thr3ads.net - Syslinux - [syslinux] load_env32 WAS: Hyper-V Gen 2 waiting for ldlinux.e64 [Oct 2015]

If this information is useful, please help other people find it:
Share via:

Geert Stappers

2015-Oct-07 23:24 UTC

[syslinux] Hyper-V Gen 2 waiting for ldlinux.e64

On Wed, Oct 07, 2015 at 05:58:51PM -0500, Clements, James
wrote:> James Clement
> > Geert Stappers
> > > On Wed, Oct 07, 2015 at 05:06:41PM -0500, Clements, James wrote:
> > > > The screen displays the following:
> > > >
> > > > PXE Network Boot using IPv4
> > > > ....
> > > >    Station IP address is 192.168.205.50
> > > >    Server IP address is 192.168.205.1
> > > >    NBP filename is efi64/syslinux.efi
> > > >    NBP filesize is 200096 Bytes
> > > >    Downloading NBP file...
> > > >    Successfully downloaded NBP file.
> > > > Getting cached packet
> > > > My IP is 192.168.205.50
> > > >
> > >
> > > So nothing with "disable UseDefaultAddress"
> > > No signs of any repeating / any loops  on the console of the
booting VM.
> > >
> > > IIRC can
http://www.syslinux.org/wiki/index.php/Development/Debugging#Syslinux_Dynamic_Debugger
> > > only be used after ldlinux being loaded ...
> > >
> >
> > Nope, nothing like that. I'm not really sure how to proceed, but I
notice on
> > your IRC channel that you advertise a 'devel 6.04-preN'
verion. Is it worth me
> > trying that?
The binaries that Gene asked the to test in the begin of this thread
are the newest. That are fresher as 6.04-preN.
> Correct me if I am wrong but if I understand correctly, syslinux.efi is
> what drives the request for ldlinux.e64. Is there a way we can enable
> some debug somehow to see why the request for ldlinux.e64 includes
> the tsize option? Or even disable the tsize option (as it appears it
> may be causing the UEFI to hang)?
> --- a/efi/main.c
+++ b/efi/main.c
@@ -1371,7 +1371,9 @@ EFI_STATUS efi_main(EFI_HANDLE image, EFI_SYSTEM_TABLE
*table)
         */
        efi_setcwd(DevicePathToStr(info->FilePath));
 
+       Print(L"fs_init start\n");
        fs_init(ops, (void *)&priv);
+       Print(L"fs_init done.\n");
 
        /*
         * There may be pending user input that wasn't processed by
@@ -1383,8 +1385,10 @@ EFI_STATUS efi_main(EFI_HANDLE image, EFI_SYSTEM_TABLE
*table)
                status = uefi_call_wrapper(in->ReadKeyStroke, 2, in,
&key);
        } while (status == EFI_SUCCESS);
 
+       Print(L"load_env32 start\n");
        if (!setjmp(load_error_buf))
                load_env32(NULL);
+       Print(L"load_env32 done.\n");
 
        /* load_env32() failed.. cancel timer and bailout */
        status = cancel_timer(timer_ev);


Untested

Geert Stappers

2015-Oct-09 14:28 UTC

head link

[syslinux] load_env32 WAS: Hyper-V Gen 2 waiting for ldlinux.e64

On Thu, Oct 08, 2015 at 01:24:26AM +0200, Geert Stappers via Syslinux
wrote:> On Wed, Oct 07, 2015 at 05:58:51PM -0500, Clements, James wrote:
> 
> > Correct me if I am wrong but if I understand correctly, syslinux.efi
is
> > what drives the request for ldlinux.e64. Is there a way we can enable
> > some debug somehow to see why the request for ldlinux.e64 includes
> > the tsize option? Or even disable the tsize option (as it appears it
> > may be causing the UEFI to hang)?
> > 
> --- a/efi/main.c
> +++ b/efi/main.c
> @@ -1371,7 +1371,9 @@ EFI_STATUS efi_main(EFI_HANDLE image,
EFI_SYSTEM_TABLE *table)
>          */
>         efi_setcwd(DevicePathToStr(info->FilePath));
>  
> +       Print(L"fs_init start\n");
>         fs_init(ops, (void *)&priv);
> +       Print(L"fs_init done.\n");
>  
>         /*
>          * There may be pending user input that wasn't processed by
> @@ -1383,8 +1385,10 @@ EFI_STATUS efi_main(EFI_HANDLE image,
EFI_SYSTEM_TABLE *table)
>                 status = uefi_call_wrapper(in->ReadKeyStroke, 2, in,
&key);
>         } while (status == EFI_SUCCESS);
>  
> +       Print(L"load_env32 start\n");
>         if (!setjmp(load_error_buf))
>                 load_env32(NULL);
> +       Print(L"load_env32 done.\n");
>  
>         /* load_env32() failed.. cancel timer and bailout */
>         status = cancel_timer(timer_ev);
> 
> 
> Untested
(James did)

It dies in load_env32.

