Date: Thu, 7 Aug 2025 19:18:21 +0100 From: Andrew Turner <andrew@FreeBSD.org> To: Jessica Clarke <jrtc27@freebsd.org> Cc: "src-committers@freebsd.org" <src-committers@FreeBSD.org>, "dev-commits-src-all@freebsd.org" <dev-commits-src-all@FreeBSD.org>, "dev-commits-src-main@freebsd.org" <dev-commits-src-main@FreeBSD.org> Subject: Re: git: 81f07332c03f - main - arm64: tidy up Top-Byte-Ignore (TBI) in the kernel Message-ID: <6BCAD216-03E1-4DA3-B0BB-542366F003DB@FreeBSD.org> In-Reply-To: <BA3B207C-6D34-4748-AE4D-DCD5AFA27789@freebsd.org> References: <202508061738.576Hc0wZ076618@gitrepo.freebsd.org> <BA3B207C-6D34-4748-AE4D-DCD5AFA27789@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
[-- Attachment #1 --] > On 6 Aug 2025, at 18:43, Jessica Clarke <jrtc27@freebsd.org> wrote: > > On 6 Aug 2025, at 18:38, Andrew Turner <andrew@freebsd.org <mailto:andrew@freebsd.org>> wrote: >> >> The branch main has been updated by andrew: >> >> URL: https://cgit.FreeBSD.org/src/commit/?id=81f07332c03fd2ac6efa8e15b1659a573d250329 >> >> commit 81f07332c03fd2ac6efa8e15b1659a573d250329 >> Author: Harry Moulton <harry.moulton@arm.com> >> AuthorDate: 2025-07-31 14:10:57 +0000 >> Commit: Andrew Turner <andrew@FreeBSD.org> >> CommitDate: 2025-07-31 14:27:06 +0000 >> >> arm64: tidy up Top-Byte-Ignore (TBI) in the kernel >> >> In preparation for TBI to be enabled for processes from 15.0 we need >> to clean up copying data between userspace and the kernel. These >> functions will check the address is within the valid userspace range, >> however as the userspace and kernel ranges may overlap when TBI is >> enabled we need to mask off the top 8 bits. >> >> Processes not using TBI are unaffected as the hardware will still >> check all bits in the address, however this will happen at the first >> load/store instruction. > > I thought Linux’s ABI was that addresses passed to the kernel must be > canonical. Is that false, or is there a reason we’re diverging from > that design choice? On Linux there is a flag to enable passing tagged pointers to the kernel. It’s needed for HWASAN to work as any pointers may be tagged, and I expect it will also be needed for MTE. I don’t know of a strong reason to not allow tagged pointers in the kernel. Andrew [-- Attachment #2 --] <html><head><meta http-equiv="content-type" content="text/html; charset=utf-8"></head><body style="overflow-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;"><br id="lineBreakAtBeginningOfMessage"><div><br><blockquote type="cite"><div>On 6 Aug 2025, at 18:43, Jessica Clarke <jrtc27@freebsd.org> wrote:</div><br class="Apple-interchange-newline"><div><meta charset="UTF-8"><span style="caret-color: rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; float: none; display: inline !important;">On 6 Aug 2025, at 18:38, Andrew Turner <</span><a href="mailto:andrew@freebsd.org" style="font-family: Helvetica; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px;">andrew@freebsd.org</a><span style="caret-color: rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; float: none; display: inline !important;">> wrote:</span><br style="caret-color: rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;"><blockquote type="cite" style="font-family: Helvetica; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; orphans: auto; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; widows: auto; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;"><br>The branch main has been updated by andrew:<br><br>URL: https://cgit.FreeBSD.org/src/commit/?id=81f07332c03fd2ac6efa8e15b1659a573d250329<br><br>commit 81f07332c03fd2ac6efa8e15b1659a573d250329<br>Author: Harry Moulton <harry.moulton@arm.com><br>AuthorDate: 2025-07-31 14:10:57 +0000<br>Commit: Andrew Turner <andrew@FreeBSD.org><br>CommitDate: 2025-07-31 14:27:06 +0000<br><br> arm64: tidy up Top-Byte-Ignore (TBI) in the kernel<br><br> In preparation for TBI to be enabled for processes from 15.0 we need<br> to clean up copying data between userspace and the kernel. These<br> functions will check the address is within the valid userspace range,<br> however as the userspace and kernel ranges may overlap when TBI is<br> enabled we need to mask off the top 8 bits.<br><br> Processes not using TBI are unaffected as the hardware will still<br> check all bits in the address, however this will happen at the first<br> load/store instruction.<br></blockquote><br style="caret-color: rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;"><span style="caret-color: rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; float: none; display: inline !important;">I thought Linux’s ABI was that addresses passed to the kernel must be</span><br style="caret-color: rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;"><span style="caret-color: rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; float: none; display: inline !important;">canonical. Is that false, or is there a reason we’re diverging from</span><br style="caret-color: rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;"><span style="caret-color: rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none; float: none; display: inline !important;">that design choice?</span><br style="caret-color: rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: normal; font-variant-caps: normal; font-weight: 400; letter-spacing: normal; text-align: start; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px; -webkit-text-stroke-width: 0px; text-decoration: none;"></div></blockquote></div><br><div>On Linux there is a flag to enable passing tagged pointers to the kernel. It’s needed for HWASAN to work as any pointers may be tagged, and I expect it will also be needed for MTE.</div><div><br></div><div>I don’t know of a strong reason to not allow tagged pointers in the kernel.</div><div><br></div><div>Andrew</div></body></html>
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?6BCAD216-03E1-4DA3-B0BB-542366F003DB>