| Message ID | 1569497622-12496-2-git-send-email-imbrenda@linux.ibm.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | s390x: SCLP error cleanup | expand |
On 26/09/2019 13.33, Claudio Imbrenda wrote: > From: Janosch Frank <frankja@linux.ibm.com> > > * All sclp codes need to be checked for page boundary violations. > * Requests over 4k are not a spec exception. > * Invalid command checking has to be done before the boundary check. > > Signed-off-by: Janosch Frank <frankja@linux.ibm.com> > Reviewed-by: Jason J. Herne <jjherne@linux.ibm.com> > --- > hw/s390x/event-facility.c | 3 --- > hw/s390x/sclp.c | 25 ++++++++++++++++++++++--- > 2 files changed, 22 insertions(+), 6 deletions(-) > > diff --git a/hw/s390x/event-facility.c b/hw/s390x/event-facility.c > index 797ecbb..6620569 100644 > --- a/hw/s390x/event-facility.c > +++ b/hw/s390x/event-facility.c > @@ -377,9 +377,6 @@ static void command_handler(SCLPEventFacility *ef, SCCB *sccb, uint64_t code) > case SCLP_CMD_WRITE_EVENT_MASK: > write_event_mask(ef, sccb); > break; > - default: > - sccb->h.response_code = cpu_to_be16(SCLP_RC_INVALID_SCLP_COMMAND); > - break; > } > } > > diff --git a/hw/s390x/sclp.c b/hw/s390x/sclp.c > index fac7c3b..76feac8 100644 > --- a/hw/s390x/sclp.c > +++ b/hw/s390x/sclp.c > @@ -213,14 +213,33 @@ int sclp_service_call(CPUS390XState *env, uint64_t sccb, uint32_t code) > cpu_physical_memory_read(sccb, &work_sccb, sccb_len); > > /* Valid sccb sizes */ > - if (be16_to_cpu(work_sccb.h.length) < sizeof(SCCBHeader) || > - be16_to_cpu(work_sccb.h.length) > SCCB_SIZE) { > + if (be16_to_cpu(work_sccb.h.length) < sizeof(SCCBHeader)) { > r = -PGM_SPECIFICATION; > goto out; > } > > - sclp_c->execute(sclp, &work_sccb, code); > + switch (code & SCLP_CMD_CODE_MASK) { > + case SCLP_CMDW_READ_SCP_INFO: > + case SCLP_CMDW_READ_SCP_INFO_FORCED: > + case SCLP_CMDW_READ_CPU_INFO: > + case SCLP_CMDW_CONFIGURE_IOA: > + case SCLP_CMDW_DECONFIGURE_IOA: > + case SCLP_CMD_READ_EVENT_DATA: > + case SCLP_CMD_WRITE_EVENT_DATA: > + case SCLP_CMD_WRITE_EVENT_MASK: > + break; > + default: > + work_sccb.h.response_code = cpu_to_be16(SCLP_RC_INVALID_SCLP_COMMAND); > + goto out_write; > + } > > + if ((sccb + work_sccb.h.length) > ((sccb & PAGE_MASK) + PAGE_SIZE)) { I think you likely miss a be16_to_cpu() around work_sccb.h.length here? > + work_sccb.h.response_code = cpu_to_be16(SCLP_RC_SCCB_BOUNDARY_VIOLATION); > + goto out_write; > + } > + > + sclp_c->execute(sclp, &work_sccb, code); > +out_write: > cpu_physical_memory_write(sccb, &work_sccb, > be16_to_cpu(work_sccb.h.length)); At least here it is swapped --------^ Thomas
