Prevent trident driver from grabbing pcnet32 hardware
Subsystems (3)
PATCH 1/1 Prevent trident driver from grabbing pcnet32 hardware
Some pcnet32 hardware erroneously has the Vendor ID for Trident. The pcnet32 driver looks for the PCI ethernet class before grabbing the hardware, but the current trident driver does not check against the PCI audio class. This allows the trident driver to claim the pcnet32 hardware. This patch prevents that.
This patch is untested on Trident 4DWAVE_DX hardware, but has been tested on pcnet32 hardware.
Thanks,
Jon
sound/oss/trident.c +2 -1
@@ -279,7 +279,8 @@static struct pci_device_id trident_pci_tbl[] = { {PCI_VENDOR_ID_TRIDENT ,PCI_DEVICE_ID_TRIDENT_4DWAVE_DX , -PCI_ANY_ID ,PCI_ANY_ID ,0 ,0 ,TRIDENT_4D_DX }, +PCI_ANY_ID ,PCI_ANY_ID ,PCI_CLASS_MULTIMEDIA_AUDIO <<8 ,0xffff00 , +TRIDENT_4D_DX }, {PCI_VENDOR_ID_TRIDENT ,PCI_DEVICE_ID_TRIDENT_4DWAVE_NX ,PCI_ANY_ID ,PCI_ANY_ID ,0 ,0 ,TRIDENT_4D_NX }, {PCI_VENDOR_ID_SI ,PCI_DEVICE_ID_SI_7018 ,
Jon Mason napsal(a):
sound/oss/trident.c
Quoted 17 lines
Some pcnet32 hardware erroneously has the Vendor ID for Trident. The pcnet32 driver looks for the PCI ethernet class before grabbing the hardware, but the current trident driver does not check against the PCI audio class. This allows the trident driver to claim the pcnet32 hardware. This patch prevents that.
This patch is untested on Trident 4DWAVE_DX hardware, but has been tested on pcnet32 hardware.
Thanks,
JonSigned-off-by Jon Mason <[email protected]>diff -r
4a7597b41d25sound/oss/trident.csound/oss/trident.c +0
-0
You should change alsa driver (sound/pci/trident/trident.c), rather than this, which will be removed soon, I guess. And, additionally, could you change that lines to use PCI_DEVICE macro?
regards,
Signature · 3 lines
Jiri Slaby www.fi.muni.cz/~xslaby ~\-/~ [email protected] ~\-/~ B67499670407CE62ACC8 22A032CC55C339D47A7E
Jiri Slaby napsal(a):
Quoted 2 lines
which will be removed soon, I guess. And, additionally, could you change that lines to use
PCI_DEVICEmacro?
plus in alsa driver converting to use macros from pci_ids for devices instead of numbers.
thanks,
Signature · 3 lines
Jiri Slaby www.fi.muni.cz/~xslaby ~\-/~ [email protected] ~\-/~ B67499670407CE62ACC8 22A032CC55C339D47A7E
Quoted 3 lines · Jiri Slaby
You should change alsa driver (
sound/pci/trident/trident.c), rather than this, which will be removed soon, I guess. And, additionally, could you change that lines to usePCI_DEVICEmacro?
This driver is not up for removal soon, as it supports a device that the alsa driver apparently doesn't (the INTERG_5050). As for PCI_DEVICE, agreed. Jon, feel like patching it up?
Cheers,
Muli
Signature · 2 lines
Muli Ben-Yehuda http://www.mulix.org | http://mulix.livejournal.com/
Quoted 8 lines · Muli Ben-Yehuda
Quoted 3 lines · Jiri Slaby
You should change alsa driver (
sound/pci/trident/trident.c), rather than this, which will be removed soon, I guess. And, additionally, could you change that lines to usePCI_DEVICEmacro?This driver is not up for removal soon, as it supports a device that the alsa driver apparently doesn't (the
INTERG_5050).
When were you going to report this?
Lee
Quoted 11 lines · Lee Revell
Quoted 8 lines · Muli Ben-Yehuda
Quoted 3 lines · Jiri Slaby
You should change alsa driver (
sound/pci/trident/trident.c), rather than this, which will be removed soon, I guess. And, additionally, could you change that lines to usePCI_DEVICEmacro?This driver is not up for removal soon, as it supports a device that the alsa driver apparently doesn't (the
INTERG_5050).When were you going to report this?
I already reported this as ALSA bug #1293.
The problem is the lack of a tester, and I'm currently inclined to schedule this driver for removal although this PCI ID is still missing - either noone is using these settop boxes anymore or someone will volunteer to test patches.
Quoted 1 line
Lee
cu
Adrian
Signature · 4 lines
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon SeedQuoted 11 lines · Adrian Bunk
Quoted 10 lines · Lee Revell
Quoted 8 lines · Muli Ben-Yehuda
Quoted 3 lines · Jiri Slaby
You should change alsa driver (
sound/pci/trident/trident.c), rather than this, which will be removed soon, I guess. And, additionally, could you change that lines to usePCI_DEVICEmacro?This driver is not up for removal soon, as it supports a device that the alsa driver apparently doesn't (the
INTERG_5050).
