* [ANNOUNCE] desc.c -- dump the i386 descriptor tables
@ 2003-04-24 3:02 Chuck Ebbert
2003-04-24 3:11 ` William Lee Irwin III
2003-04-24 8:53 ` Gabriel Paubert
0 siblings, 2 replies; 7+ messages in thread
From: Chuck Ebbert @ 2003-04-24 3:02 UTC (permalink / raw)
To: linux-kernel
Sample output:
desc.c -- dump linux descriptor tables, version 0.50
GDT at c0306280, 32 entries:
#00: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
#01: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
#02: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
#03: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
#04: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
#05: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
#06: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
#07: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
#08: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
#09: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
#10: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
#11: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
#12: base=00000000 limit=ffff flags=cf9b <G=1 P=1 S=1 DPL=0 Code>
#13: base=00000000 limit=ffff flags=cf93 <G=1 P=1 S=1 DPL=0 Data>
#14: base=00000000 limit=ffff flags=cffb <G=1 P=1 S=1 DPL=3 Code>
#15: base=00000000 limit=ffff flags=cff3 <G=1 P=1 S=1 DPL=3 Data>
#16: base=c0353800 limit=00eb flags=008b <G=0 P=1 S=0 DPL=0 Busy TSS>
#17: base=c0307c70 limit=0027 flags=0082 <G=0 P=1 S=0 DPL=0 LDT>
#18: base=00000000 limit=0000 flags=c09a <G=1 P=1 S=1 DPL=0 Code>
#19: base=00000000 limit=0000 flags=809a <G=1 P=1 S=1 DPL=0 Code>
#20: base=00000000 limit=0000 flags=8092 <G=1 P=1 S=1 DPL=0 Data>
#21: base=00000000 limit=0000 flags=8092 <G=1 P=1 S=1 DPL=0 Data>
#22: base=00000000 limit=0000 flags=8092 <G=1 P=1 S=1 DPL=0 Data>
#23: base=00000000 limit=0000 flags=409a <G=0 P=1 S=1 DPL=0 Code>
#24: base=00000000 limit=0000 flags=009a <G=0 P=1 S=1 DPL=0 Code>
#25: base=00000000 limit=0000 flags=4092 <G=0 P=1 S=1 DPL=0 Data>
#26: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
#27: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
#28: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
#29: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
#30: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
#31: base=c0355600 limit=00eb flags=0089 <G=0 P=1 S=0 DPL=0 Available TSS>
IDT at c0353000, 256 entries:
#000[0x00]: flags=8f<P,DPL0,trp> sel=60<GDT#12,RPL0> off=c0109840
#001[0x01]: flags=8e<P,DPL0,int> sel=60<GDT#12,RPL0> off=c01098d0
#002[0x02]: flags=8e<P,DPL0,int> sel=60<GDT#12,RPL0> off=c01098fc
#003[0x03]: flags=ef<P,DPL3,trp> sel=60<GDT#12,RPL0> off=c0109978
#004[0x04]: flags=ef<P,DPL3,trp> sel=60<GDT#12,RPL0> off=c0109984
#005[0x05]: flags=ef<P,DPL3,trp> sel=60<GDT#12,RPL0> off=c0109990
#006[0x06]: flags=8f<P,DPL0,trp> sel=60<GDT#12,RPL0> off=c010999c
#007[0x07]: flags=8f<P,DPL0,trp> sel=60<GDT#12,RPL0> off=c0109898
#008[0x08]: flags=85<P,DPL0,tsk> sel=f8<GDT#31,RPL0> off=00000000
#009[0x09]: flags=8f<P,DPL0,trp> sel=60<GDT#12,RPL0> off=c01099a8
#010[0x0a]: flags=8f<P,DPL0,trp> sel=60<GDT#12,RPL0> off=c01099b4
#011[0x0b]: flags=8f<P,DPL0,trp> sel=60<GDT#12,RPL0> off=c01099c0
#012[0x0c]: flags=8f<P,DPL0,trp> sel=60<GDT#12,RPL0> off=c01099cc
#013[0x0d]: flags=8f<P,DPL0,trp> sel=60<GDT#12,RPL0> off=c01099d8
#014[0x0e]: flags=8e<P,DPL0,int> sel=60<GDT#12,RPL0> off=c01099f0
#015[0x0f]: flags=8f<P,DPL0,trp> sel=60<GDT#12,RPL0> off=c0109a08
#016[0x10]: flags=8f<P,DPL0,trp> sel=60<GDT#12,RPL0> off=c0109880
#017[0x11]: flags=8f<P,DPL0,trp> sel=60<GDT#12,RPL0> off=c01099e4
#018[0x12]: flags=8f<P,DPL0,trp> sel=60<GDT#12,RPL0> off=c01099fc
#019[0x13]: flags=8f<P,DPL0,trp> sel=60<GDT#12,RPL0> off=c010988c
...
