Message ID | 4EC373F8.6080300@cn.fujitsu.com |
---|---|
State | New |
Headers | show |
----- Original Message ----- > Hi, all > > 'virsh dump' can not work when host pci device is used by guest. We have > discussed this issue here: > http://lists.nongnu.org/archive/html/qemu-devel/2011-10/msg00736.html > > We have determined to introduce a new command dump to dump memory. > The core file's format can be elf. > > I created a kdump-elf vmcore, and found that it can be used by both > crash and gdb: > > # gdb /usr/lib/debug/lib/modules/2.6.32-71.el6.x86_64/vmlinux > /work/core/vmcore > GNU gdb (GDB) Red Hat Enterprise Linux (7.2-48.el6) > Copyright (C) 2010 Free Software Foundation, Inc. > License GPLv3+: GNU GPL version 3 or later > <http://gnu.org/licenses/gpl.html> > This is free software: you are free to change and redistribute it. > There is NO WARRANTY, to the extent permitted by law. Type "show > copying" > and "show warranty" for details. > This GDB was configured as "x86_64-redhat-linux-gnu". > For bug reporting instructions, please see: > <http://www.gnu.org/software/gdb/bugs/>... > Reading symbols from > /usr/lib/debug/lib/modules/2.6.32-71.el6.x86_64/vmlinux...done. > [New Thread 1691] > [New <main task>] > #0 sysrq_handle_crash (key=99, tty=0x0) at drivers/char/sysrq.c:130 > 130 drivers/char/sysrq.c: No such file or directory. > in drivers/char/sysrq.c > (gdb) bt > #0 sysrq_handle_crash (key=99, tty=0x0) at drivers/char/sysrq.c:130 > #1 0xffffffff8130d822 in __handle_sysrq (key=99, tty=0x0, > check_mask=<value optimized out>) at drivers/char/sysrq.c:521 > #2 0xffffffff8130d8de in write_sysrq_trigger (file=<value optimized > out>, buf=<value optimized out>, count=2, ppos=<value optimized > out>) at drivers/char/sysrq.c:599 > #3 0xffffffff811cf31e in proc_reg_write (file=<value optimized out>, > buf=0x7fdabafea000 <Address 0x7fdabafea000 out of bounds>, count=2, > ppos=<value optimized out>) > at fs/proc/inode.c:207 > #4 0xffffffff8116c818 in vfs_write (file=0xffff88003c7bb740, > buf=0x7fdabafea000 <Address 0x7fdabafea000 out of bounds>, > count=<value optimized out>, pos=0xffff88003767ff48) > at fs/read_write.c:347 > #5 0xffffffff8116d251 in sys_write (fd=<value optimized out>, > buf=0x7fdabafea000 <Address 0x7fdabafea000 out of bounds>, count=2) > at fs/read_write.c:399 > #6 0xffffffff81013172 in ?? () at arch/x86/kernel/entry_64.S:487 > #7 0x0000000000000246 in ?? () > #8 0x00000000ffffffff in ?? () > #9 0x00007fdabafde700 in ?? () > #10 0x000000000000000a in ?? () > #11 0x0000000000000001 in ?? () > #12 0x0000000000000002 in ?? () > #13 0x0000000000000001 in ?? () > #14 0x00000030f80d4230 in ?? () > #15 0x0000000000000033 in ?? () > #16 0x0000000000010206 in ?? () > #17 0x00007fff8a126470 in ?? () > #18 0x000000000000002b in ?? () > #19 0xffff8800374f5000 in ?? () > #20 0xffff88003c6f9000 in ?? () > #21 0x0000000000000080 in ?? () > #22 0xffff880037680080 in ?? () > #23 0xffffffff00000014 in ?? () > #24 0x0000000000000000 in ?? () > (gdb) q > # crash -s /usr/lib/debug/lib/modules/2.6.32-71.el6.x86_64/vmlinux /work/core/vmcore > crash> bt > PID: 1691 TASK: ffff88003711d520 CPU: 0 COMMAND: "bash" > #0 [ffff88003767fae0] machine_kexec at ffffffff8103695b > #1 [ffff88003767fb40] crash_kexec at ffffffff810b8f08 > #2 [ffff88003767fc10] oops_end at ffffffff814cbbd0 > #3 [ffff88003767fc40] no_context at ffffffff8104651b > #4 [ffff88003767fc90] __bad_area_nosemaphore at ffffffff810467a5 > #5 [ffff88003767fce0] bad_area at ffffffff810468ce > #6 [ffff88003767fd10] do_page_fault at ffffffff814cd740 > #7 [ffff88003767fd60] page_fault at ffffffff814caf45 > [exception RIP: sysrq_handle_crash+22] > RIP: ffffffff8130d566 RSP: ffff88003767fe18 RFLAGS: 00010096 > RAX: 0000000000000010 RBX: 0000000000000063 RCX: 0000000000000000 > RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000063 > RBP: ffff88003767fe18 R8: 0000000000000000 R9: ffffffff815106c0 > R10: 0000000000000001 R11: 0000000000000000 R12: 0000000000000000 > R13: ffffffff8179e6c0 R14: 0000000000000286 R15: 0000000000000007 > ORIG_RAX: ffffffffffffffff CS: 0010 SS: 0018 > #8 [ffff88003767fe20] __handle_sysrq at ffffffff8130d822 > #9 [ffff88003767fe70] write_sysrq_trigger at ffffffff8130d8de > #10 [ffff88003767fea0] proc_reg_write at ffffffff811cf31e > #11 [ffff88003767fef0] vfs_write at ffffffff8116c818 > #12 [ffff88003767ff30] sys_write at ffffffff8116d251 > #13 [ffff88003767ff80] system_call_fastpath at ffffffff81013172 > RIP: 00000030f80d4230 RSP: 00007fff8a126470 RFLAGS: 00010206 > RAX: 0000000000000001 RBX: ffffffff81013172 RCX: 0000000000000400 > RDX: 0000000000000002 RSI: 00007fdabafea000 RDI: 0000000000000001 > RBP: 00007fdabafea000 R8: 000000000000000a R9: 00007fdabafde700 > R10: 00000000ffffffff R11: 0000000000000246 R12: 0000000000000002 > R13: 00000030f8379780 R14: 0000000000000002 R15: 00000030f8379780 > ORIG_RAX: 0000000000000001 CS: 0033 SS: 002b > crash> > > I wrote a sample(not finished). It only can works on x86_64(both host and guest) > I use it to create a core file: > # readelf -h /tmp/vm2.save > ELF Header: > Magic: 7f 45 4c 46 02 01 01 00 00 00 00 00 00 00 00 00 > Class: ELF64 > Data: 2's complement, little endian > Version: 1 (current) > OS/ABI: UNIX - System V > ABI Version: 0 > Type: CORE (Core file) > Machine: Advanced Micro Devices X86-64 > Version: 0x1 > Entry point address: 0x0 > Start of program headers: 64 (bytes into file) > Start of section headers: 0 (bytes into file) > Flags: 0x0 > Size of this header: 64 (bytes) > Size of program headers: 56 (bytes) > Number of program headers: 9 > Size of section headers: 0 (bytes) > Number of section headers: 0 > Section header string table index: 0 > # readelf -l /tmp/vm2.save > > Elf file type is CORE (Core file) > Entry point 0x0 > There are 9 program headers, starting at offset 64 > > Program Headers: > Type Offset VirtAddr PhysAddr > FileSiz MemSiz Flags Align > NOTE 0x0000000000000238 0x0000000000000000 0x0000000000000000 > 0x00000000000002c8 0x00000000000002c8 0 > LOAD 0x0000000000000500 0xffffffff81000000 0x0000000001000000 > 0x000000001f000000 0x000000001f000000 0 > LOAD 0x000000001f000500 0x0000000000000000 0x0000000000000000 > 0x0000000001000000 0x0000000001000000 0 > LOAD 0x0000000020000500 0x0000000000000000 0x0000000020000000 > 0x0000000000020000 0x0000000000020000 0 > LOAD 0x0000000020020500 0x0000000000000000 0x0000000020870000 > 0x0000000000010000 0x0000000000010000 0 > LOAD 0x0000000020030500 0x0000000000000000 0x0000000020850000 > 0x0000000000020000 0x0000000000020000 0 > LOAD 0x0000000020050500 0x0000000000000000 0x0000000020840000 > 0x0000000000010000 0x0000000000010000 0 > LOAD 0x0000000020060500 0x0000000000000000 0x0000000020040000 > 0x0000000000800000 0x0000000000800000 0 > LOAD 0x0000000020860500 0x0000000000000000 0x0000000020020000 > 0x0000000000020000 0x0000000000020000 0 > > I can use crash to anaylze the file, but I can not use gdb to anaylze it. > # gdb /usr/lib/debug/lib/modules/2.6.32-71.el6.x86_64/vmlinux /tmp/vm2.save > GNU gdb (GDB) Red Hat Enterprise Linux (7.2-48.el6) > Copyright (C) 2010 Free Software Foundation, Inc. > License GPLv3+: GNU GPL version 3 or later > <http://gnu.org/licenses/gpl.html> > This is free software: you are free to change and redistribute it. > There is NO WARRANTY, to the extent permitted by law. Type "show > copying" > and "show warranty" for details. > This GDB was configured as "x86_64-redhat-linux-gnu". > For bug reporting instructions, please see: > <http://www.gnu.org/software/gdb/bugs/>... > Reading symbols from > /usr/lib/debug/lib/modules/2.6.32-71.el6.x86_64/vmlinux...done. > [New <main task>] > [New <main task>] > #0 0x8103be8b00000000 in ?? () > (gdb) bt > #0 0x8103be8b00000000 in ?? () > Cannot access memory at address 0x8170dec800000000 > (gdb) q > > My first and the most important question is that: Is there necessary > to continue this work? > > The attachment is the sample patch. > > Thanks > Wen Congyang From an enterprise/support point of view, the wholesale replacement of the current use of the savevm dumpfile format by "virsh dump" with this ELF style format would be a *huge* improvement. Dave Anderson
