra <[email protected]> Link: http://lkml.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2017-08-28perf c2c: Fix remote HITM detection for SkylakeJiri Olsa1-2/+9
Skylake introduced new mem_remote bit in union perf_mem_data_src [1]. It applies to any other memory level to express Remote unknown level, as is reported by Skylake. Adding this extra check to c2c_decode_stats to properly decode remote HITMs on Skylake. [1] http://lkml.kernel.org/r/[email protected] Signed-off-by: Jiri Olsa <[email protected]> Acked-by: Andi Kleen <[email protected]> Cc: Alexander Shishkin <[email protected]> Cc: David Ahern <[email protected]> Cc: Joe Mario <[email protected]> Cc: Namhyung Kim <[email protected]> Cc: Peter Zijlstra <[email protected]> Link: http://lkml.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2017-08-28perf tools: Fix static build with newer toolchainsJiri Olsa1-1/+7
We can't pass --dynamic-list list into static build anymore, because compilers starts to scream about that. Fedora 26 started to fail build with following error: $ make LDFLAGS=-static ... /usr/bin/ld: dynamic STT_GNU_IFUNC symbol `strcmp' with pointer equality in `/usr/lib/gcc/x86_64-redhat-linux/7/../../../../lib64/libc.a(strcmp.o +)' can not be used when making an executable; recompile with -fPIE and relink with -pie There's no sense for --dynamic-list in static build, because there's no .dynsym table in static binary. Consequently the traceevent plugins have never worked with static build, but it was quietly passed by. To fix this in future I think we should add support to compile plugins within the perf binary directly for static build. Reported-and-Tested-by: Arnaldo Carvalho de Melo <[email protected]> Signed-off-by: Jiri Olsa <[email protected]> Link: http://lkml.kernel.org/n/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2017-08-28perf stat: Fix path to PMU formats in documentationJack Henschel1-2/+2
As defined in tools/perf/util/pmu.c, the EVENT_SOURCE_DEVICE_PATH is /sys/bus/event_source/devices/ (no traling 's' in event_source) This patch corrects the path in the perf stat documentation Signed-off-by: Jack Henschel <[email protected]> Cc: Alexander Shishkin <[email protected]> Cc: Jack Henschel <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: [email protected] Link: http://lkml.kernel.org/r/[email protected] Signed-off-by: Arnaldo Carvalho de Melo <[email protected]>
2017-08-28Merge 4.13-rc7 into staging-nextGreg Kroah-Hartman7-9/+37
We want the staging and iio fixes in here to handle the merge issues. Signed-off-by: Greg Kroah-Hartman <[email protected]>
2017-08-28Merge 4.13-rc7 into char-misc-nextGreg Kroah-Hartman7-9/+37
We want the binder fix in here as well for testing and merge issues. Signed-off-by: Greg Kroah-Hartman <[email protected]>
2017-08-27Merge tag 'ntb-4.13-bugfixes' of git://github.com/jonmason/ntbLinus Torvalds1-0/+4
Pull NTB fixes from Jon Mason: "NTB bug fixes to address an incorrect ntb_mw_count reference in the NTB transport, improperly bringing down the link if SPADs are corrupted, and an out-of-order issue regarding link negotiation and data passing" * tag 'ntb-4.13-bugfixes' of git://github.com/jonmason/ntb: ntb: ntb_test: ensure the link is up before trying to configure the mws ntb: transport shouldn't disable link due to bogus values in SPADs ntb: use correct mw_count function in ntb_tool and ntb_transport
2017-08-26Merge branch 'x86-urgent-for-linus' of ↵Linus Torvalds1-1/+25
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip Pull x86 fixes from Ingo Molnar: "Two fixes: one for an ldt_struct handling bug and a cherry-picked objtool fix" * 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: x86/mm: Fix use-after-free of ldt_struct objtool: Fix '-mtune=atom' decoding support in objtool 2.0
2017-08-25selftests: lib.mk: change RUN_TESTS to print messages in TAP13 formatShuah Khan1-3/+9
Change common RUN_TESTS to print messages in user friendly TAP13 format. This change add TAP13 header at the start of RUN_TESTS target run, and prints the resulting pass/fail messages with test number information in the TAP 13 format for each test in the run tests list. This change covers test scripts as well as test programs. Test programs have an option to use ksft_ API, however test scripts won't be able to. With this change, test scripts can print TAP13 format output without any changes to individual scripts. Test programs can provide TAP13 format output as needed as some tests already do. Tests that haven't been converted will benefit from this change. Tests that are converted benefit from the test counts for all the tests in each test directory. Running firmware tests: make --silent -C tools/testing/selftests/firmware/ run_tests Before the change: modprobe: ERROR: could not insert 'test_firmware': Operation not permitted ./fw_filesystem.sh: /sys/devices/virtual/misc/test_firmware not present You must have the