git.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Junio C Hamano <gitster@pobox.com>
To: "Nguyễn Thái Ngọc Duy" <pclouds@gmail.com>
Cc: git@vger.kernel.org, michelbach94@gmail.com
Subject: Re: [PATCH 7/5] fsck: use streaming interface for large blobs in pack
Date: Tue, 12 Jul 2016 11:39:45 -0700	[thread overview]
Message-ID: <xmqq7fcq66q6.fsf@gitster.mtv.corp.google.com> (raw)
In-Reply-To: <20160710104555.27478-1-pclouds@gmail.com> (=?utf-8?B?Ik5n?= =?utf-8?B?dXnhu4VuIFRow6FpIE5n4buNYw==?= Duy"'s message of "Sun, 10 Jul 2016 12:45:55 +0200")

Nguyễn Thái Ngọc Duy  <pclouds@gmail.com> writes:

> For this particular operation, not unpacking the blob and letting
> check_sha1_signature, which supports streaming interface, do the job
> is sufficient.

That reasoning is sound, I would think, but...

> We will call the callback function "fn" with NULL as "data". The only
> callback of this function is fsck_obj_buffer(), which does not touch
> "data" at all if it's a blob.

... this may be a bit too magical that the assumption needs to be
left in an in-code comment to warn people against temptation to
using "data" over there, not just where the type of verify_fn is
declared.  Essentially, you are changing the expectation from
existing functions of that type.

> Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com>
> ---
>  pack-check.c | 15 +++++++++++++--
>  pack.h       |  1 +
>  2 files changed, 14 insertions(+), 2 deletions(-)
>
> diff --git a/pack-check.c b/pack-check.c
> index 1da89a4..0777766 100644
> --- a/pack-check.c
> +++ b/pack-check.c
> @@ -105,6 +105,8 @@ static int verify_packfile(struct packed_git *p,
>  		void *data;
>  		enum object_type type;
>  		unsigned long size;
> +		off_t curpos;
> +		int data_valid = 0;
>  
>  		if (p->index_version > 1) {
>  			off_t offset = entries[i].offset;
> @@ -116,8 +118,17 @@ static int verify_packfile(struct packed_git *p,
>  					    sha1_to_hex(entries[i].sha1),
>  					    p->pack_name, (uintmax_t)offset);
>  		}
> -		data = unpack_entry(p, entries[i].offset, &type, &size);
> -		if (!data)
> +
> +		curpos = entries[i].offset;
> +		type = unpack_object_header(p, w_curs, &curpos, &size);
> +		unuse_pack(w_curs);

This made me wonder where this "unuse" comes from; w_curs's in-use
count is incremented by calling unpack_object_header() and we are
done with that access at this point, hence we have unuse here.

> +		if (type != OBJ_BLOB || size < big_file_threshold) {
> +			data = unpack_entry(p, entries[i].offset, &type, &size);
> +			data_valid = 1;

This codepath slurps the data in-core to hash and data is later
freed, i.e. non-blob objects and small ones are handled as before.

> +		}
> +
> +		if (data_valid && !data)
>  			err = error("cannot unpack %s from %s at offset %"PRIuMAX"",
>  				    sha1_to_hex(entries[i].sha1), p->pack_name,
>  				    (uintmax_t)entries[i].offset);

Otherwise, we'd go to check_sha1_signature() with map==NULL.  And
that is exactly what we want---map==NULL is the way we tell the
function to use the streaming interface to check.

Good.

> diff --git a/pack.h b/pack.h
> index 3223f5a..0e77429 100644
> --- a/pack.h
> +++ b/pack.h
> @@ -74,6 +74,7 @@ struct pack_idx_entry {
>  
>  
>  struct progress;
> +/* Note, the data argument could be NULL if object type is blob */
>  typedef int (*verify_fn)(const unsigned char*, enum object_type, unsigned long, void*, int*);
>  
>  extern const char *write_idx_file(const char *index_name, struct pack_idx_entry **objects, int nr_objects, const struct pack_idx_option *, const unsigned char *sha1);

