[syzbot] riscv/fixes test error: lost connection to test machine

Wed Aug 3 22:35:57 PDT 2022

On Tue, 31 May 2022 at 16:10, Alexandre Ghiti
<alexandre.ghiti at canonical.com> wrote:
> On Sat, May 28, 2022 at 10:31 AM Dmitry Vyukov <dvyukov at google.com> wrote:
>>
>> On Sat, 28 May 2022 at 10:09, Alexandre Ghiti <alex at ghiti.fr> wrote:
>> >
>> > On 5/27/22 19:12, Dmitry Vyukov wrote:
>> > > On Fri, 27 May 2022 at 19:04, Dmitry Vyukov <dvyukov at google.com> wrote:
>> > >> On Fri, 27 May 2022 at 16:01, Alexandre Ghiti
>> > >> <alexandre.ghiti at canonical.com> wrote:
>> > >>> On Friday, May 27, 2022 at 3:55:24 PM UTC+2 Dmitry Vyukov wrote:
>> > >>>> On Fri, 27 May 2022 at 15:50, Alexandre Ghiti
>> > >>>> <alexand... at canonical.com> wrote:
>> > >>>>> On Friday, May 27, 2022 at 3:02:01 PM UTC+2 Dmitry Vyukov wrote:
>> > >>>>>> On Fri, 27 May 2022 at 14:55, syzbot
>> > >>>>>> <syzbot+2c5da6... at syzkaller.appspotmail.com> wrote:
>> > >>>>>>> Hello,
>> > >>>>>>>
>> > >>>>>>> syzbot found the following issue on:
>> > >>>>>>>
>> > >>>>>>> HEAD commit: c932edeaf6d6 riscv: dts: microchip: fix gpio1 reg property..
>> > >>>>>>> git tree: git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux.git fixes
>> > >>>>>>> console output: https://syzkaller.appspot.com/x/log.txt?x=1418add5f00000
>> > >>>>>>> kernel config: https://syzkaller.appspot.com/x/.config?x=aa6b5702bdf14a17
>> > >>>>>>> dashboard link: https://syzkaller.appspot.com/bug?extid=2c5da6a0a16a0c4f34aa
>> > >>>>>>> compiler: riscv64-linux-gnu-gcc (Debian 10.2.1-6) 10.2.1 20210110, GNU ld (GNU Binutils for Debian) 2.35.2
>> > >>>>>>> userspace arch: riscv64
>> > >>>>>>>
>> > >>>>>>> IMPORTANT: if you fix the issue, please add the following tag to the commit:
>> > >>>>>>> Reported-by: syzbot+2c5da6... at syzkaller.appspotmail.com
>> > >>>>>> The CONFIG_KASAN_VMALLOC allows riscv kernel to boot, but now Go
>> > >>>>>> processes started crashing with:
>> > >>>>>>
>> > >>>>>> 1970/01/01 00:06:55 fuzzer started
>> > >>>>>> runtime: lfstack.push invalid packing: node=0xffffff5908a940 cnt=0x1
>> > >>>>>> packed=0xffff5908a9400001 -> node=0xffff5908a940
>> > >>>>>> fatal error: lfstack.push
>> > >>>>>> runtime stack:
>> > >>>>>> runtime.throw({0x30884c, 0xc})
>> > >>>>>> /usr/local/go/src/runtime/panic.go:1198 +0x60
>> > >>>>>> runtime.(*lfstack).push(0xdb3850, 0xffffff5908a940)
>> > >>>>>> /usr/local/go/src/runtime/lfstack.go:30 +0x1a8
>> > >>>>>>
>> > >>>>>> Go runtime tries to shove some data into the upper 16 bits of pointers
>> > >>>>>> assuming they are unused.
>> > >>>>>> However, the original pointer node=0xffffff5908a940 suggest riscv now
>> > >>>>>> has 56-bit users-space address space?
>> > >>>>>
>> > >>>>> Yes, sv57 was merged recently.
>> > >>>>>
>> > >>>>>> Documentation/riscv/vm-layout.rst claims 48-bit pointers:
>> > >>>>>> "
>> > >>>>>> The RISC-V privileged architecture document states that the 64bit addresses
>> > >>>>>> "must have bits 63–48 all equal to bit 47, or else a page-fault exception will
>> > >>>>>> occur.":
>> > >>>>>
>> > >>>>> Thanks for pointing that, I extracted that from the specification before sv57 was specified, I'll fix that.
>> > >>>>>
>> > >>>>> The current kernel code will use sv57 as it is supported and advertised by qemu, and to my knowledge, you can't downgrade to sv48 unless by re-compiling qemu using the following:
>> > >>>>>
>> > >>>>> diff --git a/target/riscv/csr.c b/target/riscv/csr.c
>> > >>>>> index 6dbe9b541f..a64b50ed75 100644
>> > >>>>> --- a/target/riscv/csr.c
>> > >>>>> +++ b/target/riscv/csr.c
>> > >>>>> @@ -637,7 +637,7 @@ static const char valid_vm_1_10_64[16] = {
>> > >>>>> [VM_1_10_MBARE] = 1,
>> > >>>>> [VM_1_10_SV39] = 1,
>> > >>>>> [VM_1_10_SV48] = 1,
>> > >>>>> - [VM_1_10_SV57] = 1
>> > >>>>> + [VM_1_10_SV57] = 0
>> > >>>>> };
>> > >>>>>
>> > >>>>> /* Machine Information Registers */
>> > >>>>>
>> > >>>>>> ...
>> > >>>>>> 0000000000000000 | 0 | 0000003fffffffff | 256 GB |
