From owner-freebsd-hackers@FreeBSD.ORG Tue Feb 5 16:48:37 2013 Return-Path: Delivered-To: hackers@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) by hub.freebsd.org (Postfix) with ESMTP id 0089BEDC; Tue, 5 Feb 2013 16:48:36 +0000 (UTC) (envelope-from alc@rice.edu) Received: from mh3.mail.rice.edu (mh3.mail.rice.edu [128.42.199.10]) by mx1.freebsd.org (Postfix) with ESMTP id CDB7F86F; Tue, 5 Feb 2013 16:48:36 +0000 (UTC) Received: from mh3.mail.rice.edu (localhost.localdomain [127.0.0.1]) by mh3.mail.rice.edu (Postfix) with ESMTP id 060454033C; Tue, 5 Feb 2013 10:48:30 -0600 (CST) Received: from mh3.mail.rice.edu (localhost.localdomain [127.0.0.1]) by mh3.mail.rice.edu (Postfix) with ESMTP id 024814033D; Tue, 5 Feb 2013 10:48:30 -0600 (CST) X-Virus-Scanned: by amavis-2.7.0 at mh3.mail.rice.edu, auth channel Received: from mh3.mail.rice.edu ([127.0.0.1]) by mh3.mail.rice.edu (mh3.mail.rice.edu [127.0.0.1]) (amavis, port 10026) with ESMTP id a1gWwsFap9aw; Tue, 5 Feb 2013 10:48:29 -0600 (CST) Received: from adsl-216-63-78-18.dsl.hstntx.swbell.net (adsl-216-63-78-18.dsl.hstntx.swbell.net [216.63.78.18]) (using TLSv1 with cipher RC4-MD5 (128/128 bits)) (No client certificate requested) (Authenticated sender: alc) by mh3.mail.rice.edu (Postfix) with ESMTPSA id E8FF54033C; Tue, 5 Feb 2013 10:48:28 -0600 (CST) Message-ID: <511137DC.20303@rice.edu> Date: Tue, 05 Feb 2013 10:48:28 -0600 From: Alan Cox User-Agent: Mozilla/5.0 (X11; FreeBSD i386; rv:17.0) Gecko/20130127 Thunderbird/17.0.2 MIME-Version: 1.0 To: Konstantin Belousov Subject: Re: dynamically calculating NKPT [was: Re: huge ktr buffer] References: <20130205151413.GL2522@kib.kiev.ua> <20130205161335.GM2522@kib.kiev.ua> In-Reply-To: <20130205161335.GM2522@kib.kiev.ua> Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: davide@freebsd.org, mdf@FreeBSD.org, alc@freebsd.org, avg@freebsd.org, rank1seeker@gmail.com, hackers@freebsd.org, Neel Natu X-BeenThere: freebsd-hackers@freebsd.org X-Mailman-Version: 2.1.14 Precedence: list List-Id: Technical Discussions relating to FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 05 Feb 2013 16:48:37 -0000 On 02/05/2013 10:13, Konstantin Belousov wrote: > On Tue, Feb 05, 2013 at 07:45:24AM -0800, mdf@FreeBSD.org wrote: >> On Tue, Feb 5, 2013 at 7:14 AM, Konstantin Belousov wrote: >>> On Mon, Feb 04, 2013 at 03:05:15PM -0800, Neel Natu wrote: >>>> Hi, >>>> >>>> I have a patch to dynamically calculate NKPT for amd64 kernels. This >>>> should fix the various issues that people pointed out in the email >>>> thread. >>>> >>>> Please review and let me know if there are any objections to committing this. >>>> >>>> Also, thanks to Alan (alc@) for reviewing and providing feedback on >>>> the initial version of the patch. >>>> >>>> Patch (also available at http://people.freebsd.org/~neel/patches/nkpt_diff.txt): >>>> >>>> Index: sys/amd64/include/pmap.h >>>> =================================================================== >>>> --- sys/amd64/include/pmap.h (revision 246277) >>>> +++ sys/amd64/include/pmap.h (working copy) >>>> @@ -113,13 +113,7 @@ >>>> ((unsigned long)(l2) << PDRSHIFT) | \ >>>> ((unsigned long)(l1) << PAGE_SHIFT)) >>>> >>>> -/* Initial number of kernel page tables. */ >>>> -#ifndef NKPT >>>> -#define NKPT 32 >>>> -#endif >>>> - >>>> #define NKPML4E 1 /* number of kernel PML4 slots */ >>>> -#define NKPDPE howmany(NKPT, NPDEPG)/* number of kernel PDP slots */ >>>> >>>> #define NUPML4E (NPML4EPG/2) /* number of userland PML4 pages */ >>>> #define NUPDPE (NUPML4E*NPDPEPG)/* number of userland PDP pages */ >>>> @@ -181,6 +175,7 @@ >>>> #define PML4map ((pd_entry_t *)(addr_PML4map)) >>>> #define PML4pml4e ((pd_entry_t *)(addr_PML4pml4e)) >>>> >>>> +extern int nkpt; /* Initial number of kernel page tables */ >>>> extern u_int64_t KPDPphys; /* physical address of kernel level 3 */ >>>> extern u_int64_t KPML4phys; /* physical address of kernel level 4 */ >>>> >>>> Index: sys/amd64/amd64/minidump_machdep.c >>>> =================================================================== >>>> --- sys/amd64/amd64/minidump_machdep.c (revision 246277) >>>> +++ sys/amd64/amd64/minidump_machdep.c (working copy) >>>> @@ -232,7 +232,7 @@ >>>> /* Walk page table pages, set bits in vm_page_dump */ >>>> pmapsize = 0; >>>> pdp = (uint64_t *)PHYS_TO_DMAP(KPDPphys); >>>> - for (va = VM_MIN_KERNEL_ADDRESS; va < MAX(KERNBASE + NKPT * NBPDR, >>>> + for (va = VM_MIN_KERNEL_ADDRESS; va < MAX(KERNBASE + nkpt * NBPDR, >>>> kernel_vm_end); ) { >>>> /* >>>> * We always write a page, even if it is zero. Each >>>> @@ -364,7 +364,7 @@ >>>> /* Dump kernel page directory pages */ >>>> bzero(fakepd, sizeof(fakepd)); >>>> pdp = (uint64_t *)PHYS_TO_DMAP(KPDPphys); >>>> - for (va = VM_MIN_KERNEL_ADDRESS; va < MAX(KERNBASE + NKPT * NBPDR, >>>> + for (va = VM_MIN_KERNEL_ADDRESS; va < MAX(KERNBASE + nkpt * NBPDR, >>>> kernel_vm_end); va += NBPDP) { >>>> i = (va >> PDPSHIFT) & ((1ul << NPDPEPGSHIFT) - 1); >>>> >>>> Index: sys/amd64/amd64/pmap.c >>>> =================================================================== >>>> --- sys/amd64/amd64/pmap.c (revision 246277) >>>> +++ sys/amd64/amd64/pmap.c (working copy) >>>> @@ -202,6 +202,10 @@ >>>> vm_offset_t virtual_avail; /* VA of first avail page (after kernel bss) */ >>>> vm_offset_t virtual_end; /* VA of last avail page (end of kernel AS) */ >>>> >>>> +int nkpt; >>>> +SYSCTL_INT(_machdep, OID_AUTO, nkpt, CTLFLAG_RD, &nkpt, 0, >>>> + "Number of kernel page table pages allocated on bootup"); >>>> + >>>> static int ndmpdp; >>>> static vm_paddr_t dmaplimit; >>>> vm_offset_t kernel_vm_end = VM_MIN_KERNEL_ADDRESS; >>>> @@ -495,17 +499,42 @@ >>>> >>>> CTASSERT(powerof2(NDMPML4E)); >>>> >>>> +/* number of kernel PDP slots */ >>>> +#define NKPDPE(ptpgs) howmany((ptpgs), NPDEPG) >>>> + >>>> static void >>>> +nkpt_init(vm_paddr_t addr) >>>> +{ >>>> + int pt_pages; >>>> + >>>> +#ifdef NKPT >>>> + pt_pages = NKPT; >>>> +#else >>>> + pt_pages = howmany(addr, 1 << PDRSHIFT); >>>> + pt_pages += NKPDPE(pt_pages); >>>> + >>>> + /* >>>> + * Add some slop beyond the bare minimum required for bootstrapping >>>> + * the kernel. >>>> + * >>>> + * This is quite important when allocating KVA for kernel modules. >>>> + * The modules are required to be linked in the negative 2GB of >>>> + * the address space. If we run out of KVA in this region then >>>> + * pmap_growkernel() will need to allocate page table pages to map >>>> + * the entire 512GB of KVA space which is an unnecessary tax on >>>> + * physical memory. >>>> + */ >>>> + pt_pages += 4; /* 8MB additional slop for kernel modules */ >>> 8MB might be to low. I just checked one of my machines with fully >>> modularized kernel, it takes slightly more than 6 MB to load 50 modules. >>> I think that 16MB would be safer, but it probably needs to be scaled >>> down based on the available phys memory. amd64 kernel could be booted >>> on 128MB machine still. >> Is there no way to not map the entire 512GB? Otherwise this patch >> could really hose some vendors. E.g. the kernel module for the OneFS >> file system is around 8MB all by itself. > No, I do not think that this patch would hose somebody with the 8MB > module, esp. if the slack is increased. Agreed. With an increase in the slack, this patch can't possibly harm anyone. On the other hand, it will eliminate the need for some people to manually tune NKPT. > But yes, I believe it is possible to note that the growth happen after > the KERNBASE point and only allocate the page tables at this region. > We would need to not update the kernel_vm_end then, probably creating > some other var to keep track of the other tail. Yes, this can probably be done. However, what Neel has is already an improvement, so I see no point in not committing it with an increase in the slack. >> I found when we moved from FreeBSD 6 to 7 that the NKPT of 32 was >> insufficient for our system to even boot so I put it back to 240 (I >> didn't want to spend a lot of time playing). At that time our module >> was loaded by the boot loader; now we do it during init to save some >> seconds on boot. But we're probably not the only ones with a large >> kernel module. >> >> Cheers, >> matthew