From: "Michael Kerrisk (man-pages)" <mtk.manpages@gmail.com> To: Andi Kleen <ak@linux.intel.com> Cc: Andi Kleen <andi@firstfloor.org>, akpm@linux-foundation.org, linux-mm@kvack.org, linux-kernel@vger.kernel.org, Hillf Danton <dhillf@gmail.com> Subject: Re: [PATCH] MM: Support more pagesizes for MAP_HUGETLB/SHM_HUGETLB v6 Date: Mon, 22 Oct 2012 17:53:45 +0200 [thread overview] Message-ID: <CAKgNAki=AL+KdYDdYnE8ZhjK-tUf5cZ163BWPe6GRM0rpi-z7w@mail.gmail.com> (raw) In-Reply-To: <20121022153633.GK2095@tassilo.jf.intel.com> On Mon, Oct 22, 2012 at 5:36 PM, Andi Kleen <ak@linux.intel.com> wrote: >> Not sure of your notation there. I assume 31..27 means 5 bits (32 >> through to 28 inclusive, 27 excluded). That gives you just 2^31 == > [27...31] (Hmm -- sleeping as I wrote that, but I at least got the end point right.) > You're right it's only 5 bits, so just 2GB. > > Thinking about it more PowerPC has a 16GB page, so we probably > need to move this to prot. > > However I'm not sure if any architectures use let's say the high > 8 bits of prot. This is all seems to make an awful muck of the API... I'm not sure whether anything is using the high 8 bits of prot, bun passing I note that there seems to be no check that the unused bits are zeroed so there's a small chance existing apps are passing random garbage there. (Of course, mmap() is hardly the only API to have that fault, and it hasn't stopped us from reusing bits in those APIs, though sometimes we've gotten bitten by apps that did pass in random garbage). >> But there seems an obvious solution here: given your value in those >> bits (call it 'n'), the why not apply a multiplier. I mean, certainly >> you never want a value <= 12 for n, and I suspect that the reasonable >> minimum could be much larger (e.g., 2^16). Call that minimum M. Then >> you could interpret the value in your bits as meaning a page size of >> >> (2^n) * M > > I considered that, but it would seem ugly and does not add that > many bits. > >> >> > So this will use up all remaining flag bits now. >> >> On the other hand, that seems really bad. It looks like that kills the >> ability to further extend the mmap() API with new flags in the future. >> It doesn't sound like we should be doing that. > > You can always add flags to PROT or add a mmap3(). Has been done before. > Or just don't do any new MAP_SECURITY_HOLEs There seems to be a reasonable argument here for an mmap3() with a 64-bit flags argument... Cheers, Michael -- Michael Kerrisk Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/ Author of "The Linux Programming Interface"; http://man7.org/tlpi/
WARNING: multiple messages have this Message-ID (diff)
From: "Michael Kerrisk (man-pages)" <mtk.manpages@gmail.com> To: Andi Kleen <ak@linux.intel.com> Cc: Andi Kleen <andi@firstfloor.org>, akpm@linux-foundation.org, linux-mm@kvack.org, linux-kernel@vger.kernel.org, Hillf Danton <dhillf@gmail.com> Subject: Re: [PATCH] MM: Support more pagesizes for MAP_HUGETLB/SHM_HUGETLB v6 Date: Mon, 22 Oct 2012 17:53:45 +0200 [thread overview] Message-ID: <CAKgNAki=AL+KdYDdYnE8ZhjK-tUf5cZ163BWPe6GRM0rpi-z7w@mail.gmail.com> (raw) In-Reply-To: <20121022153633.GK2095@tassilo.jf.intel.com> On Mon, Oct 22, 2012 at 5:36 PM, Andi Kleen <ak@linux.intel.com> wrote: >> Not sure of your notation there. I assume 31..27 means 5 bits (32 >> through to 28 inclusive, 27 excluded). That gives you just 2^31 == > [27...31] (Hmm -- sleeping as I wrote that, but I at least got the end point right.) > You're right it's only 5 bits, so just 2GB. > > Thinking about it more PowerPC has a 16GB page, so we probably > need to move