>> > >>>>>> user-space virtual memory, different per mm
>> > >>>>>> "
>> > >>>> There is no kernel config to force SV48/39, right?
>> > >>>
>> > >>> No, we rely on what the hardware advertises, if it supports sv57, we'll go for sv57, if not, we'll try sv48...etc. I had some patches to force the downgrade by using the device tree but they never got merged though.
>> > >> +original CC list
>> > >>
>> > >> FTR sent Go runtime change to support SV57:
>> > >> https://go-review.googlesource.com/c/go/+/409055
>> > >
>> > >
>> > > Is CONFIG_CMDLINE broken on riscv?
>> > > I am running with:
>> > >
>> > > CONFIG_CMDLINE="earlyprintk=serial net.ifnames=0
>> > > sysctl.kernel.hung_task_all_cpu_backtrace=1 ima_policy=tcb
>> > > nf-conntrack-ftp.ports=20000 nf-conntrack-tftp.ports=20000
>> > > nf-conntrack-sip.ports=20000 nf-conntrack-irc.ports=20000
>> > > nf-conntrack-sane.ports=20000 binder.debug_mask=0
>> > > rcupdate.rcu_expedited=1 no_hash_pointers page_owner=on
>> > > sysctl.vm.nr_hugepages=4 sysctl.vm.nr_overcommit_hugepages=4
>> > > secretmem.enable=1 sysctl.max_rcu_stall_to_panic=1
>> > > msr.allow_writes=off dummy_hcd.num=2 smp.csd_lock_timeout=300000
>> > > watchdog_thresh=165 workqueue.watchdog_thresh=420
>> > > sysctl.net.core.netdev_unregister_timeout_secs=420 panic_on_warn=1"
>> >
>> >
>> > This command line is 608-character long, but we are still stuck with the
>> > default COMMAND_LINE_SIZE to 512, I imagine that it is the problem. I
>> > had proposed a patch last year to bump that to 1024, but it never got
>> > merged
>> > https://lore.kernel.org/lkml/CAEn-LTqTXCEC=bXTvGyo8SNL0JMWRKtiSwQB7R=Pc4uhxZUruA@mail.gmail.com/T/#m4b45019dc0f5573f2a50c1f6007c5109fa35efff
>>
>>
>> risc-v maintainers, please merge it now.
>> I would even suggest 2048:
>>
>> git grep "define COMMAND_LINE_SIZE" arch/
>> arch/alpha/include/uapi/asm/setup.h:#define COMMAND_LINE_SIZE   256
>> arch/arc/include/asm/setup.h:#define COMMAND_LINE_SIZE 256
>> arch/arm/include/uapi/asm/setup.h:#define COMMAND_LINE_SIZE 1024
>> arch/arm64/include/uapi/asm/setup.h:#define COMMAND_LINE_SIZE   2048
>> arch/ia64/include/uapi/asm/setup.h:#define COMMAND_LINE_SIZE    2048
>> arch/m68k/include/uapi/asm/setup.h:#define COMMAND_LINE_SIZE 256
>> arch/microblaze/include/uapi/asm/setup.h:#define COMMAND_LINE_SIZE      256
>> arch/mips/include/uapi/asm/setup.h:#define COMMAND_LINE_SIZE    4096
>> arch/parisc/include/uapi/asm/setup.h:#define COMMAND_LINE_SIZE  1024
>> arch/powerpc/include/uapi/asm/setup.h:#define COMMAND_LINE_SIZE 2048
>> arch/s390/include/asm/setup.h:#define COMMAND_LINE_SIZE CONFIG_COMMAND_LINE_SIZE
>> arch/sparc/include/uapi/asm/setup.h:# define COMMAND_LINE_SIZE 2048
>> arch/sparc/include/uapi/asm/setup.h:# define COMMAND_LINE_SIZE 256
>> arch/um/include/asm/setup.h:#define COMMAND_LINE_SIZE 4096
>> arch/x86/include/asm/setup.h:#define COMMAND_LINE_SIZE 2048
>> arch/xtensa/include/uapi/asm/setup.h:#define COMMAND_LINE_SIZE  256
>>
>>
>> It's also interesting how the kernel handles overflow. Imagine one
>> adds that_critical_security_feature=1 to the end of an existing long
>> line.
>
>
> Your comment rang a bell and I searched in my old patchsets: I had submitted a patch [1] to output a warning in case of an overflow and to correctly truncate the command line to avoid such issues: it was taken with another series [2] which was actually never merged...My bad on this one, I followed my patch in the series but not the series itself.
>
> I'll try to re-submit it because I agree the current behaviour is really wrong.
>
> [1] https://lore.kernel.org/lkml/ee702ff7-f43c-745c-4157-b1cba53bb0b2@ghiti.fr/T/
> [2] https://lore.kernel.org/linux-devicetree/c3d52a6e1423d9d27c59ad7ab945929b09f74866.1617375802.git.christophe.leroy@csgroup.eu/T/

FTR I've merged the Go fix for SV57:
https://go-review.googlesource.com/c/go/+/409055
but it will only appear in Go 1.20.

And we still have the command line length issue for reviving syzbot testing.