85xx: Additional fixes and cleanup of MP code
PATCH 1/1 85xx: Additional fixes and cleanup of MP code
* adjust __spin_table alignment to match ePAPR v0.94 spec
* loop over all cpus when determing who is up. This fixes an issue if the "boot cpu" isn't core0. The "boot cpu" will already be in the cpu_up_mask so there is no harm
* Added some protection in the code to ensure proper behavior. These changes are explicitly needed but don't hurt:
- Added eieio to ensure the "hot word" of the table is written after all other table updates have occurred.
- Added isync to ensure we don't prefetch loading of table entries until we a released
These issues we raised by Dave Liu.
---
For 1.3.3
cpu/mpc85xx/mp.c | 6 +++++- cpu/mpc85xx/release.S | 3 ++- 2 files changed, 7 insertions(+), 2 deletions(-)
cpu/mpc85xx/mp.c +5 -1
@@ -103,6 +103,10 @@ int cpu_release(int nr, int argc, char *argv[]) } table[BOOT_ENTRY_ADDR_UPPER ] = (u32)(boot_addr >>32 ); + +/* ensure all table updates complete before final address write */ +eieio (); + table[BOOT_ENTRY_ADDR_LOWER ] = (u32)(boot_addr &0xffffffff );return 0 ;
@@ -153,7 +157,7 @@ static void pq3_mp_up(unsigned long bootpg)/* wait for everyone */ while (timeout) {int i; -for (i =1 ; i <CONFIG_NR_CPUS ; i++) { +for (i =0 ; i <CONFIG_NR_CPUS ; i++) {if (table[i *NUM_BOOT_ENTRY +BOOT_ENTRY_ADDR_LOWER ]) cpu_up_mask |= (1 << i); };
cpu/mpc85xx/release.S +2 -1
@@ -114,6 +114,7 @@ __secondary_start_page: lwz r4,ENTRY_ADDR_LOWER(r10) andi. r11,r4,1 bne 2b + isync /* get the upper bits of the addr */ lwz r11,ENTRY_ADDR_UPPER(r10)
@@ -169,7 +170,7 @@ __secondary_start_page: mtspr SPRN_SRR1,r13 rfi - .align 3 + .align L1_CACHE_SHIFT .globl __spin_table __spin_table: .space CONFIG_NR_CPUS*ENTRY_SIZE
Signature · 1 line
1.5.4.1
PATCH 1/1 85xx: Additional fixes and cleanup of MP code
Quoted 11 lines · In message <[email protected]> you wrote:
* adjust
__spin_tablealignment to match ePAPR v0.94 spec
* loop over all cpus when determing who is up. This fixes an issue if the "boot cpu" isn't core0. The "boot cpu" will already be in thecpu_up_maskso there is no harm
* Added some protection in the code to ensure proper behavior. These changes are explicitly needed but don't hurt:
- Added eieio to ensure the "hot word" of the table is written after all other table updates have occurred.
- Added isync to ensure we don't prefetch loading of table entries until we a released
Quoted 3 lines
These issues we raised by Dave Liu.
Signed-off-by Kumar Gala <[email protected]>
Will Andy send a pull request or shall I apply directly?
Best regards,
Wolfgang Denk
Signature · 7 lines
DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd at denx.de The management question ... is not _whether_ to build a pilot system and throw it away. You _will_ do that. The only question is whether to plan in advance to build a throwaway, or to promise to deliver the throwaway to customers. - Fred Brooks, "The Mythical Man Month"
PATCH 1/1 85xx: Additional fixes and cleanup of MP code
Quoted 14 lines · Wolfgang Denk
In message <[email protected]>
Quoted 12 lines · you
* adjust
__spin_tablealignment to match ePAPR v0.94 spec
* loop over all cpus when determing who is up. This fixes an issue if
the "boot cpu" isn't core0. The "boot cpu" will already be in thecpu_up_maskso there is no harm
* Added some protection in the code to ensure proper behavior. These changes are explicitly needed but don't hurt:
- Added eieio to ensure the "hot word" of the table is written after all other table updates have occurred.
- Added isync to ensure we don't prefetch loading of table entries until we a released
Quoted 5 lines
Quoted 3 lines
These issues we raised by Dave Liu.
Signed-off-by Kumar Gala <[email protected]>Will Andy send a pull request or shall I apply directly?
just apply directly.
- k