uclibc test suite with gnu 2017.03 (was Re: [Fwd: Re: nsim latest free version])
A message that is not in the archive (sent off-list or not yet received): [email protected]
A message that is not in the archive (sent off-list or not yet received): [email protected]
Hi Waldemar,
Quoted 19 lines
Quoted 10 lines
Hi Waldemar,
Quoted 5 lines · Waldemar Brodkorb
Hi Alexey,
I have some issues lately running the testsuite in NSIM. Is there may be some new version available?
How I can get an updated version?Let me check with our nSIM devs.
But could you please elaborate a bit what is your problem?It stalls / deadlocks here:
PASS tst-cancel23
FAIL tst-cancel25 got 1 expected 0
Timed out: killed the child process
PASS tst-cancel3
FAIL tst-cancel4 got 1 expected 0
Timed out: killed the child process
.... tst-cancel5
Quoted 6 lines
Seems to happen since the toolchain update to 2017.03
(binutils/gcc). ARCv2 little endian started with the old script:
https://urldefense.proofpoint.com/v2/url?u=https-3A__cgit.openadk.org_cgi_cgit_openadk.git_tree_scripts_nsim.sh&d=DwIBaQ&c=DPL6_X_6JkXFx7AXWqB0tg&r=lq
deeSSEes0GFDDl656eViXO7breS55ytWkhpk5R81I&m=8GRVuMbTElbNMAj_Fn7DTePw5-sVRsJ3lki2x-ofbzo&s=l2eNCzqvmvtPr63NgA0Uv-t7ze5J0eJeAq9ZcsDgu_Y&e=With attached uClibc-ng.config.
Can you please provide the full boot log if possible - I could successfully run the "old" test suite with 2017.03 !
Also I presume there's no real evidence to attribute this is an nSIM issue - it could very well be a Linux kernel or tools issue.
Can you share your vmlinux which we can try on nsim here !
Thx,
-Vineet
Hi Vineet,
Vineet Gupta wrote,
Quoted 26 lines
Hi Waldemar,
Quoted 24 lines
Quoted 10 lines
Hi Waldemar,
Quoted 5 lines · Waldemar Brodkorb
Hi Alexey,
I have some issues lately running the testsuite in NSIM. Is there may be some new version available?
How I can get an updated version?Let me check with our nSIM devs.
But could you please elaborate a bit what is your problem?It stalls / deadlocks here:
PASS tst-cancel23
FAIL tst-cancel25 got 1 expected 0
Timed out: killed the child process
PASS tst-cancel3
FAIL tst-cancel4 got 1 expected 0
Timed out: killed the child process
.... tst-cancel5Seems to happen since the toolchain update to 2017.03
(binutils/gcc). ARCv2 little endian started with the old script:
https://urldefense.proofpoint.com/v2/url?u=https-3A__cgit.openadk.org_cgi_cgit_openadk.git_tree_scripts_nsim.sh&d=DwIBaQ&c=DPL6_X_6JkXFx7AXWqB0tg&r=lq
deeSSEes0GFDDl656eViXO7breS55ytWkhpk5R81I&m=8GRVuMbTElbNMAj_Fn7DTePw5-sVRsJ3lki2x-ofbzo&s=l2eNCzqvmvtPr63NgA0Uv-t7ze5J0eJeAq9ZcsDgu_Y&e=
Quoted 8 lines
Quoted 1 line
With attached uClibc-ng.config.
Can you please provide the full boot log if possible - I could successfully run the "old" test suite with 2017.03 !
Also I presume there's no real evidence to attribute this is an nSIM issue - it could very well be a Linux kernel or tools issue.
Can you share your vmlinux which we can try on nsim here !