On 2011-11-16 14:29, Dave Anderson wrote: > > > ----- Original Message ----- >> Hi, all >> >> 'virsh dump' can not work when host pci device is used by guest. We have >> discussed this issue here: >> http://lists.nongnu.org/archive/html/qemu-devel/2011-10/msg00736.html >> >> We have determined to introduce a new command dump to dump memory. >> The core file's format can be elf. >> >> I created a kdump-elf vmcore, and found that it can be used by both >> crash and gdb: >> >> # gdb /usr/lib/debug/lib/modules/2.6.32-71.el6.x86_64/vmlinux >> /work/core/vmcore >> GNU gdb (GDB) Red Hat Enterprise Linux (7.2-48.el6) >> Copyright (C) 2010 Free Software Foundation, Inc. >> License GPLv3+: GNU GPL version 3 or later >> <http://gnu.org/licenses/gpl.html> >> This is free software: you are free to change and redistribute it. >> There is NO WARRANTY, to the extent permitted by law. Type "show >> copying" >> and "show warranty" for details. >> This GDB was configured as "x86_64-redhat-linux-gnu". >> For bug reporting instructions, please see: >> <http://www.gnu.org/software/gdb/bugs/>... >> Reading symbols from >> /usr/lib/debug/lib/modules/2.6.32-71.el6.x86_64/vmlinux...done. >> [New Thread 1691] >> [New <main task>] >> #0 sysrq_handle_crash (key=99, tty=0x0) at drivers/char/sysrq.c:130 >> 130 drivers/char/sysrq.c: No such file or directory. >> in drivers/char/sysrq.c >> (gdb) bt >> #0 sysrq_handle_crash (key=99, tty=0x0) at drivers/char/sysrq.c:130 >> #1 0xffffffff8130d822 in __handle_sysrq (key=99, tty=0x0, >> check_mask=<value optimized out>) at drivers/char/sysrq.c:521 >> #2 0xffffffff8130d8de in write_sysrq_trigger (file=<value optimized >> out>, buf=<value optimized out>, count=2, ppos=<value optimized >> out>) at drivers/char/sysrq.c:599 >> #3 0xffffffff811cf31e in proc_reg_write (file=<value optimized out>, >> buf=0x7fdabafea000 <Address 0x7fdabafea000 out of bounds>, count=2, >> ppos=<value optimized out>) >> at fs/proc/inode.c:207 >> #4 0xffffffff8116c818 in vfs_write (file=0xffff88003c7bb740, >> buf=0x7fdabafea000 <Address 0x7fdabafea000 out of bounds>, >> count=<value optimized out>, pos=0xffff88003767ff48) >> at fs/read_write.c:347 >> #5 0xffffffff8116d251 in sys_write (fd=<value optimized out>, >> buf=0x7fdabafea000 <Address 0x7fdabafea000 out of bounds>, count=2) >> at fs/read_write.c:399 >> #6 0xffffffff81013172 in ?? () at arch/x86/kernel/entry_64.S:487 >> #7 0x0000000000000246 in ?? () >> #8 0x00000000ffffffff in ?? () >> #9 0x00007fdabafde700 in ?? () >> #10 0x000000000000000a in ?? () >> #11 0x0000000000000001 in ?? () >> #12 0x0000000000000002 in ?? () >> #13 0x0000000000000001 in ?? () >> #14 0x00000030f80d4230 in ?? () >> #15 0x0000000000000033 in ?? () >> #16 0x0000000000010206 in ?? () >> #17 0x00007fff8a126470 in ?? () >> #18 0x000000000000002b in ?? () >> #19 0xffff8800374f5000 in ?? () >> #20 0xffff88003c6f9000 in ?? () >> #21 0x0000000000000080 in ?? () >> #22 0xffff880037680080 in ?? () >> #23 0xffffffff00000014 in ?? () >> #24 0x0000000000000000 in ?? () >> (gdb) q >> # crash -s /usr/lib/debug/lib/modules/2.6.32-71.el6.x86_64/vmlinux /work/core/vmcore >> crash> bt >> PID: 1691 TASK: ffff88003711d520 CPU: 0 COMMAND: "bash" >> #0 [ffff88003767fae0] machine_kexec at ffffffff8103695b >> #1 [ffff88003767fb40] crash_kexec at ffffffff810b8f08 >> #2 [ffff88003767fc10] oops_end at ffffffff814cbbd0 >> #3 [ffff88003767fc40] no_context at ffffffff8104651b >> #4 [ffff88003767fc90] __bad_area_nosemaphore at ffffffff810467a5 >> #5 [ffff88003767fce0] bad_area at ffffffff810468ce >> #6 [ffff88003767fd10] do_page_fault at ffffffff814cd740 >> #7 [ffff88003767fd60] page_fault at ffffffff814caf45 >> [exception RIP: sysrq_handle_crash+22] >> RIP: ffffffff8130d566 RSP: ffff88003767fe18 RFLAGS: 00010096 >> RAX: 0000000000000010 RBX: 0000000000000063 RCX: 0000000000000000 >> RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000063 >> RBP: ffff88003767fe18 R8: 0000000000000000 R9: ffffffff815106c0 >> R10: 0000000000000001 R11: 0000000000000000 R12: 0000000000000000 >> R13: ffffffff8179e6c0 R14: 0000000000000286 R15: 0000000000000007 >> ORIG_RAX: ffffffffffffffff CS: 0010 SS: 0018 >> #8 [ffff88003767fe20] __handle_sysrq at ffffffff8130d822 >> #9 [ffff88003767fe70] write_sysrq_trigger at ffffffff8130d8de >> #10 [ffff88003767fea0] proc_reg_write at ffffffff811cf31e >> #11 [ffff88003767fef0] vfs_write