Commit Graph

126 Commits

Author SHA1 Message Date
Anthony Liguori
7267c0947d Use glib memory allocation and free functions
qemu_malloc/qemu_free no longer exist after this commit.

Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
2011-08-20 23:01:08 -05:00
Artyom Tarasenko
cf973e469b set ELF_HWCAP for SPARC and SPARC64
setting ELF_HWCAP fixes dynamic library loading for Linux/sparc64
This patch allows loading busybox from Debian 6 initrd

Signed-off-by: Artyom Tarasenko <atar4qemu@gmail.com>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2011-07-14 17:41:29 +00:00
Laurent ALFONSI
14322bad88 linux-user: Define AT_RANDOM to support target stack protection mechanism.
The dynamic linker from the GNU C library v2.10+ uses the ELF
auxiliary vector AT_RANDOM [1] as a pointer to 16 bytes with random
values to initialize the stack protection mechanism.  Technically the
emulated GNU dynamic linker crashes due to a NULL pointer
derefencement if it is built with stack protection enabled and if
AT_RANDOM is not defined by the QEMU ELF loader.

[1] This ELF auxiliary vector was introduced in Linux v2.6.29.

This patch can be tested with the code above:

    #include <elf.h>       /* Elf*_auxv_t, AT_RANDOM, */
    #include <stdio.h>     /* printf(3), */
    #include <stdlib.h>    /* exit(3), EXIT_*, */
    #include <stdint.h>    /* uint8_t, */
    #include <string.h>    /* memcpy(3), */

    #if defined(__LP64__) || defined(__ILP64__) || defined(__LLP64__)
    #    define Elf_auxv_t Elf64_auxv_t
    #else
    #    define Elf_auxv_t Elf32_auxv_t
    #endif

    main(int argc, char* argv[], char* envp[])
    {
        Elf_auxv_t *auxv;

        /* *envp = NULL marks end of envp. */
        while (*envp++ != NULL);

        /* auxv->a_type = AT_NULL marks the end of auxv. */
        for (auxv = (Elf_auxv_t *)envp; auxv->a_type != AT_NULL; auxv++) {
            if (auxv->a_type == AT_RANDOM) {
                int i;
                uint8_t rand_bytes[16];

                printf("AT_RANDOM is: 0x%x\n", auxv->a_un.a_val);
                memcpy(rand_bytes, (const uint8_t *)auxv->a_un.a_val, sizeof(rand_bytes));
                printf("it points to: ");
                for (i = 0; i < 16; i++) {
                    printf("0x%02x ", rand_bytes[i]);
                }
                printf("\n");
                exit(EXIT_SUCCESS);
            }
        }
        exit(EXIT_FAILURE);
    }

Changes introduced in v2 and v3:

    * Fix typos + thinko (AT_RANDOM is used for stack canary, not for
      ASLR)

    * AT_RANDOM points to 16 random bytes stored inside the user
      stack.

    * Add a small test program.

Signed-off-by: Cédric VINCENT <cedric.vincent@st.com>
Signed-off-by: Laurent ALFONSI <laurent.alfonsi@st.com>
Signed-off-by: Riku Voipio <riku.voipio@iki.fi>
2011-06-21 20:30:09 +03:00
Peter Maydell
f3ed1f5d47 linux-user: Handle images where lowest vaddr is not page aligned
Fix a bug in the linux-user ELF loader code where it was not correctly
handling images where the lowest vaddr to be loaded was not page aligned.
The problem was that the code to probe for a suitable guest base address
was changing the 'loaddr' variable (by rounding it to a page boundary),
which meant that the load bias would then be incorrectly calculated
unless loaddr happened to already be page-aligned.

Binaries generated by gcc with the default linker script do start with
a loadable segment at a page-aligned vaddr, so were unaffected. This
bug was noticed with a binary created by the Google Go toolchain for ARM.

We fix the bug by refactoring the "probe for guest base" code out into
its own self-contained function.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Riku Voipio <riku.voipio@iki.fi>
2011-06-21 20:29:01 +03:00
Cédric VINCENT
b947527941 linux-user: Fix the load of ELF files that have no "useful" symbol
This patch fixes a "double free()" due to "realloc(syms, 0)" in the
loader when the ELF file has no "useful" symbol, as with the following
example (compiled with "sh4-linux-gcc -nostdlib"):

    .text
    .align 1
    .global _start
    _start:
        mov     #1, r3
        trapa   #40     // syscall(__NR_exit)
        nop

The bug appears when the log (option "-d") is enabled.