$ git grep load_env
core/diskstart.inc:             pm_call load_env32
core/elflink/load_env32.c:void load_env32(com32sys_t * regs __unused)
core/extern.inc:        ; elflink/load_env32.c
core/extern.inc:        extern load_env32, pm_env32_run
core/include/core.h:extern void load_env32(com32sys_t *regs);
core/isolinux.asm:              pm_call load_env32
core/isolinux.asm:              jmp kaboom              ; load_env32() should
never return. If
core/pxelinux.asm:              pm_call load_env32
core/pxelinux.asm:              jmp kaboom              ; load_env32()
shouldn't return. If it does, then kaboom!
efi/main.c:     Print(L"load_env32 start\n");
efi/main.c:             load_env32(NULL);
efi/main.c:     Print(L"load_env32 done.\n");
efi/main.c:     /* load_env32() failed.. cancel timer and bailout */

In which source file should the next debug print statement go?

Groeten
Geert Stappers
-- 
Leven en laten leven

Geert Stappers

2015-Oct-26 19:54 UTC

head link

[syslinux] load_env32

On Fri, Oct 09, 2015 at 04:28:58PM +0200, Geert Stappers via Syslinux
wrote:> On Thu, Oct 08, 2015 at 01:24:26AM +0200, Geert Stappers via Syslinux
wrote:
> > --- a/efi/main.c
> > +++ b/efi/main.c
> > @@ -1371,7 +1371,9 @@ EFI_STATUS efi_main(EFI_HANDLE image,
EFI_SYSTEM_TABLE *table)
> >          */
> >         efi_setcwd(DevicePathToStr(info->FilePath));
> >  
> > +       Print(L"fs_init start\n");
> >         fs_init(ops, (void *)&priv);
> > +       Print(L"fs_init done.\n");
> >  
> >         /*
> >          * There may be pending user input that wasn't processed
by
> > @@ -1383,8 +1385,10 @@ EFI_STATUS efi_main(EFI_HANDLE image,
EFI_SYSTEM_TABLE *table)
> >                 status = uefi_call_wrapper(in->ReadKeyStroke, 2,
in, &key);
> >         } while (status == EFI_SUCCESS);
> >  
> > +       Print(L"load_env32 start\n");
> >         if (!setjmp(load_error_buf))
> >                 load_env32(NULL);
> > +       Print(L"load_env32 done.\n");
> >  
> >         /* load_env32() failed.. cancel timer and bailout */
> >         status = cancel_timer(timer_ev);
> > 
> > 
> > Untested
> 
> (James did)
> 
> It dies in load_env32.
> 
> $ git grep load_env
> core/diskstart.inc:             pm_call load_env32
> core/elflink/load_env32.c:void load_env32(com32sys_t * regs __unused)
> core/extern.inc:        ; elflink/load_env32.c
> core/extern.inc:        extern load_env32, pm_env32_run
> core/include/core.h:extern void load_env32(com32sys_t *regs);
> core/isolinux.asm:              pm_call load_env32
> core/isolinux.asm:              jmp kaboom              ; load_env32()
should never return. If
> core/pxelinux.asm:              pm_call load_env32
> core/pxelinux.asm:              jmp kaboom              ; load_env32()
shouldn't return. If it does, then kaboom!
> efi/main.c:     Print(L"load_env32 start\n");
> efi/main.c:             load_env32(NULL);
> efi/main.c:     Print(L"load_env32 done.\n");
> efi/main.c:     /* load_env32() failed.. cancel timer and bailout */
> 
> 
> In which source file should the next debug print statement go?
--- a/core/elflink/load_env32.c
+++ b/core/elflink/load_env32.c
@@ -119,6 +119,7 @@ void load_env32(com32sys_t * regs __unused)
                NULL
        };
 
+        printf("Starting %s elf module subsystem...\n", ELF_MOD_SYS);
        dprintf("Starting %s elf module subsystem...\n", ELF_MOD_SYS);
 
        if (strlen(CurrentDirName) && !path_add(CurrentDirName)) {


Didn't show up in testresult posted
at http://www.syslinux.org/archives/2015-October/024433.html

Further advice is appreciated


Groeten
Geert Stappers
-- 
Leven en laten leven

Possibly Parallel Threads

Search for more possibly parallel threads

Syslinux - Oct 2015 - load_env32 WAS: Hyper-V Gen 2 waiting for ldlinux.e64

[syslinux] Hyper-V Gen 2 waiting for ldlinux.e64

[syslinux] load_env32 WAS: Hyper-V Gen 2 waiting for ldlinux.e64

[syslinux] load_env32

Possibly Parallel Threads