From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([140.186.70.92]:40041) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1RXitP-00049J-FU for qemu-devel@nongnu.org; Mon, 05 Dec 2011 19:28:20 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1RXitO-0006ud-Fy for qemu-devel@nongnu.org; Mon, 05 Dec 2011 19:28:19 -0500 Received: from cantor2.suse.de ([195.135.220.15]:40277 helo=mx2.suse.de) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1RXitO-0006uZ-A2 for qemu-devel@nongnu.org; Mon, 05 Dec 2011 19:28:18 -0500 From: Alexander Graf Date: Tue, 6 Dec 2011 00:44:55 +0100 Message-Id: <1323128695-11292-1-git-send-email-agraf@suse.de> Subject: [Qemu-devel] [PATCH] linux-user: map lower in address space List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel Developers Cc: Peter Maydell , Riku Voipio While trying to compile Java I can into situations where there was simply no virtual address space left for a 32-bit guest to take. For example when Java tried to allocate 1GB of heap. Part of the problem is that we're starting to map things at 0x40000000. This is a bit high. Taking that number down would give us a lot of free virtual address space which means we'd be able to squeeze more stuff in. Signed-off-by: Alexander Graf --- linux-user/mmap.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/linux-user/mmap.c b/linux-user/mmap.c index 7d846f3..8453c0d 100644 --- a/linux-user/mmap.c +++ b/linux-user/mmap.c @@ -226,7 +226,7 @@ static int mmap_frag(abi_ulong real_start, /* Cygwin doesn't have a whole lot of address space. */ # define TASK_UNMAPPED_BASE 0x18000000 #else -# define TASK_UNMAPPED_BASE 0x40000000 +# define TASK_UNMAPPED_BASE 0x10000000 #endif static abi_ulong mmap_next_start = TASK_UNMAPPED_BASE; -- 1.6.0.2