Date: Fri, 23 Feb 2024 10:55:51 -0700 From: Warner Losh <imp@bsdimp.com> To: "Bjoern A. Zeeb" <bzeeb-lists@lists.zabbadoz.net> Cc: FreeBSD FS <freebsd-fs@freebsd.org> Subject: Re: really slow problem with nvme Message-ID: <CANCZdfoBc7bW6so8H4=dD-XogPp0uQKr1pZjRvSH5%2B8HEGMTCg@mail.gmail.com> In-Reply-To: <5n117579-8259-3on3-90pn-79o8n52q4q32@yvfgf.mnoonqbm.arg> References: <5n117579-8259-3on3-90pn-79o8n52q4q32@yvfgf.mnoonqbm.arg>
next in thread | previous in thread | raw e-mail | index | archive | help
--0000000000000203660612104608 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Fri, Feb 23, 2024, 10:46=E2=80=AFAM Bjoern A. Zeeb < bzeeb-lists@lists.zabbadoz.net> wrote: > Hi, > > this is a Samsung SSD 970 EVO Plus 1TB nvme and gpart and newfs > were already slow (it took like two hours for newfs). > > Here's another example now: > > # /usr/bin/time mkdir foo > 1.82 real 0.00 user 0.00 sys > > How does one debug this? > What filesystem? Sounds like UFS but just making sure. . So what's the link speed and number of lanes? If it's bad i might reseat (though that might not help) that looks good... Though I'd bet money that this is an interrupt issue. I'd do a vmstat. -i to watch how quickly they accumulate... How old is the drive? Fresh install? Do other drives have this same issue in the same slot? Dies this drive have issues in other maxhines or slots? Warner --=20 > Bjoern A. Zeeb r15:7 > > --0000000000000203660612104608 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable <div dir=3D"auto"><div><br><br><div class=3D"gmail_quote"><div dir=3D"ltr" = class=3D"gmail_attr">On Fri, Feb 23, 2024, 10:46=E2=80=AFAM Bjoern A. Zeeb = <<a href=3D"mailto:bzeeb-lists@lists.zabbadoz.net">bzeeb-lists@lists.zab= badoz.net</a>> wrote:<br></div><blockquote class=3D"gmail_quote" style= =3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi,<br> <br> this is a Samsung SSD 970 EVO Plus 1TB nvme and gpart and newfs<br> were already slow (it took like two hours for newfs).<br> <br> Here's another example now:<br> <br> # /usr/bin/time mkdir foo<br> =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A01.82 real=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A00.00 user=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A00.00 sys<br> <br> How does one debug this?<br></blockquote></div></div><div dir=3D"auto"><br>= </div><div dir=3D"auto">What filesystem? Sounds like UFS but just making su= re. .</div><div dir=3D"auto"><br></div><div dir=3D"auto">So what's the = link speed and number of lanes?=C2=A0 If it's bad i might reseat (thoug= h that might not help) that looks good...</div><div dir=3D"auto"><br></div>= <div dir=3D"auto">Though I'd bet money that this is an interrupt issue.= I'd do a vmstat. -i to watch how quickly they accumulate...</div><div = dir=3D"auto"><br></div><div dir=3D"auto">How old is the drive? Fresh instal= l? Do other drives have this same issue in the same slot? Dies this drive h= ave issues in other maxhines or slots?</div><div dir=3D"auto"><br></div><di= v dir=3D"auto">Warner</div><div dir=3D"auto"><br></div><div dir=3D"auto"><d= iv class=3D"gmail_quote"><blockquote class=3D"gmail_quote" style=3D"margin:= 0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> -- <br> Bjoern A. Zeeb=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 = =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0r15:7<br> <br> </blockquote></div></div></div> --0000000000000203660612104608--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CANCZdfoBc7bW6so8H4=dD-XogPp0uQKr1pZjRvSH5%2B8HEGMTCg>