[Xen-devel] Re: [POWERPC/IA64] Updates required due to loader changes

[Horms <horms@xxxxxxxxxxxx>]

On Thu, Aug 24, 2006 at 08:12:02AM +0100, Ian Campbell wrote:
> On Thu, 2006-08-24 at 11:17 +0900, Horms wrote:
> > Sorry for being so naieve, but where is the staging tree?
> 
> It's an internal tree which gets regression tested and then copied
> automatically to the public xen-unstable tree.
> 
> The code in question is in the xen-unstable tree now.
> 
> > I included elf notes infastructure in the kexec patches that I have
> > posted to this list serveral times. It sounds like there is probably
> > some infastructure overlap. I'd like to take a look at what you have
> > done to see if it fits the needs of kexec so we can avoid code duplication.
> 
> I didn't know you guys had ELF infrastructure in the kexec patches, I
> hope my stuff is useful for you...

Hi Ian,

I took a look over the changes, and unforunately there doesn't seem
to be much overlap. This is for two main reasons:

  1. kexec is mainly concerned with crash notes, rather than
     generic notes (I was a bit confused when I read your original post).

  2. kexec creates these notes from within the hypervisor,
     but I beleive that your code has linux create the notes,
     while the hypervisor just reads them.

I'd really appreciate it if you could take a moment to cast your eyes
over my elf code. 

Below is the x86 code to fill in the crash notes, and below that
is include/xen/elfcore.h. This comprises the bulk of the elf code
that I have. I can come up with a full patch if you are interested.

The x86 fill-in code was basically copied from i386 linux kexec,
if there is room for improvement there, then its probably something
that also wants fixing in Linux.

The include/xen/elfcore.h. portion is based on linux, translated
to the xen style of elf code - which apparently does not have
the same origin as the linux code. I'd appreciate feedback on
the style here.

-- 
Horms
  H: http://www.vergenet.net/~horms/
  W: http://www.valinux.co.jp/en/

=== begin x86 fill-in snippet ===
static u32 *append_elf_note(u32 *buf, char *name, unsigned type, void *data,
                                                               size_t data_len)
{
        Elf_Note note;

        note.namesz = strlen(name) + 1;
        note.descsz = data_len;
        note.type   = type;
        memcpy(buf, &note, sizeof(note));
        buf += (sizeof(note) +3)/4;
        memcpy(buf, name, note.namesz);
        buf += (note.namesz + 3)/4;
        memcpy(buf, data, note.descsz);
        buf += (note.descsz + 3)/4;

        return buf;
}

static void final_note(u32 *buf)
{
        Elf_Note note;

        note.namesz = 0;
        note.descsz = 0;
        note.type   = 0;
        memcpy(buf, &note, sizeof(note));
}
=== end x86 fill-in snippet ===

=== begin include/xen/elfcore.h ===
/******************************************************************************
 * include/xen/elfcore.h
 * 
 * Created By: Horms
 *
 * Based heavily on include/linux/elfcore.h from Linux 2.6.16
 * Naming scheeme based on include/xen/elf.h (not include/linux/elfcore.h)
 *
 */

#ifndef __ELFCOREC_H__
#define __ELFCOREC_H__

#include <xen/types.h>
#include <xen/elf.h>
#include <public/xen.h>

#define NT_PRSTATUS     1
#define NT_XEN_DOM0_CR3 0x10000001 /* XXX: Hopefully this is unused,
                                           feel free to change to a 
                                           better/different value */

typedef struct
{
    int signo;                       /* signal number */
    int code;                        /* extra code */
    int errno;                       /* errno */
} ELF_Signifo;

/* These seem to be the same length on all architectures on Linux */
typedef int ELF_Pid;
typedef struct {
        long tv_sec;
        long tv_usec;
} ELF_Timeval;
typedef unsigned long ELF_Greg;
#define ELF_NGREG (sizeof (struct cpu_user_regs) / sizeof(ELF_Greg))
typedef ELF_Greg ELF_Gregset[ELF_NGREG];

/*
 * Definitions to generate Intel SVR4-like core files.
 * These mostly have the same names as the SVR4 types with "elf_"
 * tacked on the front to prevent clashes with linux definitions,
 * and the typedef forms have been avoided.  This is mostly like
 * the SVR4 structure, but more Linuxy, with things that Linux does
 * not support and which gdb doesn't really use excluded.
 */
typedef struct
{
    ELF_Signifo pr_info;         /* Info associated with signal */
    short pr_cursig;             /* Current signal */
    unsigned long pr_sigpend;    /* Set of pending signals */
    unsigned long pr_sighold;    /* Set of held signals */
    ELF_Pid pr_pid;
    ELF_Pid pr_ppid;
    ELF_Pid pr_pgrp;
    ELF_Pid pr_sid;
    ELF_Timeval pr_utime;        /* User time */
    ELF_Timeval pr_stime;        /* System time */
    ELF_Timeval pr_cutime;       /* Cumulative user time */
    ELF_Timeval pr_cstime;       /* Cumulative system time */
    ELF_Gregset pr_reg;          /* GP registers */
    int pr_fpvalid;              /* True if math co-processor being used.  */
} ELF_Prstatus;

#endif /* __ELFCOREC_H__ */

/*
 * Local variables:
 * mode: C
 * c-set-style: "BSD"
 * c-basic-offset: 4
 * tab-width: 4
 * indent-tabs-mode: nil
 * End:
 */
=== end include/xen/elfcore.h ===

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel