The NVMe spec allows the Table BIR (TBIR) and PBA DIR (PBIR) to be 0, 4, or 5. The existing NVMe driver basically only has support for 4, perhaps under the assumption that BAR4 is 64-bit and also occupies BAR5. This change adds support for BAR5, covering the case where BAR4 and BAR5 might both be present and 32-bit, where the Table BIR might be 4 and the PBA BIR might be 5, or vice versa. The NVMe spec (in the SR-IOV section) also permits VFs to use BIR=2, so I haven't added stricter checks on which BIR will be permitted by the driver. This enables FreeBSD on Google Compute Engine C4 Machines. MFC after: 3 days Reviewed by: imp Sponsored by: Google Co-authored-by: Matt Delco <delco@google.com> Signed-off-by: Jasper Tran O'Leary <jtranoleary@google.com> Differential Revision: https://reviews.freebsd.org/D53140 (cherry picked from commit 7b32f4f0a7fe9b1b2f5a3905ca15f656713255ad)
In particular, don't try to byteswap the values as 64-bit integers and always print a non-empty version as a string. Reviewed by: chuck, imp Sponsored by: Chelsio Communications Differential Revision: https://reviews.freebsd.org/D44121 (cherry picked from commit 7485926e09a08fbfe83a9bc908e7d43aaca4c172)
KIOXIA CD8 SSDs routinely take ~25 seconds to delete non-empty namespace. In some cases like hot-plug it takes longer, triggering timeout and controller resets after just 30 seconds. Linux for many years has separate 60 seconds timeout for admin queue. This patch does the same. And it is good to be consistent. Sponsored by: iXsystems, Inc. Reviewed by: imp MFC after: 1 week Differential Revision: https://reviews.freebsd.org/D42454 (cherry picked from commit 8d6c0743e36e3cff9279c40468711a82db98df23)
When running nvme passthrough commands through the ioctl interface
memory is mapped with vmapbuf() but not unmapped. This results in leaked
memory whenever a process executes an nvme passthrough command with a
data buffer. This can be replicated with a simple c function (error
checks skipped for brevity):
void leak_memory(int nvme_ns_fd, uint16_t nblocks) {
struct nvme_pt_command pt = {
.cmd = {
.opc = NVME_OPC_READ,
.cdw12 = nblocks - 1,
},
.len = nblocks * 512, // Assumes devices with 512 byte lba
.is_read = 1, // Reads and writes should both trigger leak
}
void *buf;
posix_memalign(&buf, nblocks * 512);
pt.buf = buf;
ioctl(nvme_ns_fd, NVME_PASSTHROUGH_COMMAND, &pt);
free(buf);
}
Signed-off-by: David Sloan <david.sloan@eideticom.com>
PR: 273626
Reviewed by: imp, markj
MFC after: 1 week
(cherry picked from commit 7ea866eb14f8ec869a525442c03228b6701e1dab)
Remove /^[\s*]*__FBSDID\("\$FreeBSD\$"\);?\s*\n/
Similar commit in current:
(cherry picked from commit 685dc743dc3b)
The SPDX folks have obsoleted the BSD-2-Clause-FreeBSD identifier. Catch up to that fact and revert to their recommended match of BSD-2-Clause. Discussed with: pfg MFC After: 3 days Sponsored by: Netflix (cherry picked from commit 4d846d260e2b9a3d4d0a701462568268cbfe7a5b)
- s/inaccessable/inaccessible/ (cherry picked from commit 6e8ab6715d3589ad5577ad5fa408f81b4a7077e9)
We only use nvme_completion_poll in the initialization path. The commands they queue and wait for finish quickly as they involve no I/O to the drive's media. These command take about 20-200 microsecnds each. Set the wait time to 1us and then increase it by 1.5 each successive iteration (max 1ms). This reduces initialization time by 80ms in cpervica's tests. Use this same technique waiting for RDY state transitions. This saves another 20ms. In total we're down from ~330ms to ~2ms. Tested by: cperciva Sponsored by: Netflix Reviewed by: mav Differential Review: https://reviews.freebsd.org/D32259 (cherry picked from commit 83581511d9476ef5084f47e3cc379be7191ae866)
The FreeBSD nvme driver has reset the nvme controller twice on attach to address a theoretical issue assuring the hardware is in a known state. However, exierence has shown the second reset is unnecessary and increases the time to boot. Eliminate the second reset. Should there be a situation when you need a second reset (for buggy or at least somewhat out of the mainstream hardware), the hardware option NVME_2X_RESET will restore the old behavior. Document this in nvme(4). If there's any trouble at all with this, I'll add a sysctl tunable to control it. Sponsored by: Netflix Reviewed by: cperciva, mav Differential Revision: https://reviews.freebsd.org/D32241 (cherry picked from commit 4b3da659bf62b0f5306b5acee9add41b84361498)
After some study of the code and the standard, I think we can just drop the pause(), unconditionally. If we're not initialized, then there's nothing to wait for from a software perspective. If we are initialized, then there might be outstanding I/O. If so, then the qpair 'recovery state' will transition to WAITING in nvme_ctrlr_disable_qpairs, which will ignore any interrupts for items that complete before we complete the reset by setting cc.en=0. If we go on to fail the controller, we'll cancel the outstanding I/O transactions. If we reset the controller, the hardware throws away pending transactions and we retry all the pending I/O transactions. Any transactions that happend to complete before cc.en=0 will have the same effect in the end (doing the same transaction twice is just inefficient, it won't affect the state of the device any differently than having done it once). The standard imposes no wait times here, so it isn't needed from that perspective. Unanswered Question: Do we may need to disable interrupts while we disable in legacy mode since those are level-sensitive. Sponsored by: Netflix Reviewed by: mav Differential Revision: https://reviews.freebsd.org/D32248 (cherry picked from commit e5e26e4a24a1142e02a9b477877e13ed0c194f36)