Skip to contentSkip to conversation

Mutilated form of Andi Kleen's AMD prefetch errata patch

PATCH Andi Kleen 2 participants 2 messages Sep 30, 2003 8 messages not in the archive

lore
Share & tools
lore

A message that is not in the archive (sent off-list or not yet received): 20030930073814.GA26649@mail.jlokier.co.uk.suse.lists.linux.kernel

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): 20030930133936.GA28876@mail.shareable.org.suse.lists.linux.kernel

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): 20030930144526.GC28876@mail.shareable.org.suse.lists.linux.kernel

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): 20030930165450.GF28876@mail.shareable.org.suse.lists.linux.kernel

A message that is not in the archive (sent off-list or not yet received): [email protected]

Andi Kleen

author
#8 lore raw
Quoted 6 lines · Dave Jones
Quoted 2 lines

Anyway, it should complain about lack of cmov not crash :)

not easy, given we execute cmov instructions before we even hit a printk. Such a test & output needs to be done in assembly in early startup.

I implemented it for long mode on x86-64.

It has to be done before the vesafb is initialized too, otherwise you cannot see the error message.

You could copy the code from arch/x86_64/boot/setup.S (starting with /* check for long mode. */) and change it to check for the CPUID bits you want. x86-64 checks a basic collection that has been determined to be the base set of x86-64 supporting CPUs. But it could be less or more.

-Andi

#9 lore raw
Followup to:  <[email protected]>
By author:    Andi Kleen <[email protected]>
In newsgroup: linux.dev.kernel
Quoted 12 lines

I implemented it for long mode on x86-64.

It has to be done before the vesafb is initialized too, otherwise you cannot see the error message.

You could copy the code from arch/x86_64/boot/setup.S (starting with /* check for long mode. */) and change it to check for the CPUID bits you want. x86-64 checks a basic collection that has been determined to be the base set of x86-64 supporting CPUs. But it could be less or more.

Do it in boot/setup.S so that you can still issue a message via the BIOS. However, don't forget you might need bug fixes/workarounds like the P6 SEP in there, too. From that perspective it would be nice to have something that can be written in C. Recent binutils actually allow gcc-generated .s-files to be assembled for a 16-bit environment (with lots of overrides.)

-hpa

Signature · 4 lines
<[email protected]> at work, <[email protected]> in private!
If you send me mail in HTML format I will assume it's spam.
"Unix gives you enough rope to shoot yourself in the foot."
Architectures needed: ia64 m68k mips64 ppc ppc64 s390 s390x sh v850 x86-64