Kernel+initramfs can be downloaded here:
https://debug.openadk.org/arc/nsim-arcv2-initramfspiggyback-kernel
Last lines of strace:
[pid 228] setrlimit(RLIMIT_CORE, {rlim_cur=0, rlim_max=0}) = 0
[pid 228] setpgid(0, 0) = 0
[pid 228] socketpair(AF_UNIX, SOCK_STREAM, 1, [3, 4]) = 0
[pid 228] setsockopt(4, SOL_SOCKET, SO_SNDBUF, [1], 4) = 0
[pid 228] getsockopt(4, SOL_SOCKET, SO_SNDBUF, [4480], [4]) = 0
[pid 228] setsockopt(4, SOL_SOCKET, SO_SNDBUF, [4480], 4) = 0
[pid 228] mmap2(NULL, 2097152, PROT_READ|PROT_WRITE,MAP_PRIVATE|MAP_ANONYMOUS|MAP_STACK, -1, 0) = 0x20082000
[pid 228] mprotect(0x20082000, 8192, PROT_NONE) = 0
[pid 228] clone(child_stack=0x202810e8,
flags=CLONE_VM|CLONE_FS|CLONE_FILES|CLONE_SIGHAND|CLONE_THREAD|CLONE_SYSVSEM|CLONE_SETTLS|CLONE_PARENT_SETTID|CLONE_CHILD_CLEARTID, parent_tidptr=0x20281590, tls=0x20281968, child_tidptr=0x20281590) =
229
strace: Process 229 attached
[pid 229] set_robust_list(0x20281594, 12) = 0 [pid 229] futex(0x181e8, FUTEX_WAKE_PRIVATE, 2147483647) = 0 [pid 229] read(3, <unfinished ...> [pid 228] futex(0x181e8, FUTEX_WAIT_PRIVATE, 0, NULL) = -1 EAGAIN
(Resource temporarily unavailable)
[pid 228] nanosleep({tv_sec=0, tv_nsec=100000000}, 0x5fa27d9c) = 0
[pid 228] openat(AT_FDCWD, "/usr/lib/libgcc_s.so.1", O_RDONLY) =
-1 ENOENT (No such file or directory)
[pid 228] openat(AT_FDCWD, "/usr/lib/libgcc_s.so.1", O_RDONLY) =
-1 ENOENT (No such file or directory)
[pid 228] openat(AT_FDCWD, "/lib//libgcc_s.so.1", O_RDONLY) = 5
[pid 228] fstat64(5, {st_mode=S_IFREG|0644, st_size=50740, ...}) =
0
[pid 228] mmap2(NULL, 8192, PROT_READ|PROT_WRITE,
MAP_PRIVATE|MAP_ANONYMOUS|MAP_UNINITIALIZED, -1, 0) = 0x2000c000
[pid 228] read(5,
"\177ELF\1\1\1\0\0\0\0\0\0\0\0\0\3\0\303\0\1\0\0\0@ \0\0004\0\0\0"..., 8192) = 8192
[pid 228] mmap2(NULL, 65536, PROT_NONE, MAP_PRIVATE|MAP_ANONYMOUS,
-1, 0) = 0x20282000
[pid 228] mmap2(0x20282000, 42156, PROT_READ|PROT_EXEC,
MAP_PRIVATE|MAP_FIXED, 5, 0) = 0x20282000
[pid 228] mmap2(0x2028e000, 8520, PROT_READ|PROT_WRITE,
MAP_PRIVATE|MAP_FIXED, 5, 0xa000) = 0x2028e000
[pid 228] close(5) = 0
[pid 228] munmap(0x2000c000, 8192) = 0
[pid 228] mprotect(0x2028e000, 8192, PROT_READ) = 0
[pid 228] getpid() = 228
[pid 228] tgkill(228, 229, SIGRTMIN <unfinished ...>
[pid 229] <... read resumed> 0x20281014, 100) = ? ERESTARTSYS (To
be restarted if SA_RESTART is set)
[pid 229] --- SIGRTMIN {si_signo=SIGRTMIN, si_code=SI_TKILL,
si_pid=228, si_uid=0} ---
[pid 229] getpid() = 228
[pid 229] futex(0x20290168, FUTEX_WAKE_PRIVATE, 2147483647) = 0
[pid 228] <... tgkill resumed> ) = 0
[pid 228] futex(0x20281590, FUTEX_WAIT, 229, NULLSeems a deadlock, I can stop the test with ctrl-c.
I do not see this issue on other architectures. See https://tests.embedded-test.org/uClibc-ng/1.0.25/
Dmesg:
uname -a
Linux openadk 4.9.32-1 #4 Thu Jun 22 22:42:15 CEST 2017 arc GNU/Linux dmroot at openadk:/ # dmesg
[ 0.000000] Linux version 4.9.32-1 (wbx at vopenadk) (gcc version 6.3.0 (GCC) ) #4 Thu Jun 22 22:42:15 CEST 2017 [ 0.000000] Memory @ 80000000 [512M] [ 0.000000] Memory @ 100000000 [1024M] Not used [ 0.000000] OF: fdt:Machine model: snps,nsim_hs [ 0.000000] earlycon: arc_uart0 at MMIO32 0xc0fc1000 (options '115200n8') [ 0.000000] bootconsole [arc_uart0] enabled [ 0.000000] archs-intc : 15 priority levels (default 14) [ 0.000000] [ 0.000000] IDENTITY : ARCVER [0x51] ARCNUM [0x0] CHIPID [0xffff] [ 0.000000] processor [0] : ARC HS38 R2.0 (ARCv2 ISA) [ 0.000000] Timers : Timer0 Timer1 [ 0.000000] ISA Extn : atomic (not used) ll64 (not used) unalign (not used) [ 0.000000] : mpy[opt 9] div_rem norm barrel-shift swap minmax swape(not used) [ 0.000000] BPU : partial match, cache:2048, Predict Table:16384 [ 0.000000] MMU [v4] : 8k PAGE, 2M Super Page (not used) JTLB 512 (128x4), uDTLB 8, uITLB 4 [ 0.000000] I-Cache : 32K, 4way/set, 64B Line, VIPT(not used) [ 0.000000] D-Cache : 16K, 2way/set, 64B Line, PIPT(not used) [ 0.000000] Peripherals : 0xc0000000 [ 