at ffffffff8116c818 >> #12 [ffff88003767ff30] sys_write at ffffffff8116d251 >> #13 [ffff88003767ff80] system_call_fastpath at ffffffff81013172 >> RIP: 00000030f80d4230 RSP: 00007fff8a126470 RFLAGS: 00010206 >> RAX: 0000000000000001 RBX: ffffffff81013172 RCX: 0000000000000400 >> RDX: 0000000000000002 RSI: 00007fdabafea000 RDI: 0000000000000001 >> RBP: 00007fdabafea000 R8: 000000000000000a R9: 00007fdabafde700 >> R10: 00000000ffffffff R11: 0000000000000246 R12: 0000000000000002 >> R13: 00000030f8379780 R14: 0000000000000002 R15: 00000030f8379780 >> ORIG_RAX: 0000000000000001 CS: 0033 SS: 002b >> crash> >> >> I wrote a sample(not finished). It only can works on x86_64(both host and guest) >> I use it to create a core file: >> # readelf -h /tmp/vm2.save >> ELF Header: >> Magic: 7f 45 4c 46 02 01 01 00 00 00 00 00 00 00 00 00 >> Class: ELF64 >> Data: 2's complement, little endian >> Version: 1 (current) >> OS/ABI: UNIX - System V >> ABI Version: 0 >> Type: CORE (Core file) >> Machine: Advanced Micro Devices X86-64 >> Version: 0x1 >> Entry point address: 0x0 >> Start of program headers: 64 (bytes into file) >> Start of section headers: 0 (bytes into file) >> Flags: 0x0 >> Size of this header: 64 (bytes) >> Size of program headers: 56 (bytes) >> Number of program headers: 9 >> Size of section headers: 0 (bytes) >> Number of section headers: 0 >> Section header string table index: 0 >> # readelf -l /tmp/vm2.save >> >> Elf file type is CORE (Core file) >> Entry point 0x0 >> There are 9 program headers, starting at offset 64 >> >> Program Headers: >> Type Offset VirtAddr PhysAddr >> FileSiz MemSiz Flags Align >> NOTE 0x0000000000000238 0x0000000000000000 0x0000000000000000 >> 0x00000000000002c8 0x00000000000002c8 0 >> LOAD 0x0000000000000500 0xffffffff81000000 0x0000000001000000 >> 0x000000001f000000 0x000000001f000000 0 >> LOAD 0x000000001f000500 0x0000000000000000 0x0000000000000000 >> 0x0000000001000000 0x0000000001000000 0 >> LOAD 0x0000000020000500 0x0000000000000000 0x0000000020000000 >> 0x0000000000020000 0x0000000000020000 0 >> LOAD 0x0000000020020500 0x0000000000000000 0x0000000020870000 >> 0x0000000000010000 0x0000000000010000 0 >> LOAD 0x0000000020030500 0x0000000000000000 0x0000000020850000 >> 0x0000000000020000 0x0000000000020000 0 >> LOAD 0x0000000020050500 0x0000000000000000 0x0000000020840000 >> 0x0000000000010000 0x0000000000010000 0 >> LOAD 0x0000000020060500 0x0000000000000000 0x0000000020040000 >> 0x0000000000800000 0x0000000000800000 0 >> LOAD 0x0000000020860500 0x0000000000000000 0x0000000020020000 >> 0x0000000000020000 0x0000000000020000 0 >> >> I can use crash to anaylze the file, but I can not use gdb to anaylze it. >> # gdb /usr/lib/debug/lib/modules/2.6.32-71.el6.x86_64/vmlinux /tmp/vm2.save >> GNU gdb (GDB) Red Hat Enterprise Linux (7.2-48.el6) >> Copyright (C) 2010 Free Software Foundation, Inc. >> License GPLv3+: GNU GPL version 3 or later >> <http://gnu.org/licenses/gpl.html> >> This is free software: you are free to change and redistribute it. >> There is NO WARRANTY, to the extent permitted by law. Type "show >> copying" >> and "show warranty" for details. >> This GDB was configured as "x86_64-redhat-linux-gnu". >> For bug reporting instructions, please see: >> <http://www.gnu.org/software/gdb/bugs/>... >> Reading symbols from >> /usr/lib/debug/lib/modules/2.6.32-71.el6.x86_64/vmlinux...done. >> [New <main task>] >> [New <main task>] >> #0 0x8103be8b00000000 in ?? () >> (gdb) bt >> #0 0x8103be8b00000000 in ?? () >> Cannot access memory at address 0x8170dec800000000 >> (gdb) q >> >> My first and the most important question is that: Is there necessary >> to continue this work? >> >> The attachment is the sample patch. >> >> Thanks >> Wen Congyang > > From an enterprise/support point of view, the wholesale replacement > of the current use of the savevm dumpfile format by "virsh dump" with > this ELF style format would be a *huge* improvement. Yes, fully agree. Would be cool if that could actually work for both crash and gdb. Looking forward! Jan
