From: Adrian Cox <adrian@humboldt.co.uk>
To: linuxppc-dev@lists.linuxppc.org
Subject: [PATCH] Workaround for 745x data corruption bug
Date: Sat, 31 Jul 2004 17:27:56 +0100 [thread overview]
Message-ID: <1091291276.987.57.camel@localhost> (raw)
[-- Attachment #1: Type: text/plain, Size: 583 bytes --]
Recently released errata documents show a new bug in all 745x family
processors. This can cause data corruption when memory is mapped
non-coherent and one of these conditions is true:
1) L2 hardware prefetch is enabled (as it is in Linux)
2) instructions and data are fetched from the same or adjacent cache
lines.
The attached patch adds a workaround, by setting CPU_FTR_NEED_COHERENT
on all 745x processors.
The bug is 7447A errata #16, 7457 errata #26, 7455 errata #33, 7450
errata #69.
Signed-off-by: Adrian Cox <adrian@humboldt.co.uk>
- Adrian Cox
Humboldt Solutions Ltd.
[-- Attachment #2: Type: text/x-patch, Size: 3070 bytes --]
===== arch/ppc/kernel/cputable.c 1.32 vs edited =====
--- 1.32/arch/ppc/kernel/cputable.c Sat Jun 19 03:39:25 2004
+++ edited/arch/ppc/kernel/cputable.c Sat Jul 31 16:51:34 2004
@@ -263,7 +263,7 @@
CPU_FTR_COMMON |
CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB |
CPU_FTR_L2CR | CPU_FTR_ALTIVEC_COMP | CPU_FTR_L3CR |
- CPU_FTR_HPTE_TABLE | CPU_FTR_SPEC7450,
+ CPU_FTR_HPTE_TABLE | CPU_FTR_SPEC7450 | CPU_FTR_NEED_COHERENT,
COMMON_PPC | PPC_FEATURE_ALTIVEC_COMP,
32, 32,
__setup_cpu_745x
@@ -274,7 +274,7 @@
CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | CPU_FTR_CAN_NAP |
CPU_FTR_L2CR | CPU_FTR_ALTIVEC_COMP | CPU_FTR_L3CR |
CPU_FTR_HPTE_TABLE | CPU_FTR_SPEC7450 | CPU_FTR_NAP_DISABLE_L2_PR |
- CPU_FTR_L3_DISABLE_NAP,
+ CPU_FTR_L3_DISABLE_NAP | CPU_FTR_NEED_COHERENT,
COMMON_PPC | PPC_FEATURE_ALTIVEC_COMP,
32, 32,
__setup_cpu_745x
@@ -284,7 +284,8 @@
CPU_FTR_COMMON |
CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | CPU_FTR_CAN_NAP |
CPU_FTR_L2CR | CPU_FTR_ALTIVEC_COMP | CPU_FTR_L3CR |
- CPU_FTR_HPTE_TABLE | CPU_FTR_SPEC7450 | CPU_FTR_NAP_DISABLE_L2_PR,
+ CPU_FTR_HPTE_TABLE | CPU_FTR_SPEC7450 | CPU_FTR_NAP_DISABLE_L2_PR |
+ CPU_FTR_NEED_COHERENT,
COMMON_PPC | PPC_FEATURE_ALTIVEC_COMP,
32, 32,
__setup_cpu_745x
@@ -294,7 +295,8 @@
CPU_FTR_COMMON |
CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB |
CPU_FTR_L2CR | CPU_FTR_ALTIVEC_COMP | CPU_FTR_L3CR |
- CPU_FTR_HPTE_TABLE | CPU_FTR_SPEC7450 | CPU_FTR_HAS_HIGH_BATS,
