* [PATCH] zram: Prevent use of unmapped buffer
@ 2012-11-27 7:26 Nitin Gupta
2012-11-27 7:52 ` Tomas M
2012-11-28 5:15 ` Minchan Kim
0 siblings, 2 replies; 6+ messages in thread
From: Nitin Gupta @ 2012-11-27 7:26 UTC (permalink / raw)
To: Greg KH
Cc: Seth Jennings, Minchan Kim, Dan Carpenter, Sam Hansen, Tomas M,
Mihail Kasadjikov, Linux Driver Project, linux-kernel
The commit c8f2f0db1 ("zram: Fix handling of incompressible pages")
introduced a bug which caused a kunmap()'ed buffer to be used in case
of partial writes where the data was found to be incompressible.
This fixes bug 50081:
https://bugzilla.kernel.org/show_bug.cgi?id=50081
Signed-off-by: Nitin Gupta <ngupta@vflare.org>
Reported-by: Mihail Kasadjikov <hamer.mk@gmail.com>
Reported-by: Tomas M <tomas@slax.org>
---
drivers/staging/zram/zram_drv.c | 39 ++++++++++++++++++++++++---------------
1 file changed, 24 insertions(+), 15 deletions(-)
diff --git a/drivers/staging/zram/zram_drv.c b/drivers/staging/zram/zram_drv.c
index fb4a7c9..f2a73bd 100644
--- a/drivers/staging/zram/zram_drv.c
+++ b/drivers/staging/zram/zram_drv.c
@@ -265,7 +265,7 @@ out_cleanup:
static int zram_bvec_write(struct zram *zram, struct bio_vec *bvec, u32 index,
int offset)
{
- int ret;
+ int ret = 0;
size_t clen;
unsigned long handle;
struct page *page;
@@ -286,10 +286,8 @@ static int zram_bvec_write(struct zram *zram, struct bio_vec *bvec, u32 index,
goto out;
}
ret = zram_decompress_page(zram, uncmem, index);
- if (ret) {
- kfree(uncmem);
+ if (ret)
goto out;
- }
}
/*
@@ -302,16 +300,18 @@ static int zram_bvec_write(struct zram *zram, struct bio_vec *bvec, u32 index,
user_mem = kmap_atomic(page);
- if (is_partial_io(bvec))
+ if (is_partial_io(bvec)) {
memcpy(uncmem + offset, user_mem + bvec->bv_offset,
bvec->bv_len);
- else
+ kunmap_atomic(user_mem);
+ user_mem = NULL;
+ } else {
uncmem = user_mem;
+ }
if (page_zero_filled(uncmem)) {
- kunmap_atomic(user_mem);
- if (is_partial_io(bvec))
- kfree(uncmem);
+ if (!is_partial_io(bvec))
+ kunmap_atomic(user_mem);
zram_stat_inc(&zram->stats.pages_zero);
zram_set_flag(zram, index, ZRAM_ZERO);
ret = 0;
@@ -321,9 +321,11 @@ static int zram_bvec_write(struct zram *zram, struct bio_vec *bvec, u32 index,
ret = lzo1x_1_compress(uncmem, PAGE_SIZE, src, &clen,
zram->compress_workmem);
- kunmap_atomic(user_mem);
- if (is_partial_io(bvec))
- kfree(uncmem);
+ if (!is_partial_io(bvec)) {
+ kunmap_atomic(user_mem);
+ user_mem = NULL;
+ uncmem = NULL;
+ }
if (unlikely(ret != LZO_E_OK)) {
pr_err("Compression failed! err=%d\n", ret);
@@ -332,8 +334,10 @@ static int zram_bvec_write(struct zram *zram, struct bio_vec *bvec, u32 index,
if (unlikely(clen > max_zpage_size)) {
zram_stat_inc(&zram->stats.bad_compress);
- src = uncmem;
clen = PAGE_SIZE;
+ src = NULL;
+ if (is_partial_io(bvec))
+ src = uncmem;
}
handle = zs_malloc(zram->mem_pool, clen);
@@ -345,7 +349,11 @@ static int zram_bvec_write(struct zram *zram, struct bio_vec *bvec, u32 index,
}
cmem = zs_map_object(zram->mem_pool, handle, ZS_MM_WO);
+ if ((clen == PAGE_SIZE) && !is_partial_io(bvec))