Signed-off-by: Cédric VINCENT <cedric.vincent@st.com>
Signed-off-by: Yves JANIN <yves.janin@st.com>
Signed-off-by: Riku Voipio <riku.voipio@iki.fi>
Reviewed-by: Richard Henderson <rth@twiddle.net>
2011-06-21 20:28:38 +03:00
Ulrich Hecht
a4c075f178 s390x: s390x-linux-user support
This patch adds support for running s390x binaries in the linux-user emulation
code.

Signed-off-by: Ulrich Hecht <uli@suse.de>
Signed-off-by: Alexander Graf <agraf@suse.de>
2011-05-20 17:35:12 +02:00
Guan Xuetao
d2fbca9422 unicore32: necessary modifications for other files to support unicore32
Signed-off-by: Guan Xuetao <gxt@mprc.pku.edu.cn>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2011-04-12 18:49:05 +00:00
Laurent Vivier
80f5ce758a linux-user: correct core dump format
This patch allows to really use the core dumped by qemu with guest
architecture tools.

- it adds a missing bswap_phdr() for the program headers
  of memory regions.

  "objdump -x" sample:

BEFORE:

0x1000000 off    0x00200000 vaddr 0x00000400 paddr 0x00000000 align 2**21
         filesz 0x00000000 memsz 0x00100000 flags ---
0x1000000 off    0x00200000 vaddr 0x00100400 paddr 0x00000000 align 2**21
         filesz 0x00000000 memsz 0x00080000 flags --- 6000000

AFTER:

    LOAD off    0x00002000 vaddr 0x00040000 paddr 0x00000000 align 2**13
         filesz 0x00000000 memsz 0x00001000 flags ---
    LOAD off    0x00002000 vaddr 0x00041000 paddr 0x00000000 align 2**13
         filesz 0x00000000 memsz 0x00000800 flags rw-

- it doesn't pad the note size to sizeof(int32_t).
  On m68k the NT_PRSTATUS note size is 154 and
  must not be rounded up to 156, because this value is checked by
  objdump and gdb.

  "gdb" symptoms:

      "warning: Couldn't find general-purpose registers in core file."

  "objdump -x" sample:

BEFORE:

Sections:
Idx Name          Size      VMA       LMA       File off  Algn
  0 note0         000001c4  00000000  00000000  000003b4  2**0
                  CONTENTS, READONLY
  1 .auxv         00000070  00000000  00000000  00000508  2**2
                  CONTENTS
  2 proc1         00100000  00000400  00000000  00200000  2**10
                  READONLY

AFTER:

Sections:
Idx Name          Size      VMA       LMA       File off  Algn
  0 note0         000001c4  00000000  00000000  000003b4  2**0
                  CONTENTS, READONLY
  1 .reg/19022    00000050  00000000  00000000  0000040e  2**2
                  CONTENTS
  2 .reg          00000050  00000000  00000000  0000040e  2**2
                  CONTENTS
  3 .auxv         00000070  00000000  00000000  00000508  2**2
                  CONTENTS
  4 load1         00000000  00040000  00000000  00002000  2**13
                  ALLOC, READONLY

Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Signed-off-by: Riku Voipio <riku.voipio@nokia.com>
2011-02-17 11:46:34 +02:00
Mike Frysinger
1af02e83c0 linux-user/elfload: add FDPIC support
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Riku Voipio <riku.voipio@nokia.com>
2011-02-09 10:33:54 +02:00
Stefan Weil
8d79de6e42 linux-user: Fix possible realloc memory leak
Extract from "man realloc":
"If realloc() fails the original block is left untouched;
it is not freed or moved."

Fix a possible memory leak (reported by cppcheck).