+ CPU_FTR_HPTE_TABLE | CPU_FTR_SPEC7450 | CPU_FTR_HAS_HIGH_BATS |
+ CPU_FTR_NEED_COHERENT,
COMMON_PPC | PPC_FEATURE_ALTIVEC_COMP,
32, 32,
__setup_cpu_745x
@@ -305,7 +307,7 @@
CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | CPU_FTR_CAN_NAP |
CPU_FTR_L2CR | CPU_FTR_ALTIVEC_COMP | CPU_FTR_L3CR |
CPU_FTR_HPTE_TABLE | CPU_FTR_SPEC7450 | CPU_FTR_NAP_DISABLE_L2_PR |
- CPU_FTR_L3_DISABLE_NAP | CPU_FTR_HAS_HIGH_BATS,
+ CPU_FTR_L3_DISABLE_NAP | CPU_FTR_HAS_HIGH_BATS | CPU_FTR_NEED_COHERENT,
COMMON_PPC | PPC_FEATURE_ALTIVEC_COMP,
32, 32,
__setup_cpu_745x
@@ -316,7 +318,7 @@
CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | CPU_FTR_CAN_NAP |
CPU_FTR_L2CR | CPU_FTR_ALTIVEC_COMP | CPU_FTR_L3CR |
CPU_FTR_HPTE_TABLE | CPU_FTR_SPEC7450 | CPU_FTR_NAP_DISABLE_L2_PR |
- CPU_FTR_HAS_HIGH_BATS,
+ CPU_FTR_HAS_HIGH_BATS | CPU_FTR_NEED_COHERENT,
COMMON_PPC | PPC_FEATURE_ALTIVEC_COMP,
32, 32,
__setup_cpu_745x
@@ -327,7 +329,7 @@
CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | CPU_FTR_CAN_NAP |
CPU_FTR_L2CR | CPU_FTR_ALTIVEC_COMP | CPU_FTR_L3CR |
CPU_FTR_HPTE_TABLE | CPU_FTR_SPEC7450 | CPU_FTR_NAP_DISABLE_L2_PR |
- CPU_FTR_HAS_HIGH_BATS,
+ CPU_FTR_HAS_HIGH_BATS | CPU_FTR_NEED_COHERENT,
COMMON_PPC | PPC_FEATURE_ALTIVEC_COMP,
32, 32,
__setup_cpu_745x
@@ -338,7 +340,7 @@
CPU_FTR_SPLIT_ID_CACHE | CPU_FTR_USE_TB | CPU_FTR_CAN_NAP |
CPU_FTR_L2CR | CPU_FTR_ALTIVEC_COMP |
CPU_FTR_HPTE_TABLE | CPU_FTR_SPEC7450 | CPU_FTR_NAP_DISABLE_L2_PR |
- CPU_FTR_HAS_HIGH_BATS,
+ CPU_FTR_HAS_HIGH_BATS | CPU_FTR_NEED_COHERENT,
COMMON_PPC | PPC_FEATURE_ALTIVEC_COMP,
32, 32,
__setup_cpu_745x
next reply other threads:[~2004-07-31 16:27 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2004-07-31 16:27 Adrian Cox [this message]
2004-08-01 2:53 ` [PATCH] Workaround for 745x data corruption bug Benjamin Herrenschmidt
2004-08-02 15:20 ` Kumar Gala
2004-08-02 21:40 ` Benjamin Herrenschmidt
2004-08-02 18:20 ` Mark A. Greer
2004-08-02 21:47 ` Sven Luther
2004-08-03 13:17 ` Brian Waite
2004-08-03 21:55 ` Mark A. Greer
2004-08-04 14:37 ` Brian Waite
2004-08-04 17:55 ` Mark A. Greer
2004-08-04 20:39 ` Adrian Cox
2004-08-04 0:45 ` Benjamin Herrenschmidt
2004-08-04 17:38 ` Brian Waite
2004-08-04 22:43 ` Benjamin Herrenschmidt
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1091291276.987.57.camel@localhost \
--to=adrian@humboldt.co.uk \
--cc=linuxppc-dev@lists.linuxppc.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.