evas_image_load.c's list was updated to match the generic
loaders, in 38dd405712.
The list used by cserve should be the same. Actually, there
should be a common function instead of code duplication here.
Summary:
Selection handlers are now implemented in elementary. So, we should remove them from edje.
The BLOCK_HANDLE mode should be removed also.
Reviewers: cedric, tasn
CC: cedric, raster
Differential Revision: https://phab.enlightenment.org/D312
Fix deffect with a missing tokken 'name' in the program description block when
generated by edje_cc.
Reviewers: cedric, seoz
CC: cedric
Differential Revision: https://phab.enlightenment.org/D323
Signed-off-by: Cedric Bail <cedric.bail@samsung.com>
In the case we didn't define the file to save to, fopen would have
been called with a NULL parameter. The result of that call is not
really defined and could have led to some crash at some point.
Fix deffect with a missing tokken 'state' in the first description block when
generated by edje_cc (It is perfectly valid to not specify the first description
name as it will always be "default").
Reviewers: cedric, seoz
Reviewed By: cedric
Differential Revision: https://phab.enlightenment.org/D310
Signed-off-by: Cedric Bail <cedric.bail@samsung.com>
Update the data_process_lookups function.
Earlier images are not deleted, but his name was
removed from the list, and it was possible to
access the image by id. Now all unused images
are deleted.
Reviewers: cedric, seoz
Differential Revision: https://phab.enlightenment.org/D309
Signed-off-by: Cedric Bail <cedric.bail@samsung.com>
Since cserve2 uses inotify to track image file updates,
it will drop its references to a specific file and all
the associated images.
Fix some logic in the deletion code.
Nothing extraordinary here.
Most potential crashes are extremely unlikely.
- Fix CID 1113444
- Fix CID 1113442
- Fix CID 1113441 (Logically dead code, can not be NULL)
- Fix CID 1113440: Explicit null dereferenced
This is actually an impossible situation.
Fixed by checking for nullity and printing out some error
messages instead of just crashing.
- Fix CID 1113439: Dereference after null check
Logically impossible code as both idxpath and datapath
must be either set or null at the same time.
Change the if logic to tell Coverity there's no bug.
- Fix CID 1113438 (Argument cannot be negative)
Fix wrong check of return value from shm_open.
- Fix CID 1113437 (Argument cannot be negative)
Fix wrong check of return value from shm_open.
- Fix CID 1113436 (Dereference null return value)
This case really shouldn't happen.
But the extra check does not hurt.
- Fix CID 1113435 (Dereference before null check)
Check for nullity after map open.
- Fix CID 1113434 (Extra sizeof expression)
Debug buggy debug tool :)
- Fix CID 1113433 (Uninitialized scalar variable)
Insignificant issue: only prints wrong debug logs :)
- Fix CID 1113431 (Uninitialized scalar value)
Check if (!found) only to print out logs. Not a big deal
if found was invalid.
- Fix CID 1039462 (Logically dead code)
Glyphs were previously using 3 shared buffers, now reduce to 2:
- Memory pool (mempool) containing the glyph drawable data
- Index table (Shared_Index / array) containing only the
indexes of the buffers in the mempool
- Glyph_Data table (array) containing the glyphs descriptors
AS WELL as the buffer indexes.
So, we just merge the two index tables into one by using directly
objects of type Glyph_Data for the referencing of the mempool
buffers.
One socket message was sent per each glyph used ... which means
a LOT of messages when text is being redrawn.
Reduce this flow of messages by triggering send() only when
50+ items are being used.
Btw, USED is a bit useless as there is no UNUSED equivalent.
Also, slightly improve debug logs.
SIGTERM and SIGQUIT don't need to be caught.
the proper signal to ask cserve2 to exit cleanly is SIGINT.
enlightenment_start (or systemd) should be responsible for
restarting in case of crash / unexpected termination.
cserve2 will not restart if killed with signal SIGINT,
as it will then exit with return code 0 (normal)
These operations have tons of side effects and it's a lot
easier to just avoid doing them. Now, repacking will always
need to happen as applications will add/delete strings and
items, but the less frequent, the better :)
Also, align most arrays & mempools to 32K instead of the
default page size (4K). This will also reduce resizes.
Try to reconnect to cserve2 if the socket connection was lost.
Resend some messages if necessary.
Images reload seems to be working.
Actually, the images don't change over time, so the clients just
keep the previous references to their images.
FONT RELOAD IS NOT WORKING:
- Crashes
- Invalid glyph data
- Infinite loop in _glyph_map_remap_check()
Root cause:
When new glyphs are requested from the server, they are added to
the mempool. So it is necessary to remap the font.
Unfortunately, in case of server reboot, we did not keep the mempool
so the old glyphs that were not requested again will not be valid.
Be more careful when selecting the original image for scaling.
Most load opts must remain the same, only smooth can be toyed with.
Also, fix alpha when scaling.
Let's reuse the logic from scalecache and call cserve2
functions when the scalecache should be used.
So, now, cserve2 server will not scale any image... This is
too computationally intensive for the server's main thread.
This is not optimal but makes a hell of a lot more sense for
the moment. (since cserve2 manages the SHM segments)
High-level problem:
cserve2 does not support load_opts properly when opening an image.
As a result, when (pre)loading a JPEG file with specific load
options (eg. w, h, region and orientation), the image buffer might
have the wrong dimensions.
So, we need to use load_opts when computing file hash key.
