UASP: USB Attached SCSI (UAS) protocol driver
Quoted 4 lines · Matthew Dharm
Your insinuations against the character of many
long-standing Linux
developers, based on their refusal to accept your
submission,
No, not submission. The fact is that no one bothered to review the code. It was simply flat out NAK-ed it in just four minutes, fronting one of their co-workers with the defunct driver already in.
It would've been a different story if Greg, Alan and you and anyone else willing, had provided a technical review, and /then/ reject it--the rejection reason would've been clearer to everyone, than taking all this thread to get to.
Greg didn't review it and rejected it 4 minutes after posting. Alan didn't review it and after I posted the 18 technical reasons (my 2nd post in this thread replying to Greg's) said this:
Quoted 1 line
Your driver may be the best on the planet - who knows -
Obviously he didn't review it.
You, after I posted the driver and the 18 technical reasons, said this:
Quoted 1 line
If, in fact, Luben's driver is somehow "better", ...
and
Quoted 1 line
... several things would have to be true. I do not know if they are true here, ...
and
Quoted 1 line
That said, unless the current driver is really bad, and Luben's is tremendously better,
These are from this very thread.
Quoted 8 lines
are simply
beyond the pale. I could just as easily make every one of these
accusations against you -- who pays your salary, for example? What are
their interests? You make claims about your motivations, but how am I to
verify any of those?
I've already stated that there had never been an agenda about uasp.c and Linux. The uasp.c driver was written over most of a Saturday (when I had time and uas.c showed itself to be defunct). Since uasp.c is a working driver and correctly implemented in terms of UAS, USB, SCSI, the kernel and it's layers, I thought it would be *helpful* and *useful* to people--that's all.
The driver was rejected flat out only after 4 minutes after submission claiming that there is a driver in the kernel that "does the exact same thing". I then provided 18 technical reasons as to why uas.c does NOT in fact do the exact same thing to which Greg responded with:
Quoted 1 line
But again, this driver is not acceptable as-is, sorry.
Why would he have to apologize? Anyway, to which you responded with not knowing which driver is good.
Quoted 7 lines
The linux community works because it is exactly that -- a community. We
have, just like any other community, standards of decorum. You have
violated those standards, repeatedly, despite several people politely
telling you the proper manner in which to behave.
When people read "the community" responses in these mailing lists, they think "Wow what a benevolent bunch of people" and "look, they just wanna help". But what this mailing list doesn't show is that this is a very close knit of people who work for the same company (or two) and who go to Linux conferences religiously and are very close buddies. Obviously uasp.c didn't stand a chance _and_ was rejected in just 4 minutes, without a technical review. As you can see in this thread the three of you kept fronting the uas.c driver despite the 18 technical reasons I posted and the fact that it's badly written, and needs to be replaced with something that works, as people will find out when UAS devices make it to market. It's all in the beginning of this thread, start reading here:
http://marc.info/?t=129165521000011&r=2&w=2
So after such a blatant refusal to even give a technical review of the driver, it had me thinking: why would that be?
Quoted 4 lines
Quoted 1 line
Did you review my driver? Did anyone else in this
thread? No, of course
Quoted 1 line
not. You didn't even do it the second time I called
your bluff in this
Quoted 8 lines
Quoted 1 line
thread. Why not? Simple, it's not what you're getting
paid to do now.
*bzzt* Wrong. I did review your driver. I found it interesting, but upon
first submission I found no compelling reason to prefer it over the old.
Your initial submission gave no such reasons.
If you reviewed it, then why did you write that you had no idea if the driver were good or not? Here is what you wrote in this very thread:
Quoted 1 line
If, in fact, Luben's driver is somehow "better", ...
and
Quoted 1 line
... several things would have to be true. I do not know if they are true here, ...
and
Quoted 1 line
That said, unless the current driver is really bad, and Luben's is tremendously better,
Your full message can be found here: http://marc.info/?l=linux-kernel&m=129186163519728&w=2.
But now you say that you had reviewed it!? Now you say you found the driver "interesting" but "no compelling reason to prefer it over the old". How about that it works? How about the 18 technical reasons I posted *before* you posted your response quoted above (which reasons you had read)? You didn't address any of them.
Quoted 1 line
It was only until, after your submission was refused,
It was flat out refused only 4 minutes after I submitted it.
Quoted 4 lines
that
you addressed
the technical merits of your driver over the existing one.
The reason I posted the 18 technical reasons, is because Greg claimed that uas.c "does the exact same thing". His response is here: http://marc.info/?l=linux-usb&m=129165544600441&w=2, he posted this only 4 minutes after I posted the driver. I therefore decided to clarify that uas.c does NOT in fact do the "exact same thing", and backed that up with the 18 technical points.
Quoted 5 lines
I will admit,
you have some valid technical points. But, they are completely overridden
by your inability to work within the standards of behavior of the community.
What does this actually mean? I just posted my uasp.c driver in this very thread and it got rejected 4 minutes afterwards without technical review. When I followed up with a technical review you say "your inability to work within the standards of behavior of the community".
Does providing no technical review and flat out rejecting a driver fall into your "inability to work within the standards of behavior of the community" definition?
Quoted 5 lines
If, at any time, you had decided to stop using ad-hominem attacks and
instead made a calm, composed, and well-reasoned statement about why the
work to fixup the existing driver would be worse than doing
1. The whole thing should be ripped out. Any good work would change 100% of the driver as uasp.c shows (essentially getting a new driver in).
2. I did make that "statement" in the 18 technical reasons of why uasp.c was better than uas.c, other than "it works" of course.
Quoted 2 lines
a drop-in
replacement, we would have listened.
I did. In the 18 technical reasons I listed in this very thread. Here is the link: http://marc.info/?l=linux-kernel&m=129185378712222&w=2
Quoted 5 lines
You came pretty
close to doing this,
but seemed completely unable to resist personal attacks against members of
this community.
And those are? I think you're priming up here what people should think.
Quoted 6 lines
And you never really addressed,
beyond a 1-line statement
stating that to fixup the existing driver you would need to replace it
wholesale, why working to improve the existing driver is a bad choice.
I said that in the 18 technical reasons I listed, link above. The existing driver is badly written, the infrastructure is just off.
Quoted 5 lines
We look to submitters to maintain their work. Yes, there are apparently
some difficulties getting Matt W. to push the ball forward on his UAS
driver.
It's simple, rip out uas.c and put in uasp.c and give people something working and let everyone contribute to something with good foundation.
Quoted 5 lines
However, dealing with you is comparable to
drinking napalm; I
would much rather deal with someone who is
less-than-responsive rather than
someone who is abrasive.
Actually, it's very easy to deal with me. When people come to me, I say "What can I do to help you?" and then I provide them a solution and always go the extra mile, and give them something constructive they can use, and I say, "here it is, and I did this extra thing and let me know what else I can do for you." Most of the time I don't wait for people to come to ask me, but proactively contribute to the end goal, surprising people with "it's already done" when they ask for it.
Luben
Quoted 15 lines · Luben Tuikov
Quoted 4 lines · Matthew Dharm
Your insinuations against the character of many
long-standing Linux
developers, based on their refusal to accept your
submission,No, not submission. The fact is that no one bothered to review the code. It was simply flat out NAK-ed it in just four minutes, fronting one of their co-workers with the defunct driver already in.
It would've been a different story if Greg, Alan and you and anyone else willing, had provided a technical review, and /then/ reject it--the rejection reason would've been clearer to everyone, than taking all this thread to get to.
Quoted 6 lines
Greg didn't review it and rejected it 4 minutes after posting. Alan didn't review it and after I posted the 18 technical reasons (my 2nd post in this thread replying to Greg's) said this:
Quoted 1 line
Your driver may be the best on the planet - who knows -
Obviously he didn't review it.
This is may be true, perhaps he didn't look at the code. This is not because he doesn't have confidence that your driver is well-written or holds a grudge against you. The reason for this NAK is policy: a driver already exists. In the kernel we work with what already exists, even if it may be more difficult than the alternative.
I looked at your code; from my uninformed perspective it looks very nice. You clearly do have a good understanding of the stack and your code looks refreshingly clean. That being said, a driver covering the same devices already exists in the tree. There may be good technical reasons why this driver is inferior, but these are merely reasons to fix the existing code, not rip it out to be replaced wholesale.
At this point you have two options: You can continue to argue that the world is against you, which as we seen makes neither side happy and only makes your life as a contributor harder, or you can try to work what you have into something that will fit with the existing codebase. I suggest the latter. Further arguing is not going to change this policy. That being said, your expertise is certainly needed and it is in everyone's best interest to ensure that the UAS driver in the kernel is the best that it can be.
Quoted 7 lines
I've already stated that there had never been an agenda about
uasp.cand Linux. Theuasp.cdriver was written over most of a Saturday (when I had time anduas.cshowed itself to be defunct). Sinceuasp.cis a working driver and correctly implemented in terms of UAS, USB, SCSI, the kernel and it's layers, I thought it would be *helpful* and *useful* to people--that's all.
Improving code is very useful. History has taught us that flat-out rewriting components just makes things messier.
Quoted 10 lines
The driver was rejected flat out only after 4 minutes after submission claiming that there is a driver in the kernel that "does the exact same thing". I then provided 18 technical reasons as to why
uas.cdoes NOT in fact do the exact same thing to which Greg responded with:Quoted 1 line
But again, this driver is not acceptable as-is, sorry.
Why would he have to apologize? Anyway, to which you responded with not knowing which driver is good.
He apologized because he knows you put a lot of time, effort, and thought into a piece of code which we cannot accept. It is unfortunate that this is the case but such is the reality of large-scale software development.
Quoted 3 lines
So after such a blatant refusal to even give a technical review of the driver, it had me thinking: why would that be?
Easy question. See above. In short, there is no sense to give a technical review of code which we ultimately won't be able use. Sometimes this sort of thing happens. If I were you, I would accept what people have been telling you and try to integrate portions of your patch into the existing UAS driver. This will cause the least pain/most benefit to all involved.
- Ben
Quoted 5 lines · Ben Gamari
Quoted 1 line
Quoted 1 line
Your driver may be the best on the planet - who
knows -
Quoted 1 line
Obviously he didn't review it.
This is may be true, perhaps he didn't look at the code.
"may be true". It is clear he didn't. What he wrote is quoted above.
Quoted 4 lines
This is not
because he doesn't have confidence that your driver is well-written or
holds a grudge against you.
Have you been authorized to speak for him?
Quoted 5 lines
The reason for this NAK is
policy: a driver
already exists. In the kernel we work with what already exists, even if
it may be more difficult than the alternative.
Nothing can be further from the truth. If you have been long enough around Linux, you'd know that many times there would be two perfectly working drivers for the same device with hundreds of thousands of those devices out in the wild, in the kernel. You'd also know that sometimes a driver is submitted and accepted and in the kernel, but it is so badly written that it gets re-written by one of the resident "linux kernel engineers" and then the latter driver replaces the one already existing, etc, etc. Add to this that it is sometimes good to offer people a choice between a defunct and unmaintained driver and a working and maintained driver.
Now is the time to do the right thing and get a working and technically correct UAS driver, when there are no UAS devices out there.
Quoted 4 lines
I looked at your code; from my uninformed perspective it looks very
nice. You clearly do have a good understanding of the stack and your
Quoted 8 lines
code looks refreshingly clean. That being said, a driver covering the
same devices already exists in the tree. There may be good technical
reasons why this driver is inferior, but these are merely reasons to fix
the existing code, not rip it out to be replaced wholesale.
It's much easier, faster and foolproof to rip out a defunct driver than to submit 30+ patches that change 100% of the driver. Moreover when there are no devices for it yet.
Quoted 3 lines
At this point you have two options: You can continue to argue that the
world is against you,
I don't argue that nor do I think that.
Quoted 5 lines
which as we seen makes neither side
happy and only
makes your life as a contributor harder, or you can try to work what you
have into something that will fit with the existing
Quoted 8 lines
codebase. I suggest
the latter. Further arguing is not going to change this policy. That
being said, your expertise is certainly needed and it is in everyone's
best interest to ensure that the UAS driver in the kernel is the best
that it can be.
The situation is as follows: the Linux band is more than happy to review patches to /their/ code, but when you submit a driver that replaces a defunct driver of theirs in the kernel, then all of a sudden no one can read code to give a technical review of your driver. So they say "we're happy to accept patches to /our/ code even if it replaces 100% of it, but your driver which does this in one go is rejected, sorry" and "I've no idea if it is good or bad".
Quoted 1 line
Improving code is very useful.
Improving code can be useful, but not to the extent of replacing it 100% in one go in 30+ patches.
A patch is supposed to fix a single bug or a feature missed, etc. For example see my recent patches posted as well as examine git history, as well as examine my patches from before there was git (starting with the SCSI layer using the slab cache for struct scsi_cmnd allocation).
Quoted 3 lines
History has taught us that
flat-out
rewriting components just makes things messier.
Nothing can be further from the truth. This isn't a whimsical rewrite. I tried to "fix" uas.c but it's just horrible and needs to be ripped out. Just one patch of mine changed 86% of the code and that's NOT right. But with that defunct driver, there was just no other way. The terminology the driver uses show the writer is not familiar with the technology. The logistical flow of the code shows the writer isn't familiar with the way a SCSI Delivery Subsystem works, UAS and USB in general. I've outlined that in my 18 technical point review earlier in this thread.
The right thing was to write a UAS driver that is correct and that it works, thus uasp.c.
Quoted 2 lines
He apologized because he knows you put a lot of time, effort, and
Again, you speak for other people. This isn't a good thing.
Quoted 1 line
thought into a piece of code which we cannot accept.
"we"? See when you write emails here, and pretend to be this benevolent selfless creature lacking all self-interest, you have to be more subtle and present an unbiased opinion. Using "we" isn't the way to do it.
Quoted 5 lines
It is
unfortunate
that this is the case but such is the reality of
large-scale software
development.
uasp.c is 1141 lines, a featherweight in drivers class. There is no "large-scale software development". uasp.c is self-contained driver. It doesn't change or modify any of the rest of the kernel. So now you can see how there is NO "large-scale software development". It doesn't affect the system other than when uasp.c is used people will notice that they can now use a UAS device.
Quoted 4 lines
Easy question. See above. In short, there is no sense to give a
technical review of code which we ultimately won't be able use.
You said it: "ultimately won't be able to use". Basically a decision has already been made. And here you have shown the problem: the biased opinion of the Linux band: you're welcome to improve /our/ code even it you completely replace it in one go in 30+ patches, but we're not accepting your driver, under any circumstances. How about give people a choice?
Quoted 4 lines
Sometimes this sort of thing happens. If I were you, I would accept what
people have been telling you and try to integrate portions of your
People? You mean a two people who have known each other for over 10 years are close knit friends who work for the same company whose company sponsors their conferences and gives them equipment.
Quoted 3 lines
patch into the existing UAS driver. This will cause the least pain/most
benefit to all involved.
The least pain/most benefit is to replace the defunct uas.c with uasp.c, so people can test the UAS devices they are developing and when those devices are released on the market, people can use them. See here: http://marc.info/?l=linux-scsi&m=129227488508299&w=2
Luben
Hi,
I had preferred to subsequently calmly stay on the sidelines after trying to argue for a beneficial compromise, but given this lack of realism I'm choosing to reply again:
Quoted 11 lines
Quoted 3 lines · Luben Tuikov
--- On Mon, 12/13/10, Matthew Dharm <[email protected]> wrote:> > Your driver may be the best on the planet - who knows - Obviously he didn't review it.
