Skip to contentSkip to conversation

serial 8250: tighten test for using backup timer

PATCH David Brownell 3 participants 5 messages Aug 26, 2008 – Sep 1, 2008

lore
Share & tools
lore
#0 lore raw
Quoted 9 lines
Quoted 5 lines

Also, how serious is the problem which is being fixed here? It _sounds_ like it's of the "fatal for people who have that hardware" variety, in which case we should get this into 2.6.27 and probably 2.6.26.x. Not sure about 2.5.26.x though - the patch doesn't apply there, but I didn't check whether this is due to functional changes.

For users of this version of this particular UART IP it is fatal. From looking at the git history it looks like the original patch went into 2.6.26 so it might also affect that kernel.

Second that: serial-8250-tighten-test-for-using-backup-timer.patch (from MMOTM) in mainline sooner-not-later seems right.

My own exposure to this is that the UART on DaVinci hardware, which TI allegedly derived from its original 16550 logic, has periodically gone from working to unusable with the mainline 8250.c ... and back and forth a bunch. Currently it's "unusable", a regression from some previous versions. With this patch from Alex, it's usable.

Of course there are a bunch of arch/arm/mach-davinci patches needed to make that platform more functional in mainline. I expect they'll probably merge in the 2.6.28-rc0 window. Meanwhile, running out of ramdisk with a serial console should at least work right! :)

- Dave

#1 lore raw
Quoted 6 lines · On Tue, 26 Aug 2008 10:45:34 -0700 David Brownell <[email protected]> wrote:
Quoted 6 lines
Quoted 5 lines

Also, how serious is the problem which is being fixed here? It _sounds_ like it's of the "fatal for people who have that hardware" variety, in which case we should get this into 2.6.27 and probably 2.6.26.x. Not sure about 2.5.26.x though - the patch doesn't apply there, but I didn't check whether this is due to functional changes.

Quoted 6 lines
Quoted 3 lines

For users of this version of this particular UART IP it is fatal. From looking at the git history it looks like the original patch went into 2.6.26 so it might also affect that kernel.

Second that: serial-8250-tighten-test-for-using-backup-timer.patch (from MMOTM) in mainline sooner-not-later seems right.

ok, if you think so...

I still have an unanswered question out there about the init_timer() handling in that area of the driver but afaict this patch didn't make any of it worse than it already is.

Quoted 11 lines

My own exposure to this is that the UART on DaVinci hardware, which TI allegedly derived from its original 16550 logic, has periodically gone from working to unusable with the mainline 8250.c ... and back and forth a bunch. Currently it's "unusable", a regression from some previous versions. With this patch from Alex, it's usable.

Of course there are a bunch of arch/arm/mach-davinci patches needed to make that platform more functional in mainline. I expect they'll probably merge in the 2.6.28-rc0 window. Meanwhile, running out of ramdisk with a serial console should at least work right! :)

#2 lore raw
Quoted 17 lines · Andrew Morton
Quoted 6 lines · On Tue, 26 Aug 2008 10:45:34 -0700 David Brownell <[email protected]> wrote:
Quoted 6 lines
Quoted 5 lines

Also, how serious is the problem which is being fixed here? It _sounds_ like it's of the "fatal for people who have that hardware" variety, in which case we should get this into 2.6.27 and probably 2.6.26.x. Not sure about 2.5.26.x though - the patch doesn't apply there, but I didn't check whether this is due to functional changes.

Quoted 6 lines
Quoted 3 lines

For users of this version of this particular UART IP it is fatal. From looking at the git history it looks like the original patch went into 2.6.26 so it might also affect that kernel.

Second that: serial-8250-tighten-test-for-using-backup-timer.patch (from MMOTM) in mainline sooner-not-later seems right.

Quoted 5 lines

ok, if you think so...

I still have an unanswered question out there about the init_timer() handling in that area of the driver but afaict this patch didn't make any of it worse than it already is.

Sorry, I forgot to follow up on that. I had a look at the init_timer() usage and it looks correct although it is rather strange. Reworking it would affect several classes of hardware, most of which I don't have available so I think it's safest to leave it as is for now.

Quoted 13 lines
Quoted 11 lines

My own exposure to this is that the UART on DaVinci hardware, which TI allegedly derived from its original 16550 logic, has periodically gone from working to unusable with the mainline 8250.c ... and back and forth a bunch. Currently it's "unusable", a regression from some previous versions. With this patch from Alex, it's usable.

Of course there are a bunch of arch/arm/mach-davinci patches needed to make that platform more functional in mainline. I expect they'll probably merge in the 2.6.28-rc0 window. Meanwhile, running out of ramdisk with a serial console should at least work right! :)

#3 lore raw
In reply to Andrew Morton
Quoted 13 lines · Andrew Morton
Quoted 10 lines · On Tue, 26 Aug 2008 10:45:34 -0700 David Brownell <[email protected]> wrote:
Quoted 9 lines
Quoted 5 lines

Also, how serious is the problem which is being fixed here? It _sounds_ like it's of the "fatal for people who have that hardware" variety, in which case we should get this into 2.6.27 and probably 2.6.26.x. Not sure about 2.5.26.x though - the patch doesn't apply there, but I didn't check whether this is due to functional changes.

For users of this version of this particular UART IP it is fatal. From looking at the git history it looks like the original patch went into 2.6.26 so it might also affect that kernel.

Quoted 8 lines
Quoted 2 lines

Second that: serial-8250-tighten-test-for-using-backup-timer.patch (from MMOTM) in mainline sooner-not-later seems right.

ok, if you think so...

I still have an unanswered question out there about the init_timer() handling in that area of the driver but afaict this patch didn't make any of it worse than it already is.

Any update on this situation? I'd really like this patch to get into 2.6.27.

Let me know if there's any further information you need.

Thanks,

Quoted 13 lines
Quoted 11 lines

My own exposure to this is that the UART on DaVinci hardware, which TI allegedly derived from its original 16550 logic, has periodically gone from working to unusable with the mainline 8250.c ... and back and forth a bunch. Currently it's "unusable", a regression from some previous versions. With this patch from Alex, it's usable.

Of course there are a bunch of arch/arm/mach-davinci patches needed to make that platform more functional in mainline. I expect they'll probably merge in the 2.6.28-rc0 window. Meanwhile, running out of ramdisk with a serial console should at least work right! :)

#4 lore raw
Quoted 18 lines · On Mon, 1 Sep 2008 14:36:56 +0100 "Will Newton" <[email protected]> wrote:
Quoted 16 lines · Andrew Morton
Quoted 12 lines · On Tue, 26 Aug 2008 10:45:34 -0700 David Brownell <[email protected]> wrote:
Quoted 9 lines
Quoted 5 lines

Also, how serious is the problem which is being fixed here? It _sounds_ like it's of the "fatal for people who have that hardware" variety, in which case we should get this into 2.6.27 and probably 2.6.26.x. Not sure about 2.5.26.x though - the patch doesn't apply there, but I didn't check whether this is due to functional changes.

For users of this version of this particular UART IP it is fatal. From looking at the git history it looks like the original patch went into 2.6.26 so it might also affect that kernel.

Second that: serial-8250-tighten-test-for-using-backup-timer.patch (from MMOTM) in mainline sooner-not-later seems right.

Quoted 7 lines
Quoted 5 lines

ok, if you think so...

I still have an unanswered question out there about the init_timer() handling in that area of the driver but afaict this patch didn't make any of it worse than it already is.

Any update on this situation? I'd really like this patch to get into 2.6.27.

yup, I plan to include the patch in my next batch for 2.6.27.