Quoted 8 lines
Quoted 1 line
When were you going to report this?
I already reported this as ALSA bug #1293.
The problem is the lack of a tester, and I'm currently inclined to schedule this driver for removal although this PCI ID is still missing - either noone is using these settop boxes anymore or someone will volunteer to test patches.
Hmm, yeah, if the hardware was only used in set top boxes there's not a lot that can be done, unless someone wants to donate one to the ALSA project.
It might be as simple as adding the PCI IDs to the driver.
Lee
Quoted 21 lines · Lee Revell
Quoted 11 lines · Adrian Bunk
Quoted 10 lines · Lee Revell
Quoted 8 lines · Muli Ben-Yehuda
Quoted 3 lines · Jiri Slaby
You should change alsa driver (
sound/pci/trident/trident.c), rather than this, which will be removed soon, I guess. And, additionally, could you change that lines to usePCI_DEVICEmacro?This driver is not up for removal soon, as it supports a device that the alsa driver apparently doesn't (the
INTERG_5050).Quoted 8 lines
Quoted 1 line
When were you going to report this?
I already reported this as ALSA bug #1293.
The problem is the lack of a tester, and I'm currently inclined to schedule this driver for removal although this PCI ID is still missing - either noone is using these settop boxes anymore or someone will volunteer to test patches.
Quoted 5 lines
Hmm, yeah, if the hardware was only used in set top boxes there's not a lot that can be done, unless someone wants to donate one to the ALSA project.
It might be as simple as adding the PCI IDs to the driver.
I agree, add the IDs to the ALSA driver and if someone using a INTERG_5050 adapter has problems....instant tester.
Quoted 21 lines · Lee Revell
Quoted 11 lines · Adrian Bunk
Quoted 10 lines · Lee Revell
Quoted 8 lines · Muli Ben-Yehuda
Quoted 3 lines · Jiri Slaby
You should change alsa driver (
sound/pci/trident/trident.c), rather than this, which will be removed soon, I guess. And, additionally, could you change that lines to usePCI_DEVICEmacro?This driver is not up for removal soon, as it supports a device that the alsa driver apparently doesn't (the
INTERG_5050).Quoted 8 lines
Quoted 1 line
When were you going to report this?
I already reported this as ALSA bug #1293.
The problem is the lack of a tester, and I'm currently inclined to schedule this driver for removal although this PCI ID is still missing - either noone is using these settop boxes anymore or someone will volunteer to test patches.
Quoted 5 lines
Hmm, yeah, if the hardware was only used in set top boxes there's not a lot that can be done, unless someone wants to donate one to the ALSA project.
It might be as simple as adding the PCI IDs to the driver.
To the snd-trident ALSA driver or to the snd-ali5451 ALSA driver?
Seriously, if it would be _that_ trivial, I'd have sent such a patch instead of opening the ALSA bug.
There's a serious amount of 5050 specific code in the OSS driver, and I'd therefore prefer "unsopported in ALSA" over "blindly adding PCI ID and most likely non-working".
Quoted 1 line
Lee
cu
Adrian
Signature · 4 lines
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon SeedQuoted 9 lines · Muli Ben-Yehuda
Quoted 3 lines · Jiri Slaby
You should change alsa driver (
sound/pci/trident/trident.c), rather than this, which will be removed soon, I guess. And, additionally, could you change that lines to usePCI_DEVICEmacro?This driver is not up for removal soon, as it supports a device that the alsa driver apparently doesn't (the
INTERG_5050). As forPCI_DEVICE, agreed. Jon, feel like patching it up?
Patches to follow.
After looking at the ALSA driver, it doesn't support PCI IDs for ALI_5451 or CYBER5050. Someone should look into porting any necessary code from sound/oss/trident.c to sound/pci/trident/trident.c
Thanks,
Jon
Quoted 11 lines · Jon Mason
Quoted 9 lines · Muli Ben-Yehuda
Quoted 3 lines · Jiri Slaby
You should change alsa driver (
sound/pci/trident/trident.c), rather than this, which will be removed soon, I guess. And, additionally, could you change that lines to usePCI_DEVICEmacro?This driver is not up for removal soon, as it supports a device that the alsa driver apparently doesn't (the
INTERG_5050). As forPCI_DEVICE, agreed. Jon, feel like patching it up?
Quoted 5 lines
Patches to follow.
After looking at the ALSA driver, it doesn't support PCI IDs for
ALI_5451or CYBER5050. Someone should look into porting any necessary code fromsound/oss/trident.ctosound/pci/trident/trident.c
CYBER5050 is discussed in ALSA bug #1293 (tester wanted). ALI_5451 is supported by the snd-ali5451 driver.