And, pass these load options to the loader slave,
and use them while OPENING the image. This will set
properly the geometry.
Fixes test "Preload and Prescale" in elementary_test.
Pass around "animated" flag for images that can be animated.
Fallback to local cache if the image is animated.
Implementing support for animated images in cserve2 does
not seem to make a lot of sense considering each frame must
be requested independently in real time,... and to be honest
there doesn't seem to be any valid use case anyway :)
Images and Files can be discovered by the client scanner before they
are valid (aka. loaded or opened). We want clients to ignore all
shared objects that are not ready yet, as they are in an undefined
state (values and memory might be invalid).
Shared string indexes are not repacked, since they live in a
memory pool (where the indexes can be reused).
So, the index in the table is equal to their ID. Add initial
test to check if the item at index n has the ID n.
Use tags such as INDX, FONT, IMAG, FILE.
These can help debugging and for robustness, as they will
be visible in the memory dump.
Update generation_id after repack as well
Also, print shared arrays usage as it is sent as first message
upon connection to the server.
Add fonts to LRU and flush LRU as needed.
Seems to work okay. Could probably be finetuned a bit.
- Create one Glyph_Data array per Font_Entry
- Copy Font_Source descriptors in shared array
- Copy font entries in shared index
- Send font index path over socket
- Merge Font_Source and Font_Data in shared memory
- Send font index path on client connect
- Repack font index after free
- Pass font shm name to clients
DBG and INF messages pollute the logs and performance reports
from valgrind, let's disable them at build-time for better
benchmarking.
Fix other compilation warnings and clean code a bit
If the client exits while the server is loading fonts,
the font entry objects are accessed after being freed.
Note: This is not a proper fix. The race condition should still
be fixed, but delaying client deletion a bit might help for a while...
- Send paths to all connected clients
- Send paths upon connection from a new client
- Client side:
- Fast find opened images
- Fast find loaded images
- Optimize load by skipping socket wait
Also, remap indexes when needed (file size changed)
When index file grows, the clients are not notified, so they
must constantly check that their mapping matches the current array
size.
- Move required struct definitions to common evas_cs2.h
- Open/close shm and map arrays to memory
Pretty stupid for now as it will close & reopen the shm files
even when unchanged.
- Check shm size and array header consistency
- Implement message send
- Send messages from server when calling repack()
Only import the C file for now.
Implement the following features:
- Shared Arrays
Store arrays of elements of fixed size in shm.
- Shared Mempool
Store random sized buffers in shm.
These buffers are indexed in a Shared Array and are
referred to using their index only.
- Shared Strings
Store strings in a shm in a way similar to Eina_Stringshare
(except strings are referred to using an int index).
- Include evas_cserve2_index.c to the compilation.
- Declare shared index functions in header file.
- Call init() and shutdown() on the shared index subsystem.
- Add find and foreach functions
- cserve2_shm_segment_request()
Creates a new shm segment (Shm_Handle) from an existing file,
without changing the current mmap.
- cserve2_shm_resize()
Grows the file and remaps in memory (new virtual address).
- cserve2_shm_id_get()
Returns the shm creation ID.
- Store result from mkstemp so we don't leak file descriptor
- No need to chmod, as we set correct umask before createing file
- Use fdopen to open file from file descriptor, not from path
This reverts commit bd5a5e7fce.
sorry about this... all fine eina mkstemp use and all... but this
*BREAKS EFREET*! ibar icons stop working (vanish). efreet no longer
issues a desktop file update event to e etc. etc. etc. ... so let's
keep code that works for now even if it uses mkstemp and not eina...
:) working is better. :)
It is quite common that in an image sets each image has different border size.
This patch permit to define the border value on a per image basis in the set.
NB: If we are going to exit data_thread_script function due to malloc
failure or fread failure, then we should close the previously fdopen'd
file so that we do not leak 'f'.
Signed-off-by: Chris Michael <cp.michael@samsung.com>
If we have file And fname, then we can call the source_fetch_file
function, HOWEVER we may only have One or the Other in which case we
still need to potentially free the other.
NB: Fixes Coverity CID1039296
Signed-off-by: Chris Michael <cp.michael@samsung.com>
In its current model, cserve2 will reopen, reload and scale an
image whenever a new scaled version is requested by the client.
Instead, we can load the original image when too many scaled
versions are requested, and place the original in the LRU.
Remove Image_Load_Opts and use Evas_Image_Load_Opts for cs2 master to
slave communication.
2 things to note:
- The struct size is bigger (but at the same time we removed a memcpy)
- The scale_load field is not passed to the slave
When opening an image, OPEN and SETOPTS were always sent
consecutively. Merging these two messages may improve the
performance a bit (not measured), but also simplify the
whole thing.
Note: cserve2 debug tools have not been fixed yet.
Since the load options are not copied but the pointer is simply
stored by the module, we need to move the struct higher in the
stack.
Signed-off-by: Cedric Bail <cedric.bail@samsung.com>
Most of the time, the SETOPT message is sent only
to carry 0 data. Let's add a flag to the OPEN message to
avoid useless socket calls.
Server-side: implement with message faking.
Also, fix image ID logic on client side:
- alloc data entry asap,
- assign image_id during call to OPEN
Signed-off-by: Cedric Bail <cedric.bail@samsung.com>
Some hashtable lookups failed because the keys were added
either with eina_stringshare_add() or e_s_add_lenth() but
looked for with e_s_add_length() only.
Signed-off-by: Cedric Bail <cedric.bail@samsung.com>