+ src = kmap_atomic(page);
memcpy(cmem, src, clen);
+ if ((clen == PAGE_SIZE) && !is_partial_io(bvec))
+ kunmap_atomic(src);
zs_unmap_object(zram->mem_pool, handle);
@@ -358,9 +366,10 @@ static int zram_bvec_write(struct zram *zram, struct bio_vec *bvec, u32 index,
if (clen <= PAGE_SIZE / 2)
zram_stat_inc(&zram->stats.good_compress);
- return 0;
-
out:
+ if (is_partial_io(bvec))
+ kfree(uncmem);
+
if (ret)
zram_stat64_inc(zram, &zram->stats.failed_writes);
return ret;
--
1.7.10.4
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [PATCH] zram: Prevent use of unmapped buffer
2012-11-27 7:26 [PATCH] zram: Prevent use of unmapped buffer Nitin Gupta
@ 2012-11-27 7:52 ` Tomas M
2012-11-28 5:15 ` Minchan Kim
1 sibling, 0 replies; 6+ messages in thread
From: Tomas M @ 2012-11-27 7:52 UTC (permalink / raw)
To: Nitin Gupta
Cc: Greg KH, Seth Jennings, Minchan Kim, Dan Carpenter, Sam Hansen,
Mihail Kasadjikov, Linux Driver Project, linux-kernel
Good work. The patch fixes a serious bug on 3.6.5, 3.6.6, 3.6.7 and 3.6.8
I hope it gets mainlined ASAP
Tomas M
slax.org
On Tue, Nov 27, 2012 at 8:26 AM, Nitin Gupta <ngupta@vflare.org> wrote:
> The commit c8f2f0db1 ("zram: Fix handling of incompressible pages")
> introduced a bug which caused a kunmap()'ed buffer to be used in case
> of partial writes where the data was found to be incompressible.
>
> This fixes bug 50081:
> https://bugzilla.kernel.org/show_bug.cgi?id=50081
>
> Signed-off-by: Nitin Gupta <ngupta@vflare.org>
> Reported-by: Mihail Kasadjikov <hamer.mk@gmail.com>
> Reported-by: Tomas M <tomas@slax.org>
> ---
> drivers/staging/zram/zram_drv.c | 39 ++++++++++++++++++++++++---------------
> 1 file changed, 24 insertions(+), 15 deletions(-)
>
> diff --git a/drivers/staging/zram/zram_drv.c b/drivers/staging/zram/zram_drv.c
> index fb4a7c9..f2a73bd 100644
> --- a/drivers/staging/zram/zram_drv.c
> +++ b/drivers/staging/zram/zram_drv.c
> @@ -265,7 +265,7 @@ out_cleanup:
> static int zram_bvec_write(struct zram *zram, struct bio_vec *bvec, u32 index,
> int offset)
> {
> - int ret;
> + int ret = 0;
> size_t clen;
> unsigned long handle;
> struct page *page;
> @@ -286,10 +286,8 @@ static int zram_bvec_write(struct zram *zram, struct bio_vec *bvec, u32 index,
> goto out;
> }
> ret = zram_decompress_page(zram, uncmem, index);
> - if (ret) {
> - kfree(uncmem);
> + if (ret)
> goto out;
> - }
> }
>
> /*
> @@ -302,16 +300,18 @@ static int zram_bvec_write(struct zram *zram, struct bio_vec *bvec, u32 index,
>
> user_mem = kmap_atomic(page);
>
> - if (is_partial_io(bvec))
> + if (is_partial_io(bvec)) {
> memcpy(uncmem + offset, user_mem + bvec->bv_offset,
> bvec->bv_len);
> - else
> + kunmap_atomic(user_mem);
> + user_mem = NULL;
> + } else {
> uncmem = user_mem;
> + }
>
> if (page_zero_filled(uncmem)) {
> - kunmap_atomic(user_mem);
> - if (is_partial_io(bvec))
> - kfree(uncmem);
> + if (!is_partial_io(bvec))
> + kunmap_atomic(user_mem);
> zram_stat_inc(&zram->stats.pages_zero);
> zram_set_flag(zram, index, ZRAM_ZERO);
> ret = 0;
> @@ -321,9 +321,11 @@ static int zram_bvec_write(struct zram *zram, struct bio_vec *bvec, u32 index,
