Re: [PULL 0/1] Seabios 20191106 patches

[Gerd Hoffmann]

On Thu, Nov 07, 2019 at 11:56:03AM +0000, Peter Maydell wrote:
> On Wed, 6 Nov 2019 at 12:26, Gerd Hoffmann <address@hidden> wrote:
> >
> > The following changes since commit 36609b4fa36f0ac934874371874416f7533a5408:
> >
> >   Merge remote-tracking branch 
> > 'remotes/palmer/tags/palmer-for-master-4.2-sf1' into staging (2019-11-02 
> > 17:59:03 +0000)
> >
> > are available in the Git repository at:
> >
> >   git://git.kraxel.org/qemu tags/seabios-20191106-pull-request
> >
> > for you to fetch changes up to 58b16e57ded751e2e8be626124aad1d46a408a33:
> >
> >   seabios: update to pre-1.13 snapshot (2019-11-06 13:23:02 +0100)
> >
> > ----------------------------------------------------------------
> > seabios: update to pre-1.13 snapshot
> >
> > ----------------------------------------------------------------
> >
> > Gerd Hoffmann (1):
> >   seabios: update to pre-1.13 snapshot
> 
> Hi; this fails 'make check' on at least
> aarch64, aarch32, FreeBSD, NetBSD, s390:
> 
> ERROR:/home/linux1/qemu/tests/boot-sector.c:161:boot_sector_test:
> assertion failed (signature == SIGNATURE): (0x00000000 == 0x0000dead)
> ERROR - Bail out!
> ERROR:/home/linux1/qemu/tests/boot-sector.c:161:boot_sector_test:
> assertion failed (signature == SIGNATURE): (0x00000000 == 0x0000dead)
> PASS 9 bios-tables-test /x86_64/acpi/q35/bridge
> Aborted (core dumped)
> /home/linux1/qemu/tests/Makefile.include:916: recipe for target
> 'check-qtest-i386' failed
> make: *** [check-qtest-i386] Error 1
> 
> the x86-64 bootsector tests seem to fail similarly.

[ full quote for seabios list ]

Re-ran test on x86-64 box -> works.
Tried on aarch64 machine -> fails.

Given the arch list above this pretty much looks like it is tcg-related,
even though a quick check with "qemu -accel tcg -cdrom /some/live/iso"
(on x86_64) doesn't show any obvious problems.

Recompiled seabios with gcc 4.8 instead of gcc 8
  -> Works on both x86-64 and aarch64.
  -> I'll redo the pull request with that.

I'll go try find the root cause next week.  On a quick glance this
looks like a bug in tcg or gcc.  In case anyone has hints what might
have caused this drop me a note.

thanks,
  Gerd