On 26.09.19 13:33, Claudio Imbrenda wrote: > From: Janosch Frank <frankja@linux.ibm.com> > > * All sclp codes need to be checked for page boundary violations. > * Requests over 4k are not a spec exception. > * Invalid command checking has to be done before the boundary check. Can we split this patch up so we fix one thing at a time? > > Signed-off-by: Janosch Frank <frankja@linux.ibm.com> > Reviewed-by: Jason J. Herne <jjherne@linux.ibm.com> > --- > hw/s390x/event-facility.c | 3 --- > hw/s390x/sclp.c | 25 ++++++++++++++++++++++--- > 2 files changed, 22 insertions(+), 6 deletions(-) > > diff --git a/hw/s390x/event-facility.c b/hw/s390x/event-facility.c > index 797ecbb..6620569 100644 > --- a/hw/s390x/event-facility.c > +++ b/hw/s390x/event-facility.c > @@ -377,9 +377,6 @@ static void command_handler(SCLPEventFacility *ef, SCCB *sccb, uint64_t code) > case SCLP_CMD_WRITE_EVENT_MASK: > write_event_mask(ef, sccb); > break; > - default: > - sccb->h.response_code = cpu_to_be16(SCLP_RC_INVALID_SCLP_COMMAND); > - break; > } > } > > diff --git a/hw/s390x/sclp.c b/hw/s390x/sclp.c > index fac7c3b..76feac8 100644 > --- a/hw/s390x/sclp.c > +++ b/hw/s390x/sclp.c > @@ -213,14 +213,33 @@ int sclp_service_call(CPUS390XState *env, uint64_t sccb, uint32_t code) > cpu_physical_memory_read(sccb, &work_sccb, sccb_len); > > /* Valid sccb sizes */ > - if (be16_to_cpu(work_sccb.h.length) < sizeof(SCCBHeader) || > - be16_to_cpu(work_sccb.h.length) > SCCB_SIZE) { > + if (be16_to_cpu(work_sccb.h.length) < sizeof(SCCBHeader)) { > r = -PGM_SPECIFICATION; > goto out; > } > > - sclp_c->execute(sclp, &work_sccb, code); > + switch (code & SCLP_CMD_CODE_MASK) { > + case SCLP_CMDW_READ_SCP_INFO: > + case SCLP_CMDW_READ_SCP_INFO_FORCED: > + case SCLP_CMDW_READ_CPU_INFO: > + case SCLP_CMDW_CONFIGURE_IOA: > + case SCLP_CMDW_DECONFIGURE_IOA: > + case SCLP_CMD_READ_EVENT_DATA: > + case SCLP_CMD_WRITE_EVENT_DATA: > + case SCLP_CMD_WRITE_EVENT_MASK: > + break; > + default: > + work_sccb.h.response_code = cpu_to_be16(SCLP_RC_INVALID_SCLP_COMMAND); > + goto out_write; > + } > > + if ((sccb + work_sccb.h.length) > ((sccb & PAGE_MASK) + PAGE_SIZE)) { > + work_sccb.h.response_code = cpu_to_be16(SCLP_RC_SCCB_BOUNDARY_VIOLATION); > + goto out_write; > + } > + > + sclp_c->execute(sclp, &work_sccb, code); > +out_write: > cpu_physical_memory_write(sccb, &work_sccb, > be16_to_cpu(work_sccb.h.length)); > >
On 9/27/19 10:51 AM, David Hildenbrand wrote: > On 26.09.19 13:33, Claudio Imbrenda wrote: >> From: Janosch Frank <frankja@linux.ibm.com> >> >> * All sclp codes need to be checked for page boundary violations. >> * Requests over 4k are not a spec exception. >> * Invalid command checking has to be done before the boundary check. > > Can we split this patch up so we fix one thing at a time? Sure, but we would end up with very small patches. Do you want that? > >> >> Signed-off-by: Janosch Frank <frankja@linux.ibm.com> >> Reviewed-by: Jason J. Herne <jjherne@linux.ibm.com> >> --- >> hw/s390x/event-facility.c | 3 --- >> hw/s390x/sclp.c | 25 ++++++++++++++++++++++--- >> 2 files changed, 22 insertions(+), 6 deletions(-) >> >> diff --git a/hw/s390x/event-facility.c b/hw/s390x/event-facility.c >> index 797ecbb..6620569 100644 >> --- a/hw/s390x/event-facility.c >> +++ b/hw/s390x/event-facility.c >> @@ -377,9 +377,6 @@ static void command_handler(SCLPEventFacility *ef, SCCB *sccb, uint64_t code) >> case SCLP_CMD_WRITE_EVENT_MASK: >> write_event_mask(ef, sccb); >> break; >> - default: >> - sccb->h.response_code = cpu_to_be16(SCLP_RC_INVALID_SCLP_COMMAND); >> - break; >> } >> } >> >> diff --git a/hw/s390x/sclp.c b/hw/s390x/sclp.c >> index fac7c3b..76feac8 100644 >> --- a/hw/s390x/sclp.c >> +++ b/hw/s390x/sclp.c >> @@ -213,14 +213,33 @@ int sclp_service_call(CPUS390XState *env, uint64_t sccb, uint32_t code) >> cpu_physical_memory_read(sccb, &work_sccb, sccb_len); >> >> /* Valid sccb sizes */ >> - if (be16_to_cpu(work_sccb.h.length) < sizeof(SCCBHeader) || >> - be16_to_cpu(work_sccb.h.length) > SCCB_SIZE) { >> + if (be16_to_cpu(work_sccb.h.length) < sizeof(SCCBHeader)) { >> r = -PGM_SPECIFICATION; >> goto out; >> } >> >> - sclp_c->execute(sclp, &work_sccb, code); >> + switch (code & SCLP_CMD_CODE_MASK) { >> + case SCLP_CMDW_READ_SCP_INFO: >> + case SCLP_CMDW_READ_SCP_INFO_FORCED: >> + case SCLP_CMDW_READ_CPU_INFO: >> + case SCLP_CMDW_CONFIGURE_IOA: >> + case SCLP_CMDW_DECONFIGURE_IOA: >> + case SCLP_CMD_READ_EVENT_DATA: >> + case SCLP_CMD_WRITE_EVENT_DATA: >> + case SCLP_CMD_WRITE_EVENT_MASK: >> + break; >> + default: >> + work_sccb.h.response_code = cpu_to_be16(SCLP_RC_INVALID_SCLP_COMMAND); >> + goto out_write; >> + } >> >> + if ((sccb + work_sccb.h.length) > ((sccb & PAGE_MASK) + PAGE_SIZE)) { >> + work_sccb.h.response_code = cpu_to_be16(SCLP_RC_SCCB_BOUNDARY_VIOLATION); >> + goto out_write; >> + } >> + >> + sclp_c->execute(sclp, &work_sccb, code); >> +out_write: >> cpu_physical_memory_write(sccb, &work_sccb, >> be16_to_cpu(work_sccb.h.length)); >> >> > >
On 27.09.19 11:14, Janosch Frank wrote: > On 9/27/19 10:51 AM, David Hildenbrand wrote: >> On 26.09.19 13:33, Claudio Imbrenda wrote: >>> From: Janosch Frank <frankja@linux.ibm.com> >>> >>> * All sclp codes need to be checked for page boundary violations. >>> * Requests over 4k are not a spec exception. >>> * Invalid command checking has to be done before the boundary check. >> >> Can we split this patch up so we fix one thing at a time? > > Sure, but we would end up with very small patches. > Do you want that? Why should I say no to easy-to-review, logically consistent, small chunks? I have shortcuts for my RB's and ACK's, so I don't have to type much ;)
On 9/27/19 11:17 AM, David Hildenbrand wrote: > On 27.09.19 11:14, Janosch Frank wrote: >> On 9/27/19 10:51 AM, David Hildenbrand wrote: >>> On 26.09.19 13:33, Claudio Imbrenda wrote: >>>> From: Janosch Frank <frankja@linux.ibm.com> >>>> >>>> * All sclp codes need to be checked for page boundary violations. >>>> * Requests over 4k are not a spec exception. >>>> * Invalid command checking has to be done before the boundary check. >>> >>> Can we split this patch up so we fix one thing at a time? >> >> Sure, but we would end up with very small patches. >> Do you want that? > > Why should I say no to easy-to-review, logically consistent, small > chunks? I have shortcuts for my RB's and ACK's, so I don't have to type > much ;) > Higher patch count for me, win - win :-)