  reply	other threads:[~2016-07-12 18:39 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-06-24 22:38 [bug] Reliably Reproducible Bad Packing of Objects Christoph Michelbach
2016-07-02  9:10 ` Duy Nguyen
2016-07-02 14:35   ` Duy Nguyen
2016-07-05 17:05 ` [PATCH 0/5] Number truncation with 4+ GB files on 32-bit systems Nguyễn Thái Ngọc Duy
2016-07-05 17:05   ` [PATCH 1/5] pack-objects: pass length to check_pack_crc() without truncation Nguyễn Thái Ngọc Duy
2016-07-12 17:16     ` Junio C Hamano
2016-07-05 17:05   ` [PATCH 2/5] sha1_file.c: use type off_t* for object_info->disk_sizep Nguyễn Thái Ngọc Duy
2016-07-12 17:20     ` Junio C Hamano
2016-07-12 19:26     ` Junio C Hamano
2016-07-05 17:05   ` [PATCH 3/5] index-pack: correct "len" type in unpack_data() Nguyễn Thái Ngọc Duy
2016-07-05 20:25     ` Johannes Sixt
2016-07-06 15:25       ` Duy Nguyen
2016-07-06 16:04         ` Junio C Hamano
2016-07-06 16:08           ` Junio C Hamano
2016-07-05 17:05   ` [PATCH 4/5] index-pack: report correct bad object offsets even if they are large Nguyễn Thái Ngọc Duy
2016-07-12 17:24     ` Junio C Hamano
2016-07-12 19:27     ` Junio C Hamano
2016-07-05 17:05   ` [PATCH 5/5] index-pack: correct "offset" type in unpack_entry_data() Nguyễn Thái Ngọc Duy
2016-07-05 18:11   ` [PATCH 0/5] Number truncation with 4+ GB files on 32-bit systems Christoph Michelbach
     [not found]   ` <1467756891.4798.1.camel@gmail.com>
     [not found]     ` <CACsJy8BDQbanGsf=3z3K-OuH0++EuqQFEB22udXJT+WZnFKSBg@mail.gmail.com>
2016-07-06 18:02       ` Christoph Michelbach
2016-07-06 18:54         ` Duy Nguyen
2016-07-10 10:41   ` Duy Nguyen
2016-07-10 10:42   ` [PATCH 6/5] pack-objects: do not truncate result in-pack object size " Nguyễn Thái Ngọc Duy
2016-07-10 10:45   ` [PATCH 7/5] fsck: use streaming interface for large blobs in pack Nguyễn Thái Ngọc Duy
2016-07-12 18:39     ` Junio C Hamano [this message]
2016-07-12 19:06       ` Junio C Hamano
2016-07-12 17:07   ` [PATCH 0/5] Number truncation with 4+ GB files on 32-bit systems Junio C Hamano
2016-07-12 18:48     ` Junio C Hamano
2016-07-12 20:38   ` Junio C Hamano
2016-07-13  6:01     ` Duy Nguyen
2016-07-13 15:43   ` [PATCH v2 0/7] " Nguyễn Thái Ngọc Duy
2016-07-13 15:43     ` [PATCH v2 1/7] pack-objects: pass length to check_pack_crc() without truncation Nguyễn Thái Ngọc Duy
2016-07-13 15:43     ` [PATCH v2 2/7] sha1_file.c: use type off_t* for object_info->disk_sizep Nguyễn Thái Ngọc Duy
2016-07-13 15:44     ` [PATCH v2 3/7] index-pack: correct "len" type in unpack_data() Nguyễn Thái Ngọc Duy
2016-07-13 15:44     ` [PATCH v2 4/7] index-pack: report correct bad object offsets even if they are large Nguyễn Thái Ngọc Duy
2016-07-13 15:44     ` [PATCH v2 5/7] index-pack: correct "offset" type in unpack_entry_data() Nguyễn Thái Ngọc Duy
2016-07-13 15:44     ` [PATCH v2 6/7] pack-objects: do not truncate result in-pack object size on 32-bit systems Nguyễn Thái Ngọc Duy
2016-07-13 15:44     ` [PATCH v2 7/7] fsck: use streaming interface for large blobs in pack Nguyễn Thái Ngọc Duy
2016-07-13 16:16     ` [PATCH v2 0/7] Number truncation with 4+ GB files on 32-bit systems Junio C Hamano

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=xmqq7fcq66q6.fsf@gitster.mtv.corp.google.com \
    --to=gitster@pobox.com \
    --cc=git@vger.kernel.org \
    --cc=michelbach94@gmail.com \
    --cc=pclouds@gmail.com \
    /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;
as well as URLs for NNTP newsgroup(s).