> ret = lzo1x_1_compress(uncmem, PAGE_SIZE, src, &clen,
> zram->compress_workmem);
>
> - kunmap_atomic(user_mem);
> - if (is_partial_io(bvec))
> - kfree(uncmem);
> + if (!is_partial_io(bvec)) {
> + kunmap_atomic(user_mem);
> + user_mem = NULL;
> + uncmem = NULL;
> + }
>
> if (unlikely(ret != LZO_E_OK)) {
> pr_err("Compression failed! err=%d\n", ret);
> @@ -332,8 +334,10 @@ static int zram_bvec_write(struct zram *zram, struct bio_vec *bvec, u32 index,
>
> if (unlikely(clen > max_zpage_size)) {
> zram_stat_inc(&zram->stats.bad_compress);
> - src = uncmem;
> clen = PAGE_SIZE;
> + src = NULL;
> + if (is_partial_io(bvec))
> + src = uncmem;
> }
>
> handle = zs_malloc(zram->mem_pool, clen);
> @@ -345,7 +349,11 @@ static int zram_bvec_write(struct zram *zram, struct bio_vec *bvec, u32 index,
> }
> cmem = zs_map_object(zram->mem_pool, handle, ZS_MM_WO);
>
> + if ((clen == PAGE_SIZE) && !is_partial_io(bvec))
> + src = kmap_atomic(page);
> memcpy(cmem, src, clen);
> + if ((clen == PAGE_SIZE) && !is_partial_io(bvec))
> + kunmap_atomic(src);
>
> zs_unmap_object(zram->mem_pool, handle);
>
> @@ -358,9 +366,10 @@ static int zram_bvec_write(struct zram *zram, struct bio_vec *bvec, u32 index,
> if (clen <= PAGE_SIZE / 2)
> zram_stat_inc(&zram->stats.good_compress);
>
> - return 0;
> -
> out:
> + if (is_partial_io(bvec))
> + kfree(uncmem);
> +
> if (ret)
> zram_stat64_inc(zram, &zram->stats.failed_writes);
> return ret;
> --
> 1.7.10.4
>
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] zram: Prevent use of unmapped buffer
2012-11-27 7:26 [PATCH] zram: Prevent use of unmapped buffer Nitin Gupta
2012-11-27 7:52 ` Tomas M
@ 2012-11-28 5:15 ` Minchan Kim
2012-11-29 1:33 ` Minchan Kim
1 sibling, 1 reply; 6+ messages in thread
From: Minchan Kim @ 2012-11-28 5:15 UTC (permalink / raw)
To: Nitin Gupta
Cc: Greg KH, Seth Jennings, Dan Carpenter, Sam Hansen, Tomas M,
Mihail Kasadjikov, Linux Driver Project, linux-kernel
Hi Nitin,
On Mon, Nov 26, 2012 at 11:26:07PM -0800, Nitin Gupta wrote:
> The commit c8f2f0db1 ("zram: Fix handling of incompressible pages")
> introduced a bug which caused a kunmap()'ed buffer to be used in case
I got confused by the descripton. :(
The descripton is not right.
The problem is to access freed memory, not accessing to kunmaped buffer.
partial I/O write.
1. uncmem = kmalloc
2. zram_decompress_page(uncmem)
3. memcpy(uncmem, user_mem)
4. lzo1x_1_compress(uncmem)
5. kfree(uncmem)
6. src = uncmem
7. memcpy(cmem, src, clen) <----- HIT
> of partial writes where the data was found to be incompressible.
>
> This fixes bug 50081:
> https://bugzilla.kernel.org/show_bug.cgi?id=50081
>
> Signed-off-by: Nitin Gupta <ngupta@vflare.org>
> Reported-by: Mihail Kasadjikov <hamer.mk@gmail.com>
> Reported-by: Tomas M <tomas@slax.org>
Good catch! Shame on me. I should have reivewed more carefully. :(
Please resend it with revised descrption and title.
I will rebase mh patchset just sent on top of this bug fix patch.
P.S) Sigh, Now code isn't clean due to partial read/write path handling.