This is may be true, perhaps he didn't look at the code. This is not because he doesn't have confidence that your driver is well-written or holds a grudge against you. The reason for this NAK is policy: a driver already exists. In the kernel we work with what already exists, even if it may be more difficult than the alternative.
One word: HOG-WASH. At least the e100 vs. eepro100 case is a clear counter-example to that. That was an eepro100 driver which has been in active mainline service for years and had many diagnostic features,
to then be replaced (one could argue that it _may_ be better indeed to replace a possibly less-maintained driver with a better-maintained one by the original hardware manufacturer).
The result, even from my tiny involvement (one card where I had lots
of trouble getting a make-it-work-again kind of patch into e100 - one
YEAR to have mainline actually support the hardware again),
is known as several hardware variants failing to work with the new driver (and I just found another hardware example in the very first semi-related Google search I did, so there must have been many more).
Not sure about current status, but I'm nevertheless very positive that it's much better now.
Witness this very problematic case as opposed to the current conflict case where there's a _new_ hardware/driver which has been hardly even used by anyone, to potentially be replaced by a (supposedly?) much better driver which has almost identical amount of history (read: _weeks_ or at most months, and both have not even ever been released yet).
Guys, please argue healthily and don't keep making up straw-man arguments (as I'm tending to believe when reading some parts of the discussion).
The way I see it there is this (likely incomplete) list of reasons to
prefer the "old", "challenged"(?) new driver:
. there's the "old", "timely submitted"
driver, and the entire thing may be an understandably much-less-than-positive situation for its author
. the author of the new one is a d**k to argue with
(may easily be true given several cases of his awfully "personal" arguing,
but in his position facing such often unbased opposition
other people might have lost some temper, too),
causing uncertainty about proper continued maintenance of the driver
. for some certain reason, it's had "more testing", its situation is "better", ...
. suddenly accepting an entirely different driver after another can
legitimately be seen as some kind of unhealthy disruption of the
development advancement
. at this point in time we've lost enough additional time arguing
that a change of drivers might now really be inconvenient
from a kernel release continuity/planning POV.
Let me tell you that I'm somewhat unconvinced of this, though.
Perhaps at this time the only thing to be said is that due to "policy" (yeah, whatever) a more timely submitted driver will remain in and the supposedly better driver will stay out, but given the history of Linux kernel drivers this is more than a bit unconvincing, as outlined above.
Of course one could argue that this policy simply is quite _new_
and didn't exist yet at the time of the other conflicts (e100, 8139, RAID adapters), and possibly has been established exactly _due_ to the many difficulties that turned up in these cases.
However it's still questionable whether it's appropriate
to apply such a policy _in this case_
given that _both_ candidates aren't entrenched (in active service) at all.
Still, I hope that even with a "negative" decision the best elements of the drivers will eventually find their way into a combined driver, with appropriate amounts of maintenance.
Andreas Mohr
Quoted 13 lines · Andreas Mohr
Hi,
I had preferred to subsequently calmly stay on the sidelines
after trying to argue for a beneficial compromise, but given this lack of realism I'm choosing to reply again:Ben Gamari wrote this:
Quoted 1 line
This is may be true, perhaps he didn't look at the
code. This is not
Quoted 1 line
because he doesn't have confidence that your driver is
well-written or
Quoted 8 lines
Quoted 1 line
holds a grudge against you. The reason for this NAK is
policy: a driver
Quoted 1 line
already exists. In the kernel we work with what
already exists, even if
Quoted 1 line
it may be more difficult than the alternative.
One word: HOG-WASH. At least the e100 vs. eepro100 case is a clear
Andreas, the way you've cut out the quote above, it seems that you're replying to something I've written, when in fact you're replying to Ben Gamari.
Above, I've corrected the quote, not to confuse people of course.
Quoted 9 lines
counter-example to that. That was an eepro100 driver which has been in active
mainline service for years and had many diagnostic features,
to then be replaced (one could argue that it _may_ be better indeed
to replace a possibly less-maintained driver with a better-maintained one
by the original hardware manufacturer).