At 11/18/2011 08:46 PM, Jan Kiszka Write: > On 2011-11-16 14:29, Dave Anderson wrote: >> >> >> ----- Original Message ----- >>> Hi, all >>> >>> 'virsh dump' can not work when host pci device is used by guest. We have >>> discussed this issue here: >>> http://lists.nongnu.org/archive/html/qemu-devel/2011-10/msg00736.html >>> >>> We have determined to introduce a new command dump to dump memory. >>> The core file's format can be elf. >>> >>> I created a kdump-elf vmcore, and found that it can be used by both >>> crash and gdb: >>> >> From an enterprise/support point of view, the wholesale replacement >> of the current use of the savevm dumpfile format by "virsh dump" with >> this ELF style format would be a *huge* improvement. > > Yes, fully agree. Would be cool if that could actually work for both > crash and gdb. Looking forward! Because the memory size for x86 machine can greater than 4G, so we should create elf64 format core file for 32bit OS. I create a vmcore: the guest OS is 32-bit, and the vmcore is elf64 format. I can use crash to anaylyze it, but gdb can not do the same thing. I create a kdump-elf64 vmcore on 32-bit machine, and gdb still can not anaylyze it. Does gdb support elf64 format core file on x86 box? Thanks Wen Congyang > > Jan >
On 2011-11-21 06:06, Wen Congyang wrote: > At 11/18/2011 08:46 PM, Jan Kiszka Write: >> On 2011-11-16 14:29, Dave Anderson wrote: >>> >>> >>> ----- Original Message ----- >>>> Hi, all >>>> >>>> 'virsh dump' can not work when host pci device is used by guest. We have >>>> discussed this issue here: >>>> http://lists.nongnu.org/archive/html/qemu-devel/2011-10/msg00736.html >>>> >>>> We have determined to introduce a new command dump to dump memory. >>>> The core file's format can be elf. >>>> >>>> I created a kdump-elf vmcore, and found that it can be used by both >>>> crash and gdb: >>>> > >>> From an enterprise/support point of view, the wholesale replacement >>> of the current use of the savevm dumpfile format by "virsh dump" with >>> this ELF style format would be a *huge* improvement. >> >> Yes, fully agree. Would be cool if that could actually work for both >> crash and gdb. Looking forward! > > Because the memory size for x86 machine can greater than 4G, so we should > create elf64 format core file for 32bit OS. > > I create a vmcore: the guest OS is 32-bit, and the vmcore is elf64 format. > I can use crash to anaylyze it, but gdb can not do the same thing. > > I create a kdump-elf64 vmcore on 32-bit machine, and gdb still can not anaylyze > it. > > Does gdb support elf64 format core file on x86 box? > Dunno, but I'm trying to pull in some interested gdb folks. Jan
Jan Kiszka <jan.kiszka@web.de> writes: > On 2011-11-21 06:06, Wen Congyang wrote: >> At 11/18/2011 08:46 PM, Jan Kiszka Write: >>> On 2011-11-16 14:29, Dave Anderson wrote: >>>> >>>> >>>> ----- Original Message ----- >>>>> Hi, all >>>>> >>>>> 'virsh dump' can not work when host pci device is used by guest. We have >>>>> discussed this issue here: >>>>> http://lists.nongnu.org/archive/html/qemu-devel/2011-10/msg00736.html>>>> >>>>> We have determined to introduce a new command dump to dump memory. >>>>> The core file's format can be elf. >>>>> >>>>> I created a kdump-elf vmcore, and found that it can be used by both >>>>> crash and gdb: >>>>> >> >>>> From an enterprise/support point of view, the wholesale replacement >>>> of the current use of the savevm dumpfile format by "virsh dump" with >>>> this ELF style format would be a *huge* improvement. >>> >>> Yes, fully agree. Would be cool if that could actually work for both >>> crash and gdb. Looking forward! >> >> Because the memory size for x86 machine can greater than 4G, so we should >> create elf64 format core file for 32bit OS. >> >> I create a vmcore: the guest OS is 32-bit, and the vmcore is elf64 format. >> I can use crash to anaylyze it, but gdb can not do the same thing. >> >> I create a kdump-elf64 vmcore on 32-bit machine, and gdb still can not anaylyze >> it. >> >> Does gdb support elf64 format core file on x86 box? >> > > Dunno, but I'm trying to pull in some interested gdb folks. Hello, IIRC, GDB supports ELF64 corefiles on x86 boxes, but only when configured with `--enable-64-bit-bfd'. Otherwise, it won't be able to properly understand the format.
