Skip to contentSkip to conversation

removal of 1980s PC-XT MFM 8bit ISA disk driver

GIT PULL Paul Gortmaker 4 participants 5 messages Feb 13, 2013 – Feb 14, 2013

lore
Share & tools
lore
#0 lore raw

Jens,

Please pull the following to get the removal of the original IBM PC-XT hard disk driver from the block layer (drivers/block/xd.c).

As near as I can tell, it hasn't seen a run time fix in over a dozen years, and with drive sizes of 10-20MB, and performance of about 128kB/s maximum, it is no surprise that it has been completely unused for well over a decade.

The removal was originally posted[1] well over a month ago, and since then, there has been nobody objecting to the removal, aside from someone who had mistakenly confused it with a completely different driver (hd.c)

I've done a test merge followed by an allmodconfig/allyesconfig for x86-64 on the "for-next" branch of:

git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-block.git

to ensure that there was no Makefile/Kconfig fallout since the 3.8-rc2 the original commit was created and tested on.

[1] https://patchwork.kernel.org/patch/1935381/

Thanks,
Paul.

---

The following changes since commit d1c3ed669a2d452cacfb48c2d171a1f364dae2ed:

Linux 3.8-rc2 (2013-01-02 18:13:21 -0800)

are available in the git repository at:

git://git.kernel.org/pub/scm/linux/kernel/git/paulg/linux.git delete-xt-disk

for you to fetch changes up to d1a6f4f19728d6e90480e53601a90fc9f6a348ad:

block: delete super ancient PC-XT driver for 1980's hardware (2013-01-04 20:17:40 -0500)

----------------------------------------------------------------
Paul Gortmaker (1):
block: delete super ancient PC-XT driver for 1980's hardware

 drivers/block/Kconfig  |   13 -
 drivers/block/Makefile |    1 -
 drivers/block/xd.c     | 1123 ------------------------------------------------
 drivers/block/xd.h     |  134 ------
 4 files changed, 1271 deletions(-)
 delete mode 100644 drivers/block/xd.c
 delete mode 100644 drivers/block/xd.h
#1 lore raw
Quoted 9 lines · On Wednesday 13 February 2013 21:04:32 Paul Gortmaker wrote:

Jens,

Please pull the following to get the removal of the original IBM PC-XT hard disk driver from the block layer (drivers/block/xd.c).

As near as I can tell, it hasn't seen a run time fix in over a dozen years, and with drive sizes of 10-20MB, and performance of about 128kB/s maximum, it is no surprise that it has been completely unused for well over a decade.

It was fixed just yesterday by Linus Walleij:
http://lkml.org/lkml/2013/2/12/541

Signature · 1 line
Ondrej Zary
#2 lore raw
Quoted 6 lines · [Re: [GIT PULL] removal of 1980s PC-XT MFM 8bit ISA disk driver] On 13/02/2013…
Quoted 5 lines · On Wednesday 13 February 2013 21:04:32 Paul Gortmaker wrote:

Jens,

Please pull the following to get the removal of the original IBM PC-XT hard disk driver from the block layer (drivers/block/xd.c).

Quoted 7 lines
Quoted 4 lines

As near as I can tell, it hasn't seen a run time fix in over a dozen years, and with drive sizes of 10-20MB, and performance of about 128kB/s maximum, it is no surprise that it has been completely unused for well over a decade.

It was fixed just yesterday by Linus Walleij:
http://lkml.org/lkml/2013/2/12/541

Ondrej,

Since you consistently try to interject into these cleanups and never address the logical points of reason given, let me repeat them once more, in the hope that you will read and consider their merit.

But before doing that though, lets look at this rather contrived "fix" that coincidentally happened to appear after the deletion was suggested.

1) it claims the following: "...the fact that it was in the kernel tree SAVED MY DATA." [caps exactly as in original]. This is patently false, as any kernel from 2.0, 2.2, 2.4, 2.6, 3.x would have done the job. In fact, using an older kernel would have better odds of success, based on simple bitrot rules and the bathtub curve. The above hype tries to FUD people into thinking drivers permanently "disappear" just because they get removed from master on mainline - totally untrue.

2) it "works" but only after you are required to physically pry parts off of the hardware, and yet this is considered reasonable?

3) we are asked to believe that it is completely reasonable to require this to be in the bleeding edge tree, yet at the same time, the platform needed to achieve is is chosen as a Pentium-MMX -- i.e. a system from 1996 running at ~200MHz -- as far from bleeding edge as physically possible.

4) the "fix" adds a udelay after each outb (vs simply redefining outb, I might add) -- which only proves that this has been unused for a long time -- i.e. since the whole SLOW_DOWN_IO/outb/0x80 cleanup.

