From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:52612) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bor36-0004LX-46 for qemu-devel@nongnu.org; Tue, 27 Sep 2016 07:59:49 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1bor31-0001Ny-3T for qemu-devel@nongnu.org; Tue, 27 Sep 2016 07:59:47 -0400 Received: from mx1.redhat.com ([209.132.183.28]:43772) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bor30-0001Nq-Qb for qemu-devel@nongnu.org; Tue, 27 Sep 2016 07:59:43 -0400 Date: Tue, 27 Sep 2016 13:59:39 +0200 From: Kevin Wolf Message-ID: <20160927115939.GI4090@noname.str.redhat.com> References: <1474970326-10271-1-git-send-email-pl@kamp.de> <1474970326-10271-2-git-send-email-pl@kamp.de> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1474970326-10271-2-git-send-email-pl@kamp.de> Subject: Re: [Qemu-devel] [PATCH V9 1/7] oslib-posix: add helpers for stack alloc and free List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Peter Lieven Cc: qemu-devel@nongnu.org, mreitz@redhat.com, pbonzini@redhat.com, mst@redhat.com, dgilbert@redhat.com, peter.maydell@linaro.org, eblake@redhat.com, rth@twiddle.net, armbru@redhat.com Am 27.09.2016 um 11:58 hat Peter Lieven geschrieben: > the allocated stack will be adjusted to the minimum supported stack size > by the OS and rounded up to be a multiple of the system pagesize. > Additionally an architecture dependent guard page is added to the stack > to catch stack overflows. > > Signed-off-by: Peter Lieven > --- > include/sysemu/os-posix.h | 27 +++++++++++++++++++++++++++ > util/oslib-posix.c | 42 ++++++++++++++++++++++++++++++++++++++++++ > 2 files changed, 69 insertions(+) > > diff --git a/include/sysemu/os-posix.h b/include/sysemu/os-posix.h > index 9c7dfdf..3cfedbc 100644 > --- a/include/sysemu/os-posix.h > +++ b/include/sysemu/os-posix.h > @@ -60,4 +60,31 @@ int qemu_utimens(const char *path, const qemu_timespec *times); > > bool is_daemonized(void); > > +/** > + * qemu_alloc_stack: > + * @sz: pointer to a size_t holding the requested usable stack size > + * > + * Allocate memory that can be used as a stack, for instance for > + * coroutines. If the memory cannot be allocated, this function > + * will abort (like g_malloc()). This function also inserts an > + * additional guard page to catch a potential stack overflow. > + * Note that the memory required for the guard page and alignment > + * and minimal stack size restrictions will increase the value of sz. > + * > + * The allocated stack must be freed with qemu_free_stack(). > + * > + * Returns: pointer to (the lowest address of) the stack memory. > + */ > +void *qemu_alloc_stack(size_t *sz); > + > +/** > + * qemu_free_stack: > + * @stack: stack to free > + * @sz: size of stack in bytes > + * > + * Free a stack allocated via qemu_alloc_stack(). Note that sz must > + * be exactly the adjusted stack size returned by qemu_alloc_stack. > + */ > +void qemu_free_stack(void *stack, size_t sz); > + > #endif > diff --git a/util/oslib-posix.c b/util/oslib-posix.c > index f2d4e9e..5745229 100644 > --- a/util/oslib-posix.c > +++ b/util/oslib-posix.c > @@ -499,3 +499,45 @@ pid_t qemu_fork(Error **errp) > } > return pid; > } > + > +void *qemu_alloc_stack(size_t *sz) > +{ > + void *ptr, *guardpage; > + size_t pagesz = getpagesize(); > +#ifdef _SC_THREAD_STACK_MIN > + /* avoid stacks smaller than _SC_THREAD_STACK_MIN */ > + long min_stack_sz = sysconf(_SC_THREAD_STACK_MIN); > + *sz = MAX(MAX(min_stack_sz, 0), *sz); > +#endif > + /* adjust stack size to a multiple of the page size */ > + *sz = ROUND_UP(*sz, pagesz); > + /* allocate one extra page for the guard page */ > + *sz += pagesz; > + > + ptr = mmap(NULL, *sz, PROT_READ | PROT_WRITE, > + MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); > + if (ptr == MAP_FAILED) { > + abort(); > + } > + > +#if defined(HOST_IA64) > + /* separate register stack */ > + guardpage = ptr + (((sz - pagesz) / 2) & ~pagesz); s/sz/*sz/ > +#elif defined(HOST_HPPA) > + /* stack grows up */ > + guardpage = ptr + sz - pagesz; Here too. I can fix both while applying the series. > +#else > + /* stack grows down */ > + guardpage = ptr; > +#endif > + if (mprotect(guardpage, pagesz, PROT_NONE) != 0) { > + abort(); > + } > + > + return ptr; > +} Kevin