Re: [Qemu-devel] [PATCH 1/2] elf-ops.h: Map into memory the ELF to load

[Paolo Bonzini <pbonzini@redhat.com>]

On 23/07/19 11:42, Stefano Garzarella wrote:
> "If writable is TRUE, the mapped buffer may be modified, otherwise it is an
> error to modify the mapped buffer. Modifications to the buffer are not visible
> to other processes mapping the same file, and are not written back to the file."
> 
> I don't know what "error" means, but reading the second part I thought
> the changes in that case were only visible at the current process.

My reading would be that the second part applies to the writable==TRUE
case.  In fact, the glib source code agrees:

      file->contents = (gchar *) mmap (NULL,  file->length,
                           writable ? PROT_READ|PROT_WRITE : PROT_READ,
                           MAP_PRIVATE, fd, 0);

meaning that we could after all just use writable == true.

Paolo

> I'll test it to understand better the behavior. If we can't touch it, then we
> have to make a copy in these cases.
> 
>> (We can't get out of this by just passing writable=true, because
>> we definitely don't want to be writing back to the underlying file.)
> Yes, I agree.