But in the end, those points are not as important as what is below. These general justification points you can find in the pending 3.9 commit 930d52c012b8e69ea87efd7562ded977ee9c9df9 (via the net-next tree). I hope you can consider these points with an open mind.

Thanks,
Paul.

-------
Allow me a preemptive justification against the inevitable comments from well meaning bystanders who suggest "why not just leave all this alone?". Dead drivers cost us all if they are left in tree. If you think that is false, then please first consider:

-every time you type "git status", you are checking to see if modifications
have been made by you to all that dead code.

-every time you type "git grep <regex>" you are searching through files
which contain that dead code that simply does not interest you.

-every time you build a "allyesconfig" and an "allmodconfig" (don't tell
me you skip this step before submitting your changes to a maintainer), you waste CPU cycles building this dead code.

-every time there is a tree wide API change, or cleanup, or file relocation,
we pay the cost of updating dead code, or moving dead code.

-daily regression tests (take linux-next as the most transparent
example) spend time building (and possibly running) this dead code.

-hard working people who regularly run auditing tools looking for lurking
bugs (sparse/coverity/smatch/coccinelle) are wasting time checking for, and fixing bugs in this dead code.

This last one is key. Please take a look at the git history for the files that are proposed for removal here. Look at the git history for any one of them ("git whatchanged --follow drivers/net/.../driver.c") Mentally sort the changes into two bins -- (1) the robotic tree-wide changes, and (2) the "look I found a real run-time bug while using this" category. You will see that category #2 is essentially empty.

Further to that, realize that drivers don't simply disappear. We are not operating in the binary-only distribution space like other OS. All these drivers remain in the git history forever. If a person is an enthusiast for extreme legacy hardware, they are probably already customizing their kernel source and building it themselves to support such systems. Also keep in mind that they could still build the 3.8 kernel exactly as-is, and run it (or a 3.8.x stable variant of it) for several more years if they were really determined to cling to these old experimental ISA drivers for some reason.

In summary, I hope that folks can be pragmatic about this, and not get swept up in nostalgia. Ask yourself whether it is realistic to expect a person would have a genuine use case where they would need to build a 3.9+ modern kernel and install it on some legacy hardware that has no option but to absolutely _require_ one of the drivers that are deleted here.
-------

Quoted 3 lines
Signature · 2 lines
-- 
Ondrej Zary
#3 lore raw
Quoted 2 lines · Paul Gortmaker

But before doing that though, lets look at this rather contrived "fix" that coincidentally happened to appear after the deletion was suggested.

Hm. Haha, I had no idea that this driver was simultaneously being considered for removal.

Quoted 2 lines

1) it claims the following: "...the fact that it was in the kernel tree SAVED MY DATA." [caps exactly as in original]. This is patently false,

I don't know what kind of debate you guys are having or what kind of background there is in this discussion, but it's not like you're using a welcoming tone to new contributors here.

Are you guys in the middle of some flamefest of the same type as the inclusionist vs deletionist power-struggle they have over at the Wikipedia every now and then?
http://en.wikipedia.org/wiki/Deletionism_and_inclusionism_in_Wikipedia

All I was doing was hacking some old hardware, and I was having fun, while retrieveing some old pascal programs.

Compare the title of a book put out by some kernel developer a while back.

Part of this mail seems to be something like review comments, I recommend you send these as replies to the individual patches?

As for whether to delete or maintain the driver I do not really care, what I know is that having it around was helpful and fun for this one hacker here.

Yours,
Linus Walleij

#4 lore raw
In reply to Paul Gortmaker
Quoted 17 lines · Paul Gortmaker

Jens,

Please pull the following to get the removal of the original IBM PC-XT hard disk driver from the block layer (drivers/block/xd.c).

As near as I can tell, it hasn't seen a run time fix in over a dozen years, and with drive sizes of 10-20MB, and performance of about 128kB/s maximum, it is no surprise that it has been completely unused for well over a decade.

The removal was originally posted[1] well over a month ago, and since then, there has been nobody objecting to the removal, aside from someone who had mistakenly confused it with a completely different driver (hd.c)

I've done a test merge followed by an allmodconfig/allyesconfig for x86-64 on the "for-next" branch of:

Quoted 6 lines

git://git.kernel.org/pub/scm/linux/kernel/git/axboe/linux-block.git

to ensure that there was no Makefile/Kconfig fallout since the 3.8-rc2 the original commit was created and tested on.

[1] https://patchwork.kernel.org/patch/1935381/

Pulled, I'm pretty sure this hasn't been used in a decade at least. In any case, it sits there nicely in history if someone wants to resurrect it.

Signature · 1 line
Jens Axboe