From bdb3daaeb2743a14df2cab364622e2f47ae25093 Mon Sep 17 00:00:00 2001 From: Wen Congyang <wency@cn.fujitsu.com> Date: Wed, 16 Nov 2011 16:06:10 +0800 Subject: [PATCH] dump sample --- Makefile.target | 1 + dump.c | 377 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ dump.h | 1 + hmp-commands.hx | 16 +++ monitor.c | 3 + qmp-commands.hx | 24 ++++ 6 files changed, 422 insertions(+), 0 deletions(-) create mode 100644 dump.c create mode 100644 dump.h diff --git a/Makefile.target b/Makefile.target index a111521..95d48a5 100644 --- a/Makefile.target +++ b/Makefile.target @@ -84,6 +84,7 @@ libobj-y += cpu_init.o endif libobj-$(TARGET_SPARC) += int32_helper.o libobj-$(TARGET_SPARC64) += int64_helper.o +libobj-y += dump.o libobj-y += disas.o libobj-$(CONFIG_TCI_DIS) += tci-dis.o diff --git a/dump.c b/dump.c new file mode 100644 index 0000000..bdec246 --- /dev/null +++ b/dump.c @@ -0,0 +1,377 @@ +/* + * QEMU live dump + * + * Copyright Fujitsu, Corp. 2011 + * + * Authors: + * + * This work is licensed under the terms of the GNU GPL, version 2. See + * the COPYING file in the top-level directory. + * + */ + +#include "qemu-common.h" +#include <unistd.h> +#include <elf.h> +#include <sys/procfs.h> +#include "cpu.h" +#include "cpu-all.h" +#include "targphys.h" +#include "monitor.h" +#include "kvm.h" +#include "dump.h" +#include "sysemu.h" + +#ifdef TARGET_X86_64 +typedef struct { + target_ulong r15,r14,r13,r12,rbp,rbx,r11,r10; + target_ulong r9,r8,rax,rcx,rdx,rsi,rdi,orig_rax; + target_ulong rip,cs,eflags; + target_ulong rsp,ss; + target_ulong fs_base, gs_base; + target_ulong ds,es,fs,gs; +} x86_64_user_regs_struct; + +static int write_elf64_note(Monitor *mon, int fd, CPUState *env, + target_phys_addr_t *offset) +{ + x86_64_user_regs_struct regs; + Elf64_Nhdr *note; + char *buf; + int descsz, note_size, name_size = 5; + const char *name = "CORE"; + int ret; + + regs.r15 = env->regs[15]; + regs.r14 = env->regs[14]; + regs.r13 = env->regs[13]; + regs.r12 = env->regs[12]; + regs.r11 = env->regs[11]; + regs.r10 = env->regs[10]; + regs.r9 = env->regs[9]; + regs.r8 = env->regs[8]; + regs.rbp = env->regs[R_EBP]; + regs.rsp = env->regs[R_ESP]; + regs.rdi = env->regs[R_EDI]; + regs.rsi = env->regs[R_ESI]; + regs.rdx = env->regs[R_EDX]; + regs.rcx = env->regs[R_ECX]; + regs.rbx = env->regs[R_EBX]; + regs.rax = env->regs[R_EAX]; + regs.rip = env->eip; + regs.eflags = env->eflags; + + /* FIXME */ + regs.orig_rax = 0; + regs.cs = 0; + regs.ss = 0; + regs.fs_base = 0; + regs.gs_base = 0; + regs.ds = 0; + regs.es = 0; + regs.fs = 0; + regs.gs = 0; + + descsz = sizeof(prstatus_t) - sizeof(elf_gregset_t) + + sizeof(x86_64_user_regs_struct); + note_size = ((sizeof(Elf64_Nhdr) + 3) / 4 + (name_size + 3) / 4 + + (descsz +3) / 4) * 4; + note = g_malloc(note_size); + + note->n_namesz = name_size; + note->n_descsz = descsz; + note->n_type = NT_PRSTATUS; + buf = (char *)note; + buf += ((sizeof(Elf64_Nhdr) + 3) / 4) * 4; + memcpy(buf, name, name_size); + buf += ((name_size + 3) / 4) * 4; + buf += descsz - sizeof(x86_64_user_regs_struct) - sizeof(int); + memcpy(buf, ®s, sizeof(x86_64_user_regs_struct)); + + lseek(fd, *offset, SEEK_SET); + ret = write(fd, note, note_size); + g_free(note); + if (ret < 0) { + monitor_printf(mon, "dump: failed to write elf prstatus.