IMHO, sooner or later, we need refactoring.
--
Kind regards,
Minchan Kim
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] zram: Prevent use of unmapped buffer
2012-11-28 5:15 ` Minchan Kim
@ 2012-11-29 1:33 ` Minchan Kim
2012-11-29 1:45 ` Nitin Gupta
0 siblings, 1 reply; 6+ messages in thread
From: Minchan Kim @ 2012-11-29 1:33 UTC (permalink / raw)
To: Nitin Gupta
Cc: Greg KH, Seth Jennings, Dan Carpenter, Sam Hansen, Tomas M,
Mihail Kasadjikov, Linux Driver Project, linux-kernel
On Wed, Nov 28, 2012 at 02:15:05PM +0900, Minchan Kim wrote:
> Hi Nitin,
>
> On Mon, Nov 26, 2012 at 11:26:07PM -0800, Nitin Gupta wrote:
> > The commit c8f2f0db1 ("zram: Fix handling of incompressible pages")
> > introduced a bug which caused a kunmap()'ed buffer to be used in case
>
> I got confused by the descripton. :(
> The descripton is not right.
> The problem is to access freed memory, not accessing to kunmaped buffer.
>
> partial I/O write.
>
> 1. uncmem = kmalloc
> 2. zram_decompress_page(uncmem)
> 3. memcpy(uncmem, user_mem)
> 4. lzo1x_1_compress(uncmem)
> 5. kfree(uncmem)
> 6. src = uncmem
> 7. memcpy(cmem, src, clen) <----- HIT
>
> > of partial writes where the data was found to be incompressible.
> >
I got it. You shouldn't mention partial write for proper description
because it was usecase for swap-over-zram so it can't make partial write.
Without partial write mention, your description is right but the problem
I mentioned above is another problem for partial write so we need a another
patch. I will cook.
Thanks.
> > This fixes bug 50081:
> > https://bugzilla.kernel.org/show_bug.cgi?id=50081
> >
> > Signed-off-by: Nitin Gupta <ngupta@vflare.org>
> > Reported-by: Mihail Kasadjikov <hamer.mk@gmail.com>
> > Reported-by: Tomas M <tomas@slax.org>
>
> Good catch! Shame on me. I should have reivewed more carefully. :(
> Please resend it with revised descrption and title.
> I will rebase mh patchset just sent on top of this bug fix patch.
>
> P.S) Sigh, Now code isn't clean due to partial read/write path handling.
> IMHO, sooner or later, we need refactoring.
>
> --
> Kind regards,
> Minchan Kim
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
--
Kind regards,
Minchan Kim
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] zram: Prevent use of unmapped buffer
2012-11-29 1:33 ` Minchan Kim
@ 2012-11-29 1:45 ` Nitin Gupta
2012-11-29 2:08 ` Minchan Kim
0 siblings, 1 reply; 6+ messages in thread
From: Nitin Gupta @ 2012-11-29 1:45 UTC (permalink / raw)
To: Minchan Kim
Cc: Greg KH, Seth Jennings, Dan Carpenter, Sam Hansen, Tomas M,
Mihail Kasadjikov, Linux Driver Project, linux-kernel
On 11/28/2012 05:33 PM, Minchan Kim wrote:
> On Wed, Nov 28, 2012 at 02:15:05PM +0900, Minchan Kim wrote:
>> Hi Nitin,
>>
>> On Mon, Nov 26, 2012 at 11:26:07PM -0800, Nitin Gupta wrote:
>>> The commit c8f2f0db1 ("zram: Fix handling of incompressible pages")
>>> introduced a bug which caused a kunmap()'ed buffer to be used in case
>> I got confused by the descripton. :(
>> The descripton is not right.
>> The problem is to access freed memory, not accessing to kunmaped buffer.
>>
>> partial I/O write.
>>
>> 1. uncmem = kmalloc
>> 2. zram_decompress_page(uncmem)
>> 3. memcpy(uncmem, user_mem)
>> 4. lzo1x_1_compress(uncmem)
>> 5. kfree(uncmem)
>> 6. src = uncmem
>> 7. memcpy(cmem, src, clen) <----- HIT
>>
>>> of partial writes where the data was found to be incompressible.
