In D24428#666449, @rainer_ultra-secure.de wrote:

Hi,

I have this slated for production in one week.
I will have to wait a couple of days to see if it fixes the hangs I have.

In D24428#664005, @papani.srikanth_microchip.com wrote:

In D24428#663851, @imp wrote:

In D24428#663568, @papani.srikanth_microchip.com wrote:

Can someone approve/review the code ?

I'll look at it this afternoon.

Do you also need someone to commit it to the FreeBSD tree too?

I would be happy if someone commit it to the repository.

In D24428#663851, @imp wrote:

In D24428#663568, @papani.srikanth_microchip.com wrote:

Can someone approve/review the code ?

I'll look at it this afternoon.

Do you also need someone to commit it to the FreeBSD tree too?

Can someone approve/review the code ?

Can someone approve/review the code ?

Updated copyright section

In D24428#662938, @yuripv wrote:

In D24428#661566, @papani.srikanth_microchip.com wrote:

In D24428#661402, @yuripv wrote:

In D24428#658934, @papani.srikanth_microchip.com wrote:

In D24428#657918, @imp wrote:

There's issues with applying this patch. Something seems to have gone amiss in its generation.

It almost applied cleanly to stable/12 branch, but not to the main branch:

% find . -name \*.rej
./sys/dev/smartpqi/smartpqi_mem.c.rej
./sys/dev/smartpqi/smartpqi_queue.c.rej
./sys/dev/smartpqi/smartpqi_defines.h.rej
./sys/dev/smartpqi/smartpqi_cam.c.rej
./sys/dev/smartpqi/smartpqi_misc.c.rej
./sys/dev/smartpqi/smartpqi_main.c.rej
./sys/dev/smartpqi/smartpqi_request.c.rej

In the main branch, it crashed patch :(.
A quick sample of the .rej files shows the diffs likely are easy to resolve by hand, but with such a large patch I'm leery to do so. Add '-l' to patch to cope with whitespace changes didn't seem to help.
So it looks like this patch needs to be regenerated and/or moved to git where patch generation and uploading is a bit more reliable.

(Also commented on a couple of nits that didn't look quite right in the copyright stuff, but that can wait for the patch to get done).

Hi,

• 12.0 stable branch and 12.2 main branch has two different source codes. I've pulled the 12.0 source code and applied the patch. (12.0 stable branch has bug fixes which is done by community but I do not see the same changes in 12.2 main branch because of that the patch is failing on 12.2 main branch).

"main" mentioned is literally main git branch, where this change should go first (and it's 14.0-CURRENT at the moment). I have HBA 1100-8i that is misbehaving under load, so I'd really like to try this patch -- could you please rebase this against main?

This patch is for 12.2 only, I will push a new patch separately to the main branch.

Is there a review/patch for main branch yet to test?

Here is the review patch for main branch.
https://reviews.freebsd.org/D29584

In D24428#661402, @yuripv wrote:

In D24428#658934, @papani.srikanth_microchip.com wrote:

In D24428#657918, @imp wrote:

There's issues with applying this patch. Something seems to have gone amiss in its generation.

It almost applied cleanly to stable/12 branch, but not to the main branch:

% find . -name \*.rej
./sys/dev/smartpqi/smartpqi_mem.c.rej
./sys/dev/smartpqi/smartpqi_queue.c.rej
./sys/dev/smartpqi/smartpqi_defines.h.rej
./sys/dev/smartpqi/smartpqi_cam.c.rej
./sys/dev/smartpqi/smartpqi_misc.c.rej
./sys/dev/smartpqi/smartpqi_main.c.rej
./sys/dev/smartpqi/smartpqi_request.c.rej

In the main branch, it crashed patch :(.
A quick sample of the .rej files shows the diffs likely are easy to resolve by hand, but with such a large patch I'm leery to do so. Add '-l' to patch to cope with whitespace changes didn't seem to help.
So it looks like this patch needs to be regenerated and/or moved to git where patch generation and uploading is a bit more reliable.

(Also commented on a couple of nits that didn't look quite right in the copyright stuff, but that can wait for the patch to get done).

Hi,

• 12.0 stable branch and 12.2 main branch has two different source codes. I've pulled the 12.0 source code and applied the patch. (12.0 stable branch has bug fixes which is done by community but I do not see the same changes in 12.2 main branch because of that the patch is failing on 12.2 main branch).

"main" mentioned is literally main git branch, where this change should go first (and it's 14.0-CURRENT at the moment). I have HBA 1100-8i that is misbehaving under load, so I'd really like to try this patch -- could you please rebase this against main?

Updated driver.