\n"); + return -1; + } + + *offset += note_size; + + return 0; +} +#endif + +typedef struct { + uint32_t ebx, ecx, edx, esi, edi, ebp, eax; + unsigned short ds, __ds, es, __es; + unsigned short fs, __fs, gs, __gs; + uint32_t orig_eax, eip; + unsigned short cs, __cs; + uint32_t eflags, esp; + unsigned short ss, __ss; +} x86_user_regs_struct; + +static int write_elf32_note(Monitor *mon, int fd, CPUState *env, + target_phys_addr_t *offset) +{ + /* TODO */ + return 0; +} + +static target_ulong get_phys_base_addr(CPUState *env, target_ulong *base_vaddr) +{ + int i; + target_ulong kernel_base = -1; + target_ulong last, mask; + + for (i = 30, last = -1; (kernel_base == -1) && (i >= 20); i--) { + mask = ~((1LL << i) - 1); + *base_vaddr = env->idt.base & mask; + if (*base_vaddr == last) { + continue; + } + + kernel_base = cpu_get_phys_page_debug(env, *base_vaddr); + last = *base_vaddr; + } + + return kernel_base; +} + +static int write_elf_header(Monitor *mon, int fd, int phdr_num, bool lma) +{ + Elf64_Ehdr elf_header; + int ret; + + memset(&elf_header, 0, sizeof(Elf64_Ehdr)); + memcpy(&elf_header, ELFMAG, 4); + elf_header.e_ident[EI_CLASS] = ELFCLASS64; + elf_header.e_ident[EI_DATA] = ELFDATA2LSB; /* FIXME */ + elf_header.e_ident[EI_VERSION] = EV_CURRENT; + elf_header.e_type = ET_CORE; + elf_header.e_machine = lma ? EM_X86_64: EM_386; + elf_header.e_version = EV_CURRENT; + elf_header.e_ehsize = sizeof(elf_header); + elf_header.e_phoff = sizeof(Elf64_Ehdr); + elf_header.e_phentsize = sizeof(Elf64_Phdr); + elf_header.e_phnum = phdr_num; + + lseek(fd, 0, SEEK_SET); + ret = write(fd, &elf_header, sizeof(elf_header)); + if (ret < 0) { + monitor_printf(mon, "dump: failed to write elf header.\n"); + return -1; + } + + return 0; +} + +static int write_elf_load(Monitor *mon, int fd, RAMBlock *block, int phdr_index, + target_phys_addr_t *offset, target_ulong base_vaddr) +{ + Elf64_Phdr phdr; + off_t phdr_offset; + int ret; + + memset(&phdr, 0, sizeof(Elf64_Phdr)); + phdr.p_type = PT_LOAD; + phdr.p_offset = *offset; + phdr.p_paddr = block->offset; + phdr.p_filesz = block->length; + phdr.p_memsz = block->length; + phdr.p_vaddr = base_vaddr; + + phdr_offset = sizeof(Elf64_Ehdr) + sizeof(Elf64_Phdr) * phdr_index; + lseek(fd, phdr_offset, SEEK_SET); + ret = write(fd, &phdr, sizeof(Elf64_Phdr)); + if (ret < 0) { + monitor_printf(mon, "dump: failed to write program header table.\n"); + return -1; + } + + lseek(fd, *offset, SEEK_SET); + ret = write(fd, block->host, block->length); + if (ret < 0) { + monitor_printf(mon, "dump: failed to write program segment.\n"); + return -1; + } + *offset += block->length; + + return 0; +} + +static int write_elf_notes(Monitor *mon, int fd, int phdr_index, + target_phys_addr_t *offset, bool lma) +{ + CPUState *env; + int ret; + target_phys_addr_t begin = *offset; + Elf64_Phdr phdr; + off_t phdr_offset; + + for (env = first_cpu; env != NULL; env = env->next_cpu) { +#ifdef TARGET_X86_64 + if (lma) { + ret = write_elf64_note(mon, fd, env, offset); + } else { +#endif + ret = write_elf32_note(mon, fd, env, offset); +#ifdef TARGET_X86_64 + } +#endif + + if (ret < 0) { + monitor_printf(mon, "dump: failed to write elf notes.\n"); + return -1; + } + } + + memset(&phdr, 0, sizeof(Elf64_Phdr)); + phdr.p_type = PT_NOTE; + phdr.p_offset = begin; + phdr.p_paddr = 0; + phdr.p_filesz = *offset - begin; + phdr.p_memsz = *offset - begin; + phdr.p_vaddr = 0; + + phdr_offset = sizeof(Elf64_Ehdr); + lseek(fd, phdr_offset, SEEK_SET); + ret = write(fd, &phdr, sizeof(Elf64_Phdr)); + if (ret < 0) { + monitor_printf(mon, "dump: failed to write program header table.\n"); + return -1; + } + + return 0; +} + +static int create_vmcore(Monitor *mon, int fd) +{ + CPUState *env; + target_ulong kernel_base = -1, base_vaddr; + target_phys_addr_t offset; + int phdr_num, phdr_index; + RAMBlock *block; + bool lma = false; + int ret; + + for (env = first_cpu; env != NULL; env = env->next_cpu) { + cpu_synchronize_state(env); + } + +#ifdef TARGET_X86_64 + lma = !!