Quoted 2 lines
Thanks,
Jon
cu
Adrian
Signature · 4 lines
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon SeedQuoted 6 lines
Quoted 3 lines
After looking at the ALSA driver, it doesn't support PCI IDs for
ALI_5451or CYBER5050. Someone should look into porting any necessary code fromsound/oss/trident.ctosound/pci/trident/trident.cCYBER5050 is discussed in ALSA bug #1293 (tester wanted).
ALI_5451is supported by the snd-ali5451 driver.
And I have a (working) ALI 5451; for whoever wants to know that.
Jan Engelhardt
Quoted 12 lines · Adrian Bunk
Quoted 11 lines · Jon Mason
Quoted 9 lines · Muli Ben-Yehuda
Quoted 3 lines · Jiri Slaby
You should change alsa driver (
sound/pci/trident/trident.c), rather than this, which will be removed soon, I guess. And, additionally, could you change that lines to usePCI_DEVICEmacro?This driver is not up for removal soon, as it supports a device that the alsa driver apparently doesn't (the
INTERG_5050). As forPCI_DEVICE, agreed. Jon, feel like patching it up?
Quoted 8 lines
Quoted 5 lines
Patches to follow.
After looking at the ALSA driver, it doesn't support PCI IDs for
ALI_5451or CYBER5050. Someone should look into porting any necessary code fromsound/oss/trident.ctosound/pci/trident/trident.cCYBER5050 is discussed in ALSA bug #1293 (tester wanted).
ALI_5451is supported by the snd-ali5451 driver.
Sorry, I grepped for PCI_DEVICE_ID_ALI_5451 not 0x5451. Sending a patch to fix that up too.
Thanks,
Jon
Quoted 1 line · Adrian Bunk
CYBER5050 is discussed in ALSA bug #1293 (tester wanted).
OK I set that bug to FEEDBACK, but it's open 5 months now and no testers are forthcoming. I think if we don't find one as a result of this thread we can assume no one cares about this hardware anymore.
I'm still not sure that just adding it to the ALSA driver and hoping it works is the best solution. Would we rather users see right away that their hardware isn't supported, or have the driver load and get no sound or hang the machine?
I think the best approach might just be to drop it in lieu of a tester. It will be trivial to add support later if someone finds one of these boxes.
Lee
Quoted 6 lines · Lee Revell
Quoted 1 line · Adrian Bunk
CYBER5050 is discussed in ALSA bug #1293 (tester wanted).
OK I set that bug to FEEDBACK, but it's open 5 months now and no testers are forthcoming. I think if we don't find one as a result of this thread we can assume no one cares about this hardware anymore.
The majority of Linux users doesn't read linux-kernel...
We might find users after the OSS driver is deprecated in a released kernel, or perhaps some months after it's removed from the kernel.
This would match my current experiences regarding my suggested removal of some OSS drivers.
Quoted 8 lines
I'm still not sure that just adding it to the ALSA driver and hoping it works is the best solution. Would we rather users see right away that their hardware isn't supported, or have the driver load and get no sound or hang the machine?
I think the best approach might just be to drop it in lieu of a tester. It will be trivial to add support later if someone finds one of these boxes.
Agreed.
Quoted 1 line
Lee
cu
Adrian
Signature · 4 lines
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon SeedQuoted 11 lines · Adrian Bunk
Quoted 8 lines
I'm still not sure that just adding it to the ALSA driver and hoping it works is the best solution. Would we rather users see right away that their hardware isn't supported, or have the driver load and get no sound or hang the machine?
I think the best approach might just be to drop it in lieu of a tester. It will be trivial to add support later if someone finds one of these boxes.
Agreed.
OK I'm just going to close that bug, the one person who seemed to know anything about it had this to say:
"The cyberpro 5050 is an old combined video+audio controller - and is/was used in some settop boxes (German Siemens Activity and also Loewe).
There will be no "desktop users" around.
And because I don't work for Loewe anymore: I don't have access to hardware.
If you ask me: leave it out!"
Anyone who finds one of these is free to reopen it.
Lee
Quoted 5 lines · Lee Revell
"The cyberpro 5050 is an old combined video+audio controller - and is/was used in some settop boxes (German Siemens Activity and also Loewe).
There will be no "desktop users" around.
The 5050 is used with some older geode systems. It uses the same core (from T-squared I believe) as trident and the other vendors. Its not very different to the 4DWAVE.
Alan
Old hardware (was Re: [PATCH] Prevent trident driver from grabbing pcnet32 hardware)
In reply to Adrian BunkQuoted 9 lines · Adrian Bunk
Quoted 4 lines · Lee Revell
Quoted 1 line · Adrian Bunk
CYBER5050 is discussed in ALSA bug #1293 (tester wanted).
OK I set that bug to FEEDBACK, but it's open 5 months now and no
testers
Quoted 2 lines
are forthcoming. I think if we don't find one as a result of this thread we can assume no one cares about this hardware anymore.
