Re: PATCH: Add x32_corenote.c

[Mark Wielaard <mjw at redhat dot com>]

Hi,

On Mon, Mar 30, 2015 at 05:27:30AM -0700, H.J. Lu wrote:
> On Thu, Mar 26, 2015 at 9:36 AM, H.J. Lu <hjl.tools@gmail.com> wrote:
> > Here are 5 x32 patches  I put a line:
> >
> > # Created by H.J. Lu <hjl.tools@gmail.com>, 2015.

Lets just use Copyright (C) H.J. Lu <hjl.tools@gmail.com>, 2015.

Also the patches still contain:
-   Copyright (C) 2008-2010, 2013 Red Hat, Inc.
+   Copyright (C) 2008-2010, 2013, 2015 Red Hat, Inc.

That is not really factual correct. If you modify a file just add
your own Copyright line instead.

> > in the new file I copied from the existing one.
> 
> I updated my x32 patch set to 7 patches.  They replaced the old 5
> patch set.

The patches look good. Thanks.
BTW I don't mind the ChangeLog entries being in the patch itself.
Whatever is more convenient for you.

One question about the new run-readelf-mixed-corenote.sh test.

> +# To reproduce this core dump, do this on x86_64 machine with Linux
> +# 3.7 or later:
> +# $ gcc -mx32 -x c <(echo 'int main () { return *(int *)0x12345678; }')
> +# $ ./a.out
> +testfiles testfile-x32-core
> +testrun_compare ${abs_top_builddir}/src/readelf -n testfile-x32-core <<\EOF
> +
> +Note segment of 2548 bytes at offset 0x234:
> +  Owner          Data size  Type
> +  CORE                 296  PRSTATUS
> +    info.si_signo: 11, info.si_code: 0, info.si_errno: 0, cursig: 11
> +    sigpend: <>
> +    sighold: <33,35,38-40,42,44-45>
> +    pid: 6885, ppid: 2792, pgrp: 6885, sid: 2792
> +    utime: 0.4294967296000, stime: 4294967296000.000000, cutime: 0.000000
> +    cstime: 0.000000

Are these timeval fields correct?

> +    orig_rax: -1, fpvalid: 1
> +    r15:                       0  r14:                       0
> +    r13:              4290830656  r12:                 4194960
> +    rbp:      0x00000000ffc0e070  rbx:                       0
> +    r11:              4145779200  r10:                       0
> +    r9:               4149627024  r8:               4149551744
> +    rax:               305419896  rcx:                 4195216
> +    rdx:              4290830668  rsi:              4290830660
> +    rdi:                       1  rip:      0x0000000000400380
> +    rflags:   0x0000000000010246  rsp:      0x00000000ffc0e070
> +    fs.base:   0x00000000f7754700  gs.base:   0x0000000000000000
> +    cs: 0x0033  ss: 0x002b  ds: 0x002b  es: 0x002b  fs: 0x0063  gs: 0x0000
> +  CORE                 124  PRPSINFO
> +    state: 0, sname: R, zomb: 0, nice: 0, flag: 0x03e803e800406600
> +    uid: 65537000, gid: 451216360, pid: 6885, ppid: 2792, pgrp: 6885
> +    sid: 2792

The uid and gid seem odd given that in the AUXV we see them being 1000:

> +  CORE                 152  AUXV
> +    SYSINFO_EHDR: 0xffd49000
> +    HWCAP: 0xbfebfbff  <fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe>
> +    PAGESZ: 4096
> +    CLKTCK: 100
> +    PHDR: 0x400034
> +    PHENT: 32
> +    PHNUM: 8
> +    BASE: 0xf7555000
> +    FLAGS: 0
> +    ENTRY: 0x400290
> +    UID: 1000
> +    EUID: 1000
> +    GID: 1000
> +    EGID: 1000
> +    SECURE: 0
> +    RANDOM: 0xffc0e2cb
> +    EXECFN: 0xffc0fff0
> +    PLATFORM: 0xffc0e2db
> +    NULL

Everything else looks good. Nice to have extra testcases covering x32.

Thanks,

Mark