| Message ID | 20200402172339.622720-1-leonardo@linux.ibm.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | [v3,1/1] ppc/spapr: Add hotremovable flag on DIMM LMBs on drmem_v2 | expand |
On Thu, Apr 02, 2020 at 02:23:40PM -0300, Leonardo Bras wrote: > On reboot, all memory that was previously added using object_add and > device_add is placed in this DIMM area. > > The new SPAPR_LMB_FLAGS_HOTREMOVABLE flag helps Linux to put this memory in > the correct memory zone, so no unmovable allocations are made there, > allowing the object to be easily hot-removed by device_del and > object_del. > > This new flag was accepted in Power Architecture documentation. > > Signed-off-by: Leonardo Bras <leonardo@linux.ibm.com> > Reviewed-by: Bharata B Rao <bharata@linux.ibm.com> The change looks reasonable. Is there a PAPR RFC I can see describing the new bit, though? > --- > Changes since v1: > - Flag name changed from SPAPR_LMB_FLAGS_HOTPLUGGED to > SPAPR_LMB_FLAGS_HOTREMOVABLE > --- > hw/ppc/spapr.c | 3 ++- > include/hw/ppc/spapr.h | 1 + > 2 files changed, 3 insertions(+), 1 deletion(-) > > diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c > index 9a2bd501aa..fe662e297e 100644 > --- a/hw/ppc/spapr.c > +++ b/hw/ppc/spapr.c > @@ -446,7 +446,8 @@ static int spapr_dt_dynamic_memory_v2(SpaprMachineState *spapr, void *fdt, > g_assert(drc); > elem = spapr_get_drconf_cell(size / lmb_size, addr, > spapr_drc_index(drc), node, > - SPAPR_LMB_FLAGS_ASSIGNED); > + (SPAPR_LMB_FLAGS_ASSIGNED | > + SPAPR_LMB_FLAGS_HOTREMOVABLE); > QSIMPLEQ_INSERT_TAIL(&drconf_queue, elem, entry); > nr_entries++; > cur_addr = addr + size; > diff --git a/include/hw/ppc/spapr.h b/include/hw/ppc/spapr.h > index 42d64a0368..93e0d43051 100644 > --- a/include/hw/ppc/spapr.h > +++ b/include/hw/ppc/spapr.h > @@ -880,6 +880,7 @@ int spapr_rtc_import_offset(SpaprRtcState *rtc, int64_t legacy_offset); > #define SPAPR_LMB_FLAGS_ASSIGNED 0x00000008 > #define SPAPR_LMB_FLAGS_DRC_INVALID 0x00000020 > #define SPAPR_LMB_FLAGS_RESERVED 0x00000080 > +#define SPAPR_LMB_FLAGS_HOTREMOVABLE 0x00000100 > > void spapr_do_system_reset_on_cpu(CPUState *cs, run_on_cpu_data arg); >
Hello David, On Fri, 2020-04-03 at 09:49 +1100, David Gibson wrote: > > The change looks reasonable. Is there a PAPR RFC I can see describing > the new bit, though? The ACR is approved and I already added the change to LoPAPR/LOPAR, it's just a matter of time to get it released. (I think it's accessible to members of OpenPower Foundation) This is the snippet of the change: At the bottom of "Table 3.11: Flag Word", add another line containing: Hot-removable Memory | 0x00000100 | If b'1', this LMB can be hot- removed. | | If b'0', this LMB may not be hot- removed Best regards, Leonardo Bras
On Thu, Apr 02, 2020 at 08:00:57PM -0300, Leonardo Bras wrote: > Hello David, > > On Fri, 2020-04-03 at 09:49 +1100, David Gibson wrote: > > > > The change looks reasonable. Is there a PAPR RFC I can see describing > > the new bit, though? > > The ACR is approved and I already added the change to LoPAPR/LOPAR, > it's just a matter of time to get it released. > (I think it's accessible to members of OpenPower Foundation) Hrm. The latest LoPAPR I can find is from 2016, so a "matter of time" isn't very encouraging. > > This is the snippet of the change: > > At the bottom of "Table 3.11: Flag Word", add another line containing: > Hot-removable Memory | 0x00000100 | If b'1', this LMB can be hot- > removed. > | | If b'0', this LMB may not be hot- > removed > > Best regards, > Leonardo Bras
