ux164 (ruffian) fixes

From: Ivan Kokshaysky (ink@jurassic.park.msu.ru)
Date: Tue Nov 21 2000 - 10:46:09 EST


Two issues preventing ruffians from booting 2.4 (with bridges patch)
were found and fixed.
 - rather trivial one: someone decided that interrupt 4
   (irq 20 from builtin scsi) is 'bogus' ;-)
 - another issue is way not trivial and cost Michal and me a
   lot of sweat. Type 1 PCI configuration space accesses (and to
   some degree type 0 accesses to the bridge itself) were subtly
   broken. In other words, these accesses were failing randomly
   causing all sorts of havoc. At some point I managed to break
   them completely just by adding one extra mb()...
   Copying the code from 2.2 solved the problem - but it's a kind
   of magic.
   Interesting, other pyxis machines do not seem to be so sensitive,
   so I guess some design problems with ux164 motherboard - all this
   looks pretty much like timing issues.

Many thanks to Michal for that painful and time consuming testing
he has done.

Ivan.

--- 2.4.0t11/arch/alpha/kernel/core_cia.c Fri Oct 27 21:55:01 2000
+++ linux/arch/alpha/kernel/core_cia.c Thu Nov 16 17:46:54 2000
@@ -119,6 +119,7 @@ conf_read(unsigned long addr, unsigned c
         stat0 = *(vip)CIA_IOC_CIA_ERR;
         *(vip)CIA_IOC_CIA_ERR = stat0;
         mb();
+ *(vip)CIA_IOC_CIA_ERR; /* re-read to force write */
 
         /* If Type1 access, must set CIA CFG. */
         if (type1) {
@@ -128,6 +129,7 @@ conf_read(unsigned long addr, unsigned c
                 *(vip)CIA_IOC_CFG;
         }
 
+ mb();
         draina();
         mcheck_expected(0) = 1;
         mcheck_taken(0) = 0;
@@ -171,6 +173,7 @@ conf_write(unsigned long addr, unsigned
         stat0 = *(vip)CIA_IOC_CIA_ERR;
         *(vip)CIA_IOC_CIA_ERR = stat0;
         mb();
+ *(vip)CIA_IOC_CIA_ERR; /* re-read to force write */
 
         /* If Type1 access, must set CIA CFG. */
         if (type1) {
@@ -180,6 +183,7 @@ conf_write(unsigned long addr, unsigned
                 *(vip)CIA_IOC_CFG;
         }
 
+ mb();
         draina();
         mcheck_expected(0) = 1;
         mcheck_taken(0) = 0;
@@ -188,7 +192,7 @@ conf_write(unsigned long addr, unsigned
         /* Access configuration space. */
         *(vip)addr = value;
         mb();
- mb(); /* magic */
+ *(vip)addr; /* read back to force the write */
 
         mcheck_expected(0) = 0;
         mb();
@@ -606,7 +610,8 @@ do_init_arch(int is_pyxis)
         *(vip)CIA_IOC_ERR_MASK = temp;
 
         /* Clear all currently pending errors. */
- *(vip)CIA_IOC_CIA_ERR = 0;
+ temp = *(vip)CIA_IOC_CIA_ERR;
+ *(vip)CIA_IOC_CIA_ERR = temp;
 
         /* Turn on mchecks. */
         temp = *(vip)CIA_IOC_CIA_CTRL;
--- 2.4.0t11/arch/alpha/kernel/sys_ruffian.c Sun Nov 5 16:16:43 2000
+++ linux/arch/alpha/kernel/sys_ruffian.c Tue Nov 21 00:25:42 2000
@@ -56,9 +56,9 @@ ruffian_init_irq(void)
         
         init_i8259a_irqs();
 
- /* Not interested in the bogus interrupts (0,3,4,6),
+ /* Not interested in the bogus interrupts (0,3,6),
            NMI (1), HALT (2), flash (5), or 21142 (8). */
- init_pyxis_irqs(0x17f0000);
+ init_pyxis_irqs(0x16f0000);
 
         common_init_isa_dma();
 }
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Thu Nov 23 2000 - 21:00:21 EST