>>>
> I got it. You shouldn't mention partial write for proper description
> because it was usecase for swap-over-zram so it can't make partial write.
> Without partial write mention, your description is right but the problem
> I mentioned above is another problem for partial write so we need a another
> patch. I will cook.
I will include explanation of both the cases: use of unmapped buffer
and freed buffer, when I resend these patches.
>
>>> This fixes bug 50081:
>>> https://bugzilla.kernel.org/show_bug.cgi?id=50081
>>>
>>> Signed-off-by: Nitin Gupta <ngupta@vflare.org>
>>> Reported-by: Mihail Kasadjikov <hamer.mk@gmail.com>
>>> Reported-by: Tomas M <tomas@slax.org>
>> Good catch! Shame on me. I should have reivewed more carefully. :(
>> Please resend it with revised descrption and title.
>> I will rebase mh patchset just sent on top of this bug fix patch.
>>
>> P.S) Sigh, Now code isn't clean due to partial read/write path handling.
>> IMHO, sooner or later, we need refactoring.
>>
>> --
>> Kind regards,
>> Minchan Kim
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at http://vger.kernel.org/majordomo-info.html
>> Please read the FAQ at http://www.tux.org/lkml/
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] zram: Prevent use of unmapped buffer
2012-11-29 1:45 ` Nitin Gupta
@ 2012-11-29 2:08 ` Minchan Kim
0 siblings, 0 replies; 6+ messages in thread
From: Minchan Kim @ 2012-11-29 2:08 UTC (permalink / raw)
To: Nitin Gupta
Cc: Greg KH, Seth Jennings, Dan Carpenter, Sam Hansen, Tomas M,
Mihail Kasadjikov, Linux Driver Project, linux-kernel
On Thu, Nov 29, 2012 at 10:45 AM, Nitin Gupta <ngupta@vflare.org> wrote:
> On 11/28/2012 05:33 PM, Minchan Kim wrote:
>>
>> On Wed, Nov 28, 2012 at 02:15:05PM +0900, Minchan Kim wrote:
>>>
>>> Hi Nitin,
>>>
>>> On Mon, Nov 26, 2012 at 11:26:07PM -0800, Nitin Gupta wrote:
>>>>
>>>> The commit c8f2f0db1 ("zram: Fix handling of incompressible pages")
>>>> introduced a bug which caused a kunmap()'ed buffer to be used in case
>>>
>>> I got confused by the descripton. :(
>>> The descripton is not right.
>>> The problem is to access freed memory, not accessing to kunmaped buffer.
>>>
>>> partial I/O write.
>>>
>>> 1. uncmem = kmalloc
>>> 2. zram_decompress_page(uncmem)
>>> 3. memcpy(uncmem, user_mem)
>>> 4. lzo1x_1_compress(uncmem)
>>> 5. kfree(uncmem)
>>> 6. src = uncmem
>>> 7. memcpy(cmem, src, clen) <----- HIT
>>>
>>>> of partial writes where the data was found to be incompressible.
>>>>
>> I got it. You shouldn't mention partial write for proper description
>> because it was usecase for swap-over-zram so it can't make partial write.
>> Without partial write mention, your description is right but the problem
>> I mentioned above is another problem for partial write so we need a
>> another
>> patch. I will cook.
>
>
> I will include explanation of both the cases: use of unmapped buffer
> and freed buffer, when I resend these patches.
>
As you know, I am going to solve lockdep problem and disksize setting problem.
Apparently, it conflict with your diet series.
So I would like to order our patches following as.
1. bug fix patch of unmapped/freed buffer - Nitin
2. diet patch based on 1 - Nitin
3. disksize setting and lockdep problem - Minchan
I will wait your two patchsets.
If you have some trouble to hold the time to make the patch, please tell me.
I will send 1 and 3 ahead so you can send 2 based on it.
--
Kind regards,
Minchan Kim
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2012-11-29 2:08 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-11-27 7:26 [PATCH] zram: Prevent use of unmapped buffer Nitin Gupta
2012-11-27 7:52 ` Tomas M
2012-11-28 5:15 ` Minchan Kim
2012-11-29 1:33 ` Minchan Kim
2012-11-29 1:45 ` Nitin Gupta
2012-11-29 2:08 ` Minchan Kim
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox