| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <20221022164154.kxcqsx5izr5yx5wj@core>
Date: Sat, 22 Oct 2022 18:41:54 +0200
From: Ondřej Jirman <megi@....cz>
To: Peter Geis <pgwipeout@...il.com>
Cc: Mark Kettenis <mark.kettenis@...all.nl>, heiko@...ech.de,
linux-rockchip@...ts.infradead.org, robh+dt@...nel.org,
krzysztof.kozlowski+dt@...aro.org, michael.riesch@...fvision.net,
frattaroli.nicolas@...il.com, s.hauer@...gutronix.de,
frank-w@...lic-files.de, ezequiel@...guardiasur.com.ar,
yifeng.zhao@...k-chips.com, jbx6244@...il.com,
devicetree@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2] arm64: dts: rockchip: rk356x: Fix PCIe register map
and ranges
On Sat, Oct 22, 2022 at 08:19:57AM -0400, Peter Geis wrote:
>
> [...]
>
> reg = <0x3 0xc0000000 0x0 0x00400000>,
> <0x0 0xfe260000 0x0 0x00010000>,
> <0x0 0xf4000000 0x0 0x00100000>;
>
> ranges = <0x01000000 0x0 0xf4100000 0x0 0xf4100000 0x0 0x00100000>,
> <0x02000000 0x0 0xf4200000 0x0 0xf4200000 0x0 0x01e00000>,
> <0x03000000 0x0 0x40000000 0x3 0x00000000 0x0 0x40000000>;
... more data:
Diff of my v2 vs the above ranges for my 2xwifi card + nvme setup:
--- switch.lspciv 2022-10-05 10:36:33.924838688 +0200
+++ switch-pg.lspciv 2022-10-22 18:30:33.412025097 +0200
@@ -5 +5 @@
- Memory behind bridge: 00000000-002fffff [size=3M] [32-bit]
+ Memory behind bridge: f4200000-f44fffff [size=3M] [32-bit]
@@ -7 +7 @@
- Expansion ROM at 300300000 [virtual] [disabled] [size=64K]
+ Expansion ROM at f4500000 [virtual] [disabled] [size=64K]
@@ -22 +22 @@
- Memory behind bridge: 00000000-002fffff [size=3M] [32-bit]
+ Memory behind bridge: f4200000-f44fffff [size=3M] [32-bit]
@@ -38 +38 @@
- Memory behind bridge: 00000000-000fffff [size=1M] [32-bit]
+ Memory behind bridge: f4200000-f42fffff [size=1M] [32-bit]
@@ -53 +53 @@
- Memory behind bridge: 00100000-001fffff [size=1M] [32-bit]
+ Memory behind bridge: f4300000-f43fffff [size=1M] [32-bit]
@@ -83 +83 @@
- Memory behind bridge: 00200000-002fffff [size=1M] [32-bit]
+ Memory behind bridge: f4400000-f44fffff [size=1M] [32-bit]
@@ -96 +96 @@
- Memory at 300000000 (64-bit, non-prefetchable) [size=16K]
+ Memory at f4200000 (64-bit, non-prefetchable) [size=16K]
@@ -111 +111 @@
- Memory at 300100000 (64-bit, non-prefetchable) [size=64K]
+ Memory at f4300000 (64-bit, non-prefetchable) [size=64K]
@@ -123 +123 @@
- Flags: bus master, fast devsel, latency 0, IRQ 80
+ Flags: bus master, fast devsel, latency 0, IRQ 76
@@ -125 +125 @@
- Memory at 300200000 (64-bit, non-prefetchable) [size=1M]
+ Memory at f4400000 (64-bit, non-prefetchable) [size=1M]
(not so dramatic differences)
But for SATA card + USB card + 2-port intel ethernet card, it's
massively better:
--- fullpci-my.lspciv 2022-10-15 17:16:55.002000065 +0200
+++ fullpci-pg.lspciv 2022-10-15 17:15:09.837000015 +0200
@@ -5 +5 @@
- Memory behind bridge: [disabled] [32-bit]
+ Memory behind bridge: f4200000-f55fffff [size=20M] [32-bit]