In D24428#657918, @imp wrote:

There's issues with applying this patch. Something seems to have gone amiss in its generation.

It almost applied cleanly to stable/12 branch, but not to the main branch:

% find . -name \*.rej
./sys/dev/smartpqi/smartpqi_mem.c.rej
./sys/dev/smartpqi/smartpqi_queue.c.rej
./sys/dev/smartpqi/smartpqi_defines.h.rej
./sys/dev/smartpqi/smartpqi_cam.c.rej
./sys/dev/smartpqi/smartpqi_misc.c.rej
./sys/dev/smartpqi/smartpqi_main.c.rej
./sys/dev/smartpqi/smartpqi_request.c.rej

In the main branch, it crashed patch :(.
A quick sample of the .rej files shows the diffs likely are easy to resolve by hand, but with such a large patch I'm leery to do so. Add '-l' to patch to cope with whitespace changes didn't seem to help.
So it looks like this patch needs to be regenerated and/or moved to git where patch generation and uploading is a bit more reliable.

(Also commented on a couple of nits that didn't look quite right in the copyright stuff, but that can wait for the patch to get done).

In D24428#644782, @rainer_ultra-secure.de wrote:

Please excuse me, I am only a user, I only try to get this working.

(f-hosting </root>) 130 # cd /usr/src
(f-hosting <src>) 0 # git apply /root/D24428.diff
/root/D24428.diff:1875: trailing whitespace.
_softs->pci_mem_handle.pqi_bhandle, _offset)
/root/D24428.diff:3930: space before tab in indent.

	DBG_FUNC("IN\n");

/root/D24428.diff:3944: space before tab in indent.

	DBG_FUNC("OUT\n");

/root/D24428.diff:6642: trailing whitespace.

• Populate hostwellness time variables in bcd format from FreeBSD format

/root/D24428.diff:7949: space before tab in indent.

		sizeof(raid_req->lun_number));

error: dev/smartpqi/smartpqi_cam.c: No such file or directory
error: dev/smartpqi/smartpqi_cmd.c: No such file or directory
error: dev/smartpqi/smartpqi_defines.h: No such file or directory
error: dev/smartpqi/smartpqi_discovery.c: No such file or directory
error: dev/smartpqi/smartpqi_event.c: No such file or directory
error: dev/smartpqi/smartpqi_helper.c: No such file or directory
error: dev/smartpqi/smartpqi_includes.h: No such file or directory
error: dev/smartpqi/smartpqi_init.c: No such file or directory
error: dev/smartpqi/smartpqi_intr.c: No such file or directory
error: dev/smartpqi/smartpqi_ioctl.h: No such file or directory
error: dev/smartpqi/smartpqi_ioctl.c: No such file or directory
error: dev/smartpqi/smartpqi_main.c: No such file or directory
error: dev/smartpqi/smartpqi_mem.c: No such file or directory
error: dev/smartpqi/smartpqi_misc.c: No such file or directory
error: dev/smartpqi/smartpqi_prototypes.h: No such file or directory
error: dev/smartpqi/smartpqi_queue.c: No such file or directory
error: dev/smartpqi/smartpqi_request.c: No such file or directory
error: dev/smartpqi/smartpqi_response.c: No such file or directory
error: dev/smartpqi/smartpqi_sis.c: No such file or directory
error: dev/smartpqi/smartpqi_structures.h: No such file or directory
error: dev/smartpqi/smartpqi_tag.c: No such file or directory

I know it's not your job - but can you walk me through what I have got to do to get this to compile on 12.2-RELEASE-p3?

In D24428#644381, @rainer_ultra-secure.de wrote:

In D24428#641982, @papani.srikanth_microchip.com wrote:

Fixed system crash while creating and deleting logical volume in a continuous loop.
Fixed where the volume size is not exposing to OS when it expands.
Added HC3 pci id's.
Fixed compiler issues in 12.2 kernel.

I can't seem to apply the patch cleanly on 12.2-RELEASE-p3.

See my comment in the PR above.

Fixed white space errors

Updated year in Copyright section.

Fixed system crash while creating and deleting logical volume in a continuous loop.
Fixed where the volume size is not exposing to OS when it expands.
Added HC3 pci id's.
Fixed compiler issues in 12.2 kernel.

In D24428#640413, @rainer_ultra-secure.de wrote:

This does not compile with 12.2.

See
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=240145

Can someone look at this?

Yes, we're working on the issues. I will update the latest driver soon.

Updated Inbox Driver version to match with latest Out-of-box driver 1x.4014.0.105

Added newly fixed system crashes

Hi All,

comments resolved

Added bug fixes which is happened recently