mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: blaisorblade_spam@yahoo.it
To: jdike@addtoit.com
Cc: linux-kernel@vger.kernel.org,
	user-mode-linux-devel@lists.sourceforge.net,
	blaisorblade_spam@yahoo.it
Subject: [patch 1/1] uml-malloc-use-vmalloc-fix-again
Date: Sun, 19 Sep 2004 11:46:01 +0200	[thread overview]
Message-ID: <20040919094602.7D09890DD@zion.localdomain> (raw)


Since we sometimes disable temporarily kmalloc_ok (don't ask me why), we
must test, inside __wrap_free, the new CAN_KFREE macro; and then we must update
every reference to kmalloc_ok to handle correctly even kfree_ok. Also encapsulate
a bit more this code.

Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade_spam@yahoo.it>
---

 uml-linux-2.6.8.1-paolo/arch/um/include/kern_util.h   |   10 ++++++-
 uml-linux-2.6.8.1-paolo/arch/um/kernel/main.c         |   24 ++++++++++++------
 uml-linux-2.6.8.1-paolo/arch/um/kernel/mem.c          |    5 ++-
 uml-linux-2.6.8.1-paolo/arch/um/kernel/physmem.c      |    2 -
 uml-linux-2.6.8.1-paolo/arch/um/kernel/process_kern.c |    6 ++--
 uml-linux-2.6.8.1-paolo/arch/um/kernel/skas/process.c |    4 +--
 uml-linux-2.6.8.1-paolo/arch/um/kernel/tt/tracer.c    |    2 -
 7 files changed, 35 insertions(+), 18 deletions(-)

diff -puN arch/um/include/kern_util.h~uml-malloc-use-vmalloc-fix-again arch/um/include/kern_util.h
--- uml-linux-2.6.8.1/arch/um/include/kern_util.h~uml-malloc-use-vmalloc-fix-again	2004-09-17 17:19:13.946898416 +0200
+++ uml-linux-2.6.8.1-paolo/arch/um/include/kern_util.h	2004-09-17 17:19:13.984892640 +0200
@@ -12,7 +12,15 @@
 extern int ncpus;
 extern char *linux_prog;
 extern char *gdb_init;
-extern int kmalloc_ok;
+
+extern int __kmalloc_ok;
+extern int __kfree_ok;
+
+#define KMALLOC_START() ({ __kmalloc_ok = __kfree_ok = 1; })
+#define KMALLOC_STOP() ({ __kmalloc_ok = __kfree_ok = 0; })
+#define KMALLOC_SAVE(save) ({ (save) = __kmalloc_ok; __kmalloc_ok = 0; })
+#define KMALLOC_RESTORE(save) ({ __kmalloc_ok = (save); })
+
 extern int timer_irq_inited;
 extern int jail;
 extern int nsyscalls;
diff -puN arch/um/kernel/main.c~uml-malloc-use-vmalloc-fix-again arch/um/kernel/main.c
--- uml-linux-2.6.8.1/arch/um/kernel/main.c~uml-malloc-use-vmalloc-fix-again	2004-09-17 17:19:13.947898264 +0200
+++ uml-linux-2.6.8.1-paolo/arch/um/kernel/main.c	2004-09-17 17:19:13.984892640 +0200
@@ -68,7 +68,7 @@ static __init void do_uml_initcalls(void
 
 static void last_ditch_exit(int sig)
 {
-	CHOOSE_MODE(kmalloc_ok = 0, (void) 0);
+	CHOOSE_MODE(KMALLOC_STOP(), (void) 0);
 	signal(SIGINT, SIG_DFL);
 	signal(SIGTERM, SIG_DFL);
 	signal(SIGHUP, SIG_DFL);
@@ -172,7 +172,11 @@ int main(int argc, char **argv, char **e
 }
 
 #define CAN_KMALLOC() \
-	(kmalloc_ok && CHOOSE_MODE((getpid() != tracing_pid), 1))
+	(__kmalloc_ok && CHOOSE_MODE((getpid() != tracing_pid), 1))
+
+/*Must we still check if we are not in the tracing thread? I'm not sure about this.*/
+#define CAN_KFREE() \
+	(__kfree_ok && CHOOSE_MODE((getpid() != tracing_pid), 1))
 
 extern void *__real_malloc(int);
 
@@ -211,6 +215,8 @@ extern unsigned long high_physmem;
 void __wrap_free(void *ptr)
 {
 	unsigned long addr = (unsigned long) ptr;
+	if (addr == 0)
+		return;
 
 	/* We need to know how the allocation happened, so it can be correctly
 	 * freed.  This is done by seeing what region of memory the pointer is
@@ -218,9 +224,10 @@ void __wrap_free(void *ptr)
 	 * 	physical memory - kmalloc/kfree
 	 *	kernel virtual memory - vmalloc/vfree
 	 * 	anywhere else - malloc/free
-	 * If kmalloc is not yet possible, then the kernel memory regions
-	 * may not be set up yet, and the variables not set up.  So,
-	 * free is called.
+	 * If kmalloc is not yet possible, then high_physmem and end_vm are 0,
+	 * so we will always use __real_free.
+	 * And since we disable sometimes __kmalloc_ok while running,
+	 * we cannot check CAN_KMALLOC. CAN_KFREE is created exactly for this.
 	 *
 	 * CAN_KMALLOC is checked because it would be bad to free a buffer
 	 * with kmalloc/vmalloc after they have been turned off during
@@ -228,14 +235,15 @@ void __wrap_free(void *ptr)
 	 */
 
 	if((addr >= uml_physmem) && (addr < high_physmem)){
-		if(CAN_KMALLOC())
+		if(CAN_KFREE())
 			kfree(ptr);
 	}
 	else if((addr >= start_vm) && (addr < end_vm)){
-		if(CAN_KMALLOC())
+		if(CAN_KFREE())
 			vfree(ptr);
 	}
-	else __real_free(ptr);
+	else
+		__real_free(ptr);
 }
 