So eepro100.c was written by Don (as was rtl8139.c) and was replaced by e100.c written by its manufacturer (same place where uas.c is coming from)? Interesting...
As I recall rtl8139.c was also replaced by 8139too.c after many years with many devices in the field, instead of "dividing the code into patches for the driver which is already in the kernel and submitting it for review". The author of 8139too.c used to submit patches to rtl8139.c (as can be seen in the source of that driver) but then decided to write his own 8139too.c and submit it and it got accepted and now that's the only driver in the kernel for that chip. The copyright comment of 8139too.c says this.
Hmm... Interesting and insightful stuff. So who are those people that can rewrite whole drivers and get theirs in and the old ones out?
Quoted 8 lines
The way I see it there is this (likely incomplete) list of reasons to
prefer the "old", "challenged"(?) new driver:
. there's the "old", "timely submitted"
driver, and the entire thing may be an understandably much-less-than-positive situation for its author
. the author of the new one is a d**k to argue with
Quoted 8 lines
(may easily be true given several cases of his
awfully "personal" arguing,
but in his position facing such often
unbased opposition
other people might have lost some temper,
too),
causing uncertainty about proper continued
maintenance of the driver
Within the last two months, I've submitted more patches to uasp.c (4-5) than the original author to uas.c (none). As well as I've submitted patches to uas.c itself (three, first two accepted, third rejected since it changed 86% of the uas.c code), lsusb, sd.c, block/blk-tag.c, etc.
Also, uasp.c is a protocol driver. The protocol will change a bit but not by that much. When this happens I'll submit new patches to update it, unless someone submits them before that. Even if no patches are submitted to support latter version of the protocol, UAS devices will be backward compatible.
In effect, this driver doesn't warrant much maintenance by virtue of its nature, but support is there if need be.
Quoted 2 lines
. for some certain reason, it's had "more testing", its situation is "better", ...
"more testing" -- do you mean uasp.c or uas.c? If you have a UAS device, try both drivers (do I/O) and see for yourself. uas.c also doesn't support the UAS protocol by definition as it doesn't support TMF and other features present in uasp.c.
Quoted 33 lines
. suddenly accepting an entirely different driver after another can
legitimately be seen as some kind of unhealthy disruption of the
development advancement
. at this point in time we've lost enough additional time arguing
that a change of drivers might now really be inconvenient
from a kernel release continuity/planning POV. Let me tell you that I'm somewhat unconvinced of this, though.Perhaps at this time the only thing to be said is that due to "policy" (yeah, whatever)
a more timely submitted driver will remain in and the supposedly better
driver will stay out, but given the history of Linux kernel drivers this
is more than a bit unconvincing, as outlined above. Of course one could argue that this policy simply is quite _new_
and didn't exist yet at the time of the other conflicts (e100, 8139, RAID adapters),
and possibly has been established exactly _due_ to the many difficulties that turned
up in these cases.
However it's still questionable whether it's appropriate to apply such a policy_inthiscase_
given that _both_ candidates aren't entrenched (in active service) at all.
Quoted 5 lines
Still, I hope that even with a "negative" decision the best elements of the
drivers will eventually find their way into a combined driver,
with appropriate amounts of maintenance.
You make some very good points, however it's a Goliath vs. David situation but this time David loses. While this may seem similar to rtl8139.c vs. 8139too.c and eepro100.c vs. e100.c, the difference is that the backing behind 8139too.c and e100.c is with with uas.c. Working code or the quality of the code is irrelevant here. uasp.c got rejected 4 minutes after submission.
uasp.c wouldn't even make it in the kernel alongside uas.c, to give people a choice at least. (and we've heard the excuse for that one before, while there had been two drivers for the same device in the kernel before)
In the immediate future distros may include uasp.c since it provides UAS support. Independent Linux vendors may do the same for the same reason. HDD/ASIC manufacturers may use uasp.c to test their UAS devices (an alternative to the Windows tools).