On 27.09.19 11:20, Janosch Frank wrote: > On 9/27/19 11:17 AM, David Hildenbrand wrote: >> On 27.09.19 11:14, Janosch Frank wrote: >>> On 9/27/19 10:51 AM, David Hildenbrand wrote: >>>> On 26.09.19 13:33, Claudio Imbrenda wrote: >>>>> From: Janosch Frank <frankja@linux.ibm.com> >>>>> >>>>> * All sclp codes need to be checked for page boundary violations. >>>>> * Requests over 4k are not a spec exception. >>>>> * Invalid command checking has to be done before the boundary check. >>>> >>>> Can we split this patch up so we fix one thing at a time? >>> >>> Sure, but we would end up with very small patches. >>> Do you want that? >> >> Why should I say no to easy-to-review, logically consistent, small >> chunks? I have shortcuts for my RB's and ACK's, so I don't have to type >> much ;) >> > > Higher patch count for me, win - win :-) > Now that's the spirit :)
diff --git a/hw/s390x/event-facility.c b/hw/s390x/event-facility.c index 797ecbb..6620569 100644 --- a/hw/s390x/event-facility.c +++ b/hw/s390x/event-facility.c @@ -377,9 +377,6 @@ static void command_handler(SCLPEventFacility *ef, SCCB *sccb, uint64_t code) case SCLP_CMD_WRITE_EVENT_MASK: write_event_mask(ef, sccb); break; - default: - sccb->h.response_code = cpu_to_be16(SCLP_RC_INVALID_SCLP_COMMAND); - break; } } diff --git a/hw/s390x/sclp.c b/hw/s390x/sclp.c index fac7c3b..76feac8 100644 --- a/hw/s390x/sclp.c +++ b/hw/s390x/sclp.c @@ -213,14 +213,33 @@ int sclp_service_call(CPUS390XState *env, uint64_t sccb, uint32_t code) cpu_physical_memory_read(sccb, &work_sccb, sccb_len); /* Valid sccb sizes */ - if (be16_to_cpu(work_sccb.h.length) < sizeof(SCCBHeader) || - be16_to_cpu(work_sccb.h.length) > SCCB_SIZE) { + if (be16_to_cpu(work_sccb.h.length) < sizeof(SCCBHeader)) { r = -PGM_SPECIFICATION; goto out; } - sclp_c->execute(sclp, &work_sccb, code); + switch (code & SCLP_CMD_CODE_MASK) { + case SCLP_CMDW_READ_SCP_INFO: + case SCLP_CMDW_READ_SCP_INFO_FORCED: + case SCLP_CMDW_READ_CPU_INFO: + case SCLP_CMDW_CONFIGURE_IOA: + case SCLP_CMDW_DECONFIGURE_IOA: + case SCLP_CMD_READ_EVENT_DATA: + case SCLP_CMD_WRITE_EVENT_DATA: + case SCLP_CMD_WRITE_EVENT_MASK: + break; + default: + work_sccb.h.response_code = cpu_to_be16(SCLP_RC_INVALID_SCLP_COMMAND); + goto out_write; + } + if ((sccb + work_sccb.h.length) > ((sccb & PAGE_MASK) + PAGE_SIZE)) { + work_sccb.h.response_code = cpu_to_be16(SCLP_RC_SCCB_BOUNDARY_VIOLATION); + goto out_write; + } + + sclp_c->execute(sclp, &work_sccb, code); +out_write: cpu_physical_memory_write(sccb, &work_sccb, be16_to_cpu(work_sccb.h.length));