Cc: Riku Voipio <riku.voipio@iki.fi>
Signed-off-by: Stefan Weil <weil@mail.berlios.de>
Signed-off-by: Riku Voipio <riku.voipio@nokia.com>
2011-02-09 10:33:53 +02:00
Richard Henderson
5d5c99300d linux-user: Protect against allocation failure in load_symbols.
Cc: malc <av1474@comtv.ru>
Signed-off-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: malc <av1474@comtv.ru>
2010-07-29 20:54:35 +04:00
Richard Henderson
bf858897b7 linux-user: Re-use load_elf_image for the main binary.
This requires moving the PT_INTERP extraction and GUEST_BASE
handling into load_elf_image.  Key this off a non-null pointer
argument to receive the interpreter name.

Signed-off-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@gmail.com>
2010-07-29 08:32:28 +02:00
Richard Henderson
8e62a71738 linux-user: Extract load_elf_image from load_elf_interp.
Moving toward a single copy of the elf binary loading code.
Fill in the details of the loaded image into a struct image_info.

Adjust create_elf_tables to read from such structures instead
of from a collection of passed arguments.  Don't return error
values from load_elf_interp; always exit(-1) with a message to
stderr.  Collect elf_interpreter handling in load_elf_binary
to a common spot.

Signed-off-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@gmail.com>
2010-07-29 08:32:28 +02:00
Richard Henderson
b9329d4b53 linux-user: Remove partial support for a.out interpreters.
At the bottom of the a.out support was the unimplemented load_aout_interp
function.  There were other portions of the support that didn't look
right; when I went to look in the Linux kernel for clarification, I found
that the support for such interpreters has been removed from binfmt_elf.
There doesn't seem to be any reason to keep this broken support in qemu.

Signed-off-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@gmail.com>
2010-07-29 08:32:28 +02:00
Richard Henderson
60dcbcb5b1 linux-user: Put the stack guard page at the top.
There are no supported stack-grows-up targets.  We were putting
the guard page at the highest address, i.e. the bottom of the stack.
Use the maximum of host and guest page size for the guard size.

Signed-off-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@gmail.com>
2010-07-29 08:32:28 +02:00
Richard Henderson
9058abdd18 linux-user: Improve consistency checking in elf headers.
Validate more fields of the elf header.  Extract those checks
into two common functions to be used in both load_elf_interp
and load_elf_binary.

Signed-off-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@gmail.com>
2010-07-29 08:32:28 +02:00
Richard Henderson
682674b841 linux-user: Load symbols from the interpreter.
First, adjust load_symbols to accept a load_bias parameter.  At the same
time, read the entire section header table in one go, use pread instead
f lseek+read for the symbol and string tables, and properly free
allocated structures on error exit paths.

Second, adjust load_elf_interp to compute load_bias.  This requires
finding out the built-in load addresses.  Which allows us to honor a
pre-linked interpreter image when possible, and eliminate the hard-coded
INTERP_MAP_SIZE value.

Signed-off-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@gmail.com>
2010-07-29 08:32:28 +02:00
Richard Henderson
991f8f0c91 linux-user: Clean up byte-swapping in elfload.c.
Remove ifdefs from code by defining empty inline functions
when byte swapping isn't needed.  Push loops over swapping
arrays of structures into the swapping functions.

Signed-off-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@gmail.com>
2010-07-29 08:32:28 +02:00
Richard Henderson
2849023112 linux-user: Define ELF_DATA generically.
The only consideration on this value is the target endianness.
The existing defines were incorrect for alpha and sh4eb.

Signed-off-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@gmail.com>
2010-07-29 08:32:28 +02:00
Richard Henderson
9955ffac9a linux-user: Reduce lseek+reads while loading elf files.
Define BPRM_BUF_SIZE to 1k and read that amount initially.  If the
data we want from the binary is in this buffer, use it instead of
reading from the file again.

Signed-off-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@gmail.com>
2010-07-29 08:32:28 +02:00
Richard Henderson
d97ef72eed linux-user: Reindent elfload.c.
Signed-off-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@gmail.com>
2010-07-29 08:32:27 +02:00
Richard Henderson
e167d46c7a Add more DT_* and AT_* constants to qemu's copy of elf.h.
Moving some PPC AT_* constants from elfload.c at the same time.

Signed-off-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@gmail.com>
2010-07-29 08:32:27 +02:00
Richard Henderson
cf129f3a8e linux-user: Handle filesz < memsz for any PT_LOAD segment.
I caught padzero not properly initializing the .bss segment
on a statically linked Alpha program.  Rather than a minimal
patch, replace the gross code with a single mmap+memset.