Quoted 8 lines
The majority of Linux users doesn't read linux-kernel...
We might find users after the OSS driver is deprecated in a released kernel, or perhaps some months after it's removed from the kernel.
This would match my current experiences regarding my suggested removal
of some OSS drivers.
While we're on the topic, what *is* the best place to solicit hardware donations for purposes of driver development?
I ask because the snd-nm256 driver is one that we know users have, but the ALSA driver has never worked (sound OK but the machine frequently locks up IIRC, see ALSA bug #305), and the device has been unavailable for years. Basically these users are screwed if we can't get a hardware sample, but I think that might mean someone has to give us a whole laptop.
Lee
Re: Old hardware (was Re: [PATCH] Prevent trident driver from grabbing pcnet32 hardware)
Quoted 6 lines · Lee Revell
I ask because the snd-nm256 driver is one that we know users have, but the ALSA driver has never worked (sound OK but the machine frequently locks up IIRC, see ALSA bug #305), and the device has been unavailable for years. Basically these users are screwed if we can't get a hardware sample, but I think that might mean someone has to give us a whole laptop.
I happen to have an old laptop with an nm256, that I would gladly part with, but it doesn't boot because of a hardware problem: it will shut itself off between 5 and 30 seconds of turning on. Last time I checked it seemed to be a not-worth-fixing problem with a lot of VAIOs of that generation. I doubt this is worth anything to anyone but if it is, let me know.
-Bob
Quoted 3 lines · Lee Revell
OK I set that bug to FEEDBACK, but it's open 5 months now and no testers are forthcoming. I think if we don't find one as a result of this thread we can assume no one cares about this hardware anymore.
Nobody reads ALSA bugzilla. You are probably right but its not clear, and if the audio is going the video component might as well go too.
Have you checked with the ARM people or the person who added support ( [email protected]) ?
It's also used in various appliances people "adjust" to run Linux - the Philips AOL-TV for example. Also on the SH reference boards for the hitachi sh series cpus (arch/sh*).
Quoted 3 lines
I think the best approach might just be to drop it in lieu of a tester. It will be trivial to add support later if someone finds one of these boxes.
Think it should stay for now at least. Pointing at ALSA bugzilla as proof of non-use is not a good test.
Alan
Quoted 7 lines · Alan Cox
Have you checked with the ARM people or the person who added support ( [email protected]) ?
It's also used in various appliances people "adjust" to run Linux - the Philips AOL-TV for example. Also on the SH reference boards for the hitachi sh series cpus (
arch/sh*).
Yes he was the person I quoted in my previous message, who recommended we drop it, but it looks like this device is more widely used than we thought.
I agree that if there's any doubt then we should not remove code.
Lee
Quoted 11 lines · Lee Revell
Quoted 1 line · Adrian Bunk
CYBER5050 is discussed in ALSA bug #1293 (tester wanted).
OK I set that bug to FEEDBACK, but it's open 5 months now and no testers are forthcoming. I think if we don't find one as a result of this thread we can assume no one cares about this hardware anymore.
I'm still not sure that just adding it to the ALSA driver and hoping it works is the best solution. Would we rather users see right away that their hardware isn't supported, or have the driver load and get no sound or hang the machine?
... or use the known working OSS driver?
Quoted 3 lines
I think the best approach might just be to drop it in lieu of a tester. It will be trivial to add support later if someone finds one of these boxes.
So you advocate removing a known working driver in favor of a known not to be working driver? nice.
Cheers,
Muli
Signature · 2 lines
Muli Ben-Yehuda http://www.mulix.org | http://mulix.livejournal.com/
Quoted 8 lines · Muli Ben-Yehuda
Quoted 7 lines · Lee Revell
Quoted 1 line · Adrian Bunk
CYBER5050 is discussed in ALSA bug #1293 (tester wanted).
OK I set that bug to FEEDBACK, but it's open 5 months now and no testers are forthcoming. I think if we don't find one as a result of this thread we can assume no one cares about this hardware anymore.
Quoted 7 lines
Quoted 4 lines
I'm still not sure that just adding it to the ALSA driver and hoping it works is the best solution. Would we rather users see right away that their hardware isn't supported, or have the driver load and get no sound or hang the machine?
... or use the known working OSS driver?
...
In my experience with scheduling OSS drivers for removal, users simply use the OSS drivers unless you tell them very explicitely that the OSS driver will go.
It shouldn't be too hard to port the support to ALSA if someone with the hardware is willing to test patches.
The goal is to get people still using OSS drivers where ALSA drivers support the same hardware to use the ALSA drivers - and if there were bugs in the ALSA drivers preventing them to switch to ALSA, to report them to the ALSA bug tracking system.
This has the following advantages:
- better ALSA drivers
- get rid of some unmaintained code in the kernel
Quoted 2 lines
Cheers,
Muli
cu
Adrian
Signature · 4 lines
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon SeedQuoted 3 lines · Adrian Bunk
In my experience with scheduling OSS drivers for removal, users simply use the OSS drivers unless you tell them very explicitely that the OSS driver will go.