#032[0x20]: flags=8e<P,DPL0,int> sel=60<GDT#12,RPL0> off=c0108da0
#033[0x21]: flags=8e<P,DPL0,int> sel=60<GDT#12,RPL0> off=c0108da8
#034[0x22]: flags=8e<P,DPL0,int> sel=60<GDT#12,RPL0> off=c0108db0
#035[0x23]: flags=8e<P,DPL0,int> sel=60<GDT#12,RPL0> off=c0108db8
#036[0x24]: flags=8e<P,DPL0,int> sel=60<GDT#12,RPL0> off=c0108dc0
#037[0x25]: flags=8e<P,DPL0,int> sel=60<GDT#12,RPL0> off=c0108dc8
#038[0x26]: flags=8e<P,DPL0,int> sel=60<GDT#12,RPL0> off=c0108dd0
#039[0x27]: flags=8e<P,DPL0,int> sel=60<GDT#12,RPL0> off=c0108dd8
#040[0x28]: flags=8e<P,DPL0,int> sel=60<GDT#12,RPL0> off=c0108de0
#041[0x29]: flags=8e<P,DPL0,int> sel=60<GDT#12,RPL0> off=c0108de8
#042[0x2a]: flags=8e<P,DPL0,int> sel=60<GDT#12,RPL0> off=c0108df0
#043[0x2b]: flags=8e<P,DPL0,int> sel=60<GDT#12,RPL0> off=c0108df8
#044[0x2c]: flags=8e<P,DPL0,int> sel=60<GDT#12,RPL0> off=c0108e00
#045[0x2d]: flags=8e<P,DPL0,int> sel=60<GDT#12,RPL0> off=c0108e08
#046[0x2e]: flags=8e<P,DPL0,int> sel=60<GDT#12,RPL0> off=c0108e10
#047[0x2f]: flags=8e<P,DPL0,int> sel=60<GDT#12,RPL0> off=c0108e18
#048[0x30]: flags=8e<P,DPL0,int> sel=60<GDT#12,RPL0> off=c0108e40
#049[0x31]: flags=8e<P,DPL0,int> sel=60<GDT#12,RPL0> off=c0108da0
#050[0x32]: flags=8e<P,DPL0,int> sel=60<GDT#12,RPL0> off=c0108e50
...
===========================================================================
=====
#define PROG_MSG " desc.c -- dump linux descriptor tables"
#define VERSION "0.50"
/*
Author: Chuck Ebbert <76306.1226ompuserve.com>
*/
#include <stdio.h>
#include <fcntl.h>
#include <sys/types.h>
typedef struct {
unsigned short limit;
unsigned int base;
} __attribute__ ((packed)) dtr_t;
typedef struct {
unsigned short off1;
unsigned short sel;
unsigned char none,flags;
unsigned short off2;
} __attribute__ ((packed)) idt_entry_t;
typedef struct {
unsigned a,b;
} __attribute__ ((packed)) gdt_entry_t;
/* read/write from kmem */
#define BUILD_IO_FN(fn_name, fn_fn) \
static int fn_name(int fd, int offset, void *buf, int size) { \
if (lseek(fd, offset, 0) != offset) return 0; \
if (fn_fn(fd, buf, size) != size) return 0; \
return size; \
}
BUILD_IO_FN(rkm, read)
BUILD_IO_FN(wkm, write)
void perr(char *err) { /* exit with err msg */
perror(err), exit(1);
}
/* read/write wrapper with err handling */
#define xkm(f, p1, p2, p3, p4, msg) { \
int _p4 = (p4); if ((f)((p1), (p2), (p3), _p4) != _p4) perr(msg); }
#define TEST_BIT(v,b) ( !!(v & (1 << b)) )
main() {
unsigned sys_call_off, sct, fn, sel, flags, a, b, base, num;
dtr_t idtr, gdtr;
idt_entry_t idte;
gdt_entry_t gdte;
int kmem, i, t, found = 0;
unsigned char *p;
printf(PROG_MSG ", version " VERSION "\n\n");
kmem = open("/dev/kmem", O_RDONLY);
if (kmem < 0) perr("open kmem");
asm("sgdt %0" : "=m" (gdtr));
base = gdtr.base, num = (gdtr.limit + 1) >> 3;
printf(" GDT at %08x, %d entries:\n\n", base, num);
for (i = 0; i < num; i++) {
xkm(rkm, kmem, base + 8*i, &gdte, sizeof(gdte), "rkm gdte");
a = gdte.a, b = gdte.b, t = (b & 0x1f00) >> 8;
printf("#%02d: base=%08x limit=%04x"
" flags=%04x <G=%d P=%d S=%d DPL=%d %s>\n", i,
(b & 0xff000000) | ((b & 0xff) << 16) | (a >> 16),
a & 0xffff, (b & 0x00ffff00) >> 8, TEST_BIT(b,23),
TEST_BIT(b,15), TEST_BIT(b,12), (b & 0x6000) >> 13,
(a == 0) & (b == 0) ? "Empty" :
t == 0x0b ? "Busy TSS" :
t == 0x09 ? "Available TSS" :
t == 0x02 ? "LDT" :
t == 0x0c ? "Call gate" :
(t & 0x1c) == 0x18 ? "Code" :
(t & 0x1c) == 0x1c ? "Conforming code" :
(t & 0x1c) == 0x10 ? "Data" :
(t & 0x1c) == 0x14 ? "Exp-down data" :
"Unknown");
}
asm("sidt %0" : "=m" (idtr));
base = idtr.base, num = (idtr.limit + 1) >> 3;
printf("\n IDT at %08x, %d entries:\n\n", base, num);
for (i = 0; i < num; i++) {
xkm(rkm, kmem, base + 8*i, &idte, sizeof(idte), "rkm idte");
sel = idte.sel, flags = idte.flags;
printf("#%03d[0x%02x]: flags=%02x<%s,DPL%d,%s>"