0.000000] Vector Table : 0x80000000 [ 0.000000] DEBUG : ActionPoint [ 0.000000] OS ABI [v4] : 64-bit data any register aligned [ 0.000000] On node 0 totalpages: 65536 [ 0.000000] free_area_init_node: node 0, pgdat 802d2f20, node_mem_map 9fdc0000 [ 0.000000] Normal zone: 288 pages used for memmap [ 0.000000] Normal zone: 0 pages reserved [ 0.000000] Normal zone: 65536 pages, LIFO batch:15 [ 0.000000] pcpu-alloc: s0 r0 d32768 u32768 alloc=1*32768 [ 0.000000] pcpu-alloc: [0] 0 [ 0.000000] Built 1 zonelists in Zone order, mobility grouping on. Total pages: 65248 [ 0.000000] Kernel command line: earlycon=arc_uart,mmio32,0xc0fc1000,115200n8 console=ttyARC0,115200n8 [ 0.000000] PID hash table entries: 2048 (order: 0, 8192 bytes) [ 0.000000] Dentry cache hash table entries: 65536 (order: 5, 262144 bytes) [ 0.000000] Inode-cache hash table entries: 32768 (order: 4, 131072 bytes) [ 0.000000] Memory: 518424K/524288K available (1176K kernel code, 79K rwdata, 240K rodata, 1400K init, 225K bss, 5864K reserved, 0K cma-reserved) [ 0.000000] SLUB: HWalign=64, Order=0-3, MinObjects=0, CPUs=1, Nodes=1 [ 0.000000] NR_IRQS:128 [ 0.000000] clocksource: ARC Timer1: mask: 0xffffffff max_cycles: 0xffffffff, max_idle_ns: 23890755578 ns [ 0.070000] Calibrating delay loop... 159.12 BogoMIPS (lpj=795648) [ 0.080000] pid_max: default: 4096 minimum: 301 [ 0.080000] Mount-cache hash table entries: 2048 (order: 0, 8192 bytes) [ 0.080000] Mountpoint-cache hash table entries: 2048 (order: 0, 8192 bytes) [ 0.080000] devtmpfs: initialized [ 0.090000] clocksource: jiffies: mask: 0xffffffff max_cycles: 0xffffffff, max_idle_ns: 19112604462750000 ns [ 0.090000] futex hash table entries: 16 (order: -6, 192 bytes) [ 0.090000] NET: Registered protocol family 16 [ 0.110000] clocksource: Switched to clocksource ARC Timer1 [ 0.110000] NET: Registered protocol family 2 [ 0.120000] TCP established hash table entries: 4096 (order: 1, 16384 bytes) [ 0.120000] TCP bind hash table entries: 4096 (order: 1, 16384 bytes) [ 0.120000] TCP: Hash tables configured (established 4096 bind 4096) [ 0.120000] UDP hash table entries: 512 (order: 0, 8192 bytes) [ 0.120000] UDP-Lite hash table entries: 512 (order: 0, 8192 bytes) [ 0.120000] NET: Registered protocol family 1 [ 2.550000] random: fast init done [ 22.880000] devtmpfs: mounted [ 22.910000] workingset: timestamp_bits=30 max_order=16 bucket_order=0 [ 23.530000] c0fc1000.serial: ttyARC0 at MMIO 0x0 (irq = 24, base_baud = 5000000) is a arc-uart [ 23.530000] console [ttyARC0] enabled [ 23.530000] bootconsole [arc_uart0] disabled [ 23.530000] NET: Registered protocol family 17 [ 23.600000] Starting Linux (built with OpenADK). [ 23.600000] Freeing unused kernel memory: 1400K (80002000 - 80160000) [ 23.600000] This architecture does not have kernel memory protection. [ 231.060000] random: crng init done
best regards
Waldemar
Hi,
Waldemar Brodkorb wrote,
Quoted 25 lines
Hi Vineet,
Vineet Gupta wrote,Quoted 22 lines
Hi Waldemar,
Quoted 20 lines
Quoted 10 lines
Hi Waldemar,
Quoted 5 lines · Waldemar Brodkorb
Hi Alexey,
I have some issues lately running the testsuite in NSIM. Is there may be some new version available?
How I can get an updated version?Let me check with our nSIM devs.
But could you please elaborate a bit what is your problem?It stalls / deadlocks here:
PASS tst-cancel23
FAIL tst-cancel25 got 1 expected 0
Timed out: killed the child process
PASS tst-cancel3
FAIL tst-cancel4 got 1 expected 0
Timed out: killed the child process
.... tst-cancel5Seems to happen since the toolchain update to 2017.03
Just checked with arc-2016.09-release and the test pass.
best regards
Waldemar
Hi Vineet,
did you find anything or should I disable the test for arc for now?
best regards
Waldemar
Hi Waldemar,
Quoted 3 lines · Waldemar Brodkorb
Hi Vineet,
did you find anything or should I disable the test for arc for now?
Eugeniy (in Cc) is looking into it so I guess it worth keeping the test for now.
-Alexey