 /*
diff -puN arch/um/kernel/mem.c~uml-malloc-use-vmalloc-fix-again arch/um/kernel/mem.c
--- uml-linux-2.6.8.1/arch/um/kernel/mem.c~uml-malloc-use-vmalloc-fix-again	2004-09-17 17:19:13.948898112 +0200
+++ uml-linux-2.6.8.1-paolo/arch/um/kernel/mem.c	2004-09-17 17:19:13.984892640 +0200
@@ -27,7 +27,8 @@ unsigned long *empty_zero_page = NULL;
 unsigned long *empty_bad_page = NULL;
 pgd_t swapper_pg_dir[1024];
 unsigned long highmem;
-int kmalloc_ok = 0;
+int __kmalloc_ok = 0;
+int __kfree_ok = 0;
 
 static unsigned long brk_end;
 
@@ -98,7 +99,7 @@ void mem_init(void)
 	max_pfn = totalram_pages;
 	printk(KERN_INFO "Memory: %luk available\n", 
 	       (unsigned long) nr_free_pages() << (PAGE_SHIFT-10));
-	kmalloc_ok = 1;
+	KMALLOC_START();
 
 #ifdef CONFIG_HIGHMEM
 	setup_highmem(end_iomem, highmem);
diff -puN arch/um/kernel/physmem.c~uml-malloc-use-vmalloc-fix-again arch/um/kernel/physmem.c
--- uml-linux-2.6.8.1/arch/um/kernel/physmem.c~uml-malloc-use-vmalloc-fix-again	2004-09-17 17:19:13.949897960 +0200
+++ uml-linux-2.6.8.1-paolo/arch/um/kernel/physmem.c	2004-09-17 17:19:13.985892488 +0200
@@ -291,7 +291,7 @@ int init_maps(unsigned long physmem, uns
 	total_pages = phys_pages + iomem_pages + highmem_pages;
 	total_len = phys_len + iomem_pages + highmem_len;
 
-	if(kmalloc_ok){
+	if(__kmalloc_ok){
 		map = kmalloc(total_len, GFP_KERNEL);
 		if(map == NULL) 
 			map = vmalloc(total_len);
diff -puN arch/um/kernel/process_kern.c~uml-malloc-use-vmalloc-fix-again arch/um/kernel/process_kern.c
--- uml-linux-2.6.8.1/arch/um/kernel/process_kern.c~uml-malloc-use-vmalloc-fix-again	2004-09-17 17:19:13.950897808 +0200
+++ uml-linux-2.6.8.1-paolo/arch/um/kernel/process_kern.c	2004-09-17 17:19:13.985892488 +0200
@@ -171,12 +171,12 @@ int copy_thread(int nr, unsigned long cl
 
 void initial_thread_cb(void (*proc)(void *), void *arg)
 {
-	int save_kmalloc_ok = kmalloc_ok;
+	int save_kmalloc_ok;
 
-	kmalloc_ok = 0;
+	KMALLOC_SAVE(save_kmalloc_ok);
 	CHOOSE_MODE_PROC(initial_thread_cb_tt, initial_thread_cb_skas, proc, 
 			 arg);
-	kmalloc_ok = save_kmalloc_ok;
+	KMALLOC_RESTORE(save_kmalloc_ok);
 }
  
 unsigned long stack_sp(unsigned long page)
diff -puN arch/um/kernel/skas/process.c~uml-malloc-use-vmalloc-fix-again arch/um/kernel/skas/process.c
--- uml-linux-2.6.8.1/arch/um/kernel/skas/process.c~uml-malloc-use-vmalloc-fix-again	2004-09-17 17:19:13.951897656 +0200
+++ uml-linux-2.6.8.1-paolo/arch/um/kernel/skas/process.c	2004-09-17 17:19:13.985892488 +0200
@@ -325,11 +325,11 @@ int start_idle_thread(void *stack, void 
 		siglongjmp(*cb_back, 1);
 	}
 	else if(n == 3){
-		kmalloc_ok = 0;
+		KMALLOC_STOP();
 		return(0);
 	}
 	else if(n == 4){
-		kmalloc_ok = 0;
+		KMALLOC_STOP();
 		return(1);
 	}
 	siglongjmp(**switch_buf, 1);
diff -puN arch/um/kernel/tt/tracer.c~uml-malloc-use-vmalloc-fix-again arch/um/kernel/tt/tracer.c
--- uml-linux-2.6.8.1/arch/um/kernel/tt/tracer.c~uml-malloc-use-vmalloc-fix-again	2004-09-17 17:19:13.952897504 +0200
+++ uml-linux-2.6.8.1-paolo/arch/um/kernel/tt/tracer.c	2004-09-17 17:19:13.986892336 +0200
@@ -310,7 +310,7 @@ int tracer(int (*init_proc)(void *), voi
 				case OP_REBOOT:
 				case OP_HALT:
 					unmap_physmem();
-					kmalloc_ok = 0;
+					KMALLOC_STOP();
 					ptrace(PTRACE_KILL, pid, 0, 0);
 					return(op == OP_REBOOT);
 				case OP_NONE:
_

                 reply	other threads:[~2004-09-19  9:52 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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=20040919094602.7D09890DD@zion.localdomain \
    --to=blaisorblade_spam@yahoo.it \
    --cc=jdike@addtoit.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=user-mode-linux-devel@lists.sourceforge.net \
    /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: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®