Share more code between load_elf_interp and load_elf_binary.

Legally, an ELF program need not have just a single .bss;
and PT_LOAD segment can have memsz > filesz.

Signed-off-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@gmail.com>
2010-07-29 08:32:27 +02:00
Paul Brook
97374d3858 Usermode exec-stack fix
When loading a shared library that requires an executable stack,
glibc uses the mprotext PROT_GROWSDOWN flag to achieve this.
We don't support PROT_GROWSDOWN.
Add a special case to handle changing the stack permissions in this way.

Signed-off-by: Paul Brook <paul@codesourcery.com>
2010-06-16 13:03:51 +01:00
Paul Brook
68a1c81686 Pre-allocate guest address space
Allow pre-allocation of the guest virtual address space in usermode emulation.

Signed-off-by: Paul Brook <paul@codesourcery.com>
2010-05-29 02:27:35 +01:00
Edgar E. Iglesias
0d5d469938 microblaze: Update elf machine nums.
189 was allocated in upstream binutils.

0xbaab was the old temporary value. Still used by some tools and the
linux kernel.

I've seen 115 in older gdb versions, but lets ignore that one.

Signed-off-by: Edgar E. Iglesias <edgar.iglesias@gmail.com>
2010-05-19 15:24:17 +02:00
Edgar E. Iglesias
e4cbd44d8e microblaze: Add linux-user core dumping support.
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@gmail.com>
2010-05-19 15:09:28 +02:00
Paul Brook
c581deda32 Better default guest_base
Avoid loading linux-user applications in address space already used/reserved
by the host.

Signed-off-by: Paul Brook <paul@codesourcery.com>
2010-05-05 16:32:59 +01:00
Paul Brook
2e9a5713f0 Remove PAGE_RESERVED
The usermode PAGE_RESERVED code is not required by the current mmap
implementation, and is already broken when guest_base != 0.
Unfortunately the bsd emulation still uses the old mmap implementation,
so we can't rip it out altogether.

Signed-off-by: Paul Brook <paul@codesourcery.com>
2010-05-05 16:32:59 +01:00
Rob Landley
7983f435e5 linux-user/ppc: use the Linux register layout
The dynamic linker converts the Linux layout to the AIX layout and is
reentrant so it won't do it a second time if it's already been
converted. In short it work just fine with either register layout.

OTOH, statically linked binaries expect a Linux layout.

Remove code converting the Linux layout to AIX layout so that all
binaries are presented the Linux Layout.

Signed-off-by: Rob Landley <rob@landley.net>
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
2010-03-28 16:52:27 +02:00
Richard Henderson
703e0e89c6 linux-user: Use RLIMIT_STACK for default stack size.
The current default stack limit of 512kB is far too small; a fair
number of gcc testsuite failures (for all guests) are directly
attributable to this.  Using the -s option in every invocation of
the emulator is annoying to say the least.

A reasonable compromise seems to be to honor the system rlimit.
At least on two Linux distributions, this is set to 8MB and 10MB
respectively.  If the system does not limit the stack, then we're
no worse off than before.

At the same time, rename the variable from x86_stack_size and
change the ultimate fallback size from 512kB to 8MB.

Signed-off-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
2010-03-27 14:16:51 +01:00
Paul Brook
b480d9b74d Fix usermode virtual address type
Usermode virtual addresses are abi_ulong, not target_ulong.

Signed-off-by: Paul Brook <paul@codesourcery.com>
2010-03-12 23:25:52 +00:00
Paul Brook
9f9f030928 Disassembler symbol lookup fix
Fix function signature for userspace disassembler symbol lookup.

