Re: [PATCH v3] Fix SEGFAULT on getting physical address of MMIO region.

[Jonathan Cameron via <qemu-devel@nongnu.org>]

On Thu, 10 Aug 2023 14:18:50 +0100
Jonathan Cameron via <qemu-devel@nongnu.org> wrote:

> On Fri, 4 Aug 2023 09:15:13 -0700
> Richard Henderson <richard.henderson@linaro.org> wrote:
> 
> > On 8/4/23 04:09, Mikhail Tyutin wrote:  
> > > Apply save_iotlb_data() to io_readx() as well as to io_writex(). This fixes
> > > SEGFAULT on qemu_plugin_hwaddr_phys_addr() call plugins for addresses inside of
> > > MMIO region.
> > > 
> > > Signed-off-by: Dmitriy Solovev<d.solovev@yadro.com>
> > > Signed-off-by: Mikhail Tyutin<m.tyutin@yadro.com>
> > > Reviewed-by: Richard Henderson<richard.henderson@linaro.org>
> > > ---
> > > v3:
> > > - commit wording update
> > > 
> > > 
> > >   accel/tcg/cputlb.c | 36 +++++++++++++++++++++---------------
> > >   1 file changed, 21 insertions(+), 15 deletions(-)    
> > 
> > Queued to tcg-next.  
> 
> I'm not 100% sure it's related but with qemu/master as of a few hours ago
> I'm getting a segfault with backtrace (can't copy from relevant machine)
> indicating that
> io_writex () is calling memory_region_dispatch_write() and getting
> a segfault on
> 
> if (mr->alias)
> 
> mr has very suspect content of 0x13001300130013
> 
> The segfault is intermittent (otherwise I get hang later in boot) and
> I can only get it to trigger with multiple cpus.
> 
> x86 max cpu and q35 machine.
Resolved by
https://gitlab.com/rth7680/qemu/-/commit/78e096b365326dd2b287e914b393d914dbae3b66
softmmu: Use async_run_on_cpu in tcg_commit

Thanks Richard.



> 
> Jonathan
> 
> 
> > 
> > 
> > r~
> >   
> 
> 
>