" sel=%02x<%s#%02d,RPL%d> off=%08x\n", i, i,
flags, flags & 0x80 ? "P" : "NP", (flags & 0x60) >> 5,
(flags & 0x1f) == 0xe ? "int" :
(flags & 0x1f) == 0xf ? "trp" : "tsk",
sel, sel & 4 ? "LDT" : "GDT", sel >> 3, sel & 3,
(unsigned)((idte.off2 << 16) | idte.off1));
}
close(kmem);
}
------
Chuck
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [ANNOUNCE] desc.c -- dump the i386 descriptor tables
2003-04-24 3:02 [ANNOUNCE] desc.c -- dump the i386 descriptor tables Chuck Ebbert
@ 2003-04-24 3:11 ` William Lee Irwin III
2003-04-24 8:53 ` Gabriel Paubert
1 sibling, 0 replies; 7+ messages in thread
From: William Lee Irwin III @ 2003-04-24 3:11 UTC (permalink / raw)
To: Chuck Ebbert; +Cc: linux-kernel
On Wed, Apr 23, 2003 at 11:02:02PM -0400, Chuck Ebbert wrote:
> Sample output:
> desc.c -- dump linux descriptor tables, version 0.50
> GDT at c0306280, 32 entries:
> #00: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
> #01: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
> #02: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
Spiffy; this should help debug various things.
-- wli
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [ANNOUNCE] desc.c -- dump the i386 descriptor tables
2003-04-24 3:02 [ANNOUNCE] desc.c -- dump the i386 descriptor tables Chuck Ebbert
2003-04-24 3:11 ` William Lee Irwin III
@ 2003-04-24 8:53 ` Gabriel Paubert
1 sibling, 0 replies; 7+ messages in thread
From: Gabriel Paubert @ 2003-04-24 8:53 UTC (permalink / raw)
To: Chuck Ebbert; +Cc: linux-kernel
On Wed, Apr 23, 2003 at 11:02:02PM -0400, Chuck Ebbert wrote:
> Sample output:
>
>
> desc.c -- dump linux descriptor tables, version 0.50
>
> GDT at c0306280, 32 entries:
>
> #00: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
> #01: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
> #02: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
> #03: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
> #04: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
> #05: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
> #06: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
> #07: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
> #08: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
> #09: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
> #10: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
> #11: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
> #12: base=00000000 limit=ffff flags=cf9b <G=1 P=1 S=1 DPL=0 Code>
> #13: base=00000000 limit=ffff flags=cf93 <G=1 P=1 S=1 DPL=0 Data>
> #14: base=00000000 limit=ffff flags=cffb <G=1 P=1 S=1 DPL=3 Code>
> #15: base=00000000 limit=ffff flags=cff3 <G=1 P=1 S=1 DPL=3 Data>
> #16: base=c0353800 limit=00eb flags=008b <G=0 P=1 S=0 DPL=0 Busy TSS>
> #17: base=c0307c70 limit=0027 flags=0082 <G=0 P=1 S=0 DPL=0 LDT>
> #18: base=00000000 limit=0000 flags=c09a <G=1 P=1 S=1 DPL=0 Code>
> #19: base=00000000 limit=0000 flags=809a <G=1 P=1 S=1 DPL=0 Code>
> #20: base=00000000 limit=0000 flags=8092 <G=1 P=1 S=1 DPL=0 Data>
> #21: base=00000000 limit=0000 flags=8092 <G=1 P=1 S=1 DPL=0 Data>
> #22: base=00000000 limit=0000 flags=8092 <G=1 P=1 S=1 DPL=0 Data>
> #23: base=00000000 limit=0000 flags=409a <G=0 P=1 S=1 DPL=0 Code>
> #24: base=00000000 limit=0000 flags=009a <G=0 P=1 S=1 DPL=0 Code>
> #25: base=00000000 limit=0000 flags=4092 <G=0 P=1 S=1 DPL=0 Data>
> #26: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
> #27: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
> #28: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
> #29: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
> #30: base=00000000 limit=0000 flags=0000 <G=0 P=0 S=0 DPL=0 Empty>
> #31: base=c0355600 limit=00eb flags=0089 <G=0 P=1 S=0 DPL=0 Available TSS>
Nice but the limit field is 20 bits (shifted left by 12 bits if G=1).