following enabled in your kernel: CONFIG_TEST_FIRMWARE=y selftests: fw_filesystem.sh [FAIL] modprobe: ERROR: could not insert 'test_firmware': Operation not permitted selftests: fw_fallback.sh [FAIL] After the change: TAP version 13 selftests: fw_filesystem.sh ======================================== modprobe: ERROR: could not insert 'test_firmware': Operation not permitted ./fw_filesystem.sh: /sys/devices/virtual/misc/test_firmware not present You must have the following enabled in your kernel: CONFIG_TEST_FIRMWARE=y not ok 1..1 selftests: fw_filesystem.sh [FAIL] selftests: fw_fallback.sh ======================================== modprobe: ERROR: could not insert 'test_firmware': Operation not permitted not ok 1..2 selftests: fw_fallback.sh [FAIL] Signed-off-by: Shuah Khan <[email protected]>
2017-08-25selftests: change lib.mk RUN_TESTS to take test list as an argumentShuah Khan2-15/+3
Change lib.mk RUN_TESTS to take test list as an argument. This will allow it to be called from individual test makefiles to run additional tests that aren't suitable for a default kselftest run. As an example, timers test includes destructive tests that aren't included in the common run_tests target. Change times/Makefile to use RUN_TESTS call with destructive test list as an argument instead of using its own RUN_TESTS target. Signed-off-by: Shuah Khan <[email protected]>
2017-08-25selftests: lib.mk: suppress "cd" output from run_tests targetShuah Khan1-1/+1
Suppress "cd" output from run_tests while running tests to declutter the test results. Running efivarfs test: make --silent -C tools/testing/selftests/efivarfs/ run_tests Before the change: skip all tests: must be run as root selftests: efivarfs.sh [PASS] /lkml/linux-kselftest/tools/testing/selftests/efivarfs After the change: skip all tests: must be run as root selftests: efivarfs.sh [PASS] Signed-off-by: Shuah Khan <[email protected]>
2017-08-25selftests: kselftest framework: change skip exit code to 0Shuah Khan1-1/+2
When a test is skipped, instead of using a special exit code of 4, treat it as pass condition and use exit code of 0. It makes sense to treat skip as pass since the test couldn't be run as opposed to a failed test. Signed-off-by: Shuah Khan <[email protected]>
2017-08-25selftests/timers: make loop consistent with array sizeBenjamin Gaignard1-1/+1
clocksource_list array is defined as char [10][30] so to initialise it we only have to iterate 10 times. Signed-off-by: Benjamin Gaignard <[email protected]> Signed-off-by: Shuah Khan <[email protected]>
2017-08-24Merge tag 'perf-core-for-mingo-4.14-20170823' of ↵Ingo Molnar32-64/+6595
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: - Expression parser enhancements for metrics (Andi Kleen) - Fix buffer overflow while freeing events in 'perf stat' (Andi Kleen) - Fix static linking with elfutils's libdf and with libunwind in Debian/Ubuntu (Konstantin Khlebnikov) - Tighten detection of BPF events, avoiding matching some other PMU events such as 'cpu/uops_executed.core,cmask=1/' as a .c source file that ended up being considered a BPF event (Andi Kleen) - Add Skylake server uncore JSON vendor events (Andi Kleen) - Add support for printing new mem_info encodings, including 'perf test' checks (Andi Kleen) - Really install manpages via 'make install-man' (Konstantin Khlebnikov) - Fix documentation for perf_event_paranoid and perf_event_mlock_kb sysctls (Konstantin Khlebnikov) Signed-off-by: Arnaldo Carvalho de Melo <[email protected]> Signed-off-by: Ingo Molnar <[email protected]>
2017-08-24Merge branch 'linus' into perf/core, to pick up fixesIngo Molnar2-3/+4
Signed-off-by: Ingo Molnar <[email protected]>
2017-08-24x86/lguest: Remove lguest supportJuergen Gross6-3625/+5
Lguest seems to be rather unused these days. It has seen only patches ensuring it still builds the last two years and its official state is "Odd Fixes". Remove it in order to be able to clean up the paravirt code. Signed-off-by: Juergen Gross <[email protected]> Acked-by: Rusty Russell <[email protected]> Acked-by: Thomas Gleixner <[email protected]> Cc: Linus Torvalds <[email protected]> Cc: Peter Zijlstra <[email protected]> Cc: [email protected] Cc: [email protected] Cc: [email protected] Cc: [email protected] Link: http://lkml.kernel.org/r/[email protected] Signed-off-by: Ingo Molnar <[email protected]>
2017-08-23selftests/bpf: add a test for a pruning bug in the verifierAlexei Starovoitov1-0/+28
The test makes a read through a map value pointer, then considers pruning a branch where the register holds an adjusted map value pointer. It should not prune, but currently it does. Signed-off-by: Alexei Starovoitov <[email protected]> [[email protected]: added test-name and patch description] Signed-off-by: Edward Cree <[email protected]> Acked-by: Daniel Borkmann <[email protected]> Signed-off-by: David S. Miller <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.