Date: Sat, 07 Jun 2025 02:19:11 -0400 From: Ian FREISLICH <ianfreislich@gmail.com> To: <freebsd-current@dino.sk>, <freebsd-current@freebsd.org> Subject: Re: UFS bad inode, mangled entry on Alder Lake-N(100) Message-ID: <197490b0f18.2762.64e08aff09ba5a21b2fc9010d26a90e5@gmail.com> In-Reply-To: <20250607074739.31a7116b@gema.dino.sk> References: <ab27e2fa-d798-47f8-876b-b79a54a36003@gmail.com> <6530f044-486d-4128-9e23-ee03e6686aa9@yamagi.org> <20250128122345.74075269.25648751.10101138@dino.sk> <6024769d-1cac-439b-874d-c624112d30c9@gmail.com> <20250607074739.31a7116b@gema.dino.sk>
index | next in thread | previous in thread | raw e-mail
[-- Attachment #1 --] On June 7, 2025 01:47:56 freebsd-current@dino.sk wrote: > On Tue, 28 Jan 2025 15:33:30 -0500 > Ian FREISLICH <ianfreislich@gmail.com> wrote: > >> On 2025-01-28 06:23, Milan Obuch wrote: > > [ snip ] > >>> It looks like the right thing, in my case adding >>> vm.pmap.pcid_enabled=0 to /boot/loader.conf helps. I consider this >>> easier than installing port for microcode update... >>> >>> That being said, could someone add some more pro/cons for those two >>> approaches? >>> >>> Additionally, I am using M.2 SATA drive at the moment. While NVMe >>> drive worked to some extent, if fsck was necessary for some reason, >>> it was unpleasant - some 'waiting for nvme reset' event occured, >>> this led to nvme drive detach, and the only way to fix it was >>> unscrew the drive, put it in USB-NVMe converter, do fsck via USB >>> drive, then mount it back into box... not acceptable. >> >> I chose microcode but that was hard to do because I only have one >> nvme slot and the installer panicked trying to install the package at >> the final part of the install. I had to install onto an SD and then >> use another FreeBSD install to do a pkg chroot install onto that >> temporary media and then use that to boot with the firmware update >> and chroot install the firmware and edit loader.conf on the nvme. >> >> The microcode update fixed it for me. I inferred from reading that >> enable PCID might have a performance advantage. > > Well, I run with microcode update and NVMe on my Alder Lake box with no > problem. Yesterday, however, the symptoms were back - bad inode error > for filesystem, fsck necessary. Probably some data loss... (not a big > problem, this was still device under test, and I can easily get the > somewhat important things from drive) > > What may be relevant is it happened after system upgrade to more recent > 14.3-STABLE from some a bit older 14.3-PRERELEASE, but I think > microcode update binary is OS independent, or is there some dependency? > > I am going to try reinstall with disabled PCID and test it again. I'm running 15.0-CURRENT #24 main-n277779-2a5841795fb7: Fri Jun 6 22:06:35 EDT 2025 and rebuilding all my ports without issue. I'm on the latest microcode (0x1d) as well. Ian [-- Attachment #2 --] <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd">; <html> <body> <div dir="auto"> <div dir="auto"><span style="font-size: 12pt;">On June 7, 2025 01:47:56 freebsd-current@dino.sk wrote:</span></div><div id="aqm-original" style="color: black;"> <div><br></div> <blockquote type="cite" class="gmail_quote" style="margin: 0 0 0 0.75ex; border-left: 1px solid #808080; padding-left: 0.75ex;"> <div dir="auto">On Tue, 28 Jan 2025 15:33:30 -0500</div> <div dir="auto">Ian FREISLICH <ianfreislich@gmail.com> wrote:</div> <div dir="auto"><br></div> <blockquote type="cite" class="gmail_quote" style="margin: 0 0 0 0.75ex; border-left: 1px solid #0099CC; padding-left: 0.75ex;"> <div dir="auto">On 2025-01-28 06:23, Milan Obuch wrote:</div> </blockquote> <div dir="auto"><br></div> <div dir="auto">[ snip ]</div> <div dir="auto"><br></div> <blockquote type="cite" class="gmail_quote" style="margin: 0 0 0 0.75ex; border-left: 1px solid #0099CC; padding-left: 0.75ex;"> <blockquote type="cite" class="gmail_quote" style="margin: 0 0 0 0.75ex; border-left: 1px solid #9933CC; padding-left: 0.75ex;"> <div dir="auto">It looks like the right thing, in my case adding</div> <div dir="auto">vm.pmap.pcid_enabled=0 to /boot/loader.conf helps. I consider this</div> <div dir="auto">easier than installing port for microcode update...</div> <div dir="auto"><br></div> <div dir="auto">That being said, could someone add some more pro/cons for those two</div> <div dir="auto">approaches?</div> <div dir="auto"><br></div> <div dir="auto">Additionally, I am using M.2 SATA drive at the moment. While NVMe</div> <div dir="auto">drive worked to some extent, if fsck was necessary for some reason,</div> <div dir="auto">it was unpleasant - some 'waiting for nvme reset' event occured,</div> <div dir="auto">this led to nvme drive detach, and the only way to fix it was</div> <div dir="auto">unscrew the drive, put it in USB-NVMe converter, do fsck via USB</div> <div dir="auto">drive, then mount it back into box... not acceptable. </div> </blockquote> <div dir="auto"><br></div> <div dir="auto">I chose microcode but that was hard to do because I only have one</div> <div dir="auto">nvme slot and the installer panicked trying to install the package at</div> <div dir="auto">the final part of the install. I had to install onto an SD and then</div> <div dir="auto">use another FreeBSD install to do a pkg chroot install onto that</div> <div dir="auto">temporary media and then use that to boot with the firmware update</div> <div dir="auto">and chroot install the firmware and edit loader.conf on the nvme.</div> <div dir="auto"><br></div> <div dir="auto">The microcode update fixed it for me. I inferred from reading that </div> <div dir="auto">enable PCID might have a performance advantage.</div> <div dir="auto"><br></div> </blockquote> <div dir="auto"><br></div> <div dir="auto">Well, I run with microcode update and NVMe on my Alder Lake box with no</div> <div dir="auto">problem. Yesterday, however, the symptoms were back - bad inode error</div> <div dir="auto">for filesystem, fsck necessary. Probably some data loss... (not a big</div> <div dir="auto">problem, this was still device under test, and I can easily get the</div> <div dir="auto">somewhat important things from drive)</div> <div dir="auto"><br></div> <div dir="auto">What may be relevant is it happened after system upgrade to more recent</div> <div dir="auto">14.3-STABLE from some a bit older 14.3-PRERELEASE, but I think</div> <div dir="auto">microcode update binary is OS independent, or is there some dependency?</div> <div dir="auto"><br></div> <div dir="auto">I am going to try reinstall with disabled PCID and test it again.</div></blockquote></div><div dir="auto"><br></div><div dir="auto">I'm running 15.0-CURRENT #24 main-n277779-2a5841795fb7: Fri Jun 6 22:06:35 EDT 2025</div><div dir="auto">and rebuilding all my ports without issue. I'm on the latest microcode (0x1d) as well.</div><div dir="auto"><br></div><div dir="auto">Ian</div> </div></body> </html>help
Want to link to this message? Use this
URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?197490b0f18.2762.64e08aff09ba5a21b2fc9010d26a90e5>