If the OSS drivers satisfy them, what's wrong with it?
Quoted 2 lines
It shouldn't be too hard to port the support to ALSA if someone with the hardware is willing to test patches.
Unfortunately, I have a different trident variant (the 5451).
Quoted 7 lines
The goal is to get people still using OSS drivers where ALSA drivers support the same hardware to use the ALSA drivers - and if there were bugs in the ALSA drivers preventing them to switch to ALSA, to report them to the ALSA bug tracking system.
This has the following advantages:
- better ALSA drivers
This one is fine.
Quoted 1 line
- get rid of some unmaintained code in the kernel
This one is irrelevant for trident - I maintain it.
Cheers,
Muli
Signature · 2 lines
Muli Ben-Yehuda http://www.mulix.org | http://mulix.livejournal.com/
Quoted 8 lines · Muli Ben-Yehuda
Quoted 3 lines · Adrian Bunk
In my experience with scheduling OSS drivers for removal, users simply use the OSS drivers unless you tell them very explicitely that the OSS driver will go.
If the OSS drivers satisfy them, what's wrong with it?
...
There are two drivers, each with it's own features and bugs.
Either the ALSA driver is better in any respect making the OSS driver simply obsolete, or there are problems in the ALSA driver that should be reported and fixed.
Removing the OSS driver forces users to report the problems with the ALSA driver making the latter better for everyone.
Quoted 2 lines
Cheers,
Muli
cu
Adrian
Signature · 4 lines
"Is there not promise of rain?" Ling Tan asked suddenly out
of the darkness. There had been need of rain for many days.
"Only a promise," Lao Er said.
Pearl S. Buck - Dragon SeedQuoted 6 lines · Adrian Bunk
Either the ALSA driver is better in any respect making the OSS driver simply obsolete, or there are problems in the ALSA driver that should be reported and fixed.
Removing the OSS driver forces users to report the problems with the ALSA driver making the latter better for everyone.
Yeah, and rm -rf'ing ftp.kernel.org is a great way to make everyone write a better operating system. Surely there's a less destructive way to do things.
In any case, I think this thread has run its course and we can agree to disagree now.
Cheers,
Muli
Signature · 2 lines
Muli Ben-Yehuda http://www.mulix.org | http://mulix.livejournal.com/
This revised version of the OSS Trident patch includes PCI_DEVICE Macro usage.
Thanks,
Jon
sound/oss/trident.c +8 -10
@@ -278,16 +278,14 @@ };static struct pci_device_id trident_pci_tbl[] = { - {PCI_VENDOR_ID_TRIDENT ,PCI_DEVICE_ID_TRIDENT_4DWAVE_DX , -PCI_ANY_ID ,PCI_ANY_ID ,0 ,0 ,TRIDENT_4D_DX }, - {PCI_VENDOR_ID_TRIDENT ,PCI_DEVICE_ID_TRIDENT_4DWAVE_NX , -PCI_ANY_ID ,PCI_ANY_ID ,0 ,0 ,TRIDENT_4D_NX }, - {PCI_VENDOR_ID_SI ,PCI_DEVICE_ID_SI_7018 , -PCI_ANY_ID ,PCI_ANY_ID ,0 ,0 ,SIS_7018 }, - {PCI_VENDOR_ID_ALI ,PCI_DEVICE_ID_ALI_5451 , -PCI_ANY_ID ,PCI_ANY_ID ,0 ,0 ,ALI_5451 }, - {PCI_VENDOR_ID_INTERG ,PCI_DEVICE_ID_INTERG_5050 , -PCI_ANY_ID ,PCI_ANY_ID ,0 ,0 ,CYBER5050 }, + {PCI_DEVICE (PCI_VENDOR_ID_TRIDENT ,PCI_DEVICE_ID_TRIDENT_4DWAVE_DX ), +PCI_CLASS_MULTIMEDIA_AUDIO <<8 ,0xffff00 ,TRIDENT_4D_DX }, + {PCI_DEVICE (PCI_VENDOR_ID_TRIDENT ,PCI_DEVICE_ID_TRIDENT_4DWAVE_NX ), +0 ,0 ,TRIDENT_4D_NX }, + {PCI_DEVICE (PCI_VENDOR_ID_SI ,PCI_DEVICE_ID_SI_7018 ),0 ,0 ,SIS_7018 }, + {PCI_DEVICE (PCI_VENDOR_ID_ALI ,PCI_DEVICE_ID_ALI_5451 ),0 ,0 ,ALI_5451 }, + {PCI_DEVICE (PCI_VENDOR_ID_INTERG ,PCI_DEVICE_ID_INTERG_5050 ), +0 ,0 ,CYBER5050 }, {0 ,} };
pcnet32 devices with incorrect trident vendor ID
In reply to Jon MasonQuoted 4 lines · Jon Mason
Some pcnet32 hardware erroneously has the Vendor ID for Trident. The
pcnet32 driver looks for the PCI ethernet class before grabbing the
hardware, but the current trident driver does not check against the
PCI audio class. This allows the trident driver to claim the pcnet32 hardware. This patch prevents that.
