On 02/20/2017 11:01 PM, Michael Ellerman wrote:
Douglas Miller <dougm...@linux.vnet.ibm.com> writes:
diff --git a/arch/powerpc/xmon/xmon.c b/arch/powerpc/xmon/xmon.c
index 9c0e17c..6249975 100644
--- a/arch/powerpc/xmon/xmon.c
+++ b/arch/powerpc/xmon/xmon.c
@@ -2334,9 +2338,49 @@ static void dump_pacas(void)
}
#endif
+static void dump_by_size(unsigned long addr, long count, int size)
+{
+ unsigned char temp[16];
+ int i, j;
+ u64 val;
+
+ /*
+ * 'count' was aligned 16. If that changes, the following
+ * must also change to accommodate other values for 'count'.
+ */
+ for (i = 0; i < count; i += 16, addr += 16) {
+ printf(REG, addr);
+
+ if (mread(addr, temp, 16) != 16) {
+ printf("Faulted reading %d bytes from 0x"REG"\n", 16,
addr);
+ return;
+ }
+
+ for (j = 0; j < 16; j += size) {
+ putchar(' ');
+ switch (size) {
+ case 1: val = temp[j]; break;
+ case 2: val = *(u16 *)&temp[j]; break;
+ case 4: val = *(u32 *)&temp[j]; break;
+ case 8: val = *(u64 *)&temp[j]; break;
+ default: val = 0;
+ }
+
+ printf("%0*lx", size * 2, val);
+ }
+ printf(" |");
+ for (j = 0; j < 16; ++j) {
+ val = temp[j];
+ putchar(' ' <= val && val <= '~' ? val : '.');
+ }
+ printf("|\n");
I know the ascii dump looks nice, but I think it's misleading. Which is
why I omitted it from my version.
eg.
0:mon> d $__kstrtab_init_task
c000000000c03ebe 696e69745f746173 6b006d6d755f6665 |init_task.mmu_fe|
c000000000c03ece 61747572655f6b65 7973006370755f66 |ature_keys.cpu_f|
c000000000c03ede 6561747572655f6b 657973006375725f |eature_keys.cur_|
c000000000c03eee 6370755f73706563 00766972715f746f |cpu_spec.virq_to|
0:mon> d8 $__kstrtab_init_task
c000000000c03ebe 7361745f74696e69 65665f756d6d006b |init_task.mmu_fe|
c000000000c03ece 656b5f6572757461 665f757063007379 |ature_keys.cpu_f|
c000000000c03ede 6b5f657275746165 5f72756300737965 |eature_keys.cur_|
c000000000c03eee 636570735f757063 6f745f7172697600 |cpu_spec.virq_to|
That second dump says at c000000000c03ebe there is a byte with the value
0x73, which prints as 'i' - but that's false.
So I've dropped the ascii printing for now because I want to sneak this
in to v4.11.
If you want to send a follow-up patch to do the ascii byte reversed that
would be nice.
cheers
I would disagree, anything printed as bytes should be in only one order
- the order it exists in memory. I maintain that the ascii dump is
correctly printed. The purpose of an ascii dump like this is to show
what is in memory. ASCII in memory has only one order.
Thanks,
Doug