[snipped]
>
> asm("sgdt %0" : "=m" (gdtr));
> base = gdtr.base, num = (gdtr.limit + 1) >> 3;
> printf(" GDT at %08x, %d entries:\n\n", base, num);
> for (i = 0; i < num; i++) {
> xkm(rkm, kmem, base + 8*i, &gdte, sizeof(gdte), "rkm gdte");
> a = gdte.a, b = gdte.b, t = (b & 0x1f00) >> 8;
> printf("#%02d: base=%08x limit=%04x"
^^^
s/%04x/%05x/
> " flags=%04x <G=%d P=%d S=%d DPL=%d %s>\n", i,
> (b & 0xff000000) | ((b & 0xff) << 16) | (a >> 16),
> a & 0xffff, (b & 0x00ffff00) >> 8, TEST_BIT(b,23),
^^^^^^^^^^
s/a & 0xffff/(a & 0xffff) | (b & 0x0f0000)/
And I'd also mask b with 0xf0ff00 to avoid the 4 higher bits of the
limit in the middle of the flags field.
Other suggestions left as an exercise to the reader:
a) distinguish 16 bit code from 32 bit code (GDT entry #19 is 16 bit code),
b) distinguish read/write from read-only and execute/read from
execute-only (are there any read-only or execute-only segments in the GDT?)
c) distinguish 16 and 32 bit expand down data (changes the upper
limit of the valid addresses, but it's never used in the GDT, so like
the conformant code it's not that important)
d) extend this for x86-64 :-)
Nothing to say about the IDT dump, though.
Gabriel
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [ANNOUNCE] desc.c -- dump the i386 descriptor tables
2003-04-24 13:59 Chuck Ebbert
@ 2003-04-24 16:25 ` Gabriel Paubert
0 siblings, 0 replies; 7+ messages in thread
From: Gabriel Paubert @ 2003-04-24 16:25 UTC (permalink / raw)
To: Chuck Ebbert; +Cc: linux-kernel
On Thu, Apr 24, 2003 at 09:59:58AM -0400, Chuck Ebbert wrote:
> Gabriel Paubert wrote:
>
>
> > > #31: base=c0355600 limit=00eb flags=0089 <G=0 P=1 S=0 DPL=0 Available TSS>
> >
> > Nice but the limit field is 20 bits (shifted left by 12 bits if G=1).
>
>
> Huh. The diagram I used was blank where the upper four limit bits belong
> so I assumed it was unused... and for some reason I was thinking you
> shifted
> left by 16 bits when G=1, so I never noticed the missing four bits. Thanks.
>
>
>
> > Other suggestions left as an exercise to the reader:
> >
> > a) distinguish 16 bit code from 32 bit code (GDT entry #19 is 16 bit code),
>
>
> BIOS?
Indeed for APM at least. If D (default operand size, bit 22 of b) bit is set
then the code segment is 32 bit code, if clear it's 16 bit.
The same bit is used for the upper limit of expand down data segments,
cleared for 64k, set for 4G.
> I now have it dumping LDTs, and should probably do at least cs:eip and
> eflags
> for each task.
LDT or TSS ?