On the subject of pcnet32 and the invalid vendor ID, you may find this interesting:
http://forums.gentoo.org/viewtopic-t-420013-highlight-trident.html
The user saw the correct vendor ID (AMD) in 2.4, but when upgrading to 2.6, it changed to Trident.
I guess this is still likely to be a hardware bug, but it demonstrates that the Linux PCI layer has something to do with it (even if it is just triggering it somehow).
Daniel
Re: pcnet32 devices with incorrect trident vendor ID
Quoted 6 lines · Daniel Drake
interesting:
http://forums.gentoo.org/viewtopic-t-420013-highlight-trident.html
The user saw the correct vendor ID (AMD) in 2.4, but when upgrading to 2.6, it changed to Trident.
It looks to me like there used to be a quirk that knew about this bug and fixed it.
The reason I say this is that the lspci -x dumps are the same -- both featuring the wrong vendor ID. Want to dig through 2.4 and look for this quirk?
Re: pcnet32 devices with incorrect trident vendor ID
Quoted 5 lines · Matthew Wilcox
It looks to me like there used to be a quirk that knew about this bug and fixed it.
The reason I say this is that the lspci -x dumps are the same -- both featuring the wrong vendor ID.
Good catch. I didn't notice this.
Quoted 2 lines
Want to dig through 2.4 and look for
this quirk?
No time for this personally.
Daniel
Re: pcnet32 devices with incorrect trident vendor ID
In reply to Matthew WilcoxQuoted 8 lines · Matthew Wilcox
Quoted 6 lines · Daniel Drake
interesting:
http://forums.gentoo.org/viewtopic-t-420013-highlight-trident.html
The user saw the correct vendor ID (AMD) in 2.4, but when upgrading to 2.6, it changed to Trident.
Quoted 6 lines
It looks to me like there used to be a quirk that knew about this bug and fixed it.
The reason I say this is that the lspci -x dumps are the same -- both featuring the wrong vendor ID. Want to dig through 2.4 and look for this quirk?
Oh -- found it. It's still in 2.6:
static voidfixup_broken_pcnet32(struct pci_dev* dev)
{
if ((dev->class>>8 == PCI_CLASS_NETWORK_ETHERNET)) {
dev->vendor = PCI_VENDOR_ID_AMD;
pci_write_config_word(dev, PCI_VENDOR_ID, PCI_VENDOR_ID_AMD);
}
}DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_TRIDENT, PCI_ANY_ID, fixup_broken_pcnet32);
Wonder why it isn't working now ... someone with a PPC box needs to check
(a) whether this function is being called and (b) if it is called, why it's not doing what it's supposed to.
Re: pcnet32 devices with incorrect trident vendor ID
Quoted 26 lines · Matthew Wilcox
Quoted 8 lines · Matthew Wilcox
Quoted 6 lines · Daniel Drake
interesting:
http://forums.gentoo.org/viewtopic-t-420013-highlight-trident.html
The user saw the correct vendor ID (AMD) in 2.4, but when upgrading to 2.6, it changed to Trident.
Quoted 6 lines
It looks to me like there used to be a quirk that knew about this bug and fixed it.
The reason I say this is that the lspci -x dumps are the same -- both featuring the wrong vendor ID. Want to dig through 2.4 and look for this quirk?
Oh -- found it. It's still in 2.6:
static void
fixup_broken_pcnet32(struct pci_dev* dev){ if ((dev->class>>8 == PCI_CLASS_NETWORK_ETHERNET)) { dev->vendor = PCI_VENDOR_ID_AMD; pci_write_config_word(dev, PCI_VENDOR_ID, PCI_VENDOR_ID_AMD); } }
Quoted 6 lines
DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_TRIDENT, PCI_ANY_ID, fixup_broken_pcnet32);Wonder why it isn't working now ... someone with a PPC box needs to check
(a) whether this function is being called and (b) if it is called, why it's not doing what it's supposed to.
I remember looking at this a while back. I think the corrected information is only being propagated to the 'vendor' file, and the write_config_word isn't actually updating the 'config' entry in sysfs.
If you remove the "#if 0" from lib/sysfs.c in pciutils, it should start reporting the corrected value in base lspci, etc.
Bill
Re: pcnet32 devices with incorrect trident vendor ID
Quoted 3 lines · Bill Nottingham
I remember looking at this a while back. I think the corrected information is only being propagated to the 'vendor' file, and the
write_config_wordisn't actually updating the 'config' entry in sysfs.
