Skip to contentSkip to conversation

AGP performance fixes

PATCH Benjamin Herrenschmidt 2 participants 3 messages Nov 9, 2005 1 message not in the archive

lore
Share & tools
lore

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

#1 lore raw
Quoted 10 lines · Linux Kernel Mailing List

tree 048db8e12c2b31ec2b88c3154da4c6a77b80028c
parent 329f7dba5f7dc3bc9a30ad00cf373d2e83115aa1
author Alan Hourihane <[email protected]> Mon, 07 Nov 2005 15:35:34 -0800 committer Dave Jones <[email protected]> Wed, 09 Nov 2005 05:43:54 -0800

[PATCH] AGP performance fixes

AGP allocation/deallocation is suffering major performance issues due to the nature of global_flush_tlb() being called on every change_page_attr() call.

.../...

*** Warning: ".global_flush_tlb" [drivers/char/agp/agpgart.ko] undefined!

Am I supposed to define something new for uninorth-agp ? Is this yet another x86-only (or worse, some x86 chipsets only concept going global ?

Ben.

#2 lore raw
Quoted 12 lines · Benjamin Herrenschmidt
Quoted 10 lines · Linux Kernel Mailing List

tree 048db8e12c2b31ec2b88c3154da4c6a77b80028c
parent 329f7dba5f7dc3bc9a30ad00cf373d2e83115aa1
author Alan Hourihane <[email protected]> Mon, 07 Nov 2005 15:35:34 -0800 committer Dave Jones <[email protected]> Wed, 09 Nov 2005 05:43:54 -0800

[PATCH] AGP performance fixes

AGP allocation/deallocation is suffering major performance issues due to the nature of global_flush_tlb() being called on every change_page_attr() call.

Quoted 6 lines

.../...

*** Warning: ".global_flush_tlb" [drivers/char/agp/agpgart.ko] undefined!

Am I supposed to define something new for uninorth-agp ? Is this yet another x86-only (or worse, some x86 chipsets only concept going global ?

change global_flush_tlb() to flush_agp_mappings() and you should be fine.

Linus has a fix for this.

Alan.

#3 lore raw
Quoted 8 lines
Quoted 3 lines

Am I supposed to define something new for uninorth-agp ? Is this yet another x86-only (or worse, some x86 chipsets only concept going global ?

change global_flush_tlb() to flush_agp_mappings() and you should be fine.

Linus has a fix for this.

Yup, it just showed up in git, thanks.

Ben.