From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:54929) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1XEXHJ-0005jc-7d for qemu-devel@nongnu.org; Tue, 05 Aug 2014 01:27:23 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1XEXHD-0001pE-IU for qemu-devel@nongnu.org; Tue, 05 Aug 2014 01:27:17 -0400 Received: from mailout2.w1.samsung.com ([210.118.77.12]:47532) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1XEXHD-0001ow-9B for qemu-devel@nongnu.org; Tue, 05 Aug 2014 01:27:11 -0400 Received: from eucpsbgm2.samsung.com (unknown [203.254.199.245]) by mailout2.w1.samsung.com (Oracle Communications Messaging Server 7u4-24.01(7.0.4.24.0) 64bit (built Nov 17 2011)) with ESMTP id <0N9T00HQDJ4OJM10@mailout2.w1.samsung.com> for qemu-devel@nongnu.org; Tue, 05 Aug 2014 06:26:48 +0100 (BST) Message-id: <53E06B29.5040806@samsung.com> Date: Tue, 05 Aug 2014 09:27:05 +0400 From: Mikhail Ilin MIME-version: 1.0 References: <53D63BE6.4090806@samsung.com> In-reply-to: <53D63BE6.4090806@samsung.com> Content-type: text/plain; charset=utf-8; format=flowed Content-transfer-encoding: 7bit Subject: Re: [Qemu-devel] [PATCH] /proc/self/maps content is not correct for a guest List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org Cc: mst@redhat.com, hutao@cn.fujitsu.com, riku.voipio@iki.fi, anthony@codemonkey.ws, pbonzini@redhat.com, afaerber@suse.de ping http://patchwork.ozlabs.org/patch/374162/ On 28.07.2014 16:02, Mikhail Ilin wrote: > Hi, > > As it was posted earlier the output of reading /proc/self/maps is not > correct for a guest. There are some issues: > > https://bugs.launchpad.net/qemu/+bug/1346784 > http://lists.nongnu.org/archive/html/qemu-devel/2014-07/msg03085.html > http://lists.nongnu.org/archive/html/qemu-devel/2014-07/msg02793.html > > The patch proposes: build /proc/self/maps doing a match against guest > memory > translation table and output only that map records which are valid for > guest > memory layout. > > Patches in mentioned threads are not relevant and are covered by the > current > patch. > > We did some local tests for i386, x86_64 and arm targets. The approach > seems correct. > > > From 8479d3dd00194975d7016eeecba13ddf453e9647 Mon Sep 17 00:00:00 2001 > From: Mikhail Ilyin > Date: Mon, 28 Jul 2014 15:40:31 +0400 > Subject: [PATCH] Build /proc/self/maps doing a match against guest memory > translation table. Output only that map records which are valid for guest > memory layout. > > Signed-off-by: Mikhail Ilyin > --- > include/exec/cpu-all.h | 2 ++ > linux-user/syscall.c | 25 ++++++++++--------------- > 2 files changed, 12 insertions(+), 15 deletions(-) > > diff --git a/include/exec/cpu-all.h b/include/exec/cpu-all.h > index f91581f..f9d132f 100644 > --- a/include/exec/cpu-all.h > +++ b/include/exec/cpu-all.h > @@ -198,6 +198,8 @@ extern unsigned long reserved_va; > #define RESERVED_VA 0ul > #endif > > +#define GUEST_ADDR_MAX (RESERVED_VA ? RESERVED_VA : \ > + (1ul << > TARGET_VIRT_ADDR_SPACE_BITS) - 1) > #endif > > /* page related stuff */ > diff --git a/linux-user/syscall.c b/linux-user/syscall.c > index a50229d..189a8c0 100644 > --- a/linux-user/syscall.c > +++ b/linux-user/syscall.c > @@ -5092,10 +5092,8 @@ static int open_self_cmdline(void *cpu_env, int fd) > > static int open_self_maps(void *cpu_env, int fd) > { > -#if defined(TARGET_ARM) || defined(TARGET_M68K) || > defined(TARGET_UNICORE32) > CPUState *cpu = ENV_GET_CPU((CPUArchState *)cpu_env); > TaskState *ts = cpu->opaque; > -#endif > FILE *fp; > char *line = NULL; > size_t len = 0; > @@ -5118,13 +5116,18 @@ static int open_self_maps(void *cpu_env, int fd) > if ((fields < 10) || (fields > 11)) { > continue; > } > - if (!strncmp(path, "[stack]", 7)) { > - continue; > - } > - if (h2g_valid(min) && h2g_valid(max)) { > + if (h2g_valid(min)) { > + int flags = page_get_flags(h2g(min)); > + max = h2g_valid(max - 1) ? max : > (uint64_t)g2h(GUEST_ADDR_MAX); > + if (page_check_range(h2g(min), max - min, flags) == -1) { > + continue; > + } > + if (h2g(min) == ts->info->stack_limit) { > + pstrcpy(path, sizeof(path), " [stack]"); > + } > dprintf(fd, TARGET_ABI_FMT_lx "-" TARGET_ABI_FMT_lx > " %c%c%c%c %08" PRIx64 " %02x:%02x %d %s%s\n", > - h2g(min), h2g(max), flag_r, flag_w, > + h2g(min), h2g(max - 1) + 1, flag_r, flag_w, > flag_x, flag_p, offset, dev_maj, dev_min, inode, > path[0] ? " " : "", path); > } > @@ -5133,14 +5136,6 @@ static int open_self_maps(void *cpu_env, int fd) > free(line); > fclose(fp); > > -#if defined(TARGET_ARM) || defined(TARGET_M68K) || > defined(TARGET_UNICORE32) > - dprintf(fd, "%08llx-%08llx rw-p %08llx 00:00 0 [stack]\n", > - (unsigned long long)ts->info->stack_limit, > - (unsigned long long)(ts->info->start_stack + > - (TARGET_PAGE_SIZE - 1)) & > TARGET_PAGE_MASK, > - (unsigned long long)0); > -#endif > - > return 0; > } >