Ah, here's an example from the box I remember working on this on (without the libpci change):
root@pseries 0000:00:0c.0]# pwd /sys/bus/pci/devices/0000:00:0c.0
[root@pseries 0000:00:0c.0]# lspci | grep 0c.0
00:0c.0 Ethernet controller: Trident Microsystems 4DWave DX (rev 26) [root@pseries 0000:00:0c.0]# lspci -n | grep 0c.0
00:0c.0 Class 0200: 1023:2000 (rev 26)
[root@pseries 0000:00:0c.0]# cat vendor0x1022
[root@pseries 0000:00:0c.0]# cat device0x2000
[root@pseries 0000:00:0c.0]# od -x config
0000000 2310 0020 4701 8002 2600 0002 0048 0000
0000020 01f0 ff00 0070 21c3 0000 0000 0000 0000
0000040 0000 0000 0000 0000 0000 0000 0000 0000
0000060 0000 10c3 0000 0000 0000 0000 1101 06ff
0000100 0000 0000 0000 0000 0000 0000 0000 0000
*
0000400
Note that the config space is different than the vendor file. This was with 2.6.9-ish, I don't have the box around any more to confirm with something more recent.
Bill
Re: pcnet32 devices with incorrect trident vendor ID
There are just a few differences between the 2.4 (1.30h) version of pcnet32.c and the 2.6 (1.30j) version, as I have tried to keep them as similar as possible.
The driver was changed in February 2004 to recognize the incorrect vendor ID so that the ppc workaround was no longer required, and so that the cards would work in non ppc systems. From the driver perspective the ppc workaround could be deleted.
On my systems lspci shows all the devices with the correct name whether it is ppc or x86.
PPC:
donf@elm3b48:/usr/src> lspci | grep Ethernet
0000:01:01.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet32 LANCE] (rev 54) 0000:21:01.0 Ethernet controller: Intel Corporation 82557/8/9 [Ethernet Pro 100] (rev 0d) 0000:62:00.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet32 LANCE] (rev 26) 0000:62:01.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet32 LANCE] (rev 26) 0000:62:02.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet32 LANCE] (rev 26) 0000:62:03.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet32 LANCE] (rev 26) 0001:21:01.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet32 LANCE] (rev 44) 0001:31:01.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet32 LANCE] (rev 26) 0001:41:01.0 Ethernet controller: Intel Corporation 82545EM Gigabit Ethernet Controller (Copper) (rev 01) 0001:61:01.0 Ethernet controller: Advanced Micro Devices [AMD] 79c978 [HomePNA] (rev 51) donf@elm3b48:/usr/src>
x86:
[donf@elm3b45 linux-2.6.14-git11]$ lspci | grep Ethernet
00:01.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet LANCE] (rev 54) 00:05.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet LANCE] (rev 44) 02:05.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet LANCE] (rev 16) 02:06.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet LANCE] (rev 36) 05:02.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet LANCE] (rev 44) 05:03.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet LANCE] (rev 54) 05:04.0 Ethernet controller: Advanced Micro Devices [AMD] 79c978 [HomePNA] (rev 51) [donf@elm3b45 linux-2.6.14-git11]$
Quoted 32 lines · Bill Nottingham
Quoted 3 lines · Bill Nottingham
I remember looking at this a while back. I think the corrected information is only being propagated to the 'vendor' file, and the
write_config_wordisn't actually updating the 'config' entry in sysfs.Ah, here's an example from the box I remember working on this on (without the libpci change):
root@pseries 0000:00:0c.0]# pwd /sys/bus/pci/devices/0000:00:0c.0
[root@pseries 0000:00:0c.0]# lspci | grep 0c.0
00:0c.0 Ethernet controller: Trident Microsystems 4DWave DX (rev 26) [root@pseries 0000:00:0c.0]# lspci -n | grep 0c.0
00:0c.0 Class 0200: 1023:2000 (rev 26)
[root@pseries 0000:00:0c.0]# cat vendor0x1022
[root@pseries 0000:00:0c.0]# cat device0x2000
[root@pseries 0000:00:0c.0]# od -x config
0000000 2310 0020 4701 8002 2600 0002 0048 0000
0000020 01f0 ff00 0070 21c3 0000 0000 0000 0000
0000040 0000 0000 0000 0000 0000 0000 0000 0000
0000060 0000 10c3 0000 0000 0000 0000 1101 06ff
0000100 0000 0000 0000 0000 0000 0000 0000 0000
*
0000400Note that the config space is different than the vendor file. This was with 2.6.9-ish, I don't have the box around any more to confirm with something more recent.
Bill
Signature · 2 lines
Don Fry [email protected]
Re: pcnet32 devices with incorrect trident vendor ID
Quoted 11 lines · Don Fry
There are just a few differences between the 2.4 (1.30h) version of
pcnet32.cand the 2.6 (1.30j) version, as I have tried to keep them as similar as possible.The driver was changed in February 2004 to recognize the incorrect vendor ID so that the ppc workaround was no longer required, and so that the cards would work in non ppc systems. From the driver perspective the ppc workaround could be deleted.