In the future though, I expect that someone will study uasp.c and submit patches to uas.c and eventually you'll see uas.c look exactly as uasp.c does now (sans the credits, similarly to what happened to the SAS code five years ago).
However, people are free to use whatever they want.
If you have a UAS device and would like to play with it, uasp.c is available on github: http://marc.info/?t=129227496000002&r=1&w=2
Luben
Quoted 1 line · Luben Tuikov
Andreas, the way you've cut out the quote above, it seems that you're replying to something I've written, when in fact you're replying to Ben Gamari.
Indeed, I had already realized that; that was a victim of me _manually_ replying to external archive content in the case of LKML (while preserving proper In-Reply-To ID).
Quoted 1 line
Hmm... Interesting and insightful stuff. So who are those people that can rewrite whole drivers and get theirs in and the old ones out?
<rhetorical question> :-P
I....
would like to say that I'm not really certain about this ;)
Quoted 1 line
Within the last two months, I've submitted more patches to
uasp.c(4-5) than the original author touas.c(none). As well as I've submitted patches touas.citself (three, first two accepted, third rejected since it changed 86% of theuas.ccode), lsusb,sd.c,block/blk-tag.c, etc.
"two months". So it's probably a longer time than what I thought ("weeks"), which probably is a large part of the reason why it got rejected outright without its fair share of chance of review,
within the current kernel release interval.
Still, it's then not entirely PC if some people - admittedly people other than the ones whose actual business it is/was to reject drivers :)) - then claim that the reason the driver got rejected is firm "policy" to always reject conflicting drivers.
Quoted 3 lines
Also,
uasp.cis a protocol driver. The protocol will change a bit but not by that much. When this happens I'll submit new patches to update it, unless someone submits them before that. Even if no patches are submitted to support latter version of the protocol, UAS devices will be backward compatible.In effect, this driver doesn't warrant much maintenance by virtue of its nature, but support is there if need be.
*noted*
Quoted 4 lines
Quoted 2 lines
. for some certain reason, it's had "more testing", its situation is "better", ...
"more testing" -- do you mean
uasp.coruas.c? If you have a UAS device, try both drivers (do I/O) and see for yourself.uas.calso doesn't support the UAS protocol by definition as it doesn't support TMF and other features present inuasp.c.
Since I indicated this compiled list to be about the "original" submission, it's hopefully obvious.
Quoted 1 line
uasp.cwouldn't even make it in the kernel alongsideuas.c, to give people a choice at least. (and we've heard the excuse for that one before, while there had been two drivers for the same device in the kernel before)
The situation of two drivers in direct conflict can get messy indeed (especially if people don't know which they are using), thus I can easily understand that.
Quoted 1 line
In the immediate future distros may include
uasp.csince it provides UAS support. Independent Linux vendors may do the same for the same reason. HDD/ASIC manufacturers may useuasp.cto test their UAS devices (an alternative to the Windows tools).
And that's one thing that I'm worrying about.
Let's say one does the "proper", "good" way to achieve driver improvements:
have the earlier submission get into released kernels and then submit patches to it in (let's say) third-steps to have it get to what a driver with combined features/functionality/stability would provide.
This would mean to have an original driver implementation in kernels, which would not last long, get updated significantly ("third-steps" above), then re-released until we ultimately hit the full combined and full-featured driver in further kernel releases.
Depending on how important API / user space stability/architecture effects may become, we could IMNSUTVHO see a situation where it can get moderately problematic on which distribution / version is using which driver development version.
All this as opposed to doing a (admittedly not really painless) emergency-yank now and cleanly releasing with UASP as the _first_ driver version (which probably is much less likely to change drastically) to ever hit a kernel release.
Quoted 1 line
If you have a UAS device and would like to play with it,
uasp.cis available on github: http://marc.info/?t=129227496000002&r=1&w=2
Given current mass-market unavailability of hardware (as indicated in the discussion): sorry, nope.
Andreas Mohr