L2T entries are used by both filters and TOE and the L2T is shared
between the base driver (LLD) and the TOM ULD. Add a flag to indicate
that the L2T is stopped, which means:
* t4_alloc_l2e and t4_l2t_alloc_switching will not allocate new entries.
* t4_tom will ignore all ARP/NDP updates from the kernel.
* Previously allocated L2T entries can still be freed.
MFC after: 1 month
Sponsored by: Chelsio Communications
atids are used by both filters and TOE and the atid table is in the base
driver (LLD). New atids cannot be allocated when the allocator is
stopped but existing ones can still be freed. It is expected that the
owners of outstanding atids will release them in their own stop
processing, before the adapter is restarted.
MFC after: 1 month
Sponsored by: Chelsio Communications
The suspend/resume/reset implementation in the base driver (LLD)
currently works when only stateless features are in use. This commit
adds basic infrastructure for stateful upper layer drivers (ULDs) to
participate in suspend/resume/reset.
* Add a uld_restart to indicate that the adapter has been restarted
after a stop and the ULD should resume operations.
* Move the existing functionality in t4_suspend/t4_resume to stop_lld and
restart_lld. Use these and the new uld restart routines everywhere the
adapter has to be stopped abruptly and restarted, namely:
1. PCIE bus suspend/resume/reset methods invoked by the kernel.
2. Manual internal-reset using driver sysctl.
3. Automatic internal-reset on a fatal error.
* Implement an alternate internal-reset for use in VMs and for testing.
Typical reset sequence is:
stop_adapter(sc);
stop_lld(sc);
stop_all_uld(sc);
set_adapter_hwstatus(sc, false);
/* hw reset takes place here. */
restart_adapter(sc);
restart_lld(sc);
set_adapter_hwstatus(sc, true);
restart_all_uld(sc);
MFC after: 1 month
Sponsored by: Chelsio Communications
Allow a TCP function block (tfb) to become the default stack only if
tfb->tfb_flags has the TCP_FUNC_DEFAULT_OK flags set. This allows a
TCP function block, that is not suitable as a default function block
to ensure that it is not set as the default via sysctl. In this case
sysctl would return EINVAL.
Reviewed by: gallatin, Peter Lei
Sponsored by: Netflix, Inc.
Differential Revision: https://reviews.freebsd.org/D45419
The NFS RFCs are pretty loose with respect to what characters
can be in a filename returned by a Readdir. However, FreeBSD,
as a POSIX system will not handle imbedded '/' or nul characters
in file names. Also, for NFSv4, the file names "." and ".."
are handcrafted on the client and should not be returned by a
NFSv4 server.
This patch scans for the above in filenames returned by Readdir and
ignores any entry returned by Readdir which has them in it.
Because an imbedded nul would be a string terminator, it was
not possible to code this check efficiently using string(3)
functions.
Reported by: Apple Security Engineering and Architecture (SEAR)
MFC after: 1 week
Use the intended TCP stack when creating a TCP endpoint instead of
creating it the endpoint the default stack first and after that
switching it to use the intended TCP stack.
Reviewed by: Peter Lei, rrs and jtl (older version)
Sponsored by: Netflix, Inc.
Differential Revision: https://reviews.freebsd.org/D45411
Implement the improved SEG.ACK validation described in RFC 5961.
In addition to that, also detect ghost ACKs, which are ACKs for data
that has never been sent.
The additional checks are enabled by default, but can be disabled
by setting the sysctl-variable net.inet.tcp.insecure_ack to a
non-zero value.
PR: 250357
Reviewed by: Peter Lei, rscheff (older version)
MFC after: 1 week
Sponsored by: Netflix, Inc.
Differential Revision: https://reviews.freebsd.org/D45894
The existing pretty-printing code fails to properly print complex
pointer types. This commit fixes this behaviour by traversing the
chain of CTF types until a base type is encountered.
Approved by: markj (mentor)
Fixes: c21bc6f3c2
The struct pretty-printing code uses the ctm_offset field in
struct ctf_member_v3 to calculate the address of a struct member.
However, the code treats this as a byte offset rather than the
offset in bits, leading to wrong values being printed.
Fix this by diving with ctm_offset by NBBY.
Approved by: markj (mentor)
Fixes: c21bc6f3c2
If the resource was allocated with want_iomap_res flag set.
Sponsored by: Serenity CyberSecurity, LLC
MFC after: 1 week
Reviewed by: manu, bz
Differential Revision: https://reviews.freebsd.org/D45905
pci_iomap_range creates a virtual mapping cookie for a PCI BAR.
As compared with pci_iomap it got extra offset parameter.
Sponsored by: Serenity CyberSecurity, LLC
MFC after: 1 week
Reviewed by: manu, bz
Differential Revision: https://reviews.freebsd.org/D45904
acpi_dev_get_first_match_dev returns the first match of ACPI device
and acpi_device_handle returns its ACPI handle.
Sponsored by: Serenity Cyber Security, LLC
MFC after: 1 week
Reviewed by: manu
Differential Revision: https://reviews.freebsd.org/D45847
It finds out if a given PCI device matches a given pci_id table.
Sponsored by: Serenity Cyber Security, LLC
MFC after: 1 week
Reviewed by: manu
Differential Revision: https://reviews.freebsd.org/D45846
devm_device_add_group creates a managed attribute group for a device.
Sponsored by: Serenity Cyber Security, LLC
MFC after: 1 week
Reviewed by: manu
Differential Revision: https://reviews.freebsd.org/D45845
__diag macros turn individual warnings and errors on and off locally,
depending on version of compiler. Add dummy implementation as drm-kmod
set warnings separately for each file.
Sponsored by: Serenity Cyber Security, LLC
MFC after: 1 week
Reviewed by: manu, emaste
Differential Revision: https://reviews.freebsd.org/D45842
Allocate to a simple poiter, use that everywhere, then return it at the
end. The code looks cleaner.
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D46043
I was testing this in several trees, and this chunk was in the tree I
tested. I didn't detect the mismerge before the commit...
Fixes: 6f1dd6071a
Sponsored by: Netflix
Make nvme_qpair_manual_complete_request take dnr as well as a
print_on_error action. Make the status word computation common between
it and nvme_qpair_manual_complete_tracker. And print the error when
we are cancelling the I/O on failure, but not when we're filtering
the I/O after we've failed. Make it private again to nvme_qpair.c.
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D46049
When the drive is failed, we can't send passthrough commands to the
card, so fail them right away. Rearrange the comments to reflect the
current failure paths in the driver.
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D46048
When we init the iosched instance, require clients to provide a schedule
function. We have almost, but not quite everything to know when it is
safe to schedule new I/O. The periph drivers, however, have all the
information, so make them do it when the I/O Scheduler needs to maybe
schedule I/O for rate limiting, etc. and use it to do that.
Sponsored by: Netflix
Reviewed by: jhb
Differential Revision: https://reviews.freebsd.org/D46038
Log outlier latency events to devd. In addition to counting, this will
allow analysis of whether the problem is confined to a specific block
range, or if it's a more general problem.
Sponsored by: Netflix
Reviewed by: jhb
Differential Revision: https://reviews.freebsd.org/D46036
The disk is nice to have at times, especially when you need the sector
size. At present, the only plans for this are related to logging.
Sponsored by: Netflix
Reviewed by: jhb
Differential Revision: https://reviews.freebsd.org/D46035
Pass the struct bio * down to the stats collection routines. This allows
us to collect better stats and report more details.
Sponsored by: Netflix
Reviewed by: jhb
Differential Revision: https://reviews.freebsd.org/D46034
Add kern.cam.DEV.UNIT.iosched.too_long (to count I/Os taking too long)
and kern.cam.DEV.UNIT.bad_latency (to set this threshold, defaults to
500ms). Each class of I/O (read, write, trim) has its own counters and
thresholds.
Sponsored by: Netflix
Reviewed by: jhb
Differential Revision: https://reviews.freebsd.org/D46033
GCC 14 defines __has_extension(c_atomic) but does not support
__c11_atomic_*, so require __clang__ for the CLANG_ATOMICS case.
Reviewed by: imp, emaste
Differential Revision: https://reviews.freebsd.org/D46001
* Convert t4_uld_list to an array. There will be at most 3 items in the
list and it's simpler to track them in an array with a fixed slot for
each ULD.
* There is no need to refcount ULDs so stop doing that.
* Add uld_ prefix to all members of uld_info.
* Rename async_event to uld_stop to match its actual purpose. Call it
for all ULDs and not just ULD_IWARP.
Reviewed by: jhb
MFC after: 1 week
Sponsored by: Chelsio Communications
Differential Revision: https://reviews.freebsd.org/D46029
Do not assume that the table starts at index 0 and is typically 4K in
size. The only thing the driver needs to verify is that its use of
F_SYNC_WR doesn't collide with the L2T hwidx range.
Reviewed by: jhb
MFC after: 1 week
Sponsored by: Chelsio Communications
Differential Revision: https://reviews.freebsd.org/D46028
Final CPL means the tid is done in the hardware and other resources
associated with it can be freed right away. There is no need to wait
for the kernel to detach the toepcb.
Reviewed by: jhb
MFC after: 1 week
Sponsored by: Chelsio Communications
Differential Revision: https://reviews.freebsd.org/D45991
The kernel used to call tod_pcb_detach when entering TIME_WAIT but that
seems to have changed, likely with the TIME_WAIT overhaul in the kernel
some time ago. Catch up by having the driver perform the detach.
The hardware does not handle TIME_WAIT so it's important to detach and
let the kernel arm the 2MSL timer to deal with it.
Reported by: Sony Arpita Das @ Chelsio
Reviewed by: jhb
MFC after: 1 week
Sponsored by: Chelsio Communications
Differential Revision: https://reviews.freebsd.org/D45990
A recent change dropped the only use of fls64() in axgbe, so drop its
definition.
Reviewed by: markj
Differential Revision: https://reviews.freebsd.org/D46030
Drop an implementation of __rounddown_pow_of_two, which triggers
Coverity warnings, and use the libkern implementation of
rounddown_pow_of_two instead.
Reviewed by: markj
Differential Revision: https://reviews.freebsd.org/D46022
All architectures define PCI_RES_BUS unconditionally now that only
NEW_PCIB is supported, so we no longer need the legacy code.
Reviewed by: jhb, andrew
Differential Revision: https://reviews.freebsd.org/D32955
All architectures enable NEW_PCIB in DEFAULTS (arm being the most recent
to do so in 121be55599 (arm: Set NEW_PCIB in DEFAULTS rather than a
subset of kernel configs")), so it's time we removed the legacy code
that no longer sees much testing and has a significant maintenance
burden.
Reviewed by: jhb, andrew, emaste
Differential Revision: https://reviews.freebsd.org/D32954
