| Message ID | 1340085758.28143.4.camel@pasglop |
|---|---|
| State | New |
| Headers | show |
Am 19.06.2012 08:02, schrieb Benjamin Herrenschmidt: > We were incorrectly g_free'ing an object that isn't allocated > in one error path and failed to release it completely in another > > This fixes qemu crashes with some cases of IO errors. > > Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> > --- > hw/spapr_vscsi.c | 4 +++- > 1 file changed, 3 insertions(+), 1 deletion(-) > > diff --git a/hw/spapr_vscsi.c b/hw/spapr_vscsi.c > index d2fe3e5..6afc3b4 100644 > --- a/hw/spapr_vscsi.c > +++ b/hw/spapr_vscsi.c > @@ -801,6 +801,7 @@ static void vscsi_got_payload(VSCSIState *s, vscsi_crq *crq) > if (crq->s.IU_length > sizeof(union viosrp_iu)) { > fprintf(stderr, "VSCSI: SRP IU too long (%d bytes) !\n", > crq->s.IU_length); > + vscsi_put_req(req); > return; > } > > @@ -808,7 +809,8 @@ static void vscsi_got_payload(VSCSIState *s, vscsi_crq *crq) > if (spapr_vio_dma_read(&s->vdev, crq->s.IU_data_ptr, &req->iu, > crq->s.IU_length)) { > fprintf(stderr, "vscsi_got_payload: DMA read failure !\n"); > - g_free(req); > + vscsi_put_req(req); > + return; Tab alert. :) /-F > } > memcpy(&req->crq, crq, sizeof(vscsi_crq)); > > > >
On Tue, 2012-06-19 at 13:31 +0200, Andreas Färber wrote:
> Tab alert. :)
Argh :-) Gotta get an emacs config for use with qemu :-)
Cheers,
Ben.
On 19.06.2012, at 22:18, Benjamin Herrenschmidt wrote: > On Tue, 2012-06-19 at 13:31 +0200, Andreas Färber wrote: >> Tab alert. :) > > Argh :-) Gotta get an emacs config for use with qemu :-) So do you want to resend? I can just fix it on the fly too. Alex
On Tue, 2012-06-19 at 22:20 +0200, Alexander Graf wrote:
> So do you want to resend? I can just fix it on the fly too.
If you can then sure, please do :-)
Cheers,
Ben.
On 19.06.2012, at 08:02, Benjamin Herrenschmidt wrote: > We were incorrectly g_free'ing an object that isn't allocated > in one error path and failed to release it completely in another > > This fixes qemu crashes with some cases of IO errors. > > Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> Thanks, applied to ppc-next. Alex
diff --git a/hw/spapr_vscsi.c b/hw/spapr_vscsi.c index d2fe3e5..6afc3b4 100644 --- a/hw/spapr_vscsi.c +++ b/hw/spapr_vscsi.c @@ -801,6 +801,7 @@ static void vscsi_got_payload(VSCSIState *s, vscsi_crq *crq) if (crq->s.IU_length > sizeof(union viosrp_iu)) { fprintf(stderr, "VSCSI: SRP IU too long (%d bytes) !\n", crq->s.IU_length); + vscsi_put_req(req); return; } @@ -808,7 +809,8 @@ static void vscsi_got_payload(VSCSIState *s, vscsi_crq *crq) if (spapr_vio_dma_read(&s->vdev, crq->s.IU_data_ptr, &req->iu, crq->s.IU_length)) { fprintf(stderr, "vscsi_got_payload: DMA read failure !\n"); - g_free(req); + vscsi_put_req(req); + return; } memcpy(&req->crq, crq, sizeof(vscsi_crq));
We were incorrectly g_free'ing an object that isn't allocated in one error path and failed to release it completely in another This fixes qemu crashes with some cases of IO errors. Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org> --- hw/spapr_vscsi.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)