@@ -7 +7 @@
- Expansion ROM at 300000000 [virtual] [disabled] [size=64K]
+ Expansion ROM at f5600000 [virtual] [disabled] [size=64K]
@@ -22 +22 @@
- Memory behind bridge: [disabled] [32-bit]
+ Memory behind bridge: f4200000-f55fffff [size=20M] [32-bit]
@@ -38 +38 @@
- Memory behind bridge: [disabled] [32-bit]
+ Memory behind bridge: f5400000-f54fffff [size=1M] [32-bit]
@@ -53 +53 @@
- Memory behind bridge: [disabled] [32-bit]
+ Memory behind bridge: f5500000-f55fffff [size=1M] [32-bit]
@@ -83 +83 @@
- Memory behind bridge: [disabled] [32-bit]
+ Memory behind bridge: f4200000-f53fffff [size=18M] [32-bit]
@@ -95 +95,4 @@
- Flags: fast devsel
+ Flags: bus master, fast devsel, latency 0, IRQ 74
+ Memory at f5480000 (32-bit, non-prefetchable) [size=8K]
+ Memory at f5482000 (32-bit, non-prefetchable) [size=8K]
+ Expansion ROM at f5400000 [virtual] [disabled] [size=512K]
@@ -97 +100 @@
- Capabilities: [50] MSI: Enable- Count=1/1 Maskable- 64bit+
+ Capabilities: [50] MSI: Enable+ Count=1/1 Maskable- 64bit+
@@ -100,0 +104 @@
+ Kernel driver in use: ahci
@@ -104 +108,2 @@
- Flags: fast devsel
+ Flags: bus master, fast devsel, latency 0, IRQ 75
+ Memory at f5500000 (64-bit, non-prefetchable) [size=4K]
@@ -106 +111 @@
- Capabilities: [90] MSI: Enable- Count=1/4 Maskable- 64bit+
+ Capabilities: [90] MSI: Enable+ Count=1/4 Maskable- 64bit+
@@ -108,0 +114 @@
+ Kernel driver in use: xhci_hcd
@@ -112 +118,3 @@
- Flags: fast devsel
+ Flags: bus master, fast devsel, latency 0
+ Memory at f4200000 (32-bit, non-prefetchable) [size=128K]
+ Memory at f4400000 (32-bit, non-prefetchable) [size=4M]
@@ -113,0 +122,2 @@
+ Memory at f4240000 (32-bit, non-prefetchable) [size=16K]
+ Expansion ROM at f4800000 [virtual] [disabled] [size=4M]
@@ -116 +126 @@
- Capabilities: [70] MSI-X: Enable- Count=10 Masked-
+ Capabilities: [70] MSI-X: Enable+ Count=10 Masked-
@@ -121,0 +132 @@
+ Kernel driver in use: igb
@@ -125,2 +136,6 @@
- Flags: fast devsel
- I/O ports at 1020 [disabled] [size=32]
+ Flags: bus master, fast devsel, latency 0, IRQ 85
+ Memory at f4220000 (32-bit, non-prefetchable) [size=128K]
+ Memory at f4c00000 (32-bit, non-prefetchable) [size=4M]
+ I/O ports at 1020 [size=32]
+ Memory at f4284000 (32-bit, non-prefetchable) [size=16K]
+ Expansion ROM at f5000000 [virtual] [disabled] [size=4M]
@@ -128 +143 @@
- Capabilities: [50] MSI: Enable- Count=1/1 Maskable+ 64bit+
+ Capabilities: [50] MSI: Enable+ Count=1/1 Maskable+ 64bit+
@@ -134,0 +150 @@
+ Kernel driver in use: igb
(Full output https://megous.com/dl/tmp/fullpci-pg.lspciv)
So it's still not testing the 0x3_0000_0000 range, but as far as I'm
concerned, it works with whatever I can throw at it (7 different
pcie devices I have and combining them behind a 4-port pcie switch).
The best reg/ranges combination so far. ;)
Tested-by: Ondrej Jirman <megi@....cz>
kind regards,
o.
> Very Respectfully,
> Peter Geis
Powered by blists - more mailing lists