[lldb-dev] Failing LIT-based lldb-mi tests
Александр Поляков via lldb-dev
lldb-dev at lists.llvm.org
Wed Aug 15 10:31:09 PDT 2018
I built an lldb myself. As far as I remember, this error has been seen in
swift's CI on machines running Linux. Adrian, could you correct me if I'm
wrong?
On Wed, Aug 15, 2018 at 8:02 PM <ted.woodward at codeaurora.org> wrote:
> That definitely points to an issue with either your install or your system.
>
>
>
> Is that an lldb you built, or one you installed with apt-get?
>
>
>
> Turning on the lldb-server logs might help. Unfortunately, I forget how to
> do that 😊. Pavel, do you remember?
>
>
>
> *From:* Александр Поляков <polyakov.alx at gmail.com>
> *Sent:* Tuesday, August 14, 2018 6:15 PM
> *To:* Ted Woodward <ted.woodward at codeaurora.org>
> *Cc:* Adrian Prantl <aprantl at apple.com>; LLDB <lldb-dev at lists.llvm.org>
> *Subject:* Re: [lldb-dev] Failing LIT-based lldb-mi tests
>
>
>
> LLDB fails with the same symptoms:
>
> lldb < 1> send packet: +
>
> lldb history[1] tid=0x0c09 < 1> send packet: +
>
> lldb < 19> send packet: $QStartNoAckMode#b0
>
> lldb < 1> read packet: +
>
> lldb < 6> read packet: $OK#9a
>
> lldb < 1> send packet: +
>
> lldb < 41> send packet:
> $qSupported:xmlRegisters=i386,arm,mips#12
>
> lldb < 124> read packet:
> $PacketSize=20000;QStartNoAckMode+;QThreadSuffixSupported+;QListThreadsInStopReply+;qEcho+;QPassSignals+;qXfer:auxv:read+#be
>
> lldb < 26> send packet: $QThreadSuffixSupported#e4
>
> lldb < 6> read packet: $OK#9a
>
> lldb < 27> send packet: $QListThreadsInStopReply#21
>
> lldb < 6> read packet: $OK#9a
>
> lldb < 13> send packet: $qHostInfo#9b
>
> lldb < 11> send packet: $qEcho:1#5b
>
> error: process launch failed: 'A' packet returned an error: -1
>
>
>
> `gdb-remote process` log is the same as for lldb-mi.
>
>
>
> On Wed, Aug 15, 2018 at 2:01 AM <ted.woodward at codeaurora.org> wrote:
>
> Yes, the communications back and forth indicate that lldb-server is
> running and lldb-mi is connected to it.
>
>
>
> My run was successful. You mentioned load on the machine; do you see any
> failed runs from lldb? Both lldb and lldb-mi make the same call
> (Target::Launch) to start the process.
>
>
>
> My log looks like this:
>
> lldb-mi < 1> send packet: +
>
> lldb-mi history[1] tid=0x33d0 < 1> send packet: +
>
> lldb-mi < 19> send packet: $QStartNoAckMode#b0
>
> lldb-mi < 1> read packet: +
>
> lldb-mi < 6> read packet: $OK#9a
>
> lldb-mi < 1> send packet: +
>
> lldb-mi < 41> send packet:
> $qSupported:xmlRegisters=i386,arm,mips#12
>
> lldb-mi < 124> read packet:
> $PacketSize=20000;QStartNoAckMode+;QThreadS
>
>
> uffixSupported+;QListThreadsInStopReply+;qEcho+;QPassSignals+;qXfer:auxv:read+#b
>
> e
>
> lldb-mi < 26> send packet: $QThreadSuffixSupported#e4
>
> lldb-mi < 6> read packet: $OK#9a
>
> lldb-mi < 27> send packet: $QListThreadsInStopReply#21
>
> lldb-mi < 6> read packet: $OK#9a
>
> lldb-mi < 13> send packet: $qHostInfo#9b
>
> lldb-mi < 363> read packet:
> $triple:7838365f36342d2d6c696e75782d676e75;ptrsize:8;distribution_id:7562756e7475;watchpoint_exceptions_received:after;endian:little;os_version:3.13.0;os_build:332e31332e302d3135332d67656e65726963;os_kernel:233230332d5562756e747520534d5020546875204a756e2031342030383a35323a3238205554432032303138;hostname:746564776f6f642d7562756e74752e7175616c636f6d6d2e636f6d;#73