On Fri, 2020-04-03 at 10:16 +1100, David Gibson wrote: > On Thu, Apr 02, 2020 at 08:00:57PM -0300, Leonardo Bras wrote: > > Hello David, > > > > On Fri, 2020-04-03 at 09:49 +1100, David Gibson wrote: > > > The change looks reasonable. Is there a PAPR RFC I can see describing > > > the new bit, though? > > > > The ACR is approved and I already added the change to LoPAPR/LOPAR, > > it's just a matter of time to get it released. > > (I think it's accessible to members of OpenPower Foundation) > > Hrm. The latest LoPAPR I can find is from 2016, so a "matter of time" > isn't very encouraging. I understand the feeling. There is a new initiative inside the OpenPower Foundation to keep these documents updated. I think we will be able to have more frequent updates of the documentation soon. Best Regards,
On Thu, Apr 02, 2020 at 08:19:34PM -0300, Leonardo Bras wrote: > On Fri, 2020-04-03 at 10:16 +1100, David Gibson wrote: > > On Thu, Apr 02, 2020 at 08:00:57PM -0300, Leonardo Bras wrote: > > > Hello David, > > > > > > On Fri, 2020-04-03 at 09:49 +1100, David Gibson wrote: > > > > The change looks reasonable. Is there a PAPR RFC I can see describing > > > > the new bit, though? > > > > > > The ACR is approved and I already added the change to LoPAPR/LOPAR, > > > it's just a matter of time to get it released. > > > (I think it's accessible to members of OpenPower Foundation) > > > > Hrm. The latest LoPAPR I can find is from 2016, so a "matter of time" > > isn't very encouraging. > > I understand the feeling. > There is a new initiative inside the OpenPower Foundation to keep these > documents updated. > > I think we will be able to have more frequent updates of the > documentation soon. Urgh. Ok, I'm sorry Leonardo, but, Nack. I will not merge a change where my only knowledge of the spec has to be based on hearsay. By all means use this as ammunition to convince IBM and/or the openpower foundation to expedite getting a workable document release process up and running.
On Fri, 2020-04-03 at 10:40 +1100, David Gibson wrote: > Urgh. Ok, I'm sorry Leonardo, but, > > Nack. I will not merge a change where my only knowledge of the spec > has to be based on hearsay. > > By all means use this as ammunition to convince IBM and/or the > openpower foundation to expedite getting a workable document release > process up and running. No problem David, of course I understand your point. I am working on this document release, and it will probably be available soon. When it is, we can visit this patch again. Best regards, Leonardo Bras
diff --git a/hw/ppc/spapr.c b/hw/ppc/spapr.c index 9a2bd501aa..fe662e297e 100644 --- a/hw/ppc/spapr.c +++ b/hw/ppc/spapr.c @@ -446,7 +446,8 @@ static int spapr_dt_dynamic_memory_v2(SpaprMachineState *spapr, void *fdt, g_assert(drc); elem = spapr_get_drconf_cell(size / lmb_size, addr, spapr_drc_index(drc), node, - SPAPR_LMB_FLAGS_ASSIGNED); + (SPAPR_LMB_FLAGS_ASSIGNED | + SPAPR_LMB_FLAGS_HOTREMOVABLE); QSIMPLEQ_INSERT_TAIL(&drconf_queue, elem, entry); nr_entries++; cur_addr = addr + size; diff --git a/include/hw/ppc/spapr.h b/include/hw/ppc/spapr.h index 42d64a0368..93e0d43051 100644 --- a/include/hw/ppc/spapr.h +++ b/include/hw/ppc/spapr.h @@ -880,6 +880,7 @@ int spapr_rtc_import_offset(SpaprRtcState *rtc, int64_t legacy_offset); #define SPAPR_LMB_FLAGS_ASSIGNED 0x00000008 #define SPAPR_LMB_FLAGS_DRC_INVALID 0x00000020 #define SPAPR_LMB_FLAGS_RESERVED 0x00000080 +#define SPAPR_LMB_FLAGS_HOTREMOVABLE 0x00000100 void spapr_do_system_reset_on_cpu(CPUState *cs, run_on_cpu_data arg);