(first_cpu->hflags & HF_LMA_MASK); + if (lma) { + kernel_base = get_phys_base_addr(first_cpu, &base_vaddr); + if (kernel_base == -1) { + monitor_printf(mon, "fd_dump: can not get phys_base\n"); + return -1; + } + } +#endif + + phdr_num = 1; /* PT_NOTE */ + QLIST_FOREACH(block, &ram_list.blocks, next) { + if (lma && kernel_base > block->offset && + kernel_base < (block->offset + block->length)) { + phdr_num++; + } + phdr_num++; + } + + ret = write_elf_header(mon, fd, phdr_num, lma); + if (ret < 0) + return -1; + + phdr_index = 0; + offset = sizeof(Elf64_Ehdr) + sizeof(Elf64_Phdr) * phdr_num; + + ret = write_elf_notes(mon, fd, phdr_index++, &offset, lma); + if (ret < 0) + return -1; + +#ifdef TARGET_X86_64 + QLIST_FOREACH(block, &ram_list.blocks, next) { + if (lma && kernel_base >= block->offset && + kernel_base < (block->offset + block->length)) { + if (kernel_base > block->offset) { + RAMBlock temp_block; + + temp_block.host = block->host + (kernel_base-block->offset); + temp_block.offset = kernel_base; + temp_block.length = block->length - (kernel_base-block->offset); + ret = write_elf_load(mon, fd, &temp_block, phdr_index++, &offset, + base_vaddr); + if (ret < 0) + return -1; + + temp_block.host = block->host; + temp_block.offset = block->offset; + temp_block.length = kernel_base-block->offset; + ret = write_elf_load(mon, fd, &temp_block, phdr_index++, + &offset, 0); + if (ret < 0) + return -1; + } else { + ret = write_elf_load(mon, fd, block, phdr_index++, &offset, + base_vaddr); + if (ret < 0) + return -1; + } + break; + } + } +#endif + + QLIST_FOREACH(block, &ram_list.blocks, next) { + if (lma && kernel_base >= block->offset && + kernel_base < (block->offset + block->length)) { + continue; + } + + ret = write_elf_load(mon, fd, block, phdr_index++, &offset, 0); + if (ret < 0) + return -1; + } + + return 0; +} + +int do_dump(Monitor *mon, const QDict *qdict, QObject **ret_data) +{ + const char *file = qdict_get_str(qdict, "file"); + const char *p; + int fd = -1; + +#if !defined(WIN32) + if (strstart(file, "fd:", &p)) { + fd = monitor_get_fd(mon, p); + if (fd == -1) { + monitor_printf(mon, "fd_dump: invalid file descriptor" + " identifier\n"); + return -1; + } + } +#endif + + if (strstart(file, "file:", &p)) { + fd = open(p, O_WRONLY | O_CREAT | O_TRUNC | O_BINARY); + if (fd < 0) { + monitor_printf(mon, "fd_dump: failed to open %s\n", p); + return -1; + } + } + + if (fd == -1) { + monitor_printf(mon, "unknown dump protocol: %s\n", file); + return -1; + } + + vm_stop(RUN_STATE_PAUSED); + if (create_vmcore(mon, fd) < 0) + return -1; + + return 0; +} + diff --git a/dump.h b/dump.h new file mode 100644 index 0000000..b5d9eb4 --- /dev/null +++ b/dump.h @@ -0,0 +1 @@ +int do_dump(Monitor *mon, const QDict *qdict, QObject **ret_data); diff --git a/hmp-commands.hx b/hmp-commands.hx index 089c1ac..ebbce8c 100644 --- a/hmp-commands.hx +++ b/hmp-commands.hx @@ -772,6 +772,22 @@ Migrate to @var{uri} (using -d to not wait for completion). ETEXI { + .name = "dump", + .args_type = "file:s", + .params = "file", + .help = "dump to file", + .user_print = monitor_user_noop, + .mhandler.cmd_new = do_dump, + }, + + +STEXI +@item dump @var{file} +@findex dump +Dump to @var{file}. +ETEXI + + { .name = "migrate_cancel", .args_type = "", .params = "", diff --git a/monitor.c b/monitor.c index 5ea35de..5df35e0 100644 --- a/monitor.c +++ b/monitor.c @@ -73,6 +73,9 @@ #endif #include "hw/lm32_pic.h" +/* for dump */ +#include "dump.h" + //#define DEBUG //#define DEBUG_COMPLETION diff --git a/qmp-commands.hx b/qmp-commands.hx index 97975a5..5cf21c5 100644 --- a/qmp-commands.hx +++ b/qmp-commands.hx @@ -485,6 +485,30 @@ Notes: EQMP { + .name = "dump", + .args_type = "file:s", + .params = "file", + .help = "dump to file", + .user_print = monitor_user_noop, + .mhandler.cmd_new = do_dump, + }, + +SQMP +dump +------- + +Dump to file. + +Arguments: None. + +Example: + +-> { "execute": "dump", "arguments": { "file": "fd:dump" } } +<- { "return": {} } + +EQMP + + { .name = "migrate_cancel", .args_type = "", .params = "", -- 1.7.1