>
> lldb-mi < 10> send packet: $vCont?#49
>
> lldb-mi < 17> read packet: $vCont;c;C;s;S#62
>
> lldb-mi < 27> send packet: $qVAttachOrWaitSupported#38
>
> lldb-mi < 4> read packet: $#00
>
> lldb-mi < 23> send packet: $QEnableErrorStrings#8c
>
> lldb-mi < 6> read packet: $OK#9a
>
> lldb-mi < 36> send packet: $QSetSTDIN:2f6465762f7074732f3235#4c
>
> lldb-mi < 6> read packet: $OK#9a
>
> lldb-mi < 37> send packet: $QSetSTDOUT:2f6465762f7074732f3235#ad
>
> lldb-mi < 6> read packet: $OK#9a
>
> lldb-mi < 37> send packet: $QSetSTDERR:2f6465762f7074732f3235#9e
>
> lldb-mi < 6> read packet: $OK#9a
>
> lldb-mi < 21> send packet: $QSetDisableASLR:1#ce
>
> lldb-mi < 6> read packet: $OK#9a
>
> lldb-mi < 23> send packet: $QSetDetachOnError:1#f8
>
> lldb-mi < 6> read packet: $OK#9a
>
> lldb-mi < 22> send packet: $QLaunchArch:x86_64#13
>
> lldb-mi < 6> read packet: $OK#9a
>
> lldb-mi < 72> send packet:
> $A62,0,2f757372322f746564776f6f642f6c6c6462
>
> 5f746573742f666163746c696e#19
>
> lldb-mi < 6> read packet: $OK#9a
>
> lldb-mi < 18> send packet: $qLaunchSuccess#a5
>
> lldb-mi < 6> read packet: $OK#9a
>
> lldb-mi < 16> send packet: $qProcessInfo#dc
>
> lldb-mi < 172> read packet:
> $pid:378a;parent-pid:3785;real-uid:34004;real-gid:c8;effective-uid:34004;effective-gid:c8;triple:7838365f36342d2d6c696e7578
>
>
>
> the rest of the log is full of qRegisterInfo and other packets from a
> successful run.
>
>
>
> *From:* aprantl at apple.com <aprantl at apple.com>
> *Sent:* Tuesday, August 14, 2018 4:50 PM
> *To:* Александр Поляков <polyakov.alx at gmail.com>
> *Cc:* Ted Woodward <ted.woodward at codeaurora.org>; LLDB <
> lldb-dev at lists.llvm.org>
> *Subject:* Re: [lldb-dev] Failing LIT-based lldb-mi tests
>
>
>
>
>
>
>
> On Aug 14, 2018, at 2:43 PM, Александр Поляков <polyakov.alx at gmail.com>
> wrote:
>
>
>
> Here is what I got from gdb-remote packet log:
>
> (gdb)
>
> lldb-mi < 1> send packet: +
>
> lldb-mi history[1] tid=0x784a < 1> send packet: +
>
> lldb-mi < 19> send packet: $QStartNoAckMode#b0
>
> lldb-mi < 1> read packet: +
>
> lldb-mi < 6> read packet: $OK#9a
>
> lldb-mi < 1> send packet: +
>
> lldb-mi < 41> send packet:
> $qSupported:xmlRegisters=i386,arm,mips#12
>
> lldb-mi < 124> read packet:
> $PacketSize=20000;QStartNoAckMode+;QThreadSuffixSupported+;QListThreadsInStopReply+;qEcho+;QPassSignals+;qXfer:auxv:read+#be
>
> lldb-mi < 26> send packet: $QThreadSuffixSupported#e4
>
> lldb-mi < 6> read packet: $OK#9a
>
> lldb-mi < 27> send packet: $QListThreadsInStopReply#21
>
> lldb-mi < 6> read packet: $OK#9a
>
> lldb-mi < 13> send packet: $qHostInfo#9b
>
> lldb-mi < 11> send packet: $qEcho:1#5b
>
> Could somebody help me with understanding of what is happening here?
>
>
>
>
>
> Just to clarify: Is this from a session that failed with the symptoms you
> described earlier?
>
> I'm not familiar with the protocol, but the fact that there are send and
> read log entries makes it sound like the communication itself is working.
>
>
>
> -- adrian
>
>
>
>
> --
>
> Alexander
>
--
Alexander
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/lldb-dev/attachments/20180815/8aa1f320/attachment-0001.html>
More information about the lldb-dev
mailing list