| Message ID | 20200929115713.1082952-1-maddy@linux.ibm.com |
|---|---|
| State | Accepted |
| Headers | show |
| Series | [1/2] hw/imc: move imc_init() towards end main_cpu_entry() | expand |
On 9/29/20 5:27 PM, Madhavan Srinivasan wrote: > imc_init() checks for the 24x7 microcode state at boot to > check whether the microcode is in proper state (running or paused). > But in a larger system, loading of 24x7 microcode by OCC gets delayed. Does this mean we are hitting on witherspoon/zz class system -OR- only in large system? > Because of this, imc_init() removes imc devices from the device tree. > > Moving imc_init() function towards end of the main_cpu_entry() > works around this. I see that you have marked this patch for stable. Can you tell me in which branch you want this fix and why? -Vasant > > Signed-off-by: Madhavan Srinivasan <maddy@linux.ibm.com> > --- > core/init.c | 6 +++--- > 1 file changed, 3 insertions(+), 3 deletions(-) > > diff --git a/core/init.c b/core/init.c > index 1289805f..da003e21 100644 > --- a/core/init.c > +++ b/core/init.c > @@ -1322,9 +1322,6 @@ void __noreturn __nomcount main_cpu_entry(const void *fdt) > /* NX init */ > nx_init(); > > - /* Init In-Memory Collection related stuff (load the IMC dtb into memory) */ > - imc_init(); > - > /* Probe PHB3 on P8 */ > probe_phb3(); > > @@ -1366,6 +1363,9 @@ void __noreturn __nomcount main_cpu_entry(const void *fdt) > /* Add the list of interrupts going to OPAL */ > add_opal_interrupts(); > > + /* Init In-Memory Collection related stuff (load the IMC dtb into memory) */ > + imc_init(); > + > /* Disable protected execution facility in BML */ > cpu_disable_pef(); >
On 9/29/20 5:27 PM, Madhavan Srinivasan wrote: > imc_init() checks for the 24x7 microcode state at boot to > check whether the microcode is in proper state (running or paused). > But in a larger system, loading of 24x7 microcode by OCC gets delayed. > Because of this, imc_init() removes imc devices from the device tree. > > Moving imc_init() function towards end of the main_cpu_entry() > works around this. > > Signed-off-by: Madhavan Srinivasan <maddy@linux.ibm.com> Thanks! Merged this series to master as of d505f403. -Vasant
diff --git a/core/init.c b/core/init.c index 1289805f..da003e21 100644 --- a/core/init.c +++ b/core/init.c @@ -1322,9 +1322,6 @@ void __noreturn __nomcount main_cpu_entry(const void *fdt) /* NX init */ nx_init(); - /* Init In-Memory Collection related stuff (load the IMC dtb into memory) */ - imc_init(); - /* Probe PHB3 on P8 */ probe_phb3(); @@ -1366,6 +1363,9 @@ void __noreturn __nomcount main_cpu_entry(const void *fdt) /* Add the list of interrupts going to OPAL */ add_opal_interrupts(); + /* Init In-Memory Collection related stuff (load the IMC dtb into memory) */ + imc_init(); + /* Disable protected execution facility in BML */ cpu_disable_pef();
imc_init() checks for the 24x7 microcode state at boot to check whether the microcode is in proper state (running or paused). But in a larger system, loading of 24x7 microcode by OCC gets delayed. Because of this, imc_init() removes imc devices from the device tree. Moving imc_init() function towards end of the main_cpu_entry() works around this. Signed-off-by: Madhavan Srinivasan <maddy@linux.ibm.com> --- core/init.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)