| Message ID | 1372438702-20491-8-git-send-email-pbonzini@redhat.com |
|---|---|
| State | New |
| Headers | show |
On 2013-06-28 18:58, Paolo Bonzini wrote: > This function is not used outside the iothread mutex, so it > can use ram_list.mru_block. Add a comment to qemu_ram_addr_from_host to document this requirement? > > Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> > --- > exec.c | 12 ++++++++++-- > 1 file changed, 10 insertions(+), 2 deletions(-) > > diff --git a/exec.c b/exec.c > index 8722420..f3d7b18 100644 > --- a/exec.c > +++ b/exec.c > @@ -1408,18 +1408,26 @@ int qemu_ram_addr_from_host(void *ptr, ram_addr_t *ram_addr) > return 0; > } > > + block = ram_list.mru_block; > + if (block && block->host && host - block->host < block->length) { > + goto found; > + } > + > QTAILQ_FOREACH(block, &ram_list.blocks, next) { > /* This case append when the block is not mapped. */ > if (block->host == NULL) { > continue; > } > if (host - block->host < block->length) { > - *ram_addr = block->offset + (host - block->host); > - return 0; > + goto found; > } > } > > return -1; > + > +found: > + *ram_addr = block->offset + (host - block->host); > + return 0; > } > > /* Some of the softmmu routines need to translate from a host pointer > Beside the nit: Reviewed-by: Jan Kiszka <jan.kiszka@siemens.com> Jan
Il 01/07/2013 20:33, Jan Kiszka ha scritto: > > This function is not used outside the iothread mutex, so it > > can use ram_list.mru_block. > > Add a comment to qemu_ram_addr_from_host to document this requirement? Right now there is hardly any documentation of what does _not_ require the iothread mutex... basically everything except qemu_safe_ram_ptr and qemu_ram_ptr_length requires it. Paolo
On 2013-07-01 22:48, Paolo Bonzini wrote: > Il 01/07/2013 20:33, Jan Kiszka ha scritto: >>> This function is not used outside the iothread mutex, so it >>> can use ram_list.mru_block. >> >> Add a comment to qemu_ram_addr_from_host to document this requirement? > > Right now there is hardly any documentation of what does _not_ require > the iothread mutex... basically everything except qemu_safe_ram_ptr and > qemu_ram_ptr_length requires it. I don't disagree regarding the current state. But that doesn't imply it has to be preserved. Jan
Il 02/07/2013 09:09, Jan Kiszka ha scritto: >> > >> > Right now there is hardly any documentation of what does _not_ require >> > the iothread mutex... basically everything except qemu_safe_ram_ptr and >> > qemu_ram_ptr_length requires it. > I don't disagree regarding the current state. But that doesn't imply it > has to be preserved. As soon as there are exec.c APIs that are BQL-safe, I'll document the state of all of them. In the meanwhile, I'll add a note that memory_region_find can be called outside the BQL. Paolo
diff --git a/exec.c b/exec.c index 8722420..f3d7b18 100644 --- a/exec.c +++ b/exec.c @@ -1408,18 +1408,26 @@ int qemu_ram_addr_from_host(void *ptr, ram_addr_t *ram_addr) return 0; } + block = ram_list.mru_block; + if (block && block->host && host - block->host < block->length) { + goto found; + } + QTAILQ_FOREACH(block, &ram_list.blocks, next) { /* This case append when the block is not mapped. */ if (block->host == NULL) { continue; } if (host - block->host < block->length) { - *ram_addr = block->offset + (host - block->host); - return 0; + goto found; } } return -1; + +found: + *ram_addr = block->offset + (host - block->host); + return 0; } /* Some of the softmmu routines need to translate from a host pointer
This function is not used outside the iothread mutex, so it can use ram_list.mru_block. Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> --- exec.c | 12 ++++++++++-- 1 file changed, 10 insertions(+), 2 deletions(-)