Signed-off-by: Paul Brook <paul@codesourcery.com>
2010-03-01 04:39:50 +00:00
Mike Frysinger
0ddbc96ec1 fix linux-user microblaze ELF_ARCH definition
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Edgar E. Iglesias <edgar.iglesias@gmail.com>
2010-01-17 09:51:03 +01:00
Nathan Froyd
7631c97ec9 linux-user: add core dump support for SH
Signed-off-by: Nathan Froyd <froydnj@codesourcery.com>
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
2009-12-19 20:29:18 +01:00
Nathan Froyd
7a93cc55e9 linux-user: add core dump support for M68K
Signed-off-by: Nathan Froyd <froydnj@codesourcery.com>
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
2009-12-19 20:29:18 +01:00
Nathan Froyd
51e526063d linux-user: add core dump support for MIPS
Signed-off-by: Nathan Froyd <froydnj@codesourcery.com>
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
2009-12-19 20:29:18 +01:00
Nathan Froyd
e2f3e741b4 linux-user: add core dump support for PPC
Signed-off-by: Nathan Froyd <froydnj@codesourcery.com>
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
2009-12-19 20:29:18 +01:00
Nathan Froyd
d049e626ad linux-user: fix ARM core dumps on opposite-endian hosts
Signed-off-by: Nathan Froyd <froydnj@codesourcery.com>
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
2009-12-19 20:29:18 +01:00
Nathan Froyd
21e807fadb linux-user: commonify definitions of target typedefs
There's no sense in separately declaring target_{elf_greg,uid,gid,pid}_t
for every architecture.  Just declare them once with appropriate
USE_UID16 handling.

Signed-off-by: Nathan Froyd <froydnj@codesourcery.com>
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
2009-12-19 20:29:18 +01:00
Nathan Froyd
dd0a365144 linux-user: fix ELF_USE_CORE_DUMP/USE_ELF_CORE_DUMP confusion
Signed-off-by: Nathan Froyd <froydnj@codesourcery.com>
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
2009-12-19 20:29:18 +01:00
Aurelien Jarno
49995e1748 linux-user: use TARGET_ABI_FMT_lx to print abi_ulong types
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
2009-12-19 20:29:18 +01:00
Richard Henderson
8a57a6c740 target-alpha: Drop bogus UNIQ initial value on Linux.
Signed-off-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
2009-12-13 20:32:36 +01:00
Riku Voipio
cf6de34aec linux-user: Update ARM hwcaps
Update ARM hwcaps to match Linux kernel 2.6.31 state

Signed-off-by: Riku Voipio <riku.voipio@iki.fi>
2009-10-16 14:37:33 +03:00
Paul Bolle
3bc0bdcaad linux-user: fix "#if 0"'d printf()
Make an "#if 0"'d printf() in load_elf_binary(), probably left to aid in
debugging, reflect what the actual code does. The current printf() will
only confuse those who "#if 1" it (it certainly confused me enough to
write this trivial patch).

Signed-off-by: Paul Bolle <pebolle@tiscali.nl>
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
2009-10-04 13:24:45 +02:00
Anthony Liguori
c227f0995e Revert "Get rid of _t suffix"
In the very least, a change like this requires discussion on the list.

The naming convention is goofy and it causes a massive merge problem.  Something
like this _must_ be presented on the list first so people can provide input
and cope with it.

This reverts commit 99a0949b72.

Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
2009-10-01 16:12:16 -05:00
malc
99a0949b72 Get rid of _t suffix
Some not so obvious bits, slirp and Xen were left alone for the time
being.

Signed-off-by: malc <av1474@comtv.ru>
2009-10-01 22:45:02 +04:00
Michael S. Tsirkin
6ece4df63b elfload: fix coding style nit
Put space between = and * when dereferencing a pointer,
to avoid confusion with old-style "*="

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2009-09-30 18:45:50 +00:00
Blue Swirl
ca20cf32ab Compile loader only once
Callers must pass ELF machine, byte swapping and symbol LSB clearing
information to ELF loader. A.out loader needs page size information, pass
that too as a parameter.

Extract prototypes to a separate file. Move loader.[ch] and elf_ops.h under hw.

Adjust callers. Also use target_phys_addr_t instead of target_ulong for
addresses: loader addresses aren't virtual.

Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2009-09-20 14:58:02 +00:00
Blue Swirl
72cf2d4f0e Fix sys-queue.h conflict for good
Problem: Our file sys-queue.h is a copy of the BSD file, but there are
some additions and it's not entirely compatible. Because of that, there have
been conflicts with system headers on BSD systems. Some hacks have been
introduced in the commits 15cc923584,
f40d753718,
96555a96d7 and
3990d09adf but the fixes were fragile.

Solution: Avoid the conflict entirely by renaming the functions and the
file. Revert the previous hacks.

Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2009-09-12 07:36:22 +00:00