this to prot. > > However I'm not sure if any architectures use let's say the high > 8 bits of prot. This is all seems to make an awful muck of the API... I'm not sure whether anything is using the high 8 bits of prot, bun passing I note that there seems to be no check that the unused bits are zeroed so there's a small chance existing apps are passing random garbage there. (Of course, mmap() is hardly the only API to have that fault, and it hasn't stopped us from reusing bits in those APIs, though sometimes we've gotten bitten by apps that did pass in random garbage). >> But there seems an obvious solution here: given your value in those >> bits (call it 'n'), the why not apply a multiplier. I mean, certainly >> you never want a value <= 12 for n, and I suspect that the reasonable >> minimum could be much larger (e.g., 2^16). Call that minimum M. Then >> you could interpret the value in your bits as meaning a page size of >> >> (2^n) * M > > I considered that, but it would seem ugly and does not add that > many bits. > >> >> > So this will use up all remaining flag bits now. >> >> On the other hand, that seems really bad. It looks like that kills the >> ability to further extend the mmap() API with new flags in the future. >> It doesn't sound like we should be doing that. > > You can always add flags to PROT or add a mmap3(). Has been done before. > Or just don't do any new MAP_SECURITY_HOLEs There seems to be a reasonable argument here for an mmap3() with a 64-bit flags argument... Cheers, Michael -- Michael Kerrisk Linux man-pages maintainer; http://www.kernel.org/doc/man-pages/ Author of "The Linux Programming Interface"; http://man7.org/tlpi/ -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2012-10-22 15:54 UTC|newest] Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top 2012-10-19 16:48 [PATCH] MM: Support more pagesizes for MAP_HUGETLB/SHM_HUGETLB v6 Andi Kleen 2012-10-19 16:48 ` Andi Kleen 2012-10-20 3:39 ` Hillf Danton 2012-10-20 3:39 ` Hillf Danton 2012-10-22 11:27 ` Michael Kerrisk 2012-10-22 11:27 ` Michael Kerrisk 2012-10-22 13:27 ` Andi Kleen 2012-10-22 13:27 ` Andi Kleen 2012-10-22 13:35 ` Andi Kleen 2012-10-22 13:35 ` Andi Kleen 2012-10-22 13:56 ` Michael Kerrisk (man-pages) 2012-10-22 13:56 ` Michael Kerrisk (man-pages) 2012-10-22 15:36 ` Andi Kleen 2012-10-22 15:36 ` Andi Kleen 2012-10-22 15:53 ` Michael Kerrisk (man-pages) [this message] 2012-10-22 15:53 ` Michael Kerrisk (man-pages) 2012-10-22 16:11 ` Andi Kleen 2012-10-22 16:11 ` Andi Kleen 2012-10-22 16:23 ` Michael Kerrisk (man-pages) 2012-10-22 16:23 ` Michael Kerrisk (man-pages) 2012-10-22 16:29 ` Andi Kleen 2012-10-22 16:29 ` Andi Kleen 2012-10-22 16:42 ` Michael Kerrisk (man-pages) 2012-10-22 16:42 ` Michael Kerrisk (man-pages) 2012-10-23 1:45 ` Benjamin Herrenschmidt 2012-10-23 1:45 ` Benjamin Herrenschmidt 2012-10-23 1:44 ` Benjamin Herrenschmidt 2012-10-23 1:44 ` Benjamin Herrenschmidt 2012-10-23 2:28 ` Andi Kleen 2012-10-23 2:28 ` Andi Kleen 2012-10-23 7:37 ` Michael Kerrisk (man-pages) 2012-10-23 7:37 ` Michael Kerrisk (man-pages) 2012-10-23 13:39 ` Andi Kleen 2012-10-23 13:39 ` Andi Kleen 2012-10-23 22:56 ` Valdis.Kletnieks 2012-10-22 21:39 ` Andrew Morton 2012-10-22 21:39 ` Andrew Morton
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to='CAKgNAki=AL+KdYDdYnE8ZhjK-tUf5cZ163BWPe6GRM0rpi-z7w@mail.gmail.com' \ --to=mtk.manpages@gmail.com \ --cc=ak@linux.intel.com \ --cc=akpm@linux-foundation.org \ --cc=andi@firstfloor.org \ --cc=dhillf@gmail.com \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-mm@kvack.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.