On my systems lspci shows all the devices with the correct name whether it is ppc or x86.
Funny, I get the following on my opteron system:
# lspci | grep Ethernet 0000:03:02.0 Ethernet controller: Broadcom Corporation NetXtreme BCM5703X Gigabit Ethernet (rev 02) 0000:0e:01.0 Ethernet controller: Trident Microsystems 4DWave DX (rev 26)
Quoted 63 lines
PPC:
donf@elm3b48:/usr/src> lspci | grep Ethernet
0000:01:01.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet32 LANCE] (rev 54) 0000:21:01.0 Ethernet controller: Intel Corporation 82557/8/9 [Ethernet Pro 100] (rev 0d) 0000:62:00.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet32 LANCE] (rev 26) 0000:62:01.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet32 LANCE] (rev 26) 0000:62:02.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet32 LANCE] (rev 26) 0000:62:03.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet32 LANCE] (rev 26) 0001:21:01.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet32 LANCE] (rev 44) 0001:31:01.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet32 LANCE] (rev 26) 0001:41:01.0 Ethernet controller: Intel Corporation 82545EM Gigabit Ethernet Controller (Copper) (rev 01) 0001:61:01.0 Ethernet controller: Advanced Micro Devices [AMD] 79c978 [HomePNA] (rev 51) donf@elm3b48:/usr/src>x86:
[donf@elm3b45 linux-2.6.14-git11]$ lspci | grep Ethernet
00:01.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet LANCE] (rev 54) 00:05.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet LANCE] (rev 44) 02:05.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet LANCE] (rev 16) 02:06.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet LANCE] (rev 36) 05:02.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet LANCE] (rev 44) 05:03.0 Ethernet controller: Advanced Micro Devices [AMD] 79c970 [PCnet LANCE] (rev 54) 05:04.0 Ethernet controller: Advanced Micro Devices [AMD] 79c978 [HomePNA] (rev 51) [donf@elm3b45 linux-2.6.14-git11]$Quoted 32 lines · Bill Nottingham
Quoted 3 lines · Bill Nottingham
I remember looking at this a while back. I think the corrected information is only being propagated to the 'vendor' file, and the
write_config_wordisn't actually updating the 'config' entry in sysfs.Ah, here's an example from the box I remember working on this on (without the libpci change):
root@pseries 0000:00:0c.0]# pwd /sys/bus/pci/devices/0000:00:0c.0
[root@pseries 0000:00:0c.0]# lspci | grep 0c.0
00:0c.0 Ethernet controller: Trident Microsystems 4DWave DX (rev 26) [root@pseries 0000:00:0c.0]# lspci -n | grep 0c.0
00:0c.0 Class 0200: 1023:2000 (rev 26)
[root@pseries 0000:00:0c.0]# cat vendor0x1022
[root@pseries 0000:00:0c.0]# cat device0x2000
[root@pseries 0000:00:0c.0]# od -x config
0000000 2310 0020 4701 8002 2600 0002 0048 0000
0000020 01f0 ff00 0070 21c3 0000 0000 0000 0000
0000040 0000 0000 0000 0000 0000 0000 0000 0000
0000060 0000 10c3 0000 0000 0000 0000 1101 06ff
0000100 0000 0000 0000 0000 0000 0000 0000 0000
*
0000400Note that the config space is different than the vendor file. This was with 2.6.9-ish, I don't have the box around any more to confirm with something more recent.
Bill
Signature · 2 lines
Don Fry [email protected]
Re: pcnet32 devices with incorrect trident vendor ID
In reply to Daniel DrakeQuoted 11 lines · Daniel Drake
On the subject of pcnet32 and the invalid vendor ID, you may find this interesting:
http://forums.gentoo.org/viewtopic-t-420013-highlight-trident.html
The user saw the correct vendor ID (AMD) in 2.4, but when upgrading to 2.6, it changed to Trident.
I guess this is still likely to be a hardware bug, but it demonstrates that the Linux PCI layer has something to do with it (even if it is just triggering it somehow).
Perhaps there is a significant different in the pcnet32.c files between the two versions. I also remember that there is some powerpc specific code in there related to MAC address detection. There are certainly differences in 2.4 and 2.6's version of the driver, maybe something is broken in the newer one when run on powerpc. I don't run gentoo and have no idea how to get a hold of the source of pcnet32.c from each of those two.
It does seem odd that only the pcnet32 has the pci ID change, but at the same time, somehow the driver is recognizing it and loading at boot time, so the ID can't be wrong at that time. Does the ID get mangled as part of what makes the MAC addresses read incorrectly on your 2.6.14? The 2.4 system shows all the cards overriding the MAC based on the PROM, which I believe is what the driver code should do on a powerpc system. On 2.6 that appears to only happen on one of the cards. At least on that device (pci 01:01) appears to agree what the MAC should be in both cases.
Perhaps the lspci being wrong is just a side effect of the real problem. Maybe the driver is broken and messing things up.
Len Sorensen