>
>
> > d) extend this for x86-64 :-)
>
>
> Itanium. 8)
Itanium is not interesting, it uses the same format as i386 for
this and only uses the LDT/GDT in ia32 emulation mode. Interrupts
are very different but I don't remember how they work right now.
But x86_64 uses the reserved bit of the segment descriptors
to define extended GDT entries (and LDT too IIRC) with 64 bit
bases and/or 64 bit code segments (have to download it again).
These entries occupy 16 bytes. The IDT format is different
(as well as the TSS format).
Gabriel
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [ANNOUNCE] desc.c -- dump the i386 descriptor tables
@ 2003-04-24 13:59 Chuck Ebbert
2003-04-24 16:25 ` Gabriel Paubert
0 siblings, 1 reply; 7+ messages in thread
From: Chuck Ebbert @ 2003-04-24 13:59 UTC (permalink / raw)
To: Gabriel Paubert; +Cc: linux-kernel
Gabriel Paubert wrote:
> > #31: base=c0355600 limit=00eb flags=0089 <G=0 P=1 S=0 DPL=0 Available TSS>
>
> Nice but the limit field is 20 bits (shifted left by 12 bits if G=1).
Huh. The diagram I used was blank where the upper four limit bits belong
so I assumed it was unused... and for some reason I was thinking you
shifted
left by 16 bits when G=1, so I never noticed the missing four bits. Thanks.
> Other suggestions left as an exercise to the reader:
>
> a) distinguish 16 bit code from 32 bit code (GDT entry #19 is 16 bit code),
BIOS?
> b) distinguish read/write from read-only and execute/read from
> execute-only (are there any read-only or execute-only segments in the GDT?)
>
> c) distinguish 16 and 32 bit expand down data (changes the upper
> limit of the valid addresses, but it's never used in the GDT, so like
> the conformant code it's not that important)
I now have it dumping LDTs, and should probably do at least cs:eip and
eflags
for each task.
> d) extend this for x86-64 :-)
Itanium. 8)
------
Chuck
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [ANNOUNCE] desc.c -- dump the i386 descriptor tables
2003-04-24 9:06 Chuck Ebbert
@ 2003-04-24 9:12 ` William Lee Irwin III
0 siblings, 0 replies; 7+ messages in thread
From: William Lee Irwin III @ 2003-04-24 9:12 UTC (permalink / raw)
To: Chuck Ebbert; +Cc: linux-kernel
wli wrote:
>> Spiffy; this should help debug various things.
On Thu, Apr 24, 2003 at 05:06:58AM -0400, Chuck Ebbert wrote:
> I forgot to mention: try comparing 2.2, 2.4 and 2.5.
> Also, this is what I've been using to look at interrupt entry
> point alignment. On 2.4 for sure, and probably on 2.5 you have
> a 1-in-8 chance of getting a pathologically badly aligned timer
> handler on 32-byte cacheline machines every time you compile
> (IRQ 0 entry address & 0x1f == 0x1c.) OTOH the pagefault handler
> has come up 8-byte aligned every time but I didn't look at the
> source to see if it's coded that way.
I had more in mind using it to verify the correctness of per-node IDT
stuff than checking optimality in general, but it does look very handy.
-- wli
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [ANNOUNCE] desc.c -- dump the i386 descriptor tables
@ 2003-04-24 9:06 Chuck Ebbert
2003-04-24 9:12 ` William Lee Irwin III
0 siblings, 1 reply; 7+ messages in thread
From: Chuck Ebbert @ 2003-04-24 9:06 UTC (permalink / raw)
To: William Lee Irwin III; +Cc: linux-kernel
wli wrote:
> Spiffy; this should help debug various things.
I forgot to mention: try comparing 2.2, 2.4 and 2.5.
Also, this is what I've been using to look at interrupt entry
point alignment. On 2.4 for sure, and probably on 2.5 you have
a 1-in-8 chance of getting a pathologically badly aligned timer
handler on 32-byte cacheline machines every time you compile
(IRQ 0 entry address & 0x1f == 0x1c.) OTOH the pagefault handler
has come up 8-byte aligned every time but I didn't look at the
source to see if it's coded that way.
------
Chuck
^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2003-04-24 16:21 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2003-04-24 3:02 [ANNOUNCE] desc.c -- dump the i386 descriptor tables Chuck Ebbert
2003-04-24 3:11 ` William Lee Irwin III
2003-04-24 8:53 ` Gabriel Paubert
2003-04-24 9:06 Chuck Ebbert
2003-04-24 9:12 ` William Lee Irwin III
2003-04-24 13:59 Chuck Ebbert
2003-04-24 16:25 ` Gabriel Paubert
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).